WHO technical manual on tobacco tax policy and administration

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION WHO technical manual on tobacco tax policy and administration ISBN 978-92-4-001918-8 (electronic version) ISBN 978-92-4-001919-5 (print version) © World Health Organization 2021 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/ rules/). Suggested citation. WHO technical manual on tobacco tax policy and administration. Geneva: World Health Organization; 2021. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Document design by Ana Sabino. iii Contents Foreword v Acknowledgements vii Acronyms viii Executive summary xi CHAPTER 1. Why this manual? 1 CHAPTER 2. Tobacco excise tax policy 11 2.1 Global overview of tobacco tax practices 11 2.2 Designing excise tax policy 18 2.3 Domestic and regional policy integration 55 2.4 New and emerging nicotine and tobacco products 59 2.5 Conclusions 75 ANNEX 2.1 Countries that apply different types of excise tax structures 89 ANNEX 2.2 Analytics of the tax base elasticity 90 ANNEX 2.3 Elements of the devices that make up ENDS/ENNDS products 91 CHAPTER 3. Tobacco tax administration 93 3.1 Introduction 93 3.2 Institutional arrangements 94 3.3 The tax compliance cycle 102 3.4 Control and enforcement 115 3.5 Tax administration of other tobacco products 152 3.6 The broader elements of a good tax system 155 3.7 Conclusions 156 ANNEX 3.1 Composition of tobacco products 166 ANNEX 3.2 Example of forestalling and countermeasures 172 CHAPTER 4. Political economy 175 4.1 SCARE tactic S: Smuggling and illicit trade 176 4.2 SCARE tactic C: Court and legal challenges 198 4.3 SCARE tactic A: Anti-poor rhetoric or regressivity 212 4.4 SCARE tactic R: Revenue reduction 217 4.5 SCARE tactic E: Employment impact 228 4.6 Earmarking tobacco tax revenues to fund health 233 ANNEX 4.1 Methods to assess the nature and size of the illicit tobacco trade 253 ANNEX 4.2 How are the tobacco tax revenues earmarked? 269 CHAPTER 5. Best practices 271 TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) 283 iv v Foreword In 1999, the World Bank’s Curbing the Epidemic was the first report by an interna- tional organization to recognize that increasing tobacco excise taxes was the most effective and cost-effective measure to reduce tobacco use and save lives. Over the two decades since, the evidence base supporting this claim, especially in low- and middle-income countries, has been steadily growing. Meanwhile, the credibility of the tobacco industry’s arguments against tobacco taxation has been slowly waning. In short, health-promoting tobacco taxation has come of age, and the evidence has consistently shown that it is a win for public health, a win for revenue and a win for the economy overall. But we must be cautioned against complacency. Although the evidence on tobacco taxation is irrefutable and there are now signs that the tide is turning on the global tobacco epidemic, tobacco taxation was, in 2018, the WHO MPOWER1 measure that was least implemented at the highest level of achievement. Even more concerningly, cigarettes have become more, rather than less, affordable in many low- and middle-income countries over the past decade. Many countries set rates at insufficient levels and increase them too infrequently, while others still use complex and inefficient taxation structures. This failure to advance tobacco taxation able to effect significant price increases constitutes a loss for governments in revenues, a loss for public health and a win for the tobacco industry. To overcome this inertia, this manual charts the way forward for policy-makers, finance officials and others involved in tobacco tax policy development. It equips them with the information and evidence needed for the realization of their coun- tries’ tobacco tax policy objectives. It also analyses the tobacco industry’s tactics for influencing the political economy of tobacco taxation and shows the limitations and exaggerations of the arguments used against tax increases. The manual serves as an update of the 2010 WHO technical manual on tobacco tax administration by adding new evidence on the successes of tobacco taxation in all parts of the world and broadening its scope to capture more material relevant to developing and implementing more effective tobacco tax policy. 1 The WHO MPOWER package of technical measures and resources that comprises (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco. vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Its contributions are particularly timely in a COVID-19-stricken world. As the pandemic has been worsened by the global burden of noncommunicable disease, and revenue is now desperately needed, taxing tobacco should be more palatable than ever. This manual shows policy-makers how to seize this unique opportunity to use tobacco taxation to build back better, save lives and strengthen health systems while increasing revenue. Dr Naoko Yamamoto Assistant Director-General UHC/Healthier Populations Division vii Acknowledgements This manual was developed under the direction of Jeremias Paul Jr and Anne-Marie Perucic. The following contributed to the content of the manual (in alphabetical order): • WHO: Evan Blecher, Annerie Bouw, Mark Goodchild, Roberto Iglesias, Juliette McHardy, Jeremias Paul Jr, Anne-Marie Perucic and Robert Totanes. • External authors: Mauricio Cardenas (Columbia University), Sophia Delipalla (University of Macedonia), Luk Joossens (tobacco control expert, Belgium), Marin Kurti (Eastern Connecticut State University), Enrique Fanta (former World Bank senior specialist), David Merriman (University of Illinois at Chicago) and Jean Tesche (University of Cape Town). WHO would like to thank the following reviewers for their invaluable comments (in alphabetical order): • External: Jo Birckmayer (Bloomberg Philanthropies), Adriana Blanco Mar- quizo (WHO Framework Convention on Tobacco Control), Frank Chaloupka (University of Illinois at Chicago), Yoni Dekker (WHO Framework Conven- tion on Tobacco Control), Jeffrey Drope (University of Illinois at Chicago), Ceren Ozer (World Bank), Corne van Walbeek (University of Cape Town), Chonlathan Visaruthvong (Ministry of Finance, Thailand) and Rodrigo Santos Feijo (WHO Framework Convention on Tobacco Control). • WHO: – Headquarters: Douglas Bettcher, Itziar Belausteguigoitia, Ranti Fayokun, Joseph Kutzin, Benn McGrady, Vinayak Prasad and Susan Sparkes. – Regional offices: Nina Dela Cruz (WPRO), Fatimah El-Awa (EMRO), Charles Frasier (EMRO), Lee Lily Joung-Eun (WPRO), Jagdish Kaur (SEARO), Elizaveta Lebedeva (EURO), Maxime Roche (AMRO) and Rosa Sandoval (AMRO). WHO would also like to thank Amal Amoune-Naal for the administrative support, Alison Goldstein for the technical editing and Janet DeLand for the copy-editing. Production of this document has been supported by a grant from Bloomberg Phi- lanthropies. The contents of this document are the sole responsibility of WHO and should not be regarded as reflecting the position of Bloomberg Philanthropies. viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ACRONYMS AEO authorized economic operator AFRO WHO Regional Office for Africa AMRO WHO Regional Office for the Americas ATO Australian Taxation Office BAT British American Tobacco CCTV closed-circuit television CIF cost, insurance and freight COP Conference of the Parties CPI consumer price index CVA Customs Valuation Agreement (Thailand) DGCE Directorate General of Customs and Excise (Indonesia) DIY do-it-yourself ECBA World Bank Extended Cost-Benefit Analysis EIU Economist Intelligence Unit EMRO WHO Regional Office for the Eastern Mediterranean ENDS electronic nicotine delivery systems ENNDS electronic non-nicotine delivery systems EU European Union EURO WHO Regional Office for Europe FDA Food and Drug Administration (United States) FET fair and equitable treatment GCC Cooperation Council for the Arab States of the Gulf GDP gross domestic product HTP heated tobacco product HMRC Her Majesty’s Revenue and Customs (United Kingdom) IARC International Agency for Research on Cancer IIA international investment agreement IMF International Monetary Fund IRS Internal Revenue Service (United States) ISO International Organization for Standardization IT information technology ITC International Tobacco Control JTI Japan Tobacco International KRA Kenya Revenue Authority LMICs low- and middle-income countries ACRONYMS ix MFN most favoured nation MOP Meeting of the Parties (to the Protocol) MPOWER (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco NCDs noncommunicable diseases NCI National Cancer Institute NT national treatment OECD Organisation for Economic Co-operation and Development OST other smoking tobacco PMI Philip Morris International PPP purchasing power parity QR quick response RGTE WHO Report on the global tobacco epidemic RYO roll-your-own SACU Southern African Customs Union SCARE (S) smuggling and illicit trade; (C) court and legal challenges; (A) anti-poor rhetoric; (R) revenue reduction; and (E) employment impact SDGs Sustainable Development Goals SEARO WHO South-East Asia Regional Office SII Internal Revenue Service of Chile TADAT Tax Administration Diagnostic Assessment Tool TTC transnational tobacco company UAE United Arab Emirates UHC universal health coverage VAT value added tax WAEMU West African Economic and Monetary Union WCO World Customs Organization WHO World Health Organization WHO FCTC WHO Framework Convention on Tobacco Control WHO ISPT WHO interactive smoking projection and target-setting tool WHO TaXSiM WHO tobacco tax simulation model WPRO WHO Regional Office for the Western Pacific WTO World Trade Organization x W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N xi Executive summary This WHO technical manual on tobacco tax policy and administration builds upon the 2010 WHO technical manual on tobacco tax administration by further detailing the strategies for effective tobacco tax policy development, design, implementation and administration. This 2021 edition also serves as an update to the 2010 manual, incorporating the latest developments in science, technology and policy, as well as providing illustrative recent examples from a variety of countries. The best practices laid out in this manual are designed to inform governments on the development of their tobacco taxation policy, facilitating the achievement of their health and revenue objectives while also supporting their overall development strategy. Tobacco taxes have long been seen as a source of revenue for governments, but as evidence of the harms caused by tobacco has accumulated over the years, public perception has evolved. Increasingly, governments, as well as the general public, are recognizing that taxation of tobacco is not only a revenue source but also an effective public health intervention to reduce tobacco consumption and its associated harms. The profile of tobacco taxation as a health policy tool has increased greatly since the publication of the 2010 WHO technical manual on tobacco tax administration. Multiple global commitments have been adopted over the past decade to address tobacco use specifically – as well as noncommunicable diseases (NCDs) and the Sustainable Development Goals (SDGs) more broadly – through tax and price measures to reduce demand for tobacco products, save lives and fund develop- ment. Global development institutions, including the World Bank, the International Monetary Fund (IMF) and major philanthropic foundations, also agree with WHO on the importance of emphasizing and strengthening tobacco taxation as a key health policy tool. The COVID-19 pandemic has further fuelled this shift in the narrative on tobacco taxation by revealing how the global economy is inextricably linked with population and planetary health. Investing in health is fundamental to any economic recovery, and fiscal policy will be a key driver in addressing the socioeconomic consequences of COVID-19. Interventions such as tobacco taxation – which leads to reduced tobacco consumption, improved population health and increased revenues for governments – should be part of a comprehensive strategy for a build back better recovery. The evidence is clear: significant increases in excise taxes that lead to price increases have consistently proven to be the most effective, as well as the most xii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cost-effective, mechanism for reducing tobacco consumption. This manual will guide readers through the necessary steps to create and implement the strongest tobacco taxation policies for their specific countries. There are many factors to consider when developing tobacco taxation policy. Understanding the market is a fundamental step before deciding which form of taxation to use. Specifically, the choice between ad valorem and specific taxation is influenced by the market structure in a given country. At the same time, tax structure also shapes the market structure. Over the past decade, there has been a trend towards countries adopting specific excise taxes or mixed excise systems that rely more heavily on the specific component, which the latest global data associate with the highest average prices. Tobacco tax structures can be simple, with one flat rate across the board, or com- plex, with multiple tiers for products with different characteristics. In 2018, 31 coun- tries used complex, multitiered structures. But evidence demonstrates that simpler excise tax structures – utilized in all high-income countries – leave the least room for industry manipulation or tax avoidance and brand/product switching by consumers. Not only is it important to set taxes at a high level to discourage consumption, specific excise tax policies must include regular adjustments to increase the tax rate so that it keeps up with inflation and income growth in a country over time. Excise tax increases should aim to reduce the affordability of tobacco products. The base on which the tax is applied is also important. For specific taxation, the tax base should be the quantity in clearly defined units. For ad valorem (or mixed) taxation, the best practice is to use the retail price as the tax base and introduce a minimum excise tax. With regard to non-tax regulations that affect the price of tobacco products, pric- ing regulation may be considered to prevent the tobacco industry from exercising differential tax shifting, which it uses to ensure that large price gaps exist between premium and cheap cigarettes. However, pricing policies cannot be used alone. If con- sidered, they should be used only as complements to significant excise tax increases. Other non-tax regulations include banning promotional discounts for tobacco products and banning the sale of single cigarettes. To assuage concerns that tax increases will increase inflation – as well as to reflect the declining trend in consump- tion of tobacco products – it is good practice to exclude tobacco products from the basket of items that are used to develop consumer price indexes. Finally, in order to make excise tax on tobacco products more effective in reducing overall tobacco use, all tobacco products must be taxed in a comparable way. Regular assessment, evaluation and monitoring of the impact of tobacco tax policies over time are essential components of effective tax policy development and analysis. Governments need to have accurate estimates of price, income and tax base elasticities in order to anticipate the impact of a tax increase on consumption and EXECUTIVE SUMMARY xiii tax revenue. Ideally, other factors such as non-price policies should also be taken into account when estimating price and income elasticities for a specific country. A variety of tools and indicators exist to measure impact and monitor progress, and these are described in Chapter 2 of this manual. When developing tobacco tax policy, it is also important to take the broader policy context into consideration at both the domestic and the regional level. Domestically, cooperation is needed across sectors to ensure that policies and interventions in the areas of agriculture, trade, finance and labour do not work against the public health objectives of tobacco control and taxation. For countries that are part of a regional bloc, regional harmonization of tobacco taxation is a useful tool to prevent tax revenue erosion, tax avoidance and tax evasion, as well as to protect population health. Tax harmonization must be designed carefully, however, to be effective. The experience of the European Union (EU) demonstrates that both a declining consumption trend and stable revenues can be achieved with harmonized minimum excise tax rates. Discussions of policy development and implementation for new and emerging nicotine and tobacco products such as heated tobacco products (HTPs) and electronic nicotine and non-nicotine delivery systems (ENDS/ENNDS), are complicated, by their constantly changing technology and market dynamics. Policies and regulations need to be developed carefully and adjusted accordingly. Where HTPs are not banned, the current recommendation is to tax them at the same level as cigarettes on a per-unit basis, regardless of tobacco content. Early evidence from the United States shows that demand for e-cigarettes, a subcategory of ENDS/ENNDS products, is possibly even more price-responsive than the demand for conventional cigarettes, meaning that taxes can be used as an effective deterrent to ENDS/ENNDS products use. While there is preliminary evidence of substitutability between conventional cigarette use and e-cigarette use, further research is needed to understand substitutability effects among users of both conventional cigarettes and ENDS/ENNDS products. It is essential to imple- ment regulation of ENDS/ENNDS products along with any tax policy to safeguard public health. In countries where they are not banned, ENDS/ENNDS products must be regu- lated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS taxation. Nicotine-containing and non-nicotine-containing e-liquids should be taxed equally. Ultimately, while the policy implications of these newer products require careful consideration, the fact remains that conventional tobacco products constitute the overwhelming share of consumption (more than 97% in 2018). Tobacco tax administration must be both efficient and effective to ensure that health objectives are met and the desired level of tax revenue is raised. Since the xiv W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation of tobacco taxation often involves numerous agencies within a country, clearly defined roles and responsibilities are essential to maximize efficiency. Coordination among the different agencies involved, as well as with neighbouring countries, is required for tobacco tax administration to be effective. Performance evaluation and accountability for competent authorities is also necessary, and many tools and indicators exist to facilitate these processes (described in Chapter 3). There are a number of steps authorities should take to ensure efficiency and effectiveness at each stage throughout the tax compliance cycle (see Chapter 3, section 3.3). Control and enforcement are the main functions of tax administration, and these can best be achieved through the use of a strategic plan and a risk-based approach. Controls can be exercised through licensing and due diligence, fiscal markings (e.g. tax stamps), tracking and tracing, implementation of anti-forestalling measures, national audits and specific controls for imports and exports, as well as for free zones and transhipment points. Once smuggling or illicit trade is detected, actions such as seizing and destroying smuggled and/or illicit tobacco and col- lecting due taxes must be taken immediately. To deter further illegal activities, a comprehensive audit must also be carried out, including all those involved in the illicit acts. Penalties and sanctions must be sufficient to deter illegal activities. The Protocol to Eliminate Illicit Trade in Tobacco Products provides invaluable guidance for tobacco tax administration, control and enforcement that is applicable even for countries that are not Parties to it. The broader elements of a good tax system include proper resourcing of competent authorities, strict rules and regulations to detect and punish corruption and a strong judiciary system capable of resolving disputes as soon as possible. In its efforts to oppose tobacco tax increases, the tobacco industry utilizes many SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric, (R) revenue reduction and (E) employment impact – to influence the political economy of tobacco. Chapter 4 provides detailed analyses of these issues with supporting evidence that belies the SCARE tactics, as well as guidance for tax and other relevant authorities on how to anticipate and respond to industry arguments. This manual also provides tools and methodologies to help tax authorities define and evaluate the problem of illicit trade of tobacco products in their countries, inde- pendent of the tobacco industry’s generally inflated estimates. Price (and tax) levels are not a key determinant of illicit trade; rather, the problem is exacerbated by the lack of governance and tax administration capacity. Refraining from increasing taxes is not the solution. Countries should instead respond with a comprehensive strategy to fight illicit trade, including undertaking independent estimates of illicit trade levels and implementing good tax administration practices such as those discussed in Chapter 3 and contained in the WHO Framework Convention on Tobacco Con- trol (WHO FCTC) Protocol to Eliminate the Illicit Trade in Tobacco Products. When it comes to court challenges, the tobacco industry is less likely to challenge excise taxes than other tobacco control measures, because taxation is a comparatively well-established regulatory measure. The industry will, however, exploit the slightest vulnerability in the design, adoption or implementation of tax measures. For this reason, measures to strengthen regulators’ legal position are described that will enable authorities to protect themselves from potential legal challenges. The industry argument of regressivity, or the notion that tobacco tax increases hurt the poor because they have to pay a larger share of their income in taxes than the rich, has two fundamental limitations. First, the notion of regressivity does not take into consideration the broader health and economic harms caused by tobacco use that exacerbate the impoverishment of lower-income smokers. These harms are actually reduced when tobacco consumption decreases following a tax increase. Second, the tobacco industry argument ignores the fact that higher tobacco taxes and prices can induce behaviour change as is reflected in the price elasticity of demand. Evidence consistently shows that lower-income smokers are more sensitive to price and therefore more likely to reduce smoking in response to a tax and price increase. Including these factors shows tobacco taxation to be, in fact, a progressive public health intervention that disproportionately benefits the poor. While essentially admitting that a tobacco tax increase may have the desired effect of reducing consumption, the industry also tries to argue that a tax increase will also reduce revenues. In fact, the price inelastic demand for tobacco makes tobacco tax increases a win-win for both public health and finance. This manual presents several country examples that demonstrate how well-designed and well-implemented tobacco tax increases lead to increases rather than decreases in revenue in the short to medium term. In addition, the reduced consumption resulting from a tax increase results in reductions of other tobacco-related government expenditures as well. The final tactic used by the tobacco industry to challenge proposed tax increases is to frame tobacco taxes as an economic rather than a public health issue. This false choice between health and jobs is based on faulty assumptions that 1) tobacco is a significant source of domestic employment; 2) job creation relies on tobacco consumption and 3) tobacco-related livelihoods are prosperous, sustainable and irreplaceable. Earmarking can be a useful tool for improving the political economy of tobacco tax increases. While the primary goal of tobacco tax increases is to reduce demand for tobacco, setting aside portions of tax revenue to fund other tobacco control efforts or relevant health programmes can help convince the public, politicians and officials of the value of significant tobacco tax increases. Earmarking can also be EXECUTIVE SUMMARY xv x vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N used to counter tobacco industry arguments about potential negative effects of tax increases – for example, by supporting tobacco farmers in transitioning to other crops. Tobacco taxes work. This is why the industry invests so much money and effort in blocking large tax increases and other effective tax policy reforms. Policy-makers must not be swayed by industry pressure but need only to follow the facts. This manual provides all the information policy-makers need to make the right deci- sions at each step of the process – from designing, evaluating, implementing and administering tax policy to refuting specious industry attacks and communicating the value of tobacco taxation to legislators and the broader population. An effectively designed and efficiently administered tobacco tax policy will not only produce the direct results of reducing tobacco consumption among smokers and raising revenue for governments, its effects will be felt much more broadly. Indeed, raising tobacco taxes is a SMART policy: it Saves lives; Mobilizes resources; Addresses health inequities; Reduces burdens on health systems; and Targets tobacco use, a major risk factor for NCDs. • Saves lives: Tobacco use is the leading cause of preventable deaths globally – it claims 8 million lives each year. Tobacco taxation is the most effective mechanism for reducing tobacco consumption and its associated health burden worldwide. • Mobilizes resources: Despite being the single most effective tobacco control measure, tobacco taxation is largely underutilized as a policy mechanism. Based on available data on the price and taxation of cigarettes, it is estimated that excise taxes on cigarettes generated a worldwide total of US$ 361 billion in revenues in 2018, including US$ 162 billion in revenues for low- and middle- income countries (LMICs). If all countries were to raise cigarette excise tax rates by the equivalent of US$ 1 per pack, the amount of excise revenue from cigarettes would increase by US$ 178–219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from such tax increases, with excise revenues increasing by 82–103%, providing governments in these countries with an extra US$ 133–167 billion. This shows the substantial revenue potential of tobacco taxes. • Addresses health inequities: Tobacco taxation and tax increases are effectively progressive or pro-poor policies because of their positive distributional impact. Lower-income smokers benefit disproportionately from reduced tobacco consumption and use in terms of health gains and income retention. • Reduces burdens on health systems: The worldwide economic cost of tobacco use was US$ 1.4 trillion in 2012. Tobacco taxes reduce tobacco-related burdens on governments and health systems through population-based preventive measures. • Targets tobacco use: Tobacco taxation directly targets and reduces tobacco use, which is a major risk factor for several deadly NCDs. In summary, significant tobacco tax increases, designed and implemented according to the latest guidance and best practices presented in this technical manual – and as a strong component of a comprehensive tobacco control strategy – will bring about substantial reductions in tobacco use and the health and economic harms it causes. EXECUTIVE SUMMARY xvii x viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 1 CHAPTER 1. Why this manual? BACKGROUND Tobacco taxes are not new. Governments around the world have been applying taxes on tobacco and tobacco products practically since the idea of excise was conceived. And rightly so: tobacco is not a necessity, it is easy to tax and the demand for it is relatively inelastic. These characteristics, along with the substantial revenues tobacco taxes generate, have made tobacco a highly appropriate object of taxation. As evidence of the harms of tobacco has accumulated over the years, the public perception of tobacco taxes has evolved. Now tobacco taxes are not only seen as a revenue source, but, more importantly, they are recognized as an effective public health intervention to reduce tobacco consumption. This trend reflects the reasons excise taxes exist in the first place – to discourage harmful behaviour and to mitigate the associated negative externalities (1–2). Many governments view tobacco taxes as a significant and stable source of rev- enue, which may explain why there is often a degree of hesitation whenever tobacco tax reform is proposed. Historically, many governments have relied on revenues from tobacco taxes and have even adjusted the level of taxation according to their revenue needs (3). However, some countries are beginning to recognize the value of applying high tobacco taxes primarily as a public health tool, viewing revenues as a secondary consideration (4). Arguments against tobacco tax hikes or improvements to the tax structure are often economic in nature: such tax changes will allegedly decrease revenues, wipe out jobs, increase illicit trade and harm local industries, among other claims. But the evidence has consistently shown that such claims are simply not true in an overwhelming majority of situations. The tobacco industry, in particular, frequently portrays this conflict as a false dichotomy between public health and the economy – as if prioritizing health comes at the expense of the economy. In fact, studies and real-world experiences have shown that increasing tobacco taxes not only improves public health but also has a net positive impact on the economy and development of a country – a true win-win scenario (5–6). As an update to the first WHO technical manual on tobacco tax administration published in 2010, this manual aims to help readers better navigate the various 2 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N issues surrounding tobacco taxes and their implementation. The primary intended audience includes policy-makers, finance officials, tax authorities, customs officials and other relevant persons/bodies involved in the formulation and implementation of tobacco tax policy. The manual provides a detailed guide to the design of tobacco tax policy and describes how to effectively administer these taxes to maximize impact. Detailed discussions of the political economy considerations and the hurdles that need to be overcome before and during implementation are included as well. The overarching goal is to equip those working in the tax policy and implementation spheres with sufficient information to help realize the health and revenue objectives of a government’s tobacco tax policy in line with its overall development strategy. THE CONSEQUENCES OF TOBACCO USE AND THE NEED FOR INTERVENTION Most people are aware that smoking and tobacco use are harmful to health, but few truly comprehend the scale of this harm. The tobacco epidemic claimed more than 100 million lives in the last century (7), with updated estimates now reaching 8 million deaths annually from tobacco use and exposure to second-hand smoke (8). As much as 80% of these deaths occur in low- and middle-income countries (LMICs) (6), revealing how the developing world carries much of the global burden. Tobacco use is a major risk factor for many chronic conditions, including heart disease, cancer, diabetes and chronic lung disease – collectively known as noncom- municable diseases (NCDs). NCDs account for about 15 million premature deaths (between ages 30 and 69) worldwide, killing people in their most productive years. As the leading cause of preventable deaths, tobacco use remains one of the foremost public health challenges of our time. The consequences of tobacco use also present enormous economic, development and social costs that wreak havoc on families, communities and societies. The annual economic cost of smoking was estimated at US$ 1.4 trillion in 2012, equivalent to 1.8% of the global gross domestic product (GDP) (9). With these figures likely to have increased since then, the massive health and economic burdens of tobacco use provide justification for governments to intervene and strictly regulate the market for tobacco products. The purview of tobacco control extends beyond the strong imperative to protect people’s health and well-being; it should also strive to contain the market failures and negative externalities of tobacco use, particularly since these effects can significantly impact a country’s development trajectory. The mounting evidence of the enduring destruction caused by tobacco in the 20th century provided compelling reasons for a strong global response, which led countries to negotiate the World Health Organization Framework Convention on Tobacco Control (WHO FCTC). The WHO FCTC came into force in 2005 as the CHAP T ER 1. WHY T HIS M ANUAL? 3 first public health treaty under the auspices of WHO. To facilitate its implementation at the country level, WHO packaged a set of demand-reduction measures directly taken from the treaty (7). These interventions, collectively known as MPOWER, are as follows: (M) monitoring tobacco use and prevention policies; (P) protecting people from tobacco smoke (smoke-free laws); (O) offering help to quit tobacco use (cessation services); (W) warning about the dangers of tobacco (including graphic pack warnings and plain packaging); (E) enforcing bans on tobacco advertising, promotion and sponsorship; and (R) raising taxes on tobacco products. Specifically, under Article 6 of the WHO FCTC, Parties recognized that price and tax measures are an effective and important means of reducing tobacco consumption for various segments of the population – in particular, among young persons (10). The severity of the tobacco epidemic and its ongoing damage to health and economies are clear justifications for governments to actively intervene and correct market failures. The scale of the burden and the rate at which lives are being destroyed necessitates urgent and aggressive action on tobacco control, using measures that most countries have committed to implementing and that are proven to be effective in reducing tobacco use. WHY TOBACCO (EXCISE) TAXES ARE CRUCIAL Among the different tobacco control interventions, raising excise taxes has been identified as the most effective as well as the most cost-effective measure to reduce consumption (6). While other interventions are certainly important components of a comprehensive tobacco control strategy, the direct impact of significant tax increases on consumption is by far the strongest. On average, a tax increase that causes prices to go up by 10% reduces consumption by 4% in high-income countries and 5% in LMICs (6). When implemented at scale, this demonstrates the enormous power of tobacco taxation and its potential to save lives. Tobacco taxes differ from other interventions in that their impact can increase and build over time – even if taxes are already relatively high, their rates need to be continuously increased to retain and amplify their effectiveness. However, this should not be taken as a suggestion that governments considering tobacco control interventions should focus solely on taxes. Taxes are even more effective when implemented as part of a comprehensive package of measures such as MPOWER, which covers distinct but complementary intervention points. Among the different taxes applied on tobacco products, excise taxes are the most significant because they raise both absolute and relative prices (6). This is important when considering health objectives, since it is the magnitude of the price increase of tobacco products that determines the reduction in consumption. An excise tax is typically applied on a limited set of products, designed to discourage their use by 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N raising the price significantly over that of other products available in the market. This is in contrast to value added taxes (VAT) or sales taxes, which apply to most goods and services. Seeking to raise the prices of tobacco products through VAT or sales taxes would fail to increase relative prices, making this an ineffective and inefficient method. Customs or import duties on tobacco products are also utilized, but their impact is waning with the global trend towards bilateral and regional agreements aimed at trade facilitation. The application of these duties varies across countries, but overall, they are not applicable to locally produced tobacco products. As a tool to increase prices, import duties cannot substitute for excise taxes, since they are not specifically designed to reduce consumption. When viewed as a public health policy tool, tobacco taxation is highly cost-effective, since it delivers significant impact yet is relatively inexpensive to implement (11). The costs of implementing tobacco taxation are much lower than those of clinical NCD interventions such as cancer treatments or maintenance medications, since the commodity and human capital requirements are less substantial (12). Moreover, increasing tobacco taxes actually generates additional revenue for a government. Tobacco taxes are also very effective in pre-empting or reducing consumption among groups of people who are especially price-sensitive – youth in particular, who are prevented from initiating a lifelong addiction if taxes and prices are sufficiently high (13). This is also true for the poor, who are more prone to catastrophic health expenditures than the wealthy are. Preventing initiation or encouraging cessation by imposing high taxes provides an escape route from the vicious cycle of tobacco use and poverty (6). A DECADE OF PROGRESS AND COMMITMENT TO ACTION Since 2010, when the first WHO technical manual on tobacco tax administration was published, numerous developments have raised the profile of tobacco taxes as an essential public health intervention. The Conference of the Parties (COP) to the WHO FCTC adopted guidelines for implementation of Article 6 of the treaty, which focuses on price and tax measures to reduce the demand for tobacco. Also within this period, three high-level meetings on the prevention and control of NCDs by the United Nations General Assembly, as well as the endorsement of the Global NCD Action Plan in 2013 by the World Health Assembly, have resulted in strong global commitments to implement measures, such as increased tobacco taxes to protect people’s health. The 2030 Agenda for Sustainable Development, which contains 17 goals known as the Sustainable Development Goals (SDGs), describes the global development strategy for the next decade. Within the SDGs, two specified targets are highly relevant for tobacco control: strengthening the implementation of the WHO FCTC (target 3.a) CHAP T ER 1. WHY T HIS M ANUAL? 5 and reducing premature mortality from NCDs by 30% (target 3.4). Furthermore, the Addis Ababa Action Agenda1, which aims to provide a global framework for financing the SDGs, also highlights tax and price measures on tobacco as key mechanisms to reduce demand and save lives while increasing domestic resources for develop- ment. Another important milestone was the 2018 entry into force of the Protocol to Eliminate Illicit Trade in Tobacco Products. These key events, along with several outcome documents and policy declarations in the area of tobacco control and the wider development sphere, have introduced tobacco taxation into the consciousness of a much larger share of policy-makers. As detailed in subsequent chapters, numerous countries have imposed sufficiently high tobacco tax rates while applying best practices in tax policy design and imple- mentation over the past decade (8, 10). For example, sustained and substantial tax increases have reduced tobacco use in LMICs such as Brazil (14), Turkey (15) and the Philippines (16). High-income countries also continued their leadership in this area, comprising 23 of the 38 countries judged to have sufficiently high tobacco taxes in 2018 (8). However, much remains to be done. The 2019 WHO report on the global tobacco epidemic (RGTE) shows that tobacco taxes are still the most underutilized tobacco control policy among the MPOWER measures (8), with only 14% of the world’s population being covered by sufficiently high tobacco taxes. Substantial progress has also been made in building the tools and evidence base for tobacco taxation. Volume 14 of the International Agency for Research on Cancer (IARC) handbooks of cancer prevention, Effectiveness of tax and price policies for tobacco control, published in 2011, is a key review of the literature published as of May 2010 on the effectiveness of tax and price policies in reducing tobacco use. The National Cancer Institute (NCI)-WHO Monograph on the economics of tobacco and tobacco control, published in 2016, details the evidence accumulated over the years from various countries, focusing not only on tax and price policies, but on all aspects of the economics of tobacco and tobacco control. In addition, numerous published studies from LMICs provide a comprehensive picture of the impact of tobacco taxation in different contexts. The updated Appendix 3 of the Global NCD Action Plan explains the cost-effectiveness of tobacco taxation (11), while the Global NCD Business Plan, Saving lives, spending less, built on this work by estimating a dollar figure for the return on investment expected from implementing the best-buy interventions for tobacco control, including taxation (12). The past decade has seen major steps forward for tobacco taxation in terms of global commitments, the number of countries implementing best practices and the 1 The Addis Ababa Action Agenda of the Third Conference on Financing for Development. Third Inter- national Conference, 13-16 July 2015, Addis Ababa, Ethiopia (https://sustainabledevelopment.un.org/ content/documents/2051AAAA_Outcome.pdf, accessed 17 February 2021). 6 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N expansion of the evidence base on tobacco taxes, particularly in LMICs. Within the development sphere, institutions such as the World Bank, the International Monetary Fund (IMF) and many other multilateral agencies are aligned with WHO on the importance of tobacco taxation and the need to improve its implementation (17–18). Considerable challenges remain; although it appears that the world is headed in the right direction, progress needs to accelerate at a much quicker pace in order to achieve the SDG targets by 2030. SHAPING A “NEW NORMAL” FOR TOBACCO TAXATION The global upheaval caused by the COVID-19 pandemic has cast an unprecedented spotlight on how well governments around the world prepared for and responded to the crisis. It has exposed glaring health systems vulnerabilities and highlighted the struggles of many countries to control the spread of the virus. But perhaps more than anything, the pandemic has demonstrated how the economy, trade, science, politics and many other aspects of our societies are very much interdependent and interconnected with the health of the population. It is clear that an individual’s state of health can significantly determine their susceptibility to disease and their ability to overcome it. People with NCDs are more vulnerable to becoming severely ill with a number of conditions, which also appears to be the case with COVID-19 (19). Tobacco use is a major risk factor for NCDs, and available research suggests that smokers are at higher risk of developing severe illness and dying from COVID-19 (20). Just as the different aspects of society are interconnected, so too are people’s health, the existence of health-promoting environments and the government policies and agencies that shape these environ- ments. This critical moment presents a unique opportunity and renewed motivation to discourage the use of harmful products such as tobacco and to further improve tobacco control measures, especially tobacco tax policy. Moving forward, a business-as-usual approach to tobacco taxation will not be sufficient. Responding to this new reality and preparing for the next pandemic entails implementing measures that promote healthier populations. Like the COVID-19 pandemic, any future pandemic will likely exacerbate health inequities, bring about more economic uncertainty and put pressure on governments’ fiscal capacities. Interventions such as higher tobacco taxes, which protect people’s health while generating more revenues and economic benefits, become even more important in such crisis situations. Given this context and the stakes involved, ministries of finance and tax authorities are in a unique and powerful position – one of saving not only livelihoods, but also lives. The importance of increasing tobacco taxes – one of the most effective public health tools available – cannot be overstated. The traditional approach of treating CHAP T ER 1. WHY T HIS M ANUAL? 7 tobacco tax exclusively as a revenue source has no place in the new normal. One cannot deny the scale of the tobacco epidemic, the necessity to correct market failures and the overwhelming evidence of tobacco taxation’s benefits to health and to the economy. The positive trend in the changing narrative around tobacco taxation needs to continue. Tobacco taxation should not be viewed in isolation from the rest of government policies, but rather as an important part of the whole, an essential piece in working towards our common goal of better health for all. OVERVIEW OF SUBSTANTIVE CHAPTERS This manual is primarily designed for policy-makers, finance officials, tax authorities and customs officials. It may also be useful for officials within health ministries or other government agencies, as well as nongovernmental organizations working in this area, including tobacco control advocates. Significant effort is made to present real-world examples and recent experiences from a wide range of countries to demonstrate success stories and lessons learned in raising tobacco taxes. A sub- stantial amount of evidence has been generated in LMICs over the past few years that supports and augments the existing evidence base, providing a much broader body of knowledge than was available when the first WHO technical manual on tobacco tax administration was released. Chapter 2 delves into the theory, practice and empirical evidence on tobacco excise tax policy, including current global trends. The chapter offers a detailed analysis of the various elements that constitute tax structure, aiming to provide policy-makers with a comprehensive understanding of the factors affecting prices, consumption and the market. It describes the key components to keep in mind when designing tobacco tax policy to maximize the impact of tax increases and improve the tax structure. The chapter also includes updated global price and tax data, specific examples from various countries and a discussion of tax base elasticity, automatic excise tax adjustments and pricing regulations, as well as descriptions of new and emerging nicotine and tobacco products, including electronic nicotine- and non- nicotine delivery systems (ENDS/ENNDS) and heated tobacco products (HTPs). Chapter 3 focuses on tobacco tax administration. It provides an in-depth discus- sion of the fundamental components that make tobacco tax collection effective and efficient, ensuring achievement of the health and revenue objectives of tax policy. It highlights the importance of cooperation among the various agencies involved in the implementation of tobacco taxes within countries and across borders. Building on country and regional experiences from previous decades, the chapter outlines specific measures and recommendations to maintain oversight of the whole tax compliance cycle. Also included are actions to facilitate control and enforcement, such as licensing, fiscal markings (e.g. tax stamps), tracking and tracing systems 8 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N and import and export controls. Finally, the discussion pivots towards the broader elements of a good tax system, such as proper resourcing of competent authorities, a strong judiciary and strict rules regarding corruption. Chapter 4 deals with the important challenges in the area of political economy when countries attempt to increase tobacco taxes or simplify the tax structure. The tobacco industry often relies on identified patterns of argumentation and tactics to obstruct such reforms, i.e. SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric (regressivity), (R) revenue reduc- tion and (E) employment impact – each of which is discussed thoroughly in this chapter. Also included is a detailed discussion of the measurement of illicit trade and a discussion of earmarking tobacco tax revenue for health purposes. Finally, Chapter 5 presents a comprehensive list of the best practices in tobacco tax policy and administration discussed throughout this manual. The list is intended to serve as a practical guide and quick reference to the salient points presented. CHAP T ER 1. WHY T HIS M ANUAL? 9 REFERENCES 1. Ranson K, Jha P, Chaloupka FJ, Nguyen SN. The effectiveness and cost-effectiveness of price and other tobacco control policies. In: Jha P, Chaloupka FJ, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000:427–447 (https://www.paho.org/hq/dmdocuments/2010/ Cost-effectiveness%20of%20price%20increases.pdf, accessed 4 February 2021). 2. Effectiveness of tax and price policies for tobacco control. Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications.iarc.fr/Book-And-Report-Series/Iarc- Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And-Price-Policies-For-Tobacco-Control-2011, accessed 4 February 2021. 3. Tobacco taxation in the United States. In: Lynch BS, Bonnie RJ, editors. Institute of Medicine (US) Committee on Preventing Nicotine Addiction in Children and Youths. Growing up tobacco free: preventing nicotine addiction in children and youths. Washington (DC): National Academies Press (US); 1994 (https://www.ncbi.nlm.nih.gov/books/NBK236771/, accessed 10 November 2020). 4. Sin tax reform. Manila: Department of Finance (Philippines); 2012 (https://www.dof.gov.ph/advocacies/ sin-tax-reform/, accessed 10 November 2020). 5. Goodchild M, Perucic AM, Nargis N. Modelling the impact of raising tobacco taxes on public health and finance. Bull World Health Organ. 2016; 94:250–7 (https://www.who.int/bulletin/ volumes/94/4/15-164707.pdf, accessed 4 February 2021). 6. The economics of tobacco and tobacco control. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute, NIH Publication No. 16-CA-8029A; 2016 (National Cancer Institute tobacco control monograph 21; https://cancercontrol.cancer.gov/ brp/tcrb/monographs/monograph-21, accessed 17 December 2020). 7. WHO report on the global tobacco epidemic, 2008: the MPOWER package. Geneva: World Health Organization; 2008 (https://www.who.int/tobacco/mpower/mpower_report_full_2008.pdf, accessed 10 November 2020). 8. WHO report on the global tobacco epidemic, 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/tobacco/global_report/en/, accessed 10 November 2020). 9. Goodchild M, Nargis N, Tursan d’Espaignet E. Global economic cost of smoking-attributable diseases. Tob Control. 2018;27:58–64 (https://tobaccocontrol.bmj.com/content/27/1/58, accessed 4 February 2021). 10. Guidelines for implementation of Article 6 of the WHO FCTC. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 4 February 2021). 11. Tackling NCDs: ‘best buys’ and other recommended interventions for the prevention and control of noncommunicable diseases. Geneva: World Health Organization; 2017 (https://apps.who.int/iris/ handle/10665/259232, accessed 10 November 2020). 12. Saving lives, spending less: a strategic response to noncommunicable diseases. Geneva: World Health Organization; 2018 (https://www.who.int/publications/i/item/WHO-NMH-NVI-18.8, accessed 4 February 2021). 13. Chaloupka FJ, Warner KE. The economics of smoking. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics. Elsevier; 2000;1(1):1539–1627. 14. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Rev Panam Salud Publica. 2016;40(4):243–9 (https://iris.paho.org/bitstream/ handle/10665.2/31306/v40n4a09_243-9.pdf?sequence=1&isAllowed=y, accessed 17 February 2021). 15. Cetinkaya V, Marquez PV. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 16. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington (DC): World Bank Group; 2016 (http://documents.worldbank.org/curated/en/638391468480878595/pdf/106777-PUB-PUBLIC- PUBDATE-7-26-2016.pdf, accessed 10 November 2020). 17. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 10 November 2020). 10 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 18. Irwin A, Marquez P, Jha P, Peto R, Moreno-Dodson B, Goodchild M, et al. Tobacco tax reform: at the crossroads of health and development – a multisectoral perspective. Washington (DC): World Bank Group; (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report- CrossRoads.pdf, accessed 2 February 2021). 19. Information note: COVID-19 and NCDs. Geneva: World Health Organization; 23 March 2020 (https:// www.who.int/publications/m/item/covid-19-and-ncds, accessed 10 November 2020). 20. WHO statement: tobacco use and COVID-19. Geneva: World Health Organization; 11 May 2020 (https:// www.who.int/news-room/detail/11-05-2020-who-statement-tobacco-use-and-covid-19, accessed 10 November 2020). 11 CHAPTER 2. Tobacco excise tax policy 2.1. GLOBAL OVERVIEW OF TOBACCO TAX PRACTICES A well-designed tax policy is key to having an effective tax policy. Any government that is planning to reform its tax policy must first understand the fundamental components of a good tax policy, as well as consider the strengths and weaknesses of different approaches to taxation, how they impact price and the requirements for tax administration. Understanding how the tobacco market operates in a country is equally important for policy-makers because of the inevitable interaction between the market and tax structures. Beyond the political considerations that strongly influence tobacco tax policy development, this chapter focuses on the technical aspects of tobacco taxation – excise tax in particular. Section 2.1 provides an overview of tobacco tax practices at the global level, focusing on the different ways countries structure excise tax. Section 2.2 emphasizes the importance of carefully designing excise tax policy, highlighting not only the significance of tax increases but also excise tax structure and its impact on prices, taking into account how market structure influences trends. This section also discusses the importance of measuring impact as another aspect of tax policy development, and it presents the crucial elements for performing measurement, as well as the relevant indicators available to monitor progress. Section 2.3 describes external policy considerations in the design phase to ensure that the goals of tobacco control and taxation are achieved. Intersectoral policy integration and coherence at the domestic level is discussed as a strategy to ensure that policies of other sectors do not inhibit or obstruct public health policy objectives. This section also reviews the current state of regional tax harmonization based on the experience of existing regional blocs and draws conclusions on the best policy approaches to preserve the public health interests of individual countries. Section 2.4 discusses new and emerging nicotine and tobacco products, in par- ticular HTPs and ENDS/ENNDS. It reviews the latest evidence on the health impacts of these products and current approaches to regulation. Key policy considerations are identified, and recommendations are provided for adopting an appropriate excise tax policy for these products. Section 2.5 summarizes the issues covered in the chapter and the key takeaways. 12 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.1.1 TAXES APPLIED ON TOBACCO PRODUCTS Taxes are classified as either direct or indirect. Direct taxes are imposed on the profit, income, property or wealth of persons or companies, whereas indirect taxes are imposed on the price of goods and services. Indirect taxes are most relevant to tobacco products taxation, because they directly influence price. A variety of types of indirect taxes can be applied to tobacco products. These include: • excise taxes – taxes that apply to a few selected commodities (they can also be applied to alcohol, fuel, sugar-sweetened beverages, etc.). • VAT or sales taxes – VAT is a multistage tax on all consumer goods and services that is applied proportionally to the price the consumer pays for a product. It is a tax on the amount by which the value of an article has been increased at each stage of its production or distribution. Some countries impose sales taxes instead of VAT. Unlike VAT, which is collected at every stage of the supply chain, sales taxes are generally levied at the point of retail on the total value of goods and services purchased. Ultimately, the consumer ends up paying the tax, whether it is a VAT or a sales tax. • import duties – taxes on selected goods imported into a country to be consumed in that country (i.e. goods that are not in transit to another country). In general, import duties are collected from the importer at the point of entry into the country. • other taxes – other indirect taxes, such as environmental taxes, that do not fall into any of the categories listed above. One of the most well-established and widely understood points in tax policy is that tobacco products should be subject to excise taxation. The focus of this chapter – and of this manual overall – is on excise taxes. They are the most important type of indirect taxes for tobacco control because they are applied directly to tobacco products and contribute the most to increasing the price of tobacco products relative to other goods and, subsequently, to reducing consumption. There are two basic types of excise taxes: • specific – levied as a monetary value per quantity of the product being taxed (e.g. 1 000 cigarettes, pack of 20 sticks, kilogram of tobacco); and • ad valorem – levied as a percentage of the value (e.g. retail price, or the producer/ex-factory price or the cost, insurance and freight [CIF] value1) of the product being taxed. These types of excise tax can be applied at a uniform or a differential (tiered) rate and on their own or in combination (i.e. a mixed system). 1 CIF is the value of an imported product as declared to customs upon entry into a territory. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 13 Given the widespread use of cigarettes – almost the only tobacco product used in some parts of the world – and the scarce availability of data for other tobacco products, this chapter focuses mainly on cigarettes. But there are a few examples and recommendations for other tobacco products, including those that are more prevalent in specific parts of the world (e.g. bidis or smokeless tobacco in South-East Asia and waterpipe tobacco in the Eastern Mediterranean region). 2.1.2 CIGARETTE TAXES AND RECENT TRENDS WORLDWIDE Tax and national income levels: the higher the income level, the higher the taxes and prices At the global level, cigarette price and tax levels correlate positively with a country’s income level: prices and taxes are higher in higher-income countries and lower as income level decreases. This trend has not changed over the years since 2008. Figure 2.1 presents the levels of price and tax by income groups for 2018, using the World Bank classification of income groups. Fig. 2.1 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by income group, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in purchasing power parity (PPP) adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Excise tax per pack Other taxes Retail price minus taxes Pr ic es a nd ta xa tio n pe r p ac k (P PP $ ) High-income Low-income 4.25 0.68 Middle-income 2.06 Total taxes: 5.30 (67.9% of pack price) Total taxes: 2.91 (58.3% of pack price) Total taxes: 1.18 (38.1% of pack price) PPP $ 7.80 PPP $ 4.99 PPP $ 3.09 Global 2.48 PPP $ 5.53 Total taxes: 3.36 (60.8% of pack price) 0 1 2 3 4 5 6 7 8 14 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tax and price levels: the higher the tax share, the higher the price Globally, cigarette prices correlate positively with tax percentage levels: as the total tax share (of which excise represents the largest part) as a percentage of retail price increases, the price of cigarettes generally also increases (see Fig. 2.2 below).2 This indicates that taxes do influence prices. Fig. 2.2 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by total tax levels, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Tax level and WHO regional classification: taxes and prices are highest in the European region, followed by South-East Asia, the Americas and the Western Pacific, with the lowest levels in the Eastern Mediterranean and African regions At the regional level (WHO regional classification), average levels of prices and taxes vary greatly. The highest level can be seen in the European region, which includes the European Union (EU) countries. The EU’s unified tax structure includes high levels of minimum taxes – which lead to high prices – and encourage member 2 This is a general trend and does not apply for every country; there are countries that have a large tax share but low prices for cigarettes. Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax ≥ 75% Total tax ≤ 25% 0 7 4 1 2 3 6 5 50% ≤ Total tax < 75% 25% ≤ Total tax < 50% Excise tax Other taxes Retail price minus taxes 4.44 2.24 1.57 PPP $ 5.07 PPP $ 5.33 PPP $ 7.07 PPP $ 2.60 0.17 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 15 countries to regularly increase their taxes to meet their obligations. However, with the current minimum level now being reached by all EU member countries, the motivation to increase excise taxes may wane. Indeed, the minimum cigarette excise amount in the EU has not been adjusted since it went into effect on 1 January 2014, and it is suffering from inflation erosion. Member States of the EU acknowledged this in June 2020 by stating that action at the EU level is required to ensure that minimum excise duty rates regain traction to effectively reduce the consumption of tobacco products and that the minimum rates of excise duties on a number of tobacco products would be increased (2). Excise taxes are lowest in the African and Eastern Mediterranean regions. And China – reported separately due to its size – has lower tax rates than the Western Pacific region (see Fig. 2.3). Fig. 2.3 Weighted average retail prices and taxation (excise and total) of most-sold brand of cigarettes, by region, 2018 Notes: China is represented separately from the Western Pacific Regional Office (WPRO) average because of its exceptionally large number of smokers compared with the number in other countries in the region. AFRO is the African Region, AMRO is the Region of the Americas, EMRO is the Eastern Mediterranean Region, EURO is the European Region, SEARO is the South-East Asia Region. Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Retail price Excise tax Total tax share % AFRO AMRO EMRO EURO SEARO WPRO w/o China China All 3. 80 4. 02 4. 24 4. 89 4. 02 1. 45 2. 89 2. 65 2. 48 7. 53 7. 27 5. 61 2. 51 1. 95 1. 05 42.2% 55.8% 61.2% 72.9% 63.2% 55.7% 60.8% 56.9% Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax share % 5. 53 0 1 2 3 4 5 6 7 8 9 10 40 30 50 60 70 80 20 16 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Global tax structures trend: more countries are adopting specific excise taxes or mixed excise systems that rely more on the specific component Since 2008, the number of countries that rely solely on ad valorem taxes or apply no excise taxes at all has decreased as more countries have adopted specific or mixed systems. More of the countries that have implemented a mixed system have increased the specific component of the tax structure relative to the ad valorem component (see Figs. 2.4 and 2.5).3 Fig. 2.4 Changes in excise tax structure, 2008–2018 Fig. 2.5 Changes in reliance on specific versus ad valorem component in mixed systems, 2008–2018 Source: (1). 3 For information about countries that applied each type of excise tax structure in 2018, see Annex 2.1. N um be r o f c ou nt ri es Specic excise Ad valorem Mixed excise No excise 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 54 56 57 63 50 45 24 23 24 21 19 15 49 47 44 42 41 55 57 59 56 60 63 62 Mixed excise Relying more on specic Relying more on ad valorem N um be r o f c ou nt ri es 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 45 22 23 23 24 22 26 27 27 27 32 35 37 50 54 56 57 63 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 17 The imposition of a minimum specific excise tax: half of the countries that have a mixed or ad valorem structure impose an excise tax floor A minimum specific excise tax ensures that at least a certain minimum amount of tax is paid, irrespective of price level. Almost half of the 101 countries that impose either ad valorem or a mixed excise for which data on minimum excise are available (47 countries) set a minimum specific excise tax.4 Nearly two thirds of those that set a minimum specific excise tax (29 countries) are high-income countries; most of them are in the EU, which requires its members to impose a minimum specific excise tax. The choice of tax base worldwide: almost half of the countries that apply a mixed or ad valorem excise system use retail price as the base Setting the base for applying a specific excise is relatively easy: most countries use a defined quantity of sticks for cigarettes, the weight in kilograms for tobacco and the weight in grams for other tobacco products (1). Different bases for ad valorem excises are applied in different countries. Nearly half of the 105 countries that implement either ad valorem or a mixed excise for which data are available (47 countries) use the retail price5 as the tax base for the ad valorem part, and most of those (28 countries) are high-income countries. Using the retail price as the base for the excise ad valorem tax is more effective than using the producer price or the CIF value. Unlike retail prices, which are easy for tax administrators to ascertain by monitoring the market, the producer price or CIF value is prone to undervaluation by producers or importers, who may pass on their margins to related parties further down the supply chain and successfully reduce their tax burden. This tactic is also known as transfer pricing. Additionally, global-level data show that the excise ad valorem on the retail price seems to lead to higher retail prices on average compared with an excise ad valorem applied on other bases, such as the producer price or CIF value (see Fig. 2.8 below). On complex tiered structures: 31 countries still apply complex, multitiered excise taxes on tobacco products As of 2018, 31 countries imposed excise taxes that varied according to defined char- acteristics of cigarettes, including price level, type of production, type of package and length of cigarette (Table 2.1). Some countries use more than one criterion to differen- tiate the tax rates. Indonesia, for example, imposes differential rates based on volume 4 This means that countries with a mixed system impose an overall minimum specific excise tax (where the yield of the specific plus the ad valorem excise cannot be below the set minimum specific excise tax), in addition to the excise on a specific component. 5 Countries that impose ad valorem on retail price exclusive of VAT are also included, since retail prices are easy to determine and VAT rates are known variables. 18 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N produced, type of cigarette and price level. In other countries, such as Member States of the EU, differential rates for cigarettes are prohibited by law, and the rate of the ad va- lorem tax and the amount of specific excise duty must be the same for all cigarettes (3). Table 2.1 Criteria used by countries for tiered excise taxes, 2018 BASE OF TIERS COUNTRY Retail price Bangladesh, Belarus, Indonesia, Jordan, Mozambique, Myanmar, Pakistan, Thailand Cigarette grade (e.g. premium, mid-grade, economy) Egypt, Japana, Mali Producer price China, Lao People’s Democratic Republic Production volume Indonesia Type filter/non-filter Belarus, Georgiaa, India, Kenya, Republic of Moldova, Nepal, Papua New Guinea hand/machine made India, Indonesia kretek/white cigarette Indonesia tobacco content (dark/ blonde or dark/light) Algeria, Bolivia (Plurinational State of ) Packaging soft/hard Mozambique, Uganda Cigarette length India, Nepal, Sri Lanka Trade (domestic/imported) Iran (Islamic Republic of ), Lebanon, Myanmar, Solomon Islands, Tonga, Uzbekistan Leaf content (domestic/imported) Fiji, United Republic of Tanzania a Japan and Georgia were using a tiered excise tax structure when these data were collected in 2018, but as of 2020, that is no longer the case. Source: (1). 2.2 DESIGNING EXCISE TAX POLICY Significantly increasing the taxes on and prices of tobacco products is the most effective and most cost-effective policy to control tobacco use (4). Increased taxes that are passed on to tobacco users as higher prices reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face challenges ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues such as addressing the SCARE6 tactics of the tobacco industry. This section provides guidance for policy-makers regarding the best tax structure to use from a health perspective, taking into consideration all the appropriate tax designs. It also proposes recommended indicators to consider when formulating policy change. 6 SCARE tactics are the tactics most commonly used by the tobacco industry when countries plan to increase tobacco taxes. They are described, and refuted, in detail in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 19 2.2.1 THE IMPORTANCE OF THE TYPE OF EXCISE TAX STRUCTURE The existing theoretical and empirical evidence on approaches to the choice of (uniform) specific and ad valorem excises is reviewed below, along with their effects on price, consumption, perceived quality and variety of tobacco products, govern- ment revenue and tax administration. The use of the word “quality” in this chapter does not refer in any way to the health impact of a tobacco product. It refers rather to the consumers’ perceptions of quality and their decision to buy a product, which they may evaluate based on the packaging, the blend used for the cigarette or anything that makes the product more appealing to them. Just to be clear, from a public health perspective, all cigarettes are equally harmful even if perceived by consumers as having higher or lower quality. The choice between ad valorem and specific taxation is influenced by the market structure, i.e. the nature and degree of competition in the market for goods and services. Although each country has its own specific characteristics, the tobacco market structure is typically a monopoly or an oligopoly where firms have the power to control prices – and hence exploit the tax structure – to their benefit. For example, China, the largest producer and consumer of tobacco products in the world (5), has a state monopoly. In Viet Nam, foreign brands are produced under licence by the state monopoly. In Thailand and Egypt, despite the presence of foreign companies, the market is dominated by the state-owned company. In Uruguay, the oligopoly is led by a domestically owned company. In Bangladesh, the oligopoly consists of domestically owned companies competing with foreign companies (6). In most of Africa, the market consists of transnational tobacco companies (7). The impact of tax structure on final price: uniform specific versus uniform ad valorem The choice between specific and ad valorem taxes is a long-standing issue in tax policy, as the level and structure of excises have different implications for the interests and goals of various groups. Given the market structure of the tobacco industry – typically a monopoly or oligopoly for most products in most countries – different excises may have different effects on government revenue, manufacturer profit, consumer price, perceived product quality and variety and tax administration (8–16). Consequently, the two types of excise taxes – specific and ad valorem – may have different implications for public health to the extent that they affect individual consumption via their impact on perceived product quality, variety and prices. Moreover, governments have the potential to influence tobacco excises to manage demand, raise revenue and promote public health. 20 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax on prices and in turn on consumption is also influenced by price and income elasticities, as well as consumer perceptions of quality (“perceived quality”) and the variety of available products, which, in turn, are closely linked to the type of tax structure adopted. Tax structure is affected by both the price elasticity of demand and the price elasticity of supply. The price elasticity of demand measures the responsiveness of consumer demand to changes in prices. The price elasticity of supply measures how sensitive producers are to changes in prices. Tobacco tax structure is also influenced by market structure. In a monopoly, the profit-maximizing firm sets the price, considering the price elasticity of demand: the lower the price elasticity (in absolute value) – i.e. the less sensitive the consumer is to price changes – the higher the price the monopolist can set. Profits are typically abnormal in a monopolist market structure, meaning total sales revenue is higher than total cost (where total cost includes a normal profit). A monopolist producer therefore receives more than the minimum reward required to invest its (physical and human) capital and undertake business risks. Economic theory predicts that in a private monopoly, prices are higher than in an oligopolistic market. This is not, however, necessarily true when the monopoly is owned by the state and the government’s objective is not straightforward profit maximization: the government might have other considerations, such as preserving jobs (e.g. in China) or keeping prices low for low-income consumers (e.g. in Egypt). Under a monopoly, an ad valorem taxation structure enables the monopolist producer to set prices lower than would be possible under a specific tax structure. This is feasible because under ad valorem taxation, when supply increases and price falls, the price reduction is not fully borne by the producer. Rather, the price reduc- tion is partly shared by the government since, as supply increases, the tax per unit of product sold falls. In other words, ad valorem taxation leads to lower prices and higher consumption relative to revenue-equivalent specific taxation. Technically, this means that the supply function is less elastic under ad valorem taxation. In contrast, under a specific taxation structure, any increase in the monopolist producer price will go to the producers as revenue, which incentivizes them to increase prices. The same logic also applies to an oligopolistic market structure, where profits again are, in general, abnormal. KEY TAKEAWAY 1 In a monopoly or an oligopoly, specific taxation incentivizes industry to set prices higher than it would with ad valorem taxation. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 21 Understanding oligopolies, however, is more complicated, since they are characterized by strategic interdependence among a few firms. This strategic interdependence extends to the relationship between industry and regulators. Industry anticipates the government’s regulatory policy – whether through tax or other interventions – and acts accordingly. For example, competitors may coordinate and lobby against a certain tax structure reform or tax rate increase. Under an oligopoly market structure, ad valorem taxation is a relatively more efficient tool for transferring part of the profits to the government as tax revenue, since it acts like both an excise and a profit tax. In contrast, a specific tax has a smaller (negative) effect on profits. This explains why we observe multinationals that are leaders in high-priced brands (e.g. Philip Morris International [PMI]) lobbying in favour of specific taxation (17). As an example, in the countries of the Cooperation Council for the Arab States of the Gulf (GCC), the tobacco industry has been trying for a long time to lobby governments to introduce a specific excise (18–21). After years of consideration and discussions on the possible introduction of excise taxes, the GCC adopted the Common Excise Tax Agreement of the States of the Gulf Cooperation Council in November 2016 (21), which introduced an ad valorem excise on tobacco products. Tobacco companies’ support for excise tax structures ultimately depends on the market segments they control in a particular country. A company selling mainly premium brands will favour specific excises, whereas a company that sells mid-priced or economy brands would favour ad valorem excise (17). When oligopolistic firms produce identical products, a specific tax has a stronger positive effect on price and is more likely to be overshifted to consumer prices than an ad valorem tax (13). Overshifting means that the price increases by more than the tax increase itself. Empirical evidence supports this (22–26). KEY TAKEAWAY 2 In an oligopoly, prices are likely to increase by more than the amount of the specific tax increase when demand is relatively inelastic. In general, demand for a product depends not only on prices but also on consumer perceptions of quality and preferences for variety. For example, the most popular brand in GCC countries is Marlboro, a premium brand (1). Consumers differ in their willingness to pay, depending on their respective perceptions of quality, which influence whether they ultimately purchase high- or low-priced brands. A tax-induced price increase can cause the following plausible responses from consumers or users of tobacco products: (1) a group of consum- ers will quit; (2) a group of consumers will reduce their overall consumption; 22 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N (3) another group, most likely high-income users, will switch to an upgraded version of the tobacco product if its relative price (compared to the cheaper brand) has been reduced, which is the case under a specific tax increase; there might also be a group of consumers in the lower income range who switch to lower-priced variants of the tobacco product if the price gap increases, as is the case under an ad valorem tax increase; and (4) another group might switch to the illegal market or buy products in a neighbouring country with a lower tax where possible. Consumers’ decisions to purchase are also affected by their preferences for variety – meaning preferences among products that consumers perceive as equal in quality but are given different characteristics by the producers to account for consumer taste preferences. Thus, it is possible that a tax increase that leads to an increase in average prices will lead to an increase in the total quantity demanded in the market because of an increase in the variety of product choices available to consumers. Variety enables new consumers to be captured, especially in an environment lacking certain regulations (e.g. without plain packaging and flavour bans). The tobacco industry was able to capture a new group of consumers when it introduced menthol cigarettes into the tobacco market. There is more than sufficient evidence that menthol cigarettes increased youth smoking initiation, increased nicotine dependence and reduced adult smoking cessation (27). To prevent this from happening in their countries, Member States of the EU have prohibited characterizing flavours other than tobacco in tobacco products (28). It is therefore important to consider the broader effects that the structure and level of an excise tax can have on average price, perceived quality and the variety of cigarette brands and other emerging substitutes. When consumers make choices based on dimensions other than quantity, the two types of tax structures are not equivalent, even in a perfectly competitive market where firms have no market power (29–30). To illustrate this point, consider a US$ 1 cost to improve consumer perceptions of quality for a tobacco product. This will lead to an equivalent price increase under specific taxation but not under ad valorem taxation. At an ad valorem rate of 20%, the price must increase by more than US$ 1, or by 1/(1 – 0.2) to cover the US$ 1 cost of improvement, due to the multiplier effect. A specific tax induces consumers to reduce the quantity demanded, but they might still choose to pay a higher price in exchange for a product that they perceive to be of better quality. An ad valorem tax, on the other hand, leads to a reduction in both quantity and perceived quality, not a substitution between them. An ad valorem tax has only an income effect and – unlike specific taxation – does not lead to substitution between perceived quality and quantity. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 23 KEY TAKEAWAY 3 Under specific taxation, the industry has incentives to create upgraded variants of tobacco products that attract new consumers and encourage consumption. When firms produce differentiated products, as the tobacco industry does, economic theory provides ambiguous results regarding specific versus ad valorem taxation. The relative effects of the two types of tax are not as straightforward as in the case of oligopolistic firms producing a homogeneous product. With differentiated products, the relative effects of the tax types depend on various assumptions: whether or not firms face symmetric costs, whether the number of firms is fixed or new firms can enter the market and the level of the tax revenue requirement. When firms face different costs, ad valorem taxes exacerbate the absolute differences in marginal costs between them. The high-cost brand is not considered a perfect substitute for the low-cost brand. A sufficiently high ad valorem tax rate may lead to a relative underproduction of the high-cost products (31). The effect of specific and ad va- lorem taxes on consumer perceptions of quality depends on market structure and the price and income elasticities of demand across various qualities. The relative price of the cheapest product does not necessarily remain unchanged or increase; it might fall (32). Empirical evidence showing that increases in the specific tax lead to a lower market share for the cheaper generic brands and an upward shift to premium brands (33–34) usually considers gradual tax increases and ignores income effects. Chaloupka et al. (35) found that in 21 EU countries that impose a mixed tax system, the price gap between premium and low-priced brands – while not reflecting the full distribution of cigarette prices – is smaller when the specific component of the mixed structure dominates.7 Although the price gaps are narrower under specific taxation, there is evidence that firms sometimes respond by introducing new, very cheap (subvalue) brands, or they exercise differential tax shifting. This practice has been evident in India for quite some time, with the Indian Tobacco Company launching a number of cheaper variants of its flagship cigarette brand, Gold Flake, to take advantage of a lower excise tax rate in the so-called microcigarette (< 60 mm length) market (36). Consequently, the cheapest end of India’s cigarette market has expanded significantly in recent years due in part to the marketing of new brand variants like Gold Flake Century. 7 The EU countries impose a mixed tax structure with a minimum tax floor. Some countries rely on the specific component more than others, but they remain within a given range (the specific component must be between 5% and 76.5% of total tax share of the weighted average price). 24 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As another example, transnational tobacco companies, which have sold ultra- low-priced brands in the United Kingdom since 2006, have managed to double their market share in a few years: their real price did not increase, since they absorbed part of the tax increases (37). The share of ultra-low-priced brands increased between 2001 and 2009 from 5% to 10%, while the market share of economy brands increased from 40% to 50% and the market share of premium brands and mid-priced brands decreased during the same period (from 35% to less than 25% for the former and from 15% to 5% for the latter). In order to keep the price of discount brands low and certain consumers in the market, firms may overshift the tax for premium products and undershift it for the lower-priced products (37–40). KEY TAKEAWAY 4 Evidence suggests that the price gap between brands is narrower under a specific tax structure. As the tobacco industry simultaneously consolidates producers and widens its portfolio of products, evidence is emerging that it is introducing cheaper brands while increasing the price of its expensive brands, therefore paradoxically widening the price gap within its products. The extent of the impact is still unclear, however, and this evidence does not negate the overall conclusion that a specific tax structure reduces price gaps. The impact of tax structure on final price: uniform specific, ad valorem and mixed systems Evidence from the 2019 RGTE (1) data suggests that the average price of the most- sold brand of cigarettes – weighted by the number of smokers – is the highest in countries implementing a mixed system that relies more on specific excise, followed by countries applying specific excise taxes only, followed by countries applying a mixed system that relies more on ad valorem and then by countries that apply ad valorem excise only (Fig. 2.6). The price is lowest in countries that have no excise at all. In past WHO reports on the global tobacco epidemic, countries that applied specific excise only had the highest price, on average. The trend may have changed partly because more countries are adopting mixed excise systems. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 25 Fig. 2.6 Weighted average price and excise for a pack of the most-sold brand of cigarettes, by excise tax structure, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Shang et al. (41) describe and compare price distributions, using data from 16 countries of the International Tobacco Control (ITC) Project that impose different cigarette tax structures. Specific uniform taxation tends to result in less variability in prices than all other structures (ad valorem tax, mixed tax, tiered tax). In general, structures other than uniform specific tax give rise to more opportunities for brand switching and tax avoidance. Reliance on complicated systems is likely to be as- sociated with wider price distribution, leading to greater tax avoidance, as there are more opportunities for substitution with cheaper brands when taxes rise. KEY TAKEAWAY 5 Evidence suggests that the tax structures most likely to lead to higher prices are uniform specific excise tax structures or mixed systems that rely more on specific excises. Mixed system relying more on specic excise Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Mixed system relying more on ad valorem excise Ad valorem excise No excise 2.21 Specic excise 2.62 3.26 0 7 6 5 4 3 2 1 1.66 Excise tax Other taxes Retail price minus taxes PPP $ 7.56 PPP $ 5.66 PPP $ 4.84 PPP $ 3.73 PPP $ 2.52 26 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax structure on final price: uniform excise versus tiered tax systems Another aspect of tax structure that impacts final price is the use of tiered taxation, i.e. tax rates that vary according to product characteristics. The characteristics can vary, from price level to the type of tobacco leaf contained in the cigarette, the size of production volume, the packaging, etc. Table 2.1 (earlier) lists the criteria used by 31 countries as the basis for different tax rates. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax (see Fig. 2.7). Fig. 2.7 Weighted average price and excise for a pack of the most-sold brand of cigarettes for countries with and without tiered taxation, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Uniform specific tax structures are likely to lead to relatively higher prices with less variability in price distribution. Compared with tiered tax structures that have differential rates based on brand characteristics, uniform taxation may reduce consumers’ incentive to switch to cheaper brands (leading to higher quit rates and lower prevalence), as well as decreasing manufacturers’ incentive to reduce their tax liabilities by changing their pricing strategies, production process or size (42–45). In Indonesia, for example, where small producers were taxed more fa- vourably, manufacturers had an incentive to reduce their scale of production but increase the number of affiliated small companies. The issue was resolved when tax authorities considered the aggregate production of all affiliated companies in the application of differential tax rates. By 2017, there were 786 active factories, while Excise uniform Price and taxation per pack of 20 sticks (PPP $) Excise tiers 3.28 1.94 Excise tax Other taxes Retail price minus taxes PPP $ 4.99 PPP $ 6.30 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 27 there had been 4 198 factories in 2006.8 Abolishing the differential tax rates would have been even more effective in removing the incentive for tax avoidance, as well as helping improve both public health and government finances. KEY TAKEAWAY 6 Evidence suggests that applying a uniform excise tax on cigarettes is not only easier to administer than tiered systems but also more likely to lead to higher cigarette prices. The impact of tax structure on final price: the significance of the choice of the tax base It is important for the excise tax to be applied to the base that leads to the greatest possible effect on price and revenue. For specific taxation, the tax base is the quantity of tobacco products. The quantity of cigarettes, cigars and bidis is measured in number of sticks; for other tobacco products, such as smokeless tobacco or roll-your-own (RYO), it is measured in the weight of the tobacco. When the tax is ad valorem, the choice of the tax base is important not only for health considerations – due to its effect on consumption – but also for tax revenue generation and industry profits. An ad valorem tax that is based on the ex-factory price (or CIF value) provides tobacco manufacturers with opportunities to reduce their tax liability, especially when they control the distribution system. Tobacco producers may sell cigarettes to distributors who are related parties at a reduced price, which then serves as the basis for calculating their ad valorem tax liability. Distributors, however, can then set high prices and share the extra profit with the producers (46). Because of the potential for such trade mispricing, the best practice is to use the retail price as the tax base and introduce a minimum excise tax per pack. Data in the 2019 WHO RGTE (1) show that, on average, the price level of a pack of cigarettes and the excise level are both much higher in countries that use retail price as the base for their ad valorem excise (Fig. 2.8). The maximum retail sales price, which includes all taxes, is used as the ad valorem tax base in the EU. That price also forms the tax base for ad valorem taxes in a growing number of LMICs, including Brazil, Egypt, Thailand, Turkey and Rwanda. 8 Indonesian Ministry of Finance, personal communication, 2017. 28 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.8 Weighted average price of the most-sold brand of cigarettes in countries that use retail price as the base for their ad valorem excise, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 7 The base on which the excise is applied is important. For specific excise, the base needs to be clearly defined (for cigarettes, cigars and bidis, it is the number of sticks; for other tobacco products, such as smokeless tobacco or RYO, it is the weight of tobacco). For ad valorem excise – where the base is typically either retail price, CIF value or producer price – evidence suggests that countries that apply the excise tax on the retail price of cigarettes tend to have higher prices than those that apply the tax on other bases. CIF and producer prices are difficult for government authorities to ascertain and are prone to undervaluation. The tax impact on final price: the significance of the minimum excise tax The use of a minimum excise tax in countries with ad valorem or mixed systems is another important factor in determining final price. On average, the price of a pack of cigarettes – as well as the excise level – is much higher in countries that impose a minimum specific excise than in those that do not (see Fig. 2.9). While more than half of the 47 countries that apply a minimum excise are members of the EU, removing EU countries from the average calculations produces the same conclusions. Ad valorem/mixed with retail price as base 3.72 PPP $ 5.01 PPP $ 6.41 Ad valorem/mixed with other base 1.79 Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 29 Fig. 2.9 Weighted average price of the most-sold brand of cigarettes in countries with and without a minimum specific excise tax, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 8 Among countries that apply an ad valorem or mixed excise tax on cigarettes, evidence suggests that those that impose a minimum specific excise tax tend to have higher prices than those that do not. The minimum excise tax also helps guarantee minimum excise revenues. Summarizing the advantages, disadvantages and impacts of the choice of excise tax structure for tobacco products Table 2.2 summarizes the characteristics of different types of tobacco excise taxes and the advantages and disadvantages of each type in relation to its impact on quantity demanded, perceived quality of brands offered, price, certainty and stability of revenue, administration and enforcement and opportunities for tax avoidance and tax evasion as they are predicted by the economic theory of imperfect competition and observed in real life. Ad valorem/mixed with minimum speci c 1.80 4.23 Ad valorem/mixed without minimum speci c Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes PPP $ 4.82 PPP $ 7.41 30 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 2.2 Characteristics of different types of tobacco excise taxes Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) TA X B A SE The unit of product (e.g. 1 000 cigarettes) The value of the product (e.g. retail, wholesale or manufacturer price) Excise is calculated on an ad valorem basis; however, if the calculated tax falls below a specified minimum amount, a specific tax rate applies Unit and value of product Both unit and value, unless the calculated tax falls below a specified minimum, in which case the tax base is the unit A D M IN IS TR AT IV E R EQ U IR EM EN TS The tax should be collected at the point of manufacturing or at the time of importation Low, as only the volume of the products needs to be ascertained Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as with a pure ad valorem regime Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises, as well as minimum specific excise tax compliance U N D ER VA LU AT IO N Not an issue Susceptible to undervaluation Provides an easy tool to prevent undervaluation of low-priced brands subject to the minimum specific excise The ad valorem part of the excise collection may be susceptible to undervalua- tion, depending on the choice of tax base The minimum specific excise prevents possible ad valorem tax base undervaluation of low-priced brands CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 31 Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IM PA C T O N P ER C EI V ED PR O D U C T Q U A LI TY Upgrading effect tends to reduce the relative tax on higher-priced brands Multiplier effect provides a disincentive to costly so-called quality improvement No incentive to upgrade higher- priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands IM PA C T O N P R IC E Tends to lead to relatively higher prices, particularly for low-priced cigarettes Tends to lead to relatively lower prices; price reductions will be subsidized if the multiplier effect is strong Tends to lead to relatively higher price increases for low-priced cigarettes An increase in the specific tax will to lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem payment if the base of the ad valorem includes excise An increase in the specific tax will lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem tax amount if the base of the ad valorem includes excise. Increases in the ad valorem and /or specific tax will raise the minimum tax paid if the minimum is a percentage of the total tax on, for example, weighted average price; they will reduce price gaps, given impact on perceived quality 32 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IN FL AT IO N The real value of the excise tax will be eroded unless the tax is adjusted in line with inflation The real value of the excise tax will be preserved as prices increase, at least to the extent that tobacco product prices follow inflation The real value of the minimum specific excise will be eroded over time unless the excise is adjusted in line with inflation The real value of the specific excise will be eroded unless the excise is adjusted in line with inflation The real value of the specific excise and the minimum specific excise will be eroded unless the excises are adjusted in line with inflation H EA LT H B EN EF IT S Will discourage consumption of tobacco products irrespective of the price band May encourage more trading down in favour of cheaper cigarettes, reducing the health benefit The minimum specific excise reduces incentives for trading down May reduce trading down Reduces trading down Source: (47). 2.2.2 OTHER TAX DESIGN CONSIDERATIONS The significance of automatic adjustments and indexation of specific tax to inflation Specific taxation does not depend on price and therefore, unlike the ad valorem tax, is not automatically adjusted for inflation. The real value of a specific tax is eroded over time as the price of the taxed product increases. Therefore, especially in countries with rapid growth in inflation, the nominal value of the specific tax must be increased regularly in order for the tax to maintain its real value. This is of great importance for both public health and public revenues, especially in countries where manufacturers do not increase prices regularly and/or low-priced tobacco products are the dominant products in the market. Table 2.3 lists countries that include automatic adjustments to their excise in order to avoid the erosion of the specific excise over time, using different units of adjustment and based on different frequencies. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 33 Table 2.3 Countries that include automatic adjustments to the specific excise COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Argentina Inflation (consumer price index [ CPI]), on a quarterly basis Armenia Minimum specific excise set to increase in the Tax Code by 15% on average between 2019 and 2021 Australia Wages – excise rates on tobacco and tobacco products increase in March and September each year, based on average weekly ordinary time earnings Bosnia and Herzegovina Specific excise rate is increased annually by at least 7.50 convertible marks per 1 000 cigarettes; minimum excise tax is increased annually to be at least 60% of the weighted average price Canada Inflation – federal tobacco tax rates are to be increased every five years, indexed to Canada’s CPI starting in 2019 Chile Inflation Colombia Specific tax set to 1 400 pesos, increased to 2100 pesos in 2018; starting in 2019, it will increase yearly by the CPI plus 4 points Costa Rica Inflation Dominican Republic Inflation, on a quarterly basis France Increase from 2017 to reach an average price for cigarettes of €10 per pack by 2020 Honduras Inflation, annually to December of the previous year Italy Minimum tax burden calculated every year in March on the basis of the weighted average price of cigarettes sold in the previous year New Zealand Inflation annually plus 10% annually from 2017 to 2020 Nicaragua Updated annually as of 1 January 2017, taking the highest among the annual devaluation of the official exchange rate of Cordoba with respect to the US dollar, published by the Central Bank of Nicaragua, and the annual inflation rate of the CPI published by the National Development Information Institute, observed in the last 12 months available North Macedonia Specific and minimum specific rate increase by 0.2 denars per cigarette on 1 July each year until 2023 Philippines Agreed tax increases and rates for specific excise tax between 2020 and 2023, with a 5% indexation thereafter Romania Inflation, annual (1 January) adjustment of the total excise according to inflation calculated on 1 October of the previous year Serbia Inflation, every six months Southern African Customs Union (SACU) – Botswana, Eswatini, Lesotho, Namibia and South Africa Inflation, on an annual basis9 Sweden Inflation 9 While the adjustment is not strictly automatic in the SACU, it is greatly informed by the inflation rate. The Treasury has some discretion. In recent years, the increases have typically been slightly above inflation. 34 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Turkey Producer Price Index Ukraine Agreement to increase the specific component 20% annually between 2019 and 2025 United Kingdom Increase by 2% above the retail price index (measure of inflation) for the 2015–2020 Parliament Source: (1). KEY TAKEAWAY 9 To avoid erosion of their specific excise tax, countries need to regularly – and, ideally, automatically – adjust the excise to inflation. The significance of automatic adjustments and indexation of specific tax to income growth In addition to the risk of erosion due to inflation, the effect of a (specific) tax can be significantly reduced if the tax is not adjusted for increases in consumer income. Income growth makes products more affordable – thereby encouraging consump- tion – especially in countries with rapid income growth. Australia is one of the rare countries that explicitly adjusts its specific excise rates according to wage growth (see Table 2.3). However, a number of countries have adopted automatic adjustments that are higher than inflation and sometimes largely cover income growth as well (see also Table 2.3). Adjusting tax for income growth contributes to increases in prices that make tobacco products less affordable (see section 2.2.3). KEY TAKEAWAY 10 The specific excise tax needs to be adjusted to reflect income growth so that tobacco products do not become more affordable over time. Measures for specific contexts: the role of pricing and other non-tax regulation Emerging evidence indicates that the tobacco industry finds ways to mitigate the impact of higher taxes on prices. For example, despite the heavy reliance on specific taxation in the United Kingdom, a price differential between premium and cheap cigarettes still exists. There is evidence that the tobacco industry does not always pass tax increases on to cheaper products (37, 48). Differential shifting among price categories is also observed in the EU (49–50), New Zealand (38) and the United States (51). Therefore, the public health community has suggested that pricing regulation could be considered as a method of eliminating inexpensive tobacco products that are often used by the young and the poor (52). Three types of pricing regulation are described below: minimum mark-up, price floor and price ceiling. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 35 Pricing regulation • Minimum mark-up – It has been argued that a minimum mark-up of wholesale tobacco prices could be a better strategy to raise prices of tobacco products than excise tax increases. Minimum mark-up laws aim to discourage the sale of products below an assessed cost by imposing a mark-up to the cost declared at different levels of the supply chain. Some studies in the United States have shown that minimum mark-ups do not increase average cigarette prices (53–54). However, a recent study of the impact of minimum mark-up/ price laws has shown that these laws are linked with higher prices, especially for the cheapest brands, and could be used as an effective tool to mitigate the impact of the industry’s price-reducing promotions (55). Another concern related to minimum mark-ups is that they can be manipulated by manufac- turers and are likely to lead to higher profits for the industry, as well as extra administrative costs for the government (56). • Price floor – A few studies suggest that setting a price floor, or a minimum price, is an alternative strategy for increasing tobacco taxes, particularly with respect to reducing health inequities (57–60). A price floor, imposed by the government or as a vertical restraint imposed by the supplier upon retailers, is a price that firms cannot legally undercut. Governments impose price floors to restrain unfair competition or, in the case of services, to increase quality. It is difficult, however, to find the right floor or to anticipate unintended conse- quences or an industry’s adjustments. A study in Malaysia, where a minimum price for cigarettes was imposed in 2010, found that the policy did not seem to have a meaningful impact on prices: licit brand prices remained well above the minimum price, while illicit brands remained well below it. This outcome may be a result of the floor being set too low or the proportion of illicit trade being high, either of which would reduce the effectiveness of the policy (52). In the EU, imposing minimum retail sale prices for cigarettes could be a breach of harmonized legislation concerning the internal market, as minimum prices would distort competition. Therefore, increasing minimum excise duties is recommended instead, to discourage consumption (61). Increasing the minimum excise duties would also result in the additional revenue going to the governments instead of contributing to industry profits. A price floor would probably lead to increased industry profits – giving the industry greater funds for its marketing strategies – and lower tax revenue for governments, reducing their ability to cover costs associated with tobacco use. By reducing price competition, the price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifica- tions). Competition among firms may prevent them from raising their prices, 36 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N but a government that imposes a price floor does this for them. Minimum pricing is likely to create windfall profits for manufacturers and retailers. It can even help manufacturers sustain a cartel. If the industry uses the money to increase promotions, advertising or grant proposals for related research, this could undo some of the potential benefits of the policy.10 Some recent evidence shows that, at least in the case of the United Kingdom, increased concentration of power among a handful of multinational corpora- tions is enabling them to undermine tax increases through increased price segmentation, and that requiring minimum prices might be a good way to address the problem. A longitudinal analysis of price data from the United Kingdom (48) has shown that despite regular excise tax increases over time, average real prices for cheaper segments of the tobacco market (in this case, cigarettes and RYO) did not increase – indicating an undershifting of the tax increases in those segments that resulted in increased sales volume. At the same time, average prices for more-expensive market segments increased, indicating overshifting of the tax increases that resulted in decreased sales volume. This industry strategy ensures that the most price-sensitive consumers remain addicted, while encouraging initiation and discouraging cessation. Furthermore, segmenting the market further by overshifting the tax increase on premium brands while undershifting it for cheaper brands mitigates the impact of declining consumption resulting from higher taxes while increasing overall industry margins and profitability. Another situation where setting minimum prices can be a useful policy is specific to the United States. Banning marketing and promotions11 is not possible under the freedom of expression protections of the Constitution of the United States (Amendment I), and it was estimated in 2008 that more than 82% of all advertising and promotional spending by the tobacco industry was focused on reducing the price of their products at the point of sale (62). This limitation on how government can set policy has paved the way for the implementation of minimum price policies in many states and cities to counter the detrimental impact of price promotions on consumption and on the tax policy itself. Huang et al. (55) found that the presence of minimum price laws was associated with higher cigarette prices. They also noted that cigarette prices were even higher than prices resulting from minimum price laws in states that also prohibit industry from engaging in other price-reduction strategies, 10 See, for example, the PMI strategy of setting up the Foundation for a Smoke-Free World and grant proposals for related research. 11 See section below on banning promotional discounts for tobacco products for further discussion about marketing and promotions. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 37 such as below-cost combination sales, using trade discounts to reduce the base cost of cigarettes and distributing below-cost coupons to consumers. In the contexts of both price segmentation and price promotions, the imposi- tion of minimum prices as a complementary policy to excise tax increases – not as a policy alternative – may help guarantee that taxes do indeed lead to the intended reduction in consumption. Nonetheless, more evidence is needed to support the effectiveness of this policy. • Price ceiling – Concerns about differential tax shifting have led to suggestions that a price cap may benefit public health by limiting the tobacco industry’s ability to reduce average prices by differentially shifting tax increases among various price segments (63–64). Because tobacco manufacturers operate across international markets, however, they could maintain low prices in one coun- try but maintain overall profitability by selling more premium products in another country. Additionally, limiting price increases does not fit the public health purpose of reducing consumption. It is worth noting that maximum retail prices are sometimes used as a base for calculating the ad valorem tax payments in countries with ad valorem or mixed tax systems. In a systematic review of the literature on non-tax policy approaches to raising prices, Golden et al. (63) hypothesized how such policies would influence price dispersion and average prices. Their study found that minimum price policies combined with promotion bans have the potential to increase average prices. This is, of course, relevant in a context where price promotions are present. From either a theoretical or a practical standpoint, however, it is clear that price policies cannot be used alone and should always be considered as complements to excise tax increases. Significantly increasing taxes is the most effective way to dissuade consumption, correcting whatever bias may exist. Significant tax increases also provide the added benefit of raising money for the government rather than profits for the tobacco industry. Nonetheless, a minimum price might help narrow the gap between cheap and pre- mium cigarettes when applied to all tobacco products to avoid product substitution. Other non-tax regulation • Banning promotional discounts for tobacco products – The sale of tobacco products at a discount rate – such as through reduced-price coupons or buy-one-get-one-free offers – encourages consumption and undermines tax increases. Such practices should be completely banned. They often exist outside the realm of the finance sector because they are considered a type of marketing – promotional discounts are usually addressed in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. 38 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N According to the 2019 RGTE (1), 118 countries out of 195 with all levels of income had such a provision implemented as of 31 December 2018. • Banning the sale of single sticks of cigarettes – Article 16 of the WHO FCTC, “Sales to and by minors,” paragraph 3, requires Parties to “prohibit the sale of cigarettes individually or in small packets which increase the affordability of such products to minors”. Some smokers opt for buying single sticks partly because of the lower im- mediate costs of buying cigarettes individually (65). Internal (unpublished) analysis of single-stick prices collected by WHO for the 2012, 2014 and 2016 editions of the WHO RGTE shows that, in fact, the aggregate price of 20 single sticks of cigarettes sold separately is generally higher than the price of a 20-cigarette pack sold in the market of a specific country. Despite this fact, single-stick sales – and sales of small-sized packs – make cigarettes accessible to consumers with limited disposable income. De Ojeda (66) found in a study conducted in Guatemala that single-cigarette sales are associated with increased cigarette accessibility for less-educated, lower-income populations and minors. Single-stick sales are also a feature of many markets in South-East Asia, including most notably Bangladesh and India, but also in other parts of the world, e.g. South Africa. Single-stick sales also reduce the impact of a tax increase, since the in- crease per stick is much smaller than the increase per pack (67). In a study investigating how smokers in New York City responded to a tax increase of US$ 1.25 per pack in 2008, Coady et al. (68) found that 15% of smokers bought more single cigarettes than they had previously.12 By allowing single stick sales, governments risk losing part of the ad valorem taxes if the tax base is the retail selling price; the retail price of single sticks is much more difficult to monitor than the retail price of packs of cigarettes, on which, for example, tax stamps with prices can be applied. An internal WHO analysis of the most recent tobacco control laws in 2018 in 195 countries found that 86 countries impose by law a ban on the sale of single sticks of cigarettes (36% of the countries are high-income, and 64% are LMICs). In addition to banning the sale of single sticks of cigarettes, 67 of the 86 countries specify a minimum size for packs of cigarettes. Most (52 countries) use the 20 cigarettes per pack standard, but minimum sizes 12 Before 2018, the use of single sticks was possible, but it has since been banned. See New York City Administrative Code. chapter 7: regulation of tobacco products, subchapter 1: Tobacco Product Regula- tion Act, §17-704.a-1. New York: New York Legal Publishing Corporation; 2020 (http://library.amlegal. com/nxt/gateway.dll/New%20York/admin/title17health/chapter7regulationoftobaccoproducts?f=tem plates$fn=default.htm$3.0$vid=amlegal:newyork_ny$anc=JD_T17C007, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 39 range from five sticks in Ghana up to 25 sticks in Papua New Guinea. Another 15 countries do not specifically ban the sale of single sticks but do specify the minimum size for packs of cigarettes. According to the WHO FCTC, in order to reduce affordability, single stick sales should be banned and a minimum number of cigarettes contained per pack should also be defined. KEY TAKEAWAY 11 A number of non-tax measures are closely connected to tax policies, including price regulations, bans on promotional discounts for tobacco products and bans on the sale of single sticks of cigarettes. The price policies discussed are (1) minimum mark-up, (2) price floors and (3) price ceilings. Current evidence does not yet demonstrate that minimum mark-ups and price floors lead to increases in average price. Nonetheless, they may be relevant in some specific contexts as complementary policies to excise tax increases. Price ceilings limit price increases, which can mitigate their impact on consumption. Price marketing strategies such as promotional discounts and the sale of single sticks undermine the effect of tax policies and should be banned. A minimum pack size should also be required by regulators. Tax increases and their possible impact on inflation At times, the inflationary impact of tax increases on cigarettes and other tobacco products is raised as an argument for not increasing these taxes. This may be a concern in countries where wages and/or a significant share of government spend- ing is indexed to inflation (e.g. for public pension payments) or where government policy is to keep inflation low. The extent to which tobacco product tax increases lead to increases in inflation depends on several factors, most notably the share of these taxes in prices and the weight tobacco prices are given in computing a price index. For example, if taxes account for 25% of tobacco product prices, a doubling of the tax (100% increase) will increase prices by 25%. If the weight given to tobacco products in the price index is 3%, the index will rise by 0.75% in response to the tax increase. As tobacco taxes account for a larger share of tobacco product prices, the inflationary impact of a tax increase will be greater. Similarly, as tobacco products are given more weight in computing a price index, a given tax increase will have a greater inflationary effect. In general, for most countries, the inflationary impact of tobacco product tax increases would be relatively small (47). 40 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Consumer price indexes have multiple purposes. They are an important economic indicator for most countries and are often a key determinant of monetary policy. Inflation rates have a direct impact on interest rates and exchange rates. In many countries, changes in wages, social security benefits and other payments are tied to inflation as measured by a price index. Price indexes are used to provide more accurate comparisons of changes in expenditures, incomes and prices for specific goods over time, as well as to allow comparisons across countries. Given the many uses of consumer price indexes and the potential inflationary impact of tobacco tax increases, some governments have developed alternatives that exclude tobacco (and sometimes other goods) for some uses. For example, since 1992, France has excluded tobacco products from the price index used for adjust- ing minimum wages (47). However, many countries continue to include tobacco product prices in their consumer price indexes. Excluding tobacco products from the basket of goods used in developing key price indexes would greatly reduce concerns about their impact on inflation. In addition, with declining consumption of tobacco products, the inclusion of their prices in key price indexes results in a distorted measure of price for many consumers. KEY TAKEAWAY 12 If governments are concerned about the potential inflationary impact of a tobacco tax increase because wages or some government spending may be tied to a price index, they can use a price index that excludes tobacco products. The importance of taxing cigarettes and other tobacco products in a comparable way While cigarettes are the most commonly used tobacco product globally, other tobacco products are as prevalent and sometimes more prevalent than cigarettes in some parts of the world. Bidis and smokeless tobacco are the main products consumed in some countries in South-East Asia – Bangladesh and India in particular – and waterpipes are widely used for smoking tobacco in the Eastern Mediterranean region (4). These products, as well as RYO, have historically been taxed much less than cigarettes (see, for example, Fig. 2.10 for Bangladesh and India, where the excise tax and prices of bidis and smokeless tobacco are much lower than those for cigarettes). This differential taxation undermines the health impact of excise taxes on tobacco products because (1) it encourages users to switch from cigarettes to the lower-taxed product (see the case of Thailand below); (2) it is not effective in reducing tobacco use in general, especially if the most widely used product in the country is not cigarettes; (3) it can encourage tax avoidance by companies that may CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 41 redefine products that are similar to cigarettes so that they fall within the lower- taxed product category (see the EU example below) and (4) it reduces the size of government revenues since those products could have been taxed at higher levels. Fig. 2.10 Price and tax of the most-sold brand of cigarettes, bidis and smokeless tobacco in Bangladesh and India, 2018 Source: (1). In Thailand, for example, the price of cigarettes has been raised quite successfully through taxation over a number of years, while taxes and prices of loose or RYO tobacco have until very recently remained unchanged. Indigenous tobacco used for RYO cigarettes has historically been exempt from excise, while foreign tobacco was taxed at a low level relative to that of manufactured cigarettes. Consequently, Thailand experienced growth in the RYO market even though cigarette consumption had been falling.13 The Thai government eventually took strong action to address this issue. First, the exemption for indigenous tobacco was removed in 2018. The Cabinet then approved an increase in the excise rate on small producers (of indigenous tobacco) from 0.005 baht per gram to 0.025 baht per gram in 2020, with another increase to 0.1 baht per gram scheduled for 2021 (69). In the EU, the minimum excise duty levels for cigars and cigarillos is significantly lower than that for cigarettes. The Member States of the EU are required to levy an 13 WHO Country Office for Thailand, personal communication, 2019. 12.8 0 190 54.9 64 24 Price Excise tax amount 80 12.8 3.8 44 Lo ca l C ur re nc y 0 50 100 150 200 Cigarettes (20 sticks) Bangladesh, taka India, rupees Bidis (20 sticks) Bidis (20 sticks) Cigarettes (20 sticks) Smokeless (20 g) 42 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N excise duty of at least €90 per 1 000 cigarettes, which should be 60% of the weighted average retail selling price of cigarettes released for consumption. For cigars and cigarillos, only €12 per 1 000 items, or an excise duty of 5% of the retail selling price, is required. As a result, the excise tax share on cigarettes is much higher in many EU countries than the share for cigars and cigarillos. In response, some companies started to market so-called borderline cigarillos. These products have characteristics similar to cigarettes but can be sold at a lower price because for excise purposes, they are considered as cigarillos. Although this issue seems to be largely solved by amendments to the definitions of these products at the EU level and a change in tax structures in some countries, it is important to be aware of the unintended incentives that can be created by large gaps in excise tax levels between product categories (70–71). For more details on industry tactics to undermine tax increases, see Box 2.1. KEY TAKEAWAY 13 To make excise tax on tobacco products more effective in reducing overall tobacco use and to avoid substitution between products, all tobacco products need to be taxed in a comparable way. The Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce the demand for tobacco) (73) recommend that all tobacco products should be taxed in a comparable way. Box 2.1 Industry tactics used to undermine tax increases Tax increases reduce the demand for tobacco products and present a threat to the tobacco industry’s high profits. The industry responds by using various strategies (17, 46, 48), including the following (46): Stockpiling (forestalling/front-loading) – Before the implementation of an an- nounced tax increase, manufacturers overproduce tobacco products, paying the pre-tax-increase rate. As a consequence, sales and tax revenue decline immediately but temporarily after the tax increase (while sales and revenues had increased sub- stantially just before the tax increase) and the industry attributes this drop in revenue to the emergence or increase of illicit trade. This practice results in tax avoidance if there is no law prohibiting it (see also the discussion on anti-forestalling in Chapter 3). Changing certain product characteristics (for example, weight or length) and/ or adjusting the production process – When tobacco products are taxed at different rates or are subject to different tax increases, the industry can, for example, re-label one type of tobacco product as another product that has a lower tax burden (as in the example of the EU above). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 43 Choosing the time of a price increase announcement strategically – The industry may raise prices in anticipation of a tax rate increase, generating extra profits in the period until the tax is actually implemented. When the tax increase is implemented, consumption and tax revenue will fall, but prices will not change, so the industry can claim that the tax policy was ineffective in reducing demand. Adopting price-discriminating strategies or price-related promotions – The industry may offer discounts, retailer rebates or added value (gifts) to tobacco purchases to minimize the loss of price-sensitive consumers. This, however, is not possible in countries where strict bans on tobacco advertising, promotions and sponsorship are implemented. Using brand proliferation (for example, launching a low-priced brand) and price segmentation – Manufacturers can choose to reduce prices of certain brands or introduce new, even cheaper ones to keep price-sensitive consumers in the market. There is evidence that firms introduce new cheaper products and use price-marking – printing the price directly on packs of tobacco products – to lock in their price (48). Such practices compromise both public health and revenue objectives. Differential shifting of tax increases across different price segments, depending on the market circumstances – The industry may increase the price of a product by more than the amount of the tax increase (tax overshifting) and blame the govern- ment for the total increase. Tax overshifting is profitable when demand is inelastic, that is, when the price increase more than offsets the reduction in sales. The industry may overshift the tax increase for higher-priced brands, which are expected to be more price inelastic than lower-priced brands. Additionally, to keep price-sensitive consumers in the market, the industry may temporarily absorb part (or all) of the tax increase on lower-priced brands. The differential tax shifting will lead to different responses in the demand for the different brands (37, 48). Lobbying government to distort interventions – Government policy might be influenced by tobacco industry lobbying, directly or indirectly. Policy-makers are not simply welfare or revenue maximizers; they also value political support. Industry lobbying might lead to adopting a favourable type of taxation, postponing tobacco tax increases or distorting the tax rate downwards (17). Article 5.3 of the WHO FCTC, “On the protection of public health policies with respect to tobacco control from commercial and other vested interests of the tobacco industry”, and its guidelines provide useful guidance on how to address tobacco industry interference. In fact, all 181 countries that are Parties to the WHO FCTC have a legal obligation to implement the requirements of Article 5.3. Having correct expectations about industry responses is important for estimating the impact of a tax increase on consumption and tax revenue. 4 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.2.3 MEASURING IMPACT AND RECOMMENDED INDICATORS Governments need to consider a number of indicators when formulating policy changes. Inappropriate assumptions about consumer behaviour, market structure and industry behaviour can lead to faulty policy analysis. Measuring impact on price and demand Consumption habits, local traditions and industry characteristics – such as the number of different brands offered, the possibilities of cross-border shopping and the presence and level of illicit trade – all affect the shape of the demand and supply of tobacco products, thereby determining the value of the price elasticities. Price elasticity, together with the industry’s pricing strategies – for example, the degree of tax shifting – and the tax share in the retail price, determine the elasticity of the tax base, regardless of whether the base is determined by quantity (for specific taxation) or transaction value (for ad valorem taxation). The importance of elasticity estimates Different types of elasticity should be considered: • price elasticity – own-price elasticity – measures the response of consumers’ demand for a product following a change in the price of the product. – cross-price elasticity – measures the response of consumers’ demand for a product when the price of another product changes. Cross-price elasticity can also occur between different brands or price segments for the same product. • income elasticity – the response of consumers’ demand for a product when their income level changes. Correct estimates of price and income elasticities are important for policy-makers who need to anticipate the impact of a tax increase on consumption and tax rev- enue. Estimates will vary depending on a number of factors, including whether responses are considered in the short run versus the long run, the functional form of the demand function used, whether factors such as addiction or tax evasion are accounted for and the way data are constructed. For example, details such as the degree of aggregation of data, whether gender- or age-specific data are used, the time span covered and which estimation procedures are used (e.g. ordinary least squares, two-stage least squares or generalized method of moments) will all affect the results of the estimate (72). Price elasticities may change over time, as well because of changes in any of the other factors affecting demand, such as income or tobacco control measures, and also because of changes in estimation techniques and the types or sources of data used. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 45 Moreover, what is of most interest is the price elasticity of total demand. A tax increase may reduce tax-paid retail sales but not necessarily total consumption. For example, smuggling can significantly bias price elasticities when the elasticities are estimated using legal sales data; not accounting for possible illicit trade might lead to overestimation. Similarly, when cross-border shopping is included, the price elasticity of demand is lower (in absolute value) (74). Estimating the total price elasticity of demand for legal and illegal consumption can be done by using cross- sectional data from nationally representative household surveys. However, this approach also has its weaknesses. For example, respondents tend to underreport their consumption of tobacco, which leads to bias in the size of demand. Price endogeneity14 is another technical problem that can be challenging to address. To comprehensively estimate the total effect of a tax increase on demand for all tobacco products as well as on tax revenue, the degree of substitutability between them needs to be estimated (55). Cross-price elasticity measures how the quantity demanded of a particular tobacco product changes when the price of another tobacco product increases. When this elasticity is positive, the products are substitutes; the higher the value of the elasticity, the closer substitutes the products are to one another. For example, positive cross-price elasticity between RYO and manufactured cigarettes implies that the demand for RYO increases as the price for cigarettes increases. Substitutability may also arise between different cigarette brands – when the relative price of economy brands increases, demand for premium brands may increase. This effect can be exacerbated when differential (tiered) taxation is ap- plied on different types of cigarettes, further widening the gap in prices between brands and segments and encouraging substitution. The substitutability between traditional and new and emerging tobacco and nicotine products is currently of great interest (see section 2.4 below). In some countries, different tobacco products can also be complementary rather than substitute goods. This means that when the price of a tobacco product increases, the demand for its complement drops because users are unlikely to use the complementary tobacco product alone. For example, a number of studies have found manufactured and indigenous bidi cigarettes to be complementary goods in India (75–76). The sign and magnitude of income elasticity vary across time, countries and demographic groups. For example, in the United States, a high-income country, income elasticity over time has changed from positive to negative, and cigarettes have switched from being a normal good to an inferior good (77–78). On the other hand, among LMICs, where prevalence of smoking tends to be relatively higher, 14 Price is endogenous because it is not an independent variable: it is estimated by dividing expenditure on tobacco by consumption of tobacco, with consumption being a dependent variable in the estimation of price elasticity. 46 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarettes might still be a normal good, with consumption increasing as income levels increase (positive income elasticity) (79–80). There are unobservable characteristics that differentiate higher-income smok- ers from lower-income smokers, such as differences in time and risk preferences, differences in associating a social stigma with smoking and differences in taste for smoking as a pleasurable activity. When these characteristics are ignored, estimates of the correlation between income and smoking-related outcomes are biased. Kenkel et al. (81), using techniques that estimate the causal effect of income on smoking among low-income adults, found that tobacco is a normal (even a luxury) good: higher income is associated with a higher probability of smoking participation and a lower probability of smoking cessation. These results are consistent with those regarding the impact of the business cycle – periods of expansion or recession in economic activity – on health be- haviour and outcomes. Ruhm (82–83), for example, found that smoking declines during temporary economic downturns and increases during economic expansions. Tarantilis et al. (84) found that estimates of income elasticities of demand in Greece were higher after the economic crisis of 2010 than before it. The financial crisis and the austerity measures shifted the demand for cigarettes downwards and turned cigarettes into a more income-elastic good. Interestingly, evidence from Germany suggests that the propensity to become a smoker significantly increases during an economic downturn. However, among those who are already smokers, cigarette consumption actually decreases (85). Ideally, when estimating price and income elasticities, the effect of non-price policies should also be accounted for. A recent study from South Africa shows that failing to take non-price policies into account will overstate the price effect (86). The NCI/WHO Monograph (4) suggests that price elasticity of demand for tobacco is on average -0.4 in high-income countries (ranging from -0.2 to -0.6). Estimates for LMICs are more variable, clustering around -0.5 (ranging from -0.2 to -0.8). A price elasticity of -0.5 means that a 10% increase in price would lead to a 5% reduction in consumption. KEY TAKEAWAY 14 Policy-makers need to know the elasticity of demand – including price elasticity (own-price and cross-price) and income elasticity – for tobacco products in their country in order to correctly assess the impacts of potential policy changes on consumption and subsequent revenues. These estimates need to be made on a regular basis to capture changes in demand over time. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 47 The importance of the tax base elasticity The tax base elasticity measures the sensitivity to a change in the tax rate of the base on which the tax is imposed – the base being tobacco consumption in the case of specific taxation and tobacco expenditure in the case of ad valorem taxation. The magnitude of the elasticity of the tax base depends on price elasticity of demand, the tax structure, the level of the tax rate and its share in price, along with the industry response through its decision to absorb, pass through or overshift the tax on to the retail price. Consumers’ preferences and income, the availability of substitutes and other non-price tobacco control measures also influence the tax base elasticity, essentially through the price elasticity of demand. The magnitude of the elasticity of the tax base also depends on social motivations, including price and tax expectations, which are ultimately impacted by successful tobacco control measures that affect consumers’ willingness to pay taxes or prices. In addition, the tax-base elasticity depends on smokers’ perceptions of the prob- ability of detection and tax enforcement when using illegal products, as well as the availability and accessibility of opportunities for tax evasion and avoidance. Finally, consumers’ willingness to pay taxes depends on their perceptions regarding the use of the tax revenue (87). Therefore, the tax base elasticity is largely influenced by government policy choices. Increasing the tax share in prices is recommended by WHO as a tool to achieve the public health objective of reduced tobacco use: a higher tax share in prices increases the tax base elasticity, all else remaining constant, and therefore increases the reduction in the tax base through the resulting reduction in smoking. However, manufacturers can be expected to attempt to manipulate the tax base elasticity through their pricing policies, such as tax shifting. As discussed earlier in this chapter, industry behaviour is itself affected by government tax policy and regulations. A number of factors need to be taken into account when considering tax pass- through. As discussed earlier, tax is more likely to be overshifted within a specific tax structure than within an ad valorem structure. There is also evidence of industry overshifting the tax for premium or expensive cigarette brands while undershifting the tax for cheaper brands. This indicates that within a given market, the industry’s decision on the extent of tax pass-through will vary based not only on the tax structure but also on the structure of the market. It will also vary by brand. But this does not give an indication about the impact of the tax increase on the average price of a tobacco product. In the context of the tax base elasticity and the impact of tax increases on revenues, it is important to assess how tax increases affect average prices. The example of South Africa is very useful here. Over the past two decades, South Africa has been consistently increasing its specific excise tax on cigarettes, which has led to large price increases. An analysis of the effect of excise tax increases on 48 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarette prices between 2001 and 2015 (26) shows that while there is evidence of tax overshifting, at least within a specified period of time, overall, the tax has been undershifted in real terms. This undershifting is due partly to increased competitive- ness in the market and partly to the introduction of low-priced brands. Of course, this encouraged some consumers to downshift their consumption to cheaper products, but it also pushed manufacturers of more expensive brands to absorb part of the tax increase to reduce the impact on price. A change in the level of the tax rate – with all other factors that influence con- sumption held constant – will result in a change in the tax revenue.15 Estimates of tax base elasticity help governments predict changes in tobacco tax revenues following a tax increase (see details in Annex 2.2). Under specific taxation: • tobacco consumption – the tax base – is expected to be price inelastic (17, 47); • prices increase by less than the tax increase, on average (there is no tax overshifting overall); and • consumption – the tax base – is also expected to be tax inelastic: the quantity of consumption falls less than proportionately to the tax increase, and the tax revenue increases. Under ad valorem taxation: • the tax base is the total consumer expenditure (or, equivalently, the industry sales revenue) on (legal) tobacco consumption – that is, the tax base under ad valorem taxation is determined by both price and quantity, which is itself a function of price; • the sign of the tax base elasticity – which can be either negative or positive – depends on the magnitude of the price elasticity of demand; • since evidence suggests that tobacco demand is price inelastic, the tax base elasticity is positive; • when the ad valorem tax rate increases, both price and quantity adjust, but quantity falls less than proportionately to the price increase, and tax revenue increases; and • a tax rate increase leads to both a higher level of revenue and a lower level of consumption; the value of the elasticity – and hence the tax revenue – increases with the degree of tax shifting. 15 This concept has been used by Laffer to argue that tax increases that are too high will reduce excise tax revenues (the so-called Laffer curve). For a detailed discussion on the Laffer curve, see section 4.4 in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 49 Taxation serves as an instrument for achieving both fiscal and public health objectives. If, after successful tobacco control interventions, prices reach levels where demand becomes elastic, the tax base is still most likely to be inelastic due to tax undershift- ing, since overshifting is not a good pricing policy when demand is elastic. In other words, a tax rate increase – in combination with non-price tobacco control measures that make consumers more sensitive to price (tax) increases – leads to decelerating but still positive marginal revenues. For an example of a tobacco taxation success story, see Box 2.2. For further details of countries’ experiences with tax increases and their impact on revenues, see section 4.4 in Chapter 4. KEY TAKEAWAY 15 Policy-makers’ key policy tool to control demand is tax. Therefore, it is essential they assess not only the impact of price on demand but, more appropriately, the impact of tax on demand: this is the tax base elasticity. The tax base elasticity is essentially determined by (1) the price elasticity of demand, (2) the degree to which the industry will pass the tax on to the retail price and (3) the tax as a share of the retail price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. Box 2.2 A tobacco taxation success story: Turkey Turkey is an example of a country that has been increasing taxes regularly and sig- nificantly over a relatively short period of time and has reaped the benefits of this policy. As shown in Fig. 2.11, the excise tax per pack of cigarettes more than doubled in real terms over 10 years, with the real price almost doubling as well. In parallel, tobacco excise revenues increased by 67% and cigarette sales decreased by 20%. Since the beginning of the country’s Health Transformation Program in 2003, Turkey has successfully increased public health spending and collected more tobacco tax revenue. According to the latest available figures, in 2015, tobacco tax revenue was equivalent to 42% of the country’s public health expenditure and 1.5% of GDP (88). 50 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.11 Tobacco excise revenue and consumption in Turkey (base year 2008), 2008–2018 Sources: Reference 1 for the price of the most-sold brand, Ministry of Finance for the sales and revenue data and IMF world economic outlook, April 2020. See https://www.imf.org/en/Publications/WEO/weo- database/2020/April for the adjustment for inflation. Impact on affordability While price increases clearly have an impact on consumption, when the effects of increasing per capita income of a population are not considered, the price impact may not be as strong as expected. Increases in a population’s income also increase its purchasing power. And, as indicated earlier, tobacco products generally behave like a normal good. Consequently, as income increases, it is expected that tobacco consumption will increase as well. To mitigate this effect, price increases (following tax increases) need to be greater than increases in income. This is where the concept of affordability comes in. Affordability examines the effects of both increasing prices and increasing incomes on consumer behaviour. A common and easy way to calculate affordability, made popular by Blecher and van Walbeek (89), is to use the percentage of GDP per capita required to buy 2000 cigarettes (or 100 packs of 20 cigarettes) in a given year. An increase in this proportion over time will indicate that cigarettes are becoming less affordable and should lead to reductions in consumption. Changes in trends in affordability of cigarettes over time help policy-makers understand how prices are evolving 0 1 2 3 4 5 6 Tu rk is h lir as p er p ac k N um ber of sticks, 10 000 000 Turkish liras 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real excise tax amount Real price Real tobacco excise revenues Cigarettes, per capita sales 1 100 500 1 700 2 300 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 51 relative to a population’s ability to purchase cigarettes and enable them to revise their policies accordingly. Recent studies in India, for example, highlight the wide price differential between manufactured cigarettes and indigenous tobacco products such as bidis and chewing tobacco, as well as the propensity for these indigenous products to become more affordable over time due to favourable or more lenient tax policies towards them (90–91). Figure 2.12 shows the change in affordability of a pack of the most-sold brand of cigarettes by country income group between 2008 and 2018. During this time period, affordability declined in almost 70% of high-income countries, while it declined in slightly more than 35% of middle-income countries and only 26% of low-income countries. Fig. 2.12 Number of countries that have experienced a change in affordability of cigarettes between 2008–2018, by income level Note: Change in affordability was computed as the least squares rate of change in the per capita GDP required to purchase 2000 cigarettes of the most-sold brand in local currency in a given year. The trend rate of growth was computed for countries with four or more years of data, including 2018. Affordability was assessed as not having changed if the least squares trend in the per capita GDP required to purchase 2000 cigarettes over the period 2008–2018 was not statistically significant at the 5% level. Source: (1). KEY TAKEAWAY 16 From a health perspective, in addition to examining the impact of a tax increase on the levels of price, demand and revenues, policy-makers should consider a tax hike that will lead to prices rising more than increases in their population’s income; a tax increase should make tobacco products less affordable to consumers so that demand will be effectively reduced. Cigarettes became less aordable Cigarettes became more aordable Aordability did not change Could not be assessed due to insucient data High income Middle income Low income 39 36 9 13 5 7 37 23 6 13 2 5 52 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Projecting impact on consumption, smoking prevalence and lives saved The WHO interactive smoking projection and target-setting tool (WHO ISPT) The WHO ISPT enables national policy-makers and tobacco control experts to explore the potential impact of proposed tobacco control policies. It uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. The WHO ISPT provides projections of (1) tobacco smoking rates and (2) tobacco-smoking-related deaths in a country under different policy settings and for different time periods. It was designed to promote multisec- toral collaboration within countries by enabling experts from various ministries (for example, health, education, finance, national statistics), civil society, academia and media to explore options for medium- and long-term tobacco control planning together with WHO experts. Use of the WHO ISPT enables strong partnerships for policy change advocacy, program development and evaluation.16 In particular, it can help policy-makers in the Ministry of Finance assess the specific contribution of tax policies – within overall tobacco control policies – towards achieving specific targets in tobacco prevalence reduction. Projecting impact on excise revenue The WHO tobacco tax simulation model (WHO TaXSiM) The WHO TaXSiM is a simple but data-intensive Excel-based tool that helps policy- makers analyse their tobacco tax policy and assess the impact of any excise tax increase or change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Using detailed data about the market – including the majority of brands found in the market, their market share and price levels and the applicable tax – and assumptions about price elasticity of demand, the WHO TaXSiM predicts the impact of tax changes on consumer prices, consumption volume and tax revenues generated by each brand and market segment for the following year. The exercise can be done for multiple years.17 By exploring market data in detail, in addition to assessing the potential rev- enue impact of changes in excise tax, the WHO TaXSiM is a useful instrument for highlighting weaknesses and opportunities in an existing tax system and market. It can also encourage policy-makers to create administrative databases that can be periodically updated to monitor the dynamics of the cigarette market. 16 The WHO ISPT is not available publicly, but WHO will work directly with interested countries upon request to use it to produce data-to-action-type plans. 17 For more information about the methodology, see https://www.who.int/tobacco/economics/tax- sim_background.pdf, accessed 29 September 2020. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 53 KEY TAKEAWAY 17 Policy-makers can use available tools to simulate the impact of tax increases on prices, consumption and revenues, as well as smoking prevalence and lives saved. Recommended indicators to monitor tobacco taxation progress MPOWER WHO publishes a biannual RGTE, which monitors global progress in tobacco con- trol. In particular, the report focuses on the implementation of the policy package MPOWER, a set of proven demand-reduction measures in line with the key provi- sions of the WHO FCTC (1). While raising taxes on tobacco (component R) is proven to be the most effective and cost-effective policy to reduce tobacco use (4), implementing the entire MPOWER package at the best practice level will reinforce the impact of R. For example, as mentioned earlier in this chapter, banning promotional discounts as part of the E measure (enforce bans on tobacco advertising, promotion and sponsorship) will favour price increases following a tax increase. If all the MPOWER tobacco control measures except R were implemented at the best practice level, all else remaining constant, revenues would be expected to decline. Thus, in order to maintain revenue levels, it is important to raise excise taxes on tobacco products regularly to compen- sate for the decline in tobacco use from the other four tobacco control measures. Tax share The main indicator in the R policy in the RGTE (1) is the total share of indirect taxes in the retail price of the most-sold brand of cigarettes.18 Countries whose most-sold brand of cigarettes has a total tax that is equal to or greater than 75% of the retail price are considered to be at the highest level of achievement. While total taxes include excise taxes, VAT (or sales taxes), import duties (when applicable) and other indirect taxes (where applicable), it is preferable to focus on excise taxes, since they are the component that most influences the relative price of tobacco. The share of excise tax in the retail price can be extracted from the RGTE database.19 The 2010 WHO technical manual on tobacco tax administration recommended making excise taxes account for at least a 70% share of excise taxes in the retail price of tobacco products (47). 18 For more details about how this indicator was compiled, see Technical Note III of the RGTE 2019 (https://www.who.int/tobacco/global_report/Technical-Note-III.pdf?ua=1). 19 See taxes and retail price for a pack of 20 cigarettes, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.1-Taxes-and-retail-price-for-a-pack- of-20-cigarette-most-sold-brand.xls?ua=1, accessed 29 September 2020). 54 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Affordability As discussed previously, the share of tax in the retail price is not enough to ensure that a policy will be successful in reducing demand. Any tax increase should lead to an increase in price that will effectively discourage consumption. While global trends indicate that a high tax share is positively correlated with a high price level (see Fig. 2.2 in section 2.1.2), this may not necessarily apply to a particular country; a tax share can be high, while at the same time tobacco products remain afford- able. For this reason, it is important to monitor not only tax increases but also whether those increases led to a price increase that is greater than income increases. As described in section 2.2.3 of this chapter, a common indicator is the percentage of GDP per capita required to buy 100 packs of 20 cigarettes in a given year.20 Other indicators As discussed in detail in sections 2.2.1 and 2.2.2, a good tax structure can make a tax policy more effective in increasing prices and decreasing affordability of tobacco products. Indicators can include whether a uniform excise is applied, whether it is a specific excise and whether it is adjusted regularly for inflation. A number of such indicators are also monitored through the RGTE and can be downloaded online.21 A tobacco tax indicator compiled in 2020 combines the various elements that form a good tobacco tax policy. The Tobacconomics Cigarette Tax Scorecard (92) rates a country’s tobacco tax policy performance based on best practices. The four components that determine the level of performance are (1) cigarette price (in PPP), (2) changes in the affordability of cigarettes over time, (3) the share of taxes (total and excise) in retail cigarette prices and (4) the structure of cigarette taxes (i.e. whether excise is applied; whether it is uniform or tiered; whether excise is specific, ad valorem or mixed; and, for the ad valorem component, if the tax is applied on the retail price and if there is a minimum specific excise and, for the specific component, if tax is automatically adjusted upwards). Each of the four components is given a score, using a five-point index, with the total score reflecting an average of the four component scores. The closer the total score is to 5, the bet- ter the tobacco tax policy performance is in a given country. While this published scorecard is currently applied only on cigarettes, it can be easily applied on other tobacco products, provided the needed data are available. 20 This indicator has also been compiled in the RGTE; see (https://www.who.int/tobacco/global_report/ Table-9.6-Affordability.xls?ua=1, accessed 29 September 2020). 21 See supplementary information on taxation, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.5-Supplementary-information-on-taxation. xls?ua=1, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 55 KEY TAKEAWAY 18 Tobacco taxation works best if implemented as part of a comprehensive MPOWER package. MPOWER is an overall indicator that incorporates all the key demand-side tobacco control measures. KEY TAKEAWAY 19 In addition to assessing the potential impact of a tax increase, policy-makers need to monitor progress over time. The share of the tax in the retail price is an indicator of progress. However, it is important to remember that an effective tax increase must translate into higher prices in order to make tobacco products less affordable. Combining all the components of a good tax policy into one scorecard can also be useful for assessing tobacco tax policy as a whole. 2.3 DOMESTIC AND REGIONAL POLICY INTEGRATION While it is essential to design tobacco tax policies with the utmost consideration of all the aforementioned factors, it is also important to consider how external factors can impact or even impede public health policy objectives. As Chapter 3 explains, cooperation among the various agencies that are directly involved in tax administration, collection and enforcement is important for effective and efficient tax policy implementation. But in the design phase, it is also essential to engage with agencies and other policy-makers that are not directly involved with taxation. Domestically, coordination is required to ensure that policies in non-health sectors do not negatively impact or even counteract tobacco control initiatives. For countries that are part of a regional bloc, harmonization of tobacco taxation is essential to protect the single market – as well as the health of the population – and to prevent tax revenue erosion, tax avoidance and tax evasion. 2.3.1 INTERSECTORAL COOPERATION ON DOMESTIC POLICY Domestic policies in agriculture, industry, trade, finance and labour have the po- tential to create or support incentives at different stages of tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. For example, subsidies provided to farmers or manufacturers involved in growing or processing tobacco can reduce prices and incentivize continued participation or even increase development in these areas, which is counterproductive to the goals of making tobacco products less affordable and reducing tobacco consumption. Multisectoral integration and policy coherence are needed at the country level to ensure that public policies and interventions in non-health sectors do not act against the intended public health impact of tobacco control and taxation. 56 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 20 Greater policy coherence in agriculture, industry, trade, finance and labour should be promoted to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. 2.3.2 REGIONAL TOBACCO TAX HARMONIZATION Policy integration is driven by the recognition that cooperation on domestic policies can substantially increase the gains from forming a regional bloc. Harmonization is desirable and may be necessary in certain areas with spillover effects, such as tax policy, the possibility of a so-called “race to the bottom” or threats to public health. Harmonization could be as simple as setting minimum standards and requirements based on global norms and best practices. Harmonization of tobacco taxation is required to ensure the establishment and proper functioning of a single market, prevent tax revenue erosion, prevent tax avoidance and tax evasion and protect people. When barriers to trade between countries are removed, harmonized tax rates support the single market because they improve the ability of consumers, producers and investors to make decisions that are not distorted by taxation but reflect real opportunity costs. Tax competition – where countries simply undercut each other’s tax rate – could prevent governments from raising sufficient funds to pursue social policy. To avoid such a race to the bottom, countries can establish minimum tax rates within the customs union (93). Even if tax competition is not present, when substantial tax differences exist in neighbouring countries, there is a clear incentive to trade across borders in order to reduce tax payments legally or illegally. The experiences of established regional economic communities offer important policy lessons, not only in terms of the general integration process but also for the process and extent of tax policy coordination. The EU implemented a successful regional tax harmonization scheme. Over the years, the focus in harmonization of tobacco taxes has broadened from the elimination of tax obstacles to the fight against harmful tax competition, tax avoid- ance and tax evasion and, more recently, to public health protection. Naturally, addressing these issues requires increasing convergence in fiscal policy and tax administration. Although price differentials still exist, setting a minimum on the share of taxes in the final price of tobacco products as well as a minimum excise tax has helped countries reach some level of harmonization. The EU experience confirms that both a declining tobacco consumption trend and stable revenues can be achieved with harmonized minimum excise rates (94). Moreover, the harmonization process has CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 57 offered opportunities for the enforcement authorities (tax or customs) to obtain information that can be used in the fight against fraud and tax evasion. On the other hand, the experience of harmonization efforts in the West African Economic and Monetary Union (WAEMU) shows how the absence of a supranational body (like the EU) or a hegemonic member state (see the SACU example below) can slow down policy integration that would benefit all member countries (95). The eight countries of the WAEMU are bound by a Tax Directive22 that requires them to impose an ad valorem excise on the CIF value or producer price of tobacco products, which is subject to under-declaration and is difficult to ascertain. Additionally, a maximum excise rate is imposed, and some members apply additional taxes to deal with this constraint. The Directive was revised in 2017 (96), but unfortunately the tax structure remains the same, and the maximum rate was not removed but rather has been increased. The Southern African Customs Union (SACU), which has five member countries, is the oldest existing customs union, established in 1910. Thanks to the hegemonic lead of South Africa, a country with a sophisticated administration system and an aggressive tobacco tax policy, SACU adopted a well-integrated tax policy that has benefited all its members (95). The GCC, established in 1981, is a regional intergovernmental political and economic union consisting of six states of the Persian Gulf. Home to one fifth of the global oil supply (97), the GCC has never relied on taxation as a source of revenue; no direct or indirect taxes were applicable in the region. Although there was no excise on tobacco products, as a customs union, the GCC countries have a common external tariff. This common tariff includes harmonized rates but also a harmonized structure. The import duty is 100% of the CIF value of tobacco products imported in the region, with a minimum tax per quantity imported. However, in recent years, to reduce their dependence on income from oil, GCC countries have considered diversifying their sources of income, including by de- veloping reliance on indirect taxes such as excise and VAT. In 2015, a decision was adopted at the 36th GCC summit meeting to implement selective taxes on all imported tobacco products and cultivated raw tobacco grown domestically (GCC Decision number 963/1). A follow-up decision in December 2016 formally agreed to the introduction of an excise tax on tobacco and other products such as sugary and energy drinks, as well as special goods (alcohol and pork meat), in all GCC countries. The decisions at the national level to implement this subregional decision came into force gradually in all GCC countries, starting with Saudi Arabia, which began 22 Directive No. 03/98/CM/WAEMU on the harmonization of Member States’ legislation of excise duties was adopted 22 December 1998. It was amended by Directive No. 03/2009/CM/WAEMU of 27 March 2009 with the objective of harmonizing excise duties within WAEMU. 58 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation in June 2017, followed by Bahrain and the United Arab Emirates (UAE) later that year, Qatar in 2018 and Oman in 2019 (1). Only Kuwait has yet to adapt its national laws accordingly. The excise introduced by the GCC countries has a structure somewhat similar to the import duty on tobacco products: the rate is 100%, but the base was changed from the CIF value to the retail price excluding taxes. The introduction of the excise led to large increases in the price of the most- sold brand of cigarettes in member countries between 2016 and 2018 – by 33% in Bahrain, more than 80% in the UAE and more than 100% in Saudi Arabia (1). In federations such as Canada and the United States – where the central govern- ment has real taxing power and some financial and regulatory control over the states or provinces – tobacco taxes are not harmonized (98–99). Even though there are significant interjurisdictional differences in taxes and prices, and tax harmonization holds great potential to reduce the scope of illicit transactions in the tobacco market, there is little evidence that Canadian provinces or individual states in the United States are interested in tobacco tax harmonization. Tax harmonization is most relevant in the context of further economic integration within a group of countries that are already part of a customs union, but it needs to be planned well to be effective. Discrepancies in law interpretation and a lack of standardization of tobacco product definitions and tax base lead to suboptimal situations. Tax rate alignment, or setting minimum rates, should come after tax structure alignment. It is important that governments support the move towards harmonization and are committed to dedicating enough financial resources and skilled personnel to oversee the entire process. KEY TAKEAWAY 21 In the context of regional economic integration and ongoing discussions regarding the possibility of harmonizing tobacco excise taxation among member countries, the experiences of existing groups can be instructive. So far, only the EU, SACU, WAEMU and, more recently, the GCC have effectively implemented a harmonized approach to excise taxation of tobacco products. Lessons learned indicate that harmonization should be planned well and should not come at the expense of tobacco control. Setting a common minimum specific excise tax, adjusted over time, is the best approach. This ensures that taxes and prices are above a minimum level, encouraging equalization of price levels and at the same time reducing affordability across countries. On the other hand, agreeing on maximum tax rates is a bad policy. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 59 Countries that wish to raise their taxes further for revenue purposes, health concerns or both should be given the space to do so. Concerns about illicit trade provoked by higher tax rates are best dealt with by strong cooperation in administration and enforcement, information sharing and adoption of new technology with common or interoperable information systems. 2.4 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In recent years, awareness of tobacco risks and harms, implementation of tobacco control provisions – especially under the WHO FCTC – and tightening of regulations have resulted in declining sales of cigarettes, primarily in high-income economies. This has changed the dynamics of the tobacco market. In response to these effective tobacco control measures, the tobacco industry has diversified its business by promot- ing a new portfolio of products, which they claim to be technological innovations that supposedly reduce the harms and risks associated with conventional tobacco products, particularly cigarettes. So-called novel tobacco products have been promoted by the tobacco industry as “cleaner alternatives,” “safer alternatives” and “reduced harm/risk products” with no smoke and no ash. On the basis of these claims, they negotiate for less-restrictive regulatory environments within countries. Some of the new products are also mar- keted or promoted for smoking cessation, despite the evidence of this outcome being inconclusive. Where these products are not banned, one of the debates in the global health community concerns the issue of their regulation and taxation. 2.4.1 HEATED TOBACCO PRODUCTS (HTPs) HTPs are tobacco products that produce aerosols containing nicotine and toxic chemicals upon heating of the tobacco or activation of a device containing the tobacco. These aerosols are inhaled by users sucking on or smoking the device. They contain the highly addictive substance nicotine (found in tobacco) as well as non-tobacco additives and are often flavoured. The tobacco in HTPs may be in the form of specially designed cigarettes (e.g. so-called heat sticks or Neo sticks) or pods or plugs. These products include IQOS from PMI, Ploom TECH from Japan Tobacco International (JTI), glo from British American Tobacco (BAT) and PAX from PAX Labs. HTPs differ not only from con- ventional cigarettes but also from ENDS – some of which are called e-cigarettes – as ENDS do not contain tobacco but rather a nicotine solution (see next subsection). However, the boundaries between the different products are becoming increasingly 60 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N difficult to define, given the emergence of so-called hybrid tobacco products that contain both nicotine solution and tobacco. HTPs are currently available in more than 40 countries and are banned in fewer than 10 countries. Even in countries where they are regulated, there is significant variation in the approaches taken to regulation. A variety of factors affect a country’s ability to control and regulate the use of HTPs, including national regulatory pow- ers, enforcement capacity regulatory frameworks, country capacity and tobacco industry interference (1). Most countries tax HTPs at a lower rate than cigarettes and on the kilogram of tobacco as a base when applying a specific or mixed excise (see Table 2.4). The use of such a base may be quite challenging for tax collection, especially because of the difficulty of checking the tobacco content in each stick. In the past, some countries taxed cigarettes per kilogram of tobacco, but today it is common practice to tax them per stick regardless of tobacco content. Table 2.4 Excise taxation of HTPs, first collected for July 2018 – updated for July 2020 OVERALL COMPARISON WITH CIGARETTES Type of excise Base unit is kg, overall rate lower than cigarettes Base unit is sticks, rate is the same as cigarettes Base unit is sticks, rate is lower than cigarettes Other Specific excise Albania, Austria, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Greece, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Montenegro, Netherlands, New Zealand, North Macedonia, Romania, Russian Federation, Slovakia, Slovenia, Sweden, United Kingdom Azerbaijan a, Japan, Ukraine b Armenia, Hungary, Jordan, Italy c, Philippines, Republic of Korea d Montenegro e, Republic of Moldova f , Serbia g Ad valorem excise (base is retail price unless specified other- wise between brackets) Spain, Switzerland Saudi Arabia and United Arab Emirates (base is retail price exclusive of excise and VAT) Indonesia h CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 61 Mixed system (ad valorem compo- nent based on retail price unless specified other- wise between brackets) France, Germany, Poland, Portugal Colombia, Georgia, Israel and West Bank and Gaza Strip (ad valorem excise base is wholesale price) a The specific excise rate applied is the same as for imported cigarettes, higher than the rate applied to domestically produced cigarettes. b The rate is the same as the minimum excise on cigarettes per 1 000 pieces. Rate and structure were effective as of 1 January 2021. c The specific excise rate is defined as 25% of the excise tax on cigarettes based on an equivalency used between cigarettes and HTPs. There are planned increases of this proportion to 40% by 2023. d In 2020 the specific excise rate was only 11% lower than cigarettes. e The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. f Specific excise rate is higher than for cigarettes but, unlike HTPs, cigarettes also face an ad valorem excise. Overall effect of excise is a slightly lower for HTPs. g The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. There is a planned phased increase of this proportion aiming equalization with cigarettes by 2025. h While cigarettes face a specific excise tax rate, HTPs face an ad valorem rate, the highest rate as defined by law, on the basis of a pre-defined minimum price. Sources: (1, 100, 101, WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021 and the Campaign for Tobacco-Free Kids website on Taxation and Price for Heated Tobacco Products https://www.tobaccofreekids.org/what-we-do/global/taxation-price/staging-tax-gap). A study by Liber (102) compared prices of HTPs and cigarettes in 34 countries and showed that while taxes have been systematically lower for HTPs than for cigarettes, prices were higher in half of the countries surveyed. KEY TAKEAWAY 22 HTPs, when taxed, are usually taxed lower than cigarettes, although they generally seem to be priced higher than cigarettes. It is important to remember that HTPs are tobacco products, and the same provisions that apply to tobacco products should apply to them as well. This is articulated in WHO’s information sheet on HTPs (103), which provides guidance on how these products should be regulated, as well as Decision FCTC/COP8(22) for novel and emerging tobacco products. Moreover, MPOWER measures, which help WHO Member States to implement the demand-reduction articles of the WHO FCTC, are applicable to HTPs, in particular, Article 6 for taxation. Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. Furthermore, HTPs contain chemicals not found in cigarette smoke, the health effects of which are not yet known. Independent assessment of industry data demonstrates that more than 20 harmful and potentially harmful chemicals 62 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N are significantly higher in HTP emissions than in cigarette smoke (104). Therefore, there is a need to learn more about these products and the health impacts of their emissions, as well as the impacts of exposure to these emissions. KEY TAKEAWAY 23 Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. From both public health and tax administration perspectives, HTPs should be taxed at the same level and in the same way as tobacco cigarettes. Some countries have already adopted this approach and are taxing HTPs at the same rate per stick as cigarettes (Azerbaijan, Colombia, Georgia, Israel, Japan, Ukraine and West Bank and Gaza Strip). Saudi Arabia and the UAE, which have recently introduced an excise tax on tobacco products as part of the GCC, are now applying the same import duty rate and excise tax structure for cigarettes and HTPs. Continuing developments in technology and changes in products have led to a recommendation to tax HTPs per unit. The definition of unit may vary by product within the HTP category. For example, one unit of IQOS is one heat stick, for Glo it is one Neo Stick and for Ploom TECH it is one tobacco pod. Governments will need to determine the exact definition of a unit for each product allowed on the market. The potential complexity of the market strongly supports limiting the types of HTPs allowed in a country and setting strict regulations to standardize the products as much as possible. Countries can also consider taxing the devices used to consume HTPs, i.e. the holder and the charger (see product description in Annex 3.1). KEY TAKEAWAY 24 HTPs are tobacco products, and they need to be treated as such. Where they are not banned, HTPs need to be strictly regulated and taxed. The recommendation is to tax them at the same level as cigarettes on a per-unit basis. Countries can also consider taxing the devices used for HTP consumption. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 63 2.4.2 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS)23 Products like ENDS and ENNDS have evolved rapidly over the past decade. ENDS heat a solution (e-liquid) containing nicotine, but not tobacco, and other chemicals that may be toxic to people’s health to create an aerosol, which is inhaled by the user. Examples of ENDS include Juul from Juul Labs, Vype from BAT and blu from Imperial Brands (1). Electronic non-nicotine delivery systems (ENNDS) are essentially the same as ENDS, but the e-liquid used generally does not contain nicotine. Upon testing, however, many so-called zero-nicotine solutions are found to contain nicotine (105–107). While generally considered as a single product class, ENDS products constitute a diverse group with potentially significant differences in the production of toxicants and delivery of nicotine. There are several coexisting types of devices for ENDS/ ENNDS on the market, including first-generation or so-called cigalikes, second- generation tank systems and even-larger third-generation or personal vaporizers. Collectively, they are also often referred to as e-cigarettes, vapes or vape pens. Other categories of ENDS include e-hookahs, e-pipes and e-cigars – hence, ENDS is an all-encompassing term for multiple product categories. Some of the products resemble their conventional tobacco counterparts – cigarettes, cigars, cigarillos, pipes or hookahs – while others are shaped more generically like pens, USB memory sticks or basic cylinders. Different forms of nicotine are also used in these ENDS, the most recent one being nicotine salts, which deliver high levels of nicotine (1). There are two types of ENDS/ENNDs products: open systems and closed systems. Open systems are devices that allow the user to buy e-liquids and fill their device with the mixtures they want (with no nicotine, different nicotine concentrations and/or flavours). Closed systems are products that come with a prefilled container (called a cartridge, pod or tank). For the past decade, divisive debates have been waged over the effectiveness of ENDS as smoking cessation aids – especially for tobacco users who are unable to give up the habit – as well as the possibility of ENDS playing a role in public health. However, the evidence remains inconclusive. Despite the tobacco industry and other related industries promoting these products as tools for quitting smoking, current evidence does not support their use as part of a population-based cessation strategy (108). Accordingly, the United States Surgeon General, in January 2020, concluded that 23 It is worth noting that ENDS are not tobacco products and not exactly new products – the technology has been around since the late 1980s (e.g. Premier, Eclipse and Accord). However, the recent generation of these products is new and has more or less piggybacked on the success of e-cigarettes. 64 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N E-cigarettes, a continually changing and heterogeneous group of products, are used in a variety of ways. Consequently, it is difficult to make generalizations about efficacy for cessation based on clinical trials involving a particular e- cigarette, and there is presently inadequate evidence to conclude that e-cigarettes, in general, increase smoking cessation (109). The evidence on the adverse health effects associated with use of ENDS is mounting, and when ENDS are used in combination with smoking – which is the practice of the majority of ENDS users (110) – the adverse health effects of two or more products are combined. However, there are insufficient data to understand the full breadth of these effects, as ENDS have not been on the market long enough for their long-term effects to be established. Nevertheless, the evidence is clear that the aerosols of the majority of ENDS and ENNDS, some of which are cancer causing chemicals. ENDS also contain nicotine, which is highly addictive. In addition, ENDS are associated with increased risk of cardiovascular diseases and lung disorders, as well as adverse effects on the developing fetus during pregnancy (108, 110). For adolescents, the use of nicotine can lead to dependence and may harm brain development. Use of ENDS could also lead to a new generation of nicotine and tobacco users, as seen in some countries, especially since these products are designed to appeal to young people. Although the specific level of risk associated with ENDS has not yet been determined conclusively, these products are undoubtedly harmful. Therefore they should be strictly regulated if allowed to be sold in domestic markets, and must be kept away from children. Taxation will be a key component of regulation, since it is an effective tool for influencing consumer behaviour. Some countries have taken the bold decision to completely ban these products. Approaches that have been taken range from partial to comprehensive bans, and ENDS/ENNDS products were banned in more than 30 countries in 2018.24 In other countries, they are regulated as, for example, consumer products, pharmaceutical products or tobacco products, or they are completely unregulated. WHO recom- mends that where ENDS/ENNDS are not banned, they should be regulated to achieve the following objectives: 1. prevent the initiation of ENDS/ENNDS by non-smokers and youth, with special attention to vulnerable groups; 2. minimize as much as possible potential health risks for ENDS/ENNDS users and protect non-users from exposure to their emissions; 3. prevent unproven health claims being made about ENDS/ENNDS; and 24 Data collected for the WHO RGTE 2019. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 65 4. protect tobacco control activities from all commercial and other vested interests related to ENDS/ENNDS, including the interests of the tobacco industry. ENDS/ENNDS present a risk to youth, who have taken up their use in high numbers in some countries, including Canada and the United States (1, 111). The Juul brand, for example, has quickly gained a significant e-cigarette market share in the United States (112–113). Its marketing and popularity have led the United States Food and Drug Administration (FDA) to raise serious concerns and to seek solutions to effectively prevent youth from taking up the use of ENDS/ENNDS (114). The city of San Francisco banned the sale of e-cigarettes in June 2019 (115). In addition to posing a risk for initiation by youth, ENDS can attract non- tobacco users or prevent current smokers from quitting. Taxation could play a role in preventing the uptake of these products, specifically among non-smokers, vulnerable groups, children and adolescents. KEY TAKEAWAY 25 The long-term health effects of ENDS/ENNDS products are still unknown, but they are clearly harmful to health. Furthermore, evidence on the effectiveness of ENDS products as a smoking- cessation aid remains inconclusive. Taxing these products could play a role in preventing their uptake, specifically among non-smokers, vulnerable groups, children and adolescents. Price elasticity of demand for ENDS products In the context of taxation, it is important to ask whether demand for ENDS is price-responsive. Preliminary evidence, although almost exclusively focused on e-cigarette data from the United States, indicates that this is the case: demand for e-cigarettes may be even more price-responsive than the demand for conventional cigarettes, so taxes can be used to deter initiation by never-users (116–123). Most of the studies of price elasticity of demand for ENDS products also demonstrate that e-cigarettes and conventional cigarettes are partial substitutes – that is, they show positive cross-price elasticity. The magnitude of the elasticity indicates the degree of substitutability between products: the higher its value, the closer the products are to being substitutes, with higher cigarette prices being associated with increased e-cigarette sales. Some of the studies also show a substitutability effect in the other direction, with increased prices for e-cigarettes leading to an increase in conven- tional cigarette use (117, 120). All of the studies show evidence of substitutability except for one (124), which differentiates between exclusive and dual users and shows no evidence of substitution between e-cigarettes and conventional cigarettes. 66 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The presence of concurrent (or dual) use – consumers using both conventional cigarettes and ENDS products – complicates results and highlights the need for more research in this area to better disentangle the different effects. Liber et al. (125) analysed sales prices in a sample of 45 countries and concluded that comparable units of conventional cigarettes cost less than disposable e-cigarettes. The units considered for pricing e-cigarettes included both the e-liquid and the rechargeable device. Taken alone, the price of e-liquids is on average much lower than that of cigarettes in high-income countries and the same in LMICs. The time needed to buy back a rechargeable device is estimated to be less than two weeks in most countries. One can argue that increasing price differentials by further increasing taxes on regular cigarettes could be effective in driving current smokers of regular cigarettes to e-cigarettes (126) as a potentially lower-risk alternative (127). However, the ef- fectiveness of ENDS as smoking cessation devices is still being debated; a study by Sweet et al. (128) shows that dual use of e-cigarettes as a potential tool for cessation was effective only in the short term. Moreover, significantly more smokers said they would quit if cigarette prices doubled and e-cigarettes were not available (122) or that they would never have become addicted to nicotine if e-cigarettes had not been so readily available (129). Once an e-cigarette user is addicted to nicotine, there is a risk of initiating traditional tobacco products use (130). In general, cessation can be better facilitated by governments via stronger implementation of the other tobacco control policies that have been proven effective at reducing use. KEY TAKEAWAY 26 Few studies are available on the price elasticity of ENDS products, and the available data come almost exclusively from the United States. These early studies indicate that demand for e-cigarettes will go down as the price of e-cigarettes increases. Generally, the results also show that cigarettes and e-cigarettes are partial substitutes, where an increase in cigarette price would increase the demand for e-cigarettes while reducing demand for cigarettes. But these results do not differentiate between people who are exclusive cigarettes or e-cigarette users and those who are users of both products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 67 Tax structure Different countries impose different tax structures on ENDS/ENNDS products25 (see details in Table 2.5). The Republic of Korea, for example, imposes a specific tax per millilitre of ENDS/ENNDS e-liquid (131), while Indonesia imposes an ad valorem tax on the retail price of the e-liquid; the maximum rate allowed by law for tobacco products (132). In the United States, there is no common way to tax e-cigarettes among the states that do tax them (133–134). The situation is similar in the EU, where new and emerging nicotine and tobacco products are not currently covered by the tobacco tax directive, and Member States may apply a national tax as they see fit under their own rules. All the EU countries that tax ENDS products apply a specific excise per millilitre of e-liquid. These different tax treatments have the potential to distort the functioning of the internal market. In February 2020, the European Commission concluded that the current provisions of the harmonized directive are no longer relevant for the taxation of ENDS and HTPs, and this is a source of concern from the internal market perspective (135). In June 2020, the Member States of the EU reiterated that it is urgent and necessary to upgrade the EU regulatory framework by harmonizing defini- tions and the tax treatment of novel products such as ENDS/ENNDS and HTPs (2). Table 2.5 Types of excises applied on ENDS/ENNDS products e-liquids globally and in individual states in the United States, as of July 2019 (updated as of July 2020 for all countries except the United States) TYPE OF EXCISE COUNTRIES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Albania, Kazakhstan, Kyrgyzstan, Pakistan, Portugal, Republic of Korea, Romania, Russian Federation, Slovenia, Sweden Azerbaijan, Cyprus, Estonia, Finland, Georgia, Greece, Hungary, Italyb, Latvia, Lithuania, Montenegro, Morocco, North Macedonia, Philippines, Serbia Ad valorem (% of retail price or import value) Bahraina Indonesia, Jordan, United Arab Emirates, Yemen TYPE OF EXCISE INDIVIDUAL STATES IN THE UNITED STATES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Delaware, Illinois (Chicago), Cook County, Louisiana, Ohio, Puerto Ricoc, Connecticutc Kansas, North Carolina, Washington, West Virginia, Wisconsin 25 The focus is on the e-liquid used for ENDS/ENNDS products. 68 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ad valorem (% of wholesale/ distributor price) Alaska (Juneau, Matanuska- Susitna Borough), California, Illinois, Maine, Maryland (Montgomery County), Minnesota, Nevada, Pennsylvania, Vermont, Washington DC, Virgin Islandsc New York Mixed New Jersey, New Mexico a Tax applied to e-shisha (or e-hookah) because e-cigarettes are banned in Bahrain. b Italy imposes differential rates for nicotine and non-nicotine containing liquids. c States in which it is unclear if only ENDS or both ENDS and ENNDS products are taxed with an excise. Sources: (1, 135, complementary data from Frank Chaloupka and WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021). Table 2.6 provides reference material on the pros and cons of different considerations for determining the tax structure and base of ENDS/ENNDS products e-liquids. Table 2.6 Excise tax options for ENDS/ENNDS products e-liquids TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Specific Volume of nicotine- containing e-liquid (regardless of concentration) 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only the volume needs to be determined 1. Difficult to compare if tax equivalencya with cigarettes is sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarettes and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed 5. May encourage more do-it- yourself (DIY) products where e-liquids are mixed by the users themselves, which increases the risk of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 69 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of e-liquid regardless of nicotine presence 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only volume needs to be determined 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Does not require laboratory capacity to detect the presence of nicotine in liquids 1. Difficult to compare if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed Volume of all e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL are affected the most; also reduces the price gap between different products 2. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 3. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 4. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 5. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to user 1. Difficult to compare with cigarettes if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed 70 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of nicotine- containing e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL affected the most; also reduces the price gap between different products 2. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as the nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 3. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 4. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to the user 1. Difficult to compare if tax equivalency with cigarettes is being sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids as there is no simple way currently available to determine whether the e-liquid contains nicotine; self-declarations by industry are not sufficient 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed Ad valorem Producer price/ CIF value of nicotine- containing e-liquid (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 71 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Producer price/ CIF value of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Retail price of nicotine- containing e-liquids (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 2. Requires capacity to monitor the market to assess market prices 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled in an e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risk of accidents, illness and death 72 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Retail price of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires capacity to monitor the market to assess retail prices 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Ad valorem with minimum specific or mixed Ad valorem + min specific 1. Volume of nicotine- containing e-liquids will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of nicotine- containing e-liquids will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up. b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products 1. Requires capacity to monitor the market to assess retail prices 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotin; (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Difficult to set a minimum specific excise amount/specific excise amount, especially if tax equivalency with cigarettes is sought 5. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 73 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Ad valorem + min specific 1. Volume of all e-liquids regardless of nicotine presence will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of all e-liquids regardless of nicotine presence will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 1. Requires capacity to monitor the market to assess retail prices 2. Difficult to set a minimum specific excise amount/specific excise amount, especially if equivalency with cigarettes is sought 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels a Tax equivalency can be measured in different ways: (1) in terms of tax burden (as % of the retail price) or (2) as the exact amount of tax for equivalent quantities (assuming an equivalency between a certain volume of e-liquid and a pack of cigarettes). b There is also the option to use the producer price/CIF value as a base for the ad valorem component, but it is a weaker option because the base is difficult to ascertain and therefore prone to undervaluation. Note: Table compiled following a WHO Expert Meeting on Taxation of Electronic Nicotine and Non-Nicotine Delivery Systems (ENDS/ENNDS), Geneva, Switzerland, 2–4 September 2019. There is currently a lack of evidence on the practical challenges being faced by countries favouring one approach over the other. Furthermore, such data are difficult to obtain because the nature of the market is constantly changing. However, a clear recommendation can be made with regard to which e-liquids to tax. As indicated in Table 2.5, some countries tax all e-liquids – whether or not they contain nicotine (ENDS and ENNDS products) – while some tax only nicotine- containing e-liquids (ENDS products). As shown in Table 2.6, there is evidence that in a number of instances, ENNDS products do contain some nicotine. Additionally, ENNDS products are not harmless (136–137). It is therefore recommended that all e-liquids be taxed for both ENDS and ENNDs products. The question of whether to employ differential taxation based on nicotine content seems reasonable from a health perspective, since nicotine is a toxic substance. However, this would likely create an additional burden for tax administrators as 74 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N they would need to determine the nicotine concentration of e-liquids on the market. Additionally, this may no longer be relevant, as advancements in technology indicate that other features of the product can influence nicotine delivery beyond the actual concentration of the e-liquid. It is now possible to increase nicotine delivery at low nicotine concentrations by increasing battery power (by reducing resistance or increasing voltage) (138). In terms of implementation, while most countries seem to have adopted a specific excise tax on ENDS/ENNDS e-liquids per millilitre, one benefit of implementing ad valorem taxation is that it seems relatively easier to regulate in the context of a large heterogeneity of products. However, it is essential that the tax be applied on the retail price value of the products, as this base is easier to ascertain than any other value that could be declared by the manufacturer. It is also important to add that regulation of the characteristics of ENDS/ENNDS products is essential, and it should be implemented along with any tax policy adopted. Regulations should include: 1. setting a maximum nicotine concentration per millilitre to safeguard public health, including reducing the risk of dependence, especially among youth; 2. setting a maximum volume for cartridges to reduce toxicants exposure and possibly limit use; 3. setting a maximum capacity for refill containers to reduce toxicants exposure and possibly limit use; 4. setting a maximum battery power to reduce the possibility of influencing nicotine and toxicant delivery; and 5. taxing nicotine regardless of its source (e.g. tobacco, eggplant, synthetic). Countries may choose to impose an excise tax on ENDS and ENNDS devices26 as well. The easiest type of tax would be an ad valorem tax based on the declared retail price. If countries choose not to impose an excise tax on these products, they should at least impose the regular VAT or sales tax rate. Imposing an excise tax on devices can be challenging from an administrative perspective, as all components need to be clearly defined and classified as devices for ENDS/ENNDS consumption. For example, if the device is assembled after importation and some parts may be used for other purposes than ENDS/ENNDS consumption, authorities may face a challenge in detecting and differentiating which component parts would be subject to excise tax and which would not. 26 See Annex 2.3 for an overview of elements of devices used in ENDS/ENNDS products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 75 KEY TAKEAWAY 27 There is currently not enough evidence to recommend one tax structure over another for ENDS/ENNDS products. However, it is clear that taxing the e-liquids used for consumption is key. The excise tax should be applied on all e-liquids, whether or not they contain nicotine. If the preferred type of excise tax is ad valorem, it should be applied to the retail price. Countries can consider taxing devices as well, but they need to adequately assess their administrative capacity to do so. Policy-makers need to be mindful of the diversity and rapid evolution of ENDS/ ENNDS products and adjust accordingly. Regulation must reflect this reality so that loopholes will not be exploited by the industry. For example, ENDS/ENNDs products include not only e-cigarettes, vapes and vape pens but also other categories such as e-hookahs, e-pipes and e-cigars. Lawmakers need to be clear about how ENDS/ENNDS products are defined so that subcategories do not fall under the radar when regulation comes into effect. Definitions will also be relevant when it comes to taxation. An unclear definition can lead to a seemingly contradictory situation, such as in Bahrain, where e-cigarettes are banned but e-hookahs are not.27 Finally, while policy-makers need to be mindful of the emergence of new products and must take appropriate actions to protect the health of their citizens, it is important to remember that the overwhelming share of nicotine consumption remains that of tobacco products, especially cigarettes. The total market value of ENDS/ENNDS and HTPs sales in 2018 was less than 2.2% of the total market value, while cigarette sales alone accounted for 91% of the same total market value (139–140). 2.5 CONCLUSIONS An overview of excise tax application globally reveals a broad variety of price and tax levels, as well as structures used for taxing tobacco products, in particular, cigarettes. Some trends, however, indicate that tax and price levels are higher among higher- income countries. The rate of taxes also matters: higher tax rates are correlated with higher prices, and higher prices change behaviour, which leads to a reduction in consumption. More countries are moving away from ad valorem taxes and towards either mixed or specific excise systems, and there are few countries that do not impose any excise tax on cigarettes. 27 In Bahrain, the Ministry of Production and Trade Decision 38 of 2013 banned e-cigarettes, while the official list of excisable products from the Ministry of Finance includes e-shishas (or e-hookahs), making them apt to be taxed and therefore considered legal. 76 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Significantly increasing the taxes and prices of tobacco products is the most effective and cost-effective policy to control tobacco use. Increased taxes – which are passed on to smokers as higher prices – reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face several challenges, ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues, as well as the perceived contribution of the tobacco sector to economic development. In designing tax policy, the tax structure adopted not only affects consumption overall, it also shapes the market structure. Ad valorem taxation incentivizes industry to set prices lower than specific taxation does. Evidence suggests that under a specific tax, the price gap between premium and lower-priced products is narrower, therefore reducing incentives for substitution to lower-priced products following a tax increase. However, as industry consolidates producers and widens its portfolio of products, new evidence indicates that the industry is introducing cheaper brands while increasing the price of its expensive brands, therefore, paradoxically, widening the price gap between its products. Evidence also suggests that prices are higher under a specific excise tax structure. Additionally, from a tax administration perspective, a specific tax is easier to imple- ment, since only the quantity produced needs to be ascertained rather than the value of the product. Another aspect of tax structure is the use of tiered taxation – that is, tax rates that vary on the basis of different product characteristics. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax. Tiered taxation encourages substitution from premium to cheaper brands, maintaining smoking prevalence and reducing the health im- pact of tax rate increases. In addition to leading to lower prices, tiered taxation is difficult to administer and creates opportunities for the tobacco industry to avoid and evade taxes. The design of a tax structure must also consider the base on which tax is applied. The choice of base should lead to the highest possible effect on price and revenue. For specific taxation, the tax base is the quantity. When the tax is ad valorem, the CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 77 choice of the tax base is important not only for health considerations, through its effect on consumption, but also for tax revenue generation, as well as industry profits. An ad valorem tax based on the producer price, or CIF value, gives tobacco manufacturers opportunities to reduce their tax liability, especially when they control the distribution system through related parties. The best practice in an ad valorem (or mixed) excise structure is to use the retail price as the tax base and introduce a minimum excise tax per pack. Other tax design considerations include the importance of using automatic adjust- ments and indexation to inflation and income growth for the specific excise tax in order to avoid erosion of the tax over time. Emerging evidence indicates that tobacco taxation does not always achieve the intended results, because the tobacco industry finds ways to circumvent it. Non- tax policies such as pricing regulation (in particular, minimum mark-ups or price floors/minimum prices) may be seen as a complementary approach to ensuring a high price level and discouraging consumption of tobacco products. So far, these policies have not proven to increase average prices. A price floor is likely to lead to increased industry profits, giving the industry greater funds for its marketing strategies (such as the introduction of new products), and lower tax revenues for governments. By reducing price competition, a price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifications). However, where powerful multinationals are operating in certain markets with presence in all market segments and with the capability to overshift a tax on some brands while undershifting the tax on others, or where price promotions cannot be banned, minimum price policies may help increase the effectiveness of tax increases. Other non-tax policies affecting price levels are those relating to promotional dis- counts for tobacco products and the sale of single sticks of cigarettes. Both should be completely banned. The ban of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Higher taxes are the most effective way to dissuade consumption, with the added benefit of raising money for the government – money that can be earmarked for health and education programs, rather than going as profits to the tobacco industry. Additionally, in order to make excise tax on tobacco products more effective in reduc- ing overall tobacco use and in line with the recommendation of the Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce 78 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N the demand for tobacco), all tobacco products need to be taxed in a comparable way; the focus should not be on cigarettes only. Tax choices and reforms have various and sometimes conflicting consequences for the market. For example, there might be a trade-off between quantity and variety or perceived quality implications. It is important for the government to recognize that firms respond strategically to changes in tax policy. Close monitoring of the market is necessary to form correct expectations about industry responses and enable estimates of the impact of a tax increase on consumption and tax revenue. To estimate the total effect of a tax increase on demand for tobacco products and tax revenue, it is important to use correct estimates of the own-price elasticity of demand, the cross-price elasticity and the income elasticity of demand. It is also important to use updated estimates of demand elasticities, as the environment within which consumers make decisions continues to change. For example, financial crises or successful tobacco control interventions can be expected to shift demand and change elasticity. Another key measure of the impact of tax policy is the tax base elasticity. Policy- makers need to be mindful of the three key components of tax base elasticity: (1) the price elasticity of demand of tobacco, (2) the share of the tax in the consumer price and (3) the degree of pass-through of the excise tax rate increase to consumer price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. It is important to acknowledge that if tax increases lead to increases in prices be- low concurrent increases in income levels, they will not be effective in reducing consumption, as tobacco remains a normal good in most countries. Policy-makers need to account for affordability when considering tax increases. They should ensure that tax increases are high enough to increase prices above income growth so that consumption goes down effectively. When designing tax policy and deciding on the right level to impose, policy-makers need to assess and project the impact of their policy decisions. Monitoring and evaluation are important. Tools for measuring impact can be very helpful, and several such tools exist. The WHO ISPT, for example, looks not only at the impact of tax policy but also at a set of tobacco control policies, and this enables national policy-makers and other tobacco control experts to explore the potential impact of CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 79 future tobacco control policies. The tool uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. More specific to tobacco tax policy, the WHO TaXSiM assesses the impact of any excise tax increase and change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the effectiveness of their policies and whether those policies have an impact on tobacco use over time. The implementation of the MPOWER package is one useful indicator for assessing tobacco control overall. Tobacco taxation works best if it is implemented as part of a comprehensive MPOWER package. The tax share in the retail price of a selected tobacco product is one indicator of the effectiveness of tax policy, but a more important one is affordability, that is, whether tax increases do lead to price increases that are above income and general price increases. A useful indicator to assess the performance of the tax policy overall is the Tobacconomics Cigarette Tax Scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). Domestic policies in agriculture, industry, trade, finance and labour all have the potential to create or support incentives at different stages in tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. Greater domestic policy coherence should be pro- moted across different sectors of the government to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. Differential tax structures and rates have the potential to distort the functioning of the internal market. Harmonization of tobacco taxation ensures the establishment and proper functioning of a single market; prevents tax revenue erosion, tax avoidance and tax evasion; and protects people’s health. In this context, tax competition, where countries simply undercut each other’s tax rate, might prevent governments from achieving their tobacco control objectives and raising sufficient funds to pursue public health policies. To avoid such a race to the bottom, countries can establish minimum tax rates on all tobacco products. A common high minimum specific excise tax is the best approach to ensure that taxes and prices are above a minimal level. 80 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N In recent years, the world has experienced the rise of new and emerging tobacco and nicotine products, including ENDS/ENNDS and HTPs, which the industry claims are safer than traditional tobacco products. The evidence so far suggests that these products could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of these products as well as initiation, cessation and switching of tobacco use behaviours among different socioeconomic groups, are not yet clear. Until more evidence for the claimed benefits of these tobacco products is available, caution should be taken in developing tax policy. Therefore, the current recommendation is for HTPs to be taxed at the same level as cigarettes on a per-unit basis regardless of tobacco content. In countries where they are not banned, ENDS/ENNDS products must be regulated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS products taxation. Nicotine- and non-nicotine-containing e-liquids should be taxed equally. If ad valorem excise is chosen as the structure, the base should be applied on the retail price. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While the evolution of new and emerging tobacco and nicotine delivery systems merits the attention of administrators and regulators, it is worth remembering that the bulk of the world’s tobacco consumption is still in conventional cigarettes. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 81 REFERENCES 1. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 2. Outcome of proceedings: Council conclusions concerning the structure and rates of excise duty applied to manufactured tobacco. Brussels: Council of the European Union; 2 June 2020 (8483/20; https://www.consilium.europa.eu/media/44235/st08483-en20.pdf, accessed 28 September 2020). 3. Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco: Article 7, paragraph 2. Brussels: OJEU. 2011;176:24–36 (https://eur-lex.europa. eu/LexUriServ/LexUriServ.do?uri=OJ:L:2011:176:0024:0036:EN:PDF, accessed 28 September 2020). 4. U.S. National Cancer Institute and World Health Organization.The economics of tobacco and tobacco control. NIH Publication No. 16-CA-8029A. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute; 2016 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/brp/tcrb/monographs/monograph-21, accessed 28 September 2020). 5. Drope J, Schluger N, Cahn Z, Drope J, Hamill S, Islami F, et al. The tobacco atlas. Atlanta: American Cancer Society and Vital Strategies; 2018 (https://tobaccoatlas.org/wp-content/uploads/2018/03/ TobaccoAtlas_6thEdition_LoRes_Rev0318.pdf, accessed 24 January 2021). 6. Sector reports, cigarettes. In: GlobalData [website]. London: John Carpenter House; 2019 (restricted access). 7. Vellios N, Ross H, Perucic AM. 2018 trends in cigarette demand and supply in Africa. PLOS ONE. 2018;13(8):e0202467 (https://doi.org/10.1371/journal.pone.0202467, accessed 28 September 2020). 8. Keen MJ. The balance between specific and ad valorem taxation. Fiscal Studies. 1998;19(1):1–37 (https://www.ifs.org.uk/fs/articles/fsmick1.pdf, accessed 24 January 2021). 9. Kay JA, Keen MJ. The structure of tobacco taxes in the European community. IFS report series. Institute for Fiscal Studies; 1982. 10. Kay JA, Keen MJ. How should commodities be taxed?: Market structure, product heterogeneity and the optimal structure of commodity taxes. Euro Econ Rev. 1983;23(3):339–58. 11. Kay JA, Keen MJ. Commodity taxation for maximum revenue. Public Finan Q. 1987;15(4):371–85. 12. Kay JA, Keen MJ. Product quality under specific and ad valorem taxation. Public Finan Q. 1991;19(2):238–47. 13. Delipalla S, Keen MJ. The comparison between ad valorem and specific taxation under imperfect competition. J Public Econ. 1992;49(3):351–67. 14. Suits DB, Musgrave RA. Ad valorem and unit taxes compared. Q J Econ. 1953;67(4):598–604. 15. Skeath SE, Trandel GA. A pareto comparison of ad valorem and unit taxes in noncompetitive environments. J Public Econ. 1994;53(1):53–71. 16. Myles G. Imperfect competition and the optimal combination of ad valorem and specific taxation. Int Tax Public Finance. 1994;3:29–44. 17. Effectiveness of tax and price policies for tobacco control. Geneva: WHO Press, World Health Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications. iarc.fr/Book-And-Report-Series/Iarc-Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And- Price-Policies-For-Tobacco-Control-2011, accessed 28 January 2021) 18. Bevan J. GCC taxation. Aylesbury (UK): Rothmans Exports Limited; 1992 (https://www.industrydocuments. ucsf.edu/docs/yxpy0198, accessed 29 September 2020). 19. Morton IT. Note from IT Morton to I Birks regarding agreement on specific taxation; 8 February 1992 (https://www.industrydocuments.ucsf.edu/docs/#id=nxpy0198, accessed 29 September 2020). 20. Galgut R. Oman. BAT (UK and Export) Limited; 16 April 1992 (https://www.industrydocuments.ucsf. edu/docs/xxpy0198, accessed 29 September 2020). 21. Common excise tax agreement of the states of the Gulf Cooperation Council [website]. Abu Dhabi: UAE Ministry of Finance; 2016 (https://www.mof.gov.ae/En/lawsAndPolitics/govLaws/Documents/ Common%20Excise%20Tax%20Agreement.pdf, accessed 29 September 2020). 82 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 22. Brock B, Choi K, Boyle RG, Moilanen M, Schillo B. Tobacco product prices before and after a statewide tobacco tax increase. Tob Control. 2016;25(2):166–73. 23. Hanson A, Sullivan R. The incidence of tobacco taxation: evidence from geographic micro-level data. Natl Tax J. 2009;62(4):677–98. 24. Delipalla S, O’Donnell O. Estimating tax incidence, market power and market conduct: the European cigarette industry. Int J Ind Organ. 2001;19(6):885–908. 25. Keeler TE, Hu TW, Barnett PG, Manning WG, Hai-Yen S. Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxation. J Health Econ. 1996;15:499–512. 26. Linegar DJ, van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2017;0:1–7 (https://tobaccocontrol.bmj.com/content/27/1/65, accessed17 February 2021). 27. Villanti AC, Collins LK, Niaura RS, Gagosian SY, Abrams DB. Menthol cigarettes and the public health standard: a systematic review. Public Health. 2017;17:983 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5747135/, accessed 24 January 2021). 28. Directive 2014/40/EU of the European Parliament and of the Council of 3 April 2014 on the approximation of the laws, regulations and administrative provisions of the member states concerning the manufacture, presentation and sale of tobacco and related products and repealing Directive 2001/37/EC: Article 7. Brussels: OJEU. 2014;57:1–38 (https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=OJ:L:2014:127:FULL&from=EN, accessed 24 January 2021). 29. Delipalla S, Keen MJ. Product quality and the optimal structure of commodity taxes. J Public Econ Theory. 2006;8:547–54. 30. Obara T, Tsugawa S. The welfare comparison of ad-valorem tax and specific tax with both quality and quantity choice of a consumer. Theor Econ Lett. 2017;7:1797–1813 (https://www.scirp.org/journal/ paperinformation.aspx?paperid=79862, accessed 24 January 2021). 31. Anderson S, de Palma A, Kreider B. Tax incidence in differentiated product oligopoly. Journal of Public Econ. 2001;81(2):173–92. 32. Razzolini L, Shughart WF, Tollison RD. On the third law of demand. Econ Inq. 2003;41:2. 33. Sobel RS, Garrett TA. Taxation and product quality: new evidence from generic cigarettes. J Polit Econ. 1997;105:880–7 (http://faculty.citadel.edu/sobel/All%20Pubs%20PDF/Generic%20Cigarettes. pdf, accessed 24 January 2021). 34. Nargis N, Fong GT, Chaloupka FJ, Li Q. The choice of discount brand cigarettes: a comparative analysis of International Tobacco Control (ITC) surveys in Canada and the United States (2002-2005). Tob Control. 2014;23:i86–i96 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/suppl_1/i86. full.pdf, accessed 24 January 2021). 35. Chaloupka FJ, Peck R, Tauras JA, Xu X, Yurekli A. Cigarette excise taxation: the impact of tax structure on prices, revenues and cigarette smoking. Cambridge: National Bureau of Economic Research; 2010 (NBER Working Paper 16287; https://www.nber.org/system/files/working_papers/w16287/w16287. pdf, accessed 24 January 2021). 36. Cigarettes in India, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 37. Gilmore AB, Tavakoly B, Taylor G, Reed H. Understanding tobacco industry pricing strategy and whether it undermines tobacco tax policy: the example of the UK cigarette market. Addiction. 2013;108(7):1317–26 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3746125/, accessed 24 January 2021). 38. Marsh L, Cameron C, Quigg R, Hoek J, Doscher C, McGree R, et al. The impact of an increase in excise tax on the retail price of tobacco in New Zealand. Tob Control. 2016;25(4):458–634. 39. Brock B, Schillo BA, Moilanen M. Tobacco industry marketing: an analysis of direct mail coupons and giveaways. Tob Control. 2015;24:505–8. 40. Chaloupka FJ, Warner KE. The economics of smoking. Handbook Health Econ. 2000;1:1539–1627. 41. Shang C, Chaloupka FJ, Fong G, Thompson M, O’Connor RJ. The association between tax structure and cigarette price variability: findings from the ITC Project. Tob Control. 2015;24:iii88–iii93 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/24/Suppl_3/iii88.full.pdf, accessed 24 January 2021) 42. Li Q, Hyland A, Fong GT, Jiang Y, Elton-Marshall T. Use of less expensive cigarettes in six cities in China: findings from the International Tobacco Control (ITC) China Survey. Tob Control. 2010;19(suppl CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 83 2):i63–i68 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/19/Suppl_2/i63.full.pdf, accessed 24 January 2021). 43. Hyland A, Laux FL, Higbee C, Hastings G, Ross H, Chaloupka FJ, et al. Cigarette purchase patterns in four countries and the relationship with cessation: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15(suppl III):iii59–iii64 (https://tobaccocontrol.bmj.com/ content/tobaccocontrol/15/suppl_3/iii59.full.pdf, accessed 24 January 2021). 44. Tsai YW, Yang CL, Chen CS, Liu, TC, Chen PF. The effect of Taiwan’s tax induced increases in cigarette prices and brand switching and the consumption of cigarettes. Health Econ. 2005;14:627–41. 45. Cummings KM, Hyland A, Lewit E, Shopland D. Use of discount cigarettes by smokers in 20 communities in the United States, 1988-1993. Tob Control. 1997;6(suppl 2):S25–S30 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/6/suppl_2/S25.full.pdf, accessed 24 January 2021). 46. Ross H, Tesche J. Undermining government tax policies: common strategies employed by the tobacco industry in response to increases in tobacco taxes. Economics of Tobacco Control Project, School of Economics, University of Cape Town and Tobacconomics, Health Policy Center, Institute for Health Research and Policy. Chicago: University of Illinois at Chicago; 2016 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC5747349/, accessed 26 January 2021). 47. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 48. Hiscock R, Branston JR, McNeill A, Hitchman SC, Partos TR, Gilmore AB. Tobacco industry strategies undermine government tax policy: evidence from commercial data. Tob Control. 2017;0:1–10 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 49. Filippidis F, Laverty A, Hone T, Been JV, Millett C. Association of cigarette price differentials with infant mortality in 23 European Union countries. JAMA Pediatr. 2017;171(11):1100–1106 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5710372/#:~:text=Cigarette%20price%20increases%20across%20 23,priced%20cigarettes%20during%20this%20period, accessed 26 January 2021). 50. López-Nicolás A, Belén Cobacho M, Fernández E. The Spanish tobacco tax loopholes and their consequences. Tob Control. 2013;22(E1):e21. 51. Ballester L, Auchincloss A, Robinson L, Mayne S. Exploring impacts of taxes and hospitality bans on cigarette prices and smoking prevalence using a large dataset of cigarette prices at stores 2001–2011, USA. Int J Environ Res Public Health. 2017;14(3):e318 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5369154/, accessed 24 January 2021). 52. Liber A, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia survey. Tob Control. 2015;24:iii83-iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed on 26 January 2021). 53. Feighery E, Ribisl K, Schleicher N, Zellers L, Wellington N. How do minimum cigarette price laws affect cigarette prices at the retail level? Tob Control. 2005;14:80–85 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/14/2/80.full.pdf, accessed 26 January 2021). 54. Tynan MA, Ribisi KM, Loomis BR. Impact of cigarette minimum price laws on the retail price of cigarettes in the USA. Tob Control. 2012;22:e1. 55. Huang J, Chriqui JF, DeLong H, Mirza M, Diaz MC, Chaloupka FJ. Do state minimum markup/price laws work? Evidence from retail scanner data and TUS-CPS. Tob Control. 2016;25:i52–i59 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i52.full.pdf, accessed 26 Jan 2021). 56. Whitehead R, Brown L, Riches E, Rennick L, Armour G, McAteer J, et al. Strengths and limitations of tobacco taxation and pricing strategies. Edinburgh: NHS Health Scotland; 2018 (http://www. healthscotland.scot/media/1829/rapid-evidence-review-strengths-and-limitations-of-tobacco- taxation-and-pricing-strategies.pdf, accessed 27 January 2021). 57. Golden S, Farrelly M, Luke D, Ribisl K. Comparing projected impacts of cigarette floor price and excise tax policies on socioeconomic disparities in smoking. Tob Control. 2016;25:i60-i66 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i60.full.pdf, accessed 27 January 2021). 58. Doogan NJ, Wewers ME, Berman M. The impact of a federal cigarette minimum pack price policy on cigarette use in the USA. Tob Control. 2018;27(2):203–08. 84 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 59. Marynak K, Xu X, Wang X, Holmes CB, Tynan MA, Pechacek T. Estimating the impact of raising prices and eliminating discounts on cigarette smoking prevalence in the United States. Public Health Rep. 2016;131:536–43 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4937114/#:~:text=We%20found%20 that%20each%20price,smokers%20aged%20%E2%89%A526%20years., accessed 27 January 2021). 60. McLaughlin I, Pearson A, Laird-Metke E, Ribisl K. Reducing tobacco use and access through strengthened minimum price laws. Am J Public Health. 2014;104(10):1844–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4937114/#:~:text=We%20found%20that%20each%20price,smokers%20aged%20 %E2%89%A526%20years., accessed 27 January 2021). 61. European Commission v. French Republic, C-197/08, European Court of Justice 4 March 2019 (https:// eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A62008CC0197, accessed on 27 January 2021). 62. Tobacco price promotion: policy responses to industry price manipulation. Boston: Public Health and Tobacco Policy Center; October 2011 (http://tobaccopolicycenter.org/documents/Price%20 Promotion%20Overview%20FINAL.pdf, accessed 28 September 2020). 63. Golden SD, Holt Smith M, Feighery EC, Roeseler A, Rogers T, Ribisl KM. Beyond excise taxes: a systematic review of literature on non-tax policy approaches to raising tobacco product prices. Tob Control. 2016;25:377–85. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/4/377.full.pdf, accessed 27 January 2021). 64. Branston J, Gilmore S. The case for Ofsmoke: the potential for price cap regulation of tobacco to raise ₤500 million per year in the UK. Tob Control. 2014;23(1):45–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3812939/, accesed 27 January 2021). 65. von Lampe K, Kurti M, Johnson J. “I’m gonna get me a loosie” understanding single cigarette purchases by adult smokers in a disadvantaged section of New York City. Prev Med Rep. 2018;12:182–5. (https:// www.sciencedirect.com/science/article/pii/S2211335518302195, accessed 27 January 2021). 66. de Ojeda A, Barnoya J, Thrasher JF. Availability and costs of single cigarettes in Guatemala. Nicotine Tob Res. 2013;15(1):83–7 . 67. Lal P, Kumar R, Ray S, Sharma N, Bhattarcharya B, Mishra D, et al. The single cigarette economy in India - a back of the envelope survey to estimate its magnitude. Asian Pac J Cancer Prev. 2015;16(13):5579-82 (http://journal.waocp.org/article_31295_9ebd3981652934bbf5d78ec066b6a39d.pdf, accessed 27 January 2021). 68. Coady MH, Chan CA, Sacks R, Mbamalu IG, Kansagra SM. The impact of tax on purchasing behaviors among New York City smokers. Am J Public Health. 2013;103(6):e54-e60 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC3698745/, accessed 27 January 2021). 69. Cigarettes in Thailand, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 70. Cigars in Germany, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 71. Report from the Commission to the Council on Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2018 (COM(2018) 17 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/report_excise_duty_manufactured_ tobacco_12012018_en.pdf, accessed 28 September 2020). 72. Gallet CA, List JA. Cigarette demand: a meta-analysis of elasticities. Health Econ. 2003;12(10):821–35. 73. Guidelines for implementation of article 6 of the WHO FCTC. Geneva: World Health Organization; 2016 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 29 September 2020). 74. DeCicca P, Kenkel D, Liu F. Excise tax avoidance: the case of state cigarette taxes. J Health Econ. 2013;32(6):1130–41 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3855906/, accessed 27 January 2021). 75. Jawad M, Lee JT, Glantz S, Millett C. Price elasticity of demand of non-cigarette tobacco products: a systematic review and meta-analysis. Tob Control; 2018;27(6):689–95 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6056338/, accessed 27 January 2021). 76. John RM. Price elasticity estimates for tobacco products in India. Health Policy Plan. 2008;23(3):200–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2800993/, accessed on 27 January 2021). 77. Wasserman J, Manning WG, Newhouse JP, Winkler JD. The effects of excise taxes and regulations on CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 85 cigarette smoking. Journal of Health Econ, Elsevier. 1991;10(1):43–64. 78. Cheng K, Kenkel D. U.S. cigarette demand: 1944-2004. B.E. Journal of Econ Analysis & Policy. 2010;10(1):10.2202/1935-1682.2438 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3820165/, accessed 28 January 2021). 79. Bobak M, Jha P, Nguyen S, Jarvis M. Poverty and smoking. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000. 80. Peck, RM. Equity issues, tobacco, and the poor. Yurekli A, de Beyer J, editors. Washington (DC): World Bank Group; 2013. (Economics of Tobacco Toolkit, Tool 6; https://openknowledge.worldbank.org/ handle/10986/16265, accessed 28 January 2021). 81. Kenkel DS, Schmeiser MD, Urban C. Is smoking inferior?: evidence from variation in the earned income tax credit. J Hum Res. 2014;49:1094–1120. 82. Ruhm, CJ. Healthy living in hard times. Journal of Health Econ. 2005;24(2):341–63. 83. Ruhm, CJ. Are recessions good for health? The Quarterly Journal of Econ. 2000;115(2):617-50. 84. Tarantilis F, Athanasakis K, Zavras D, Vozikis A, Kyriopoulos I. Estimates of price and income elasticity in Greece. Greek debt crisis transforming cigarettes into a luxury good: an econometric approach. BMJ Open. 2015;5(1):e004748. (https://bmjopen.bmj.com/content/bmjopen/5/1/e004748.full.pdf, accessed 28 January 2021). 85. Kaiser M, Reutter M, Sousa-Poza A, Strohmayer K. Smoking and the business cycle: evidence from Germany. Discussion Paper Series. Bonn, Germany: IZA Institute of Labour Economics; 2017 (http:// ftp.iza.org/dp10953.pdf, accessed 28 September 2020). 86. Tingum EN, Mukong AK, Mdege N. The effects of price and non-price policies on cigarette consumption in South Africa. Tob Induc Dis. 2020;18(July):62 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC7398599/, accessed 28 January 2021). 87. Vardavas CI, Filippidis F, Agak, I, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the importance of earmarking revenue to health care and tobacco control. Tob Induc Dis. 2012;10(1):21 (http://www. tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the-revenue-to- health-care-and-tobacco,66110,0,2.html, accessed 28 January 2021). 88. Marquez P, Catinkaya V. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 89. Blecher EH, van Walbeek CP. An international analysis of cigarette affordability. Tob Control. 2004;13:339– 46 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/13/4/339.full.pdf, accessed 28 January 2021). 90. Goodchild M, Sinha P, Munish VG, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/handle/10665/331836, accessed 28 January 2021). 91. John RM, Dauchy E. Trends in affordability of tobacco products before and after the transition to GST in India. Tob Control Online First. 20 March 2020. 92. Chaloupka FJ, Drope J, Siu E, Vulovic V, Stoklosa M, Mirza M, et al. Tobacconomics cigarette tax scorecard. Chicago: Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/636/uic-tobacco-scorecard-report-final.pdf, accessed 28 January 2021). 93. Kanbur R, Keen MJ. Jeux sans frontieres: tax competition and tax coordination when countries differ in size. Am Econ Rev. 1993;83(4):877–92 (https://www.researchgate.net/publication/4901220_Jeux_ Sans_Frontieres_Tax_Competition_and_Tax_Coordination_When_Countries_Differ_in_Size, accessed 28 January 2021). 94. Bouw A. Tobacco taxation in the European Union: an overview (English). Washington (DC): World Bank Group; 2017 (http://documents.worldbank.org/curated/en/493581492415549898/Tobacco- taxation-in-the-European-Union-an-overview, accessed on 28 September 2020). 95. Blecher E, Drope J. The rewards, risks and challenges of regional tobacco tax harmonization. Tob Control. 2014;0:1–5. 96. Directive No.n° 01/2017/CM/UEMOA of 22 December 2017. Paris: Droit-Afrique; 2017 (https://www. 86 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N droit-afrique.com/uploads/UEMOA-Directive-2017-01-droits-accise-tabacs.pdf, accessed 29 September 2020). 97. Mirzoev T, Zhu L, YangY, Pescatori A, Matsumoto A, Callen T. The future of oil and fiscal sustainability in the GCC region. Middle East and Central Asia Research Department. Washington (DC): International Monetary Fund; 2020 (https://www.imf.org/-/media/Files/Publications/DP/2020/English/FOFSGCCEA. ashx, accessed 28 January 2021). 98. State cigarette excise tax rates & rankings. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/factsheets/0097.pdf, accessed 28 September 2020). 99. A map and table comparing cigarette prices in Canada (March 2018). Toronto: Non-Smokers’ Rights Association/Smoking and Health Action Foundation; 2018 (https://nsra-adnf.ca/wp-content/ uploads/2018/03/180302_map_and_table.pdf, accessed 4 February 2021). 100. European Commission excise duty rates on manufactured tobacco. Brussels: European Commission; 31 December 2017 (https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco- energy/excise-duties-tobacco_en, accessed 1 July 2018). 101. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2019 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study-on-the-tobacco-taxation-directive-2019_en.pdf, accessed 28 September 2020). 102. Liber AC. Heated tobacco products and combusted cigarettes: comparing global prices and taxes. Tob Control; 2018;28(6). 103. Heated tobacco products: information sheet - 2nd edition. Geneva: World Health Organization; 2020 (https://www.who.int/publications/i/item/WHO-HEP-HPR-2020.2, accessed 28 January 2021). 104. Kass Lempert L, Glantz S. Analysis of FDA’s IQOS marketing authorisation and its policy impacts. Tob Control 2020;0:1–9. 105. Chivers E, Janka M, Franklin P, Mullins M, Larcombe A. Nicotine and other potentially harmful compounds in “nicotine-free” e-cigarette liquids in Australia. Med J Aust. 2019;210(3):118–9 (https://onlinelibrary. wiley.com/doi/epdf/10.5694/mja2.12059, accessed 28 September 2020). 106. Wenzl T, Zelinkova Z, Regueiro J, Giri A. Scientific-technical support activities to DG TAXUD-C-2 on the option to include e-cigarettes within the scope of excisable goods for the Impact Assessment on a possible revision of Directive 2011/64/EU. Administrative Arrangement TAXUD/2015/DE/329. Luxembourg: Publications Office of the European Union; 2017 (https://publications.jrc.ec.europa.eu/ repository/bitstream/JRC105828/kjna29062enn.pdf, accessed 28 September 2020). 107. Lisko JG, Tran H, Stanfill SB, Blount BC, Watson CH. Chemical composition and evaluation of nicotine, tobacco alkaloids, pH, and selected flavors in e-cigarette cartridges and refill solutions. Nicotine Tob Res. 2015;17(10):1270–78. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573955/, accessed 28 January 2021). 108. Ghebreyesus TA. Progress in beating the tobacco epidemic. Lancet. 17 August 2019;394:548–49 (https:// www.thelancet.com/pb-assets/Lancet/pdfs/S0140673619317301.pdf?hss_channel=tw-27013292, accessed 29 September 2020). 109. Smoking cessation, a report of the Surgeon General. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2020. (https://www.hhs.gov/sites/default/ files/2020-cessation-sgr-executive-summary.pdf, accessed 8 October 2020). 110. E-cigarettes are harmful to health. Geneva: World Health Organization. 5 February 2020 (https://www. who.int/news-room/detail/05-02-2020-e-cigarettes-are-harmful-to-health, accessed 29 September 2020). 111. Shan L, Manzione LC, Azagba S. Psychological well-being and dual-use of cigarettes and e-cigarettes among high school students in Canada. J Affect Disord. 2020;265:357–63 112. King BA, Gammon DG, Marynak KL, Rogers T. Electronic cigarettes sales in the United States, 2013- 2017. JAMA. 2018;320(13):1379-80 (https://jamanetwork.com/journals/jama/fullarticle/2705175, accessed 28 January 2021). 113. LaVito A. Popular e-cigarette Juul’s sales have surged almost 800 percent over the past year [website]. CNBC. 2 July 2018 (https://www.cnbc.com/2018/07/02/juul-e-cigarette-sales-have-surged-over-the- past-year.html, accessed 28 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 87 114. Statement from FDA Commissioner Scott Gottlieb, M.D., on new enforcement actions and a youth tobacco prevention plan to stop youth use of, and access to, JUUL and other e-cigarettes. Washington (DC): US Food and Drug Administration; 23 April 2018 (www.fda.gov/NewsEvents/Newsroom/ PressAnnouncements/ucm605432.htm, accessed 29 September 2020). 115. Paul K. San Francisco becomes first US city to ban sale of e-cigarettes. The Guardian. 25 June 2019 (https://www.theguardian.com/us-news/2019/jun/25/san-francisco-e-cigarettes-vaping-juul-ban, accessed 29 September 2020). 116. Pesko MF, Courtemanche CJ, Maclean JC. The effects of traditional cigarette and e-cigarette tax rates on adult tobacco product use. J Risk Uncertain. 2020;60:229–258. 117. Cotti CD, Courtemanche CJ, Maclean JC, Nesson ET, Pesko MJ, Tefft N. The effects of e-cigarettes taxes on e-cigarettes prices and tobacco product sales: evidence from retail panel data. Cambridge: National Bureau of Economic Research; 2020 (NBER Working Paper 26724; https://www.nber.org/ system/files/working_papers/w26724/w26724.pdf, accessed 28 January 2021). 118. Saffer H, Dench DL, Grossman M, Dave DM. E-cigarettes and adult smoking: evidence from Minnesota. Cambridge: National Bureau of Economic Research; 2019 (NBER Working Paper 26589; https://www. nber.org/system/files/working_papers/w26589/w26589.pdf, accessed 28 January 2021). 119. Huang J, Gwarnicki C, Xu X, Caraballo RS, Wada R, Chaloupka FJ. A comprehensive examination of own- and cross-price elasticities of tobacco and nicotine replacement products in the U.S. Prev Med. 2018;117:107–14 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6195827/, accessed 28 September 2020). 120. Zheng Y, Zhen C, Dench D, Nonnemaker JM. U.S. demand for tobacco products in a system framework. Health Econ. 2017;26:1067–86. 121. Stoklosa M, Drope JM, Chaloupka FJ. Prices and e-cigarette demand: evidence from the European Union. Nicotine Tob Res. 2016;18(10):1973–80. 122. Grace RC, Kivell BM, Laugesen M. Estimating cross-price elasticity of e-cigarettes using a simulated demand procedure. Nicotine Tob Res. 2015;17(5):592–8. (https://www.researchgate.net/ publication/270291421_Estimating_Cross-Price_Elasticity_of_E-Cigarettes_Using_a_Simulated_ Demand_Procedure, accessed 28 January 2021). 123. Huang J, Tauras J, Chaloupka FJ. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob Control. 2014;23:iii41–iii47 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/23/suppl_3/iii41.full.pdf, accessed 17 February 2021). 124. Cheng KW, Chaloupka FJ, Shang C, Ngo A, Fong GF, Borland R, et al. Prices, use restrictions and electronic cigarette use—evidence from wave 1 (2016) US data of the ITC Four Country Smoking and Vaping Survey. Addiction. 2019;114(suppl 1):115–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC6661221/, accessed 28 January 2021). 125. Liber AC, Drope JM, Stoklosa M. Combustible cigarettes cost less to use than e-cigarettes: global evidence and tax policy implications. Tob Control. 2016;0:1–6 (https://www.researchgate.net/ publication/299473181_Combustible_cigarettes_cost_less_to_use_than_e-cigarettes_Global_ evidence_and_tax_policy_implications, accessed 28 January 2021). 126. Chaloupka FJ, Sweanor D, Warner KE. Differential taxes for differential risks — toward reduced harm from nicotine yielding products. N Engl Journal Med. 2016;373(7):594-7. 127. Goniewicz ML, Knysak J, Gawron M, Kosmider L, Sobczak A, Kurek J, et al. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob Control. 2014;23:133–9. 128. Sweet L, Brasky TM, Cooper S, Doogan N, Hinton A, Klein EG, et al. Quitting behaviors among dual cigarette and e-cigarette users and cigarette smokers enrolled in the tobacco user adult cohort. Nicotine Tob Res. 2019;21(3):278–84 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6379027/, accessed 28 September 2020). 129. Wills TA, Knight R, Williams RJ, Pagano I, Sargent JD. Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics. 2014;135(1):e43–e51 (https://www. ncbi.nlm.nih.gov/pmc/articles/PMC4279062/, accessed 28 January 2021). 130. Electronic Nicotine Delivery Systems: report by WHO. Geneva: World Health Organization; 2014. (FCTC/ COP/6/10 Rev.1; https://apps.who.int/gb/fctc/PDF/cop6/FCTC_COP6_10Rev1-en.pdf, accessed 28 January 2021) 88 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 131. WHO report on the global tobacco epidemic 2017: monitoring tobacco use and prevention policies. Geneva: World Health Organization; 2017 (https://www.who.int/tobacco/global_report/2017/en/, accessed 28 January 2021). 132. Regulation of the Minister of Finance of the Republic of Indonesia, number 146, PMK.010/2017 on excise tax rates of tobacco products. Jakarta: Ministry of Finance Indonesia; 2017. 133. Gourdet CK, Chriqui JF, Chaloupka FJ. A baseline understanding of state laws governing e-cigarettes. Tob Control. 2014;23:iii37-iii40 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4078672/, accessed 28 January 2021). 134. An act to ensure that nicotine products are taxed equally. 127th Maine Legislature: first regular session-2015. H.P. 670, Legislative Document No. 973. 2015 (https://www.mainelegislature.org/legis/ bills/getPDF.asp?paper=HP0670&item=1&snum=127&PID=, accessed 28 January 2021). 135. Evaluation of the Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco. Commission Staff Working Document. Brussels: European Commission; 2020 (SWD(2020) 32 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/10- 02-2020-tobacco-taxation-report.pdf, accessed 28 September 2020). 136. Electronic nicotine delivery systems and electronic non-nicotine delivery systems (ENDS/ENNDS): report by WHO. Geneva: World Health Organization; 2016. (FCTC/COP/7/11; https://www.who.int/ fctc/cop/cop7/FCTC_COP_7_11_EN.pdf, accessed 28 September 2020). 137. E-cigarette use among youth and young adults: a report of the Surgeon General. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2016 (https://e-cigarettes. surgeongeneral.gov/documents/2016_SGR_Full_Report_non-508.pdf, accessed 28 September 2020). 138. Wagener TL, Floyd EL, Stepanov I, Driskill LM, Frank SG, Meier E, et al. Have combustible cigarettes met their match? The nicotine delivery profiles and harmful constituent exposures of second-generation and third-generation electronic cigarette user. Tob Control 2017;26:e23–e28 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/26/e1/e23.full.pdf, accessed 28 September 2020). 139. Vapour products data. In: Euromonitor International [website]. London: Euromonitor International; 2019 (restricted access). 140. Global market data: segment insights. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 89 ANNEX 2.1 Table A2.1 Countries that apply different types of cigarette excise tax structures, 2018 SPECIFIC EXCISE (65 COUNTRIES) AD VALOREM EXCISE (42 COUNTRIES) MIXED EXCISE (63 COUNTRIES) NO EXCISE (15 COUNTRIES) Albania, Andorra, Australia, Azerbaijan, Barbados, Belarus, Belize, Bolivia (Plurinational State of ), Burundi, Canada, Cook Islands, Dominica, Ecuador, Eswatini, Fiji, Gambia, Honduras, Iceland, India, Indonesia, Jamaica, Japan, Jordan, Kazakhstan, Kenya, Kiribati, Kyrgyzstan, Lesotho, Malaysia, Mauritius, Mongolia, Mozambique, Myanmar, Namibia, Nepal, New Zealand, Nicaragua, Norway, Pakistan, Palau, Papua New Guinea, Peru, Philippines, Republic of Korea, Saint Lucia, Saint Vincent and the Grenadines, Samoa, Seychelles, Singapore, Solomon Islands, South Africa, Sri Lanka, Suriname, Tajikistan, Timor-Leste, Tonga, Trinidad and Tobago, Uganda, United Republic of Tanzania, USA, Uruguay, Uzbekistan, Vanuatu, Yemen, Zimbabwe Argentina, Armenia, Bahrain, Bangladesh, Benin, Burkina Faso, Cabo Verde, Cambodia, Cameroon, Chad, Comoros, Côte d’Ivoire, Cuba, Democratic Republic of the Congo, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Ghana, Grenada, Guatemala, Guinea-Bissau, Liberia, Madagascar, Mali, Mauritania, Niger, Panama, Paraguay, Saint Kitts and Nevis, Saudi Arabia, Senegal, Sierra Leone, Sudan, Syrian Arab Republic, Togo, Turkmenistan, Tuvalu, United Arab Emirates, Venezuela (Bolivarian Republic of ), Viet Nam, Zambia Algeria, Austria, Belgium, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Central African Republic, Chile, China, Colombia, Congo, Costa Rica, Croatia, Cyprus, Czechia, Denmark, Dominican Republic, Egypt, El Salvador, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iran (Islamic Republic of ), Ireland, Israel, Italy, Lao People’s Democratic Republic, Latvia, Lebanon, Lithuania, Luxembourg, Malta, Mexico, Montenegro, Morocco, Netherlands, Nigeria, North Macedonia, Poland, Portugal, Republic of Moldova, Romania, Russian Federation, Rwanda, Sao Tome and Principe, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Thailand, Tunisia, Turkey, Ukraine, United Kingdom, West Bank and Gaza Strip Afghanistan, Angola, Antigua and Barbuda, Democratic People’s Republic of Korea, Iraq, Kuwait, Libya, Maldives, Marshall Islands, Micronesia (Federated States of ), Nauru, Niue, Oman,a Qatar,a Somalia a This table shows the status of cigarette excise tax structures as of July 2018 and does not account for changes occurring after that date, in particular for the cases of Qatar and Oman, which introduced excise on tobacco in January 2019 and June 2019, respectively. Source: WHO RGTE. 90 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 2.2 THE ANALYTICS OF THE TAX BASE ELASTICITY Assume tax revenue R = tsQ or R = tvPQ , where Q is the quantity consumed, ts is the specific tax, tv is the ad valorem tax and P is the consumer price. The following equations can help to illustrate the different components of the tax base elasticity. Under a specific excise regime, change in revenue depends essentially on the change in consumption: where R is the tobacco tax revenue, is the specific excise tax and is the tobacco tax base elasticity. The tax base elasticity is made of: where ε, the price elasticity = , is the degree of pass-through of the specific excise tax rate increase on consumer price and is the tax-price ratio. Under an ad valorem excise regime, change in revenue depends essentially on the change in tobacco expenditure: where R is the tobacco tax revenue, tav is ad valorem excise tax and ηav is the tobacco tax base elasticity. The tax base elasticity here is made of: where is the degree of pass-through of the ad valorem excise tax rate increase on consumer price, is the tax-price ratio and ε the price elasticity = . ∂R = Q (1+ηs)∂ ts ηs = ε ts P ∂P ∂ts tav P ηav = (1 + ε) tav P ∂P ∂tav∂P ∂tav ∂P ∂ts ΔQ ΔP P Q ΔQ ΔP P Q ts P ∂R = P Q (1+ηav)∂tav CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 91 ANNEX 2.3 ELEMENTS OF THE DEVICES THAT MAKE UP ENDS/ENNDS PRODUCTS The main components of any ENDS/ENNDS kit include essentially: • USB charger (not a car charger) • Inbuilt battery Additionally, For open systems • Tanks (refillable containers) with removable atomizer (often sold bundled with atomizers) • Clearomizers/refillable pods (no removable atomizer) • E-liquid For closed systems • Disposable e-cigarettes: not rechargeable, thrown away after e-liquid is finished • Nondisposable e-cigarettes: – Pre-filled cartomizers (cartridges designed to go with the cigalike kit) – Pre-filled tank refills/pods (pods or cartridges designed to go with the prefilled tank/pod kits) Some definitions: • Atomizer: uses a heating element to vaporize the e-liquid • Cartomizer: combines the cartridge/tank and the atomizer • Clearomizer: same as cartomizer, uses different technology • Cartridge/tank/pod: container that includes the e-liquid In summary, ENDS/ENNDS product devices include the following: • USB charger (not a car charger) • Inbuilt battery • Disposable e-cigarettes • Atomizer • Cartomizer/clearomizer • Cartridge/tank/pod with or without atomizer • Pre-filled cartridge/tank/pod (for closed systems, includes e-liquid) • E-liquid (added in the cartridge/tank/pod in open systems) Source: ECigIntelligence, 2020. Information also obtained from vaping websites, including https://www. misthub.com/blogs/vape-tutorials/76788357-tutorial-atomizer-vs-cartomizer-vs-clearomizer, http:// www.bestclearomizer.com/clearomizer-vs-cartomizer-vs-atomizer/, https://wayofleaf.com/accessories/ vapes/atomizer-vs-clearomizer-vs-cartomizer, https://wayofleaf.com/accessories/vapes/atomizer-vs- clearomizer-vs-cartomizer, accessed 15 July 2020. 92 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 93 CHAPTER 3. Tobacco tax administration 3.1 INTRODUCTION Imposing excise taxes on tobacco products usually serves more than one purpose. Governments often find themselves balancing interests between financial and public health objectives. Both objectives can best be achieved by an efficient and effective competent authority with strong technical capacity to enforce and collect taxes. A competent authority is the agency, organization or department that is legally as- signed to complete a particular activity; in the case of administering tobacco taxes, the competent authority is often a tax administration, revenue authority, customs department or ministry of finance. Article 6 of the WHO FCTC (1), along with its guidelines (2), provides a solid foundation for sound tax administration. As stated under section 1.5 in the guidelines, tobacco tax systems should be efficient and effective. They should be structured to minimize the costs of compliance and administration, while ensuring that the desired level of tax revenue is raised and health objec- tives are achieved. Efficient and effective administration of tobacco tax systems enhances tax compliance and collection of tax revenue while reducing tax evasion and the risk of illicit trade. Efficiency in tax administration refers to minimizing the costs per unit of tax revenue collected. It is measured by comparing the resources used with the revenues gener- ated. Effectiveness in tax administration refers to a high level of compliance – also described as taxpayers meeting their obligations. Thus, an efficient and effective competent authority collects the tax at a minimum cost while ensuring conformity to the rules. Tobacco taxation is the single most effective tobacco control measure for re- ducing tobacco use and is best implemented as part of a comprehensive tobacco control plan (3). Illicit trade – including smuggling and illicit manufacturing – and tax avoidance undermine the effectiveness of tax policies and their objectives (4). The impact on illicit trade is often cited by opponents of tax increases, who argue that increasing taxes increases illicit trade. They contend that illicit trade can lead 94 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N to lower revenues for governments and lower prices. The challenge faced by the competent authorities is to ensure that due taxes are declared and collected on all tobacco products that are manufactured in and/or imported into its jurisdiction, while at the same time detecting tobacco products that are illegally manufactured in and/or imported into its jurisdiction, stopping such activity and prosecuting the responsible parties. This chapter describes the shared characteristics of good tax administrations, including best practices based on country experiences. It regularly refers to the WHO FCTC, and – given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade – draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (5). Any practice or conduct prohibited by law and related to production, shipment, receipt, possession, distribution, sale or purchase of tobacco products – including any practice or conduct intended to facilitate such activity – is considered as illicit trade (Article 1). The objective of the Protocol is to eliminate and prevent all forms of illicit trade in tobacco products. At the same time, the Protocol includes measures for tobacco tax administration based on international best practices, which makes it relevant for all countries, even those that are not Parties to it. The Protocol was adopted at the fifth session of the COP to the WHO FCTC in 2012 and entered into force on 25 September 2018. As indicated in the Preamble, it was developed in response to the increasing international illicit trade in tobacco products (5). The Protocol covers three main areas: (1) measures to control the supply chain (Part III); (2) measures dealing with offences, including sanctions (Part IV) and (3) international cooperation (Part V). Different provisions of the Protocol are discussed in detail throughout this chapter, and section 3.4 is devoted specifically to control and enforcement. 3.2 INSTITUTIONAL ARRANGEMENTS Competent authorities that collect taxes effectively in an efficient way share a number of attributes. The organizational structures of these authorities contain clearly defined roles, responsibilities and rules for coordination among relevant bodies. Moreover, competent authorities collect data regularly and manage information needed for assessing risks. The key to successful risk management is to share this information among relevant authorities both within a country and between countries. Effective and efficient competent authorities also regularly evaluate their performance and accountability according to key performance indicators to identify areas for improve- ment. These characteristics are discussed in greater detail in the following sections. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 95 3.2.1. CLEARLY DEFINED ROLES AND RESPONSIBILITIES OF COMPETENT AUTHORITIES The designation of competent authorities for the implementation and enforcement of tax laws – including clear definitions of the boundaries of authority among numerous agencies within a country – is essential for efficient collection of taxes. Areas where different agencies need to cooperate and share data must also be defined. Overlap of activities by different authorities leads to inefficient use of resources, whereas gaps create opportunities for fraud, leading to ineffective tax laws. The importance of clearly defined roles and responsibilities applies not only to tax authorities and customs but also to law enforcement agencies, including police and border control forces. The implementation and enforcement of taxation is organized differently in various countries. The most common structure separates customs and tax administration. The trend since the 1990s, however, has been to combine these functions into one agency, such as Her Majesty’s Revenue and Customs (HMRC) in the United King- dom, SUNAT in Peru1 and AFIP in Argentina.2 Several countries have increased coordination between tax and customs by creating a revenue secretariat and also implementing systems to share tax records as a single taxpayer account. Coordina- tion between tax policy and tax administration authorities has also increased. One can think of combining both into one department within the ministry of finance or ensure that tax administration authorities are consulted during the tax policy process. Some tasks, such as licensing, may be handled by other ministries such as the ministries of health, agriculture or trade. For example, the Ministry of Health of Brunei and the Health Science Authority of Singapore are responsible for the licensing of importers of tobacco products (6). In some federal countries, including Colombia and the United States, excise taxes – including tobacco taxes – are collected and enforced by local or state tax administrations. Other countries have organized the administration of national taxes by establishing a single unified revenue body. Particularly in larger economies, that body is often responsible for both direct and indirect taxes, including excise taxes, and reports to the ministry of finance. All the functions needed for effective and efficient tax administration are established within these bodies (7). No matter what the institutional arrangements may be, it is vital that the agencies cooperate and exchange information and that their competencies find their basis in law. More information on this topic is provided in section 3.2.2. 1 Law Decreto Supremo 061-2002-PCM - Disponen fusión por absorción de la Superintendencia Na- cional de Administración Tributaria – SUNAT con la Superintendencia Nacional de Aduanas - Aduanas [Supreme decree year 2002 about the merger between Tax and Customs Administration]. Lima: El Peruano, 12, July 2002 (in Spanish) (http://www.sunat.gob.pe/legislacion/sunat/ds061-2002-PCM.pdf, accessed 13 November 2020). 2 Administracion Federal de Ingresos Publicos, Decreto 618/1997 [Federal Administration of Public Revenue, Decree 618] (in Spanish) (http://servicios.infoleg.gob.ar/infolegInternet/an- exos/40000-44999/44432/norma.htm, accessed 13 November 2020). 96 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Many countries, however, have separate bodies for the collection of taxes and customs duties. A 2015 survey of 135 tax administrations worldwide found that only 36% of them were responsible for both tax administration and customs ad- ministration (8). In most countries, customs authorities are more likely to collect excise duties on imports, and in many countries, VAT or sales tax is collected jointly with tobacco tax, particularly for imported products. This simplifies controls and creates synergy by unifying common processes and procedures, resulting in cost savings for tax administrations and taxpayers. The involvement of multiple bodies in tax collection requires especially good collaboration and information-sharing to ensure efficient and effective collection of taxes and duties. KEY TAKEAWAY 1 Institutional arrangements with clearly defined roles and responsibilities – designed to prevent overlaps and voids – contribute to effective and efficient tax administration. 3.2.2. EFFECTIVE COORDINATION AMONG RELEVANT BODIES Coordination at the national level Coordination among relevant bodies is key to effective tobacco tax administration. This means not only clearly defined roles and responsibilities, as described in the previous section, but also coordination among the competent authority, customs and those responsible for formulating, analysing and implementing tax policy. Regardless of the institutional arrangements – whether the responsible parties are all within the ministry of finance or in separate government agencies – all parties need to cooperate and exchange information to optimize tax collection and enforcement of tax policy. In practice, this means that information should be shared among, for example, customs, local government units that issue licences and health authori- ties – particularly those that regulate the sale of tobacco products. For tax authorities, the most relevant information concerning excise taxes in- cludes the identity of taxpayers and those involved in the trade of tobacco (import and export data, licences, criminal records, tax returns, bank statements, etc.); the category, quantity, value and location of manufactured goods; and the movement of those goods until all taxes are paid. Legal impediments to obtaining this informa- tion – such as bank secrecy or privacy regulations – should be kept in mind, and where needed, exceptions for fiscal procedures should be incorporated into law. Seizure data are also a valuable source of information; more details on this are provided in section 3.4. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 97 Tax authorities should regularly coordinate with law enforcement agencies – such as the police and border control forces, depending on a country’s laws – to properly monitor tobacco-related activities and enforce the tax laws. Often, the competent tax authority and customs authorities work in close cooperation with anti-fraud teams.3 Coordination and sharing of information can be required in legislation or regulations to ensure a streamlined process and avoid confusion. This can be done on an ad hoc basis as needed or with formal planned exchanges of information and regular meetings. It is recommended that at least a legal basis for exchange or access to information among government bodies be established to prevent claims during legal procedures that evidence was obtained unlawfully. Some countries go beyond exchanging information and cooperation. In the Neth- erlands, for example, customs authorities not only carry out work for the Ministry of Finance, they also carry out non-fiscal tasks for seven other departments, including the Ministry of Agriculture, Nature and Food Quality; the Ministry of Justice and Security; and the Ministry of Foreign Affairs (9). These activities are often based on bilateral agreements between the Ministry of Finance and the other departments. In other countries, such as the United States and Canada, Customs and Border Protection are not part of the Ministry of Finance; they are part of the Department of Homeland Security in the United States and the Ministry of Public Safety and Emer- gency Preparedness in Canada. These agencies also carry out many non-fiscal tasks. Along with the implementation of new tobacco control and tax laws, several countries have also created high-level committees to ensure good coordination and implementation of the laws. Led by health and finance ministries, committees ensure coordination and fine-tuning to achieve desired results. Botswana, Chile, Colombia, Indonesia and Senegal, among other countries, have successfully started with coordina- tion, planning and monitoring of tobacco laws’ implementation through periodic com- mittee meetings. The committees usually include representatives from the ministries of health, finance, tax and customs, police, transport and, in some cases, education. Coordination across borders Effective approaches to control smuggling in tobacco products require interventions at the borders of jurisdictions and therefore must involve the border agencies. However, with the globalization of trade, there is a need for close coordination not only between tax and border control authorities but also between different jurisdictions. Recent cases have demonstrated that an absence of formal cooperation frameworks may expose a market to financial crime, including money-laundering and financing of terrorism (10). 3 See, for example, Focus on tax fraud. Customs administration of the Netherlands, tax and customs administration. 2017;2 (https://customsnl-insight.nl/article/309563676, accessed 3 October 2020). 98 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Accession to international cooperation agreements such as the Protocol, the Organisation for Economic Co-operation and Development (OECD) Multilateral Convention on Mutual Administrative Assistance in Tax Matters and other regional arrangements will contribute greatly to the effective exchange of information and cooperation among enforcement agencies. An effective exchange of market data and information from participating jurisdictions can prevent potential cross-border crimes and loss of domestic revenue. International cooperation reinforces domestic measures to stop illicit trade and raise much-needed revenues. Parties to the Protocol have a commitment to cooperate with one another and to share information to meet their obligations under the Protocol (Article 20). The Protocol itself is the legal instrument that allows Parties to cooperate and share information across borders. Authorities of governments that are not Parties to the Protocol or another coopera- tion agreement that represents a legal instrument to exchange information could conclude a mutual assistance agreement or exchange of information agreement to guide the procedures under which information exchange can take place effectively. The Revised Kyoto Convention of 2010 promulgated by the World Customs Organization (WCO) recommends that jurisdictions that enter into bilateral agree- ments require the other jurisdiction to provide pre-arrival information on goods bound for their customs territory. A survey of 87 WCO members in 2013 found that the vast majority of customs administrations had the legal authority to share information related to the supply chain of tobacco products with other administrations (11). Some economic blocs have also established harmonized legislation applying to all of their Member States to provide administrative cooperation to efficiently cooperate on tax matters (12). Coordination can include the establishment of a special agency to ensure the safety and proper functioning of external borders, such as the European Border and Coast Guard Agency, also known as Frontex (from the French frontières extérieures, “external borders”). In some of the Frontex-led operations, EU and non-EU countries cooperate together with international organizations to target cross-border crime, including the smuggling of cigarettes and raw tobacco (13). Criminals who engage in illicit trade of tobacco products are usually also en- gaged in related criminal activities such as bribery, money laundering, corruption, obstruction of justice and even financing of terrorist organizations (14). A number of international treaties provide the legal framework for addressing such conduct through mechanisms that tackle illicit trade from a criminal justice perspective, such as the United Nations Convention against Transnational Organized Crime, the United Nations Convention against Corruption and the International Convention for the Suppression of the Financing of Terrorism. Table 3.1 summarizes the types of structures available for such coordination. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 99 Table 3.1 Structures for coordinating mechanisms TYPE OF COORDINATION BASIS INVOLVED ACTORS National coordination Agreements with a basis in law between national agencies Customs authorities, ministries of finance and those responsible for formulating, analysing and implementing tax policy; law enforcement agencies, such as police and border control forces; and anti-fraud teams Agreements between ministries or a basis in law or regulation on the establishment of high- level committees Ministry of health, finance, revenue, justice, transport and sometimes education and enforcement entities such as customs and police Bilateral coordination Bilateral cooperation agreements National governments Regional coordination Regional arrangements such as • Harmonized legislation applying to all Member States of an economic bloc to provide administrative cooperation in taxation to efficiently cooperate on tax matters • Regulation to jointly establish a special agency to ensure the safety and functioning of external borders EU Member States, the European Border and Coast Guard Agency (Frontex), customs, law and border enforcement agencies International coordination International treaties or conventions such as • The Protocol • OECD multilateral Convention on Mutual Administrative Assistance in Tax Matters • United Nations Convention against Transnational Organized Crime • United Nations Convention against Corruption • International Convention for the Suppression of the Financing of Terrorism Parties to international treaties and conventions, law and border enforcement agencies KEY TAKEAWAY 2 Regardless of differing institutional arrangements, coordination and cooperation within a country and across jurisdictions are essential to optimize tax collection and enforcement of tax policy. 100 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 3.2.3. EVALUATION OF PERFORMANCE AND ACCOUNTABILITY Key strategic indicators are useful for assessing the performance of a competent authority. Performance indicators can include measures such as net revenue col- lected, total expenditures compared with budgeted amounts, the ratio of costs to collection, measures of filing and payment compliance and taxpayer satisfaction (15). Several international organizations, including the IMF, the World Bank, the Inter-American Development Bank and OECD have developed tools to evaluate tax and customs with key performance indicators. This section provides information on some of the indicators that are particularly useful for measuring performance related to tobacco taxes, including the cost of collection ratio, tax gap analysis and tax revenue targets. Cost of collection ratio Collection costs vary among countries. The cost of collection ratio is the total ex- penditure as a percentage of the total net taxes collected. This ratio is often used as a measure of efficiency and effectiveness of competent authorities. In Table 3.2, the cost of collection ratio is calculated for country groups by income level, based on an annual IMF survey. The numbers in the table give an indication of resources used and revenues collected for taxes in general. The same definition of cost of collection was used for all countries. The tax revenue excludes VAT and excise taxes on imported products, so it reflects internal taxes only: personal and corporate income taxes, VAT and excise on domestic production. Customs duties are also not included. The results show the differences among countries at various income levels. Other contributing factors include differences between tax systems, economic situations and compliance levels. Table 3.2 Cost of collection ratio in 2015 per 100 units (ratio of average recurrent budget to revenue collecteda) GROUP (SAMPLE SIZE) 2015 Low-income countries (6) 1.3 Lower-middle-income countries (15) 1.6 Upper-middle income countries (18) 0.9 High-income countries (36) 0.9 All (76) 1.1 a Does not include VAT or excise on imports Source: (Reference 8, Appendix Table 12). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 101 As one would expect, given lower levels of automation and resources, the ratio is higher for low- and lower-middle-income countries, greater than 1.0 (more than 1.0 currency unit needed to collect 100 currency units). The ratios for upper-middle income and high-income countries are below 1.0, indicating more efficient and/or effective collection systems. The cost of collection might be less relevant for taxes that are introduced with other than solely financial objectives, such as influencing a change in behaviour. In particular, in the case of excise taxes applied on tobacco products, the cost of tax collection does not reveal the full picture. If excise tax rates are increased substan- tially – or at least increased above inflation and income growth – consumption will be reduced. As a result, health care costs will be reduced due to reduced tobacco- related mortality and morbidity and increased productivity. These savings are not factored into the ratio of cost of collection to revenue, but governments do benefit from these lower expenses overall. Nevertheless, the cost of collection can be used as an indicator of the efficiency of a competent authority. Tax gap analysis Tax gap analysis is another method of determining how effectively taxes on tobacco products are collected. The tax gap is the difference between the tax due and the tax that is collected. For example, the theoretical tax due under an ad valorem tax on the retail price of cigarettes would be the average price of a pack of cigarettes multiplied by the number of packs sold (estimated from household expenditure surveys, for example) multiplied by the tax rate. This outcome can then be compared to the actual revenues collected (16). The effectiveness of tax collection can also be determined by using the macro- economic input-output matrix, measuring the added value of the economic sector – tobacco in this case – and the theoretical VAT due and then comparing the result with the real VAT collection. This methodology is valid for measuring domestic tax evasion (more information on the use of this method to estimate illicit trade is provided in Chapter 4, section 4.1). Tax revenue target The performance of a competent authority can also be evaluated by determining whether the tax revenue target has been met, if mandated, for a given tax period. Although revenue forecasts are often used as targets, caution is advised. Forecast revenues could include assumptions such as economic growth, inflation and amount collected. Forecasting is a good practice, however, and competent authorities should provide input to the government for the forecasting of revenues to improve the quality of the estimates. Competent authorities should monitor the actual collections 102 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N in comparison with the forecasted revenues, but the theoretical base may not be attainable for a variety of reasons. In addition, a revenue target could provide an incentive for some customs and competent authorities to simply aim to reach the target amount, rather than making efforts to collect the maximum amount possible with the available resources. 3.3 THE TAX COMPLIANCE CYCLE For any tax, there are associated compliance, control and enforcement processes. The compliance cycle usually includes registration and licensing, tax declarations, recordkeeping, storage in warehouses, duty suspension, collection of tax and tax refunds. Figure 3.1 illustrates the typical stages of the tax compliance cycle. Fig. 3.1 Tax compliance cycle 3.3.1 REGISTRATION AND LICENSING Along with regulating and ensuring the integrity of those who deal with controlled substances or goods, the main objective of licensing is to regulate the supply chain. Licensing is a powerful tool for obtaining more information and securing the supply chain of tobacco products. Parties to the Protocol are committed to licensing the manufacturing, import and export of tobacco products and manufacturing equipment (Article 6). In addition, Parties are committed to endeavouring to license – as considered appropriate – the persons involved in the growing of tobacco and the retailing, transporting, wholesal- ing, brokering, warehousing and distribution of tobacco products or manufacturing Registration and licensing (renewal) Tax declaration Authorities: audit and control Taxpayers: recordkeeping Audit and control of information provided in tax declaration Payment and collection of tax Refund Licensed activities, for example: import, production, transport, storage, export, etc. 1 2 3 456 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 103 equipment (Article 6). To ensure an effective licensing system, Parties shall monitor and collect, where applicable, any licence fees that may be levied and consider using them in effective administration and enforcement of the licensing system, for public health or for any other related activity in accordance with national law. If feasible, each Party shall require that retailers and tobacco growers – except for traditional growers working on a noncommercial basis – maintain complete and accurate records of all relevant transactions in which they engage, in accordance with its national law (Article 9.4). Article 6.3(b) of the Protocol provides a list of information to be requested from the applicant of the licence, including: • relevant identity information on the applicant • business location of the manufacturing unit or warehouse and production capacity • detailed list of tobacco products and equipment used • description of where the manufacturing equipment will be installed and used • documentation or declaration of any criminal records • information on bank accounts to be used for transactions and payments • description of intended use and intended market of sale of the tobacco products. To make it easier for authorities to collect all the information they need, rules of confidentiality could be exempted in the licensing process. Licences can be general – covering all activities requiring a licence – or issued for each activity separately, such as different licences for manufacturing, importing and retail. A general licence is less burdensome for the licensing authority, whereas licences for each type of activity offer greater control but at the cost of more adminis- tration (17). The cost of implementing the licensing system should be proportionate to the potential impact of the system. Not only should the type of licences be taken into consideration, the process and information needed to obtain a licence should be carefully considered to ensure proportionality. The more stringent the process is – in terms of the information required and the obligations the system imposes on licensees – the more burdensome the regime will be on both businesses and the authorities who must administer and enforce it. The more information is collected, the higher the compliance and administrative burden will be. It is recommended that the added value of the information be balanced with the additional compliance, administrative and/or enforcement burden. The level of stringency should be decided with consideration of factors such as the level of risk of the activity and the availability of enforcement capacity. A more stringent regime might be justifiable for activities that pose a higher risk for the 104 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government in terms of potential loss of tax revenues – such as the import, production and handling of excisable products on which the excise taxes have not yet been paid. Authorities could consider setting licence fees at a high enough level to cover the costs of administering and enforcing the system. For an example of a system that relies mainly on licensing and permissions, see the case study of Australia in Box 3.1. Wholesalers, distributors and retailers of tobacco products could also be required to obtain a licence before they can engage in the trade of those products. This would enable the competent authority to require reports on, for example, transactions relating to the purchase and sale of tobacco products. Moreover, it would allow the authorities to complete the audit trail of the entire supply chain and to obtain data that will help tax and health policy-makers properly and effectively monitor tobacco products. Governments could also require a licence for entities dealing with raw materials or growing tobacco, including farmers. If licensing of tobacco farmers is deemed appropriate and subsequently required in a country, the farmers have to identify and register their farm areas and location to obtain a licence. The benefit of requiring licences for farmers is that the control of the legitimate supply chain is extended to the identification of the source of the raw material for tobacco products. It also makes it more difficult to divert raw tobacco from the licit to the illicit supply chain. Licences are issued by different agencies across the world. In Brazil, for example, the Health Surveillance Agency is responsible for providing licences. Operators need to obtain approval of the layout of manufacturing and warehousing facilities before they can operate. In addition, they must demonstrate how they will comply with other laws and regulations – for example, by showing the design of product packaging, including the pack, carton and master case. The factory location must be identified before manufacturers can obtain a licence. Finally, a licence is required for the importation of machinery to produce tobacco products (18). Licences can be a source of useful information if authorities establish the informa- tion that applicants must supply in order to obtain the licence. Such information could include the quantity, price and how the tobacco harvests are disposed, as well as the identity of the buyers. It is recommended that an effective licensing regime collect information to establish both the identity and characteristics of applicants by requiring criminal records on relevant offences, such as previous noncompliance with tobacco licences or fraud. To avoid loopholes for monitoring raw tobacco, importers of tobacco leaf could also be licensed or at least required to register and report information on quanti- ties, sources and sales. In some countries, this information is already collected by a government agency other than tax authorities, for example, by the ministry of agriculture. Duplication of requirements and reporting should be avoided through CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 105 legislation and coordination among agencies. If licensing of (small-scale) farmers is difficult to implement, subsequent purchasers (first processors) in the supply chain could be licensed and regulated instead. Licensing first processors is often less burdensome to enforce for competent authorities because, in general, there are far fewer first processors than there are growers. For example, in the EU, between 50 and 100 first processors have been identified, compared with 55 000 farmers (19). Countries could also consider requiring registration of persons or entities engaged in the manufacture and import or sale of materials used for the manufacturing of tobacco products, such as cigarette papers, tobacco leaves, additives, adhesives, acetate or any other type of filters used for cigarettes, tipping paper and cellophane or plastic wraps, as well as materials for packing the cigarettes into packs, reams and master cases. In addition, tobacco manufacturers could be required to obtain a licence before they can purchase these materials. The Parties to the Protocol should decide on appropriate measures, depending on research as to whether key inputs that are essential for manufacturing of tobacco products exist and can be identified and subject to effective controls. KEY TAKEAWAY 3 The objective of licensing is to regulate and secure the supply chain. It is a powerful tool for obtaining information for verification, further investigation and audits. Ideally, all persons involved in the growing of tobacco and retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. Licensing requisites Based on case studies and best practices – including experiences from managing bonded warehouses where the value of merchandise or suspended duties or taxes is high – the following kinds of information could be required to obtain a licence, in particular, for producers, warehouses and distributors of tobacco products: • certification of safety of installations, perimeter security for production and storage (may include CCTV [closed-circuit television]4 access for tax administration) • certification of financial solvency • detailed online, real-time inventory of tobacco products and main raw materi- als, accessible by tax administration • electronic accounting systems 4 The term “closed-circuit television” is used generically to describe surveillance camera systems. 106 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • detailed lists of owners and managers • banking and other financial records • periodic electronic reports of transactions for tobacco products • anytime tax administration right of entry for inventories • mandatory electronic tax returns and payments • mandatory prior-to-arrival customs declarations for tobacco products • declarations of compliance with the tax stamp system (if applicable) • for those involved in import or export, authorized economic operator (AEO) certification • proof of compliance with the bond or guarantee regime • agreement to finance reasonable cost of inspections and tracking and tracing. Box 3.1 Case study of licensing in Australia Australia has taken an approach to controlling tobacco taxes that differs from that in many other countries. It has not used fiscal marks or tracking and tracing.5 Instead, it administers tobacco taxes through licensing and permission-based systems aimed at facilitating operations by lower-risk entities while preventing or tightly controlling commerce involving higher-risk entities. The domestic tax agency, the Australian Taxation Office (ATO), is responsible for most of the controls. These controls cover tobacco that is grown or manufactured and imported as finished goods or as leaf for manufacturing in Australia. In fact, the legal tobacco market in Australia consists only of imported finished tobacco products. In 2006, all tobacco-growing licences were cancelled by the ATO because manufacturers switched to cheaper leaf from external suppliers. The last domestic cigarette manufacturers closed in 2015 and 2016, and there has been no legal domestic tobacco growing or manufacture since then. The ATO administers all other functions relating to the import of tobacco and tobacco products, including licensing of bonded warehouses used to store imported products and issuance of permissions to undertake movement of bonded tobacco products between licensed bonded warehouses or to places of export. Importers must apply for a licence for a bonded warehouse to store imported tobac- co. The applicant must meet general criteria such as fitness, recordkeeping and security.6 These criteria are designed to ensure that only low-risk entities are able to enter the ex- cise tax system. Risk levels are also kept at an acceptable level through provisions allow- ing the suspension or cancellation of licences, subject to appeal. Licences are valid for a 5 See sections 3.4.4 and 3.4.5 for detailed discussions on fiscal marks and tracking and tracing. 6 The entity must not have been charged with an offence under the Excise Act or any Commonwealth, State or Territory Act that carries a penalty in excess of US$ 105 000 in the previous 12 months (or convicted in the previous 10 years), has shown a history of compliance with tax law in the previous four years, has had no previous cancellation of a licence, has adequate financial resources and is not in receivership. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 107 three-year period, with automatic renewal for licensees with demonstrated compliance. The permission system relies on post-transaction audits of commercial records. Criteria used to assess risk include the size of the duty liability, the compliance record of both parties and the possibility of diversion into the market. When there is a perceived risk of revenue loss, the application can be denied or a financial security deposit can be required. Exports of tobacco products are also subject to an export declaration process with the Australian Border Force. An approved export declaration is required for the products to be able to leave the country. Following recommendations from a government task force in 2017, the status of tax-suspended, bonded tobacco was eliminated as of 1 July 2019. In addition, an import licensing regime was introduced, and commercial tobacco imports without a licence are banned. Importers are required to identify their duty liabilities at import and make immediate payment; there are no credit terms available. Full payment of duties and taxes to the Australian Border Force are required prior to a release of tobacco products into the country. Sources: (20–21). As mentioned above, certification as an AEO could be requested as part of the licensing process. Most customs authorities are familiar with the concept of AEOs. Created by the WCO, AEO principles were initially focused on security concerns (22). Having a special licensing regime for operators of the tobacco supply chain is recom- mended due to the special nature of the product. For countries that have no system in place, AEO certification could be a starting point for setting up such a regime. An AEO is defined by the WCO SAFE Framework of Standards (22) as a party involved in the international movement of goods – in whatever function – that has been approved by, or on behalf of, a national customs administration as complying with WCO or equivalent supply chain security standards. AEOs include, inter alia, manufacturers, importers, exporters, brokers, carriers, consolidators, intermediaries, ports, airports, terminal operators, integrated operators, warehousers and distributors. For many years – in some cases, even since the 1970s – customs administrations have been increasingly involved in the security of the international trade supply chain. More recently, customs administrations have developed security programmes in a global context. The AEO is part of these programmes, and in 2005, the WCO adopted the SAFE Framework of Standards. Since then, a number of traders have been required to make substantial investments in order to obtain AEO status and must continue to invest to maintain that status. The AEO program is also recognized by the Trade Facilitation Agreement, a multilateral agreement signed by 174 countries (23). 108 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Some regional blocs have further specified the standards for AEOs and provide clear and well-structured information on their websites to guide and encourage operators to apply for AEO status. A good example of this practice is the website of the Revenue Commissioners of the Republic of Ireland, which contains the in- formation shown in Box 3.2. Box 3.2 AEO: Republic of Ireland Tax and Customs What are AEOs? AEO status is a certified standard authorization issued by customs administrations in the European Union (EU). It certifies that an economic operator has met certain standards in relation to: • safety and security • systems to manage commercial records • compliance with customs rules • financial solvency • practical standards of competence or professional qualifications. This is primarily a trade facilitation measure that recognizes reliable operators and encourages best practices in the international supply chain. As an AEO, an operator could benefit from: • recognition worldwide as a safe, secure and compliant business partner in international trade; • lower risk scores in risk analysis systems when profiling; • priority treatment if physical controls are conducted; • mutual recognition of AEO programmes under Joint Customs Cooperation Agreements, which could result in faster movement of goods through third- country borders; • reduced data sets for entry and exit summary declarations (this applies only to AEO safety and security); • easier access to simplified procedures; • reduction or waiver of comprehensive guarantees. The conditions for AEO status apply to all businesses regardless of size. Manufacturers, exporters, freight forwarders, warehouse keepers, clearance agents, carriers and importers may all apply for AEO status. Source: (24). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 109 3.3.2 DATA COLLECTION, DECLARATIONS AND ACCOUNTING The effectiveness of risk analysis depends on the quality and reliability of the available data. This is also the case for risk analysis in relation to tobacco taxes. Obtaining reliable data can be a challenge in many countries, but the use of electronic sys- tems to collect and manage data is increasing in most competent authorities. The introduction of VAT in many countries around the world has greatly improved the availability of data that can be used for tobacco tax analysis, since reporting is done along the supply chain on, for example, the value, quantity of goods and transaction date. Most countries applying excise duties also have a VAT system in place. In addition, more countries are becoming Parties to the Protocol. With the implementation of the Protocol, more data will become available because countries will be obliged to implement, among other measures, licensing systems with report- ing requirements and tracking and tracing systems. More information on tracking and tracing systems is provided in section 3.4. The obligations of the Protocol will also assist in monitoring the stock of tobacco products. Ideally, all entities involved in the tobacco product supply and distribution chains should be licensed and required to record every transaction that occurs. As this might be burdensome for both tax authorities and taxpayers, the use of automated and electronic systems is recommended in order to decrease the costs of compliance. An accurate inventory system for all raw materials, machinery, goods in process and finished products can be required. It is even more important to have good recordkeeping of the required data. As the volume of reported data increases, a good information technology (IT) system will be needed. The use of IT for periodic tax declarations, accounting, inventory and financial data is critical for obtaining accurate information and decreasing costs for the entire reporting system. Most countries now have some level of automation that can facilitate data analysis. An emerging trend is the use electronic invoices, issued by traders, as part of online real-time information for tax administration. Countries generally start by using electronic invoices at public utility companies and then later expand the use to large companies. Electronic invoices minimize the use of paper, contribute to automated recordkeeping and give accurate and timely information about transactions for tax administration. Several countries began using electronic invoices for companies on a voluntary basis and later made their use mandatory, especially for large companies with a high number of transactions – including the tobacco industry. Electronic invoices have been implemented successfully in EU countries and almost all Latin American countries, as well as several Asian countries.7 7 Electronic Invoicing in Latin America: English Summary of the Spanish Document; Inter-American Development Bank, Inter-American Center of Tax Administrations, 2018 (https://publications.iadb.org/ publications/english/document/Electronic-Invoicing-in-Latin-America.pdf ). 110 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N To verify that information is accurate, competent authorities could systematically cross-check declared information against third-party information (e.g. from banks, financial institutions, employers) or match the data with the information in registers of other government agencies. Processes of cross-checking and data matching could also be automated to minimize the administrative burden (25). KEY TAKEAWAY 4 Reliable data are essential for effective risk analysis. While obtaining these data can be challenging, electronic systems can help reduce the burden by automating procedures of data collection and cross-checking of information with different sources. 3.3.3 RECORDKEEPING Parties to the Protocol are committed to requiring, as appropriate, that all persons or entities engaged in the supply chain of tobacco, tobacco products and manu- facturing equipment keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of tobacco products (Article 9). Relevant information includes market volumes, trends, forecasts of tobacco products and quantities of tobacco products and manufacturing equipment kept in stock in tax and customs warehouses in transit, transhipment and under duty suspension. This information should be required from the persons and entities engaged in the supply chain and submitted to the competent authority on a regular basis, as provided for in the law. The competent authority can use the submitted information to monitor compliance with tobacco regulations and payment of taxes. A registry with this level of detail can realistically be kept only in electronic form. Records must provide full accountability for materials used in the production of tobacco products. The intention is that tax authorities and manufacturers should be able to reconcile the production quantities with the inputs used in production – thereby providing confidence that no unrecorded or illicit production has occurred. Obligations should also be imposed on suppliers of key inputs to show that supply is commensurate with demand (17). KEY TAKEAWAY 5 To monitor compliance and payment of taxes, all persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions, as well as details of materials used in the production of tobacco products. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 111 3.3.4 WAREHOUSING, STORAGE AND DISTRIBUTION According to Article 6.2 of the Protocol, all Parties shall endeavour to license persons involved in any wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Maintaining a system of authorization allows the authorities to carry out controls in production and storage facilities to ensure that taxes are paid (2). The approval process to obtain an authorization could include an evaluation of the layout of the plant or warehouse, the machinery that will be used and the flow of production, warehousing and shipping, including the points of entry and exit of raw materials and finished products. The basic method of monitoring production and ensuring that only tax-paid products are released to the market from the premises is to identify the production facilities and to control the entry and exit points. From time to time, the competent authority should conduct a physical inventory of the goods contained therein to check whether all documentation was duly prepared and approved and to determine the accuracy and completeness of the records kept. If the jurisdiction requires tax stamps to be placed on the tobacco products, only products with the proper stamps affixed can be withdrawn. Generally, tobacco products for which the required taxes have not been paid and, if required, fiscal marks have not been affixed should not be allowed into warehouses. For practical reasons, many countries allow suspension of excise duties, meaning that prior authorized persons can produce, send, receive and store tobacco products on which the excise duty has not yet been paid. The relevant authorities could also require that products on which the taxes have been paid should not be stored in the same areas as the products under duty suspension. Obviously, products under suspension of payment of excise duties are at high risk, which could justify stricter requirements for production, trade, storage and handling. Australia, which has a strict system of licensing and requirements for permission to move tobacco products, has migrated to a new system that eliminates bonded warehouses from the supply chain as of 1 July 2019. Importers are required to have an import licence and to pay excise taxes on cigarettes immediately upon import (see Box 3.1 in section 3.3.1). 3.3.5 DUTY SUSPENSION Many countries require authorization of natural or legal persons (as authorized warehouse keepers) to produce, process, hold, receive and dispatch products sub- ject to excise duty during their business. Producing, processing, holding, receiving and dispatching excise goods often take place under suspension of the excise duty. Guarantees can be requested from authorized persons to secure the payment of taxes. Features of such a system may include strict criteria for granting authorization, 112 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N warehouse pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. The use of a computerized system for monitoring movements of excise goods under suspension of excise duty can be a control as well. Different licences for products under duty suspension could also be considered. This would make enforcement easier and less burdensome for both authorities and operators. In general, it is recommended to allow the handling of excise goods under suspension of duties only if strict criteria are met. Such criteria could include pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. In principle, the movements of tobacco products should also be covered by the tracking and tracing system. Considering the high risk related to these products, additional monitoring could be considered appropriate, such as a computerized system monitoring the movements of excise goods under suspension of excise duty. In the design of such a system, it is recommended that close attention be paid to customs procedures for import and export to ensure alignment and avoid a vacuum in monitoring. An example of a computerized system is the EU’s Excise Movement and Control System, which follows the movement of all excise products – including manufactured tobacco products – for which excise taxes have not been paid. The system records the movement in real time and is thereby an important tool for combatting fraud. In addition, this system is indispensable for the exchange of information and co- operation between the relevant authorities of Member States of the EU (26). Finally, authorization is required before tobacco products can be produced, imported or stored under suspension of excise duties (27). KEY TAKEAWAY 6 Products under duty suspension of excise taxes are at a higher risk of tax evasion, which can justify stringent measures such as requesting guarantees to ensure the payment of taxes, additional licensing requirements, compliance with computerized systems to monitor the movement of excise goods under suspension and on-site authorization and audits. 3.3.6 COLLECTION OF TAXES To reduce the complexity of tax collection systems, it is recommended that excise taxes be imposed at the point of manufacture, import or release from storage or production warehouses for consumption. This is common practice in the majority of countries that impose excise taxes. Collecting taxes at this level of the supply chain CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 113 greatly limits the number of taxpayers and thus the resources needed to control them. Encouraging taxpayers to use electronic payment methods can also increase the chances of collecting all taxes. The same applies to requiring guarantees for certain high-risk activities, such as the handling of goods under duty suspension. Many countries decide on a case- by-case basis the level of the guarantee, depending on the situation of the requestor and the level of risk (quantity or value and potentially due excise taxes) that the regular business activities represent in a given time frame. Some countries allow a reduction of guarantees for operators with a track record of good compliance. It should be noted that a guarantee is not a limitation of the liability; taxpayers can still be requested and liable to pay an amount far above the level of the guarantee. Tax payments should be required by law to be remitted at fixed intervals after sales or on a fixed date each month (2). Many countries have a specialized collection enforcement unit that works full-time on the collection of taxes. It is important to have a stop-filer or payment control that can act immediately when noncompliance occurs, by sending a message and phone call of late declaration or late payment to the taxpayer. This increases the likelihood of keeping taxpayers compliant. If nondeclaration or nonpayment persists, the bond or guarantee could be executed. Another reason for collecting excise taxes around the time of production or import is that quantities can be monitored more effectively at these points. There are different options for monitoring the supply chain of tobacco products. The decision about what kind of monitoring system to use depends on the country’s financial, technical and human resources. The weakest form of monitoring is in- dustry self-declaration. Activities to verify compliance and ensure the collection of the full amount of taxes due can include, for example, physical checks, audits, cross-checking of declared information with third-party data and inspection of administration and recordkeeping. In general, in countries with poor administration systems, enforced compliance is carried out by imposing physical control over the production or manufacturing process. The cost of physical control increases when there is a potential for fraud by excise officers. However, fraud can be diminished significantly when excise officers are rotated frequently among different locations and supervisors make surprise visits. Historically, some countries (e.g. India) have posted tax administration staff at production facilities to monitor production and removals. In India, a staff member of the competent authority is placed in cigarette and large bidi manufacturing facili- ties around the clock. Each officer records the daily production and the quantity of cigarettes/bidis that leaves the factory and reports to the next officer.8 8 Ministry of Finance India, personal communication, 2009. 114 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A better option is to monitor production remotely. The competent authority can require the installation of CCTV cameras in strategic places throughout the manufacturing and warehousing facilities. With these, the authority can establish a central command post from which the facilities and activities can be continu- ously monitored and documented. In addition, the competent authority can carry out physical inventory controls from time to time and – if electronic invoices are implemented – cross-checking between invoices and declared inventory. This is also an effective way to prevent collusion between staff of a competent authority and manufacturers or importers. For example, in 2015, the Bureau of Internal Revenue of the Philippines required all tobacco companies to install CCTV cameras in their production lines and warehouses. This decision was taken in response to large seizures of untaxed cigarettes, with the objective of monitoring production to ensure the payment of all taxes. The collection process must also be supported by IT systems. These systems must provide for transparency and accuracy to ensure a safe process for the flow of payments from taxpayers to the tax treasury. Most countries have implemented automated electronic systems for tax payments linked to each declaration, for both domestic and import. It is key for tax administrations to have a comprehensive agreement with the banking system in order to obtain lower transaction costs, if applicable. Some countries have implemented a state payment web portal that allows citizens to pay their taxes and other fees such as county fees, fees for car permits and licences and agricultural, health and environmental fees online. KEY TAKEAWAY 7 Excise taxes should be imposed at the point of manufacture, import or release from storage or production warehouses for consumption, to ensure that quantities can be monitored effectively. This also reduces the complexity of tax collection systems by limiting the number of taxpayers and thus the resources needed to control them. 3.3.7 TAX REFUNDS Refunds for VAT, excise taxes and customs duties are a common process in most countries, under the principle that consumption taxes are not exported. Frequency and methods of refund vary by country. It is common to have monthly refunds (if there are exports during the period), and the reimbursements may be sent directly to the exporter or reserved as a credit to pay other taxes. An alternative used by some countries that have a high volume of exports is a so-called zero rate, or suspension, meaning that indirect taxes (VAT, excise taxes and customs duties) are suspended for the whole chain – from import of raw materials to production and packing until CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 115 export. This regime requires a special licensing process. Since the tobacco industry has an export component, the refund process for this sector requires special atten- tion for tax administration. 3.4 CONTROL AND ENFORCEMENT Control and enforcement are the main functions of tax administration. In fact, most tax laws include the objective “to control and enforce tax compliance” and, for customs, “to control and enforce tax and duty payments at the border” or similar phrases. The Protocol provides guidance for control and enforcement of tobacco taxes. Efficient and effective competent authorities often have a strategic plan to ensure compliance, a risk-based approach to identify the problematic points in the chain and the ability to direct resources accordingly to high-risk or high-value areas. Tasks that can play a role in control and enforcement include controlling the registration and licensing process, due diligence, verifying declarations and collec- tion of taxes. Production and distribution controls including tracking and tracing, fiscal markings, audits and import and export controls all play a role in control and enforcement. This section describes the main activities for improving control and enforcement, focusing on the tobacco supply chain. The procedures and penalties that can be enacted once illicit trade in tobacco has been detected are also discussed. 3.4.1 CONTROL AND ENFORCEMENT PLANNING Strategic plan In modern tax administrations, it is common to have a strategic plan, with control and enforcement as pillars. Appropriate control of the compliance cycle is key to keeping taxpayers in compliance and preventing illicit trade and tax avoidance. For this reason, most tax administrations focus a majority of their resources on preventive policy. Some examples of this can be found in the strategic plans of the United Kingdom and the United States’ Internal Revenue Service (IRS). In the United Kingdom, HMRC has had a well-developed strategic plan for years. A key pillar of the plan focuses on keeping taxpayers compliant. This is the concept of prevention: controlling initial minor noncompliant behaviour for the majority of taxpayers, while using strong enforcement for the minority on the noncompliant side (28). The strategic plan of the IRS has a similar approach, with a focus on control. If noncompliance is detected, data analysis and behavioural insights are used to identify the best way to address noncompliance. Early intervention or self-correction are examples of ways to address detected noncompliance. The IRS also highlights the importance of resolving noncompliance to ensure taxpayer confidence in the tax system and protecting the integrity of the system (29). 116 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Risk-based approach Following the establishment of a strategic plan, an enforcement and control plan must be drafted. This plan should include definitions of the activities that will be enforced, the taxpayers upon whom they will be enforced and the circumstances under which they will be enforced, as well as allocating resources for staffing, audit- ing, infrastructure and IT. Targets must also be defined, including the number of interventions and the amount of additional collected revenue or reduction of tax evasion. Several tax administrations elaborate annual plans with periodic perfor- mance reviews aimed at improving results, and they correct allocations and targets as needed. Clear targeting of interventions is needed for better results, more efficient use of resources, lower costs for taxpayers and more effective collection. In other words, the point is to focus interventions on those who have a higher probability of noncompliance. Using a risk-based approach can be particularly beneficial. Tax risk management is a key element of control strategy in modern tax admin- istration. A risk is a possible threat to reaching objectives such as collecting taxes in an effective and efficient way for competent authorities. Risk assessment is the process of analysing risks and deciding on the best way to manage an identified risk. The responses can vary from acceptance to mitigation to avoidance. Proper risk assessment allows competent authorities to use their available resources most efficiently and to become more effective in dealing with risks. It can be used to improve compliance by identifying taxpayers or types of activities with a high risk of noncom- pliance. Groups of taxpayers with the same characteristics often have similar risks. Groups with a high risk of noncompliance could then be subject to greater review. Areas of potentially greater risk of noncompliance in the tobacco supply chain include import, export and transfers to and from warehouses, particularly when they take place under duty suspension. Gathering risk-related information from internal and external sources is a best practice in compliance risk management. Such sources could include third-party information (e.g. from banks, credit card companies, transport companies), studies on taxpayer behaviour and research on compliance issues, tax gap analysis, tax audits and declarations (30). Risk management uses these different sources of data along with algorithms to find patterns of high noncompliance. Risk analysis can indicate reduced risk as well. Lower-risk areas are likely to need less governance to ensure compliance, which allows for resources to be directed elsewhere. Risk assessment can therefore help with strategic allocation of limited resources to the areas of greatest risk while at the same time reducing the burden on lower-risk taxpayers. Risk management has always been done by competent authorities, but data availability and statistical methods to identify patterns have changed the way risks CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 117 can be assessed. Although many risk assessment systems are still done manually or include manual elements, the use of intensive data techniques allows systematic, deeper and more targeted analysis (31). Modern risk assessment makes use of electronic data on taxpayers, tax payments, declarations from other taxes, such as VAT, and third-party information. With these data, tax authorities can identify indicators that suggest where further activities might be required to ensure compliance. For example, VAT invoices can be used to match reported purchases of inputs of tobacco leaf to sales invoices of tobacco leaf wholesalers. For taxpayers (i.e. those who are licensed and provide required reports), competent authorities can create a business analytics program to determine whether the data reported are consistent on each side of the transactions. Moreover, in countries with a VAT system, competent authorities can compare the data reported by taxpayers under the VAT system with data reported under the tobacco excise tax system to detect any inconsistency. VAT invoices can also be used to verify inputs and sales data. If VAT is collected at all levels of the supply chain, it is easier for govern- ments to monitor the supply chain for the enforcement of excise duty obligations. Regular surveys on tobacco consumption that use the same methodology can also provide indications about the level of compliance with excise tax policy. A sudden drop in revenue that is not reflected in consumption data could be an indication of illicit manufacturing, illegal imports, cross-border shopping or forestalling. In addition, seizure data can provide valuable information on areas and activities at high risk of noncompliance. The structure of tobacco tax policies should also be taken into account when con- ducting risk analyses. If excise tax rates are increased, there might be a greater risk of forestalling or front-loading (see the discussion on anti-forestalling later in this section). Differentiated excise tax rates based on product or packaging characteristics – such as distinctions between soft and hard packs or filter and nonfilter cigarettes – are also prone to manipulation by operators, which could affect tax revenues. One of the options to mitigate these risks is to amend the excise tax policy and apply a uniform tax rate. For customs transactions, the use of risk management is a key element in target- ing merchandise and support declarations to be inspected. Before the 1990s, most customs agencies used random criteria for selecting targets for inspection. Since that time, many countries have implemented risk-based approaches for selecting inspections. Historical data on importers and trade communities, complemented by artificial intelligence technology, show that risk management tools dramatically increased the effectiveness of physical inspections. Most modern customs agencies have implemented such techniques, allowing for more effective control processes while facilitating smoother processes for those transactions that are in compliance. Box 3.3 details some of the recent changes in risk management processes. 118 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.3 Changes in risk management The OECD developed Fig. 3.2 to show the framework and key steps for understanding compliance risks in 2004 (32). The same approach is still used to identify, assess and prioritize risk. However, many competent authorities now use new technologies and advanced data analytics, along with more information sources, including external data from banks, employers and sales invoices for VAT, for example. Fig. 3.2 Compliance risk management process Source: (32). The methods of identifying risks and the analysis of compliance behaviour have also changed. Traditionally, competent authorities used audits to identify high-risk cases. With more diverse and better data, competent authorities can now use more evidence-based approaches to examine risk patterns. Success of compliance activi- ties is now more often measured in terms of their impact on the overall compliance environment, rather than only on increased revenues.9 9 For more information on effective risk management with several indicators and a checklist of questions, see the Tax Administration Diagnostic Assessment Tool (68). OPERATING CONTEXT Assess and prioritize risks Evaluate compliance outcomes: - Registration - Filing - Reporting - Payment Monitor performance against plan Analyse compliance behaviour (causes, options for treatment) Determine treatment strategies Identify risks Plan and implement strategies CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 119 In Indonesia, use of the compliance risk management process reduced the share of illicit trade in total consumption of cigarettes from 12% to 3%. More information on this can be found in the case study of Indonesia in Box 3.13 later in this chapter. Understanding the products – as well as the supply and distribution chains – al- lows competent authorities to identify which areas along a chain pose the greatest risk and therefore require more resources. Detailed information on the composition of selected tobacco products is given in Annex 3.1. KEY TAKEAWAY 8 Risk analysis helps identify the points of intervention that have higher probabilities of noncompliance. A risk-based approach with targeted interventions allows for better results and more efficient use of resources to ensure effectiveness of tax collection. 3.4.2 CONTROLS OVER THE TOBACCO SUPPLY CHAIN As defined in Article 1 of the Protocol, the supply chain covers the manufacture of tobacco products and manufacturing equipment – as well as their import or export – and may be extended, where relevant, to one or more of the following activities when so decided by a Party: 1. retailing of tobacco products 2. growing of tobacco, with the exception of traditional small-scale growers, farmers and producers 3. transporting of commercial quantities of tobacco products or manufacturing equipment 4. wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Article 4.1 of the Protocol requires parties to “adopt and implement effective mea- sures to control or regulate the supply chain of tobacco products to prevent, deter, detect, investigate and prosecute illicit trade in such goods and to cooperate with one another to this end”. Concrete measures to regulate the supply chain, as well as best practices in this regard, are discussed further below. Figure 3.3 shows the main places for reporting and monitoring along the supply chain: import, ex-factory and removals from warehouses. Manufacturers could be required to report imported inputs at the border, as importers of finished products do. If components are subject to licensing, information can be required as part of the licensing process. The arrows in Fig. 3.3 represent transporting, which is also part of the supply chain. 120 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 3.3 Cigarette supply chain from manufacture or import to retail sale 3.4.3 LICENSING10 AND DUE DILIGENCE A licensing system is effective only if it is properly controlled. Most tax administra- tions have experience with licensing processes for excise taxes on products such as alcoholic beverages and energy products. It is strongly recommended that lessons learned with the licensing process of such products be applied when implementing and enforcing tobacco-related licensing. Licensing provides timely and accurate data that can serve as the basis for audits because it identifies and controls legitimate operators. For new operators, the process to obtain a licence could include visits and verification of production factories, storage facilities and distribution premises. Countries that have no licensing system in place and would like to start applying licences could allow a transitional period for existing operators to comply with the new licensing requirements. The process of licensing control must be carried out and updated periodically, in particular by controlling the validity of bonds or guarantees, the proper functioning of the required systems (CCTV, for example) and recordkeeping. 10 Licensing is discussed here in the context of due diligence and enforcement. Details about how licensing can be set up and what information could be requested are presented in section 3.3.1. Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blend d. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: wher he tobac o will be roll d into st cks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 121 Where licences are required, the law should include a provision specifying that purchases from unlicensed suppliers – or sales to unlicensed purchasers – are not allowed. This means that both suppliers and purchasers would need to verify those with whom they are doing business. This requirement provides enforcement au- thorities with an entry point to enforce the licensing system at both ends. Also, a licensing requirement for manufacturing equipment assists authorities in identifying and prosecuting illegal manufacturing of tobacco products, reducing the burden of proof substantially. In many countries, the presence of manufacturing equipment is not sufficient proof that illegal manufacturing is taking place; the machinery has to be in operation and producing illegal tobacco products when authorities inspect the location. With a licensing requirement, however, the presence of machinery without a licence is sufficient for authorities to act. The validity of licences should be time-limited, requiring renewals or reapplica- tion, to maintain a high level of control. Adherence to the conditions required for a licence should be controlled by the authorities, and penalties for noncompliance – for example, suspension or withdrawal of a licence – should be severe enough to act as a deterrent (33). Regulations for licensing should provide for inspection of the licensee’s products and premises, with penalties for noncompliance, which could include criminal and civil prosecution for serious or repeated offences. As stated in Article 6.3(a) of the Protocol, Parties need to establish or designate a single authority or multiple authorities to issue, renew, suspend, revoke and/or cancel licences. In accordance with Article 7 of the Protocol, persons engaged in the supply chain are required by law to conduct due diligence before and during business relationships. They also must report to the competent authorities any evidence that a customer is engaged in activities in contravention of its obligations arising from the Protocol. This requirement includes customer identification, monitoring of sales to ensure that the quantities are commensurate with demand for such products within the intended market and taking measures to ensure compliance. Knowledge of the demand of a market is indispensable for determining if there is a case of oversupplying. If the supply of tobacco products to a lower-taxing foreign market exceeds the demand, it creates a higher risk that these products will be smuggled back into a higher-taxing country, undermining the objectives and effectiveness of the higher-taxing jurisdiction. In the past, some governments decided to impose a fine on tobacco companies if the quantities supplied were significantly higher than the demand and the risk of being smuggled back into their jurisdiction was judged to be high (34,35).11 11 Excise duty rates applied in all the EU countries can be found on the European Commission’s webpage: https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco-energy/ excise-duties-tobacco_en. 122 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 9 Licensing helps to identify and control legitimate operators. The data obtained from licensing can serve as a basis for audits. Licences should be controlled on a regular basis and updated periodically to ensure their validity. 3.4.4 FISCAL MARKINGS (E.G. TAX STAMPS) Fiscal markings are another important tool for controlling and monitoring pro- duction and import of tobacco products. Their use is generally considered to be appropriate for increasing compliance with tax laws. Fiscal markings can also be of help in distinguishing between genuine and illicit tobacco products. Tax stamps or other fiscal markings affixed to packs of cigarettes or tobacco products facilitate the collection of excise taxes, as well as audits and enforcement actions. The pres- ence of fiscal markings enables both the competent authority and the public to monitor whether the taxes on tobacco products were properly paid. It thus assists the competent authority in investigating illicit trade and prosecuting violations. Fiscal markings include tax stamps, enhanced tax stamps (banderols) and digital tax stamps. Examples of fiscal marks are tobacco stamps, tax stamps, excise stamps, tax stickers and banderols. Box 3.4 presents details on the different types and features of tax stamps. Tobacco products for export are often required to be marked that they are for export. Box 3.5 provides useful information regarding the International Organization for Standardization (ISO) standard for excise tax stamps. The terminology “fiscal mark” holds no indication of the characteristics of the mark. A fiscal marking is affixed to each pack of tobacco product. Requiring a standard package size can facilitate the application of the markings (2). Fraudsters can be deterred from attempting to re-use fiscal markings (in particular stamps) by having the marking affixed to each pack of cigarettes (or other tobacco product) before the pack is wrapped with cellophane (36). In most cases, tax stamps are purchased by the producer or importer and applied to each product sold as proof of excise tax payment (33). Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Box 3.4 Types and features of tax stamps Over time, tax stamps and markings have become more sophisticated. In the past, tax stamps were often paper-based and easy to counterfeit. New tax stamps use additional security features to make them more difficult to counterfeit. Authentica- tion solutions against counterfeiting can utilize various security features, including: • overt features – features that can be verified by the naked eye; CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 123 • covert features – features that can be authenticated only by using dedicated and specialized electronic readers; • semi-covert features – features requiring a simple tool that does not involve extensive training; and • forensic features – features that can be identified through laboratory analysis. Tax markings can be either physical or digital: • physical markings – the information is contained in the document or device attached to the package. • digital markings – information is obtained through a link with a database and by decrypting with the tools and keys used for creation of the data. The term “digital tax stamp” sometimes leads to confusion, as some paper-based stamps with digital components are also described as digital tax stamps. Tax markings that are fully digital do not contain information in the document or device attached to the package. It is probably too simplistic to say that digital tax stamps are more secure than paper- based tax stamps. Both types have advantages and weaknesses. For instance, both physical and digital tax stamps can be weak or strong on security features. Neverthe- less, the management, production, sales, transport and monitoring of physical tax stamps require increased attention. Tax stamps have the same value as banknotes and are a possible attraction for theft, loss and fraud. In Belgium, for example, the Court of Audit severely criticized the lack of control of the production and stock management of tax stamps in 2015. The Court of Au- dit concluded that tax stamps issuance should operate under recognized security practices and procedures relative to the security risk associated with the various production, distribution and issuance processes. Moreover, it was noted that new printing technology of digital tax stamps on packs may facilitate stock management and lead to less fraud. In March 2016, Belgium changed its stamps. The printing became an in-house process by the financial federal government department. The new stamps are still printed with a watermark, but they also have a digital component. The change resulted in a cost reduction by standardizing the sizes and optimizing the production process. Some of the more advanced fiscal marking technologies include embedded threads and watermarks; special inks and coatings, such as so-called invisible inks, holograms and foils; and calculated or changeable content. Because of their enhanced security features, these stamps can be more expensive than traditional stamps. In the state of California in the United States, the traditional stamps cost US$ 0.42 per 1 000 stamps. The cost of the first generation of high-tech stamps was 10 times higher, at US$ 4.77 per 1 000 stamps. This price nearly doubled for the second-generation encrypted stamp, to US$ 8.20 per 1 000 stamps. Nevertheless, California collected 124 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N about US$ 450 million of additional tax revenue in the first decade following the implementation of encrypted tax stamps. This additional revenue was far greater than the costs of implementation and enforcement. Other jurisdictions have also revised their tax stamps to incorporate new technolo- gies. The state of Michigan, for instance, replaced heat-applied cigarette tax stamps with digital pressure-applied stamps in 2015. Michigan deployed a tax stamp with several overt and covert security features and a unique quick response (QR) code and serial number. QR codes (machine-readable codes consisting of an array of black and white squares, typically used for storing URLs or other information) can have purposes beyond tracking and tracing. The QR code can be read by consumers with a smartphone or tablet application to access information on smoking-cessation programs, report violations of the state’s youth access policies, connect to a tip line to report noncompliant packs and learn about the harms from illicit tobacco sales and purchases. Enforcement authorities can validate stamps using the smartphone- based eTRACS (Electronic Tax Reporting and Audit Compliance System). As part of the system’s implementation, the Michigan State Police department created teams of enforcement officers in each of the state’s seven districts and the state Department of Treasury created its own enforcement team. Sources: (37–42). Box 3.5 The ISO standard on excise tax stamps The ISO published its excise tax stamp standard (ISO/TC 292/SC) in October 2018. The purpose of the ISO standard is to assist tax and finance authorities in enhancing compliance with excise tax regulations. A tax stamp is defined as a visible tax stamp, label or mark placed on certain types of consumer goods to show that the applicable excise tax has been paid. The ISO standard applies to tax stamps that are physical in nature – not to digital markings, which are directly printed on to packs without a physical component. “Authentication” in this standard refers to the authentication of the tax stamp, not the product on which the tax stamp is affixed. In other words, authentication of a tax stamp on a cigarette pack means that the tax stamp is authentic but does not guarantee that the pack is authentic. In addition, control measures are needed at the time of the application of the stamp to verify the conformity of the tax stamp with the corresponding product. The standard provides guidance on the content, security, issuance and examination of physical tax stamps used to indicate that the required taxes have been paid and that the tax stamp is authentic. The use of stamps to facilitate tracking and tracing within the supply chain is not described. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 125 Specifically, the ISO standard deals with the following issues: • defining the functions of a tax stamp • identifying and consulting with stakeholders • planning the procurement process and selection of suppliers • the design and construction of tax stamps • the overt and covert security features that provide protection of the tax stamp • the finishing and application processes for the tax stamp • security of the tax stamp supply chain • serialization and unique identifier codes for tax stamps • examination of tax stamps • monitoring and assessing tax stamp performance. A stamp may fulfil many functions, but the core business of tax stamps is to ensure and facilitate the collection of revenue. The tax stamp must use a combination of security features. The tax authority should ensure that the tax stamp can be authen- ticated and that counterfeit, altered, tampered or otherwise fraudulent tax stamps can be detected. The standard provides detailed information on the different components of the tax stamp such as the substrate, inks, adhesives, laminate, authentication or security features and the unique identifier that should enable checks on the payments of the required tax. The process of procurement is discussed in detail in the standard. The tax author- ity should ensure that the procurement process is open, transparent and meets the sustainability objectives. The tax authority should set out the goals and requirements to give tendering organizations more leeway in proposing optimum solutions that might be different from those the authority would specify. The standard is not prescriptive; rather, it provides a catalogue of options. It does not, for instance, recommend specific security features, but it does describe the different types of features that are necessary for a tax stamp to be secure. Tax officials still need to make decisions and choose the option that suits them best, but the standard remains recommended reading for those who would like to introduce tax stamp programmes in their jurisdiction. ISO standards are not freely available but can be purchased at the ISO Store (www.ISO.org) or from an ISO national member body. Source: (43). According to Article 8 of the Protocol, each Party shall require that unique, secure and nonremovable identification markings – such as codes or stamps – are affixed to or form part of all unit packets, packages and any outside packaging of cigarettes for the implementation of the tracking and tracing system within a period of five 126 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N years, and of other tobacco products within a period of 10 years, of entry into force of the Protocol for that Party. The Protocol specifies that at least the following information shall form part of the unique marking: • date and location of manufacture • manufacturing facility • product description • where available, the intended market of retail sale. In several countries, QR codes are used as fiscal markings for tobacco and alcohol tax control. Each stamp has a unique identifier code and a QR code. The data stored in the QR code provide the following product information: • manufacturer • production location • stamp order date • tax status and class • brand • intended market • unique identifier (serial number). KEY TAKEAWAY 10 The use of fiscal markings is generally considered to be an appropriate tool for increasing compliance with tax laws. Fiscal markings can also be helpful for distinguishing between genuine and illicit tobacco products. 3.4.5 TRACKING AND TRACING A tracking and tracing system assists authorities in determining the origin of tobacco products – and the point of diversion, if applicable – as well as monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to enable authorities to have information on all transactions through the entire tobacco product supply chain until duties are paid or other obligations are discharged. Traceability is not used only for tobacco products. It is also used to improve the supply chain function, as in the case of parcel services, as well as for product safety reasons, to manage potential product recalls and for regulatory reasons. Tracking is the process that monitors where a product is at all times while also creating a time and location record for all movements. Tracing is the ability to identify the past locations of a product, so that the product’s route can be followed CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 127 back to its origin (44). In other words, traceability is “the ability to trace the history, application or location of an object” (45). A tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it is possible to unambiguously register that product’s movements. Other necessary characteristics include the ability to share the registered move- ment information and to authenticate products. This enables a product’s status to be captured through the supply chain and its history to be identified and verified retrospectively. According to Article 8.4.1 of the Protocol, Parties should require the following information to be available: • the date and location of manufacture • the manufacturing facility • the machine used • the production shift or time of manufacture • the name, invoice, order number and payment records of the first customer not affiliated with the manufacturer • the product description and intended market of retail sale • any warehousing and shipping • the identity of any known subsequent purchaser • the intended shipment route, date, destination, point of departure and consignee. A good tracking and tracing system enables the government to properly monitor the supply chain, improves its ability to ensure collection of the proper duties and taxes, provides it with the ability to authenticate whether the identification marking is genuine and matches the product and improves its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. The following elements are required for an effective tracking and tracing system (46): • A serialized unique identification marking for each package of product. These identifiers are a distinctive combination of numbers, letters or both. They cannot be predictable or used more than once. The representation of the identifier on the package can be human-readable (letters or numbers) or machine-readable (barcodes). Generation of codes and encryption that are part of a tobacco industry patent should be excluded. • A data carrier with the serialized unique identifier and other information such as date and location of manufacture, manufacturing facility, product description and, where available, the intended retail market. This informa- tion should be readable by authorized agencies of any Party to the Protocol. The data carrier should comply with quality standards and be suitable for 128 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N high-speed production lines. Two-dimensional barcodes, for example, meet these standards and are readable with inexpensive equipment. • A link and parent-child relationship (called aggregation) between different packaging units that offers the option to trace a pallet without the need to scan all the packs and master cases of that pallet. • Recordkeeping of all shipping and receiving events along the supply chain. This includes, for example, the departure location and the arrival location, as well as the involved operators. International standards from the ISO are recommended for the capture and exchange of data and events. • The use of international standards for key information that is encoded in the data carrier (5). An example of a unique and internationally recognized identifier for products is a Global Trade Item Number. The following details on information storage and sharing are drawn from various sections of the Protocol. Data and events along the supply chain must be stored in an independent database that is controlled by competent government authorities. At the global level, national and/or regional databases can be interconnected to facilitate international inquiries by competent authorities. Parties to the Protocol agree to establish a global information-sharing focal point located at the Conven- tion Secretariat of the WHO FCTC, accessible to all Parties, enabling them to make enquiries and receive relevant information. Each Party shall ensure that the information recorded under paragraph 5 of Article 8 of the Protocol is accessible to the global information-sharing focal point on request, subject to paragraph 9, through a standard electronic secure interface with its national and/or regional central point. The global information-sharing focal point shall compile a list of the competent authorities of Parties and make the list available to all Parties. The cost of tracking and tracing systems is a concern for many countries, but as indicated in paragraph 14 of Article 8 of the Protocol, jurisdictions may require the tobacco industry to bear any costs associated with putting in place the tracking and tracing system in a country (46). In Brazil, the cost for cigarette manufacturers was US$ 0.0185 per pack (42). In Kenya, the cost for manufacturers was US$ 0.024 per pack (42). Along with considering the characteristics of a tracking and tracing system in selecting a particular one, it is important to avoid conflicts of interest, ensure fair and transparent dealing with suppliers, implement a zero-tolerance policy for corruption or anti-competitive behaviour and ensure compliance. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, which deals with industry interference, and Article 8 of the Protocol. Article 8.13, which states that “each Party shall ensure that its competent authorities, in participating CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 129 in the tracking and tracing regime, interact with the tobacco industry and those representing the interests of the tobacco industry only to the extent strictly necessary in the implementation of this Article”. Box 3.8 provides a cautionary example of a tracking and tracing system that is not compliant with Article 5.3 of the WHO FCTC. While the Protocol contains a great deal of information on the requirements that a tracking and tracing system should meet, questions come up in relation to the implementation of such systems. To achieve the objectives of the Protocol, the Meeting of the Parties (MOP) to the Protocol, as the governing body of the treaty, has the prerogative to establish subsidiary bodies, such as expert groups and working groups. In decision FCTC/MOP1(6), the MOP established a working group for the development and implementation of tracking and tracing systems in accordance with Article 8 of the Protocol, including the global information-sharing focal point (Article 8.1) and unique identification markings for cigarette packets and pack- ages (Article 8.3), to further elaborate on the next steps. The working group will produce a comprehensive report compiling good practices and experiences on the implementation of tracking and tracing systems, as well as unique identification markings for cigarette packets and packages at national or regional levels. The working group was also given a mandate to prepare a conceptual analysis of how a global information-sharing focal point could be set up. Implementing a complete tracking and tracing system with fiscal markings takes time. In most of the countries that have already implemented tracking and tracing, it took several years from starting with the legal framework to final implementation. Several hurdles need to be overcome: • Legal framework approval is usually delayed by the tobacco industry. • Knowledge of tracking and tracing and associated technologies is scarce at tax administrations. • Tender and bidding processes are complex. • Coordination between domestic tax authorities and customs is weak. Even though the process might be lengthy, the investment in a tracking and tracing system will be repaid with the amount of tobacco taxes that are not lost due to evasion. When implementing a new tracking and tracing system, tax administration should ask for collaboration and technical assistance from intergovernmental organizations and countries that have successfully implemented such systems, in order to speed up and ensure success of the process. Examples of tracking and tracing systems implementation in Chile, Kenya and the EU are detailed in Boxes 3.8, 3.9 and 3.10. For countries that already have fiscal markings in place, the potential interaction between the markings and the implementation of a tracking and tracing system should be taken into account. Further information on this interaction is presented in Box 3.6. 130 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.6 Tracking and tracing and fiscal markings It is becoming increasingly common for stamps to contain some tracking and trac- ing features, such as unique identification markings and basic information on the product that describes the company, tax status or the product itself. The intention is to mark each pack with a unique identification marking so it can be monitored from the point of production to the retailer, including each step in between, thereby creating a complete time and location history. Although a tax stamp could meet the requirements of Article 8 of the Protocol and have tracking and tracing features, in general, the focus of tax stamp systems differs from that of tracking and tracing systems. Tracking and tracing is more than the unique, secure and nonremovable identification markings on the packages of tobacco products. It implies reading or scanning the codes; linking the codes between packs, cartons, master cases and pallets; uploading the information to a database; recording of any shipping and receiving events along the supply chain; and interconnecting the different databases. While new tax stamp programs contain tracking and tracing features, they are primarily intended to facilitate tax collection on the domestic market and not to track duty-suspended cross-border trade or the export of products. The focus of tax stamp systems is on authenticity and the proof that taxes are paid. The focus of tracking and tracing systems is on unique identification and on control of the movements in the supply chain by monitoring and investigating the past and future location of products. Tax stamp programs focus on stock management, verification (that the stamps correspond to the product) and authentication (that the stamps are genuine), while the focus of tracking and tracing systems is on the origin, intended route, first customer and final destination. The focus of tax stamps is primarily on individual packs intended for the duty-paid domestic market, while the focus of tracking and tracing systems is on all packaging (packs, cartons, master cases, pallets) and certainly – but not exclusively – for the duty-suspended export market. Nevertheless, sometimes there can be synergies. For example, the EU countries that require a tax stamp or national identification mark for fiscal purposes have the option to use it as the security feature for tracking and tracing purposes, provided that the requirements are met. In summary, tax stamps can be converted to or be part of a tracking and tracing system when the converted system provides aggregation between packs, cartons and master cases and records all movement along the sup- ply chain. For the export market, a unique identification marking should be added. Sources: (39–42). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 131 Box 3.7 What not to do: use the industry solution to tracking and tracing Codentify is a serialization system used to produce cigarette pack markers for the purpose of verifying whether cigarette packs are legal. It was patented by PMI but subsequently licensed at no cost to other major cigarette manufacturers. In 2016, Codentify was transferred to Inexto, which is an affiliate of the French group Impala. What is the problem with Codentify/Inexto? The main issue is that Codentify/Inexto’s links to the tobacco industry make it incom- patible with the Protocol, which came into force in September 2018. The Protocol specifies that obligations assigned to a Party shall not be performed by or delegated to the tobacco industry. Additionally, many elements indicate it is an ineffective means of authentica- tion. For example, the 12-character digital codes generated by Codentify can be easily duplicated or cloned and used as originals on either a counterfeit or genuine pack, which can then pass the system’s basic verification test. The codes are also produced by relatively unsecured, commercially available equipment and do not include high-security features capable of protecting the authenticity of identifier numbers. Systems that use multilayered, advanced security solutions that enable distributors, retailers, customers and authorities to identify noncompliant products are more secure. Another problem is that Codentify/Inexto cannot track products as efficiently as other available systems. It requires a much larger enforcement capacity to achieve the same detection rates as other systems that are not linked to the tobacco indus- try. Authorities would have to inspect significantly more packs marked under the Codentify system than is necessary under some other systems to achieve the same certainty of not missing a fraudulent pack. In addition, not all stakeholders will be able to verify that a pack marked under the Codentify system is genuine, while other available systems do offer this possibility. Source: (47). 132 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.8 A successful tobacco traceability system: SITRAF, Chile The Servicio de Impuestos Internos (SII, Internal Revenue Service of Chile) has success- fully coordinated and implemented a tracking and tracing system for tobacco products. Application of a compliance management model In the framework of a compliance management model, a traceability system for tobacco products utilizes a structural measure to reduce tax evasion. It is estimated that evasion of taxes (VAT and excise taxes) in the cigarette market in Chile amounts to 16.6% of the country’s total market annually – approximately US$ 300 million. The traceability system implemented in Chile (SITRAF (TAB2)) allows authorities to know in a certain and timely manner the quantities of cigarettes produced or imported into the country. Moreover, it helps authorities to distinguish between counterfeit products and original products that did not comply with payment of the tax. In 2018, the implementation of the traceability system was awarded to a company through public bidding. The company is in charge of the implementation and operation of the system for five years, according to a contract signed with the SII, and must maintain a team of 20 people available for the project. Direct markings are applied to items produced in Chile for national consump- tion, and stamps are used for imported products. For both types of product, the marking is based on a data matrix code, which is printed using security ink that is distinguishable from any other type of ink with specific devices that are provided by the awarded company. Although products for export are not subject to marking, they are controlled and accounted for by the traceability system. For national production, devices are installed on each production line that rec- ognize the type of pack being produced, print a unique code on each pack and then read it (activation) to save all the information on servers located in the production plant. This information is transmitted to the central servers of the system and then to the SII. It is also available for on-site inspection. In the case of imported products, the stamps must be acquired in Chile by each importer and then sent to its producer abroad, which is responsible for adhering them to each pack of cigarettes prior to wrapping the packs with cellophane, us- ing applicators on the production lines. Once the cigarettes enter Chile, the tax determination process has been completed in the service (Provisional Free Transit Guide) and the corresponding taxes have been paid to customs (Import Declara- tion), the importer must enter the data on the stamps used by the importer on the platform of the traceability system. After validation, the stamps can be activated in the system – that is, they are recognized as valid for commercialization. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 133 In addition to the devices provided by the company, SII has developed a smart- phone application for verification by citizens. Although the application cannot verify the authenticity of the ink used, it is able to verify whether a code is correctly gener- ated and display the information contained in the traceability system for the code (brand, variety, quantity of cigarettes, products/importer) so that the taxpayer can verify its consistency. Progress of the compliance management model Implementation of the traceability system has required a coordinated effort both within the service – for the generation of instructions, procedures and computer developments – and with other institutions, such as the National Customs Service and the Ministry of Health. It is an unprecedented project in terms of coordinating the implementation of the system in the production lines of the different tobacco companies in the country and the provider company. Some of the main milestones of the project are: September 2014: Law 20,780 on Tax Reform establishes an obligation to implement the system within a term of six months, after the publication of the resolution determining the obligated tax payers. May 2015: Resolution No. 47 determines obligated taxpayers. June 2015: Circular No. 47 describes obligation to incorporate stamps or distinctive marks as a traceability mechanism. February 2016: Law 20,899 on Tax Reform simplifies the definition of the system, allowing the system to be outsourced or provided by the SII, in addition to making the type of traceability more flexible. August 2016: Traceability system regulation D.S. 1,027 is issued (published on 28 December 2016). March 2017: Exempt Resolution No. 49 of the Ministry of Finance authorizes the SII to outsource all or part of the traceability system. June 2017: Bidding bases in public market are published. February 2018: Tender is awarded to selected company. June 2018: Decision of contract is made by General Comptroller of the Republic. August 2018: Resolution No. 61 determines taxpayers obliged to apply the trace- ability system. August 2018: Holding of first workshops for detailed definitions of the project, with the participation of Customs, Ministry of Health, provider and SII. September – October 2018: Visits of plants to coordinate with producers and define adaptations to production lines for system implementation. January 2019: Resolutions No. 6, 7 and 8 passed, with response to a request for an 134 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N extension of producers; start of system implementation in all production lines in the country. February 2019: Resolution No. 16 establishes a term to commercialize the remaining stock without marking. 2019: Resolution No. 24, with request for extension to importer, includes training of customs staff. March 2019: The traceability system is started up. June 2019: Stamp process begins for all imported cigarettes. Currently, the system is installed and operating in all production lines in the coun- try, placing traceability markings on virtually 100% of the cigarettes produced and imported. During 2019, the traceability system enabled controlling approximately 1 175 million packs: 744 million produced for national consumption, 409 million produced for export and 22 million imported yearly. Source: (48). Box 3.9 Case study of Kenya’s implementation of a tracking and tracing system Kenya’s current tracking and tracing system was preceded by a series of reforms in both tax structure and administration of excise taxes. The reforms included electronic cargo monitoring of exports, which allowed for automatic monitoring and reporting. The system appears to be highly effective because it requires less capacity and is less prone to manipulation than earlier systems. The experience of Kenya shows that a lower-middle-income country can successfully implement a sophisticated system capable of decreasing illicit trade. It also shows the importance of other measures such as strengthening enforcement, increasing cooperation and communication among different agencies and increasing penalties for noncompliance. Illicit trade, as measured by the Kenya Revenue Authority (KRA), was estimated to be around 15% of total consumption in the market during the initial reform period. After the introduction of the new system in 2015, it dropped to 5%. Timeline of the major reforms: 2003: In this period, the paper tax stamps used had a unique identifier and were colour-coded to indicate the type of product. Regular compliance checks were in- troduced. In 2007, the cost of a stamp was 2.124 Kenyan shillings or US$ 0.023 per pack. However, the stamps were found to be easily counterfeited and could not be linked to specific brands. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 135 2008: The KRA proposed a tracking and tracing system and increasing tax rates. The new system was introduced gradually. 2010: Enhanced security features, including ultraviolet markings, were added to the paper stamps. The stamps were to be clearly visible when packs were displayed for sale and placed so that opening a pack would destroy the stamp. The stamps were verified at four different points along the supply chain. The costs were just slightly higher than those of the previous stamps at US$ 0.024 per pack. Licensing was introduced for domestic manufacturers, subject to annual renewal. Importers were required to register with the KRA. Licences required submission of details on the company directors, inventories and equipment, accounting systems, input-to-production ratios and brands produced. Penalties for noncompliance were increased and included up to three years in prison. An electronic cargo tracking system was launched. Electronic seals were affixed on containers or trucks, and GPS technology was used for tracking. A bond was payable on exports to cover excise and VAT taxes. The bond was released only when the goods reached the final destination and taxes were paid. Verification involving both countries of the business deal takes place at the bor- ders. The electronic system provides information about the departure and arrival of the goods and the disarming of the seals. Authorities in the importing country are notified before the shipment leaves the domestic production facility. The system reduces the number of checkpoints and staff needed and generates arrival reports that can be verified with VAT refund requests. As a result of these changes, three factories and seven of the 10 importers were shut down due to noncompliance. Exports to Côte d’Ivoire, Eritrea, Mali and Sudan stopped because companies could not provide evidence that the goods reached the final destination and taxes were paid. More than US$ 11 million in excise tax losses was recovered in 2011. The KRA estimated that illicit trade dropped to 8%. 2013: A contract was signed to introduce a tracking and tracing system for tobacco and alcohol, the Excisable Goods Management System, in April. The system added production counting, tracking and tracing, stock control, processing and other data collection to the existing system. Infrastructure requirements included high-speed broadband internet at production facilities, warehouses, the KRA and ports, along with reliable power or backup generators at those points. Implementation was planned in three stages: • Stage 1 – A new electronic digital stamp with a unique identifier was introduced. It included a data matrix code plus overt markings (holograms, fluorescent fibres, a security link for KRA authentication and visible two-dimensional codes for verification and activation), semi-covert markings (UV features, fluorescent prints detectable by specialized devices, mini text printing for retailers and 136 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N distributors) and forensic taggants for use in prosecutions. The stamp also included human-readable codes for verification by short message service using the KRA web portal. • Stage 2 – Control and monitoring systems were automated in February 2014. Manufacturers had to install photosensitive readers on production lines, with data automatically sent to the KRA in real time. Each stamp was activated and associated with a brand and package size on the line. The KRA database is automatically updated every 15 minutes. • Stage 3 – Market surveillance began, with 83 officers given powers to seize illicit cigarettes and make arrests. The officers were equipped with hand-held devices that transmitted data to the KRA for authentication. Distributors and retailers became liable for selling products without an excise stamp and were subject to fines plus prison sentences of up to three years for noncompliance. In 2016, a smartphone application became available with which the public could authenticate cigarette packs. Importers must now buy digital stamps and send them to export facilities in other countries to be affixed. Tax liability is due at removal from a factory or at import. The electronic cargo monitoring system is still in effect. 2016: The Excise Duty and Tax Procedures Acts clarified new obligations and penalties. 2017: A new integrated customs management system was launched. The KRA estimates that illicit trade levels are now around 5%. The current, more comprehensive digital system is cheaper than the previous paper tax stamp system. Manufacturers pay for the production monitoring system, but it counts as a busi- ness expense on corporate tax returns. In 2018, two manufacturers and 10 licensed importers were operating in Kenya. In 2018, aggregation between the markings of packs, cartons and master cases had not yet been implemented but was expected to be forthcoming. Source: (36) Box 3.10 Case study of the new EU tracking and tracing system, May 2019 Cigarette smuggling and other forms of illicit trade in the EU is estimated to cause a loss of €10 billion in revenue annually. In 2018, 4.2 million packs (20 sticks per pack) of illegal cigarettes were seized by customs in the EU. Illicit tobacco production was also increasing: an illegal factory in Ireland, dismantled in 2018, was capable of producing 250 000 cigarettes per hour. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 137 The EU tobacco control policy is described in the Tobacco Products Directive and is influenced by the Protocol. Article 15 of the Directive calls for the traceability of cigarettes and RYO tobacco products by May 2019 and of other tobacco products by May 2024. The EU tracking and tracing system is sufficiently flexible to be imple- mented at both the regional and the single-country level. Countries can choose among providers as long as the basic requirements are met. The policy provides a high level of protection against any attempts at manipulating the data. The report- ing obligations cover all the economic operators involved in the manufacture and distribution of tobacco products. The EU system requires all unit packets of tobacco products to be marked with a unique identifier Information on the movements of those products is to be stored by third-party data storage providers. The suppliers of the unique identifiers and data stor- age are to be financially and legally independent from the tobacco industry. The data are to be fully accessible to authorities of EU Member States for enforcement purposes. The generation of unique identifiers, as well as all other codes required for pre- registration of economic operators, facilities and machines, will be done at the Member State level by designated identifier issuers. Manufacturers and importers are required to supply information relating to the product and production lines when requesting unique identifiers from the issuers. The issuers will then generate and deliver batches of unique identifiers. On the production line, manufacturers of tobacco products will complete each unique identifier with a marking indicating the date and time. The unique identifier will be a machine-readable, optical, one- or two-dimensional barcode. An anti-tampering device, capable of creating an unalterable independent record of the verification process, must have been installed previously. This additional record will be accessible to public authorities for potential investigation and inspection. Unit packets, as well as aggregated packages such as cartons, master cases or pallets, can be tracked and traced throughout the supply chain. Tracking is also allowed at an aggregated packaging level as long as unit packets remain traceable. During transport, each dispatch and arrival up to the final dispatch to the first retail outlet must be recorded and reported. All recorded information must be submitted to the independent third-party data storage facility, generally within three hours, and 24 hours before dispatch and transloading. Costs, including operational costs, are shifted to the tobacco industry, in line with Article 8 of the Protocol. The EU system of tobacco traceability and security features became operational on 20 May 2019. Sources: (36, 49–50). 138 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 11 A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as monitoring and controlling the movement of tobacco products and their legal status. 3.4.6 ANTI-FORESTALLING “Forestalling” is a term that describes increases in the production or stock of products in anticipation of a tax increase (2). Other terms referring to this practice include “stockpiling” and “front-loading”. Forestalling occurs when manufacturers or importers increase their tax-paid stock or oversupply the market by increasing production or imports in order to pay the previous lower rate. It reduces and delays the effective- ness of tax measures. The effective starting date of the new tax rate will be delayed, revenues will be lower and the possible effect on prices and thus consumer behaviour will also be postponed. To illustrate how anti-forestalling measures function, an example is presented in Annex 3.2. A legal basis must exist for anti-forestalling measures; otherwise, the govern- ment cannot prevent the industry from forestalling. Legal measures to deal with forestalling include (51–52): 1. Limiting the amount of tobacco products that can be released subject to the old tax rate and levying the new tax on the products exceeding that limit. 2. Levying the new tax rate on all goods that are still in stock and not yet sup- plied to the final consumer. 3. Limiting the number of tax stamps issued at the rate that was in effect before the increase or limiting the time that products with a tax stamp with the old rate can be sold. 4. Requiring producers and importers to buy new tax stamps annually or after a tax increase. Under the first three measures, the competent authority determines the limit for taxation at the previous (lower) rate. The quantity allowed may be based on the shelf life of tobacco products – around six months for cigarettes – or normal inventory levels, such as an average over the previous three years. The first measure, limiting the amount of tobacco products that can be released, requires resources from the competent authority for enforcement. Authorities may decide to post inspectors in each production facility, but even without posting inspectors, procedures are necessary for determining when the allowed quantity has been exceeded and what subsequent actions to take. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 139 The second measure, levying the new tax on goods in stock and not yet supplied to the final consumer, might be difficult to implement. The competent authority is required to monitor the manufacturing process and, at the very least, to conduct a stock-taking of all players in the supply chain – including the various manufacturers, importers, wholesalers and retailers – in a very short window of time. If monitoring covers only the stock of manufacturers and importers, this measure could easily be circumvented by ensuring the stock is sold to others in the supply chain or by setting up separate distribution companies to purchase the stock. Controlling stock at the retail level is burdensome and might not be administratively feasible given the large number of cigarette retailers. It becomes even more burdensome if there is no licensing requirement for retailers because they will first have to be identified. The third measure accomplishes the same thing as the first if a country uses tax stamps. The fourth also requires tax stamps and is simpler for the competent authority but somewhat more burdensome for the tobacco companies, since stamps must be purchased every year. Box 3.11 provides examples of anti-forestalling measures in EU countries. Box 3.11 Examples of anti-forestalling measures in EU countries Several EU Member States have taken measures to limit forestalling. A cautious ap- proach seems to be required for designing such measures to ensure that they comply with EU legislation and the general principles of EU law – in particular, the principle of proportionality. No disputes have occurred concerning the right of initiative of EU Member States to implement anti-forestalling measures. Nevertheless, several EU Member States had to defend their measures in front of the Court of Justice of the European Union, the institution that ensures all national legislation is in line with EU law and a consistent application of that law (53). The Court acknowledged that anti-forestalling measures are appropriate to combat tax evasion and tax avoidance. Moreover, the Court emphasized that fiscal legislation is an important and effective instrument for discouraging consumption of tobacco products and therefore for protecting public health (54–55). However, the measures taken should be proportion- ate to the objectives. The principle of proportionality means that only the action needed to achieve the objective should be taken, and it should not exceed what is necessary. This principle regulates the measures taken within the EU and is included in the Treaty on the European Union. The Court demanded that Portugal amend its legislation to ensure compliance with the principle of proportionality. Belgium, Estonia and Hungary were also urged to change their anti-forestalling measures to bring them in line with EU legislation (56–57). 140 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The majority of the EU Member States have anti-forestalling measures in place, but there is no harmonization of these measures. The following are some examples:12 • Portugal limits the quantity of cigarettes that can be released in the last four months of a year to the average of the previous 12 months plus 10%. In addition, manufacturers and importers must sell cigarette packages with a tax marking of the preceding calendar year within three months. For other tobacco products, longer limits apply. • Denmark limits the number of tax stamps issued before a tax increase at the old rate to 20% more than are usually purchased in the two months before the end of the year. • In Poland, tax stamps are valid only for the current calendar year, and cigarettes with the old stamp can be sold only through February of the following year. • In Romania, companies must apply for approval to release for consumption from the customs office. Source: (58). KEY TAKEAWAY 12 Forestalling reduces and delays the effectiveness of tax measures. Implementing anti-forestalling measures can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. 3.4.7 ADDITIONAL NATIONAL AUDITS AND CONTROLS In addition to the previously described measures, several periodic audits and controls could be implemented to increase compliance with tax laws. The most common audits and controls are the following: • Cost audit – The cost audit method provides expected VAT and tobacco tax collection by simulating the intermediate and final cost of cigarettes. It starts with the inventories of raw materials and estimates added values and final cost, then matches the results with real collection from the tobacco supply chain. Annex 3.1 provides more information about the components that make up some selected (tobacco) products. • Transfer pricing audit – To ensure companies pay their fair share of tax, prices of transactions between related companies should be assessed, and when prices are not in line with the market conditions, they should be corrected. Companies that operate at the international level (transnational companies), including many tobacco companies, can manipulate import or export prices 12 Considering the frequency of court cases concerning anti-forestalling incidents, these measures might have been replaced or amended. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 141 of merchandise or raw material to related companies or branches in other countries, with the objective of lowering profits in countries with higher tax rates, and can transfer those profits to countries with lower taxes. • Price and market monitoring – Retail price surveys can provide information about variance from the market price in certain locations, highlighting areas of potential tax avoidance or illicit trade. Physical control of such locations requires rapid response teams, as implemented in the Philippines. To monitor the tax compliance of its taxpayers, the competent authority in the Philippines needs to understand the tobacco market; it must have information on brands, market segments and prices of products. This information enables authori- ties to estimate the impact of tax and price changes on consumer behaviour and revenues. Market data can be analysed as part of risk management and anti-fraud analysis to determine whom to investigate for noncompliance and when to do so. Sales data can be triangulated to validate other data sources, such as household surveys on prevalence. Market data and trends are also useful indicators for determining whether there is a case of oversupplying. See also section 3.4.3. • Consumer control – Involving the public via awareness campaigns has also been shown to be effective. Consumers have the right to be assured that the products available in the market are authentic and come from legitimate sources. Thus, it is in the consumers’ interest to understand and be able to verify that they are buying genuine products. The features of the fiscal marks on tobacco products should help consumers distinguish between genuine and illicit products. Some countries – Kenya, for example (see the case study in Box 3.9) – use a smartphone application to allow anyone to check both covert and overt features and to report any cigarettes with incorrect markings. Other countries, such as the Netherlands, have developed a smartphone application that allows anyone to report a suspected case of excise tax fraud. • Cross-check controls – Competent authorities should consider using multiple sources to obtain market data and determine if these data are consistent with tax declarations. VAT declarations can be used to verify that suppliers and purchasers of raw materials and final products are reporting the same amounts. Bank information can be used to verify both sides of transactions along the supply chain. Any discrepancy can alert the competent authority to conduct further investigation for possible illicit trade of tobacco or tax evasion. 142 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 13 Several different types of periodic audits and controls that can be carried out to increase compliance, including cost audits, transfer pricing audits, price and market monitoring, consumer controls and cross-check controls. 3.4.8 IMPORT AND EXPORT CONTROLS Parties to the Protocol should allow import and export of tobacco products and manufacturing equipment only by duly licensed natural persons or legal entities (Article 6.1). A well-known strategy used by fraudsters is to declare products for export so that no duties are due to the country of export. These products are subse- quently transported through other countries, using the in-transit regime that allows temporary suspension of duties until the goods arrive at their final destination. Before arriving at their final destination – where the excise duties would be due – the goods disappear or are lost while being diverted to the illegal supply chain. The goods may never leave the country, or they may be smuggled back into the country from which they were exported without declaring or paying duties. This risk of loss of revenue can be mitigated by requiring a guarantee or bond, which will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless the required fiscal marking (such as a tax stamp or export label) is affixed on the pack, according to the national law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Illicit tobacco trade could be decreased significantly if the various competent authorities that have jurisdiction over manufacturers and exporters of tobacco prod- ucts and manufacturing equipment would provide the competent authority at the destination with prior information when a shipment has been authorized and is about to take place. The information could include the name of the consignee, a description of the item shipped and the quantity. Also, the competent authority at the destination should inform the competent authority having jurisdiction over the shipper that a shipment was received along with the pertinent information relating to the shipment. A good IT system is also required for import and export. Electronic processing of prior-to-arrival manifest and import declarations is recommended. Most countries have implemented an online customs system to process import and export declara- tions, including all required data such as country of origin or export, description of merchandise, value, weight, cargo insurance, carrier, importer or exporter and broker identification, detailed tax duties to pay and final destination. The World Trade Organization (WTO) Trade Facilitation Agreement provides sev- eral tools for better controls, including collaboration between customs administrations, CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 143 exchange of information, use of non-invasive devices and prior-to-arrival import declaration (23). Non-invasive detection equipment at customs posts is highly capable of detect- ing contraband merchandise. The most common tools are X-ray scanners that are used for small parcels, containers, trucks and trains. Most modern ports have also implemented the use of X-ray scanners, and such technology is improving the speed of controls to as little as two minutes per container. Although the cost of scan- ners is declining, it remains inaccessibly high for countries with limited resources. Fortunately, scanners are often available for lease, making them accessible for tax administrations in those countries. Less sophisticated and less costly detection equipment includes endoscopes, mirrors, night vision equipment, cameras and automatic licence plate readers (33). A still less expensive alternative is the use of dogs, which can be trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied in order to reduce contra- band. These include the separation of processing operations from the sealed storage of taxed and untaxed products. Physical and direct control by officials of the excise authority during a part or the whole of an operation can be applied (for example, physical escort of the transit consignment from border to border by individual trucks or in a convoy, or application of radio or satellite tracking systems such as GPS-enabled devices to goods, conveyances, vehicles or containers). Control at borders is essential and should include integrated technology and cooperation with agencies at the border station. Frontline officers should be sup- ported by appropriate intelligence, guidance and supervision from management, as well as technical aids to enforcement. Within a country, mobile excise control units are helpful for verifying excisable goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. These operations require close coordination between police, border guards and other public services. Exports also require special attention, in particular if VAT and tax refunds are granted to the export of tobacco products. Validating the real exit in the declared amount is essential to avoid illegal re-entry to a territory and the improper refund of taxes. For any tax refund, an audit including tax credit information must be carried out. The audit may include the invoices for the whole chain involved in the export, including tobacco farmers, first processors, manufacturers, wholesalers, storage and transport. 14 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 14 To ensure control of import and export, it is recommended that only duly licensed persons and entities be allowed to import and export tobacco products and manufacturing equipment. 3.4.9 FREE ZONES AND TRANSHIPMENT POINTS The term “free zone” is very broad and can refer to a number of different types of areas. The Financial Action Task Force listed the following types in its 2010 report (59): free trade, export processing, enterprise, free ports, foreign trade, special economic zones and bonded warehouses. A number of these areas can include tobacco manufacturing and trade. By definition, controls such as regulation and oversight within free zones are less strict than in other areas. This can make them appealing to persons involved in illegal cigarette manufacturing or trade (17). In fact, illicit activities related to free zones (not limited to tobacco) are regularly documented by organizations that recognize the linkage. These activities include money laundering, tax evasion and trade in counterfeit goods or other illicit goods (60). A report from the European Parliament (61) referring to free ports in particular, mentioned that the motivation for using them included a “high degree of secrecy and deferral of import duty and indirect taxes”. The report even proposed the “urgent phasing out of free ports”. In the European Parliament report, free ports are free zones that function as (semi-) permanent storage areas for high-value goods. The Protocol includes a time-bound provision of effective controls on all manu- facturing and transactions of tobacco products in free zones (Article 12). Free zones are defined as a part of the territory of a Party where goods are considered to be outside the customs territory for import duties and taxes (Article 1.5). This is the same definition used in the International Convention on the Simplification and Harmonization of Customs Procedures (Revised Kyoto Convention) (62). Parties to the Protocol must implement effective controls in free zones within three years of entry into force of the Protocol. For countries not yet Parties to the Protocol, stringent controls of manufacturing and transactions involving tobacco products in free zones are an important component of an effective and efficient tax administration. One of the measures for dealing with free zones within the Protocol includes implementing “effective controls on all manufacturing of, and transactions in, tobacco and tobacco products, in free zones, by use of all relevant measures as provided in this Protocol”. As indicated in an Interpol report (17), a significant vulnerability of free zones is the fact that different economic operations (e.g. manufacturing, assembly, re-packaging and warehousing) take place outside the control of authorities. It is therefore essential for customs administrations to exercise their authority in free zones CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 145 to effectively identify and fight illicit trade in tobacco products. Relevant measures listed in the Protocol should be applied. These include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a track- ing and tracing regime. Removing exemptions on excise taxes is an additional way to increase control and remove incentives for using free zones as a means for tax evasion. Indonesia, for example, imposes excise taxes on cigarette manufacturing in its free trade zone.13 Parties to the Protocol shall also prohibit the intermingling of tobacco products with non-tobacco products in a single container or any other such similar transporta- tion unit when removed from free zones. Finally, each Party is to “adopt and apply control and verification measures to the international transit or transhipment of tobacco products and manufacturing equipment in conformity with the provisions of the Protocol”. Article 13 of the Protocol, which covers all duty-free sales of tobacco products, requires Parties to the WHO FCTC to consider prohibiting or restricting the sale to or import by international travellers of tax-free or duty-free tobacco products, as mentioned in Article 6 of the WHO FCTC. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect government revenues by creating a loophole in the tax structure (2). KEY TAKEAWAY 15 Customs administrations should exercise their authority in free zones to prevent different economic operations from taking place outside the control of authorities. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a tracking and tracing regime. 3.4.10 PROCEDURES AFTER DETECTING ILLICIT TRADE OF TOBACCO The procedures described in previous sections are intended to increase compliance and to prevent illicit trade. When smuggling or illicit trade is detected – through, for example, audits, tracking and tracing systems, verification of declarations or border control – actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. To deter further illegal behaviour, a comprehensive audit of everyone and everything involved in the illicit acts must also be carried out. Assets and vessels involved in the illicit activity can be seized, and financial accounts can be frozen. Some countries, including the United Kingdom, Canada and Chile, have also adopted a strategy known as “follow the 13 Indonesian Ministry of Finance, personal communication, January 2020. 146 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N money” to obtain more information on those who finance illicit trade. This strategy is intended to have a further-reaching effect by targeting those who finance the transport, production and storage of illicit products. The United Kingdom, Canada and Chile all created special teams tasked with identifying and targeting the offenders. Furthermore, Article 18 of the Protocol provides for the confiscation and destruction of tobacco, tobacco products and manufacturing equipment. One of the difficulties faced by competent authorities in exercising the authority to seize and forfeit products and/or equipment used in the manufacture or distribution of tobacco products is the cost of keeping or storing the goods and/or machinery before destruction. Thus, the law should also provide for a mechanism and timetable for the disposal and/or destruction of seized and forfeited goods or machinery, while prescribing a mechanism by which these properties can still be presented as admissible evidence in a judicial proceeding. Boxes 3.12 and 3.13 provide examples of successful efforts to combat illicit trade in the United Kingdom and Indonesia. KEY TAKEAWAY 16 As soon as smuggling or illicit trade in tobacco products is detected, actions such as collecting taxes and seizing and destroying smuggled and/or illicit tobacco must be taken. Box 3.12 The United Kingdom’s experience in fighting illicit trade in tobacco products In 2000, illicit cigarettes accounted for 22% of the cigarette market in the Unit- ed Kingdom. To deal with the problem, Her Majesty’s Customs and Excise14 implemented a major anti-smuggling effort. The strategy was refreshed with ad- ditional resources and measures in 2011 and reviewed in 2015. The result was a steady decline in the illicit cigarette market to 10% by 2013/2014. The measures taken were comprehensive and included hiring 1 000 new customs officers and investigators. In addition, tobacco supply chain legislation was introduced, aimed at discouraging tobacco manufacturers from facilitating smuggling. Tougher sanctions included increased fines of up to £5 million levied on a manu- facturer, criminal prosecution with sentences up to seven years, confiscation of assets as part of the proceeds of the crime, payment of duty on the confiscated goods plus penalties up to 100% of the duty, prohibition of the sale of tobacco products for 14 By the time of the renewed strategy, the respective bodies were the HMRC and the United Kingdom Border Agency. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 147 up to six months, unlimited fines for selling tobacco without the United Kingdom duty-paid fiscal mark after 13 March 2015, referral for withdrawal of the transporter’s licence, use of immigration sanctions to refuse entry to the United Kingdom for tobacco smugglers and civil action, including bankruptcy. The cost of these measures was £209 million over the first three years of the program and around £100 million annually by 2008/2009. This figure covers only HMRC and excludes any costing of the United Kingdom Border Agency. In 2013/2014, tobacco tax revenues were £9.5 billion. After a review of the strategy in 2015, the controls on raw tobacco were strength- ened by the introduction of an approval system in 2017. Anyone who manufactures, purchases, acquires, owns or is in the possession of a tobacco products manufacturing machine must be licensed with customs as of 1 August 2018 (63). The United Kingdom ratified the Protocol on 27 June 2018. It was the 40th country to ratify, which was the trigger point for the Protocol to enter into force. Fig. 3.4 Estimate of the illicit cigarette market and United Kingdom tax-paid consumption Sources: (16, 64). Bi lli on c ig ar et te s UK tax paid consumption Illicit market 20 00 -01 20 01 -02 20 02 -03 20 03 -04 20 04 -05 20 05 -06 20 06 -07 20 07 -08 20 08 -09 20 09 -10 20 10 -11 20 11 -12 20 12 -13 20 13 -14 20 14 -15 20 15 -16 20 16 -17 20 17 -18 0 10 20 30 40 50 60 148 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.13 A success story: Indonesia reduced illegal cigarettes from 12% to 3% of the market In some countries, most of the illegal cigarettes are imported, but in Indonesia most of them are produced within the country by unregistered manufacturers that are usually home-based and relatively small. In several territories, specifically on the Island of Java, illegal cigarettes have been produced for generations. This practice is supported by the availability of raw tobacco materials and cloves, as well as cheap labour costs, especially for female workers. Indonesia also produces cigarette products that are not available in most other countries. These products – known as handmade clove cigarettes (sigaret kretek tangan) – contain cloves, and the production process covers blending, rolling and packing. Parts of the process are done by hand, and 99% of the labourers are women. The strategy for combatting the illegal cigarette trade in Indonesia is divided into two main parts: preventive actions and responsive actions. Indonesia’s success in effectively tackling illicit trade is attributed to the following key factors. Monitoring and surveillance Preventive actions consist of administrative measures – such as issuance of permits and the excise stamp purchasing mechanism – that use risk management by optimizing the Excise Service Information System (ExSis). With this IT system, the Directorate General of Customs and Excise (DGCE) can oversee both daily transactions and daily production from factories. When information of suspicious activities is obtained, DGCE can suspend the purchase of excise stamps. The efforts to fight the illegal cigarette trade also invite stakeholders to be involved by supplying information regarding high-risk areas and regional governments. Strategic communications and community involvement DGCE continuously disseminates information and conducts public education to fight illegal cigarettes. These efforts are conducted every year, using a special campaign slogan. In 2019, the slogan was “Gempur Rokok Ilegal” (“Fight Illegal Cigarettes”). Key performance indicators for DGCE units and offices Parallel with the above-mentioned preventive actions, DGCE also continuously conducts responsive actions: enforcement, investigation and audit activities in cigarette factories. Enforcement activities are planned and measured by consider- ing the limited human resources and the large scale of the monitored territories. To demonstrate the effectiveness of administrative and enforcement measures in curbing the trade of illegal cigarettes, both types of activities are translated into key CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 149 performance indicators for all DGCE working units and offices, including regional offices and personnel. Use of technology and intelligence Coordination between different DGCE offices responsible for monitoring the produc- tion and marketing of cigarettes is maintained by sophisticated IT applications that enable efficient distribution of information and investigation activities. The applica- tions are the Customs Intelligence and Tactical Centre for data and analysis and the Centre for Command and Control (Pusat Komando dan Pengendalian/Puskodal) for ensuring that sea patrols work effectively and efficiently. Independent evaluation To evaluate the efforts and activities to reduce the circulation of illegal cigarettes in Indonesia (e.g. cigarettes without stamps, with fake stamps or with used stamps), in 2016, the government commissioned the University of Gadjah Mada in Yogyakarta to conduct a survey using a stratified random sampling method. To maintain objectiv- ity and independence, an independent body from this well-known university was appointed to conduct the survey. The survey results showed that the level of illegal cigarette circulation in Indonesia was 12.1% of total consumption. In 2018, the DGCE commissioned the University of Gadjah Mada to conduct another survey. Results showed that circulation of illegal cigarettes had been reduced to 7.0%. In 2019, using the same method the university used, DGCE conducted a survey that showed a reduction to 3.0%. Fig. 3.5 shows the results of the surveys. Fig. 3.5 Share of illicit trade in total cigarette consumption in Indonesia, 2016–2019 Source: Customs and excise department, Ministry of Finance, Indonesia, personal communication, 2020. 2016 2017 2018 2019 12.1% 10. 9% 7% 3% 150 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The results of the actions taken can be used as feedback for DGCE in developing excise policies for both service and enforcement, including policies for excise tariffs. DGCE will provide recommendations for an optimum policy format for controlling consumption, maintaining labour protections, optimizing revenue and, most impor- tantly, constantly reducing the consumption of illegal cigarettes on a national level. 3.4.11 PENALTIES Penalties and sanctions must be sufficient to deter illegal activities. Otherwise, finan- cial penalties may simply be paid as a cost of doing business while the illegal activity continues. The Protocol specifies commitments for Parties and provides information on best practices for non-Parties. Article 14.1 in Part IV of the Protocol requires each Party to establish unlawful activities, including manufacturing, wholesaling, broker- ing, selling, transporting, distributing, storing, shipping and importing or exporting tobacco products or manufacturing equipment without the payment of applicable duties or taxes or without using fiscal stamps or other required markings or labels. Articles 14.2 and 15 of the Protocol mandate Parties to determine which of the types of unlawful conduct set out in Article 14.1 shall be criminal offences. Parties must adopt legislative and other measures to give effect to such determinations, as well as to define whether the liability for committing illicit trade in tobacco is a criminal, civil or administrative offence. Article 17 further provides that the Parties shall consider adopting measures as needed to authorize competent authorities to levy penalties in an amount proportionate to lost taxes and duties resulting from the commission of illicit trade. Box 3.14 provides a case study of how Colombia used penalties to fight illicit trade. Box 3.14 The use of penalties to combat illicit trade in Colombia In 2017, the specific tax on cigarettes in Colombia was doubled, increasing from COL$ 700 per pack in 2016 to COL$ 1 400 in 2017. The tax rate was tripled from 2016 levels in 2018, reaching COL$ 2 100 per pack. A provision was added to increase taxes annually after 2018 at the rate of inflation plus 4%. In 2015, before the tax increase, Law 1762 introduced a number of measures to fight illicit trade more effectively. The length of imprisonment for dealing in contraband cigarettes was increased from 3–5 years to 4–12 years. Moreover, government officials who facilitate illicit trade – or anyone involved in transporting or retail sales of illicit tobacco – face similar prison terms. The law allows vehicles used for smuggling to be confiscated, and penalties were increased for illicit trade that is conducted through areas such as special economic zones. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 151 Under the law, illicit trade is considered to be a source of money laundering, which means that the Financial Intelligence Unit can use the same methods it uses to investigate other illegal financial activities. This practice is not common. The law specifically created new sanctions related to alcohol and tobacco excise tax evasion, including the seizure of goods, fines, closure of retail outlets and the suspension or cancellation of licences, authorizations or registries. Arrests and seizures have increased under the new law. Indeed, since its enactment, law authorities reported that between 2016 and 2018, five criminal organizations were dismantled, 53 individuals were apprehended and 72 assets were confiscated. In addition, 2 236 individuals were apprehended and 503 vehicles transporting smuggled goods were confiscated, as transport of such goods is now also considered a crime under the law. More importantly, thanks to the large tax increases, consumption decreased while revenues increased substantially in 2017 and 2018 (see Fig. 3.6). It is estimated that illicit trade in cigarettes in five Colombian cities in 2016 constituted 3.5% of total consumption, a much lower estimate than the industry data suggest. In 2017, after nine months of the tax increase implementation, a similar study found that illicit cigarettes remained low, at 6.4% of total consumption. Fig. 3.6 Packs sold and tobacco tax revenue before and after the tax increase in Colombia, 2016–2018 Sources: (65–67 and Ministry of Finance, Colombia (Direccion de Apoyo Fiscal), personal communication, 2020). 2016 2017 2018 Cigarette sales, million packs Cigarette excise revenue, million US$ current 0 100 200 300 400 500 600 700 800 673.7 521 446.4 195.1 301.2 386.2 33.7% reduction 97.9% increase 152 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N For consumers in possession of illicit tobacco, the minimum penalty should be confiscation and destruction of the illicit tobacco products found in their possession, with payment required for the unpaid tax and duties on those products. In the state of California in the United States, it is illegal to possess a tobacco product on which taxes are due and not yet paid. The burden of proving that taxes have been paid is on those who have the products in their possession. The provi- sion is enforced by the California Department of Tax and Fee Administration and local law enforcement agencies. A violation is a misdemeanour, with a maximum fine of US$ 5 000 and/or up to one year in prison. Illegal packages are subject to seizure and forfeiture. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are using this type of legislation to address illicit trade of tobacco. Asset confiscation and increased penalties for involvement in illicit trade are becoming more common as well. Withholding or even confiscation of trucks involved in smuggling is also common in several countries. KEY TAKEAWAY 17 Penalties and sanctions imposed should be sufficient to deter illegal tobacco trade activities. Penalties should be levied in amounts proportionate to lost taxes and duties resulting from illicit trade. 3.5 TAX ADMINISTRATION OF OTHER TOBACCO PRODUCTS In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. However, there is a lack of standard- ization of other products and sometimes large informal markets. For example, it is estimated that two thirds of waterpipe tobacco in the EU is non-duty-paid (19). Other tobacco products – such as bidis in South-East Asia, waterpipe tobacco in the Eastern Mediterranean region and snus in Sweden – are considered part of a country’s traditions. This sometimes leads to situations where governments are hesitant to strongly regulate and tax these products. Some products, such as kreteks (clove cigarettes) in Indonesia and bidis in India and Bangladesh, are mainly sold in one market. Other tobacco products are more likely to be produced by hand on a small scale, making it difficult to detect and collect taxes on them. The same applies to RYO tobacco, which can be produced on a small scale by hand or with the use of small machinery. The trade in raw tobacco and small-scale home production of tobacco often take place outside of monitoring and control systems (19). As mentioned in section 3.3.1, countries have found various solutions to address this problem, including prior approval for purchase or sale of raw materials and reg- istering, authorizing or licensing of all operators and growers that handle raw tobacco. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 153 3.5.1 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In principle, adding a product to an existing tax framework is not likely to impose significant costs. It is reasonable to expect that challenges similar to those faced in dealing with conventional tobacco products will be faced in the collection of taxes on new products, as market players will attempt to use loopholes in tax regulation to avoid or evade taxes whenever possible. However, new challenges are expected to arise when those new products involve rapidly changing technology and where their market dynamics are widely unknown. Furthermore, taxation of new tobacco products may require additional capacity, as a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. 3.5.2 HEATED TOBACCO PRODUCTS (HTPs) Many countries apply a specific excise on tobacco products according to tobacco weight (see Table 2.4 in Chapter 2). With HTPs, this is likely to impose a challenge, since assessing the content of tobacco in a heated tobacco stick will be an additional burden. From a tax administration perspective, it will be easier for authorities to apply taxes per stick or per unit, as is done for cigarettes. 3.5.3 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS) PRODUCTS Some countries tax only nicotine-containing e-liquids while others tax both nicotine- and non-nicotine-containing e-liquids. Taxing only nicotine-containing e-liquids re- quires laboratory capacity to detect the presence of nicotine (see Table 2.5 of Chapter 2). Self-declarations by industry are not sufficient, since some e-liquids labelled as nicotine-free have been found to contain nicotine (see section 2.4.2). Therefore, it is simpler to tax both nicotine- and non-nicotine-containing e-liquids. One likely challenge of taxing all e-liquids will be the capacity to detect and differentiate whether the e-liquids used in ENDS/ENNDS are falsely declared as being for other purposes at the import and manufacturing levels. More information on advantages and disadvantages of different excise tax policies is given in Table 2.6 of Chapter 2. The challenge in taxing the other components of ENDS/ENNDS products is their diversity (see section 2.4.2 of Chapter 2) and the possibility that some parts may be used for other purposes (e.g. in batteries). As indicated earlier, rapidly changing technology and the lack of control and knowledge of the market make taxation of ENDS/ENNDS devices challenging. It may be for this reason that the majority of countries that tax those products address only the e-liquids. When applying a tax on these newer products, countries should be aware that many customers buy their products online. It is therefore recommended that countries 154 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N deciding to tax these products draw up a proper implementation plan, including how taxes will be collected on imported products and online sales. Online cross- border sales of tobacco products are not permitted in all countries. Several EU countries have banned such sales, which include online sales from retail outlets to consumers in another country. Of course, such bans makes sense only if there is also capacity to enforce them. As in the case of taxing tobacco products, the following actions will be important to more effectively impose taxes on these products: 1. implementing strong enforcement mechanisms such as licensing, recordkeep- ing and control of the supply chain, which can include but is not limited to: a) imposing strict licensing of retailers, importers and manufacturers; ideally, licensing of all those involved in the supply chain and developing a tracking and tracing regime for ENDS/ENNDS products and HTPs (to share cost, this can be done in tandem with the system developed for cigarettes); b) exercising the right to set the frequency and type of audits or controls; c) exercising the right to confiscate goods; and d) imposing sanctions such as penalties, fines and/or withdrawal of licences (if applicable) if legislation is not respected. Specific to ENDS/ENNDS products: 2. implementing highly consequential sanctions for producers who declare nicotine-containing e-liquids as “non-nicotine-containing”; and 3. requiring a fee (contribution to the costs) for laboratory tests when a new product is brought on the market or when there are significant modifications to an existing one. More information on the policy options to apply excise taxes on ENDS/ENNDS is given in Chapter 2, section 2.4.2, Table 2.6. KEY TAKEAWAY 18 In principle, the administration of taxes on new and emerging nicotine products and tobacco products should be similar to that for cigarettes. Due to the lack of standardization of these products, however, a rapid and constantly evolving understanding of them and their supply chain will be required to achieve effective and efficient administration of taxes. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 155 3.6 THE BROADER ELEMENTS OF A GOOD TAX SYSTEM 3.6.1 PROPER RESOURCING OF COMPETENT AUTHORITIES In addition to legal tools and a legal basis on which to act and enforce, the authori- ties in charge of implementing excise tax laws should be provided with sufficient resources to hire the necessary staff to properly implement and enforce them. The necessary staffing could encompass multiple agencies and will often require coop- eration between agencies, since some aspects – such as regulation, licensing and border control – may be performed by agencies other than the competent authority. Staff of competent authorities need the necessary tools, equipment, training and supplies to carry out their functions. This requirement includes the means to build or purchase and maintain a software system that will allow taxpayers to submit required information electronically. Electronic filing has benefits for both taxpayers and authorities. It minimizes the compliance cost for taxpayers and can therefore support voluntary compliance (68). To identify risks of noncompliance, the software system should offer competent authorities the ability to analyse the data submitted by taxpayers and cross-check it with data from other taxes – such as VAT – and third-party sources, such as banks and household surveys. Another option for authorities is to make more efficient use of existing resources. For example, an authority could optimize the risk management system by switching to a risk-based approach: resources could be saved by auditing taxpayers who are more likely to be noncompliant based on risk analysis rather than auditing all of them. Other problems that challenge the effective functioning of a competent authority are lack of a coherent strategy and problems with professionalism related to lack of training or corruption (69). Having a strategy avoids directing resources towards less- important areas. The strategy should always be aligned with the objectives, so that com- petent authorities can identify which steps they should take and in which order they should take them to reach these objectives. A strategy is indispensable to prioritizing and organizing resources so that identified issues or risks can be addressed efficiently. 3.6.2 CORRUPTION Competent authorities should implement tax laws with integrity and have strict rules and regulations for detecting corruption. Strict rules and regulations should also be in place for the punishment of both agency personnel and taxpayers who engage in corrupt practices. Corruption within a competent authority results in the improper monitoring of tax compliance and is one of the causes of the proliferation of illicit trade in tobacco products. It also erodes confidence in competent authorities and ultimately in governments overall. In addition to effective laws and regulation, strong internal audits covering prevention, investigation and sanctions should be 156 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implemented. To improve prevention, a risk map should be created that highlights areas of misconduct and possible leakages. An action plan to update controls should be established to improve areas of weakness detected in procedures and systems. The audits should also be scheduled regularly. Prepared internal auditors with pow- ers to conduct investigations are necessary. Sanctions for corruption, including administrative sanctions and criminal prosecution, must be strong. 3.6.3 A STRONG JUDICIARY The judicial system should be honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. The appeals process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially in the context of illicit trade. KEY TAKEAWAY 19 Broader elements of a good tax system include (1) proper resourcing of competent authorities to hire staff and obtain necessary equipment and systems, (2) strict rules and regulations to detect and punish corruption among both agency personnel and taxpayers and (3) ensuring that the judicial system is honest and independent, with disputes being solved as quickly as possible. 3.7 CONCLUSIONS Policies are more effective if they are properly implemented and enforced. Com- petent authorities have a key role in the achievement of financial and public health objectives of excise taxes. Given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade, this chapter draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (i.e. the Protocol). The Protocol provides a blueprint of measures to address the problem of illicit trade and can be used as a model even by countries that are not Parties to it. Qualities of an effective and efficient tax administration include institutional ar- rangements where roles and responsibilities of competent authorities are clearly defined to avoid overlap and voids. Additionally, effective collaboration among relevant bodies must be facilitated. At the national level, within any organizational arrangement, it is vital that agencies cooperate and exchange information and that their competences find their basis in law. A legal basis for exchange or access to information between government bodies should be ensured. At the international level, especially for border control, the role of customs is key, and access to international CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 157 cooperation agreements such as the Protocol is very useful. An organizational tax administration structure must include a system of performance evaluation and accountability through pre-defined key indicators. To ensure compliance, the accuracy of information for the tax compliance cycle is key, including clear and straightforward taxpayer registration and licensing, declara- tion, recordkeeping, warehousing, distribution, collection and tax refund processes. • Licensing is a powerful tool for obtaining information and securing the supply chain of tobacco products. Ideally, all persons involved in the growing of tobacco and the retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. • Collecting as much information as possible on the business of tobacco and recording all transactions are key to reducing tax evasion, but this may be burdensome for authorities. The use of IT for periodic tax declarations, ac- counting, inventory and financial information is critical for obtaining accurate information and can help decrease the cost of the whole reporting system. • Recordkeeping should be ensured. All persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of those products. • Maintaining a system of authorization for warehousing allows the authori- ties to carry out controls in production and storage facilities to ensure that taxes are paid. Ideally, bonded warehouses should be eliminated from the supply chain. • Duty suspension – which is often applied during the producing, processing, holding, receiving and dispatching of excise goods – should be granted only if strict criteria are met (e.g. for granting authorization, warehouse pre-authoriza- tion visits, adequate stock control measures, checking the origin of excise prod- ucts and the entire production process and coding and marking of products). • To limit the number of taxpayers a competent authority has to manage, tax collection should take place close to the point of production and import. • Refunds for VAT, excise taxes and customs duties are common in most coun- tries, under the principle that taxes are not exported. The refund process must be closely monitored to avoid opportunities for tax evasion. Control and enforcement – key components of tax administration – include a number of measures to secure the supply chain: licensing and due diligence, fiscal markings, 158 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N track and trace, anti-forestalling measures, audits and controls, import and export control and attention to free zones and transhipment points. Control and enforce- ment need to be included as pillars in the strategic plan of the tax administration. Enforcement and control plans must be designed to define the activities and taxpayers that are subject to enforcement and to allocate staffing, auditing, infrastructure and IT resources. Targets must be defined, including the number of interventions and any additional collection or reduction of tax evasion. This includes choosing interventions for those who have a higher probability of noncompliance (the risk- based approach). In the tobacco supply chain, import, export and transfers to and from warehouses may be areas at greater risk of noncompliance. • Licensing provides timely and accurate data that can serve as the basis for audits, since it identifies and controls legitimate operators. The process of licensing control must be carried out and updated periodically – in particular, by controlling the validity of bonds or guarantees and the proper functioning of the required systems and recordkeeping. Where licences are required, the law should include a provision that disallows purchases from unlicensed suppliers or sales to unlicensed purchasers. This means that both suppliers and purchasers will need to verify those with whom they are doing business. This will substantially help to reduce the burden of proof for authorities. In addition, to maintain a high level of control, the validity of licences should be limited in time, making renewals or reapplication required. • Another important measure for controlling and monitoring production and import of tobacco products is the use of fiscal markings (e.g. tax stamps). In addition to increasing compliance with tax laws, fiscal markings can help distinguish between genuine and illicit tobacco products. The use of fis- cal marks enables both the competent authority and the public to monitor whether the taxes on tobacco products have been properly paid. In addition to locally produced and imported products, tobacco products for export should also be required to be marked, but with an indication that they are for export. Requiring a standard package size can facilitate the application of fiscal markings. To lower the chance that fraudsters attempt to re-use fis- cal markings (in particular, stamps) the marking should be applied to each pack of cigarettes (and other tobacco products) before the pack is wrapped with cellophane. Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Fiscal markings should include several security features to make them more difficult to counterfeit. These can include overt, covert, semi-covert and/or forensic features. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 159 • Tracking and tracing systems assist authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as in monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to provide authorities with information on all transactions throughout the entire tobacco product supply chain until duties are paid or other obligations are discharged. Any tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it becomes possible to unambiguously register that product’s movements. A good tracking and tracing system enables the government to properly monitor the supply chain, improve its ability to ensure collection of the proper duties and taxes, authenticate whether the identification marking is genuine and matches the product, improve its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. To reduce the financial burden of implementing such a system, jurisdictions could require the tobacco industry to bear the cost. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, and governments should ensure that the system is independent from the tobacco industry. While the objectives of fiscal markings and tracking and tracing systems are different, stamps increasingly contain tracking and tracing features. • Implementing legal measures to prevent forestalling can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. Forestalling, stockpiling or front-loading occur when manufacturers or im- porters increase their tax-paid stock or oversupply the market by increasing production or imports before a tax increase in order to pay the previous lower rate. • Periodic audits and controls can be implemented to increase compliance. These include cost audits, transfer price audits, price and market monitoring, consumer controls and cross-check controls. • Import and export of tobacco products and manufacturing equipment should be allowed only for duly licensed natural persons or legal entities. The risk of loss of revenue can be mitigated by requiring a guarantee or bond that will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless required fiscal markings (such as tax stamps or export labels) are affixed on the pack, according to the law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Exchange of information between jurisdictions on the movement of goods can also reduce the risk of evasion. 160 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Non-invasive detection equipment (such as X-ray scanners) can be used at customs posts to detect contraband merchandise. A cheaper alternative is the use of dogs that are trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied to reduce contraband. Such measures include the separation of processing operations from the sealed storage of taxed and untaxed products. Within a country, mobile excise control units are helpful in verifying excis- able goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. Physical control operations require close coordination between police, border guards and other public services. • Controls such as regulation and oversight are usually less strict in free zones and transhipment points. This can make free zones appealing to persons involved in illegal cigarette manufacturing or trade. Customs administrations should exercise their authority in free zones to effectively identify and fight illicit trade in tobacco products. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing tracking and tracing regimes and removing exemptions from excise taxes. Other actions include the prohibition of intermingling of tobacco products with non-tobacco products in a single container or other similar transportation unit when the products are removed from free zones. Sale of tax-free or duty-free tobacco products to international travellers should be prohibited, as these sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect govern- ment revenues by creating a loophole in the tax structure. Procedures after detection of illicit trade of tobacco products should be clearly defined. If smuggling or illicit trade is detected through audits, tracking and tracing systems, verification of declarations or border control, actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. It is also important that penalties and sanctions be sufficient to deter illegal activities. Low financial penalties may simply be paid as a cost of doing business while the illegal activity continues. The minimum penalty for consumers in possession of illicit tobacco products should be confiscation and destruction of the products found in their possession and required payment for the unpaid tax and duties on those products. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are taking advantage of this type of legislation and using it to address illicit trade of tobacco as well. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 161 In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. The challenge for taxation of other products includes the lack of standardization of those products and sometimes large informal markets. Knowledge of the product and the supply chain greatly help to facilitate effective tax administration. The trade in raw tobacco and small-scale home production of RYO and other products such as bidis often takes place outside of monitoring and control systems. The best way to address this challenge is to enforce prior approval for purchase or sale of raw materials and a requirement to register, obtain an authorization or license all operators and growers handling raw tobacco. In principle, adding new and emerging nicotine and tobacco products to an exist- ing tax framework is not expected to impose significant costs. It is reasonable to expect that similar challenges will be faced in the collection of taxes on these newer products, as market players will attempt to use the current loopholes in tax regulation to avoid or evade taxes on these products whenever possible. However, challenges are expected to arise, as newer products involve rapidly changing technology, and their market dynamics are widely unknown. Furthermore, additional capacity may be required, since a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. Because the newer nicotine and tobacco products are widely purchased online, countries deciding to tax these products should draw up a proper implementation plan that includes rules on how taxes will be collected on imported products and online sales. Online cross-border sales are not permitted in some countries. The elements of a good tax system include (1) proper resourcing of competent authorities sufficient for hiring the necessary staff to properly implement and en- force excise tax laws; (2) implementation of tax laws with integrity and with strict rules and regulations to detect corruption and for the punishment of both agency personnel and taxpayers who are engaged in corrupt practices; and (3) ensuring that the judicial system is honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. 162 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N REFERENCES 1. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 2. WHO FCTC guidelines for implementation of Article 6 Price and tax measures to reduce the demand for tobacco. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/treaty_instruments/ Guidelines_Article_6_English.pdf?ua=1, accessed 5 October 2020). 3. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 4. Joossens L, Raw M. From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–234 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 5 October 2020). 5. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 6. Reyes IP. Measures to control the tobacco supply chain in the ASEAN: a review of excise tax stamps, track and trace mechanisms, and licensing systems in the ASEAN region. Bangkok: Southeast Asia Tobacco Control Alliance; 2014 (https://seatca.org/dmdocuments/SEATCA%20MEASURE%20BOOK%20. pdf, accessed 5 October 2020). 7. Italy’s tax administration: a review of institutional and governance aspects. Paris: Organisation for Economic Co-operation and Development; 2016 (https://www.oecd.org/tax/administration/italy- tax-administration-a-review-of-institutional-and-governance-aspects.pdf, accessed 5 October 2020). 8. Crandall W, Gavin E, Masters A. ISORA 2016: understanding revenue administration. Washington (DC): International Monetary Fund; 2019 (Departmental Paper Series No. 19/05; (https://www.imf.org/en/ Publications/Departmental-Papers-Policy-Papers/Issues/2019/03/07/ISORA-2016-Understanding- Revenue-Administration-46337, accessed 5 October 2020). 9. Dutch customs in 2018. Heerlen: Customs Administration of the Netherlands; 2019 (https://over-ons. belastingdienst.nl/wp-content/uploads/2019/04/Dutch-Customs-in-2018.pdf, accessed 8 October 2020). 10. ESAs transform the way competent authorities cooperate with each other on AML/CFT matters [press statement]. European Banking Authority; 16 December 2019 (https://eba.europa.eu/esas-transform- way-competent-authorities-cooperate-each-other-amlcft-matters, accessed 5 October 2020). 11. Illicit trade report 2013. Brussels: World Customs Organization; 2014 (http://www.wcoomd.org/-/ media/wco/public/global/pdf/topics/enforcement-and-compliance/activities-and-programmes/ illicit-trade-report/itr_2013_en.pdf?la=en, accessed 5 October 2020). 12. Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation and repealing Directive 77/799/EEC. Brussels: OJEU; 2011. (L 64/1; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:32011L0016&from=en, accessed 29 January 2021). 13. 439 stolen cars seized in 12 days as part of Frontex-led operation. Frontex News Release; 14 October 2019 (https://frontex.europa.eu/media-centre/news-release/439-stolen-cars-seized-in-12-days-as- part-of-frontex-led-operation-vvvirI, accessed 5 October 2020). 14. Focus on: the illicit trafficking of counterfeit goods and transnational organized crime. Vienna: United Nations Office on Drugs and Crime; 2014 (https://www.unodc.org/documents/counterfeit/FocusSheet/ Counterfeit_focussheet_EN_HIRES.pdf, accessed 5 October 2020). 15. Crandall W. Revenue administration: performance measurement in tax administration. Washington (DC): International Monetary Fund; 2010 (Technical Notes and Manuals 10/11; https://www.imf.org/ external/pubs/ft/tnm/2010/tnm1011.pdf, accessed 7 October 2020). 16. Measuring tax gaps 2019 edition – tax gap estimates for 2017-18. London: HM Revenue and Customs; 2019 (https://webarchive.nationalarchives.gov.uk/20200701215139/https://www.gov.uk/government/ statistics/measuring-tax-gaps, accessed 29 January 2021). 17. Countering illicit trade in tobacco products: a guide for policy-makers. Lyon: INTERPOL; 2014 (Legal handbook series; https://www.aph.gov.au/DocumentStore.ashx?id=704d9168-2aa7-4df4-bb81- CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 163 0fe1f357dfa6&subId=408832, accessed 29 January 2021). 18. Resolução- RDC No. 226 de 30 de Abril de 2018 [Resolution of April 30, 2018]. Ministério da Saúde/ Agência Nacional de Vigilância Sanitária/Diretoria Colegiada; 2018 (in Portuguese) (https://www.in.gov. br/materia/-/asset_publisher/Kujrw0TZC2Mb/content/id/12527411/do1-2018-05-02-resolucao-rdc- n-226-de-30-de-abril-de-2018-12527407, accessed 17 February 2021). 19. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 20. Preece R. Addressing the illicit flow of tobacco products in Australia. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 21. Excise on tobacco [website]. Canberra: Australian taxation office; 2019 (https://www.ato.gov.au/ business/excise-on-tobacco/, accessed 5 February 2021). 22. Safe Framework of Standards. Brussels: World Customs Organization; 2018 (http://www.wcoomd. org/-/media/wco/public/global/pdf/topics/facilitation/instruments-and-tools/tools/safe-package/ safe-framework-of-standards.pdf?la=en, accessed 29 January 2021). 23. The Trade Facilitation Agreement. Geneva: World Trade Organization; 2017 (https://www.wto.org/ english/docs_e/legal_e/tfa-nov14_e.htm, accessed 29 January 2021). 24. Authorised economic operator (AEO) [website]. Kildare: Irish Tax and Customs; 2020 (https://www. revenue.ie/en/customs-traders-and-agents/authorised-economic-operators/index.aspx, accessed 7 October 2020). 25. Performance outcome area 1: integrity of the registered taxpayer base. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www. tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 26. Commission implementing regulation (EU) 2016/323 of 24 February 2016 laying down detailed rules on cooperation and exchange of information between member states regarding goods under excise duty suspension pursuant to Council Regulation (EU) No 389/2012. Brussels: OJEU; 2018 (http://data. europa.eu/eli/reg_impl/2016/323/2018-02-15, accessed 7 October 2020). 27. Council Directive 2008/118/EC concerning the general arrangement for excise duty repealing Directive 92/12/EEC. Brussels: OJEU; 2009 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:3 2008L0118&from=en, accessed 29 January 2021). 28. Our strategy. London: HM Revenue & Customs, 2017 (https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/629941/HMRC-Strategy.pdf, accessed 31 January 2021). 29. Strategic plan FY2018-2019. Washington (DC): Internal Revenue Service, 2017 (https://www.irs.gov/ pub/irs-pdf/p3744.pdf, accessed 31 January 2021). 30. Performance outcome area 2: effective risk management. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/ assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 31. Tax administration 2017: comparative information on OECD and other advanced and emerging economies. Paris: Organisation for Economic Co-Operation and Development; 2017. 32. Compliance risk management: managing and improving tax compliance. Paris: Organisation for Economic Co-Operation and Development; 2017 (https://www.oecd.org/tax/administration/33818656. pdf, accessed 7 October 2020). 33. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 34. Evans P. U.K. Fines British American Tobacco for Oversupplying Cigarettes in Belgium. The Wall Street Journal. 13 November 2014 (https://www.wsj.com/articles/u-k-fines-british-american-tobacco-for- oversupplying-cigarettes-in-belgium-1415873470, accessed 31 January 2021). 35. Doward J. BAT fined for oversupplying tobacco in low-tax European jurisdictions. The Guardian. 164 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 16 November 2014 (https://www.theguardian.com/business/2014/nov/16/bat-fined-for-oversupplying- tobacco-in-low-tax-european-jurisdictions, accessed 31 January 2021). 36. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021) 37. Cour des comptes. Accises sur les produits du tabac, Bruxelles [Court of Auditors, Excise duties on tobacco products, Brussels]. Bruxelles: Cour des comptes; 2015 (in French) (https://www.ccrek.be/ docs/2015_27_AccisesProduitsDuTabac.pdf, accessed 7 October 2020). 38. Chaloupka FJ, Edwards SM, Ross H, Diaz M, Kurti M, Xu X, et al. Preventing and reducing illicit tobacco trade in the United States. Atlanta: Centers for Disease Control and Prevention; 2015 (https://www. cdc.gov/tobacco/stateandcommunity/pdfs/illicit-trade-report-121815-508tagged.pdf, accessed 29 January 2021). 39. The Tax Stamp Forum 2015. Hexham: Reconnaissance International (https://10times.com/tax-stamp- forum-miami, accessed 10 September 2020). 40. Tax stamps: a technical study and market report. London: Reconnaissance international; 2012. 41. ЕАНС/2013/Health/ll concerning the provision of an analysis and feasibility assessment regarding EU systems for tracking and tracing of tobacco products and for security features. Brussels: European Commission, 2015 (https://ec.europa.eu/health/sites/health/files/tobacco/docs/2015_tpd_tracking_ tracing_frep_en.pdf, accessed 29 January 2021). 42. Ross H. Controlling illicit tobacco trade: international experience. Economic research informing tobacco control policy. Cape Town: University of Cape Town; 2015 (https://tobacconomics.org/ uploads/misc/2015/05/Ross_International_experience_05.28.15.pdf, accessed 18 December 2020). 43. Security and resilience — authenticity, integrity and trust for products and documents — guidelines for the content, security, issuance and examination of excise tax stamps. Geneva: International Organization for Standardization; 2018. 44. Borkowski F, Twomey C. European Union: an update on EU policies. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 45. Quality management systems — Fundamentals and vocabulary. ISO 9001:2015(en). Geneva: International Organization for Standardization; 2015 (https://www.iso.org/obp/ui/#iso:std:iso:9001:ed- 5:v1:en, accessed 16 February 2021). 46. Protocol to Eliminate Illicit Trade in Tobacco Products: questions & answers. Geneva: World Health Organization; 2019 (https://www.who.int/fctc/protocol/faq/en/, accessed 7 October 2020). 47. Ross H, Eads M, Yates M. Why governments cannot afford Codentify to support their tracking and tracing solutions. Tob Control. 2018;27(6):706–08. 48. Ser vicio de Impuestos Internos, Chi le, Noticias, 28 May 2019. (http://www.si i .c l/ noticias/2019/280519noti01er.htm, accessed 20 October 2020). 49. Directive 2014/40/EU of the European Parliament and of the Council. OJEU. 2014;127:1–38 (https:// ec.europa.eu/health/sites/health/files/tobacco/docs/dir_201440_en.pdf, accessed 7 October 2020). 50. Report on the EU customs enforcement of intellectual property rights: results at the EU border. Luxembourg: Publications Office of the European Union, 2019 (https://ec.europa.eu/taxation_customs/ sites/taxation/files/2019-ipr-report.pdf, accessed 7 October 2020). 51. Matthews P. Forestalling ahead of property tax changes. London: Office for Budget Responsibility; 2016 (OBR Working Paper No. 10; https://obr.uk/docs/dlm_uploads/Working-paper-No.10-1.pdf, accessed 29 January 2021). 52. Ross H, Tesche J, Vellios N. Undermining government tax policies: common strategies employed by the tobacco industry in response to tobacco tax increases. Prev Med. 2017;105S:S19–S22 (https://www. sciencedirect.com/science/article/pii/S0091743517302165?via%3Dihub, accessed 17 February 2021). 53. European Commission v Portugal supported by Belgium, Estonia and Poland, Case C-126/15. Judgement of the Court (Fifth Chamber); 29 June 2017 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?u ri=CELEX:62015CJ0126&from=EN, accessed 31 January 2021). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 165 54. Amalia Valesko v Zollamt Klagenfurt, Case C-140/05. Judgement of the Court (Second Chamber); 5 October 2006 (http://curia.europa.eu/juris/showPdf.jsf;jsessionid=02DEE17CC11845BAE0F42FC7 CFD24C85?text=&docid=65599&pageIndex=0&doclang=en&mode=lst&dir=&occ=first&part=1&c id=2591459, accessed 29 January 2021). 55. European Commission v French Republic, Case C-197/08. Judgement of the Court (Third Chamber); 4 March 2010 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:62008CJ0197&qid =1612223322830&from=EN, accessed 31 January 2021). 56. Commission requests that Belgium amend its legislation on tobacco: letter of formal notice. Brussels: European Commission; 8 November 2018. (Memo 18/6247; https://ec.europa.eu/commission/ presscorner/detail/EN/MEMO_18_6247, accessed 29 January 2021). 57. Commission requests that Belgium amend its legislation on tobacco, MEMO/18/6247 of 8 November 2018. 58. Pederson H, Floristean A, Iseppi L, Dawkins R, Smith C, Mørup C, et al. Study on the measuring and reducing of administrative cost for economic operators and tax authorities and obtaining in parallel a higher level of compliance and security in imposing excise duties on tobacco products. Brussels: European Commission; 2014 (https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/ ramboll-tobacco-study.pdf, accessed 3 October 2020). 59. Money laundering vulnerabilities of free trade zones. Paris: Financial Action Task Force; 2010. https:// www.fatf-gafi.org/media/fatf/documents/reports/ML%20vulnerabilities%20of%20Free%20Trade%20 Zones.pdf, accessed 7 October 2020). 60. Omi K. ‘Extraterritoriality’ of free zones: the necessity for enhanced customs involvement. World Customs Organization; 2019 (WCO research paper no. 47; http://www.wcoomd.org/-/media/wco/public/global/ pdf/topics/facilitation/ressources/permanent-technical-committee/225-226/itemixb_fzs_wco_e. pdf?la=en, accessed 29 January 2021). 61. European Parliament resolution of 26 March 2019 on financial crimes, tax evasion and tax avoidance. European Parliament; 2019 (2018/2121(INI)); http://www.europarl.europa.eu/doceo/document/TA- 8-2019-0240_EN.html, accessed 7 October 2020). 62. International Convention on the Simplification and Harmonization of Customs Procedures. Brussels: World Customs Organization; 2008 (http://www.wcoomd.org/Topics/Facilitation/Instrument%20 and%20Tools/Conventions/pf_revised_kyoto_conv/Kyoto_New, accessed 7 October 2020). 63. Evidence Notice 2004: tobacco duty – tobacco products manufacturing machine licensing scheme. London: HM Revenue & Customs, 2018 (https://www.gov.uk/government/publications/excise-notice- 2004-tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme/excise-notice-2004- tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme, accessed 31 January 2021). 64. Tackling illicit tobacco: from leaf to light: the HM Revenue & Customs and Border Force strategy to tackle tobacco smuggling. London: HM Revenue and Customs; 2015 (https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/418732/Tackling_illicit_ tobacco_-_From_leaf_to_light__2015_.pdf, accessed 7 October 2020). 65. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 66. Cardenas M. Case studies in illicit tobacco trade: Colombia. Chicago: University of Illinois at Chicago; 2020 (Tobacconomics Fact Sheet; https://tobacconomics.org/files/research/606/UIC_Colombia-Illicit- Trade-Fact-Sheet_v1.4.pdf, accessed 29 January 2021). 67. Maldonado N, Llorente BA, Escobar D, Iglesias RM. Smoke signals: monitoring illicit cigarettes and smoking behaviour in Colombia to support tobacco taxes. Tob Control. 2019;29:s243-s248 ( https:// tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/s243.full.pdf, accessed 29 January 2021). 68. Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 69. Pellechio A, Tanzi V. The reform of tax administration. Washington (DC): International Monetary Fund; 1995 (Working Paper no. 95/22; https://www.elibrary.imf.org/view/IMF001/07242- 9781451843941/07242-9781451843941/07242-9781451843941_A001.xml?language=en&redirect=true, accessed 29 January 2021). 166 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.1 COMPOSITION OF TOBACCO PRODUCTS To implement and enforce tobacco taxes in the most efficient way, competent authori- ties should be familiar with all of the components of tobacco products, including each of the raw materials used in their manufacture, production inputs and tobacco manufacturing machinery. Knowledge of the components of excisable products and machinery provides valuable information to identify activities at high risk for non- compliance, implement measures to ensure all taxes are paid and prevent illicit trade. Nicotine, non-nicotine and tobacco products and their component parts In most countries, the ministry of finance determines tax policy, including which tobacco products are taxed, while the ministry of health is responsible for product and use regulation. This could lead to different definitions of the same product, depending on which ministry is responsible for a given law or regulation. Wherever possible, a clear and common definition should be developed to simplify procedures and avoid confusion. Tobacco products take various forms, and not all may be regulated or subject to excise tax in a specific jurisdiction. In addition to cigarettes, other traditional tobacco products include smokeless tobacco – such as chewing tobacco, snuff and snus – as well as bidis and kreteks (clove cigarettes), which can be hand-rolled or manufactured, pipes, hookah or waterpipe and cigars. Cigarettes It is important to understand the materials and component parts of the tobacco products most commonly used in a particular country. Cigarettes are the most common and significant tobacco products in terms of volume and tax revenues in most jurisdictions. A cigarette stick is composed of: • the tobacco blend of various types of tobacco plant (leaves and stem and other plant parts) and additives (including flavours); • the cigarette paper used to wrap the tobacco blend to make up the tobacco rod; • the acetate filter that forms the white portion at the tip of a filtered cigarette, which is in direct contact with the smoker’s mouth; • the tipping paper or wraps around the filter; and • the adhesive that secures the cigarette paper around the tobacco blend and the tipping (1). Each manufacturer follows a specific process to produce cigarettes. Aside from the tobacco blend, manufacturers also vary the size of cigarette paper and tipping paper and the length of acetate filter used per stick (1). In some countries, these elements are standardized. In an ideal regulatory framework, a manufacturer would CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 167 be required to submit information on the specific process for each brand and variant of the tobacco product that it manufactures to the competent authorities as part of the licensing requirement (see section 3.4.3). Authorities could, for example, require manufacturers to submit this information in order to obtain a licence. The minimum requirements of the administration and the information that should be included could be laid down in law or lower regulation to ensure that authorities have the information in their possession for all licensed manufacturers. This information contributes to verifying whether a company is reporting the actual quantity of cigarettes manufactured for sale and sold cigarettes by comparing the amount of materials used for production and the quantities used per cigarette with the total number of manufactured cigarettes. Eventually, this information also contributes to validating whether the taxes are properly paid. Figure A3.1 shows the component parts of a typical machine-made traditional cigarette. Fig. A3.1 Component parts of a machine-made cigarette Source: Author’s compilation. Photo by Walter Klerx. Not all parts of tobacco products are subject to the same level of control. According to Article 6.5 of the Protocol, five years following the entry into force of this Protocol, the MOP shall ensure at its next session that evidence-based research is conducted to ascertain whether any key inputs exist that are essential to the manufacture of tobacco products, are identifiable and can be subject to an effective control mechanism. On the basis of such research, the MOP shall consider appropriate action. In addition to the component parts of tobacco products, materials needed for packaging a specific number of sticks into a pack of cigarettes, usually 20 per pack, can be monitored. These materials include the foil paper, the package paper (which could bear the brand name, design and health warnings), the fiscal marking (if required) and the plastic or cellophane wrap. A fixed number of packs of cigarettes, Filter Tipping paper Cigarette paper Tobacco Tobacco rodFiltration zone 168 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N normally 10 packs, are packed into cartons, also called reams. These cartons are usu- ally made of soft paperboard or cardboard, possibly with branding, and are wrapped in plastic wrap or cellophane. Fifty cartons are packed in master cases, which are made of sturdier and thicker paperboard and stacked on pallets (usually 50 master cases to a pallet). An effective regulatory framework would require manufacturers and importers to provide information to the competent authorities on packaging and design, as well as the number of sticks per pack, carton and master case. Prior approval for purchase or sale of materials used in the cigarette production process can also be required. In the Philippines, suppliers of such raw materials, including those providing tobacco papers and filter components, are required to have a licence (2). In some of the Member States of the EU, raw tobacco is also subject to fiscal and legal requirements. For example, in Slovakia and Poland, raw tobacco can be handled only by authorized operators. While authorized operators do not have to pay excise duties on raw tobacco, if raw tobacco is detected by an unauthorized operator, excise duties will be due. Hungary, Italy and the United Kingdom require registration or authorization for all operators and growers handling raw tobacco (3). In addition to knowing the quantities of inputs required to produce a specific amount of a regulated product (e.g. cigarettes), the competent authority also needs to understand the supply, manufacturing and distribution chains to be able to properly monitor, regulate and determine whether taxes have been paid (see also Fig. 3.3). Novel and emerging nicotine, non-nicotine and tobacco products In recent times, new products have been introduced to several markets, namely, ENDS, ENNDS and HTPs. ENDS usually comprise a nicotine-containing e-liquid but do not contain to- bacco. ENNDS are essentially the same but do not (ostensibly) contain nicotine. The WHO COP requested the Convention Secretariat to invite Parties to monitor and report on scientific, regulatory and market developments such as initiation, cessation, advertising and promotion of ENDS and ENNDS. Furthermore, the COP requested WHO to report on the development of methods by regional and international standards-development organizations for the testing and measuring of contents and emissions of these products (4). There are different types of e-cigarettes – the most common type of ENDS and ENNDS – and currently there are four generations of products. However, they can be divided into two broad categories: open systems and closed systems. Both types of e-cigarette use a wick and a heat source to generate an aerosol. The wick is saturated with e-liquid, and a microprocessor is used to control operations (not all include this). Some e-cigarettes also have an LED light to imitate the burning end of a conventional cigarette (5). Fig. A3.2 presents examples of open and closed systems. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 169 E-liquid pod cover Rechargeable battery Heating element (heats solution, aerosolizing nicotine) Mouthpiece E-liquid pod Fig. A3.2 Examples of open and closed systems of ENDS/ENNDs products Open ENDS/ENNDS system (e-cigarette) Closed ENDS system (e-cigarette) Source: (6). Unlike ENDS/ENNDS, HTPs do contain tobacco. HTPs produce aerosols containing nicotine and toxic chemicals when tobacco is heated or when a device containing tobacco is activated (7). HTPs are composed of two elements: the sticks or pods that contain the tobacco and the device used to heat the tobacco. Both are necessary for the product to be used. Fig. A3.3 shows an example of a heated tobacco product. HTPs are tobacco products and are therefore subject to the regulatory measures contained in the WHO FCTC. Rechargeable battery Power button (to start vaping) Mouthpiece Atomizer / Heating element (heats solution, aerolizing nicotine) E-liquid tank (refillable e-liquid nicotine tank) 170 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. A3.3 Components of an HTP Holder: Tobacco stick: Note: PLA: polyactic acid, MPF: mouthpiece filter. Sources: (8-9). More information on tax administration of other tobacco products is presented in section 3.5. Tobacco stick Holder Charger Casing Control Electronics Battery Heating Blade Heatstick total lenght: 45mm MPF (7 mm) PLA (18 mm) Tobacco plug (12 mm) Hollow acetate tube (8 mm) Outer paper Diameter max 7.42 mm Tipping paper CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 171 REFERENCES 1. Memorandum by Gallaher Group Plc: the tobacco industry and the health risks of smoking. Select Committee on Health Minutes of Evidence. London: UK Parliament, 2000 (https://publications. parliament.uk/pa/cm199900/cmselect/cmhealth/27/0011323.htm, accessed 7 October 2020). 2. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 3. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco.. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 4. Electronic nicotine delivery systems and electronic non-nicotine delivery systems. Geneva: World Health Organization; 2016 (Decision FCTC/COP7(9) of the Conference of the Parties to the WHO Framework Convention on Tobacco Control; https://www.who.int/fctc/cop/cop7/FCTC_COP7_9_EN.pdf?ua=1, accessed 7 October 2020). 5. Brown CJ, Cheng JM. Electronic cigarettes: product characterization and design considerations. Tob Control. 2014;23:ii4-ii10 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3995271/pdf/ tobaccocontrol-2013-051476.pdf, accessed 31 January 2021). 6. E-cigarettes and vapor products [webpage]. King County, Washington (USA); 2019 (https://www. kingcounty.gov/depts/health/tobacco/data/e-cigarettes.aspx, accessed 9 October 2020). 7. Heated tobacco products. Geneva: World Health Organization; 2020 (WHO/HEP/HPR/2020.2 Information sheet; https://apps.who.int/iris/bitstream/handle/10665/331297/WHO-HEP-HPR-2020.2- eng.pdf?sequence=1&isAllowed=y, accessed 7 October 2020). 8. Tobacco heating system (IQOS) briefing document. Silver Spring: US Federal Drug Administration; 2018 (https://www.fda.gov/media/110377/download, accessed 31 January 2021). 9. Premarket tobacco product application: technical project lead review. Silver Spring: US Federal Drug Administration; 2017 (https://www.fda.gov/media/124247/download, accessed 7 October 2020). 172 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.2 EXAMPLE OF FORESTALLING AND COUNTERMEASURES This example is hypothetical but inspired from the situation in the Philippines. The amounts and prices have been altered, however, and it is assumed that the normal inventory kept by a tobacco manufacturer is two months.  The excise tax imposed on a pack of cigarettes in the current year is US$ 3.00; it will be increased to US$ 3.30 at the beginning of the new fiscal year (January in this example). The monthly production of Brand Y cigarettes of company X, which it declares for tax purposes, is as follows: MONTH OF THE CURRENT YEAR PACKS OF CIGARETTES January 10 000 000 February 10 500 000 March 9 900 000 April 11 000 000 May 10 200 000 June 10 600 000 July 9 700 000 August 10 100 000 September 9 900 000 October 10 100 000 November 20 000 000 December 25 000 000 Since the normal inventory is two months, the quantity in the two months prior to the implementation of the new excise tax rate is disregarded. The shelf life of tobacco products is approximately six months. The average of the six months prior to November is computed to obtain the quantity presumed to be produced or imported if there was no tax increase. The quantity from May to October (inclusive) divided by 6 is 10 100 000 packs. Thus, any quantity produced beyond 10 100 000 packs for the months of November and December (the months prior to the implementation of the new tax rate) is assessed using the new tax rate. In this example, 10 100 000 of the packs produced in November will be taxed at the old rate of $3.00, and 9 900 000 packs will be taxed at the new rate of $3.30. For December, 10 100 000 packs will be taxed at $3.00, while 14 900 000 packs will be taxed at $3.30. Without imposing these measures, the government would have been deprived of the excise tax increase on 24 800 000 packs. In addition, the effect of the increase on prices and consumers would have been delayed by approximately two months. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 173 In countries using tax stamps, the withholding of the issuance of stamps is a well-known approach to counter forestalling. Another practical solution is to allow the competent authorities to request advances from the industry to cover revenue shortfalls, provided there is a legal basis for such requests. 174 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 175 CHAPTER 4. Political economy As with any proposed government action, policy-makers need to navigate the political environment of tobacco taxation at every stage of policy development, implementa- tion and administration. While every country’s distinct history, culture, systems and structural forces shape its unique political landscape, there are some universal themes when it comes to tobacco control and particularly tobacco taxation. These themes boil down to the distribution of money, power and resources. The tobacco industry, as both a political and economic player, understands these themes well. The industry has been effective in using principles of the political economy of tobacco taxation in its efforts to block important advancements in tobacco control. Nevertheless, the savvy policy-maker can see through the industry arguments by considering who benefits from industry-favoured policy measures and interven- tions. The industry’s challenges to tobacco tax policies can be organized into the five categories of SCARE tactics. This chapter provides a road map to help policy-makers navigate the political economy of tobacco taxation through each of these themes. The first five sections dissect the tobacco industry framing of each issue, pinpoint- ing the flaws in each argument, identifying the extent to which each concern has merit and suggesting how a responsible government can address each one. These discussions are supported by unbiased evidence from independent, peer-reviewed research, as well as specific examples from country experiences. Sections 4.1 through 4.5 on SCARE tactics will equip policy-makers with the tools they need to proceed with confidence that their tobacco tax policy – developed and implemented fol- lowing the guidelines spelled out in this technical manual – will bring about the greatest health and economic benefits for their constituents, regardless of industry attempts to thwart them. Section 4.6 further buoys policy-makers’ efforts to ensure the beneficial impacts of their policies, as it describes how earmarking can improve the political economy of tobacco taxation by funding programmes and initiatives that promote and support the health and well-being of the population. 176 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.1 SCARE TACTIC S: SMUGGLING AND ILLICIT TRADE 4.1.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases will inevitably result in an increase in the illicit trade in tobacco products (1–2). They claim that higher tax rates and higher prices strengthen the financial incentives for criminal enterprises to supply cigarettes from lower-tax jurisdictions, boost domestic tax evasion and encourage smokers to seek cheaper illegal cigarettes. The industry also challenges the argument that tobacco tax hikes increase government revenue by claiming that the presence of an illicit tobacco market will actually reduce revenue collection following a tax increase. More recent versions of this argument – adapted to address public health concerns about tobacco use – claim that illicit market growth also offsets reductions in smoking prevalence that would otherwise be brought about by tobacco tax increases. In summary, the tobacco industry and its allies claim that raising tobacco taxes are ineffective – and even counterproductive – because they are circumvented by illicit markets, which prevents the government from achieving its public health objective and reduces rather than increases tax revenues. When a country considers a proposal to increase tobacco excises, the tobacco industry and its allies frequently make exaggerated claims about the size and scope of illicit tobacco trade in that country. Opponents of tobacco tax increases argue that price differentials are the exclusive – or at least the dominant – cause of illicit trade. Influenced by this fear-inducing faulty diagnosis, tax authorities frequently find it difficult to make decisions about tobacco taxes. However, the industry diagnosis always contains the same erroneous elements. First, the illicit trade in a country is frequently less than the industry portrays it to be, and the country’s tax enforcement policy towards tobacco products is rarely unique or in any way different from the norm in the country (3). Second, the scale of illicit trade in tobacco is not exclusively or even primarily determined by tax or price differentials. Typically, it results from a set of governance problems characterized by government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, lack of dissuasive sanctions and/or weak norms regarding participation in illegal and informal markets (4–7). This section provides guidance for tax and other relevant authorities on how to respond to the tobacco industry SCARE tactic that increasing tobacco taxes will lead to smuggling and illicit trade in their countries. Tax authorities need to know the nature, causes and extent of illicit trade so that they can define the problem properly and formulate an appropriate response. This chapter addresses the available tools for better defining and understanding specific illicit trade problems, particularly tools that facilitate independent assessment of the magnitude of that trade. Improvements CHAP T ER 4. PO LI T I C AL ECO N OMY 177 to governance within the realm of tax authorities – such as best practices in tobacco tax administration and policies to improve the effectiveness of fiscal regulations or norms regarding participation in informal and illegal markets – are discussed in Chapter 3. This section first describes the nature of the illicit tobacco trade to highlight some of its complexities and identify complementary policies for tackling the problem. Next, evidence that calls into question the link between illicit trade and high prices or tax rate changes is discussed. Finally, to help tax authorities assess their own situation, several different methodologies are presented to estimate the scope of the illicit tobacco trade and to evaluate estimates of that trade for a particular country or tax jurisdiction. 4.1.2 THE NATURE AND EXTENT OF THE ILLICIT TOBACCO TRADE The WHO FCTC defines illicit trade as: any practice or conduct prohibited by law and which relates to production, ship- ment, receipt, possession, distribution, sale or purchase, including any practice or conduct intended to facilitate such activity (8). Non-duty-paid tobacco products found in a jurisdiction (i.e. through littered-pack surveys) could be the result of either of two related but distinct activities: tax evasion and tax avoidance. Tax evasion is a set of unlawful actions seeking the non-payment of tobacco taxes and duties, whereas tax avoidance comprises legal actions with the purpose of avoiding payment of some or all taxes, such as bringing an amount of cigarettes up to the legal allowance from a lower- into a higher-tax jurisdiction. Tax avoidance is not illegal and is therefore not considered part of illicit trade in tobacco products.1 The focus of this section is on tax evasion activities, which can occur in the movement across borders or in domestic production and distribution. When tax evasion happens across borders, it is known as smuggling (9) and can be done on a large scale or a small scale. Tax evasion in the domestic market can be partial, when licensed and authorized producers or distributors comply with only part of their tax obligations, or total, when the whole production and distribution system is illegal and out of sight of tax administrators (5, 10–14).2 Large-scale tax evasion schemes can be run by different types of producers and their associated distributors, such as the transnational tobacco companies (TTCs) and their national subsidiaries, 1 Tax avoidance practices –common among states in the United States and countries in the EU – are not analysed in this section. 2 Tax evasion is normally considered as illicit manufacturing in the literature (14). 178 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N other local tobacco companies producing their own brands and illegal factories that normally counterfeit other brands or produce illicit brands. Large-scale smuggling involves, for example, taking advantage of tax-free zones and mislabelling shipping products prior to or during transit (11) or using so- phisticated clandestine networks. This form of tax evasion is systemic and can be carried out by TTCs (12), by local companies producing in countries with low tax enforcement, e.g. Paraguay (15) that feed neighbouring countries and regional illicit hubs through a network of clandestine distributors (15) or by companies located in tax-free zones like Jebel Ali and Dubai in the UAE (16) or such zones in Russia or Cyprus (17). The origins, routes and quantities of large-scale smuggling frequently change as affected countries or markets react by strengthening enforcement and seeking bilateral cooperation with the jurisdictions of origin. For example, the 2013 comprehensive strategy of the EU (10) enhanced bilateral cooperation with major source and transit countries of illicit cigarettes coming to Europe, including Russia, Belarus and Ukraine. These agreements improved day-to-day cross-border coopera- tion, reduced illegal flows and introduced gradual tobacco excise rate adjustments in those three countries to bring them to European levels. Small-scale smuggling (also known as ant smuggling or bootlegging) is the cross-border trafficking of cigarettes in quantities that are larger than the allowable limits (e.g. two cartons) but smaller than large shipments (e.g. truckloads, cargo containers), normally for the purpose of selling at a profit (11). This type of illicit trade may exist in places where there are opportunities within neighbouring tax jurisdictions. For example, small-scale smuggling is commonly done by individuals living in French and German provinces near lower-taxed countries (e.g. Belgium, Luxembourg, Switzerland, Spain, Poland and Czechia) (18). Counterfeiting is a form of illicit manufacturing that involves the production of tobacco products (including packaging and tobacco filler) without the approval of the trademark holder (13). Another product of illicit manufacturing is so-called cheap or illicit whites. Cheap whites are branded (e.g. Jin Ling) or unbranded cigarettes that are legally or illegally produced3 and knowingly sold in the illicit market (17). Cheap whites are not usually produced by TTCs (17, 19).4 They are produced by small tobacco produc- tion companies in one country and often sold in illegal markets of neighbouring 3 Ross et al. (17) analysed this issue and found that the sale to the first purchaser is usually legal. Their analysis covers the production in free zones (i.e. in the UAE, Russia and Cyprus) and production exported from Viet Nam, Indonesia and China. In those cases, there is no need to make the first sale illegally. However, cheap white production in Paraguay is sold to domestic distributors, and most of those sales are completely illegal. 4 Ross et al. (17) and Gilmore et al. (19) identify some cheap white brands sold by TTCs, such as President (PMI), produced in Ukraine, and Esse (Korea Tobacco & Ginseng Company, KT&G), produced in Indonesia. CHAP T ER 4. PO LI T I C AL ECO N OMY 179 countries. For example, in Paraguay, cheap whites are produced on a large scale by a few companies under the guise that they are marketed domestically, but a large share is smuggled into Uruguay and Brazil (4). Iglesias et al. (20) showed how TTCs’ cheap brands were illicitly shipped through Paraguay to be sold in the Brazilian and Argentine markets in the 1990s. This contributed to increased production of cheap whites in Paraguayan firms, which continued the illicit business even after Brazilian legislation obstructed the illegal activity of the TTCs. Domestic tax evasion is a pervasive phenomenon, particularly in LMICs. Partial tax evasion in tobacco products can be found at any level of tax rates or prices and is generally the result of defective legislation or weak tax enforcement.5 Complete or total tax evasion occurs when producers and distributors are clandestine or when there are serious institutional challenges to tax enforcement between two tax jurisdictions, such as between the United States and Native American Reservations. Evidence of illicit manufacturing has increased in recent years in several places in the world, including the EU (10) and Brazil (21). TTCs were predominant in illicit trade activity until the end of the 20th century, and even with the entrance of new actors into the illicit business, TTCs have not entirely exited. Gilmore et al. (22) analysed industry-funded data and seizure data and concluded that TTCs are still involved in illicit trade in Europe, despite the Anti-Contraband and Anti-Counterfeit Agreements (the “Agreements”) signed between the four TTCs and the EU (23).6 Using industry-funded data, Gilmore et al. show that 58% of illicit EU cigarettes can be attributed to the four main TTCs. When seizure data are used, 69% to 73% of illicit EU cigarettes can be attributed to these firms (22). It is always difficult to assess the extent of the global illicit tobacco trade because of its illegality, its global and changing nature and problems with data collection (24). Before the 21st century, when TTCs were almost unique actors in the large-scale smuggling of well-known cigarette brands, the difference between global exports and imports of cigarettes could provide a good approximation of the size of this problem globally (7). However, with the growth of illicit manufacturing in general, the manufacturing of cheap whites and the illegal movements of those products 5 This occurs when licensed and authorized producers underreport actual quantities and sell the non- duty-paid produced quantities through illegal channels. It can also include instances when producers do not report quantities at all, as in many ad valorem systems of LMICs. 6 From 2000 on, the European Commission and 10 Member States launched court cases regarding smuggling and money laundering against several TTCs. To end the court cases, the Anti-Contraband and Anti-Counterfeit Agreements were signed, which required the TTCs to exercise stringent control over their supply chain (through tracking and tracing, due diligence and anti-money-laundering and reporting obligations), share operational intelligence with Member States and the EU and pay penalties for seizures, as well as annual payments over a period of 12 years. The agreement with PMI has ended, the one with JTI will end in 2022 and the others with Imperial Tobacco and BAT will run until 2030 (23). 180 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N over the past two decades, trade statistics are no longer as useful as they were in the past. Joossens et al. (25) tried to estimate the size of the global illicit cigarette market by adding different types of national estimations prepared around 2007. They found that the estimated size was 657 billion cigarettes per year, or 11.6% of the global cigarette market. According to Joossens et al., illicit trade costs govern- ments US$ 40.5 billion in tax revenue worldwide, and eliminating illicit tobacco trade would recover US$ 13 billion in immediate revenue in high-income countries and US$ 18.3 billion in LMICs.7 Descriptions of the types of illicit trade are useful for developing the first com- ponent of a strategy to fight it: assess the nature and size of the problem. Table 4.1 presents all the main components of a strategy to fight illicit trade. To make progress in this first component – knowing the problem – authorities could use and adapt existing instruments of health surveillance or seek partnerships with academia and independent specialists to investigate the issues involved, using different methodolo- gies (see subsection 4.1.4 and Annex 4.1 on methodologies to assess the nature and size of the problem). Knowing the nature of the problem requires the cooperation of different government actors – for example, to investigate both the financial and criminal operations of organized crime behind the illicit trade. The gathering of qualitative information on the nature of the illicit trade should start simultaneously with the statistical work of measuring the magnitude of the problem. Table 4.1 Components of a strategy to fight the illicit tobacco trade 1. Assess the nature and size of the problem Use and adapt existing health surveillance and other existing national surveys to assess the problem Seek partnerships with academia and independent specialists to find ways to rigorously study illicit trade Use financial and police investigations to identify and fight organized crime operating in illicit trade 2. Start identifying and implementing appropriate country-specific policies and strategies to address illicit trade Improve tax and customs administration to close the legal and administrative loopholes facilitating illicit trade Implement other appropriate policies to deal with country-specific problems 3. Become a Party and/or implement the Protocol to Eliminate Illicit Trade in Tobacco Products Adapt the Protocol supply-chain control obligations Adjust national penalties for illicit trade offences Seek and build international cooperation 7 The WCO publishes an Illicit Trade Report annually, with the main characteristics and trends of illicit flows in key products, including tobacco, using data based on customs seizures. CHAP T ER 4. PO LI T I C AL ECO N OMY 181 Methodologies available to estimate the nature and size of illicit trade are discussed in subsection 4.1.4. This is the first step for dealing with SCARE tactic S. Chapter 3 discusses at length the relevant tax administration measures and best practices to minimize opportunities for illicit trade in tobacco products. Table 4.2 presents examples of appropriate policies and strategies targeted to address specific types of illicit trade in addition to the best practices described in Chapter 3. After completing the first step of this strategy, tax, health and justice authorities should discuss how to face country-specific problems, considering not only tax and customs administration measures but also social, law enforcement and international cooperation policies and strategies. Table 4.2 Suggested policies and strategies to address country-specific illicit trade problems MAIN TYPE OF ILLICIT TRADE IN THE JURISDICTION PROBLEMS POLICIES/STRATEGIES TO USE Bootlegging Neighbouring low-tax jurisdiction Bilateral negotiations to harmonize tobacco tax systems Difficulty of controlling people’s movements in countries with extensive land borders Identify and establish suitable social protection or employment policies for targeted populations in border regions Extensive land border with multiple accesses Bilateral cooperation with law enforcement and border control forces, monitoring of access routes to main consumption markets Large-scale smuggling from neighbouring jurisdiction Neighbouring low-tax jurisdiction and difficulties in controlling borders Bilateral negotiations to harmonize tobacco tax systems and bilateral law enforcement cooperation Producers and distributors in the lower-tax jurisdiction aiming to supply the high-tax jurisdiction Bilateral cooperation to harmonize tax systems and control producers and distributors in the origin country, create conditions for legal exports and taxed imports Large-scale smuggling from a third country or tax-free zones Producers and distributors aiming to supply non-duty-paid tobacco products wherever possible Customs and other forms of international cooperation to control and monitor exports from identified areas Domestic tax evasion Existence of many small informal or semi-formal producers Encourage business concentration through producer associations and cooperatives, create incentives for formalization and establish licensing rules and basic electronic information systems for raw material and production 182 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Underreporting from formal producers Improve tax administration with policies such as basic electronic information systems for inputs and production, establish neutral procedures to verify production, improve audit systems, increase third-party information on inputs and production of tobacco products Clandestine factories Law enforcement investigation of commercial associations with raw- material and machine producers and distributors Governments should try to identify the incentives and governance problems that encourage and allow illicit trade movement inside their country. As seen in Table 4.2, the design and implementation of policies to deal with those problems do not depend exclusively on tax and customs authorities; they also depend on the efforts of the police and law enforcement, the Justice Department and the judicial apparatus. In other words, a great deal of coordination and consultation among different types of government bodies and expertise is needed to produce an adequate response. It is also clear from Table 4.2 that domestic tax evasion by formal producers can be tackled by tax authorities and is mainly related to the supply-chain-control provisions of the Protocol. The Protocol (26) builds upon and complements Article 15 of the WHO FCTC, which addresses means of countering illicit trade in tobacco products as a key aspect of a comprehensive tobacco control strategy. The Protocol is a blueprint of measures to deal with this problem, and its provisions should be part of any strategy for fighting the illicit market. It is a legally binding treaty in its own right that entered into force on 25 September 2018. As described in detail in Chapter 3, the Protocol has three main lines of action: supply-chain controls, recommendations on how to treat unlawful conduct related to the illicit tobacco trade and suggested mechanisms to seek and build international cooperation to fight that trade. Countries can start implementing Article 15 of the WHO FCTC and the appropriate polices or strate- gies recommended by the Protocol even before acceding to it, selecting those most suitable to the nature and extent of their particular problem. Such transitional work will facilitate the eventual implementation of the Protocol, because any plan to correct loopholes in tax and customs practices will bring government authorities closer to the best practices recommended in the Protocol. 4.1.3 DETERMINANTS OF TAX EVASION: THE ROLE OF PRICE LEVELS The argument that price and tax rates are the main determinants of the illicit tobacco trade has persuaded some governments (e.g. Uruguay and Georgia in the past) to avoid policies that may lead to cigarette price increases (e.g. excise tax rate increases) (4). CHAP T ER 4. PO LI T I C AL ECO N OMY 183 Some governments (e.g. Canada in 1994, Brazil in 1999 and Pakistan in 2017) (20, 27) have even reduced tax rates in attempt to reduce the illicit trade. The wider scholarly literature demonstrates that illicit trade is not a monocausal phenomenon (7) but is the result of many factors, most of them related to gover- nance issues. Government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, weak norms regarding participation in illegal and informal markets and conflicts between neighbouring countries (5) all contribute to the existence and growth of the illicit tobacco trade. It is difficult to isolate the role of price from each of the other factors because (1) obtaining prices and quantity measures of illicit trade is inherently challenging; (2) in most countries, there are many cigarette brands, and prices vary between and even among brands; and (3) there is a lack of good measures to deal with nonprice factors affecting illicit trade, such as government corruption and ineffective criminal justice. These constraints make it challenging to develop rigorous empirical evidence about how price and other factors affect illicit trade. Despite these fundamental challenges, the economic literature has produced credible evidence that price is always only one factor – and often not the most important factor – determining the extent of illicit trade. Many econometric studies about the influence of price and other factors have focused on cross-border shopping (or small-scale bootlegging from low- to high-tax jurisdictions), given the availability in the United States and Europe8 of sales data for low- and high-tax jurisdictions, classified in a convenient way by geographical zones – i.e. close to or far from the borders. Those studies attempted to explain the illicit trade flows or the relatively higher sales in low-tax jurisdictions as a function of price and tax differentials between the lower-tax and surrounding higher-tax jurisdictions, after controlling for other important factors affecting cross-border sales such as proximity to borders and levels of corruption (6, 7, 11, 28).9 The main conclusion of the studies is that illicit trade flows are not linked solely to price (29). Some show a significant effect of price differentials together with other factors, but others do not find significant price differential effects. The important policy implication of these analyses is that decreasing tobacco tax rates and real prices in higher-tax jurisdictions could have minimal or no effect on illicit market shares.10 8 This was a traditional strand of the literature in the United States on trade among states, and to a lesser extent in European countries, most of which used conventional but inaccurate illicit trade measurements. 9 Recently, PMI-Altria financed some studies of factors affecting cross-border sales. One of those studies, Prieger and Kulic (28), criticized Merriman et al. (2000) (9) and arrived at the conclusion that in cross-border shopping, price differentials are important for determining the magnitude of illicit trade. 10 Brazil decreased tax rates and real prices at the beginning of the 21st century to fight illicit trade coming from lower-tax jurisdictions. After this action, however, the government lost revenues, and the size and scope of illicit trade remained unaltered, according to industry sources (20). 184 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Observational and case studies provide information that may improve public policy even when they are unable to produce compelling evidence of causal relation- ships. Some observational studies have correlated price levels with illicit market shares, using large samples of countries. Joossens et al. (25) found that countries with high taxes and prices normally have lower shares of illicit trade than countries with lower tax shares and prices. In their sample, high-income countries generally have relatively high cigarette prices and tax shares, but their favourable results (i.e. lower levels of illicit trade) are related to effective tax administration and lower corruption levels. In contrast, LMICs generally have lower prices and tax shares, along with significant illegal market shares. Joossens et al. attribute difficulties in fighting illicit trade to weak tax and customs administrations and, in most cases, institutional and legal challenges (25). Figure 4.1.1 illustrates the relationship between price and illicit trade, using the price (in US$) per pack of the most-sold brand of cigarettes and the estimated level of illicit trade for 94 countries in 2018.11 There is no apparent unique association between the two variables. Running a linear regression with retail price as the explana- tory variable and share of illicit trade as the dependent variable shows an inverse, but not statistically significant, relationship between price and illicit market share.12 Figure 4.1.1 illustrates some particular cases: • Many countries with low prices (i.e. lower than US$ 2 per pack) have the highest levels of illicit trade in the sample, e.g. Brazil (BRA) ($1.33 and 46.3% illicit share), Pakistan (PAK) ($0.39 and 40%), Ethiopia (ETH) ($0.55 and 32.9%), Ghana (GHA) ($1.06 and 29%) and Cameroon (CMR) ($0.89 and 25%). • In contrast, many of the countries with prices between US$ 4 and $8 – which could be considered high enough for financial incentives to operate – have illicit trade shares of less than 10% of total consumption. These countries include the Republic of Korea (KOR) ($4.02 and 0.8%), Czechia (CZE) ($4.31 and 2.9%) and Sri Lanka (LKA) ($6.89 and 1.6%). • All countries that have very high prices – higher than US$ 8 – except for Ireland, register illicit trade shares below 20%. These countries include France (FRA) ($9.39 and 17.8%), Switzerland (CHE) ($8.71 and 5.5%), Singapore (SGP) ($10.35 and 3.7%) and Norway (NOR) ($14.51 and 9.6%). 11 National estimates of the magnitude of illicit trade are controversial. The tobacco industry’s numbers overestimate the problem and are based on questionable methodologies. Estimates with a rigorous and transparent methodology are not available for a large sample of countries for the same year. In order to compare price levels with illicit market shares, Euromonitor’s estimations of illicit market share were selected, for two reasons: they are comparable estimates for a large sample of countries in a given year, and no one could argue that they are biased towards tobacco control’s points of view. The use of Euromonitor data does not imply that WHO fully agrees with all the details and methodologies used to obtain them. 12 Other factors must be taken into account to transform this observational analysis into a rigorous analysis of cause and effect. CHAP T ER 4. PO LI T I C AL ECO N OMY 185 Fig. 4.1.1 Share of illicit trade versus retail price of the most-sold brand of cigarettes in US$, by country, 2018 Note: The extent of illicit trade in cigarettes is measured by Euromonitor as the estimated quantity of illegal cigarettes consumed in a country divided by the estimated total consumption of cigarettes in that country. Sources: (27, 30). 51 2 3 4 6 7 8 9 10 11 12 13 14 1615 0 10 20 25 15 5 30 35 40 45 50 MYS BRA ECU ECUPAK ETH PAN CRI URY GRCIND GTM SLV ARE LVAHND NGA ZAF CAN FRA LBNAGO IRL DOM GHA CMR Retail price, USD per pack Ili ci t t ra de % IRQ VNM COL MMR AUT SWE EST PER ESP ROM POL CHL ISR GBR KEN LAO KHM EGY MKD TZA THA TUN BGD UZB UKR GEO KAZ CHN KWT HRVJOR CZE SVK KOR JPN BLR SRB TUR OMN FIN NLD CHE SGP SAU USA LKA DNK DEU ITA HUN SVN PRT BOL DZAIDN CIV BEL MAR ARG AUS NOR NZL LTU BIH MEX BGRPHLAZE RUS 186 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As indicated in numerous studies and analyses of illicit trade in tobacco products (4–5), the most effective way to tackle the problem is not to forgo tax increases but rather to strengthen the capacity to fight the trade. Therefore, it is important to consider the relationship between good governance and illicit trade. The more capacity a country has to counter illicit trade in general, the lower the level of that trade will be. An index compiled by the Economist Intelligence Unit (EIU), the Global Illicit Trade Environment Index, measures countries’ structural capacity to fight illicit trade overall. The EIU indicator is a combination of four indicators or categories designed to assess countries’ performance in those areas; the closer the overall indicator is to 100, the better the country’s capacity to fight illicit trade. The four categories are:13 1. government policy, which measures the government’s commitment to pro- actively monitoring and preventing illicit trade; 2. supply and demand, which measures the extent to which the domestic en- vironment discourages or encourages supply and demand for illicit goods; 3. transparency and trade, which measures transparency and the degree of governance applicable to free-trade zones and transhipments; and 4. customs environment, which measures how effectively customs services facilitate legitimate trade while at the same time preventing illicit trade. Figure 4.1.2 illustrates the relationship between the EIU indicator and the estimated level of illicit trade in cigarettes in a set of countries. There is an inverse and statisti- cally significant relationship between the indicator and the estimated level of illicit trade in cigarettes. This suggests that as the capacity to fight illicit trade in general increases, the illicit trade in cigarettes falls.14 13 For more details about this indicator, visit http://illicittradeindex.eiu.com/. 14 The association was significantly different from zero at a 90% confidence level, using a linear regression between the two variables. CHAP T ER 4. PO LI T I C AL ECO N OMY 187 Fig. 4.1.2 Share of illicit trade versus the EIU indicator in 70 countries, by country, 2018 Sources: (30–31). % Il ic it tr ad e ci ga re tt es MYS 0 10 15 5 20 25 30 35 40 45 50 55 60 EIU GIT indicator 20 3010 40 50 60 70 80 90 JPN KOR HRV SAU UKR BLR KHM LAO MMR IRQ GTM DOM VNM MAR SRB PER BGR MEX ROM COL ZAF BIH PHL IDN ITA PRT HUN BEL DEU ESP ARG AUS SWECHL TW POL ISR AUTLTU FRA CAN IRL LVA ARE GRC IND URY CRI PAN PAK ECU BRA EST GBR FIN SVN NLD SVK TUN RUS DZA TUR CHNKAZ THA CZE USA SGP NZL DNK 188 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Rigorous independent research has established that despite the challenges of illicit trade, taxation of tobacco products is an effective public health intervention that substantially reduces tobacco use and generates government revenue (5). Further, when cigarette taxes increase, governments generate higher revenue and consumption is reduced (32–33). However, ineffective tax administration can allow illicit trade to grow and can undermine some of the benefits of tobacco taxation by making cheaper cigarettes available. For example, the average street price of smuggled cigarettes in Malaysia is 55% lower than its legal tax-paid equivalent (34). Illicit tobacco trade also reduces government tax revenue and may increase health costs associated with smoking and costs associated with policing. 4.1.4 MEASURING ILLICIT TRADE IN TOBACCO PRODUCTS The magnitude of illicit trade is a powerful argument in tax policy discussions, and for this reason the tobacco industry funds estimation of illicit trade in countries or regions of particular interest to itself (i.e. Project Sun and Project Star in the EU and Oxford Economics in East Asia). However, a recent systematic review of industry data on illicit trade finds substantial methodological weaknesses in industry-commissioned reports (24). Furthermore, Blecher et al. (35) argue that industry-funded studies tend to systematically overestimate the size of illicit trade to persuade authorities to abandon tobacco tax reforms. Independent researchers have also uncovered inconsistencies in tobacco-industry-funded estimates (36). Some examples of inflated industry-linked illicit trade estimates are given in Table 4.3, which compares peer-reviewed and independent studies with estimates funded by the tobacco industry. Because some countries have several industry estimates from different sources or years, Table 4.3 presents the estimate included in the article that published the independent study, because it was considered as representative and adequate to illustrate the overestimation. In all cases, the industry estimates exceed those of the independent studies. Measuring the scale of illicit trade can be a daunting task for governments because different methods are employed by independent researchers, governments and the tobacco industry. Nonetheless, it is worth investing in these studies because they drive policy discussions and can be used to evaluate the impact of policies (e.g. tax increases, plain packaging and health warnings). CHAP T ER 4. PO LI T I C AL ECO N OMY 189 Table 4.3 Illicit market share estimated in independent studies compared with estimates in tobacco-industry-funded studies COUNTRY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDEPENDENT STUDIES SOURCE OF THE INDEPENDENT STUDY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDUSTRY- FUNDED STUDIES INSTITUTION RESPONSIBLE FOR THE INDUSTRY- FUNDED STUDIES Colombia 3.5% of the total market in five cities, 2016 Maldonado et al., 2018 (37) 13% of the total market, 2014 FND and INVAMER, 2015 Chile 16.3% of the total market in the Metropolitan Region of Santiago, 2017 Paraje et al., 2020 (38) 24.3% of the total market, 2017 Observatorio del Comercio Ilícito BATC, 2017 Brazil 28.8% of the total market, 2014 Iglesias et al, 2017 (39) 34%, of the total market, 2014 BAT public statement, 2015 Mexico 8.8% of the total market in eight major cities, 2017 Saenz de Miera Juarez et al., 2020 (40) 16.6% of the total market, 2012 Confederación de Cámaras Industriales, 2012 As shown in Table 4.4, methodologies to measure illicit trade can be grouped into three types: (1) direct measurement; (2) residual methods and (3) expert opinion (12). Direct measurements rely on evidence directly linked to actual illicit behaviour and pack observation; residual methods infer evasion based on theory and evidence about consumption and legal sales; and expert opinion distills information garnered from talking to individuals with the most direct knowledge of the tobacco market. Each method has advantages and disadvantages. No single method is unambigu- ously superior to others, but direct measurement and residual methods are more conducive to determining the size of the illicit market, whereas expert opinion could provide insight into the details of the market’s operations.15 Table 4.4 presents the relative amount of resources and the degree of expertise required to implement each main measurement method, as well as the primary purpose, data collection characteristics, sampling features and unit of analysis. A brief description of each of the methods is presented in Annex 4.1. Merriman (11) and Ross (9) provide more expansive details. There is no simple selection rule for deciding what measurement method to use. The major factors to consider when selecting a method or methods include (1) the nature and characteristics of the illicit trade problem (i.e. where and how the 15 In interviews with experts from the tobacco industry, provisions of Article 5.3 of the FCTC and its Guidelines need to be followed. 190 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N problem manifests and whether domestic tax evasion or illegal inflows of foreign brands or a combination of both predominates), (2) previously collected data, (3) available budget and (4) expertise of available analysts. Available budget and staff skills are often the main restrictions that governments face. Therefore, Table 4.4 orders the measurement methods according to resources needed and available expertise. For example, residual methods and expert opinion can provide crude but useful estimates at low cost and require the lowest levels of technical sophistication. Another low-cost option for countries that employ popula- tion health surveillance surveys is to add questions to measure illicit trade, such as brand name, value and quantities of the last purchase. In contrast, the direct measurement approach often requires sophisticated research designs and expensive (and time-consuming) field research. Table 4.4 Overview of resources and expertise needed and main purpose of measurement methods METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Seizures (D) $ Low Identify trends in types of products, transporta- tion methods, points of entry and brand names Secondary data use Non- probability Shipments Use of existing health sur- veillance surveys – self- reported consump- tion (D) $ Low, only additional questions Size of illicit trade, adding or improving questions on brands, value and quantities of the last purchase Additional primary data collection Probability Individuals Gap analysis (R) $ Medium Provides a measure of changes in illicit trade Secondary data use Universe Nations Econo- metric modelling (R) $ High Estimation of price elasticity of substitution from tax- paid to illicit products Secondary data use Universe Geography Expert interviews (E) $ Low Characteristics of the illicit trade Primary data collection Non- probability Individuals Smoker intercepts and pack observa- tion surveys (D) $$$ Medium Size and characteristics of illicit trade, probability-based sample to be representative of population Primary data collection Probability Individuals CHAP T ER 4. PO LI T I C AL ECO N OMY 191 METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Pack return and swap surveys (D) $$$ Medium Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Littered- pack surveys (D) $$$ Medium Size and characteristics of illicit trade, compa- rability with industry estimation using empty-pack surveys Primary data collection Probability Individuals Covert purchases (D) $$$ Medium Type of products and trade channels of illicit trade Primary data collection Probability Geography Self-report consumer surveys (D) $$$ High Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Notes: Universe includes total population; D = direct measurement, R = residual method, E = expert opinion. Scale for resource costs assuming a moderately sized study (e.g. a representative study of a region of several million): $ (cheapest) – weeks of skilled labour hours; $$ (moderately expensive) – 1 to 2 months of skilled labour hours; and $$$ (most expensive) – 6 to 12 months of skilled and unskilled labour hours. A more detailed description of the different measurement methods is given in Annex 4.1. To further assist responsible authorities in deciding which method to select, Table 4.5 presents the key characteristics of each of the measurement methods, along with the main advantages and disadvantages of each. Countries may begin with methods that require fewer resources and less skills to obtain an overview of the problem. Seizures – which are a by-product of law enforcement efforts – pro- vide a first step, and countries can analyse the information obtained (origin of the products, brands, location, etc.) and report the results to increase public awareness of the problem.16 Alternatively, countries can add questions related to illicit trade to existing and funded health surveillance surveys conducted regularly by health surveillance authorities and statistical authorities. In that way, cooperation in using existing measurement methods between health authorities – the tobacco control office and health surveillance unit – tax and customs authorities and the national 16 Seizures are useful for obtaining qualitative information about the illegal activity, but they have to be treated very cautiously in projecting the size of the problem. Countries may think they have a very large problem because they have competent authorities doing an extraordinary job at finding illicit goods. On the other hand, countries can have less-efficient authorities making few seizures, and in these environments, seizures tell nothing about the size and nature of the problem. 192 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N statistical office could be a starting point for identifying the nature and size of the illicit trade problem in the country. Direct observation of packs has been increasingly implemented in many LMICs, through different types of surveys such as intercepts of smokers or retailers, pack return, littered-pack inspections and covert purchases of cigarettes. These activities have expanded the skills of independent researchers and academia and increased knowledge of these methods. Also, increasingly cheaper digital technologies allow interviewers to take pictures and record pack characteristics in direct observation surveys or in larger national self-report consumer surveys. Table 4.5 Key characteristics, advantages and disadvantages of illicit trade measurement methods METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Seizures (D) Statistics of tobacco products confiscated by local and national authorities Readily available from law enforcement agencies May not provide a representative picture of the size and/or nature of illicit trade Using existing health surveillance surveys to obtain self-reported consumption (D) Adding or improving questions about brand names, quantities, prices, locale of purchase and other factors Produces good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Gap analysis (R) Compare self-reported consumption data with observed (usually administrative) data about tax-paid sales When quality data are available, is simple and easily reproduced (providing for measurements over time) and explainable Data on tax-paid sales and/or consumption are frequently inaccurate and in many cases do not provide information on the size of the illicit market, but only on changes over time Econometric modelling (R) Estimated according to the difference between tax-paid sales and predicted consumption given by the model Because it is consistent with a long tradition of economic theory, empirical estimates can be evaluated Requires high-quality data on a variety of important variables over a period of time and advanced econometric modelling expertise CHAP T ER 4. PO LI T I C AL ECO N OMY 193 METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Expert interviews (E) Experts include researchers (e.g. in economics, criminal justice and public health), journalists, tax and enforcement specialists, product manufacturers and wholesalers Useful for identifying the nature of and trends in the marketplace (e.g. venues where illicit cigarettes are sold, modes of entry), and the interviews can be useful for defining the method to assess the size of the illicit trade Information obtained may not be generalizable, and expert knowledge may be outdated or limited by the experts’ experience; also, experts often have strong biases Smoker/retailer intercepts and pack observation surveys (D) Examining the packs of smokers and cigarette retailers, convenience or probability-based sample Is direct and objective, and smokers do not suffer from any value judgements The difficulty of identifying areas representative of the tobacco use population and sampling important subpopulations such as elderly and immobile smokers, but household surveys could overcome sampling issues Pack return and swap surveys (D) Also a pack observation study using survey sampling techniques to examine smokers’ pack characteristics May decrease the stigma associated with traditional smoking surveys In LMICs, survey distribution may be unreliable because of the mail delivery system Littered pack surveys (D) Also known as empty-discarded-pack surveys; publicly discarded packs bear characteristics (e.g. tax stamps, public health warnings) that indicate whether they are tax compliant Yields estimates that are less likely to be biased from issues of social desirability, recall error and confidentiality Significant budgets could be needed to employ field researchers to collect, code and analyse the data; surveys do not provide information about the smoker and the price paid Covert purchases (D) Uses covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes Directly identifies sources of illicit cigarettes It is difficult to create a sampling frame of retailers for illicit sources or to know what smokers are actually buying and how much Self-report consumer surveys (D) Surveys can be distributed to individuals or households, using various modes of distribution Good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Notes: D = direct measurement, R = residual method, E = expert opinion. 194 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ultimately, when capacity allows, more solid estimates will need to be made using more than one methodology at a given point in time. Ideally, estimates will be made on a regular basis in order to assess the evolution of illicit trade over time and its possible connection to policy changes. 4.1.5 CRITIQUING STUDIES THAT MEASURE THE SIZE OF THE ILLICIT MARKET Measuring illicit trade is a challenge for researchers, industry and governments, because the trade is, by definition, hidden from plain sight. Buyers and dealers sometimes go to great lengths to ensure that their participation in illicit activity is concealed. Over the years, researchers and government agencies have been increas- ingly interested in estimating the size of illicit markets and identifying effective interventions. As consumers of research, governments should critically examine available studies and evaluate them on their scientific rigor and methodological transparency. Measurement issues are particularly acute with respect to the illicit tobacco trade because it is a politicized topic. High estimates may raise questions about the tobacco industry’s ability to control the supply chain, its involvement in illicit diversion, the impact of taxation policies and the effectiveness of enforcement strategies. While the industry has portrayed itself as taking an active stance in measuring and fighting illicit trade (e.g. Project Star, conducted by KPMG LLC but paid for by PMI, later followed by Project Sun), in the past it has used smuggling as a strategy to enter closed markets – for example, in China and Russia (19, 41). Govern- ments should carefully scrutinize evidence about the illicit tobacco trade produced by industry or quasi-industry sources and are advised to seek alternative evidence. Quasi-industry reports are studies commissioned by the industry but published by private research companies (e.g. Ernst and Young, Oxford Economics) (11). Characteristics of good analyses One of the main characteristics of a good analysis is scientific rigor, which involves the use of relevant theoretical frameworks, sound statistical methods and examination of the robustness of findings (e.g. sensitivity analyses). High-quality research reports provide transparent explanations about their methodology and statistical analysis steps undertaken, as well as supplementary analyses that established the robustness of the findings. For example, Joossens et al. (25) clearly describe the data sources used (limitations and advantages and where they can be found) and calculations performed on the number of lives that would be saved if the global market share of illicit cigarettes was eliminated. Explanations should be detailed enough to allow future researchers to scrutinize the analysis and replicate the findings. Replicability is another hallmark of good science. For example, littered-pack studies should detail where and when data collection took place, how many packs CHAP T ER 4. PO LI T I C AL ECO N OMY 195 were collected per geographical unit, the protocol of identifying the illicit packs (e.g. characteristics of the warning labels, brands, tax stamps, etc.) and details of statistical analyses. There should also be explanations of the representativeness of the selected geographical areas. Failure to provide this depth of information may call into question the generalizability of a study and whether there are faults with the chosen method. In the context of policy decisions regarding illicit trade, the most useful data provide information about a representative sample of individuals and geographies. Studies that are limited to, for example, one group of individuals based on specific characteristics or a given geography may yield biased information. Research reports also should be clear about the study’s limitations. For example, studies that measure illicit trade often do not measure product counterfeiting and do not include non- cigarette tobacco products in their estimates (42). When statistical estimates are included, they should provide confidence intervals as well as point estimates to account for uncertainty resulting from simple random chance (11). Characteristics of flawed analyses Flawed analyses can convolute and distort scientific knowledge about illicit trade. Flaws usually manifest in the data, methodology, statistical analysis and/or interpreta- tion of the results (11). Studies may be purposefully designed with methodological flaws to yield high or low estimates of the trade. For example, research showing that illicit trade constitutes a large share of the total market may be used to support arguments that taxes cause sharp increases in illicit trade, whereas lower estimates may be used to support arguments that certain governmental interventions (e.g. increased retail inspections) are effective. Pressures to skew data may also be tied to funding. For example, high estimates can sway governments to provide more resources for law enforcement activities. Analyses can be purposefully skewed by using data sources or data collection methods that will provide biased estimates. Flawed studies sometimes provide incomplete or inaccurate descriptions of their methodology. They may lack detail regarding the quality of the data used or information about how the data were collected and analysed. For example, a common weakness in industry-funded research on discarded packs is that the methods of collection and forensic analysis are not reported, ostensibly because they are “proprietary” information (43). However, these methodological details are key to assessing whether a study’s findings are biased by sampling error, model misspecification, measurement error, non-response or other flaws. It may be impossible to assess measurement error if researchers fail to disclose questions included in a survey instrument. Survey items used to measure the illicit tobacco trade may be imprecise. For example, asking respondents the frequency with 196 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N which they purchase “cheap” cigarettes may yield biased estimates, particularly if consumers can purchase cigarettes at discounted prices by using coupons. To more ac- curately measure tax evasion, surveys must include questions about the location of last purchase, purchase price, presence of public health warning labels and brand names. There are other ways that flawed studies can inadvertently or purposefully distort estimates of illicit trade. For example, data collectors can intentionally oversample areas known to be hot spots of illicit sales or sites that residents from lower-tax jurisdictions visit. Researchers can collect discarded cigarette packs close to the borders of countries with lower taxes to (inadvertently or purposefully) demonstrate the undesired side effects of tax policies. Studies published in non-peer-reviewed or lightly peer-reviewed outlets such as edited book volumes or policy briefs should be viewed with more scepticism than those published in highly regarded peer- reviewed outlets. 4.1.6 CONCLUSIONS Globally, the illicit tobacco trade continues to be a major concern for tax admin- istrators because of the challenges it generates to collecting higher revenues as well as the challenges to accurate and independent measurement. Industry figures provide distorted conclusions regarding the extent of the problem – frequently with a monocausal explanation of the link between illicit trade and tobacco taxation. Illicit trade comprises multisystemic issues and requires multiple strategies. Worldwide, countries at different levels of economic development have implemented a variety of effective measures to combat the illicit trade in tobacco products. The Philippines and the United Kingdom, for example, have addressed illicit trade as part of their overall tobacco tax reform (4). Price (and tax) levels are not a key determinant of illicit trade, the presence of which is exacerbated by the lack of tax administration capacity. Refraining from increasing taxes is not the solution; countries should instead respond with a com- prehensive strategy that includes at least these three main components: 1. It should identify – independently from the industry – the nature and dimen- sions of the problem. It is necessary to assess scientifically and with the best statistical practices the size of the illicit trade to understand the characteristics and scope of the problem. 2. It should identify and implement appropriate policies and strategies targeted at addressing the specific type of illicit trade the country is experiencing. It should address directly the country-specific institutional and/or governance challenges – as well as the lack of multilateral coordination that can exacer- bate illicit trade – and improve tax and customs administration practices as described in Chapter 3. CHAP T ER 4. PO LI T I C AL ECO N OMY 197 3. It should implement best practices contained in the WHO FCTC Protocol to Eliminate the Illicit Trade in Tobacco Products and accede to the Protocol if the country is not yet a Party. There are proper methods and policies with which to address the illicit tobacco trade. If countries start implementing the appropriate policies, they can raise tobacco taxes and reap health and revenue benefits even in the presence of illicit trade. 198 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.2 SCARE TACTIC C: COURT AND LEGAL CHALLENGES 4.2.1 INTRODUCTION The tobacco industry views well-designed and significant tax increases as a threat to the profit, growth and long-term sustainability of its business. As noted by PMI in 1985: Of all the concerns there is one – taxation – that alarms us the most. While [other restrictions] … do depress volume, in our experience taxation depresses it much more severely (44). The industry is, however, less likely to launch direct legal challenges to excise taxes than to other tobacco control measures (see Box 4.2.1 for details), because taxation – and excise tax in particular – is a comparatively well-established regulatory measure; in many jurisdictions, taxes have been levied on tobacco products for more than a century. There is also less unanimity in opposition to taxes among tobacco industry actors, because differences in the market position of different tobacco companies affect their interests in tax policy. This, in turn, decreases the likelihood that they will act collectively on the issue (45). BAT’s stated strategy in the early 1990s was to influence governments with regard to the level and structure of tobacco taxation in order to promote market growth and to secure competitive advantage (46). Nevertheless, tobacco industry actors will still legally challenge, or at least legally threaten, significant tax measures when vulnerabilities in their design, adoption or implementation are apparent. Box 4.2.1 Court and legal challenges to tobacco tax measures Evidence suggests that the tobacco industry and its allies instigate fewer legal actions against tax measures than against other tobacco control measures: 1. The Campaign for Tobacco-Free Kids’ tobacco control laws database contains only a handful of cases concerning tobacco tax measures, but hundreds on other tobacco control topics. This pattern can also be seen in a 2018 review of tobacco control legal challenges that examined this and two other databases to select 96 cases relevant to the question of the WHO FCTC’s usefulness in litigation (47). Only 6 of these 96 cases were challenges related to tax measures. CHAP T ER 4. PO LI T I C AL ECO N OMY 199 2. A 2013 systematic review of empirical studies on tobacco industry interference with tobacco tax policy found that only 9 of 36 relevant articles reported the specific use of litigation as a tobacco industry tactic (1). All 9 concerned constitutional challenges to earmarking provisions for tobacco tax initiatives in the United States (1). 3. A 2015 study on industry interference in LMICs cited legal challenges to tobacco control measures in 15 countries as examples of industry interference, but none of the challenges concerned a tobacco tax measure (48). 4. A 2016 analysis of papers published in systematic reviews of industry inter- ference with tax and marketing measures found that only 5 of 65 papers concerning tobacco tax related to the use of litigation or threats of litigation to interfere with tobacco tax measures (49). The tobacco industry makes extensive use of legal experts (1, 50–52) who study all relevant laws and regulations closely to determine their likely and arguable boundar- ies for the purpose of manipulating regulations and regulators (1, 50–52). Based on this expert advice, tobacco companies know when regulations remain within the bounds of both international and domestic obligations but can still argue that legally permissible tobacco control measures would be defeated in litigation if passed (48, 51–52). As the threat of a legal challenge alone can be used to the industry’s advantage, recourse to litigation is seldom needed or desirable (1, 45, 48, 51, 53–56). Even when litigation is launched, the objective may be to delay or weaken a measure rather than to win on the merits of the case (1, 45, 48, 53). To counter actual and threatened legal challenges, policy-makers need to be aware of relevant legal obliga- tions when preparing and implementing tobacco control measures. Fortunately, the tobacco industry playbook is relatively predictable. Tax and other tobacco control measures can thus be designed to strengthen the regulators’ legal position against genuine threats and enable them to dismiss baseless industry threats. 4.2.2 COUNTRY EXPERIENCES WITH LEGAL CHALLENGES TO TOBACCO TAXATION Legal obligations that are relevant to tobacco taxation include those under do- mestic law and international instruments such as international trade agreements and international investment agreements (IIAs).17 Some of the legal issues that a tax measure may encounter are outlined in Table 4.6. Case studies from various countries illustrate how these legal issues have and have not been avoided in the 17 Relevant international trade agreements include the WTO Agreement and custom unions such as the EU, the East African Customs Union and Mercosur. Relevant IIAs include bilateral investment treaties and the investment chapters in free trade agreements and within custom unions. 200 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N passage, design and implementation of tobacco taxes. These issues are not the norm, however, and should not give rise to undue apprehension. The case studies are rated as positive, mixed or negative based on the extent to which the legal decision upheld the taxation measure in question. Table 4.6 Potential legal issues for tobacco tax measures VULNERABILITIES LEGAL OBLIGATIONS CASE STUDIES Inadequate consultation and other procedural vulnerabilities Domestic procedural law 1, 2 Due process protections for investors under IIAs None Procedural requirements under WTO Agreements and Custom Unions 3 Discrimination against imports or investors Nondiscrimination obligations under WTO Agreements and Customs Unions 8, 9, 10 Nondiscrimination obligations under IIAs 11 Investment incentives or inducements Arbitration mechanisms under investor-state contracts 12 Fair and equitable treatment clauses of IIAs None Other substantive breaches Constitutional rights and restrictions on taxation 4 Statutory restrictions on the imposition of taxation 6 Expropriation clauses of IIAs 5 Ultra vires (the scope of legal authority) 7 Avoiding procedural vulnerabilities in tax laws Procedural defects can be avoided by taking great care in progressing and imple- menting regulatory or legislative provisions. Procedural concerns pose a dilemma for tobacco control regulators. Article 5.3 of the WHO FCTC and the COP guide- lines for its implementation state that policy-makers and regulators should interact with the tobacco industry only when and to the extent strictly necessary (57). For taxation measures, interaction might be necessary because consultative and de- liberative processes could be prescribed under domestic constitutional provisions and procedures for good governance, due process requirements of IIAs and some international trade agreements. The tobacco industry may use these requirements as leverage to delay, distort or hijack the rule-making process in contravention of Article 5.3. Accordingly, interactions with the tobacco industry should be limited to strictly necessary consultation conducted in a transparent or public manner but with care that this does not come at the expense of a measure’s defensibility. The proper balance will depend on the jurisdiction in question, since constitutional, statutory and applicable international legal obligations vary. CHAP T ER 4. PO LI T I C AL ECO N OMY 201 CASE STUDY 1 (MIXED): Industry manipulation of legislative procedures In 2012, a bill stipulating, among other things, the creation of a new specific excise tax on cigarettes passed its final reading in Costa Rica’s Legislative Assembly. Passage of the bill had, however, proceeded under “urgency” and notwithstanding a pending constitutional enquiry (a constitutional query is meant to prevent passage of a bill).18 ISSUE MAJORITY DECISION MINORITY DECISION LESSON Whether the court could consider the enquiry despite passage of the bill and the effect the bill’s passage could have despite the enquiry. The enquiry was taken up by the Supreme Court’s Constitutional Division’s majority (58). The signing and publication of the bill by the executive was suspended by the Constitutional Division pending their decision on the merits of the case – which, in the end, found any question of the bill’s constitutionality baseless (58). The enquiry was inadmissible by reason of having been filed too late and notice of its filing having not been received by the legislature prior to the reading of the bill (58). In disagreement with the majority, the minority held that the court could not consider the enquiry or suspend the bill’s signing by the executive – the final step in becoming law. This challenge demonstrates how the tobacco industry’s defenders may attempt to frustrate and impede a tax measure’s passage. In this case, the challenge seemed to have been a delaying tactic, as it was posted on the same day as the final reading of the bill. Its authors may have either wanted its pending nature to cause the legislature to delay or, as occurred, to create conditions for a procedural and constitutional challenge in the absence of delay. All the grounds of the challenge itself were found to be without merit. Although such frivolous challenges cannot be prevented, they can and should be anticipated to ensure that they do not lead to a tax measure’s defeat. 18 “Urgency” is a procedure under which a bill is progressed through a legislature in an expedited fashion. 202 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N CASE STUDY 2 (POSITIVE): Adhering to domestic procedural requirements Kenya’s tobacco control regulations required the tobacco industry to pay a levy to compensate the state for health care and other negative externalities of smoking. In a 2016 challenge brought against these regulations, the plaintiff, BAT, was unsuccessful on every count (59–60). Even though the levy was not considered a tax measure by the court, the case study is instructive on how regulators may safeguard tax measures against procedural challenges. ISSUES LAWS AND ARGUMENTS DECISIONS LESSONS Whether the government’s consultations on the measure were adequate. Asserting that the Constitution and the Statutory Instruments Act together meant that “appropriate consultations with persons who are likely to be affected” were required because of the measure’s likely substantial effect on business. BAT claimed that this standard was not met. Kenya’s government claimed that it was under no obligation to undertake special or extensive consultation with the tobacco industry. The judge found in favour of Kenya’s government, noting that (1) the requirement to consult does not imply that any particular view needs to prevail; (2) dissatisfaction with the level of consultation is not decisive; (3) on the facts, industry was allowed, and often invited, to send representatives to all relevant public consultative meetings and parliamentary committee hearings; and (4) consultation on the regulations was adequate (59). The tobacco industry carefully scrutinizes legislative and regulatory processes for defects. In this case, Kenyan government officials appropriately distanced themselves from the tobacco industry by not permitting its representatives special consideration but did permit their attendance at public meetings and the ability to submit their views under usual procedures. In this way, both the principles behind WHO FCTC Article 5.3 and the requirement for consultation under Kenyan law were observed. CHAP T ER 4. PO LI T I C AL ECO N OMY 203 Avoiding procedural issues in tax administration CASE STUDY 3 (NEGATIVE): Contravening procedural requirements in international obligations In 2010, a WTO panel held that Thailand violated the Customs Valuation Agreement (CVA) by the process it used to value cigarettes that Phillip Morris (PM) Thailand imported into the country from a related party, PM Philippines. Customs values are important as they are the tax base for tariffs and can feed into the base for other taxes levied against the value of the good, such as ad valorem excise taxes and VAT. Transaction values declared by PM Thailand were rejected by Thai tax authorities as influenced by the relationship between the parties and a customs value determined by deduction was substituted (61). ISSUE LEGAL OBLIGATION DECISION LESSONS Whether Thailand adequately consulted with PM Philippines before rejecting its declared transaction value (61). The CVA requires good faith exchange of reasons and information, with opportunities for response (61). Thailand had failed to properly explain its reasons for rejecting the transaction value, as well as its belief that price was influenced by the relationship between the two parties (61). This was a violation of the CVA. Thailand did not appeal these findings. Thailand’s authorities needed to take greater care in their dealings with the tobacco industry to ensure they met the pertinent procedural obligations. In this instance, a specific and high standard of consultation – the provision of detailed reasons and an opportunity for response – was prescribed by the CVA and Thailand failed to meet it. 204 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ensuring compliance with substantive requirements Rules found in domestic and international law also establish substantive obliga- tions. This subsection focuses on substantive obligations found in domestic law. International obligations concerning discrimination and investment incentives are considered in the next subsections. CASE STUDY 4 (POSITIVE): Tax measure found to be consistent with the Constitution The Chilean government introduced a substantial increase in tobacco and fuel excise, and in 1995, a coalition of taxpayers brought a Constitutional challenge to the measure (62). ISSUE DECISION LESSON Whether the tax was, per article 19 of the Chilean Constitution, “obviously disproportionate or unjust” (62). The excise tax increase did not violate the Constitution, as it was neither confiscatory nor manifestly irrational. Generally applicable excise taxes are not vulnerable to challenges for being excessive, unfair or disproportionate. CASE STUDY 5 (POSITIVE): Failure to grant tax rebates not an expropriation under an IIA This case study is an example of a claim for breach of an expropriation clause in an investment treaty. Such clauses protect foreign investors against measures that can be construed as directly or indirectly seizing an investment or depriving it of its value (63). In the case, an investor was, for more than a decade, denied tax rebates by the Mexican government. This affected the profitability of the business of purchasing and reselling Mexican cigarettes abroad, and the investor brought the claim to an investment agreement arbitral tribunal in 2002 (64). ISSUE DECISION LESSONS Whether Mexico’s failure to grant rebates to the investor exceeded the bounds of valid regulation to constitute indirect expropriation of the investor’s investment (65). There was no expropriation. The arbitral tribunal noted that not all business problems are violations: the investor had no right to participate in the “grey market” export of cigarettes and there were sound reasons to restrict that market (65). Further, the investor was able to participate in other business ventures and actually continued to have business success (65). Claims of indirect expropriation made under IIAs are unlikely to be successful, as generally applicable tax measures are a legitimate form of regulation. A mere loss of profit will not suffice. Claims of expropriation will not succeed unless a substantial or significant deprivation of the investment results. CHAP T ER 4. PO LI T I C AL ECO N OMY 205 CASE STUDY 6 (NEGATIVE): A regulation contrary to superior domestic legislation In 2011, an Indonesian tobacco industry association group, FORMASI, challenged a new excise regulation. Since 2009, the government had been implementing a tiered specific excise tax system based on a set of characteristics (size of production, type of cigarettes and price levels). In 2011, excise rates were increased in nearly all of the 19 tiers, but the reference prices were not accordingly adjusted. This gave rise to a legal issue. ISSUE DECISION LESSONS Whether new excise regulations breached a 57% ceiling for the rate of excise on the retail sale price of tobacco products under the superior Excise Law (66–70). The challenge specified that excise exceeded this ceiling for hand-rolled domestic clove cigarettes (kreteks) (68, 71). The Court found in favour of the tobacco industry association, and the government was required to immediately revoke the 2011 regulation. (69–70). It is advisable to stay within the rules and be aware of legal hierarchies – including superior domestic legislation. The tobacco industry scrutinizes all increases in tobacco taxes. In this case, a breach of a legislative requirement for a single category of tobacco product resulted in Indonesia suffering lost revenue and a setback in its efforts to reduce tobacco consumption. Ensuring a tax measure is within an authority’s legal power A tax measure is ultra vires when it goes beyond the legal power of the enacting body. As with case study 6, this is a legal issue that involves legal hierarchies. In ultra vires cases, however, instead of centring on conflict between inferior and superior law, the issue is whether an authority that enacts a tax measure is authorized to do so. This issue may arise when a tax measure is enacted by a subnational jurisdiction or by an executive acting under a statutory delegation. CASE STUDY 7 (NEGATIVE): Tobacco taxation contrary to the Australian Constitution ISSUE DECISION LESSONS Whether New South Wales’ licensing and penalty fees regime constituted an excise tax by other means contrary to the Australian Constitution’s exclusive grant of that power to the federal government (72). The court found that state licensing fees were excise taxes and that this was contrary to the Australian Constitution (72). Authorities enacting tobacco tax measures must act within the scope of their legal power. 206 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Avoiding discrimination against imports and foreign investments Although inherently discriminatory, customs duties may be used subject to the agreed upper limits in a country’s trade agreements. Excise and other taxes designed with the aim of raising tobacco prices to reduce demand and advance human health should be origin-neutral: they should not seek to tax local products less than foreign products or aim to treat foreign products differently from one another. Tobacco tax measures are pursuing objectives other than health when they aim to raise the price of imports more than that of local products or seek to burden favoured market participants less than others. Solely health-protective tobacco taxes will not ordinarily violate Articles III:2 and I:1 of the GATT (the WTO’s General Agreement on Tariffs and Trade), which prohibit discriminatory taxation (in light of general exceptions). Nor will solely health-protective tobacco taxes directly violate anti-discrimination protections for investors found in the national-treatment (NT), most-favoured-nation (MFN), expropriation and fair-and-equitable-treatment (FET) clauses of IIAs (63, 73–74). It is possible to make claims for breach of international obligations on grounds other than discrimination, but such claims are generally highly unlikely to succeed. CASE STUDY 8 (NEGATIVE): BAT v Uganda (2017 East African Court of Justice) DISCRIMINATION LESSON Uganda established a higher level of excise taxes on imported cigarettes – including those from Partner states of the East African Customs Union (75) – than on local cigarettes. Its implementation was discrimination contrary to Article 15 of the Customs Union Protocol (75). Differential taxation explicitly based on origin can be construed as protectionist discrimination in violation of international obligations. The tobacco industry can also turn to international trade agreements outside of the WTO – in particular, customs union mechanisms. Difficulties arise when ostensibly origin-neutral and health-protective tobacco taxes result in dissimilar taxation of tobacco products (73). Discrimination does not exist simply because there is dissimilar taxation – the taxation must adversely impact imported goods more than local products, the imports of one nation more than another or a particular investor’s products more than comparable products. Where dissimilar taxation between product categories results in discrimination, the tax will ordinarily still be lawful if the dissimilar taxation is based solely on a legitimate regulatory distinction between the product categories in question.19 19 The precise applicable rules vary depending on the nature of the legal obligations in question. Under the GATT, dissimilar taxation of like or directly competitive products can be justified based on scientifically grounded distinctions between products under Article III:2 and, in the alternative, discrimination that is necessary under the explicit carve-out for health-protective measures, Article XX(b) (73). For the MFN and NT clauses of IIAs, differential taxation can be argued as nondiscriminatory on the basis that difference in harm means the products are not “alike” or, in the alternative, discrimination is justified based on scientific evidence of differences in harm and rational reasons for the health-protective role of differential taxation (63). CHAP T ER 4. PO LI T I C AL ECO N OMY 207 Where discrimination is inadvertent, lack of an intention to discriminate is not sufficient as a defence for breach of obligations under IIAs or the GATT (73, 76). Policy-makers should carefully scrutinize measures to determine: 1. whether an aspect of a tax measure’s design or implementation may be more to the detriment of imports or foreign investors than of local products or domestic investors; 2. whether the potentially discriminatory aspect of the tax measure serves any useful purpose in supporting the tax measure (i.e. it is needed to achieve the health goal); 3. whether there is any reasonable alternative that could achieve the same effect without the potential for discrimination; (i.e. it is indispensable) and 4. when it is needed and indispensable there is a good chance that it will be defensible. The case studies below provide examples of discrimination arising in connection with a tobacco tax measure. CASE STUDY 9 (NEGATIVE): Thailand – Customs and Fiscal Measures on Cigarettes from the Philippines (2010 WTO panel) The facts of this case are presented in case study 3. This case study examines claims of discrimination rather than the procedural issues. DISCRIMINATION EXPLANATION LESSONS Thailand implemented its policy for determining the tax base for VAT on cigarettes inconsistently (61). Thailand applied a methodology in fixing the tax base, in particular a marketing cost component, of imported cigarettes that differed from that for local products (61). This resulted in the marketing cost component for the imported cigarettes being higher than it would have been under the general methodology. This difference in treatment was insufficiently justified and therefore considered discriminatory. As there is potential for inadvertent discrimination when the base for an ad valorem tax is fixed, tax base determinations must be consistent and well- reasoned (61). This case study demonstrates how policy- makers need to take care in designing and implementing ad valorem taxes to ensure they are nondiscriminatory and legally defensible. 208 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Thailand’s VAT rebate policy imposed a potentially higher tax burden and also created more burdensome administrative requirements for imported cigarettes (61). Resellers of cigarettes produced by a government entity were granted an exemption from VAT (61). Although resellers of imported cigarettes would be eligible for a tax credit on their VAT, this was not an automatic process (61). The distinct treatment of resellers of imported cigarettes and those of local cigarettes resulted in the risk that there would be a higher VAT burden for the former (61). The distinct treatment also imposed an additional administrative burden on resellers of imported cigarettes and altered conditions of competition (61). Rules for the collection and enforcement of tax obligations should be the same, or as similar as practicable, in both form and effect for domestic and imported tobacco products. CASE STUDY 10 (NEGATIVE): Dominican Republic – Measures Affecting the Importation and Internal Sale of Cigarettes (2004 WTO panel; 2005 WTO Appellate Body) Under article XX(d) of the GATT, discrimination that is necessary to secure com- pliance with a legitimate tax measure will be justified provided there is no less- discriminatory alternative. In this case, this justification was used unsuccessfully. DISCRIMINATION EXPLANATION LESSONS The Dominican Republic’s tax stamp regulations were discriminatory towards imported goods (77). Under the regulations, all cigarette packs had to be affixed with tax stamps, but imported cigarettes were to be affixed with tax stamps under the supervision of local tax authorities following importation, while locally manufactured cigarettes could be affixed with a tax stamp in the course of production. This de facto distinction between local and imported products modified the conditions of competition to the detriment of imported cigarettes by (1) increasing costs for importers and (2) impairing the aesthetics of imported products (77). The panel did not consider this discrimination justified: it was not necessary for the enforcement of tax measures, because less restrictive alternatives were available such as permitting importers to affix tax stamps during the course of production (Dominican Republic – measures affecting) (77). The panel’s findings were upheld on appeal (78). Policies crafted to ensure compliance with tax measures need to also be nondiscriminatory. Discrimination claims can arise when compliance costs are higher for imports than for local products and this de facto distinction is avoidable. It is important to consider whether less burdensome alternatives may achieve the same objective. CHAP T ER 4. PO LI T I C AL ECO N OMY 209 CASE STUDY 11 (NEGATIVE): Feldman Karpa v Mexico (2002 ICSID [International Centre for Settlement of Investment Disputes] Arbitral Tribunal) Arbitral tribunals have accepted differences in treatment accorded to investors protected by IIAs when there is a legitimate connection between the distinctions drawn and public welfare objectives (76). The facts of this case are presented in case study 5. This case study examines aspects of the case involving the investor’s claim of discrimination, rather than the substantive issue of expropriation. Claims of discrimination are made on different grounds than claims for expropriation, which is why the case was decided differently on this claim. DISCRIMINATION LESSONS Denial of foreign investors’ claims for tax rebates. Tax rebate claims were granted to similar local investors (65), which was a violation of an IIA’s national treatment clause (65). Foreign and local investors must be treated similarly, and consistent and well-documented policies must be used to guide administrative decisions. The denial of the rebates may have been justified, but the government was unable to establish this due to a lack of documentation. Avoiding the investment incentives trap Investor-state contracts between the tobacco industry and governments should be avoided. They are not merely “contractual” in the domestic law sense, as even in the absence of an applicable IIA, they can be internationalized to provide inves- tors the right to (1) remove dispute settlement from the state’s court in favour of independent arbitration and (2) remove the dispute from the state’s legal framework in favour of general principles of law (63, 76). Commitments under these clauses cannot, therefore, be legislatively moderated or extinguished, nor can liability be limited within domestic courts that may be more likely to favour the state’s right to regulate in favour of public health (76). Investor-state contracts and other noncontractual inducements can be further internationalized by umbrella clauses within IIAs. Such clauses make reneging on undertakings assumed towards investors a breach of the IIA (76). Moreover, even in the absence of an umbrella clause, contracts and inducement can underpin a claim for legitimate expectation and breach of fair and equitable treatment and can also strengthen an investor’s claim for indirect expropriation (63). Arbitral awards make clear that although taxes can be expected to vary and tobacco will be regulated, investors can have the legitimate expectation that states will abide by formal inducements and written contractual undertakings. A common clause within investor-state contracts, the stabilization clause, is ruinous to evidence-based tobacco control’s most effective measure: excise tax in- creases. Stabilization clauses purport to freeze specific domestic law from the time 210 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of investment (63). Seemingly less onerous, economic equilibrium clauses require contracting states to compensate for regulatory changes that negatively affect an investment’s value (63). There is little difference in effect between these two types of clauses: liability for the cost of breaching an equilibrium economic clause can be onerous enough to make it fiscally challenging and politically unpalatable. CASE STUDY 12: An investor-state contract A state entered into an investment agreement with a TTC in 2001 on the privatiza- tion of its state-owned tobacco enterprise and creation of a joint venture. This investment was to provide economic benefits under the agreement: the joint venture would increase exports and profit using the TTC’s cash and expertise while also ensuring prioritization of local employment, manufacturing and resources. The final investor-state contract included a form of economic equilibrium clause under which any increase in the excise tax rates applied to the company’s tobacco products before a set date would be compensable. While the agreement was not removed from the state’s law, it provided for independent arbitration in case of a dispute over its compensation. In addition, there is a bilateral investment treaty between the host state and another state in which the TTC’s subsidiary has residence that includes a FET clause – this could buttress, if needed, the protection provided by the stand-alone arrangements of the investor-state contract. There were similar less-formal inducements offered to a separate TTC. The extent to which incentives have been granted to the tobacco industry is unknown, but contracts and inducements are likely to be offered in the context of the privatization of state-owned tobacco interests and in dealings between investors and state-owned tobacco enterprises (63). Although countries have been entrapped by their incentives to industry, the investor-state contract provides the clearest example of how undertakings and inducements with the tobacco industry under- mine tobacco control (56, 79–80). States should avoid offering industry incentives and, in particular, entering into contractual undertakings with the industry. More systematically, government should consider avoiding IIAs that elevate incentives and inducements above sensible and reasonable regulation. 4.2.3 CONCLUSIONS Health-protective and origin-neutral tobacco excise taxes are legally defensible, and industry threats are usually baseless. There are, however, certain rules governing procedure, design and consultation that governments may need to consider: 1. Governments should be aware of the standard of consultation required under do- mestic law and any applicable international obligations (case studies 1, 2 and 3). CHAP T ER 4. PO LI T I C AL ECO N OMY 211 It is important to distance the tobacco industry from the policy-making process to the extent that this is permissible. Do not grant the industry special consideration, but do ensure that it is consulted with as required – for example, by providing public meetings, timely information and the ability to submit industry views – while being aware of potential procedural manipulation (case studies 1, 2 and 3). 2. Excise tax is generally safe from challenges that claim it is confiscation or expropriation under domestic or international law (case studies 4 and 5). But express limits on taxation can be found in other laws or a country’s constitution or in the limits of the power to tax granted to an authority (case studies 6 and 7). 3. Explicit and de facto discrimination against foreign tobacco products or investors must be avoided in the design, implementation or enforcement of tax measures (case studies 8, 9, 10 and 11). Legal issues may arise not from the tax measure itself, but rather from ancillary measures that support its implementation (case studies 9 and 10). 4. Explicit differentiation between products based on their effect on health may be challenged as discrimination if it falls heaviest on imported products and has to be justified on the basis of evidence of impact on health and a lack of alternatives. 5. Investment incentives in the form of inducements or contractual undertak- ings should not be offered, as these may be binding (case study 12) or may ground a challenge under an IIA; they are also contrary to the WHO FCTC Article 5.3 Guidelines. 212 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.3 SCARE TACTIC A: ANTI-POOR RHETORIC (REGRESSIVITY) 4.3.1 INTRODUCTION In their efforts to lobby against tax increases, the tobacco industry and its affiliates often claim that increases in tobacco taxation will hurt the poor (81–82). This argu- ment is based on the concept of regressivity in relation to taxation. Conceptually, a tax can be regressive if it means that lower-income people pay a greater proportion of their household income to meet the tax burden than do wealthy people. In other words, the tax burden tends to be relatively higher for lower-income households than for middle- and high-income households. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use. Second, higher tobacco taxes and prices can induce behaviour change among the population, as reflected in the price elasticity of demand (83–84). In combination, these broader considerations effectively make tobacco taxation a progressive – rather than regressive – public health intervention. 4.3.2 REGRESSIVITY AND THE BROADER PERSPECTIVE In a narrow sense, tobacco taxation can be seen as regressive because lower-income people must allocate a relatively greater proportion of their household income than wealthy people to pay for tobacco products when those products become more expensive following a tax increase. In many countries, people from lower-income groups use tobacco more than other people (85). A systematic literature review by WHO found a robust association between lower income and a higher prevalence of current smoking among adults, both men and women (86). This finding was consistent across three decades of studies, across most geographic regions and across countries of different income classifications. For example, in India, high rates of tobacco use – i.e. use by more than 30% of the adult population – are found only in lower-income states such as Assam and Odisha, where net state domestic product is still below 100 000 rupees per capita (see Fig. 4.3.1) (87). CHAP T ER 4. PO LI T I C AL ECO N OMY 213 Fig. 4.3.1 Relationship between adult tobacco use and net state domestic product per capita in states and union territories of India, 2016–2017 Source: (87). However, this finding does not account for broader health and economic factors that determine the full impact on households. Tobacco taxation can in fact be viewed as a progressive – or pro-poor – policy when these wider considerations are properly ac- counted for and explained. In terms of health concerns, the relatively high use of tobac- co among low-income populations translates into a much greater burden of tobacco- attributable diseases for these populations, including higher morbidity and mortality. Low-income groups are also less able to afford medical care to treat tobacco- attributable diseases, and large out-of-pocket medical expenditures can further impoverish many families. Consequently, many poor individuals do not get or even seek the medical care they need. One study found that in Bangladesh, 55% of patients diagnosed with a tobacco-attributable illness did not seek further medical care. This lack of health care utilization was attributed in part to prohibitively high out-of-pocket treatment costs (88). The combination of high rates of tobacco use and lack of access to affordable medical care means that tobacco use measurably contributes to the poverty rate in a number of high-tobacco-burden countries, including China and India (89–90). A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India 214 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The poor are also known to be more price-sensitive than the wealthy; lower-income smokers exhibit higher price elasticities than their higher-income counterparts. This is demonstrated in recent studies by the World Bank, findings of which are shown in Fig. 4.3.2 (91). The poor respond more strongly to higher tobacco taxes and prices by reducing their use of tobacco products more than others, and thus they benefit disproportionately in terms of avoiding tobacco-related deaths, diseases and associated medical costs. A similar conclusion was drawn in a systematic review of the population impact of tobacco control policies on socioeconomic inequities in high-income countries at the late stage of the tobacco epidemic (92). The review found 16 relevant studies relating to taxation, only one of which found a regressive association between tax and the social economic gradient (seven found a progressive impact, while the others produced mixed results). Fig. 4.3.2 Price elasticity of tobacco consumption, medium estimate, by decile Source: (91). This wider economic perspective is explained in the World Bank’s Extended Cost- Benefit Analysis (ECBA) framework, which assesses the distributional impact of tobacco tax increases on health, among other factors (82, 83). That is, the ECBA framework looks beyond the simple or partial definition of regressivity (i.e. impact El as tic it y Deciles -1.2 1 2 3 4 5 6 7 8 9 10 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina CHAP T ER 4. PO LI T I C AL ECO N OMY 215 on household expenditure by income levels) to capture the full distribution of benefits, including improved health and income. The ECBA framework has been applied in studies of various countries, including Bangladesh, Bosnia and Herzegovina, Chile, Indonesia, Republic of Moldova, South Africa, the Russian Federation, Ukraine and Viet Nam. The evidence from these studies supports the view that effective tobacco tax policies can generate pro-poor and welfare-improving outcomes. When reductions in medical expenditures and additional years of working life that result from lower smoking-related mortality are taken into account, the overall policy of tobacco tax increases becomes progressive rather than regressive (see Fig. 4.3.3) (84). A similar conclusion has been reached in studies of high-income countries, such as the United States, where a tobacco tax increase was enacted in 2009 (93). Fig. 4.3.3 Impact of a 100% price increase, with medium elasticities, by deciles Source: (91). Tobacco tax increases will also often lead wealthier smokers to contribute relatively more than poorer smokers to the overall amount of tax revenue collected. This is because poorer smokers reduce their consumption the most, since they are more price- sensitive and wealthier smokers also tend to purchase premium (higher-priced and In co m e ga in s (% ) Deciles -1 2 3 54 6 7 8 9 10 0 1 2 3 4 5 1 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina 216 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N taxed) tobacco products (94). Hence, higher tobacco taxes can be seen as progressive in terms of additional revenue collection and health gains resulting from cessation, as well as from preventing the young from taking up smoking in the first place. One study from China suggests that a 50% tax increase would raise US$ 703 billion over 50 years, with just 14% of this increase being borne by smokers in the lowest income quintile (95). In addition, the tax increase would yield a savings of US$ 24 billion in expenditures on tobacco-related diseases, with about 28% of these savings being enjoyed by smokers in the lowest income quintile. The ECBA framework actually presents a rather conservative profile of the net benefits of raising tobacco taxes, since it does not include other sources of gain, such as reduced harm from exposure to second-hand smoke, increased productivity and the potential for poor households to benefit from social programmes funded through increased tax revenues (96). Assessments of the distributive impact of the 2009 tobacco tax increase in the United States found that the overall progressivity of the increase was enhanced by the tax rev- enue being used to expand health insurance coverage for children of low- and middle- income families (97). Accounting for this expanded coverage added to the progressiv- ity of the overall legislative package, the bottom line being that the impacts are positive for lower-income quintiles and greatest, on average, for low-income households (93). Similarly, a large proportion of the tobacco tax revenues from the Philippines’ so-called Sin Tax Reform was used to subsidize universal health coverage (UHC) for poor and near-poor families. Globally, 37 countries are known to earmark some tobacco tax revenues for health programs, with many of these programs indirectly benefiting the poor and less-advantaged disproportionately more than other groups (27) (for details on earmarking, see section 4.6). 4.3.3 CONCLUSIONS Contrary to the perception of tobacco taxation being regressive, it is a strong pro- poor policy when the broader economic impacts are taken into consideration. The tax burden is not a complete indicator of regressivity, since it does not include the negative health and economic impacts of tobacco-attributable diseases or the positive impacts of behaviour change in response to tax and price increases. The health and economic burdens of tobacco-attributable diseases fall dispropor- tionately on the poor, who tend to have higher tobacco use and are also the least able to afford the necessary medical care. Because the poor tend to be more price-sensitive, they curtail their use and consumption more significantly than wealthier smokers in response to tax increases, which in turn reduces their downstream health and economic costs. Tobacco taxation can be made even more progressive by earmarking or allocating tobacco tax revenues for social goods and services that benefit the poor (see section 4.6). CHAP T ER 4. PO LI T I C AL ECO N OMY 217 4.4 SCARE TACTIC R: REVENUE REDUCTION 4.4.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases result in reduced tax revenues for the government. According to them, the reduction in revenues is caused either by substitution to cheaper, lower-taxed or smuggled tobacco products or by reductions in consumption overall (98–99). The tobacco industry often refers to the Laffer curve to make this argument. According to this curve, revenues increase along with tax rates up to a certain point, after which further increasing tax rates leads to declining revenues. When considering tobacco taxes, the tobacco industry assumes that countries are already approaching or are even beyond the critical tax rate level (98). However, the argument rests on a narrow theoretical and empirically unsubstanti- ated foundation (98–100). The price inelastic demand for tobacco and the relatively low tax share in prices in many countries explain the win-win for public health and finance, i.e. that declines in consumption and increases in revenues can occur simultaneously (98, 101). Furthermore, many country examples (see case studies below) demonstrate that well-designed and well-implemented tobacco tax increases lead to increases in revenue, at least in the short to medium term (98, 100). Although consumption will diminish with a tobacco tax increase, the percentage increase in excise tax per unit is greater than the percentage decrease in tobacco consumption, cancelling out at least some of the effect of reduced consumption on revenue (98–99). A change in the tax rate, with all other factors influencing consumption kept constant, corresponds to a change in the tax revenue and is represented by a move- ment along the Laffer curve. As the tax rate changes, so does the elasticity of the tax base; each point on the Laffer curve corresponds to a different tax base elasticity. When one or more of the other factors changes, this affects the position of the curve, and the tax base elasticity changes at a given tax rate. For example, a successful smoke-free policy or advertising ban that reduces the demand for tobacco shifts the curve down, reducing the tax revenue potential for each tax rate. To demonstrate that few, if any, countries are beyond the revenue-maximizing point on the Laffer curve, Table 4.7 shows the revenue impact of increasing excise taxes under different scenarios, using different price elasticities of demand, different levels of tax increases and different starting tax shares, depending on country income levels. This is the tax base elasticity approach from which the Laffer curve is derived (for more details, see section 2.2.3 and Annex 2.2). The total and excise tax shares shown are weighted averages for each country income group, calculated from the RGTE dataset. The revenue gains were simulated using progressive levels of excise tax increases (25%, 50%, 75% and 100%) and varying price elasticities of demand (-0.4 to -1.2). 218 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.7 Percent increase in excise revenues under different scenarios of tax levels, tax increases and price elasticities20 Income group Total tax as % of retail price Excise tax as % of retail price Increase in excise tax Increase in excise revenue when price elasticity of demand is: -0.4 -0.6 -0.8 -1.0 -1.2 LOW INCOME 38% 22% 25% 22% 20% 19% 17% 16% 50% 43% 39% 36% 33% 29% 75% 63% 57% 52% 46% 41% 100% 82% 74% 66% 59% 51% MIDDLE INCOME 58% 41% 25% 19% 17% 14% 11% 9% 50% 37% 31% 26% 20% 15% 75% 54% 45% 36% 27% 19% 100% 71% 57% 45% 34% 23% HIGH INCOME 68% 55% 25% 18% 15% 11% 8% 5% 50% 35% 27% 21% 14% 8% 75% 50% 39% 29% 19% 10% 100% 65% 50% 36% 23% 11% Source: Authors’ calculations using data from the RGTE (27).21 Substantial revenue increases occurred in all the scenarios that were considered in the simulation. These results reaffirm much of what is already known, i.e. that higher tax increases generate higher revenue gains, and that these gains increase with the increasing inelasticity of demand. Even when demand is relatively price elastic (-1.2), the simulation predicts a gain in revenue. The tax share in price also affects revenue potential. The lower the tax share in price, the larger the revenue potential. This suggests that revenue reductions as a result of an excise tax increase will occur only if the scenario is extreme (i.e. a very elastic demand coupled with a very high current tax share). It is important to note that the vast empirical literature 20 These projections use 2018 data from 185 countries. The countries were classified according to World Bank income group, with the average total tax share, excise tax share and VAT/sales tax share for each country weighted according to the number of current adult cigarette smokers. To calculate the projected revenue for each stated elasticity, it was assumed that there would be full pass-through of the excise tax increase, along with constant percentages of non-excise taxes (VAT/sales tax) as a share of the retail price. The consequent changes in price were multiplied against the respective elasticities to derive the expected change in consumption. The projected revenues could be easily computed by multiplying the new consumption figures against the increased excise tax rates. 21 These calculations do not take into account brand substitution (cross-price elasticities), income ef- fects or illicit trade. The excise tax was assumed to be a specific tax, while the non-excise taxes (VAT and others) were bundled and treated as an ad valorem tax with retail price as the tax base. The difference between retail price minus all taxes was also assumed to be constant, with full pass-through of the tax increase to consumers. CHAP T ER 4. PO LI T I C AL ECO N OMY 219 shows tobacco to be universally inelastic; thus the extreme scenario should not be given credence by policy-makers. Furthermore, as demonstrated by the data in Table 4.7, tax shares in most countries are relatively low and reinforce the revenue potential of tobacco tax increases. The revenue potential of tobacco taxes is indeed quite significant. It is estimated that in 2018, excise taxes on cigarettes generated a total of US$ 361 billion in revenues worldwide, including US$ 162 billion in LMICs. If all countries were to raise excise rates by the equivalent of US$ 1 per pack of cigarettes, the amount of excise revenue would increase by between US$ 178 billion and US$ 219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from these tax increases, with excise revenues in these countries increasing by US$ 133 billion to US$ 167 billion, or by 82–103%.22 Revenue reduction in the countries examined was due to other causes, not the tax increase per se. For example, Tonga significantly increased its excise tax on cigarettes in 2016 and saw a very sharp decrease in its consumption (40% decrease), followed by a revenue decrease. This occurred because 20% of smokers switched to an untaxed, cheap local loose tobacco product called Tapaka Tonga (102). The lesson learned was that Tonga needed to tax all its tobacco products at the same level to avoid substitution to lower-price/untaxed tobacco products. Another example of revenue decrease that was not related to tax increases but rather to tax administration mismanagement is the case of South Africa (see explanation in the case study later in this section). Finally, declines in revenue due to long-term declining trends in tobacco use should not be confused with being beyond the revenue-maximizing point of the Laffer curve. For example, in the United Kingdom, where long-term declines in tobacco use are being experienced, a nominal decline in revenues occurred between 2017 and 2018 even though excise taxes remained unchanged. Conversely, even countries with very high tobacco excise rates experience increases in revenues as a result of tobacco tax increases (see the case study of Australia below) (98, 100). This suggests that few countries, if any, are beyond the revenue-maximizing point on the Laffer curve. Tobacco consumption is expected to be tax inelastic, even if demand becomes effectively price elastic as a result of successful tobacco control interventions. Taxation serves as an instrument for both fiscal and public health objectives. If after successful tobacco control interventions, prices reach levels where demand is elastic, the tax base is still most likely to be inelastic due to tax undershifting, since overshifting is not a good pricing policy when demand is elastic (for a more detailed discussion on the shifting of tax, see section 2.2.2). In other words, a tax rate increase in combination 22 Goodchild M, Perucic AM, Paul J. Tobacco taxation as a strategy to achieve global targets for smoking prevalence. Unpublished manuscript. October 2020. 220 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N with non-price tobacco control measures, which make consumers more sensitive to price (tax) increases, leads to declining but still positive marginal revenues. In the long run, tobacco control policies, including price and tax measures, may be so successful in reducing consumption that revenues will plateau or fall. This is ultimately the long-term policy goal. Reducing the impact of the policies and ending the global tobacco epidemic is the aim of tobacco control and not something to be avoided. However, ending the global tobacco epidemic is unfortunately not foreseen in the short to medium term; therefore, governments can currently rely upon tobacco taxes as a reliable source of revenue (103). 4.4.2 THE REVENUE IMPACT OF EXCISE TAX INCREASES: CASE STUDIES The following case studies illustrate four key points: (1) large and regular tax increases result in large and consistent revenue increases; (2) countries with high taxes and falling prevalence of tobacco use can still increase revenue with tax increases; (3) countries that reduce taxes experience revenue declines; and (4) countries that increase taxes in the face of illicit trade still increase revenue. Large and regular tax increases usually mean large and consistent revenue increases South Africa’s experience shows how successive tax increases, well above inflation and year after year, generate additional revenues even after taxes have been increased substantially. After two decades of declining real revenue in the 1970s and 1980s as real excise per pack declined, South Africa implemented successive excise tax increases from 1994 until 2011 (Fig. 4.4.1) (98, 104). After adjusting for inflation, this resulted in a real excise tax revenue increase of 245% (98). Revenues began to plateau from 2012 as tax increases stalled. They began to decline after 2015 – not due to tax increases, however, but due to a dramatic decline in administrative capacity and enforcement measures exacerbated by large-scale corruption in the government, including the tax administration authority (105). The rapid and catastrophic decline in tax administration and enforcement has been the subject of much attention (106). CHAP T ER 4. PO LI T I C AL ECO N OMY 221 Fig. 4.4.1 Real excise tax per pack of cigarettes and real excise tax revenue in South Africa, 1961–2020 Source: Data shared by University of Cape Town, 2020. Similarly, the Philippines provides a compelling example of how large and regular tax increases alongside reforms to tax structure can lead to large and consistent revenue increases – in this case, also through an accompanying reform to the tax structure (Fig. 4.4.2) (98). The 2012 Sin Tax Law consolidated the country’s four tax tiers into two by 2013 and established a uniform structure by 2017. The same law provided for large, progressive increases across the board, but in particular for the lowest tax categories (98). Not only were the revenue gains substantial, they exceeded all the projections for 2013–2017 made prior to the law’s passage (98). Excise tax per pack Excise tax revenue Ra nd s pe r p ac k (c on st an t 2 02 0 ra nd s) Excise revenue (constant 2020 rands) 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11 20 13 20 15 20 17 20 19 0 0 5 6 10 10 15 14 20 18 222 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.2 Real excise tax per pack of cigarettes (multitier, two-tier, unitary) and total tobacco real excise tax revenue in the Philippines, 2012–2018 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2012 as the base year. Sources: (107, 108 and data shared by the Philippines Department of Finance, September 2020). Ukraine is another example of a country that has regularly increased taxes over the past 10 years and has experienced increased revenues along with decreases in consumption and the number of smokers. Figure 4.4.3 the shows the trends in excise tax, revenues, cigarette sales and number of smokers in 2008–2017. Increases in excise rates were consistently accompanied by increases in revenues. In 2014–2015, excise tax was not increased above inflation (and inflation, especially in 2015, was very high, at 48.7%), so real values of excise and revenues went down. But it is evident from the data that revenues closely follow the path of excise levels even when sales go down. 12 Re al e xc is e ta x pe r p ac k, P hi lip pi ne p es os (2 01 2 ba se ) Billions (Philippine pesos) in tobacco real excise tax revenue (2012 base) 00 20 60 100 140 2012 HIGH PREMIUM SIN TAX LAW RA 10354 TRAIN LAW RA 10963 UNITARY RATE 2013 2014 2015 2016 2017 2018 5 10 15 20 25 30 32 .1 9 68 .6 6 71 .0 9 93 .5 2 87 .2 0 96 .0 1 11 4. 39 24.4 25.4 26.2 26.8 26.9 27.5 11.7 16 23.1 19.6 LOW Real tobacco excise revenues Dierent levels of real tobacco excise tax per pack MEDIUM 7.6 2.7 28.3 CHAP T ER 4. PO LI T I C AL ECO N OMY 223 Fig. 4.4.3 Average real cigarette excise tax rates, real cigarettes excise tax revenues (base year 2008) and cigarette sales and number of cigarette smokers in Ukraine, 2008–2017 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2008 as the base year. Source: Data provided by Konstantin Krasovsky, July 2020. Countries with high tax and falling prevalence of tobacco use can still increase revenue with tax increases Countries with already high tobacco taxes and rapidly diminishing tobacco use can still increase revenue by increasing taxes (98, 109). Australia has implemented comprehensive tobacco control policies and enacted consistent tobacco tax increases on top of what were already some of the highest tax rates in the world (see Fig. 4.4.4). Between 2001 and 2010, revenue increased with increasing tax rates, but in real terms (inflation-adjusted) it remained static (109). Then, in 2010, a 25% excise tax increase was introduced, with large annual increases scheduled from 2013 onward (98, 109). The result of this tax policy has been consistent and large increases in revenue year after year for nearly a decade, even when the increases were being made on already high tax rates.23 23 The apparent reduction in revenues in 2012 and 2013 was due to a change in the source of the data for 2001–2011 and 2012–2016. Data for 2012 and 2013 do not include customs duty, while all other years do. 73 95 125 112 88 82 75 76 67 11.8 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Number of daily smokers, in millions Real tobacco excise revenue, in billions of Ukrainian hryvnia Real average excise per cigarette packs of 20, Ukrainian hryvnia Number of taxed cigarettes (sales), in billions of sticks 0.6 3.6 10.1 9.2 8.7 8.6 8.4 8.1 7.3 6.2 6.5 6.3 7.8 10.3 11.2 12.1 13 1.4 2.2 2.5 2.9 3.5 3.2 2.7 3.3 4 12.7 13.4 74 9.7 224 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.4 Real excise tax and customs duty per stick and real total revenue (all tobacco products) in Australia, 2001–2019 Notes: Rates published by Australian Taxation Office and Australia Department of Immigration and Border Protection, adjusted using Australian Bureau of Statistics Consumer Price Index rates. The 2011–2012 and 2012–2013 figures do not include customs duty, which explains the apparent decline in revenue. Using official disclosures, Scollo and Bayly estimate that duties in these years were $7397.2 and $7687.2 respectively (110). Sources: (109, 110). Countries that reduced taxes and saw revenues decline Prior to 1982, Canada lowered taxes on cigarettes and experienced declining revenues as well as increased smoking – particularly among youth. Subsequent fivefold in- creases in cigarette taxes between 1982 and 1992 resulted in more revenue, increases in retail price and substantial reductions in consumption, with teenage smoking declining by nearly two thirds (5). In the early 1990s, a growing illicit trade in ciga- rettes emerged in which Canadian cigarettes exported to the United States were then smuggled back into Canada (5). The tobacco industry – which was later found to be complicit in and profiting from this illicit trade – sought to frame Canada’s high tax rates as the cause of smuggling (111–112) and succeeded in convincing the federal government, as well as six provincial governments, to make massive reductions in the tobacco tax (111–112). As a result, federal tax revenues fell significantly – more than twice as much as the government had predicted – and smoking rates among both adults and youth began to increase (5, 112). The Canadian government later changed its strategy, and the federal excise tax was restored, resulting in increased Specic excise/ duty per stick (cigarettes/cigars less than 0.8 g) Total revenue (all tobacco products) To ta l c us to m s/ du ty re ve nu e (in a tio n, a dj us te d, 20 19 A us tr al ia n do lla rs , m ill io ns ) Total excise/duty per stick (ination, adjusted, 2019 A ustralian dollars) 0 3 000 6 000 9 000 12 000 $0.8 $0.6 $0.4 $0.2 $0 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 CHAP T ER 4. PO LI T I C AL ECO N OMY 225 revenues and decreased smoking (5). Canada’s focus then shifted to using customs enforcement, rather than tax rates, as the best means of countering illicit trade (5). Countries that increased taxes in the face of illicit trade and still increased revenue As discussed in section 4.1, the tobacco industry exploits illicit trade as a strategy to undermine tobacco tax policy, with the goal of deterring governments from increasing tobacco taxes. The narrative that has been created is that higher tobacco tax rates result in increased illicit trade and undermine the policy goals by resulting in lower (or no) declines in tobacco use or lower (or no) increases or even decreases in revenue. However, as shown in section 4.1, the empirical evidence does not sup- port the industry arguments. Furthermore, the evidence shows that the industry and its allies have consistently overstated and exaggerated the scale and extent of illicit trade (see section 4.1). As was the case in Canada, Brazil’s tobacco tax policy suffered from a fear that the illicit market would expand unless it was undercut by price competition in the legal market, which it was thought could be best encouraged through tax cuts (20). Real excise tax rates declined from 1999 until the mid 2000s, as nominal increases were below the rate of inflation. This resulted in declines in real tax revenues (20). In these years, the tobacco industry used the tax cuts to increase profit margins rather than decrease prices and outcompete the illicit market, while also exaggerating the size and scope of the illicit trade problem (20). This caused the industry’s argument on illicit trade and revenue to lose credibility and resulted in increases in tax rates from 2007 onwards, with a major reform passed in 2011 (20). Tobacco excise rates and minimum prices were scheduled by the law to increase at levels above expected inflation from 2011 until 2015 (20). This resulted in substantial increases in the tobacco excise per pack, as well as overall revenue, which by 2015 had more than doubled from its low point in 2013 – equating to more than 50% in real terms (see Fig. 4.4.5). The success of this reform shows that revenues can be increased by higher rates despite the presence of a sizeable illicit market (113). More recent data show that revenues in Brazil declined in 2015 and 2016, coinciding with an increase in illicit trade, but also with an exceptionally bad economic recession that saw GDP decline by more than 3% in those years. 226 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.5 Average real excise tax per pack, real excise tax revenues and domestic cigarette sales in Brazil, 1999–2014 Notes: Data based on domestic sales and tobacco excise revenues, Federal Revenue Secretariat, indexed to 2013 Brazilian reals, using Consumer Price Index. Revenue collection indexed to 2013 reals, using Consumer Price Index. Source: (20). Improvements in tax administration and enforcement can also generate increases in revenues. In Kenya, several measures, including fiscal markings and, later, an advanced tracking and tracing system, improved collection, resulting in increases in both legal sales and tax revenues and a reduction in illicit sales (114). Moreover, these examples of poor governance indicate that attention should be focused on countries where a significant loss in administrative and enforcement capacity un- dermined revenue collection. 4.4.3 CONCLUSIONS The tobacco industry uses revenue concerns as a SCARE tactic to avoid, dilute and/or delay tobacco tax increases. The argument that higher taxes will decrease revenue is theoretically plausible, but real-world examples have demonstrated that this has not occurred. Furthermore, simulations show that even large tax increases in current average tax shares yield substantial revenue gains. The use of the Laffer curve by the tobacco industry should be challenged and refuted. The relatively price inelastic nature of cigarette demand combined with the Real excise tax amount per pack Domestic sales Real excise tax revenue D om es tic s al es (b ill io n pa ck s) Re al e xc is e ta x re ve nu es (b ill io n 20 13 re ai s) Real excise tax am ount per pack (2013 reais) 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 0 1 2 3 4 5 6 0 1 2 3 CHAP T ER 4. PO LI T I C AL ECO N OMY 227 low tax share and no overshifting of the tax means that most – if not all – countries are still far from the revenue-maximizing point, indicating that increases in taxes will lead to increases in revenues. The case studies in this section refute each of the tobacco industry’s arguments regarding alleged potential revenue loss due to tax increases. The experiences of South Africa, the Philippines and Ukraine demonstrate that large and regular tax increases result in large and consistent revenue increases. Well-designed tax structures have also proven to play an important role in generating revenues. The experience of Australia shows that even countries with already high tax rates and declining prevalence of tobacco use can increase revenues with regular, large tax increases. The experience of Canada warns against following the advice of the tobacco industry to decrease taxes as a way to fight illicit trade. It demonstrates clearly that decreasing tobacco taxes will decrease revenue and encourage consumption, rather than counteract illicit trade. The experience of Brazil shows that countries with substantial illicit trade issues can still increase revenue by increasing taxes. Finally, in the few cases where revenue decreases were seen, the reasons for the decreases were not strictly linked to tax increases. This was the case in Tonga, where the increase in tax was applied only to cigarettes and not to their close substitute, loose tobacco – leading smokers to switch products. In South Africa, a decrease in revenue was the result of the weakening of government institutions. And in Ukraine, real revenues decreased only during the two years when taxes were not increased above inflation. 228 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.5 SCARE TACTIC E: EMPLOYMENT 4.5.1 INTRODUCTION In opposing tax increases, the tobacco industry often seeks to frame tobacco taxes as an economic rather than a public health issue (5, 48, 53, 115). Particular emphasis is placed on the alleged threat that tax increases pose to employment in tobacco farm- ing and manufacturing, as well as other related industries (5). This so-called choice between health and jobs is, however, based on several false premises, including (5): 1. tobacco is a significant source of jobs within the context of broader labour markets, and domestic tobacco tax increases will have a drastic effect on domestic employment (48, 53, 116–117); 2. tobacco consumption is an indispensable engine for job creation (5, 48, 54); and 3. tobacco provides highly prosperous, sustainable and irreplaceable livelihoods (5, 53, 118). In reality, the relationship between tobacco taxation and employment is consider- ably more complex than the industry makes it out to be. In fact, there is ample evidence to show that tobacco taxes are a win-win for public health and the fiscal space, without measurable risks to employment. 4.5.2 THE LINK BETWEEN TOBACCO EMPLOYMENT AND TOBACCO TAX RATES Tobacco farming, production and manufacturing (including hand-rolling in some countries, most of them in South-East Asia) constitute a small proportion of the labour force, even in countries where the industry is most heavily concentrated (5, 48, 103, 116). Employment in tobacco farming and manufacturing has been declining globally due to advances in technology, trade liberalization, market consolidation and the privatization of formerly state-owned tobacco companies (5, 103, 119). These same trends have led to the heavy concentration of tobacco growing and manufacturing in only a handful of countries – and within these countries, often in only a small number of regions (103, 119–121). Even in those countries that lead in tobacco growing and manufacturing, tobacco’s overall share of total agricultural and manufacturing employment is relatively small and is often decreasing as efficiencies in production reduce labour intensity (5, 117–118, 122). Similarly, the industry’s claim that tobacco taxes reduce employment is exag- gerated and typically overlooks wider trends driving tobacco industry employment. Indeed, tobacco industry developments and innovations have played a greater role in the reduction of employment in the tobacco industry than have tobacco control policies (103). Despite industry claims that tobacco taxes can affect employment, CHAP T ER 4. PO LI T I C AL ECO N OMY 229 characteristics of the location of production – such as market size, labour costs, growing conditions and leaf preferences – have much more to do with tobacco industry interests than with the tobacco tax rate (103, 123). Moreover, jobs in countries that produce tobacco primarily for export are not greatly affected by reductions in local consumption resulting from tax increases (5, 103, 116, 123). Finally, it has been demonstrated that tobacco tax increases do not have a significant effect on employment in the retail sector, as most retail businesses sell other goods (103). Estimates of the gross employment impact of tobacco tax hikes demonstrate that job losses that do occur can be more than compensated for by increases in revenue. A 2018 World Bank study estimated that in Indonesia, for example, an ambitious tax reform that would simplify tiers and increase prices by close to 50% would reduce gross employment in the tobacco manufacturing sector by less than 0.5% (a loss of 2 914 jobs). The government could provide income support to the displaced workers (for example, through training, temporary transport/mobility or income support) with less than 2% of the revenue gained from the tax increase (117). Similarly, a 2019 study by Bangladesh’s National Board of Revenue estimated that a substantial increase in tobacco taxation would cause 7 012 lost jobs, but that the total income associated with these job losses in the bidi industry would amount to only 3.5% of the revenue gained (120). Accordingly, increased revenue can more than compensate for the expenditure of supporting those who lose jobs and need to acquire new skills before transitioning to new employment (120). Box 4.5.1 Employment fears deployed to frustrate tobacco tax reform in Indonesia In 2017, the Indonesian Ministry of Finance decided to implement tobacco tax in- creases by 2019 and tier simplification by 2021 (124). This resolution was, however, abandoned within a year, after a concerted campaign by tobacco industry actors and their allies to reframe the increase as an economic issue with a focus on, among other things, the effect the tobacco tax increase would have on employment (124). This defeat for the tobacco tax initiative came despite estimations of how the loss of income associated with lost jobs would be dwarfed by the additional revenue gained by the tax (117). Earlier analysis had estimated an overall large net positive impact on employment from tobacco tax increases (125), which illustrates how evidence that challenges assumptions around the negative socioeconomic impacts of tobacco control tends to be discounted (126). In this case, tobacco industry arguments seem to have resonated strongly with politicians from the electoral districts of West Java, East Java, Central Java and West Nusa Tenggara, where employment in tobacco farming and manufacturing is concentrated (124). Although tobacco manufacturing 230 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N represented only 5.13% of total manufacturing employment, the concentration of the job and economic activity meant that arguments regarding employment were particularly salient (117,119). Accordingly, tobacco industry arguments that tobacco taxation would negatively impact employment and farmer livelihoods prevailed, despite strong opposing evidence (119). Concentration of tobacco industry activity within countries poses particular obstacles to overcoming industry arguments on employment and needs to be given careful attention. 4.5.3 THE EVIDENCE ON THE NET EFFECT OF TOBACCO TAX INCREASES ON EMPLOYMENT A proper analysis of the effect of tobacco tax increases on employment must examine their impact on net or economywide employment. Decreases in expenditures on tobacco associated with tobacco control do not mean that expenditures simply disappear; rather, they are redistributed towards consumption of other goods and services, thereby generating employment elsewhere in the economy (5, 103, 123). Similarly, though the effect of higher tobacco taxes on net consumption is arguably more ambiguous, revenues from this intervention do generate spending, invest- ment and employment in public services such as health and education (5). Tobacco control polices usually have a marginal neutral or positive effect on net employment, particularly in countries that are net importers of raw or manufactured tobacco products, as expenditures on these imported items tend to flow out of the country (5). Export-oriented tobacco producers are less sensitive to local demand and are not significantly affected by domestic tobacco tax measures, which likely have a near-neutral net impact (5). In some cases, the net employment impact is a very small negative number, typically less than 1% (127–128). A recent study estimated that in the United Republic of Tanzania – a large tobacco-producing and exporting country – a 30% reduction in smoking prevalence would result in a net employment decline of just 0.5% across the economy as a whole (129). A similar study of Pakistan found that, with some variance depending on where spending was redistributed from tobacco consumption, the overall net effect on employment from a significant reduction in expenditure on cigarette employment – 1 billion rupees – would be a gain of between 6 651 and 5 803 jobs (122). This increase would occur because expenditure on cigarettes produces much less employment in the broader economy than expenditure on food and education (122). In the United Republic of Tanzania, as elsewhere, increased revenue could be used to assist those who lose employment with transitioning to new livelihoods. CHAP T ER 4. PO LI T I C AL ECO N OMY 231 4.5.4 THE VIABILITY OF BETTER LIVELIHOODS In arguing against tobacco tax increases, the tobacco industry advances the myth that people employed in tobacco production – particularly tobacco farming, but also manufacturing – lack any other prospect for a comparably attractive livelihood. However, studies based on extensive survey data in Indonesia, Kenya, Malawi, the Philippines and Zambia have shown that despite needing to commit significant amounts of labour to their crop, tobacco farmers often suffer losses rather than gain profits (119, 121, 123, 130–132). Furthermore, the Indonesian studies demonstrate that tobacco farming has a negative impact on household income and opportunity compared with the experience of other farming households that have given it up (119). Declines in consumption as a result of tobacco tax increases are gradual and susceptible to the same progressive adaptation that has occurred for decades (5, 103). While there will be a need in some countries for the government to help farmers transition to other crops or industries in the longer term, this process will not be a major short-term shock to employment or the wider economy (123). Because tobacco growing and manufacturing can be concentrated in just a few locations within a country, job losses within the tobacco industry might have a disproportionate effect in one location, while employment gains from reduced consumption may be spread across the whole country (120, 123). A study of the employment effects of tobacco tax increases in Bangladesh estimated that up to 60% of all job losses would occur in only two districts – among the poorest in the country – due to the high level of industry concentration (120). Studying the need for support, as well as the means of delivery and funding of support, is particularly necessary in these circumstances. Beyond the need to ensure equity and support employment, a failure to provide for targeted relief can exacerbate fear of job losses and may prove fatal to a tobacco tax proposal (120). Box 4.5.2: Supporting alternative livelihoods in the Philippines The Philippines earmarked 15% of the revenue from a 2012 increase in tobacco taxes to supporting economically viable alternative livelihoods for tobacco farmers and workers (5). Tobacco farming in the Philippines is regionally concentrated, and the tobacco industry had previously been successful in deploying concern for smallholder tobacco farmers to undermine tobacco control measures (136). The provision of economic support was a politically effective countermeasure to tobacco industry SCARE tactics and eased the tax increase’s passage. Given the Philippines’ integration with global tobacco markets and demand, tobacco farmers have not been seriously affected by the tobacco tax increase and 232 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N reduced domestic demand (108, 132). Nevertheless, transfers to tobacco-growing regions from the earmarked tax have been substantial (108). These funds are support- ing gradual transitions to alternative livelihoods, with farmers being encouraged to take up alternative crops, as well as establishing infrastructure, such as market-to-farm roads, that will make these alternative crops more economically viable (108, 133–134). Supporting alternative livelihoods for farmers and other tobacco workers is important because it can offset the political effect of industry arguments, even though domestic tax increases usually have only a modest and gradual effect on employment. There are various models for supporting alternative livelihoods when employment in the tobacco sector gradually diminishes due to decreases in either global or national demand. The Philippines is exemplary, but many other countries have either implemented or experimented with supporting crop transitions. Turkey’s alternative crop pro- gramme, implemented in anticipation of the privatization of the country’s cigarette monopoly, has proven effective in supporting many tobacco farmers’ move to other crops (135). Smaller-scale crop substitution projects in Kenya and Yunnan Province in China have shown how financial, regulatory and infrastructure support from government can contribute to crop transitions (5, 53). Argentina, Bangladesh, Mexico and the state of Maryland in the United States provide additional case studies of how governments can support these transitions (5, 136). 4.5.5 CONCLUSIONS The tobacco industry exaggerates the importance of tobacco employment and over- states the impact that domestic demand reduction due to local taxes will have on tobacco farmers serving a global market. The industry also simplifies employment’s relationship with taxation by focusing only on gross employment in tobacco, which ignores the reality that expenditures on tobacco do not disappear but rather are redistributed for other consumption that can produce a similar or higher number of jobs. Many detailed studies have found that tobacco growing is much less profitable and sustainable than the tobacco industry claims. Tobacco farmers throughout the world have successfully transitioned to other crops, although the transition often requires temporary or additional support from the government or other stakeholders. The extent of such support is moderated by the reality that transition from tobacco to other crops is a long-term consideration. CHAP T ER 4. PO LI T I C AL ECO N OMY 233 4.6 EARMARKING TOBACCO TAX REVENUES TO FUND HEALTH 4.6.1 INTRODUCTION Earmarking tax revenues involves the separation of all or a portion of revenue from a tax or group of taxes to be put aside for a specific purpose (137). Globally, more than 80 countries earmark for health (138), and 37 earmark tobacco tax revenues for health (27). There are two main types of earmarks: hard – also called substantive – and soft, or symbolic (139). Hard earmarks link the expenditure with a revenue source in legislation. This can limit funding if the earmarked revenues are the main source of funding, or it can cause surpluses to accrue wastefully when more revenues are raised than may be expended for the earmarked purpose. Soft earmarks include dedicated funds or commitments to use funds for a particular purpose. They are not necessarily legally binding. For example, in France, the ma- jority of tobacco tax revenue is used to fund social security (which includes health insurance and health care), but there is no hard, formal earmark (140). Earmarks can also be some combination of hard and soft. In the Philippines, tobacco tax earmarks are legally binding, but earmarked revenues go to the general fund, and the Department of Health must submit an annual budget for covered programs as part of its budget request. Earmarking is a broad and contentious topic that goes beyond the specifics of tobacco tax earmarking. Discussions on the topic fall within the ambit of public financial management, and earmarking generally is not encouraged. From a tobacco control perspective, however, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. It is a tool to improve the political economy of tobacco taxation; it is a secondary issue only, after the primary goal of reducing demand for tobacco. One way to use earmarking to improve the political economy of tobacco taxation is to link the payment of tax by tobacco users to benefits they will receive through the funding of complementary tobacco control programmes, such as cessation support, or through increased funding for health programmes on which they will rely disproportionately. This is known as the benefit principle. Earmarking for tobacco control makes sense, as its financial cost is relatively small and tobacco tax reduces demand more effectively when implemented within a package of complementary tobacco control measures. Another way earmarking improves the political economy of tobacco taxation is by safeguarding against any perceived or potential negative ramifications of the tax itself. This is important for neutralizing erroneous but often convincing tobacco industry arguments against effective tobacco tax policies. For example, the Philippines earmarks the bulk of the additional revenues from sin taxes for the health insurance premiums of the poor. In addition, a portion 234 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the country’s tobacco tax revenues is earmarked to provide for the economic well-being of tobacco growers and tobacco growing regions, with the general aim of promoting economically viable alternatives to tobacco farming and manufacturing as a safeguard against the potential for reduced domestic tobacco demand (141). Tobacco tax earmarks are complex, however, and care is required when determin- ing whether a particular earmark is needed on the grounds of political economy and justified on the grounds of equity and economic efficiency. In assessing tobacco tax earmarks, many of the same criteria that have been used to assess the appropriateness of generic earmarks also apply. In the rest of this section, these criteria are set out and matched with reasons for the ability of well-designed tobacco tax earmarks to fulfil them. The types and structures of tobacco tax earmarking are explored alongside descriptions of country experiences to provide guidance on how tobacco tax earmarks are used, when they are justified and the best ways to design them. 4.6.2 CIRCUMSTANCES IN WHICH EARMARKS MAY BE SUITABLE Scepticism about earmarking is both long-standing and justified, but much of the debate concerns earmarking generally and is not specifically concerned with the merits of tobacco tax earmarking (138). The main concerns raised about earmark- ing are listed in Table 4.8, accompanied by suggestions for how earmarks may be structured to address these concerns. Table 4.8 Concerns about earmarking and suggested safeguards to avoid the concerns MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Democratic accountability and oversight: earmarks undermine democratic processes by impeding legislative and executive oversight over expenditure. Establishing proper oversight and accountability procedures is important to ensure funds are not mismanaged (138). Additionally, if a soft earmark structure, which transfers revenue to the general fund from which it is then allocated, is adopted, this will not be a concern. Budget rigidity: earmarking may create budget rigidity that can lead to inefficient allocation of resources (138). An earmark’s particular design determines how much rigidity is introduced (138). Flexible soft earmarks are less prone to introducing rigidity than hard earmarks. Concerns about rigidity can be reduced by the inclusion of a sunset clause that ensures that the earmark is automatically discontinued or reviewed after a set period of time has elapsed (138). A further safeguard is to establish the earmark as a waterfall account, with any excess revenue over a set amount being allocated to the general fund. CHAP T ER 4. PO LI T I C AL ECO N OMY 235 MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Fragmentation: earmarking can result in fragmented and uncoordinated expenditures. This means policies complementary to the earmarked purpose but outside of its purview may be unfunded (138, 142). This is a legitimate concern. The negatives of fragmentation cannot be entirely eliminated, but they may be outweighed by the other merits of tobacco tax earmarking. That said, proposals for tobacco tax earmarks should be scrutinized to ensure that the funded purpose is at least cost-effective. Decreased equity: equity will decrease if individual access to benefits is narrowly defined according to payments made. This issue is not likely to arise with tobacco tax earmarks but is conceivable and something that should be guarded against in an earmark’s design. Capture by special interests: because earmarks are often the result of political expediency, an earmarked purpose may be determined by powerful special interests promoting a tax’s passage rather than careful prioritization of resources (138). Well-designed earmarks will guarantee funding for underresourced programmes and high- priority programmes. While the above concerns may be valid and design does matter, tobacco tax and other health-promoting taxes are not subject to the same concerns when it comes to the justifiability of earmarking their revenue (138, 143). Some of the factors that distinguish tobacco tax earmarks from more general critiques of earmarking are listed in Table 4.9 (138). Table 4.9 Concerns about earmarking and distinguishing factor for tobacco tax earmarks GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Procyclicality: earmarked revenues are often procyclical and susceptible to booms and busts (138–139, 142). Tobacco tax revenues are generally not cyclical (they are recession-proof ), and revenue is predictable relative to most other indirect and direct taxes (103). Budget rigidity Tobacco tax earmarks necessarily involve only a relatively small proportion of the budget; therefore, the effect of any rigidity will be relatively insignificant. Partly because of the relatively small amounts involved, there is only limited real-world evidence of tobacco tax earmarks having introduced harmful rigidity (143).24 24 See also the subsection on the amount of money associated with tobacco tax earmarks in section 4.6.3. 236 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Capture by special interests All earmarks should be scrutinized to ensure that their funded purpose is cost-effective. In the case of tobacco tax earmarks, however, political economy considerations may mean that it is sufficient for a low-priority purpose to be funded if the funding will unlock the political will needed for effective tobacco tax increases. In these cases, special interests are being purposefully catered to in order to ensure that tobacco tax increases occur. Of course, arguments against tobacco taxes and tobacco tax earmarking are led by special interests such as the tobacco industry (144–147). Insufficient revenue: the earmarked revenue source may become insufficient for funding its purpose (139, 142). Even though revenue may decrease in the long term when more tobacco users quit, such decrease is expected to be gradual. See Section 4.4 for details on how, with effective design, revenue will generally increase even with declining consumption. In addition to these reasons why general concerns about earmarking do not fully apply to well-designed tobacco tax earmarks, there are a number of compelling reasons for tobacco tax earmarking to finance tobacco control or public health that argue in favour of its implementation: • Significant increases in excise taxes are the most effective, as well as the most cost-effective mechanism for reducing consumption, but they are best implemented as a part of a package of complementary tobacco control measures, such as the WHO MPOWER package. Earmarking tobacco tax revenue for interventions that may not be funded otherwise can strengthen overall tobacco demand reduction (148). • The political economy of tobacco tax increases also makes earmarks attractive: – People have been shown to be more supportive of tobacco tax increases when they know the revenues will be used for targeted social programmes (143, 149–150). Earmarking tax revenue for health or tobacco control frames tobacco tax as a public health intervention in the minds of the public, which may otherwise view it as merely a revenue source (138). Research has shown that using earmarking to link a tobacco tax to health can also help raise awareness about the dangers of tobacco use (143). – When tobacco tax revenue is earmarked for programmes that benefit vulnerable groups, the tax becomes more equity-enhancing. Although lower socioeconomic groups and young adults receive disproportionate health and economic benefits from tobacco tax increases over the medium term, these groups will expend a greater share of their income in the CHAP T ER 4. PO LI T I C AL ECO N OMY 237 short term because of tobacco taxes. Earmarking tobacco tax revenue for programmes such as UHC or cessation services that provide immediate benefits to these groups neutralizes some critiques of tobacco taxation (e.g. the 2009 United States federal excise tax increase and the 2012 Philippines Sin Tax Reform illustrate how equity-enhancing earmarking facilitated passage of substantial tax rises) (93, 138, 143, 150). 4.6.3 EARMARKING PRACTICES AND COUNTRY EXAMPLES Earmarking tax revenues for health is a common practice in 80 countries. In 2018, 37 countries from all regions of the world earmarked tobacco tax revenues for health purposes.25 Case studies in the political economy of tobacco tax earmarking In 2012, the Philippines comprehensively reformed tobacco and alcohol excise taxes. Tobacco taxes were increased significantly, and numerous tax tiers were reduced to only one tier by 2017. Although increasing revenue was a foremost motive for some officials, the reform was explicitly framed around boosting UHC funding and advancing public health by reducing alcohol and tobacco consumption. Earmark- ing of tax revenue for UHC was essential to the political compromise that made this trailblazing tax increase a reality. It ensured that the increase, which may have otherwise been perceived as regressive, was framed as a progressive public health measure in the public imagination, while also appeasing tobacco growers and their political representatives. Earmarking was also important because the earmark en- sured high-level support for the tax by achieving a key political priority (151). Its soft-earmark structure meant it was not a blank cheque to the Ministry of Health, and this addressed concerns within the Ministry of Finance. Similarly, in Australia, earmarking of revenue helped overcome community objections to tobacco taxes and tobacco control more generally that resulted from the tobacco industry’s sponsorship of sports and the arts in the 1980s. Attempts to completely ban tobacco advertising and sponsorship had been unsuccessful due to strong pressure from sports, arts and racing lobbies that claimed that a ban would harm these activities. States, starting with Victoria, responded by earmarking funding for Health Promotion Foundations that took over the tobacco industry’s sponsorship activities and also paid for antismoking campaigns. In 1997, these earmarks ended after a High Court ruling that the Constitution did not allow states to collect excise taxes. However, in recognition of the successful work of the Foundations, the federal 25 Details about earmarked taxes by country are provided at https://www.who.int/tobacco/global_report/ Table-9-4-Use-of-earmarked-tobacco-taxes.xls?ua=1. 238 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government began funding them directly from the federal budget (152). Although this example may be difficult to replicate precisely, it shows how earmarks with built-in sunset clauses for piloting cost-effective interventions can potentially graduate to funding from the general budget once they have proven their effectiveness. It also more generally shows how earmarks can disarm community objections, reframe tobacco tax increases and unlock the political will needed to advance effective tobacco control measures. Structures for managing earmarked tobacco taxes A 2016 review of nine countries’ tobacco tax revenue earmarking experiences identified three arrangements for governance and allocating revenue (151). Table 4.10 presents some examples of these allocation arrangements. In some countries, earmarked tobacco tax revenues are combined with alcohol tax revenues. Table 4.10 Illustrative arrangements for allocating earmarked tobacco tax revenues POSSIBILITIES FOR ALLOCATING TOBACCO TAX EARMARK REVENUE Forms of budget allocation Revenue goes to the general fund and is later assigned to the official actor(s) specified in the earmark. Revenues do not go through the general budget but are instead paid into a separate account belonging to the official actor(s) specified in the earmark. Earmarked tax revenue is paid directly to the account of the entity managing an autonomous or semi- autonomous fund. Examples In the Philippines, revenue goes to the general fund before being allocated to the Ministry of Health following submission of a budget for its use (140). In Romania, revenue goes directly into a Ministry of Health account that is distinct from the general fund. In Panama, revenue is paid into subaccounts of the three recipient agencies (the Ministry of Health, the National Cancer Institute and the Customs Authority). In Thailand, ThaiHealth directly receives the earmarked revenues in its own account. In Viet Nam, the Viet Nam Tobacco Control Fund receives the revenues directly into a subaccount it manages but that belongs to the Ministry of Health. Source: (151). Where is the money being spent? Earmarked tobacco tax revenues are used for a variety of health purposes, including tobacco control, health promotion and UHC. A wide variety of other programmes have also been funded with earmarks from tobacco taxes, including disaster relief (e.g. hospital medical supplies and equipment to treat COVID-19 in India), youth pro- grams, sports and craft jobs in Yemen, social cohesion in Morocco, health and social programs in areas dependent on tobacco growing in Argentina, health promotion CHAP T ER 4. PO LI T I C AL ECO N OMY 239 and tobacco control in Thailand and alternative livelihood programs for tobacco farmers as well as economic projects in tobacco-growing provinces in the Philippines. Table 4.11 shows the three main categories of health programmes to which ear- marked tobacco tax revenue is allocated, as well as a fourth miscellaneous category, with country-specific examples for each.26 Table 4.11 Programmes to which earmarked tobacco tax revenue is allocated TOBACCO CONTROL NCD PREVENTION AND CONTROL PROGRAMMES (otherwise indicated between brackets) HEALTH COVERAGE EXPANSION (e.g. through health insurance coverage) OTHER, MORE GENERAL OR UNSPECIFIED HEALTH PROGRAMMES Costa Rica, Côte d’Ivoire, the Islamic Republic of Iran, Madagascar, Panama (tobacco cessation and fighting illicit trade), Switzerland, Viet Nam Cook Islands, Costa Rica, Mauritania (anti-cancer research), Palau (NCD prevention only), Panama (National Institute of Oncology), Paraguay Colombia, Congo, Egypt, Palau, Philippines Algeria, Argentina, Bangladesh, Botswana, Cabo Verde, Chad (programmes delivering antiretroviral drugs), Colombia (sports), Comoros (sports, hospital emergencies), Congo (sports), Côte d’Ivoire (AIDS programme), El Salvador, Estonia (sports), Guatemala, Indonesia, the Islamic Republic of Iran (sports), Ireland, Jamaica, Lithuania (sports), Madagascar (sports), Morocco, Nepal, Paraguay (sports), Republic of Korea (health promotion), Romania, Thailand (health promotion), United States, Yemen (sports) Note: Countries appear in more than one column when their earmarked tax revenues are used in more than one specific health programme. Source: (27). The amount of money associated with tobacco tax earmarks Case studies of the experiences of nine countries in tobacco tax earmarking show that earmarked funds are relatively small in comparison with government spend- ing on health (see Table 4.12) and, consequently, even smaller in terms of GDP. Therefore, the argument that tobacco tax earmarks would introduce rigidity into public financial management may not apply. 26 Details about how the tobacco tax revenues are earmarked are given in Annex 4.2. 240 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.12 Proportion of earmarked tobacco tax funds in government expenditures COUNTRY ESTIMATED ANNUAL TOTAL FUNDS FROM EARMARKED TAX ANNUAL FUNDS FROM TOBACCO TAX EARMARKS as a % of general government expenditure on health in 2013 Botswana 2014–2015: 4 million pula (US$ 0.48 million) NA Egypt 2013–2014: 392 million Egyptian pounds (US$ 52.06 million); earmarked taxes only 1.8% of total taxes on cigarettes 1.086% Iceland 2014: 108.3 million kronor (US$ 0.89 million) 0.083% Panama 2014: US$ 27.8 million 1.322% Philippines 2014: 50.18 billion Philippine pesos (US$ 1.18 billion) NA Poland 2013: 1 million złoty (US$ 0.316 million) from general budget  0.001% Romania 2014: 1.1 million lei (US$ 0.33 million); 14.4% of total health budget 0.004% Thailand 2014: 4064.74 million baht (US$ 125.15 million); 1.78% of Ministry of Health budget and 1.84% of National Health Security Fund 0.932% Viet Nam 2014: 299.171 billion dong (US$ 13.91 million); 0.5% of national health budget 0.335% Source: (151). 4.6.4 CONCLUSIONS Despite the initial principled resistance to earmarking by some ministries of finance, experience has shown that the use of revenue from tobacco taxes and other taxes on the consumption of products that have negative externalities can ensure political as well as public support. Successful earmarking needs a well-developed structure for the use of funds for health purposes. Even intergovernmental organizations that are opposed to earmarking (e.g. the IMF) have acknowledged the justifiability of well-designed tobacco tax earmarks when revenue is directed to specific cost- effective programmes (153–154). The amounts of tobacco tax revenue effectively earmarked for health have been relatively small and could hardly introduce the feared rigidity in government budgets. Moreover, in some countries, those funds have helped to implement much-needed health programmes (e.g. Australia, the Philippines, Thailand). More governments are considering this option as a stable medium-term source of secure funding for programmes such as tobacco control. The payoffs will be seen in the future as fewer people fall ill and less medical care for tobacco-related illnesses is needed. In Australia, CHAP T ER 4. PO LI T I C AL ECO N OMY 241 an earmarked tax was used to fund a needed and underresourced programme that proved to be successful, effective and impactful; the programme is now sustainably funded, embedded in the federal budget. Earmarking is desirable in a particular political economy when it enables the implementation of effective tobacco taxation that will increase price and reduce consumption. It will, however, also be rational as a matter of public financial man- agement, economic efficiency and democratic governance when concerns such as the following are considered. Although not every question needs an affirmative answer, policy-makers who can answer yes to many of the following questions will likely be considering an effective and rational tobacco tax earmark:27 • Does the tobacco tax earmark’s purpose rationally connect with the recipient programme’s purpose? Earmarks that fund tobacco control or other health programmes are more economically rational under the benefit principle than those that fund unrelated programmes such as childhood education, even when the popularity of the unrelated programmes may make a tax increase politically palatable. • Does the tobacco tax earmark’s amount rationally connect with the needs of the recipient programme? Earmarked funds that cannot be absorbed by the recipient programme are, in effect, money taken away from other needs. • When a tobacco tax earmark funds health programmes, is this clearly com- municated to the public to ensure that the framing of the tobacco tax increase as a health measure reinforces the demand-reduction effect? • Is the programme being funded by the tobacco tax earmark a politically neglected but highly cost-effective or crucially needed programme that, once established as a proof of concept, has a chance of being funded out of the general budget? • Does the tobacco tax earmark’s purpose rationally connect with the effects of the tax itself? Earmarks that fund programmes that disproportionately benefit lower socioeconomic groups or that fund alternative livelihoods for former tobacco workers and farmers will have equity-enhancing effects that will reinforce the already progressive nature of tobacco taxes. • Is the scope of the earmark’s purpose narrow enough that it can be funded mostly from the tobacco tax earmark, to ensure that the revenue is additive and does not merely substitute for spending that would otherwise come from the general fund? • Does the design of the tobacco tax earmark provide for flexibilities that ensure that windfall revenue collection is not squandered on a purpose already saturated with overfunding? 27 Adapted from and informed by References 138–139, 143, 150. 242 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Does the design of the tobacco tax earmark include a sunset clause that triggers its automatic end or review? The presence of such a clause will guard against inefficiency in allocation arising and being left uncorrected due to political inertia. CHAP T ER 4. PO LI T I C AL ECO N OMY 243 REFERENCES 1. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tob Control. 2013; 22(2): 144–153 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 31 January 2021). 2. LaFaive MD, Drenkard S, Nesbit T. Strong link between cigarette tax and illegal smuggling rates [website]. Midland: Mackinak Center for Public Policy; 2018 (https://www.mackinac.org/strong- link-between-cigarette-tax-and-illegal-smuggling-rates#:~:text=An%20increase%20of%2011%20 percentage,packs%20to%2030.6%20million%20packs.&text=Tax%20evasion%20and%20avoidance%20 is,high%20excise%20taxes%20on%20cigarettes., accessed 31 January 2021). 3. Ross H, Blecher E. Illicit trade in tobacco products need not hinder tobacco tax policy reforms and increases. Chicago: University of Illinois at Chicago; 2019 (Tobacconomics White Paper; https:// tobacconomics.org/research/illicit-trade-in-tobacco-products-need-not-hinder-tobacco-tax- policy-reforms-and-increases-white-paper/#:~:text=additional%20related%20research-,Illicit%20 Trade%20in%20Tobacco%20Products%20Need%20Not%20Hinder%20Tobacco,Reforms%20 and%20Increases%20( White%20Paper)&text=The%20paper%20cites%20research%20 that,consumption%2C%20and%20improves%20overall%20health, accessed 31 January 2021). 4. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 12 October 2020). 5. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 6. Joossens L, Chaloupka FJ., Merriman D, Yürekli A. Issues in smuggling of tobacco products. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac00Box 370064B00PUBLIC0.pdf, accessed 18 February 2021). 7. Merriman D, Yürekli A., Chaloupka, FJ. How big is the worldwide cigarette smuggling problem. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac 00Box370064B00PUBLIC0.pdf, accessed 18 February 2021). 8. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 9. Merriman D. Understand, measure, and combat tobacco smuggling. Washington (DC): World Bank Group; 2003 (Economics of tobacco toolkit 7; https://openknowledge.worldbank.org/bitstream/ handle/10986/16267/805800NWP070Sm0Box0379807B00PUBLIC0.pdf?sequence=1&isAllowed=y, accessed 31 January 2021). 10. Progress report on the implementation of the Commission Communication. Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products – a comprehensive EU strategy. Brussels: European Commission; 2017 (COM(2017) 235 final; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:52017DC0235&from=en, accessed 31 January 2021). 11. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 12. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Reuter P, Majmundar M. (editors). Washington (DC): National Academies Press; 2015. 13. von Lampe K. The illegal cigarette trade. In: Natarajan M, editor. International crime and justice. New York: Cambridge University Press. 2010:148–54. 14. Joossens L, Raw M, From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–34 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 31 January 2021). 15. Gomis B, Lee K, Carrillo Botero N, Shepherd P, Iglesias RM. “We think globally”: the rise of Paraguay’s 24 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tabacalera del Este as a threat to global tobacco control. Global Health. 2018;14(1):110 (https:// globalizationandhealth.biomedcentral.com/articles/10.1186/s12992-018-0412-3, accessed 31 January 2021). 16. Communication from the commission to the Council and the European Parliament: Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products - A comprehensive EU Strategy. Brussels: European Commission; 2013 (https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52013DC0324&from=EN, accessed 31 January 2021). 17. Ross H, Vellios N, Smith KC, Ferguson J, Cohen JE. A closer look at ‘cheap white’ cigarettes. Tob Control. 2016;25(5):527–31 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/527.full.pdf, accessed 31 January 2021). 18. Nagelhout GE, van den Putte B, Allwright S, Mons U, McNeill A, Guignard R, et al. Socioeconomic and country variations in cross-border cigarette purchasing as tobacco tax avoidance strategy. Findings from the ITC Europe Surveys. Tob Control. 2014;23:i30–i38 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/23/suppl_1/i30.full.pdf, accessed 1 February 2021). 19. Gilmore AB, Rowell A, Gallus S, Lugo A, Joossens L, Sims M. Towards a greater understanding of the illicit tobacco trade in Europe: a review of the PMI funded ‘Project Star’ report. Tob Control. 2014;23:e51–e61 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/e1/e51.full.pdf, accessed 1 February 2021). 20. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Revista Panam Salud Publica. 2016;40(4):243–9 (https://www.scielosp.org/pdf/ rpsp/2016.v40n4/243-249, accessed 1 February 2021). 21. Operação Huno combate mercado clandestino de cigarros em sete Estados [Operação Huno fights clandestine cigar market in seven states]. Brasilia: Federal Police of Brazil; 2015 (in Portuguese) (http:// www.pf.gov.br/agencia/noticias/2015/11/operacao-huno-combate-mercado-clandestino-de-cigarros- em-seis-estados, accessed 12 October 2020). 22. Gilmore AB, Gallagher AWA, Rowell A. Tobacco industry’s elaborate attempts to control a global track and trace system and fundamentally undermine the Illicit Trade Protocol. Tob Control. 2019;28:127–40 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/28/2/127.full.pdf, accessed 1 February 2021). 23. Technical assessment of the experience made with the Anti-Contraband and Anti-Counterfeit Agreement and general release of 9 July 2014 among Philip Morris International and affiliates, the Union and its Member States. Brussels: European Commission; 2016 (SWD (2016) 44 final; https://ec.europa.eu/anti- fraud/sites/antifraud/files/technical_assessment_pmi_24022016_en.pdf, accessed 1 February 2021). 24. Gallagher AWA, Evans-Reeves KA, Hatchard JL, Gilmore AB. Tobacco industry data on illicit tobacco trade: a systematic review of existing assessments. Tob Control. 2019;28(3):334–45 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/28/3/334.full.pdf, accessed 1 February 2021). 25. Joossens L, Merriman D, Ross H, Raw M. How eliminating the global illicit cigarette trade would increase tax revenue and save lives. Paris: International Union Against Tuberculosis and Lung Disease; 2009 (https://theunion.org/sites/default/files/2020-08/How%20Eliminating%20the%20Global%20Illicit%20 Cigarette%20Trade%20Would%20Increase%20Tax%20Revenue%20and%20Save%20Lives%20-%20 EN.pdf, accessed 1 February 2021). 26. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 27. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 28. Prieger JE, Kulic J. Cigarette taxes and illicit trade in Europe. Econ Inquiry. 2018;56(3):1706–23 (https:// papers.ssrn.com/sol3/papers.cfm?abstract_id=2718519, accessed 18 February 2021). 29. Stoklosa M. Prices and cross-border cigarette purchases in the EU: evidence from demand modelling. Tob Control. 2020;29:55–60. 30. Passport: illicit trade in tobacco products. In: Euromonitor International [website]. London: Euromonitor International; 2018 (restricted access). 31. The Global Illicit Trade Environment Index. London: The Economist Intelligence Unit; 2018 (http:// illicittradeindex.eiu.com/documents/EIU%20Global%20Illicit%20Trade%20Environment%20Index%20 CHAP T ER 4. PO LI T I C AL ECO N OMY 245 2018%20-%20Overall%20Results%20White%20Paper%20June%206%20FINAL.pdf, accessed 1 February 2021). 32. Chaloupka FJ, Yürekli A, Fong GT. Tobacco taxes as a tobacco control strategy. Tob Control. 2012;21:172– 80 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/172.full.pdf, accessed 1 February 2021). 33. Chernick H, Merriman D. Using littered pack data to estimate cigarette tax avoidance in NYC. Natl Tax J. 2013;66(2):635–68 (https://www.ntanet.org/NTJ/66/3/ntj-v66n03p635-668-littered-pack-data- cigarette-tax.pdf, accessed 1 February 2021). 34. Liber AC, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia Survey. Tob Control. 2015;24:iii83–iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed 1 February 2021). 35. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 36. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 37. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 38. Paraje G, Araya D, Drope J. Illicit cigarette trade in metropolitan Santiago de Chile. Tob Control 2020;29:68–73. 39. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 40. Saenz de Miera Juarez BS, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020(0):1-7 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 1 February 2021). 41. Lee K, Collin J. ‘Key to the future’: British American Tobacco and cigarette smuggling in China. PLoS Medicine. 2006;3(7):e228 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1502159/pdf/pmed.0030228. pdf, accessed 2 February 2021). 42. Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38– i43 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i38.full.pdf, accessed 2 February 2021). 43. Aziani A, Kulick J, Norman N, Prieger JE. Empty discarded pack data and the prevalence of illicit trade in cigarettes; 2017. (https://ssrn.com/abstract=2906015, accessed 12 October 2020). 44. Philip Morris International. The Perspective of PM International on Smoking and Health Issues. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1985. Bates no. 2023268328-2023268349 (https://www.industrydocuments.ucsf.edu/tobacco/docs/#id=pxpb0040, accessed 5 February 2021). 45. Mandal S, Gilmore A, Collin J, Weishaar H, Smith K, McKee M. Block, amend, delay: tobacco industry efforts to influence the European Union’s Tobacco Products Directive (2001/37/EC). Smokefree Partnership/Cancer Research UK; 2009 (http://www.smokefreepartnership.eu/images/Reports-and- Position-Papers/EU_TI_TPD_report_May_2012.pdf, accessed 15 October 2020). 46. Bishop DJ. Key area paper: indirect taxation – British American Tobacco. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1994. Bates no. 503911771-503911780 (http:// bat.library.ucsf.edu/data/h/f/k/hfk11a99/hfk11a99.pdf, accessed 15 October 2020). 47. Zhou S, Liberman JD, Ricafort E. The impact of the WHO Framework Convention on Tobacco Control in defending legal challenges to tobacco control measures. Tob Control. 2009:28:s113-s118 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/28/Suppl_2/s113.full.pdf, accessed 2 February 2021). 48. Gilmore AB, Fooks G, Drope J, Bialous SA, Jackson RR. Exposing and addressing tobacco industry 246 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N conduct in low-income and middle-income countries. Lancet. 2015;385:1029–43. 49. Ulucanlar S, Fooks G, Gilmore A. The policy dystopia model: an interpretive analysis of tobacco industry political activity. PLoS Med. 2016:13(9):e1002125 (https://journals.plos.org/plosmedicine/ article?id=10.1371/journal.pmed.1002125, accessed 2 February 2021). 50. Apollonio D, Glantz S. Tobacco manufacturer lobbying to undercut minimum price laws: an analysis of internal industry documents. Tob Control. 2020;29:e10-e17 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC7374022/pdf/nihms-1590806.pdf, accessed 2 February 2021). 51. Crosbie E, Glantz S. Tobacco industry argues domestic trademark laws and international treaties preclude cigarette health warning labels, despite consistent legal advice that the argument is invalid. Tob Control. 2012;23:e7 (https://www.researchgate.net/publication/233769722_Tobacco_industry_ argues_domestic_trademark_laws_and_international_treaties_preclude_cigarette_health_warning_ labels_despite_consistent_legal_advice_that_the_argument_is_invalid, accessed 2 February 2021). 52. Puig S. Tobacco litigation in international courts. Harv Int Law J. 2016;57(2):1–68. 53. Crooked nine: nine ways the tobacco industry undermines health policy. New York: Vital Strategies; 2019 (https://exposetobacco.org/wp-content/uploads/2019/09/Crooked-9-STOP.pdf, accessed 2 February 2021). 54. Balwicki L, Stokłosa M, Balwicka-Szczyrba M, Tomczak W. Tobacco industry interference with tobacco control policies in Poland: legal aspects and industry practices. Tob Control. 2016;25(5):521–26 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/521.full.pdf, accessed 2 February 2021). 55. Sud SR, Brenner JE, Shaffer ER. Trading away health: the influence of trade policy on youth tobacco control. J Pediatrics. 2015;166(5):1303–07 (https://www.jpeds.com/action/showPdf?pii =S0022-3476%2815%2900017-7, accessed 2 February 2021). 56. Lee S, Ling PM, Glantz SA. The vector of the tobacco epidemic: tobacco industry practices in low and middle-income countries. Cancer Cause Control. 2012;23:117–29 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC3332051/pdf/nihms-367996.pdf, accessed 2 February 2021). 57. Zhou S, Liberman J. The global tobacco epidemic and the WHO Framework Convention on Tobacco Control—the contributions of the WHO’s first convention to global health law and governance. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018 (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3339718, accessed 18 February 2021). 58. Legislative Consultation with Constitutional Division of the Supreme Court, No. 2012-003918. Constitutional Division of the Supreme Court of Costa Rica; 20 March 2012 (https://www. tobaccocontrollaws.org/files/live/litigation/830/CR_Legislative%20Consultation%20with%20.pdf, accessed 2 February 2021). 59. British American Tobacco Kenya Ltd v Ministry of Health, Petition No. 143 of 2015. High Court of Kenya; 2016 (https://www.tobaccocontrollaws.org/files/live/litigation/2643/KE_British%20American%20 Tobacco%20Kenya.pdf, accessed 2 February 2021). 60. British American Tobacco v Ministry of Health, et al, Civil Appeal No. 112 of 2016. Kenyan Court of Appeal; 2017 (https://www.tobaccocontrollaws.org/files/live/litigation/2521/KE_British%20 American%20Tobacco%20Ltd%20v.pdf, accessed 2 February 2021). 61. Thailand – customs and fiscal measures on cigarettes from the Philippines: Report of the Panel. Geneva: World Trade Organization; 2010 (WT/DS371/R; https://docs.wto.org/dol2fe/Pages/SS/directdoc. aspx?filename=Q:/WT/DS/371R.pdf&Open=True, accessed 2 February 2021). 62. Tribunal Constitucional de Chile [Constitutional Court of Chile] (1995 Rol No. 219-95; https://www. tribunalconstitucional.cl/descargar_sentencia2.php?id=3074, accessed 4 April 2020.) 63. McGrady B. International investment law. In: Voon T, Mitchell AD, Liberman J, editors. Regulating tobacco, alcohol and unhealthy foods: the legal issues. New York: Routledge; 2014. 64. Marvin Roy Feldman Karpa v United Mexican States, Case No. ARB(AF)/99/1 International Centre for Settlement of Investment Disputes; 2002 (https://www.italaw.com/sites/default/files/case-documents/ ita0319.pdf, accessed 2 February 2021). 65. Feldman Karpa v Mexico (2002 ICSID Arbitral Tribunal) (https://www.italaw.com/sites/default/files/ case-documents/ita0319.pdf, accessed 18 February 2021). 66. Undang-Undang Nomor 11 Tahun 1995 tentangCukai (Lembaran Negara Republik Indonesia Tahun1995 CHAP T ER 4. PO LI T I C AL ECO N OMY 247 Nomor 76) (Excise Law,in Indonesian). Article 5. 67. Excise Regulation (2011) Reference No. PMK Number 167/PMK.011/2011 (in Indonesian) (https:// peraturan.bpk.go.id/Home/Details/145897/pmk-no-167pmk0112011, accessed 3 February 2021). 68. Mahkamah Agung. Indonesia Supreme Court, Case No. 73/PDT/2018/PT SBY (https://putusan3. mahkamahagung.go.id/direktori/putusan/74efff9cfe18d60990fcde19583ecd5f.html, accessed 6 February 2021). 69. Maya D. Analisis kebijakan kenaikam tarif cukai dan penyesuaian batasan harga jual eceran hasil tembakau [Policy analysis on excise tax increases and price limit adjustments retail tobacco products]; 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-09/S45937-Deacy%20Maya accessed 3 March 2020). 70. Indriawati EN, Susanti L, Qurnia MA. Kebijakan tarif cukai hasil tembakau terhadap perekonomian [Tobacco excise tariff policy on the economy]; 2014 (in Indonesian). (https://www.academia. edu/6611977/MAKALAH_CHT_kel_6, accessed 15 October 2020). 71. Arroyani DW. Evaluasi Kebijakan Penetapan Tarif Cukai Hasil Tembakau Berdasarkan PMK [Evaluation of Tobacco Excise Tariff Determination Policy Based on PMK].167/PMK.011/2011. 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-08/S44664-Dwira%20Wanti%20Arroyan, accessed 15 October 2020). 72. Ngo Ngo Ha and Another v. State of New South Wales and Others (Matter No S 45 of 1996); Walter Hammond & Associates Pty Ltd v. State of New South Wales and Others (Matter No S 165 of 1996). Constitutional law; (1997) 146 ALR 355: 5 August 1997 (https://staging.hcourt.gov.au/assets/ publications/judgments/1997/029--NGO_NGO_HA_AND_ANOR_v_STATE_OF_NEW_SOUTH_WALES_&_ ORS_(Matter_No_S_45_of_1996);_WALTER_HAMMOND--(1997)_146_ALR_355.html, accessed 2 February 2021). 73. McGrady B. Trade and public health: the WTO, tobacco, alcohol, and diet. New York: Cambridge University Press; 2011. 74. Giorgetti C. Health and international investment law. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018. 75. British American Tobacco v Attorney General and Center for Human Rights and Development. Constitutional Court of Uganda Petition No.46 of 2016; 2017 (https://media.ulii.org/files/judgments/ ugcc/2019/4/2019-ugcc-4_0.pdf, accessed 18 February 2021). 76. Lim CL, Ho J, Paparinskis M. International investment law and arbitration: commentary, awards and other materials. Cambridge: Cambridge University Press; 2018. 77. Dominican Republic – measures affecting the importation and internal sale of cigarettes : Report of the Panel. Geneva: World Trade Organization; 2004 (WT/DS302/R; https://docs.wto.org/dol2fe/ Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302R.pdf&Open=True, accessed 21 February 2021) 78. Dominican Republic — measures affecting the importation and internal sale of cigarettes: Report of the Appellate Body. Geneva: World Trade Organization; 2005 (WT/DS302/AB/R; https://docs.wto. org/dol2fe/Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302ABR.pdf&Open=True, accessed 21 February 2021). 79. Gilmore A. Tobacco and transition: understanding the impact of transition on tobacco use and control in the former Soviet Union [Thesis]. London: London School of Hygiene and Tropical Medicine, June 2005 (https://escholarship.org/uc/item/3rw2c04c, accessed 2 February 2021). 80. MacKenzie R, Collin J, Sopharo C, Sopheap Y. ‘Almost a role model of what we would like to do everywhere’: British American Tobacco in Cambodia. Tob Control. 2004;13(suppl II): ii112–ii117 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/13/suppl_2/ii112.full.pdf, accessed 2 February 2021). 81. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tobacco Control 2013;22(2):144-53 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 21 February 2021). 82. Price and tax. In: TobaccoTactics [website]. Bath: University of Bath; 2020 (https://tobaccotactics.org/ topics/price-and-tax/, accessed 15 December 2020). 83. Vulovic V, Chaloupka FJ. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation—commentary. Tob Control. 24 June 2020. 248 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 84. Verguet S, Kearns PKA, Rees VW. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation. Tob Control. 23 June 2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/08/11/tobaccocontrol-2019-055315. full.pdf, accessed 2 February 2021).. 85. Palipudi KM, Gupta PC, Sinha DN, Andes LJ, Asma S, McAfee T. Social determinants of health and tobacco use in thirteen low and middle income countries: evidence from Global Adult Tobacco Survey. PLoS ONE. 2012;7(3):e33466 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3306395/pdf/ pone.0033466.pdf, accessed 2 February 2021). 86. Systematic review of the link between tobacco and poverty. Geneva: World Health Organization; 2014 (http://www.who.int/tobacco/publications/economics/9789241507820/en, accessed 21 October 2020). 87. Goodchild M, Sinha P, Gill Munish V, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/bitstream/handle/10665/331836/whoseajph2020v9i1p73. pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 88. Faruque GM, Ahmed M, Huq I, Parven R, Wadood SN, Chowdhury SR, et al. The economic cost of tobacco use in Bangladesh: a health cost approach. Dhaka: Bangladesh Cancer Society; 2020 (https://www. researchgate.net/publication/342445467_The_Economic_Cost_of_Tobacco_Use_in_Bangladesh_A_ Health_Cost_Approach, accessed 2 February 2021). 89. John RM, Sung H, Max WB, Ross H. Counting 15 million more poor in India, thanks to tobacco. Tob Control. 2011;20:349–52 (https://www.researchgate.net/publication/49809290_Counting_15_million_ more_poor_in_India_thanks_to_tobacco, accessed 2 February 2021). 90. Liu Y, Rao K, Hu TW, Sun Q, Mao Z. Cigarette smoking and poverty in China. Soc Sci Med. 2006;Dec;63(11):2784–90. 91. Fuchs A, González Icaza F, Paz D. Distributional effects of tobacco taxation – a comparative analysis. Washington (DC): World Bank Group; 2019 (Policy research working paper 8805; https://openknowledge. worldbank.org/bitstream/handle/10986/31534/WPS8805.pdf?sequence=4&isAllowed=y, accessed 2 February 2021). 92. Smith CE, Hill SE, Amos A. Impact of population tobacco control interventions on socioeconomic inequalities in smoking: a systematic review and appraisal of future research directions. Tob Control. 29 September 2020 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/09/28/ tobaccocontrol-2020-055874.full.pdf, accessed 2 February 2021). 93. Furman J. Six lessons from the U.S. experience with tobacco taxes – World Bank conference “Winning the tax wars: global solutions for developing countries”. Washington (DC): Obama White House Archives; 24 May 2016 (https://obamawhitehouse.archives.gov/sites/default/files/page/files/20160524_cea_ tobacco_tax_speech.pdf, accessed 21 October 2020). 94. Önder Z, Yürekli A. Who pays the most cigarette tax in Turkey. Tob Control. 2016;25(1):39–45. 95. Verguet S, Gauvreau CL, Mishra S, MacLennan M, Murphy SM, Brouwer ED, et al. The consequences of tobacco tax on household health and finances in rich and poor smokers in China: an extended cost-effectiveness analysis. Lancet Glob Health. 2015;3:e206–16 (https://www.thelancet.com/action/ showPdf?pii=S2214-109X%2815%2970095-1, accessed 2 February 2021). 96. Fuchs Tarlovsky A, González Icaza F. The welfare and distributional effects of increasing taxes on tobacco in Vietnam. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/ curated/en/758391561477648302/pdf/The-Welfare-and-Distributional-Effects-of-Increasing-Taxes- on-Tobacco-in-Vietnam.pdf, accessed 2 February 2021). 97. Expanding the global tax base: taxing to promote public goods: tobacco taxes. Washington (DC): World Bank Group; 2016 (http://pubdocs.worldbank.org/en/799981480947591462/Taxing-to-Promote- Public-Goods-tobacco-taxation-chapter-conference-25-nov-2016-final-version-pmarquez-002.pdf, accessed 2 February 2021). 98. Blecher E. Increasing tobacco taxes significantly will increase revenues. Chicago: University of Illinois at Chicago; 2018. (Tobacconomics policy brief; https://tobacconomics.org/files/research/470/Laffer- Curve-Policy-Brief_v2.2.pdf, accessed 2 February 2021). 99. Savedoff W, Lopert R. Sugar, rum, and tobacco: domestic resource mobilization for low-income countries through excise taxes. Washington (DC): Centre for Global Development; 2020 (CGD Policy Paper CHAP T ER 4. PO LI T I C AL ECO N OMY 249 164, https://www.cgdev.org/sites/default/files/PP164-Savedoff-Health-Taxes-DRM.pdf, accessed 2 February 2021). 100. Lavado R, Postolovska I, Sotomayor RE. Beyond the perceived regressivity of higher tobacco taxes, turning short-term losers into long-term winners? In: Tobacco reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 101. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 102. Tonga NCD related taxation policy assessment preliminary findings. Presentation made at the innovations and actions against NCDs conference ADB headquarters, Manila, 17 July 2018. 103. Moreno-Dodson B. Raising tobacco taxes: why, what, and how: where public health and economics converge. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e- library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 104. Fuchs A, Del Carmen G, Mukong AK. Long-run impacts of increasing tobacco taxes: evidence from South Africa. Policy Research Working Paper 8369. Washington [DC]: World Bank; 2018 (https://openknowledge. worldbank.org/bitstream/handle/10986/29497/WPS8369.pdf?sequence=4&isAllowed=y, accessed 18 February 2021) . 105. van Walbeek C. Case studies in illicit tobacco trade: South Africa. Tobacconomics fact sheet. Chicago: University of Chicago, Tobacconomics, Health Policy Center; 2020 (https://tobacconomics.org/files/ research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 106. van Loggerenberg, J. Tobacco Wars. Cape Town: NB Publishers; 2019. 107. Marquez P. Global tobacco control: a development priority. In: Tobacco tax reform at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads. pdf, accessed 2 February 2021). 108. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington [DC]: World Bank; 2017 (https:// tobacconomics.org/files/research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 109. Scollo M, Bayly M. Revenue from tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia. org.au/chapter-13-taxation/13-6-revenue-from-tobacco-taxes-in-australia, accessed 21 October 2020). 110. Scollo M, Bayly M. Tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia.org.au/ chapter-13-taxation/13-2-tobacco-taxes-in-australia, accessed 21 October 2020). 111. Breton E, Richard L, Gagnon F, Jacques M, Bergeron P. Fighting a tobacco-tax rollback: a political analysis of the 1994 cigarette contraband crisis in Canada. Journal of Pub Health Policy. 2006;27:77–99. 112. Fanta E, Garcia M. Preventing and controlling illicit tobacco flows. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB- Report-CrossRoads.pdf, accessed 2 February 2021). 113. Goodchild M, Perucic AM, Zheng R, Blecher E, Paul J. The health impact of raising tobacco taxes in developing countries. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/ taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 114. Ross H. Tracking and tracing tobacco products in Kenya. Prev Med. 2017;105:S15–S18 (https://reader. elsevier.com/reader/sd/pii/S0091743517301457?token=424D9EC1D42A9C69BDB55B90090C3F7853AA 20A4970D0E379C91EE64C73A273A0FA949557F4F120A0B636701581EC055, accessed 2 February 2021). 115. Mamudu H, Hammond R, Glantz S. Tobacco industry attempts to counter the world bank report curbing the epidemic and obstruct the WHO framework convention on tobacco control. Soc Sci & 250 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Med. 2008;67(11):1690–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2662513/pdf/nihms81164. pdf, accessed 2 February 2021). 116. Gilmore A. Understanding the vector in order to plan effective tobacco control policies: an analysis of contemporary tobacco industry materials. Tob Control. 2012;21:119–26 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/21/2/119.full.pdf, accessed 2 February 2021). 117. The economics of tobacco taxation and employment in Indonesia: policy implications technical brief. Washington (DC): World Bank Group; 2018 (https://openknowledge.worldbank.org/bitstream/ handle/10986/29814/126158-WP-P154568-PUBLIC-add-series-WBGPNIndoEmploymentUPDATEFin alWeb.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 118. Sahadewo GA, Drope J, Li Q, Nargis N, Witoelar F. Tobacco or not tobacco: predicting farming households’ income in Indonesia. Tob Control. 2020. 119. The economics of tobacco farming in Indonesia: health, population, and nutrition global practice. Washington (DC): World Bank Group, 2017 (http://documents1.worldbank.org/curated/ en/161981507529328872/pdf/120307-REVISED-PUBLIC-WBGIndoEconomicsTobaccoFarming.pdf, accessed 2 February 2021). 120. The revenue and employment outcome of biri taxation in Bangladesh. Dhaka: National Board of Revenue, Government of the People’s Republic of Bangladesh; 2019 (https://www.who.int/docs/ default-source/searo/bangladesh/pdf-reports/cat-2/biri-study-report-03-12-2019.pdf?sfvrsn=b8fea69_2, accessed 2 February 2021). 121. Magati P, Lencucha R, Li Q, Drope J, Labonte R, Appau A, et al. Costs, contracts and the narrative of prosperity: an economic analysis of smallholder tobacco farming livelihoods in Kenya. Tob Control. 2019;28(3):268–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6512316/pdf/nihms-1012064. pdf, accessed 2 February 2021). 122. Ghaus K, Iqbal MA, Sabir M, Saleem W, Ali A, Aamir N. Macroeconomic impacts of tobacco use in Pakistan. Karachi: The Social Policy and Development Centre; 2018 (https://tobacconomics.org/files/ research/483/Macroeconomic-impact-of-Tobacco-in-Pakistan.pdf, accessed 2 February 2021). 123. Ross H. The effects of tobacco taxes on employment and labor productivity. In: Tobacco tax reform: at the crossroads of health and development. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021) 124. Bigawanto M. Tobacco industry interference undermined tobacco tax policy in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance (SEATCA); 2019 (https://seatca.org/dmdocuments/ Indonesia%20TII%20in%20Tax.pdf, accessed 2 February 2021). 125. Ahsan A, Wyono N. Analysis of the impact of higher cigarettes prices on employment in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance; 2007 (https://seatca.org/dmdocuments/An%20 Analysis%20of%20the%20Impact%20of%20Higher%20Cigarette%20Prices%20on%20Employment%20 in%20Indonesia.pdf, accessed 2 February 2021). 126. Higashi H, Khuong TA, Ngo AD, Hill PS. Evidence and decision making: tobacco control policy and legislation in Vietnam. Int J of Health Plann. 2013;2(1):e72–e94. 127. Buck D, Raw M, Godfrey CA, Sutton M. Tobacco and jobs: the impact of reducing consumption on employment in the UK. York: University of York; 1995 (https://www.researchgate.net/ publication/4814523_Tobacco_and_jobs_the_impact_of_reducing_consumption_on_employment_ in_the_UK, accessed 2 February 2021). 128. Petkova R, Ileva R and Rusenov A. Tobacco control policies and tobacco industry employment in Bulgaria. In: Economic, social and health issues in tobacco control. Report of a WHO international meeting, Kobe, Japan, 3-4 December 2001. Geneva: World Health Organization; 2003. 129. Jha V, Narayanan BG, Wadhwa D, Tesche J. Economic and environmental effects of reduction in smoking prevalence in Tanzania. Tob Control; 2020;29:24–8 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/29/1/24.full.pdf, accessed 2 February 2021). 130. Lencucha R, Makoka D, Drope J, Appau A, Labonté R, Li Q, et al. Costs, revenues and profits: an economic analysis of smallholder tobacco farmer livelihoods in Malawi. Tob Control. 2017;26(6)634–40 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC5957074/pdf/nihms965842.pdf, accessed 2 February 2021). 131. Goma FM, Labonté R, Drope J, Li Q, Zulu R, Kangwa E. The economics of tobacco farming in Zambia: CHAP T ER 4. PO LI T I C AL ECO N OMY 251 tobacco farmers survey report. Lusaka: University of Zambia and American Cancer Society; 2019 (https://www.cancer.org/content/dam/cancer-org/research/economic-and-healthy-policy/zambia- report-7-16-19.pdf, accessed 28 October 2020). 132. Chavez JJ, Drope J, Li Q, Aloria MJ. The economics of tobacco farming in the Philippines. Quezon City: Action for Economic Reforms and Atlanta: American Cancer Society, 2016 (https://www.cancer. org/content/dam/cancer-org/research/economic-and-healthy-policy/the-economics-of-tobacco- farming-in-the-philippines.pdf, accessed 22 October 2020). 133. Augstin CP, Cardenas RP, Cortez JB, Gimongala GC, Ramos AR. The effects of the sin tax reform law of 2012 to tobacco farmers of Amulung, Cagayan. IJARMSS. 2017;6(12):57–75. 134. Mercado A. Tobacco tax: good for the tobacco farmers, too. Business World; 2019 [website] (https:// www.bworldonline.com/tobacco-tax-good-for-the-tobacco-farmers-too/, accessed 25 October 2020). 135. Yürekli A, Önder Z, Elibol M, Erk N, Cabuk A, Fisunoglu M, et al. The economics of tobacco and tobacco taxation in Turkey. Paris: International Union Against Tuberculosis and Lung Disease; 2010 (https:// www.who.int/tobacco/en_tfi_turkey_report_feb2011.pdf, accessed 2 February 2021). 136. Latif E, Warner V. Legislation and smoking prevention. In: Loddenkemper R, Kreuter M, editors. The Tobacco Epidemic: progress in Respiratory Research. 2015;42:136–148. 137. Tandon A, Cashin C. Assessing public expenditure on health from a fiscal space perspective. Washington (DC): World Bank Group; 2010 (http://documents1.worldbank.org/curated/en/333671468330890417/ pdf/560530WP0Box341penditureFiscalSpace.pdf, accessed 2 February 2021). 138. Cashin C, Sparkes S, Bloom D. Earmarking for health: from theory to practice. Geneva: World Health Organization; 2018 (https://apps.who.int/iris/bitstream/handle/10665/255004/9789241512206-eng. pdf?sequence=1, accessed 2 February 2021). 139. Wilkinson M. Paying for public spending: is there a role for earmarked taxes? Fiscal Studies. 1994;15(4):119–35 (https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1475-5890.1994.tb00213.x, accessed 2 February 2021). 140. ITAF: impôts et taxes affectées à la protection sociale. Les impôts et taxes affectés (ITAF) [taxes and duties allocated to social protection]. Vie publique: au coeur du débat public [website]. Paris: Directorate of Legal and Administrative Information; 2018 (in French) (https://www.vie-publique.fr/fiches/22010- impots-et-taxes-affectees-itaf-protection-sociale, accessed 31 October 2020). 141. Seventeenth congress of the Philippines, Metro Manila, RA 11346, 23 July 2018 (https://www. officialgazette.gov.ph/downloads/2019/07jul/20190725-RA-11346-RRD.pdf, accessed 31 October 2020). 142. McCleary W. The earmarking of government revenue: a review of some World Bank experience. World Bank Res Obs. 1991;6(1):81–104 (http://documents1.worldbank.org/curated/en/109871468764672767/ pdf/multi-page.pdf, accessed 2 February 2021). 143. Strategic investment of tobacco tax revenue. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/global/pdfs/en/strategic_investment_tobacco_tax_revenue. pdf, accessed 31 October 2020). 144. Bialous SA, Glantz SA. Arizona’s tobacco control initiative illustrates the need for continuing oversight by tobacco control advocates. Tob Control. 1999;8:141–51 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/8/2/141.full.pdf, accessed 2 February 2021). 145. Goldman LK, Glantz SA. The passage and initial implementation of Oregon’s measure 44. Tob Control. 1999;8:311–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1763939/pdf/v008p00311.pdf, accessed 2 February 2021). 146. Heiser PF, Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health. 1997;87(6):968–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380932/pdf/amjph00505-0082. pdf, accessed 2 February 2021). 147. Landman A, Bialick P. Tobacco industry involvement in Colorado. Chicago: American Lung Cancer Association; 2004 (https://escholarship.org/uc/item/9d880448#main, accessed 18 February 2021). 148. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010. (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 149. Vardavas CI, Filippidis FT, Agaku IT, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the 252 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N importance of earmarking the revenue to health care and tobacco control. Tob Induc Dis. 2012;10:21 (http://www.tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the- revenue-to-health-care-and-tobacco,66110,0,2.html, accessed 31 October 2020). 150. Bird RM. Tobacco and alcohol excise taxes for improving public health and revenue outcomes: marrying sin and virtue?. Washington (DC): World Bank Group; 2015 (Policy Research Working Paper 7500; https://openknowledge.worldbank.org/bitstream/handle/10986/23464/ Tobacco0and0al0ing0sin0and0virtue00.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 151. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 152. The establishment and use of dedicated taxes for health. Manila: World Health Organization Western Pacific Regional Office; 2004 (https://www.who.int/health_financing/documents/dedicated_taxes. pdf?ua=1, accessed 2 February 2021). 153. Teja RS. The case for earmarked taxes: theory and example. Washington (DC): International Monetary Fund; 1988 (IMF Working Paper; https://www.elibrary.imf.org/doc/IMF001/14514- 9781451922677/14514-9781451922677/Other_formats/Source_PDF/14514-9781455212002.pdf, accessed 2 February 2021). 154. Keen M. Taxation and development – again. Washington (DC): International Monetary Fund; 2012(IMF Working Paper WP/12/220O; (https://www.imf.org/external/pubs/ft/wp/2012/wp12220.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 253 ANNEX 4.1 METHODS TO ASSESS THE NATURE AND SIZE OF THE ILLICIT TOBACCO TRADE A4.1 DIRECT MEASUREMENT A4.1.1 SMOKER INTERCEPT AND PACK OBSERVATION SURVEYS Illicit trade can be measured directly by examining the cigarette packs of smokers. The smokers themselves can provide information on purchasing patterns, brand preferences and prices paid. Researchers can select individuals or retailers to survey based on a convenience sample (i.e. a sample that may not be representative) or a probability-based sample (i.e. a sample selected to be statistically representative of an underlying population). Data collected from a pack could reveal whether the pack is compliant or non- compliant with the local tax laws. Information can be obtained from objective markings such as brands, public health warning labels, tax stamps, foreign language labels or duty-free labels. During these stops, researchers can record demographic information (e.g. age and gender of the smoker), smoking-related history (e.g. number of cigarettes smoked per day) and price information. This is helpful in understand- ing the profile of smokers who are able and willing to avoid cigarette taxes. Pack observations can be used in conjunction with population-based household surveys to obtain population-based estimates of the illicit tobacco trade (1,2). For example, as part of a regular national health survey, Kaplan et al. conducted a cross-sectional study of smokers in Turkey, using a face-to-face interviewer-administered survey and pack observation (3). They were able to collect sociodemographic, lifestyle and medical details along with pack observations as part of the study protocol. Advantages and disadvantages of smoker intercept and pack observation surveys A primary advantage of conducting pack observation is that it is direct and objective, and smokers are not subject to any value judgements (2). Paired with survey data, pack observation can appropriately account for respondents who are not residents of the area in which they are surveyed (4). Disadvantages include the difficulty of identifying areas that are representative of the tobacco use population and the difficulty of sampling important subpopulations such as elderly and immobile smok- ers. Also, surveys conducted in the daytime may discount the number of youthful smokers who are in school. Another disadvantage is that a sizeable number of smokers may refuse to show their last-purchased pack (2). Kaplan et al. found that 24% of smokers sampled in Turkey did not show their cigarette pack to the study interviewer (3). This issue may be mitigated by asking users to provide information on the brand purchased, whether any public health warnings were posted and the 254 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N price paid (5). Although the responses are based on recall, they may still yield useful information. For example, Joossens et al. allowed smokers who did not show their packs to provide self-reported information and found no statistical differences in illicit packs between these respondents and those who did show their packs (2). Another obstacle to accurate measurement based on smoker intercepts is the inability to determine the tax payment of smokers who purchase single cigarettes, since these smokers are generally not given packs. However, information can still be captured in self-report surveys by asking smokers to report the brand purchased and price paid. Key study for readers to refer to for additional guidance: Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–447. A4.1.2 PACK RETURN AND SWAP SURVEYS Pack return and pack swap surveys fall within the broader category of pack ob- servation studies that use survey sampling techniques to examine smokers’ pack characteristics and to determine whether they are tax compliant. For these surveys, the unit of analysis is the individual. The main differences between pack swap and pack return surveys is that swap surveys offer the smoker a replacement pack, whereas pack returns are built into mail surveys and allow respondents to mail in their unopened packs. Pack swap and pack return surveys use probability and nonprobability sampling procedures. Probability sampling allows researchers to generalize to the broader population. Governments can use this method to rapidly assess the availability of illicit products in a given geographic area or to measure the share of the illicit market- place. Rapid assessment may be performed in instances where there is an emerging tobacco product (e.g. a new cheap white brand) or suspected counterfeiting of tax stamp features. Rapid assessment using a convenience sampling strategy could place researchers near busy intersections where they could ask smokers for permission to look at their cigarette packs or to take photographs that could be analysed later. A population-based study requires a sample that closely mirrors the tobacco use population. Advantages and disadvantages of pack return and pack swap surveys Pack swap and pack return surveys may help to overcome the stigma associated with traditional smoking surveys. For example, when researchers ask smokers to see their cigarette packs (or when they take photographs), no value judgements are made. These surveys are good rapid-assessment tools that can be used to examine CHAP T ER 4. PO LI T I C AL ECO N OMY 255 the effectiveness of physical features of a pack designed to deter illicit trade (e.g. packs that have tracking and tracing technology or high-tech stamps). In addi- tion, they can be supplemented with population-based tobacco use surveys. When coupled with such survey data, these methods allow researchers to obtain relevant information about the context of illicit purchases, including, for example, sources (e.g. street, peer networks, retail stores) and prices. Mail-in surveys are filled out in the comfort of the respondent’s home without the presence of family members or passers-by, which may assure them that responses will be kept confidential. A potential disadvantage is that smokers who purchase both illicit cigarettes and tax-paid cigarettes may disproportionately mail back compliant packs. In addition, in LMICs, this mode of survey distribution may be unreliable because of issues associated with mail delivery systems. Key study for readers to refer to for additional guidance: Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61-66. A4.1.3 LITTERED-PACK SURVEYS Littered-pack surveys, also known as empty discarded pack surveys, are used pre- dominantly in high- and middle-income countries (e.g. the United States, France, Canada, New Zealand, Mexico and Poland). This unobtrusive method relies on the premise that smokers publicly discard packs (e.g. on streets, sidewalks and in public trash cans). The packs bear characteristics that illustrate whether they are tax compliant (e.g. tax stamps, health warnings). For example, an Albanian health warning label on a cigarette pack discarded in Greece provides evidence that the pack was destined for the Albanian market. The pack may have been smuggled into Greece by criminal entrepreneurs or it may have been brought by a visitor. Collecting discarded packs from a representative geographic sample and examining these characteristics can provide estimates of tax compliance. Operationally, this data collection method uses an ecological approach whereby geographies are the units of analysis. Geographical units are meant to represent the smokers in the city/ country and can be administratively defined (e.g. by the country’s census bureau or transit zones) or may reflect researcher-defined neighbourhoods (e.g. half-mile buffer zones near bus stops or activity spaces). Researchers in Canada have expanded the littered-pack method to include col- lection and analysis of cigarette butts on 25 postsecondary campuses. The cigarette butts provide information on the brands sold (or lack thereof) and allow researchers to distinguish between legal and illegal products (6). A recent innovative expansion 256 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the littered-pack methodology is the collection of packs from cigarette retailers. John and Ross collected empty packs of tobacco products from a sample of registered and unregistered retailers in India (7). Collecting packs from retailers was relevant given that single cigarettes dominate the illicit market in India. Smokers who buy single cigarettes would be unable to provide a pack in a pack swap or street intercept survey, so collecting littered packs from the ground would undercount sales of single cigarettes. The feasibility of this method is dependent on the relationship between researchers and retailers (enhanced trust) and the efforts taken to ensure confidentiality. In some countries, retailer compliance with this research method might be strained because of concerns regarding confidentiality, since retailers may face criminal and civil penalties, depending on the research findings. Advantages and disadvantages of littered-pack surveys Littered-pack surveys are generally advantageous for governments because they facilitate comparison with industry estimates. This is one of the most-preferred methods because it yields estimates that are less likely to be biased due to issues of social desirability, recall error and confidentiality that plague survey research, and they are much less expensive than face-to-face interviews used in smoker intercept or household surveys. However, there are some issues regarding these surveys, espe- cially in high-income countries, including the inability to differentiate between tax avoidance and tax evasion (8). For example, a pack in Berlin that bears a Vietnamese tax stamp may have been smuggled in mass quantity or brought in by a temporary visitor. Researchers have circumvented this issue and broadened the umbrella to measure cigarette tax noncompliance considering the potential biases introduced by tourism. Another disadvantage of littered-pack surveys is that larger budgets are needed to employ field researchers to collect, code and analyse the data. Not all countries employ tax stamps on their cigarette packaging, which may make it difficult to measure tax compliance. These surveys also can underestimate the markets in low-income countries such as India, where the main item of illegal trade is single cigarettes (7). The surveys may also overestimate illicit trade if littering behaviour is correlated with willingness to engage in illicit trade. Finally, littered-pack surveys and butt collections provide information on the proportion of butts and packs that are illegal, not the proportion of smokers that purchase illegal cigarettes (6). Key studies for readers to refer to for additional guidance: Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38–i43. Merriman D. The micro-geography of tax avoidance: evidence from littered cigarette CHAP T ER 4. PO LI T I C AL ECO N OMY 257 packs in Chicago. Am Econ J Econ Policy. 2010;2(2):61–84. Stoklosa M., Paraje G., Blecher E., A Toolkit on Measuring Illicit Trade in Tobacco Products. A Tobacconomics and American Cancer Society Toolkit. Chicago, IL:Tobacconomics, Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/621/uic-illicit-trade-tool-kit-eng-v2.0-2. pdf, accessed 18 February 2021). A4.1.4 SELF-REPORT POPULATION SURVEYS Self-report surveys, when distributed to a representative sample of the population, can provide meaningful data on the prevalence of tax noncompliance. The surveys can be distributed to individuals or households in various ways, including face-to-face, telephone, mail and internet. Questions that specifically address illicit purchases can be added as supplementary questions to existing health or tobacco surveys. Some countries include such questions in their adult and youth tobacco surveys to estimate tax evasion/avoidance. For example, Canada’s annual Youth Smoking Survey asks smokers about the frequency of their purchases of First Nations/Native brand cigarettes (9-10). Davis et al. used data from the New York Adult Tobacco Survey to measure the source of purchase of the last cigarette pack purchased (i.e. Native American Reservations, lower-tax neighbouring states or countries, toll-free telephone numbers, the internet, duty-free shops) and the price paid (11). Twenty- eight nations currently use surveys to measure tax noncompliance as part of the ITC Project (12). Similar analyses can be conducted using questions from the Global Adult Tobacco Use Surveys. For example, Iglesias et al. used the Brazil Global Adult Tobacco Use Surveys to compare self-reported prices with a defined threshold retail price to estimate the proportion of illicit cigarette use among smokers in Brazil (13). Countries are encouraged to use existing global health surveys or to incorporate similar types of questions pertaining to illicit trade in their annual health surveys. Asking respondents about price paid per pack (including taxes), brand name and location where cigarettes were purchased (e.g. duty free shop, unlicensed vendor, internet) can contribute to a better understanding of the illicit tobacco trade. Advantages and disadvantages of self-report surveys Self-report surveys can be repeated over time to measure purchasing trends and progress associated with increases in cigarette taxation. Well-designed surveys can also provide generalizable estimates at the national level. Depending on the size of the sample, a self-report survey can provide comparable data across geographies that can help governments target resources. For example, findings that illicit cigarettes are more common in urban areas could lead to additional education campaigns and targeted enforcement. 258 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Limitations of self-report surveys include the possibility of bias due to the social stigma associated with participating in the illicit trade, which could lead survey respondents to underreport participation. Additionally, surveys may be unable to gauge whether individuals are associated with tax avoidance versus tax evasion. Finally, there is evidence that self-report household surveys may underrepresent smokers. Key studies for readers to refer to for additional guidance: Callaghan RC, Veldhuizen S, Ip D. Contraband cigarette consumption among adolescent daily smokers in Ontario, Canada. Tob Control. 2011;20(2):173–174. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany (NY): New York State Department of Health, Tobacco Control Program; 2006. A4.1.5 COVERT-PURCHASES SURVEYS A number of studies in high-, middle- and low-income countries use covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes in public and semi-private spaces (14-17). This method is also used by the tobacco industry in the United States to identify retailers who sell counterfeit cigarettes (18). Covert-purchases surveys do not provide estimates of the size of the illicit trade (i.e. market volume). Instead, they serve as a surveillance tool to identify where illicit cigarettes are sold and the extent to which they have infiltrated legal businesses. For example, a covert-purchases survey can examine whether illicit cigarettes are sold through legal retailers. It can also be used to measure compliance with emerging tobacco control policies that focus on, for example, product standardization or new regulations on flavours (e.g. plain packaging or bans on flavoured tobacco products). Covert-purchases surveys use trained researchers to visit a selected sample of retailers and directly purchase or inquire about the availability of illicit tobacco products. Retailers are not informed about the goals of the studies. Methods for determining the availability of illicit product vary. For example, in some studies, covert buyers do not directly inquire about illicit products. Instead, they purchase packs of tobacco products, paying full price, to determine whether retailers are selling illicit products under the guise that they are licit (14). The research team then examines the packs to determine whether they are legal. In the United States, researchers have observed that some consumers are paying full price for illicit untaxed packs smuggled from lower-tax states (14). Other research protocols directly ask retailers for illicit products (14,16-17). In Guatemala, Arevalo et al. specifically asked retailers for “imported cigarettes” (17). The ways covert buyers ask for illicit products may also vary geographically. For example, in some countries covert buyers may ask for “cheaper” packs or for illicit whites such as Jin Ling. CHAP T ER 4. PO LI T I C AL ECO N OMY 259 Advantages and disadvantages of covert-purchases surveys Covert purchasing allows researchers to directly identify sources of illicit cigarettes. It also allows them to measure and test the dynamics between buyer and seller. For example, researchers can experiment to see if repeated attempts to purchase products increase the likelihood of purchase (known as the familiarity protocol) (16). One methodological challenge associated with covert purchases is that it is difficult to create a sampling frame for illicit sources because some may be unknown (e.g. pubs or homes). The traditional approach is to make purchases in legal outlets, which may bias estimates. Another issue with this method is that it is difficult for buyers (also called raters) to purchase products if they are unfamiliar with the seller or do not fit the typical demographics of purchasers. Therefore, researchers using covert- purchases surveys must have detailed knowledge of the marketplace, including the ways individuals specifically ask for illicit tobacco products, and they must know whether they mirror the demographics of the neighbourhood. For example, in a study of South Bronx smokers, von Lampe et al. found that smokers looked for certain clues to assess whether they were being sold illicit cigarettes (19). Overall, this method can be quite costly because it requires training researchers, travelling to retailers and purchasing product. Covert-purchases surveys do not enable researchers to estimate the level of illicit trade, but they can provide information on availability of supply. Key studies for readers to refer to for additional guidance: Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Global Health. 2016;12(1):78. A4.1.6 SEIZURES OF GOODS Seizures are the result of enforcement activity carried out by local, national and international organizations that confiscate tobacco products that are illegally manu- factured, transported and sold. Seizures are meant to reduce the profits associated with illicit trade by confiscating proceeds (e.g. cash, cars or houses) and the tools of the trade (e.g. print and tobacco machinery). Seizures can occur at various points in the supply chain. Seizure data are often tallied by the responsible agencies and used to measure program effectiveness or as justification for requesting additional resources (e.g. personnel). Some of the data may be supplied to international customs organiza- tions, including the WCO (20). The quality of recordkeeping varies. For example, some agencies may maintain criminal files in databases that detail dates of seizure, brand names and laboratory testing. 260 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Seizures provide preliminary data on the scope of criminal activity and can help identify key trends to guide law enforcement agencies’ efforts. For example, seizures can identify trends on the modus operandi of smugglers, including transporta- tion methods (e.g. sea cargo versus trucks), point of entry and brand preference. Seizure statistics can also be used as a preliminary test to measure the efficacy of interventions. For example, Stoklosa and Ross used seizure data from the Canadian province of Nova Scotia to test the impact of a 2015 menthol ban. He found no statistically significant change in the number of menthol cigarettes seized before and after the ban (1). Advantages and disadvantages of seizures of goods Generally, seizure statistics can be readily obtained from law enforcement agencies through formal requests to agency gatekeepers (e.g. public information officers). Seizure data, however, generally do not provide a representative picture of illicit activity. For example, certain geographies may yield higher seizures because that is where the bulk of operations are being conducted. Police agencies may focus on certain geographies (e.g. locations near borders) rather than randomly inspecting, and their findings may be limited to those specific regions. Seizure data may also be skewed by the type of investigation procedures utilized. Large seizures may be the result of long-term investigations (i.e. wiretaps or culling confidential infor- mants), while smaller seizures may come from anti-smuggling cases that involve cross-border purchases of low quantities of cigarettes (less than 1 000) (2). Seizures can also be skewed by industry cooperation with law enforcement agencies. For example, the tobacco industry may be more likely to support law enforcement on counterfeit seizures rather than smuggling cases because counterfeiting impacts their brand integrity. A4.2 RESIDUAL METHODS Because the illicit tobacco trade is often decentralized, it can be difficult to observe directly. However, researchers are sometimes able to make inferences about its size without direct observation by comparing observed tobacco tax revenues with the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. The difference between observed and hypothetical revenues is called the residual and can be used as an indicator of the magnitude of illicit trade. Even when the residual is only an approximate measure, changes in its size may be a reliable indicator of changes in the size of the trade. When actual tobacco tax revenues are reliably observed, the main challenge for residual methods is that of producing accurate estimates of the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. CHAP T ER 4. PO LI T I C AL ECO N OMY 261 A4.2.1 GAP ANALYSIS Gap analysis is the preferred residual methodology because it is intuitive, straight- forward and relatively easy to explain to policy-makers and the general public, and it has been widely employed in government studies (4). Researchers using gap analysis compare survey-based self-reported consumption data with observed (usually administrative) data on tax-paid sales. The basic premise is that if both self-reports and observed data are accurate, any difference between reported consumption and tax-paid sales can be explained by legal imports of non-taxed cigarettes (such as duty-free sales), exports of taxed cigarettes, tax evasion or tax avoidance. The greatest research challenge in implementing gap analysis – as with most residual methods – is obtaining reliable and accurate estimates of tobacco consump- tion. In its simplest implementation, gap analysis calculates the residual as the difference (which should be a minimum of zero) between the amount of tobacco consumption reported in surveys and tax-paid sales, which are generally available from administrative sources, minus exports. This simple calculation, however, is generally flawed, since surveys of reported tobacco consumption underestimate true consumption. Underreporting of tobacco consumption may result from survey respondents’ reticence about disclosing behaviour that is viewed as unhealthy and potentially socially undesirable. Depending on the legal and cultural context, cer- tain groups (e.g. women or youth) may be more likely than others to underreport consumption. Other groups (e.g. rebellious young men and women) may accurately report or even overestimate consumption. Reuter and Majmundar measured actual consumption by total national taxed sales in the United States, where both legal untaxed imports and exports of taxed tobacco are widely believed to be very small, and found that the ratio of self-reported consumption to actual consumption was only 65% (4). After incorporating this survey underreporting into their analyses and considering the evidence from their gap analysis and the literature, Reuter and Majmundar found that the illicit market in the United States, which largely consists of avoidance or evasion of subnational state taxes, is between 8.5% and 21% of consumption (4). The higher range of the estimate is consistent with prior estimates using population-based pack observation studies (21). Researchers using gap analysis for countries or regions where legal imports of untaxed tobacco (such as duty-free products) or (legal or illegal) exports of taxed cigarettes are more significant should attempt to incorporate data about, or estimates of, these factors into their calculations. Legal untaxed imports of tobacco should be subtracted from reported consumption (after adjustment for underreporting), and exports of taxed tobacco should be subtracted from taxed sales. Obtaining data about legal untaxed imports and exports of taxed tobacco may be challenging, because these imports and exports may be the result of decentralized decisions of individual 262 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N travellers as they cross tax borders. Data on these activities will not necessarily be collected through normal administrative activities. Despite these challenges, gap analyses may still prove useful. For example, if researchers have reason to believe that misreporting of tobacco consumption and the amounts of legal untaxed imports and taxed exports are relatively stable or follow known trends (e.g. are falling) over time, gap analyses can be used to provide estimates or lower (or upper) bounds on illicit trade when several years of data on taxed sales and reported consumption are available. Data sources may be country tax administrators who have access to sales data and health departments that have access to population-level studies of reported tobacco con- sumption. In this context, multiple years of data on reported consumption and tax-paid sales can allow researchers to estimate changes in the size of the illicit trade even when it is difficult to measure the absolute level. Paraje used the 2008 Global Adult Tobacco Survey and the 2013 National Health Survey to measure reported tobacco consumption in Brazil (22-23). Advantages and disadvantages of gap analysis A major advantage of gap analysis is that when quality data are available, it is simple, easily reproduced and explainable to policy-makers and the general public. How- ever, high-quality data on reported consumption may not be available, especially in low-income countries. In many cases, gap analysis does not provide reliable information on the size of the illicit market but only on changes in the size over time (22). Additionally, some low-income countries may not have reliable estimates of tax-paid cigarette sales, and secondary data repositories of cigarette sales may not be transparent about their methodology (24). Another disadvantage of gap analysis is that it generally cannot be used to obtain separate estimates of tax avoidance and tax evasion. Biased estimates may also result if surveys of tobacco consumption are not representative of the population (25). Moreover, it is generally not possible to quantify the precision of the estimates or uncertainty associated with the estimates, because of both statistical uncertainty resulting from the use of samples to imperfectly represent populations (e.g. the share of the population that smokes) and uncertainty about key facts such as the degree to which survey respondents understate their tobacco consumption. Key studies for readers to refer to for additional guidance: Szklo A, Iglesias RM, Carvalho de Souza M, Szklo M, Maria de Almeida L. Trends in illicit cigarette use in Brazil estimated from legal sales, 2012–2016. Am J Public Health. 2018;108(2):265–269. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine and Tob Res. 2019;21(8):1079–86 CHAP T ER 4. PO LI T I C AL ECO N OMY 263 A4.2.2 ECONOMETRIC MODELLING There is a long tradition of using data to estimate parameters of demand functions that relate the quantity of goods consumed to the prices faced by consumers, their incomes and other variables. Because of the addictive nature of tobacco – and because of important public health and public policy concerns relating to tobacco use – economists have paid particular attention to the estimation of cigarette demand functions (26). As the literature on this topic developed, it became apparent that taxed tobacco sales would be a biased indicator of tobacco consumption if some consumers obtained their tobacco in illicit markets. Similarly, the price of cigarettes in the legal market might overestimate the price paid by consumers if some sales were not tax-paid. While economists generally cannot observe sales in the illicit market, they have been able to develop models that predict conditions under which consumers avoid tobacco taxes. They reason that the relative size of illicit tobacco markets depends primarily on two variables: the relative price of taxed and untaxed consumption and the ease of obtaining lower-cost (untaxed) tobacco. Other variables, includ- ing the social stigma from evading tax laws and the perceived relative quality of illicit tobacco, could also influence the demand for it. While illicit trade cannot be directly observed, it can be estimated from the difference between tax-paid sales and predicted consumption. Tax-paid sales can be less than predicted consump- tion when retailers or consumers evade taxes. They can be greater if some tax-paid cigarettes are bought within the jurisdiction and then consumed in areas where after-tax prices are higher. Econometric modelling estimates of illicit trade must be tailored to the situation in the country that is being studied, and therefore the data requirements may differ substantially from case to case. Researchers using this method should be familiar with the literature and should also understand the conditions in the areas they are researching. They must always include some measure of tobacco consumption or sales and some measure of the price of tobacco in the home country, as well as other variables (e.g. income) that are known to affect the demand for tobacco. It is also generally necessary to include variables that measure the availability and relative price of illicit tobacco, which can often be measured by comparing tobacco taxes in the home country with those in areas that are the source of illicit tobacco. Advantages and disadvantages of econometric modelling The major advantage of econometric modelling is that it is consistent with a long tradition of economic theory and practice, and the quality of the modelling techniques and empirical estimates can therefore be evaluated against widely accepted criteria. Empirical analyses provide estimates of price elasticities, income elasticities and price elasticities of tax avoidance. A substantial literature base makes it possible to 264 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N rigorously quantify uncertainty about the estimates and to test their robustness to various assumptions made in the modelling process. Estimates can be compared to other estimates available in the literature, and the results of these analyses can be used to simulate the impacts of policy changes (including tax and/or enforcement increases) on both consumption and tax avoidance. Because this methodology provides direct estimates of the uncertainty of the results, researchers can specify their level of confidence in the findings. A shortcoming of econometric modelling is that it requires high-quality data on a variety of important variables over a period of time, as well as advanced econo- metric modelling expertise. Also, because results from the econometric models are based on statistical inference and economic theory rather than direct observation (e.g. the proportion of packs without tax stamps), it can be difficult to explain to policy-makers and the general public. Key studies for readers to refer to for additional guidance: Becker GS, Grossman M, Murphy KM, (1994). An empirical analysis of cigarette addiction. Amer Econ Review. 1994;84(3):396–418. Schafferer C, Yeh CY, Chen SH, Lee JM, Hsieh CJ. A simulation impact evaluation of a cigarette excise tax increase on licit and illicit cigarette consumption and tax revenue in 36 European countries. Public Health. 2018;162:48–57. A4.2.3 EXPERT OPINION (KEY-INFORMANT SURVEYS AND INTERVIEWS) Insight on illicit trade dynamics can come from experts in the field, including researchers (e.g. in economics, criminal justice and public health), taxation depart- ments, enforcement agencies, product manufacturers, wholesalers and retailers. Other key informants include journalists and academics who have secured confi- dential informants. Experts can provide novel information about emerging trends (e.g. new smuggling routes). In some cases, researchers can obtain interviews with incarcerated or active offenders (27-29). For example, researchers studying cigarette smuggling in eastern Africa conducted interviews with more than 150 Ugandan tobacco smugglers (29). Experts can be queried through surveys or semi-structured interviews. When sampling frames are available (e.g. directories of tax department employees), surveys are more expedient than interviews. However, when experts are hard to find, non- random sampling strategies coupled with interviews are recommended. Identifying experts may require recruiting a gatekeeper who is tasked with helping researchers find additional experts; or purposive sampling, where individuals are identified based on set criteria (e.g. they are taxation experts employed by local governments) (30). CHAP T ER 4. PO LI T I C AL ECO N OMY 265 Advantages and disadvantages of key-informant surveys and interviews Informant interviews can be a useful starting point for identifying trends in the marketplace (e.g. venues where illicit cigarettes are sold or modes of entry). One of the disadvantages of relying on informants is that the information solicited from them may not be generalizable. Expert knowledge may be outdated or limited by the informants’ experience. Furthermore, the opinions of experts are subjective and may be biased by the experts’ employment status and the sampling methods used. For example, persons working in law enforcement may overestimate the extent of bootlegging in order to secure additional funding for future operations. Similarly, manufacturers looking to defeat taxes may overestimate the illicit trade to illustrate the links between taxation and illicit behaviour. Alternatively, tobacco control advocates may underestimate illegal market measures in order to support the argument that taxes do not increase illicit trade. Key studies for readers to refer to for additional guidance: Joossens L, Raw M. Cigarette smuggling in Europe: who really benefits? Tob Control. 1998;7:66–71. doi:10.1136/tc.7.1.66 PMID: 9706757. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20(3):226–232. A4.3 MIXED AND MULTIMETHOD STUDIES Given the shortcomings of the aforementioned methods for assessing the nature and size of the illicit tobacco trade, governments may want to validate their findings by using mixed or multiple methodologies. Mixed methods use two methodological paradigms, qualitative and quantitative, as tools for exploration and explanation. For example, mixed method studies can use littered-pack surveys to measure the size of the market along with self-report surveys of smokers to understand patterns of purchasing, including sources, frequency and social norms. For example, Stoklosa and Ross estimated the share of the illicit market in Poland using a population- based self-report survey and a littered-pack survey (1). Using both types of survey simultaneously enables governments to assess their validity in estimating the size of the illicit market. Alternatively, governments can employ multimethod research, i.e. the use of multiple methods that are similar in tradition (e.g. focus groups and semi-structured interviews) (31). Saenz de Miera et al. used face-to-face interviews (households), litter collection and observation of single-stick sellers, which enabled them not only to cross-validate the two major methodologies, but also to see if the brand of the single stick was a good measure of licit versus illicit trade (33). 266 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Advantages and disadvantages of mixed and multimethod studies Mixed and multimethods studies enable researchers to check the validity of their findings. Multiple methods are preferred in contexts where illicit trade estimates are politicized. For example, low estimates may be challenged by the tobacco indus- try, while high estimates may be challenged by tobacco control researchers and/or proponents. Mixed and multiple methods (e.g. littered-pack surveys and informant interviews) can enable governments to understand the situational context in which the illicit trade operates, including the actors involved and venues of sale. One disadvantages of using mixed and multiple methods is cost. Governments that are constrained by tight budgets may choose to use a single method that provides the most accurate information. However, given the issues faced by each method, this may not be feasible – each method has limitations. Instead, governments can pair a high-cost method with a lower-cost method (e.g. pairing interviews with empty pack surveys, or law enforcement seizure data with face-to-face consumer surveys). Key study for readers to refer to for additional guidance: Zaloshnja E, Ross H, Levy DT. The impact of tobacco control policies in Albania. Tob Control. 2010;19:463–468. CHAP T ER 4. PO LI T I C AL ECO N OMY 267 REFERENCES 1. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 2. Joossens L, Lugo A, La Vecchia C, Gilmore AB, Clancy L, Gallus S. Illicit cigarettes and hand-rolled tobacco in 18 European countries: a cross-sectional survey. Tob Control. 2014;23:e17–e23 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC3812425/pdf/nihms491463.pdf, accessed 2 February 2021). 3. Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–7. 4. Reuter P, Majmundar M. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Washington (DC): National Academies Press; 2015. 5. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 6. Barkans M, Lawrance KA. Contraband tobacco on post-secondary campuses in Ontario, Canada: analysis of discarded cigarette butts. BMC Pub Health. 2013;13(1):335 (https://bmcpublichealth. biomedcentral.com/articles/10.1186/1471-2458-13-335, accessed 2 February 2021). 7. John RM, Ross H. Illicit cigarette sales in Indian cities: findings from a retail survey. Tob Control. 2018;27(6), 684–688. 8. Davis KC, Grimshaw V, Merriman D, Farrelly MC, Chernick H, Coady MH, et al. Cigarette trafficking in five northeastern US cities. Tob Control. 2014;23(e1):e62–e68. 9. Callaghan RC, Veldhuizen S, Leatherdale S, Murnaghan D, Manske S. Use of contraband cigarettes among adolescent daily smokers in Canada. CMAJ. 2009;181(6-7):384–6 (https://www.cmaj.ca/content/ cmaj/181/6-7/384.full.pdf, accessed 2 February 2021). 10. Guindon GE, Burkhalter R, Brown KS. Levels and trends in cigarette contraband in Canada. Tob Control. 2017;26(5):518–25 (https://www.researchgate.net/publication/307890937_Levels_and_trends_in_ cigarette_contraband_in_Canada, accessed 2 February 2021). 11. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany: New York State Department of Health; 2006 (https://www. health.ny.gov/prevention/tobacco_control/docs/cigarette_purchasing_patterns.pdf, accessed 2 February 2021). 12. International Tobacco Control Evaluation Project (2018). Surveys. Waterloo: University of Waterloo; 2018 (http://www.itcproject.org/surveys, accessed 12 October 2020). 13. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 14. Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. 15. Brown J, Welding K, Cohen JE, Cherukupalli R, Washington C, Ferguson J, et al. An analysis of purchase price of legal and illicit cigarettes in urban retail environments in 14 low-and middle-income countries. Addiction. 2017;112:1854–60 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5600117/pdf/ADD- 112-1854.pdf, accessed 2 February 2021). 16. Scollo M, Bayly M, Wakefield M. Availability of illicit tobacco in small retail outlets before and after the implementation of Australian plain packaging legislation. Tob Control. 2015;24(e1):e45–e51. 17. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Globalization and Health. 2016;12(1):78 (https://globalizationandhealth. biomedcentral.com/articles/10.1186/s12992-016-0219-z, accessed 2 February 2021). 18. Phillip Morris USA Inc. v. Shalabi. United States, District Court, C.D. California; 2004 (https://www. casemine.com/judgement/us/5914b6cbadd7b0493477b3da, accessed 2 February 2021). 19. von Lampe K, Kurti M, Johnson J, Rengifo AF. ‘I wouldn’t take my chances on the street’ navigating illegal cigarette purchases in the South Bronx. J Res Crime Delinq. 2016;53(5):654–80 (https://www. 268 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N researchgate.net/publication/298515449_I_Wouldn’t_Take_My_Chances_on_the_Street_Navigating_ Illegal_Cigarette_Purchases_in_the_South_Bronx, accessed 2 February 2021). 20. Customs and Tobacco Report 2009. Brussels: World Customs Organization; 2009. (http://www.wcoomd. org/en/media/newsroom/2010/june/~/media/83967DFEB9F74D388924A4C61F279DC4.ashx, accessed 12 October 2020). 21. Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ, et al. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61–i66 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3984758/pdf/nihms567943.pdf, accessed 2 February 2021). 22. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine Tob Res. 2019;21(8):1079–86. 23. National Survey of Health. Instituto Brasileiro de Geografia e Estatística; 2013 (in Portuguese) (https:// www.ibge.gov.br/en/statistics/social/health/16840-national-survey-of-health.html?=&t=downloads, accessed 27 November 2020). 24. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 25. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 26. Chaloupka FJ, Warner KE. The economics of smoking. In: Arrow KJ and Intriligator MD, editors. Handbook of Health Economics. Amsterdam: Elsiver; 2000. pp1539–1627. 27. Antonopoulos GA. Cigarette smugglers: a note on four ‘unusual suspects’. Glob. Crime. 2007;8(4):393–8 (https://www.researchgate.net/publication/248955367_Cigarette_Smugglers_A_Note_on_ Four_’Unusual_Suspects’, accessed 2 February 2021). 28. Antonopoulos GA. The Greek connection(s): the social organization of the cigarette-smuggling business in Greece. Eur J Criminol. 2008;5(3):263–88 (https://www.researchgate.net/publication/249752218_ The_Greek_ConnectionsThe_Social_Organization_of_the_Cigarette-Smuggling_Business_in_Greece, accessed 2 February 2021). 29. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20:226–32 (https://www.researchgate.net/publication/49809289_ Blood_cigarettes_Cigarette_smuggling_and_war_economies_in_central_and_eastern_Africa, accessed 2 February 2021). 30. Babbie, ER. The basics of social research. Boston: Cengage Learning; 2013. 31. Tashakkori A, Teddlie C., editors. Sage handbook of mixed methods in social & behavioral research. New York: Sage; 2010. 32. Saenz de Miera Juarez B, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 269 ANNEX 4.2 HOW ARE THE TOBACCO TAX REVENUES EARMARKED? The introduction of earmarking of tobacco tax revenue is almost always combined with an increase in excise taxes (or a new surcharge) rather than reallocation of existing revenues (1). Table A4.1 provides examples of the different approaches used by several countries to earmark tobacco tax revenues. Table A4.1 Approaches used to earmark tobacco tax revenues FUNDING SOURCES/ TYPE OF TAX EXAMPLES OF TAX BASE AND RATES As part of the excise system (tobacco, alcohol) Specific Republic of Korea: 841 won (US$ 0.75) per pack or 29% of the specific excise rate Costa Rica: 467.8 cólones (US$ 0.83) per pack or 100% of the specific excise rate Congo: 20 CFA francs (US$ 0.036) per pack or 50% of the specific excise rate Ad valorem Colombia: 10% of retail price (equivalent to 100% of the ad valorem rate) New levy (surcharge on the existing excise or completely new levy) Specific Egypt: additional 0.75 Egyptian pounds (US$ 0.042) per pack Ad valorem Thailand: surcharge of 2% over the excise tax base Botswana: new tobacco levy of 30% of the cost of production or CIF Percentage of excise revenue Cook Islands: 50% of revenues from the excise tax on tobacco Guatemala: 100% of revenues from the excise tax on tobacco Note: Conversions of amounts from the local currency were made using the official exchange rates from the IMF as of 31 July 2018 (date of the data collection). Source: (2). REFERENCES 1. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 2. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 270 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 271 CHAPTER 5. Best practices in tobacco tax policy and administration TAX POLICY Use excise tax increases to achieve the public health goal of reducing the death and diseases caused by tobacco use Extensive research has clearly demonstrated the effectiveness of higher tobacco product taxes and prices in reducing tobacco use and its harmful consequences, particularly among the poor and the young. In fact, tobacco excise tax increases are the single most effective and cost-effective policy for reducing tobacco use. Excise taxes are the most significant taxes applied on tobacco products because of their ability to raise both absolute and relative prices. Tobacco excise tax increases also generate sizeable new revenues that will be sustained in the short to medium term. In the long term, continued increases in tobacco taxes – coupled with implementa- tion of other evidence-based tobacco control policies and programmes – will lead to even larger reductions in tobacco use and its consequences. Include significant tobacco excise tax increases as part of a comprehensive strategy to reduce tobacco use Governments should adopt a comprehensive tobacco control strategy that includes objectives for reducing adult tobacco use and preventing youth tobacco use. Rais- ing excise taxes significantly is the most effective, as well as the most cost-effective, measure for reducing consumption. When combined with other demand reduction interventions, the impact of tax increases on tobacco use is even stronger. Such interventions include comprehensive smoke-free policies in all public spaces, total bans on tobacco advertising, promotion and sponsorship by tobacco companies, large graphic health warnings about the consequences of tobacco use, plain packaging, broad efforts to help current users quit and mass media public education campaigns. Implementation of a comprehensive strategy to reduce tobacco use leads to greater reductions in the harmful consequences of tobacco use, builds public and political support for higher taxes and maximizes the effectiveness of tax increases in achieving public health objectives. 272 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Involve the competent authority from the start when considering the revision of a tax policy Competent authorities such as tax administrations and customs authorities are key partners in the effective implementation of a tax policy. Policy-makers need to ensure that those competent authorities are consulted and involved in the tax policy revision process so that their concerns about the impacts of policy change on enforcement can be taken into account from the beginning. This can also help identify and address possible loopholes early on in the enforcement process. Coordination among relevant bodies, including close cooperation and sharing of information, will optimize enforcement of tax policy and tax collection. To streamline the process of cooperation and exchanges of information, a basis in law needs to be established. Additionally, the involvement of tax administration authorities in the entirety of the tax revision policy process is important to ensure effective implementation of the policy. Promote greater policy coherence across sectors such as agriculture, industry, trade, finance and labour Greater multisectoral integration and policy coherence is needed at the country level to achieve effective health improvements. In particular, it is important to ensure that public policies and interventions in non-health sectors (e.g. agriculture, industry, trade, finance and labour) do not act against the intended public health impact of tobacco control and taxation (such interventions include providing subsidies to tobacco growing or manufacturing). TAX DESIGN Tax structure matters and simpler is better Complex tax structures are difficult to administer, create opportunities for tax avoid- ance and evasion and are less effective than simpler structures in achieving public health and revenue goals. Simplifying the structure of tobacco excise taxes will facilitate tax administration, reduce tax avoidance and evasion, enhance revenues and have a greater impact on tobacco use by reducing incentives to substitute among tobacco products or brands in response to tax increases. Countries with multiple tiers of tobacco tax rates based on product characteristics (e.g. price level, length, weight, type of tobacco) should reduce and eventually eliminate these differential tax rates. An appropriate transition strategy is to reduce the variations in tax rates over time with the aim of implementing a uniform tax (i.e. a single rate applies whether excise is ad valorem or specific) on a given tobacco product. Applying a uniform tax to all brands of a given tobacco product also sends a clear message that they are equally harmful. CHAP T ER 5. B E S T PR AC T I CE S 273 Rely more on specific tobacco excises to drive price increases Greater reliance on specific excise taxes maximizes the impact of tobacco taxes on public health by reducing the gap in prices between premium and low-priced alternatives and limiting opportunities for users to switch down in response to tax increases. For countries that currently rely on an ad valorem tax, an appropriate first step would be to shift to a mixed system by adding a sizeable specific component or introducing a high minimum specific excise tax (an excise tax floor). For countries that rely on a mix of ad valorem and specific taxes, the specific tax component should be increased regularly so that it accounts for a greater share of the total excise tax. Increase tobacco taxes significantly to reduce the affordability of tobacco products To maximize the public health impact of higher tobacco taxes while at the same time generating higher revenues, governments should significantly raise taxes to increase prices and reduce the affordability of tobacco products. In many LMICs, tobacco use increases with incomes, and since incomes rise faster than tobacco product prices, these products are becoming more affordable. To reduce afford- ability, tax increases need to result in real price increases that are higher than the increases in real incomes. Where revenue increases are a goal, rely on regular excise tax increases If governments want to increase tobacco revenues, they must increase excise taxes regularly. From the tax revenue perspective, the important determinant is the tax base elasticity, which has three key components: the price elasticity of demand of tobacco, the share of the tax in the retail price and the degree of pass-through of the excise tax rate increase on to retail price. Tax increases will increase revenues at least in the short to medium term, because demand is price inelastic, tax levels are generally low as a proportion of retail prices and the pass-through of tax increases on to retail prices is unlikely to be higher than the tax increase itself (i.e. there is no overshifting). In addition, increasing tax rates is the only policy measure that can reverse reduced revenues in a declining market that has strong tobacco control policies. Automatically adjust specific tobacco taxes for inflation and income growth Unless specific tobacco taxes are regularly adjusted, their real value will fall over time as general price levels increase. When this happens, their effectiveness in reducing tobacco use will be diminished. Governments should establish a mechanism for automatically adjusting specific taxes to keep pace with inflation. Recently, some governments have begun to extend this indexation to include income growth as well, further ensuring that tobacco does not become more affordable over time. 274 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Pricing regulations cannot be considered an alternative to excise tax. However, in some specific contexts, pricing regulations could be used in conjunction with excise taxes to help ensure the effective implementation of tax increases In certain contexts where increasing taxes is challenging or the tax structure is weak, non-tax policies such as pricing regulation (specifically, minimum mark-ups and price floors/minimum prices) may be seen as a second-best alternative to ensure a high price level and dissuade consumption of tobacco products. These policies, however, do not necessarily lead to the desired price level, nor do they protect consumers and government from industry manipulation. However, in the context of powerful multinationals that sell brands across all market segments and could easily undershift a tax increase to cheaper brands – or where price promotions cannot be banned – minimum price policies may help increase the effectiveness of tax increases, especially if the minimum prices are increased regularly. Implement nontax policies affecting price levels, such as banning promotional discounts for tobacco products and the sale of single sticks of cigarettes The banning of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Do not allow concerns about the inflationary impact of higher tobacco taxes to deter tax increases Given that wages or some government spending may be tied to a price index, govern- ments can reduce concerns about the inflationary impact of a tobacco tax increase by using a price index that excludes tobacco products. TAX PARITY Tax all tobacco products in a comparable way Increasing excise taxes on some tobacco products but not on others results in changes in the relative prices of these other products. This induces substitution towards relatively less-expensive products – for example, from expensive manufactured cigarettes to other, cheaper tobacco products such as RYO tobacco, bidis, cheroots or chewing tobacco. As a result, the overall reduction in tobacco use is smaller than it would have been had all taxes increased by comparable amounts. Comparable increases in the taxes on all tobacco products maximize the public health impact of tobacco tax increases by minimizing opportunities for substitution. Moreover, increases in taxes on all tobacco products will generate larger increases in revenues. CHAP T ER 5. B E S T PR AC T I CE S 275 Strictly regulate new and emerging tobacco and nicotine products where they are not banned and impose an excise tax In recent years, the world has been experiencing the rise of new and emerging tobacco and nicotine products including ENDS, ENNDS and HTPs. The tobacco industry claims these new products are safer than traditional tobacco products, but the evidence so far suggests that they could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of newer products – as well as initiation, smoking cessation and switching behaviour among different socioeconomic groups – are not yet clear. Best practices for taxing new and emerging tobacco and nicotine products, based on current knowledge, are that: 1. HTPs should be taxed at the same level as cigarettes and, in terms of structure, through a specific excise per unit regardless of tobacco content. HTPs contain tobacco and should be treated as a tobacco product. 2. ENDS/ENNDS products should be taxed in a manner that discourages up- take by youth and non-users. Nicotine- and non-nicotine-delivery systems containing e-liquids should be taxed equally. 3. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While these newer products create additional challenges for tobacco control, it is important to remember that cigarettes remain by far the predominant tobacco product and that raising taxes and prices on cigarettes – and thereby reducing their use – should remain the top priority. MONITORING AND EVALUATION Know your market Know your market well. The type of tax structure you choose and the impacts it will have on consumption and tax revenue are shaped by the particular dynamics of your market. Understanding the nature and degree of competition in your market is vital to selecting the appropriate type of tax structure and policies to achieve your public health and revenue objectives. This knowledge will also facilitate more accurate estimates of the impacts of a tax increase, as well as better anticipation of industry responses. Assess the impact of your policies to design and implement the most effective tobacco excise tax policies Monitoring and evaluation are essential for effective tobacco taxation, and they 276 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N should be built into the initial design – or redesign – of tobacco tax policies. A number of tools exist to help policy-makers pre-emptively assess the effects of a proposed tobacco control policy on consumption, smoking prevalence and lives saved. The WHO TaXSiM uses target simulations to assist governments in predict- ing how specific tax changes will impact consumer prices, consumption and tax revenue in their market. Adopt indicators that help you measure improvements in tax policy and its impact Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the improvement of their policies and determine if those policies have an impact on tobacco use over time. The tax share of the retail price for a particular tobacco product is a key indicator that should be used in conjunction with an af- fordability indicator. A recommended target for countries to aspire to is to have an excise tax that represents at least 70% of the retail price of tobacco products. Another useful indicator to assess the performance of the tax policy overall is the use of a tax scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). TAX ADMINISTRATION Implement best practice approaches in general tax administration to make tobacco tax administration more effective and efficient Best practice approaches include (1) defining clearly the roles and responsibilities of competent authorities, (2) ensuring effective coordination among relevant bodies at the national and international levels and (3) undertaking evaluation of performance and accountability against pre-agreed indicators to identify points for improvement. Ensure compliance and accuracy of information on the tax compliance cycle To achieve this, implement the following actions: • Require licences for manufacturing, importing, exporting, retailing, growing, transporting, wholesaling, brokering, warehousing and distributing tobacco products. This will help secure the supply chain while obtaining valuable infor- mation, e.g. through access to companies’ accounting and inventory systems. • Make sure all persons and entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment keep complete and accu- rate records of all relevant transactions and details of materials used in the production of tobacco products. CHAP T ER 5. B E S T PR AC T I CE S 277 • Ensure that tax declarations collect as much information as possible on the taxpayer. • Collect taxes close to the point of production and import to limit the number of taxpayers a competent authority needs to manage. • Maintain a system of authorization for warehousing to carry out controls in production and storage facilities to ensure that taxes are paid. • Use electronic methods, through the best available IT, for declarations and collection of taxes. This allows for cross-check of information provided in dec- larations with information from other government agencies and third parties. Ensure control and enforcement on the supply chain To achieve this, implement the following actions: • Include control and enforcement as a fundamental pillar in the strategic plan of the tax administration overall. • Use a risk-based approach by choosing defined targets for enforcement and control, such as those who have a higher probability of noncompliance. • In the licensing process, ensure that purchases from unlicensed suppliers or sales to unlicensed purchasers are not allowed. Ensure also that the validity of licences is limited in time and require renewals or reapplication to maintain a high level of control. • Use tax stamps with strong security features to reduce the risk of stamp counterfeiting. These markings facilitate the collection of excise taxes, audits and enforcement actions. • Implement a tracking and tracing system for tobacco products. A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, and monitoring and control- ling the movement of tobacco products and their legal status. • Implement anti-forestalling measures so that forestalling does not delay a tax increase and its intended effect on revenues and consumer behaviour. • Control import and export of tobacco products and manufacturing equipment by allowing only duly licensed natural persons or legal entities to conduct such activities. • Strengthen border control, e.g. by utilizing non-invasive tools such as X-ray scanners and dogs to detect tobacco products. • Limit or tightly control and, ideally, ban activities related to production and trade of tobacco products in tax-free zones to avoid opportunities for tax evasion. • Prohibit intermingling of tobacco products with non-tobacco products in a single container or any other similar transportation unit when removed from tax-free zones. 278 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Prohibit the sale to or import by international travellers of tax-free or duty- free tobacco products. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and adversely affect government revenues by creating a loophole in the tax structure. Clearly define procedures to follow after detecting illicit trade of tobacco • Take immediate action to seize and destroy smuggled and/or illicit tobacco and collect due taxes. • Ensure certain, swift and severe sanctions for those caught engaging in illicit trade in tobacco products, such as penalties, fines and withdrawal of licences. It can also be effective to consider illicit trade in tobacco products by law as a source of money-laundering. Become a Party to and/or implement the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products The WHO FCTC Protocol provides a blueprint of best practices and policies for dealing with illicit trade and should be part of any strategy to fight it. Implement, to the extent possible, the same rules and regulations for tax administration and enforcement for all tobacco products, as well as new and emerging nicotine and tobacco products Implement broad policies for ensuring a good tax system that will trickle down to good tax administration of tobacco products by: • ensuring proper resourcing of competent authorities; • having strict rules and regulations to detect corruption and to punish both personnel and taxpayers who are engaged in corrupt practices; and • ensuring a strong judicial system that is independent in fact and in perception, where disputes are solved quickly. The appeal process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially for illicit trade. POLITICAL ECONOMY Beyond the technical soundness of best practices in tax policy and administration, a critical factor in advancing tobacco taxes is the ability to get the political buy-in of the highest instances in the government. One key strategy is to address concerns around the political economy of tobacco taxation, which are often exploited by the tobacco industry to block major reforms. CHAP T ER 5. B E S T PR AC T I CE S 279 SCARE tactics The tobacco industry uses SCARE tactics to dissuade governments from implement- ing tobacco tax increases. These include smuggling and illicit trade (S), court and legal challenges (C), anti-poor rhetoric (A), revenue reduction (R) and employment impact (E). Best practices for countering these tactics are described below. S: Smuggling and illicit trade Do not allow concerns over the impact of increasing excise taxes on illicit trade in tobacco affect your decision to increase them. Rely on your own estimates of the level and nature of illicit trade and not on the industry’s estimates. Illicit trade in tobacco products continues to be a major concern for tax administrators because of the difficulties associated with accurate and independent measurement of it, as well as with its elimination. Industry figures provide a distorted understanding of the extent of the problem, along with a monocausal explanation of the link between illicit trade and tobacco taxation. It is therefore recommended that governments (1) assess independently and with the best statistical practices the size of the illicit trade to assess the scope of the problem; (2) address directly the country-specific institutional and/or governance challenges, including multilateral coordination, and improve tax and customs administrations practices; and (3) implement best practices to fight illicit trade, contained in the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products. Ideally, accede to the Protocol if not yet a Party. C: Court and legal challenges Do not let tobacco industry threats of court and legal challenges to tax increases or reforms prevent you from improving your tax policy. Closely follow legal requirements for design, procedure and consultation to strengthen your legal position and minimize the possibility that any challenge will be raised. Health-protective and non-discriminatory tobacco excise taxes are legally defensible, and industry threats will usually be baseless. Your legal position can be strengthened, however, by exercising care with a tax measure’s procedure, design and consultation: (1) determine the standard of consultation required under domestic law and any applicable international obligations; (2) distance the tobacco industry from the policy-making process to the extent that this is permissible; (3) avoid unnecessary and unjustified discrimination towards foreign tobacco products or investors in the design, implementation or enforcement of a tax measure; and (4) do not offer investment incentives in the form of inducements or contractual undertakings, as these may be binding in and of themselves or grounds for a challenge under an international investment agreement. 280 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A: Anti-poor rhetoric Do not allow concerns about the regressivity of higher tobacco taxes prevent tobacco tax increases. In recent years, there has been an overwhelming increase in the evidence concern- ing the positive distributional impact of tobacco taxes and tax increases. Indeed, tobacco taxation and tax increases are actually a progressive or pro-poor policy once these wider considerations are properly accounted for. In its effort to lobby against tax increases, the tobacco industry often claims that tobacco taxation will hurt the poor. This argument is based on the concept of regressivity in relation to taxation. Conceptually, a tax is regressive if it means lower-income people must pay a relatively greater proportion of their household income to meet the tax liability than wealthy people. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use that are disproportionately experienced by lower socioeconomic groups. Second, higher tobacco taxes and prices can induce behavioural change in the population, as reflected in the price elasticity of demand, which means that lower-income smokers will curtail their smoking the most and thus will benefit disproportionately in terms of health gains from reduced tobacco consumption and use. In fact, these broader considerations make tobacco taxation a progressive, rather than regressive, public health intervention. R: Revenue reduction Do not let fears of potential revenue reductions prevent you from increasing excise taxes on tobacco products. Tax increases, even in countries with already high taxes, bring in additional revenue. Arguments by tobacco control opponents that tax increases will not result in increases in revenue are unfounded. The relatively price inelastic nature of cigarette demand, combined with the low tax share and no overshifting of the tax, means that for most, if not all, countries, increases in revenues will accompany increases in taxes. If tax increases are carefully designed and tax administration is functional, it is extremely unlikely that tax increases will lead to revenue decreases. E: Employment impact Do not allow concerns about employment impact to prevent tobacco tax increases. The tobacco industry often seeks to frame tobacco taxes as an economic issue rather than a public health issue. Particular emphasis is placed on the alleged threat tax increases pose to employment in tobacco farming and manufacturing, as well as related industries. This so-called choice between health and jobs, however, is largely based on exaggeration. The tobacco industry exaggerates the importance of tobacco CHAP T ER 5. B E S T PR AC T I CE S 281 employment relative to total national employment and overstates the impact that domestic demand reduction from local taxes will have on tobacco farmers serving a global market. The argument used by the industry also ignores the fact that expendi- tures on tobacco do not disappear but rather are redistributed to other consumption that can produce a similar or higher number of jobs. Case studies demonstrate the possibility and methods for governments to support farmers in transitioning to other crops that provide similar and often better returns with greater sustainability. Earmarking Consider earmarking tobacco tax revenues for health-focused programmes, especially if it helps advance tobacco control efforts and, more specifically, efforts to implement large tobacco tax increases and tax reforms. This could have the additional benefit of funding health programmes where they are poorly funded or not prioritized. From a tobacco control perspective, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. Earmarking is a tool to improve the political economy of tobacco taxation; it is only a secondary issue, after the primary goal of reducing demand for tobacco. Evidence shows that public support for higher tobacco taxes is greater when at least some of the increased revenues are explicitly used to support health-focused programmes. Current evidence shows that the amounts effectively earmarked for health have been relatively small and therefore unlikely to introduce rigidity in government budgets. At the same time, in some countries, those funds have helped to imple- ment much needed underresourced health programmes. The payoffs will be seen in the future as fewer people fall ill and need less medical care for tobacco-related illnesses. Earmarking tobacco tax revenues to fund high-burden/low-priority health programmes could pave the way for raising awareness about the importance of such programmes and their effectiveness, thereby convincing governments to redefine their priorities and commit to including the programmes in their regular budget. 282 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) Focus on tobacco taxation’s purposes Tobacco tax policy should aim not only to increase revenues but also to decrease consumption and improve health. To both raise revenue and reduce consumption, you need to (1) simplify tobacco tax structures, (2) significantly increase rates to impact price levels, and (3) regularly adjust rates to at least account for inflation and income growth. Analyse your tax structure and identify its weaknesses You need to analyse and identify the problems of your current tax structure to know which steps to consider next. Which tax structure do you have: specific, ad valorem, mixed, or no excise? Identify the improvements to be made to the existing tax policy/structure Your present tax structure and tax situation will provide you with the steps you would ideally take next to achieve the aims in Step 1. Specific: 1. Ensure that the tax automatically adjusts for inflation and income growth effects. 2. Ensure that all price promotions are banned. Ad Valorem: 1. Ensure that the tax base of the ad valorem is retail price. 2. Introduce a high specific excise component (and a minimum specific excise). 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. Mixed: 1. Ensure that the tax base of the ad valorem component is retail price. 2. Ensure that you are using a high specific excise component and a minimum specific excise. 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. No Excise: 1. Introduce a high specific excise. 2. Ensure that the rate automatically adjusts for inflation and income growth effects. 3. Ensure that all price promotions are banned. Assess tobacco taxation’s political economy Reform must begin with an assessment of tobacco taxation’s political economy: (1) learn from past successes and failures – what went wrong, what went right, what you can do differently this time; (2) assess the reform’s strengths and weaknesses, likely opportunities and risks; (3) determine who the main supporters and opponents of reform inside and outside of government have been and may be, based on past reforms and current situation; and (4) anticipate argu- ments that will be used against the reform (refer to SCARE tactics). STEP 1 STEP 2 STEP 3 STEP 4 Prepare a plan for realizing the reform Focusing on the overall aims identified in Step 1, the steps for achiev- ing them identified in Steps 2 and 3 and the political economy around this reform as identified in Step 4, prepare your plan: 1. Be clear on the non-negotiable objectives for the reform and the trade-offs you are prepared to make to realize them. 2. Develop a plan to approach potential allies and win them over to the reform efforts. 3. Develop the counterarguments that will be needed in response to the SCARE arguments identified earlier. 4. Prepare the evidence you will need ahead of time. To do this, get support from academics and relevant intergovernmental agencies. Mobilize a coalition for reform 1. Formulate a strategic communications plan: aim for political support both at the highest levels and among the public (framing tobacco taxation as a health issue has helped win political support in many countries). 2. Identify champions in government: ensure that finance and health officials are on the same page; involve implementing departments, such as enforcement agencies, from the start. 3. Mobilize allies from academia, civil society and the private sector to counter the anticipated pushback from the tobacco industry, its proxies and its allies. Monitor and evaluate To make the most well-informed policy decisions, a reform effort should be monitored to assess its overall impact and its effect on key indicators; this will help identify issues to be fixed while also creating a strong evidence base for further reform efforts. Get and analyse the relevant data to better understand the market situation and its dynamics: 1. Monitor the market and its evolution (e.g. retail prices, duty-paid sales, market shares). 2. Get regular estimates of price elasticity (including cross-price elasticity), income elasticity and tax base elasticity to evaluate any changes in tobacco demand. Use relevant tools to assess the impact of the tax policy on consumption and revenue: 1. Use specific tools on the impact of excise on price, consumption and revenue (e.g. the WHO TaXSiM). 2. Use global tools to assess the tax increase’s impact on prevalence (e.g. the WHO ISPT). Monitor key indicators closely to assess improvements over time: 1. Tax as a percentage of retail price. 2. Change in affordability of tobacco products over time. 3. Change in the tobacco tax scorecard, which combines a mix of best practices in tax policy. 4. Change in sales, prevalence and illicit trade in tobacco products. 5. Improvements in MPOWER package achievement. STEP 5 STEP 6 STEP 7

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION WHO technical manual on tobacco tax policy and administration ISBN 978-92-4-001918-8 (electronic version) ISBN 978-92-4-001919-5 (print version) © World Health Organization 2021 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/ rules/). Suggested citation. WHO technical manual on tobacco tax policy and administration. Geneva: World Health Organization; 2021. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Document design by Ana Sabino. iii Contents Foreword v Acknowledgements vii Acronyms viii Executive summary xi CHAPTER 1. Why this manual? 1 CHAPTER 2. Tobacco excise tax policy 11 2.1 Global overview of tobacco tax practices 11 2.2 Designing excise tax policy 18 2.3 Domestic and regional policy integration 55 2.4 New and emerging nicotine and tobacco products 59 2.5 Conclusions 75 ANNEX 2.1 Countries that apply different types of excise tax structures 89 ANNEX 2.2 Analytics of the tax base elasticity 90 ANNEX 2.3 Elements of the devices that make up ENDS/ENNDS products 91 CHAPTER 3. Tobacco tax administration 93 3.1 Introduction 93 3.2 Institutional arrangements 94 3.3 The tax compliance cycle 102 3.4 Control and enforcement 115 3.5 Tax administration of other tobacco products 152 3.6 The broader elements of a good tax system 155 3.7 Conclusions 156 ANNEX 3.1 Composition of tobacco products 166 ANNEX 3.2 Example of forestalling and countermeasures 172 CHAPTER 4. Political economy 175 4.1 SCARE tactic S: Smuggling and illicit trade 176 4.2 SCARE tactic C: Court and legal challenges 198 4.3 SCARE tactic A: Anti-poor rhetoric or regressivity 212 4.4 SCARE tactic R: Revenue reduction 217 4.5 SCARE tactic E: Employment impact 228 4.6 Earmarking tobacco tax revenues to fund health 233 ANNEX 4.1 Methods to assess the nature and size of the illicit tobacco trade 253 ANNEX 4.2 How are the tobacco tax revenues earmarked? 269 CHAPTER 5. Best practices 271 TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) 283 iv v Foreword In 1999, the World Bank’s Curbing the Epidemic was the first report by an interna- tional organization to recognize that increasing tobacco excise taxes was the most effective and cost-effective measure to reduce tobacco use and save lives. Over the two decades since, the evidence base supporting this claim, especially in low- and middle-income countries, has been steadily growing. Meanwhile, the credibility of the tobacco industry’s arguments against tobacco taxation has been slowly waning. In short, health-promoting tobacco taxation has come of age, and the evidence has consistently shown that it is a win for public health, a win for revenue and a win for the economy overall. But we must be cautioned against complacency. Although the evidence on tobacco taxation is irrefutable and there are now signs that the tide is turning on the global tobacco epidemic, tobacco taxation was, in 2018, the WHO MPOWER1 measure that was least implemented at the highest level of achievement. Even more concerningly, cigarettes have become more, rather than less, affordable in many low- and middle-income countries over the past decade. Many countries set rates at insufficient levels and increase them too infrequently, while others still use complex and inefficient taxation structures. This failure to advance tobacco taxation able to effect significant price increases constitutes a loss for governments in revenues, a loss for public health and a win for the tobacco industry. To overcome this inertia, this manual charts the way forward for policy-makers, finance officials and others involved in tobacco tax policy development. It equips them with the information and evidence needed for the realization of their coun- tries’ tobacco tax policy objectives. It also analyses the tobacco industry’s tactics for influencing the political economy of tobacco taxation and shows the limitations and exaggerations of the arguments used against tax increases. The manual serves as an update of the 2010 WHO technical manual on tobacco tax administration by adding new evidence on the successes of tobacco taxation in all parts of the world and broadening its scope to capture more material relevant to developing and implementing more effective tobacco tax policy. 1 The WHO MPOWER package of technical measures and resources that comprises (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco. vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Its contributions are particularly timely in a COVID-19-stricken world. As the pandemic has been worsened by the global burden of noncommunicable disease, and revenue is now desperately needed, taxing tobacco should be more palatable than ever. This manual shows policy-makers how to seize this unique opportunity to use tobacco taxation to build back better, save lives and strengthen health systems while increasing revenue. Dr Naoko Yamamoto Assistant Director-General UHC/Healthier Populations Division vii Acknowledgements This manual was developed under the direction of Jeremias Paul Jr and Anne-Marie Perucic. The following contributed to the content of the manual (in alphabetical order): • WHO: Evan Blecher, Annerie Bouw, Mark Goodchild, Roberto Iglesias, Juliette McHardy, Jeremias Paul Jr, Anne-Marie Perucic and Robert Totanes. • External authors: Mauricio Cardenas (Columbia University), Sophia Delipalla (University of Macedonia), Luk Joossens (tobacco control expert, Belgium), Marin Kurti (Eastern Connecticut State University), Enrique Fanta (former World Bank senior specialist), David Merriman (University of Illinois at Chicago) and Jean Tesche (University of Cape Town). WHO would like to thank the following reviewers for their invaluable comments (in alphabetical order): • External: Jo Birckmayer (Bloomberg Philanthropies), Adriana Blanco Mar- quizo (WHO Framework Convention on Tobacco Control), Frank Chaloupka (University of Illinois at Chicago), Yoni Dekker (WHO Framework Conven- tion on Tobacco Control), Jeffrey Drope (University of Illinois at Chicago), Ceren Ozer (World Bank), Corne van Walbeek (University of Cape Town), Chonlathan Visaruthvong (Ministry of Finance, Thailand) and Rodrigo Santos Feijo (WHO Framework Convention on Tobacco Control). • WHO: – Headquarters: Douglas Bettcher, Itziar Belausteguigoitia, Ranti Fayokun, Joseph Kutzin, Benn McGrady, Vinayak Prasad and Susan Sparkes. – Regional offices: Nina Dela Cruz (WPRO), Fatimah El-Awa (EMRO), Charles Frasier (EMRO), Lee Lily Joung-Eun (WPRO), Jagdish Kaur (SEARO), Elizaveta Lebedeva (EURO), Maxime Roche (AMRO) and Rosa Sandoval (AMRO). WHO would also like to thank Amal Amoune-Naal for the administrative support, Alison Goldstein for the technical editing and Janet DeLand for the copy-editing. Production of this document has been supported by a grant from Bloomberg Phi- lanthropies. The contents of this document are the sole responsibility of WHO and should not be regarded as reflecting the position of Bloomberg Philanthropies. viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ACRONYMS AEO authorized economic operator AFRO WHO Regional Office for Africa AMRO WHO Regional Office for the Americas ATO Australian Taxation Office BAT British American Tobacco CCTV closed-circuit television CIF cost, insurance and freight COP Conference of the Parties CPI consumer price index CVA Customs Valuation Agreement (Thailand) DGCE Directorate General of Customs and Excise (Indonesia) DIY do-it-yourself ECBA World Bank Extended Cost-Benefit Analysis EIU Economist Intelligence Unit EMRO WHO Regional Office for the Eastern Mediterranean ENDS electronic nicotine delivery systems ENNDS electronic non-nicotine delivery systems EU European Union EURO WHO Regional Office for Europe FDA Food and Drug Administration (United States) FET fair and equitable treatment GCC Cooperation Council for the Arab States of the Gulf GDP gross domestic product HTP heated tobacco product HMRC Her Majesty’s Revenue and Customs (United Kingdom) IARC International Agency for Research on Cancer IIA international investment agreement IMF International Monetary Fund IRS Internal Revenue Service (United States) ISO International Organization for Standardization IT information technology ITC International Tobacco Control JTI Japan Tobacco International KRA Kenya Revenue Authority LMICs low- and middle-income countries ACRONYMS ix MFN most favoured nation MOP Meeting of the Parties (to the Protocol) MPOWER (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco NCDs noncommunicable diseases NCI National Cancer Institute NT national treatment OECD Organisation for Economic Co-operation and Development OST other smoking tobacco PMI Philip Morris International PPP purchasing power parity QR quick response RGTE WHO Report on the global tobacco epidemic RYO roll-your-own SACU Southern African Customs Union SCARE (S) smuggling and illicit trade; (C) court and legal challenges; (A) anti-poor rhetoric; (R) revenue reduction; and (E) employment impact SDGs Sustainable Development Goals SEARO WHO South-East Asia Regional Office SII Internal Revenue Service of Chile TADAT Tax Administration Diagnostic Assessment Tool TTC transnational tobacco company UAE United Arab Emirates UHC universal health coverage VAT value added tax WAEMU West African Economic and Monetary Union WCO World Customs Organization WHO World Health Organization WHO FCTC WHO Framework Convention on Tobacco Control WHO ISPT WHO interactive smoking projection and target-setting tool WHO TaXSiM WHO tobacco tax simulation model WPRO WHO Regional Office for the Western Pacific WTO World Trade Organization x W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N xi Executive summary This WHO technical manual on tobacco tax policy and administration builds upon the 2010 WHO technical manual on tobacco tax administration by further detailing the strategies for effective tobacco tax policy development, design, implementation and administration. This 2021 edition also serves as an update to the 2010 manual, incorporating the latest developments in science, technology and policy, as well as providing illustrative recent examples from a variety of countries. The best practices laid out in this manual are designed to inform governments on the development of their tobacco taxation policy, facilitating the achievement of their health and revenue objectives while also supporting their overall development strategy. Tobacco taxes have long been seen as a source of revenue for governments, but as evidence of the harms caused by tobacco has accumulated over the years, public perception has evolved. Increasingly, governments, as well as the general public, are recognizing that taxation of tobacco is not only a revenue source but also an effective public health intervention to reduce tobacco consumption and its associated harms. The profile of tobacco taxation as a health policy tool has increased greatly since the publication of the 2010 WHO technical manual on tobacco tax administration. Multiple global commitments have been adopted over the past decade to address tobacco use specifically – as well as noncommunicable diseases (NCDs) and the Sustainable Development Goals (SDGs) more broadly – through tax and price measures to reduce demand for tobacco products, save lives and fund develop- ment. Global development institutions, including the World Bank, the International Monetary Fund (IMF) and major philanthropic foundations, also agree with WHO on the importance of emphasizing and strengthening tobacco taxation as a key health policy tool. The COVID-19 pandemic has further fuelled this shift in the narrative on tobacco taxation by revealing how the global economy is inextricably linked with population and planetary health. Investing in health is fundamental to any economic recovery, and fiscal policy will be a key driver in addressing the socioeconomic consequences of COVID-19. Interventions such as tobacco taxation – which leads to reduced tobacco consumption, improved population health and increased revenues for governments – should be part of a comprehensive strategy for a build back better recovery. The evidence is clear: significant increases in excise taxes that lead to price increases have consistently proven to be the most effective, as well as the most xii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cost-effective, mechanism for reducing tobacco consumption. This manual will guide readers through the necessary steps to create and implement the strongest tobacco taxation policies for their specific countries. There are many factors to consider when developing tobacco taxation policy. Understanding the market is a fundamental step before deciding which form of taxation to use. Specifically, the choice between ad valorem and specific taxation is influenced by the market structure in a given country. At the same time, tax structure also shapes the market structure. Over the past decade, there has been a trend towards countries adopting specific excise taxes or mixed excise systems that rely more heavily on the specific component, which the latest global data associate with the highest average prices. Tobacco tax structures can be simple, with one flat rate across the board, or com- plex, with multiple tiers for products with different characteristics. In 2018, 31 coun- tries used complex, multitiered structures. But evidence demonstrates that simpler excise tax structures – utilized in all high-income countries – leave the least room for industry manipulation or tax avoidance and brand/product switching by consumers. Not only is it important to set taxes at a high level to discourage consumption, specific excise tax policies must include regular adjustments to increase the tax rate so that it keeps up with inflation and income growth in a country over time. Excise tax increases should aim to reduce the affordability of tobacco products. The base on which the tax is applied is also important. For specific taxation, the tax base should be the quantity in clearly defined units. For ad valorem (or mixed) taxation, the best practice is to use the retail price as the tax base and introduce a minimum excise tax. With regard to non-tax regulations that affect the price of tobacco products, pric- ing regulation may be considered to prevent the tobacco industry from exercising differential tax shifting, which it uses to ensure that large price gaps exist between premium and cheap cigarettes. However, pricing policies cannot be used alone. If con- sidered, they should be used only as complements to significant excise tax increases. Other non-tax regulations include banning promotional discounts for tobacco products and banning the sale of single cigarettes. To assuage concerns that tax increases will increase inflation – as well as to reflect the declining trend in consump- tion of tobacco products – it is good practice to exclude tobacco products from the basket of items that are used to develop consumer price indexes. Finally, in order to make excise tax on tobacco products more effective in reducing overall tobacco use, all tobacco products must be taxed in a comparable way. Regular assessment, evaluation and monitoring of the impact of tobacco tax policies over time are essential components of effective tax policy development and analysis. Governments need to have accurate estimates of price, income and tax base elasticities in order to anticipate the impact of a tax increase on consumption and EXECUTIVE SUMMARY xiii tax revenue. Ideally, other factors such as non-price policies should also be taken into account when estimating price and income elasticities for a specific country. A variety of tools and indicators exist to measure impact and monitor progress, and these are described in Chapter 2 of this manual. When developing tobacco tax policy, it is also important to take the broader policy context into consideration at both the domestic and the regional level. Domestically, cooperation is needed across sectors to ensure that policies and interventions in the areas of agriculture, trade, finance and labour do not work against the public health objectives of tobacco control and taxation. For countries that are part of a regional bloc, regional harmonization of tobacco taxation is a useful tool to prevent tax revenue erosion, tax avoidance and tax evasion, as well as to protect population health. Tax harmonization must be designed carefully, however, to be effective. The experience of the European Union (EU) demonstrates that both a declining consumption trend and stable revenues can be achieved with harmonized minimum excise tax rates. Discussions of policy development and implementation for new and emerging nicotine and tobacco products such as heated tobacco products (HTPs) and electronic nicotine and non-nicotine delivery systems (ENDS/ENNDS), are complicated, by their constantly changing technology and market dynamics. Policies and regulations need to be developed carefully and adjusted accordingly. Where HTPs are not banned, the current recommendation is to tax them at the same level as cigarettes on a per-unit basis, regardless of tobacco content. Early evidence from the United States shows that demand for e-cigarettes, a subcategory of ENDS/ENNDS products, is possibly even more price-responsive than the demand for conventional cigarettes, meaning that taxes can be used as an effective deterrent to ENDS/ENNDS products use. While there is preliminary evidence of substitutability between conventional cigarette use and e-cigarette use, further research is needed to understand substitutability effects among users of both conventional cigarettes and ENDS/ENNDS products. It is essential to imple- ment regulation of ENDS/ENNDS products along with any tax policy to safeguard public health. In countries where they are not banned, ENDS/ENNDS products must be regu- lated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS taxation. Nicotine-containing and non-nicotine-containing e-liquids should be taxed equally. Ultimately, while the policy implications of these newer products require careful consideration, the fact remains that conventional tobacco products constitute the overwhelming share of consumption (more than 97% in 2018). Tobacco tax administration must be both efficient and effective to ensure that health objectives are met and the desired level of tax revenue is raised. Since the xiv W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation of tobacco taxation often involves numerous agencies within a country, clearly defined roles and responsibilities are essential to maximize efficiency. Coordination among the different agencies involved, as well as with neighbouring countries, is required for tobacco tax administration to be effective. Performance evaluation and accountability for competent authorities is also necessary, and many tools and indicators exist to facilitate these processes (described in Chapter 3). There are a number of steps authorities should take to ensure efficiency and effectiveness at each stage throughout the tax compliance cycle (see Chapter 3, section 3.3). Control and enforcement are the main functions of tax administration, and these can best be achieved through the use of a strategic plan and a risk-based approach. Controls can be exercised through licensing and due diligence, fiscal markings (e.g. tax stamps), tracking and tracing, implementation of anti-forestalling measures, national audits and specific controls for imports and exports, as well as for free zones and transhipment points. Once smuggling or illicit trade is detected, actions such as seizing and destroying smuggled and/or illicit tobacco and col- lecting due taxes must be taken immediately. To deter further illegal activities, a comprehensive audit must also be carried out, including all those involved in the illicit acts. Penalties and sanctions must be sufficient to deter illegal activities. The Protocol to Eliminate Illicit Trade in Tobacco Products provides invaluable guidance for tobacco tax administration, control and enforcement that is applicable even for countries that are not Parties to it. The broader elements of a good tax system include proper resourcing of competent authorities, strict rules and regulations to detect and punish corruption and a strong judiciary system capable of resolving disputes as soon as possible. In its efforts to oppose tobacco tax increases, the tobacco industry utilizes many SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric, (R) revenue reduction and (E) employment impact – to influence the political economy of tobacco. Chapter 4 provides detailed analyses of these issues with supporting evidence that belies the SCARE tactics, as well as guidance for tax and other relevant authorities on how to anticipate and respond to industry arguments. This manual also provides tools and methodologies to help tax authorities define and evaluate the problem of illicit trade of tobacco products in their countries, inde- pendent of the tobacco industry’s generally inflated estimates. Price (and tax) levels are not a key determinant of illicit trade; rather, the problem is exacerbated by the lack of governance and tax administration capacity. Refraining from increasing taxes is not the solution. Countries should instead respond with a comprehensive strategy to fight illicit trade, including undertaking independent estimates of illicit trade levels and implementing good tax administration practices such as those discussed in Chapter 3 and contained in the WHO Framework Convention on Tobacco Con- trol (WHO FCTC) Protocol to Eliminate the Illicit Trade in Tobacco Products. When it comes to court challenges, the tobacco industry is less likely to challenge excise taxes than other tobacco control measures, because taxation is a comparatively well-established regulatory measure. The industry will, however, exploit the slightest vulnerability in the design, adoption or implementation of tax measures. For this reason, measures to strengthen regulators’ legal position are described that will enable authorities to protect themselves from potential legal challenges. The industry argument of regressivity, or the notion that tobacco tax increases hurt the poor because they have to pay a larger share of their income in taxes than the rich, has two fundamental limitations. First, the notion of regressivity does not take into consideration the broader health and economic harms caused by tobacco use that exacerbate the impoverishment of lower-income smokers. These harms are actually reduced when tobacco consumption decreases following a tax increase. Second, the tobacco industry argument ignores the fact that higher tobacco taxes and prices can induce behaviour change as is reflected in the price elasticity of demand. Evidence consistently shows that lower-income smokers are more sensitive to price and therefore more likely to reduce smoking in response to a tax and price increase. Including these factors shows tobacco taxation to be, in fact, a progressive public health intervention that disproportionately benefits the poor. While essentially admitting that a tobacco tax increase may have the desired effect of reducing consumption, the industry also tries to argue that a tax increase will also reduce revenues. In fact, the price inelastic demand for tobacco makes tobacco tax increases a win-win for both public health and finance. This manual presents several country examples that demonstrate how well-designed and well-implemented tobacco tax increases lead to increases rather than decreases in revenue in the short to medium term. In addition, the reduced consumption resulting from a tax increase results in reductions of other tobacco-related government expenditures as well. The final tactic used by the tobacco industry to challenge proposed tax increases is to frame tobacco taxes as an economic rather than a public health issue. This false choice between health and jobs is based on faulty assumptions that 1) tobacco is a significant source of domestic employment; 2) job creation relies on tobacco consumption and 3) tobacco-related livelihoods are prosperous, sustainable and irreplaceable. Earmarking can be a useful tool for improving the political economy of tobacco tax increases. While the primary goal of tobacco tax increases is to reduce demand for tobacco, setting aside portions of tax revenue to fund other tobacco control efforts or relevant health programmes can help convince the public, politicians and officials of the value of significant tobacco tax increases. Earmarking can also be EXECUTIVE SUMMARY xv x vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N used to counter tobacco industry arguments about potential negative effects of tax increases – for example, by supporting tobacco farmers in transitioning to other crops. Tobacco taxes work. This is why the industry invests so much money and effort in blocking large tax increases and other effective tax policy reforms. Policy-makers must not be swayed by industry pressure but need only to follow the facts. This manual provides all the information policy-makers need to make the right deci- sions at each step of the process – from designing, evaluating, implementing and administering tax policy to refuting specious industry attacks and communicating the value of tobacco taxation to legislators and the broader population. An effectively designed and efficiently administered tobacco tax policy will not only produce the direct results of reducing tobacco consumption among smokers and raising revenue for governments, its effects will be felt much more broadly. Indeed, raising tobacco taxes is a SMART policy: it Saves lives; Mobilizes resources; Addresses health inequities; Reduces burdens on health systems; and Targets tobacco use, a major risk factor for NCDs. • Saves lives: Tobacco use is the leading cause of preventable deaths globally – it claims 8 million lives each year. Tobacco taxation is the most effective mechanism for reducing tobacco consumption and its associated health burden worldwide. • Mobilizes resources: Despite being the single most effective tobacco control measure, tobacco taxation is largely underutilized as a policy mechanism. Based on available data on the price and taxation of cigarettes, it is estimated that excise taxes on cigarettes generated a worldwide total of US$ 361 billion in revenues in 2018, including US$ 162 billion in revenues for low- and middle- income countries (LMICs). If all countries were to raise cigarette excise tax rates by the equivalent of US$ 1 per pack, the amount of excise revenue from cigarettes would increase by US$ 178–219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from such tax increases, with excise revenues increasing by 82–103%, providing governments in these countries with an extra US$ 133–167 billion. This shows the substantial revenue potential of tobacco taxes. • Addresses health inequities: Tobacco taxation and tax increases are effectively progressive or pro-poor policies because of their positive distributional impact. Lower-income smokers benefit disproportionately from reduced tobacco consumption and use in terms of health gains and income retention. • Reduces burdens on health systems: The worldwide economic cost of tobacco use was US$ 1.4 trillion in 2012. Tobacco taxes reduce tobacco-related burdens on governments and health systems through population-based preventive measures. • Targets tobacco use: Tobacco taxation directly targets and reduces tobacco use, which is a major risk factor for several deadly NCDs. In summary, significant tobacco tax increases, designed and implemented according to the latest guidance and best practices presented in this technical manual – and as a strong component of a comprehensive tobacco control strategy – will bring about substantial reductions in tobacco use and the health and economic harms it causes. EXECUTIVE SUMMARY xvii x viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 1 CHAPTER 1. Why this manual? BACKGROUND Tobacco taxes are not new. Governments around the world have been applying taxes on tobacco and tobacco products practically since the idea of excise was conceived. And rightly so: tobacco is not a necessity, it is easy to tax and the demand for it is relatively inelastic. These characteristics, along with the substantial revenues tobacco taxes generate, have made tobacco a highly appropriate object of taxation. As evidence of the harms of tobacco has accumulated over the years, the public perception of tobacco taxes has evolved. Now tobacco taxes are not only seen as a revenue source, but, more importantly, they are recognized as an effective public health intervention to reduce tobacco consumption. This trend reflects the reasons excise taxes exist in the first place – to discourage harmful behaviour and to mitigate the associated negative externalities (1–2). Many governments view tobacco taxes as a significant and stable source of rev- enue, which may explain why there is often a degree of hesitation whenever tobacco tax reform is proposed. Historically, many governments have relied on revenues from tobacco taxes and have even adjusted the level of taxation according to their revenue needs (3). However, some countries are beginning to recognize the value of applying high tobacco taxes primarily as a public health tool, viewing revenues as a secondary consideration (4). Arguments against tobacco tax hikes or improvements to the tax structure are often economic in nature: such tax changes will allegedly decrease revenues, wipe out jobs, increase illicit trade and harm local industries, among other claims. But the evidence has consistently shown that such claims are simply not true in an overwhelming majority of situations. The tobacco industry, in particular, frequently portrays this conflict as a false dichotomy between public health and the economy – as if prioritizing health comes at the expense of the economy. In fact, studies and real-world experiences have shown that increasing tobacco taxes not only improves public health but also has a net positive impact on the economy and development of a country – a true win-win scenario (5–6). As an update to the first WHO technical manual on tobacco tax administration published in 2010, this manual aims to help readers better navigate the various 2 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N issues surrounding tobacco taxes and their implementation. The primary intended audience includes policy-makers, finance officials, tax authorities, customs officials and other relevant persons/bodies involved in the formulation and implementation of tobacco tax policy. The manual provides a detailed guide to the design of tobacco tax policy and describes how to effectively administer these taxes to maximize impact. Detailed discussions of the political economy considerations and the hurdles that need to be overcome before and during implementation are included as well. The overarching goal is to equip those working in the tax policy and implementation spheres with sufficient information to help realize the health and revenue objectives of a government’s tobacco tax policy in line with its overall development strategy. THE CONSEQUENCES OF TOBACCO USE AND THE NEED FOR INTERVENTION Most people are aware that smoking and tobacco use are harmful to health, but few truly comprehend the scale of this harm. The tobacco epidemic claimed more than 100 million lives in the last century (7), with updated estimates now reaching 8 million deaths annually from tobacco use and exposure to second-hand smoke (8). As much as 80% of these deaths occur in low- and middle-income countries (LMICs) (6), revealing how the developing world carries much of the global burden. Tobacco use is a major risk factor for many chronic conditions, including heart disease, cancer, diabetes and chronic lung disease – collectively known as noncom- municable diseases (NCDs). NCDs account for about 15 million premature deaths (between ages 30 and 69) worldwide, killing people in their most productive years. As the leading cause of preventable deaths, tobacco use remains one of the foremost public health challenges of our time. The consequences of tobacco use also present enormous economic, development and social costs that wreak havoc on families, communities and societies. The annual economic cost of smoking was estimated at US$ 1.4 trillion in 2012, equivalent to 1.8% of the global gross domestic product (GDP) (9). With these figures likely to have increased since then, the massive health and economic burdens of tobacco use provide justification for governments to intervene and strictly regulate the market for tobacco products. The purview of tobacco control extends beyond the strong imperative to protect people’s health and well-being; it should also strive to contain the market failures and negative externalities of tobacco use, particularly since these effects can significantly impact a country’s development trajectory. The mounting evidence of the enduring destruction caused by tobacco in the 20th century provided compelling reasons for a strong global response, which led countries to negotiate the World Health Organization Framework Convention on Tobacco Control (WHO FCTC). The WHO FCTC came into force in 2005 as the CHAP T ER 1. WHY T HIS M ANUAL? 3 first public health treaty under the auspices of WHO. To facilitate its implementation at the country level, WHO packaged a set of demand-reduction measures directly taken from the treaty (7). These interventions, collectively known as MPOWER, are as follows: (M) monitoring tobacco use and prevention policies; (P) protecting people from tobacco smoke (smoke-free laws); (O) offering help to quit tobacco use (cessation services); (W) warning about the dangers of tobacco (including graphic pack warnings and plain packaging); (E) enforcing bans on tobacco advertising, promotion and sponsorship; and (R) raising taxes on tobacco products. Specifically, under Article 6 of the WHO FCTC, Parties recognized that price and tax measures are an effective and important means of reducing tobacco consumption for various segments of the population – in particular, among young persons (10). The severity of the tobacco epidemic and its ongoing damage to health and economies are clear justifications for governments to actively intervene and correct market failures. The scale of the burden and the rate at which lives are being destroyed necessitates urgent and aggressive action on tobacco control, using measures that most countries have committed to implementing and that are proven to be effective in reducing tobacco use. WHY TOBACCO (EXCISE) TAXES ARE CRUCIAL Among the different tobacco control interventions, raising excise taxes has been identified as the most effective as well as the most cost-effective measure to reduce consumption (6). While other interventions are certainly important components of a comprehensive tobacco control strategy, the direct impact of significant tax increases on consumption is by far the strongest. On average, a tax increase that causes prices to go up by 10% reduces consumption by 4% in high-income countries and 5% in LMICs (6). When implemented at scale, this demonstrates the enormous power of tobacco taxation and its potential to save lives. Tobacco taxes differ from other interventions in that their impact can increase and build over time – even if taxes are already relatively high, their rates need to be continuously increased to retain and amplify their effectiveness. However, this should not be taken as a suggestion that governments considering tobacco control interventions should focus solely on taxes. Taxes are even more effective when implemented as part of a comprehensive package of measures such as MPOWER, which covers distinct but complementary intervention points. Among the different taxes applied on tobacco products, excise taxes are the most significant because they raise both absolute and relative prices (6). This is important when considering health objectives, since it is the magnitude of the price increase of tobacco products that determines the reduction in consumption. An excise tax is typically applied on a limited set of products, designed to discourage their use by 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N raising the price significantly over that of other products available in the market. This is in contrast to value added taxes (VAT) or sales taxes, which apply to most goods and services. Seeking to raise the prices of tobacco products through VAT or sales taxes would fail to increase relative prices, making this an ineffective and inefficient method. Customs or import duties on tobacco products are also utilized, but their impact is waning with the global trend towards bilateral and regional agreements aimed at trade facilitation. The application of these duties varies across countries, but overall, they are not applicable to locally produced tobacco products. As a tool to increase prices, import duties cannot substitute for excise taxes, since they are not specifically designed to reduce consumption. When viewed as a public health policy tool, tobacco taxation is highly cost-effective, since it delivers significant impact yet is relatively inexpensive to implement (11). The costs of implementing tobacco taxation are much lower than those of clinical NCD interventions such as cancer treatments or maintenance medications, since the commodity and human capital requirements are less substantial (12). Moreover, increasing tobacco taxes actually generates additional revenue for a government. Tobacco taxes are also very effective in pre-empting or reducing consumption among groups of people who are especially price-sensitive – youth in particular, who are prevented from initiating a lifelong addiction if taxes and prices are sufficiently high (13). This is also true for the poor, who are more prone to catastrophic health expenditures than the wealthy are. Preventing initiation or encouraging cessation by imposing high taxes provides an escape route from the vicious cycle of tobacco use and poverty (6). A DECADE OF PROGRESS AND COMMITMENT TO ACTION Since 2010, when the first WHO technical manual on tobacco tax administration was published, numerous developments have raised the profile of tobacco taxes as an essential public health intervention. The Conference of the Parties (COP) to the WHO FCTC adopted guidelines for implementation of Article 6 of the treaty, which focuses on price and tax measures to reduce the demand for tobacco. Also within this period, three high-level meetings on the prevention and control of NCDs by the United Nations General Assembly, as well as the endorsement of the Global NCD Action Plan in 2013 by the World Health Assembly, have resulted in strong global commitments to implement measures, such as increased tobacco taxes to protect people’s health. The 2030 Agenda for Sustainable Development, which contains 17 goals known as the Sustainable Development Goals (SDGs), describes the global development strategy for the next decade. Within the SDGs, two specified targets are highly relevant for tobacco control: strengthening the implementation of the WHO FCTC (target 3.a) CHAP T ER 1. WHY T HIS M ANUAL? 5 and reducing premature mortality from NCDs by 30% (target 3.4). Furthermore, the Addis Ababa Action Agenda1, which aims to provide a global framework for financing the SDGs, also highlights tax and price measures on tobacco as key mechanisms to reduce demand and save lives while increasing domestic resources for develop- ment. Another important milestone was the 2018 entry into force of the Protocol to Eliminate Illicit Trade in Tobacco Products. These key events, along with several outcome documents and policy declarations in the area of tobacco control and the wider development sphere, have introduced tobacco taxation into the consciousness of a much larger share of policy-makers. As detailed in subsequent chapters, numerous countries have imposed sufficiently high tobacco tax rates while applying best practices in tax policy design and imple- mentation over the past decade (8, 10). For example, sustained and substantial tax increases have reduced tobacco use in LMICs such as Brazil (14), Turkey (15) and the Philippines (16). High-income countries also continued their leadership in this area, comprising 23 of the 38 countries judged to have sufficiently high tobacco taxes in 2018 (8). However, much remains to be done. The 2019 WHO report on the global tobacco epidemic (RGTE) shows that tobacco taxes are still the most underutilized tobacco control policy among the MPOWER measures (8), with only 14% of the world’s population being covered by sufficiently high tobacco taxes. Substantial progress has also been made in building the tools and evidence base for tobacco taxation. Volume 14 of the International Agency for Research on Cancer (IARC) handbooks of cancer prevention, Effectiveness of tax and price policies for tobacco control, published in 2011, is a key review of the literature published as of May 2010 on the effectiveness of tax and price policies in reducing tobacco use. The National Cancer Institute (NCI)-WHO Monograph on the economics of tobacco and tobacco control, published in 2016, details the evidence accumulated over the years from various countries, focusing not only on tax and price policies, but on all aspects of the economics of tobacco and tobacco control. In addition, numerous published studies from LMICs provide a comprehensive picture of the impact of tobacco taxation in different contexts. The updated Appendix 3 of the Global NCD Action Plan explains the cost-effectiveness of tobacco taxation (11), while the Global NCD Business Plan, Saving lives, spending less, built on this work by estimating a dollar figure for the return on investment expected from implementing the best-buy interventions for tobacco control, including taxation (12). The past decade has seen major steps forward for tobacco taxation in terms of global commitments, the number of countries implementing best practices and the 1 The Addis Ababa Action Agenda of the Third Conference on Financing for Development. Third Inter- national Conference, 13-16 July 2015, Addis Ababa, Ethiopia (https://sustainabledevelopment.un.org/ content/documents/2051AAAA_Outcome.pdf, accessed 17 February 2021). 6 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N expansion of the evidence base on tobacco taxes, particularly in LMICs. Within the development sphere, institutions such as the World Bank, the International Monetary Fund (IMF) and many other multilateral agencies are aligned with WHO on the importance of tobacco taxation and the need to improve its implementation (17–18). Considerable challenges remain; although it appears that the world is headed in the right direction, progress needs to accelerate at a much quicker pace in order to achieve the SDG targets by 2030. SHAPING A “NEW NORMAL” FOR TOBACCO TAXATION The global upheaval caused by the COVID-19 pandemic has cast an unprecedented spotlight on how well governments around the world prepared for and responded to the crisis. It has exposed glaring health systems vulnerabilities and highlighted the struggles of many countries to control the spread of the virus. But perhaps more than anything, the pandemic has demonstrated how the economy, trade, science, politics and many other aspects of our societies are very much interdependent and interconnected with the health of the population. It is clear that an individual’s state of health can significantly determine their susceptibility to disease and their ability to overcome it. People with NCDs are more vulnerable to becoming severely ill with a number of conditions, which also appears to be the case with COVID-19 (19). Tobacco use is a major risk factor for NCDs, and available research suggests that smokers are at higher risk of developing severe illness and dying from COVID-19 (20). Just as the different aspects of society are interconnected, so too are people’s health, the existence of health-promoting environments and the government policies and agencies that shape these environ- ments. This critical moment presents a unique opportunity and renewed motivation to discourage the use of harmful products such as tobacco and to further improve tobacco control measures, especially tobacco tax policy. Moving forward, a business-as-usual approach to tobacco taxation will not be sufficient. Responding to this new reality and preparing for the next pandemic entails implementing measures that promote healthier populations. Like the COVID-19 pandemic, any future pandemic will likely exacerbate health inequities, bring about more economic uncertainty and put pressure on governments’ fiscal capacities. Interventions such as higher tobacco taxes, which protect people’s health while generating more revenues and economic benefits, become even more important in such crisis situations. Given this context and the stakes involved, ministries of finance and tax authorities are in a unique and powerful position – one of saving not only livelihoods, but also lives. The importance of increasing tobacco taxes – one of the most effective public health tools available – cannot be overstated. The traditional approach of treating CHAP T ER 1. WHY T HIS M ANUAL? 7 tobacco tax exclusively as a revenue source has no place in the new normal. One cannot deny the scale of the tobacco epidemic, the necessity to correct market failures and the overwhelming evidence of tobacco taxation’s benefits to health and to the economy. The positive trend in the changing narrative around tobacco taxation needs to continue. Tobacco taxation should not be viewed in isolation from the rest of government policies, but rather as an important part of the whole, an essential piece in working towards our common goal of better health for all. OVERVIEW OF SUBSTANTIVE CHAPTERS This manual is primarily designed for policy-makers, finance officials, tax authorities and customs officials. It may also be useful for officials within health ministries or other government agencies, as well as nongovernmental organizations working in this area, including tobacco control advocates. Significant effort is made to present real-world examples and recent experiences from a wide range of countries to demonstrate success stories and lessons learned in raising tobacco taxes. A sub- stantial amount of evidence has been generated in LMICs over the past few years that supports and augments the existing evidence base, providing a much broader body of knowledge than was available when the first WHO technical manual on tobacco tax administration was released. Chapter 2 delves into the theory, practice and empirical evidence on tobacco excise tax policy, including current global trends. The chapter offers a detailed analysis of the various elements that constitute tax structure, aiming to provide policy-makers with a comprehensive understanding of the factors affecting prices, consumption and the market. It describes the key components to keep in mind when designing tobacco tax policy to maximize the impact of tax increases and improve the tax structure. The chapter also includes updated global price and tax data, specific examples from various countries and a discussion of tax base elasticity, automatic excise tax adjustments and pricing regulations, as well as descriptions of new and emerging nicotine and tobacco products, including electronic nicotine- and non- nicotine delivery systems (ENDS/ENNDS) and heated tobacco products (HTPs). Chapter 3 focuses on tobacco tax administration. It provides an in-depth discus- sion of the fundamental components that make tobacco tax collection effective and efficient, ensuring achievement of the health and revenue objectives of tax policy. It highlights the importance of cooperation among the various agencies involved in the implementation of tobacco taxes within countries and across borders. Building on country and regional experiences from previous decades, the chapter outlines specific measures and recommendations to maintain oversight of the whole tax compliance cycle. Also included are actions to facilitate control and enforcement, such as licensing, fiscal markings (e.g. tax stamps), tracking and tracing systems 8 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N and import and export controls. Finally, the discussion pivots towards the broader elements of a good tax system, such as proper resourcing of competent authorities, a strong judiciary and strict rules regarding corruption. Chapter 4 deals with the important challenges in the area of political economy when countries attempt to increase tobacco taxes or simplify the tax structure. The tobacco industry often relies on identified patterns of argumentation and tactics to obstruct such reforms, i.e. SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric (regressivity), (R) revenue reduc- tion and (E) employment impact – each of which is discussed thoroughly in this chapter. Also included is a detailed discussion of the measurement of illicit trade and a discussion of earmarking tobacco tax revenue for health purposes. Finally, Chapter 5 presents a comprehensive list of the best practices in tobacco tax policy and administration discussed throughout this manual. The list is intended to serve as a practical guide and quick reference to the salient points presented. CHAP T ER 1. WHY T HIS M ANUAL? 9 REFERENCES 1. Ranson K, Jha P, Chaloupka FJ, Nguyen SN. The effectiveness and cost-effectiveness of price and other tobacco control policies. In: Jha P, Chaloupka FJ, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000:427–447 (https://www.paho.org/hq/dmdocuments/2010/ Cost-effectiveness%20of%20price%20increases.pdf, accessed 4 February 2021). 2. Effectiveness of tax and price policies for tobacco control. Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications.iarc.fr/Book-And-Report-Series/Iarc- Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And-Price-Policies-For-Tobacco-Control-2011, accessed 4 February 2021. 3. Tobacco taxation in the United States. In: Lynch BS, Bonnie RJ, editors. Institute of Medicine (US) Committee on Preventing Nicotine Addiction in Children and Youths. Growing up tobacco free: preventing nicotine addiction in children and youths. Washington (DC): National Academies Press (US); 1994 (https://www.ncbi.nlm.nih.gov/books/NBK236771/, accessed 10 November 2020). 4. Sin tax reform. Manila: Department of Finance (Philippines); 2012 (https://www.dof.gov.ph/advocacies/ sin-tax-reform/, accessed 10 November 2020). 5. Goodchild M, Perucic AM, Nargis N. Modelling the impact of raising tobacco taxes on public health and finance. Bull World Health Organ. 2016; 94:250–7 (https://www.who.int/bulletin/ volumes/94/4/15-164707.pdf, accessed 4 February 2021). 6. The economics of tobacco and tobacco control. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute, NIH Publication No. 16-CA-8029A; 2016 (National Cancer Institute tobacco control monograph 21; https://cancercontrol.cancer.gov/ brp/tcrb/monographs/monograph-21, accessed 17 December 2020). 7. WHO report on the global tobacco epidemic, 2008: the MPOWER package. Geneva: World Health Organization; 2008 (https://www.who.int/tobacco/mpower/mpower_report_full_2008.pdf, accessed 10 November 2020). 8. WHO report on the global tobacco epidemic, 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/tobacco/global_report/en/, accessed 10 November 2020). 9. Goodchild M, Nargis N, Tursan d’Espaignet E. Global economic cost of smoking-attributable diseases. Tob Control. 2018;27:58–64 (https://tobaccocontrol.bmj.com/content/27/1/58, accessed 4 February 2021). 10. Guidelines for implementation of Article 6 of the WHO FCTC. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 4 February 2021). 11. Tackling NCDs: ‘best buys’ and other recommended interventions for the prevention and control of noncommunicable diseases. Geneva: World Health Organization; 2017 (https://apps.who.int/iris/ handle/10665/259232, accessed 10 November 2020). 12. Saving lives, spending less: a strategic response to noncommunicable diseases. Geneva: World Health Organization; 2018 (https://www.who.int/publications/i/item/WHO-NMH-NVI-18.8, accessed 4 February 2021). 13. Chaloupka FJ, Warner KE. The economics of smoking. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics. Elsevier; 2000;1(1):1539–1627. 14. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Rev Panam Salud Publica. 2016;40(4):243–9 (https://iris.paho.org/bitstream/ handle/10665.2/31306/v40n4a09_243-9.pdf?sequence=1&isAllowed=y, accessed 17 February 2021). 15. Cetinkaya V, Marquez PV. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 16. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington (DC): World Bank Group; 2016 (http://documents.worldbank.org/curated/en/638391468480878595/pdf/106777-PUB-PUBLIC- PUBDATE-7-26-2016.pdf, accessed 10 November 2020). 17. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 10 November 2020). 10 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 18. Irwin A, Marquez P, Jha P, Peto R, Moreno-Dodson B, Goodchild M, et al. Tobacco tax reform: at the crossroads of health and development – a multisectoral perspective. Washington (DC): World Bank Group; (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report- CrossRoads.pdf, accessed 2 February 2021). 19. Information note: COVID-19 and NCDs. Geneva: World Health Organization; 23 March 2020 (https:// www.who.int/publications/m/item/covid-19-and-ncds, accessed 10 November 2020). 20. WHO statement: tobacco use and COVID-19. Geneva: World Health Organization; 11 May 2020 (https:// www.who.int/news-room/detail/11-05-2020-who-statement-tobacco-use-and-covid-19, accessed 10 November 2020). 11 CHAPTER 2. Tobacco excise tax policy 2.1. GLOBAL OVERVIEW OF TOBACCO TAX PRACTICES A well-designed tax policy is key to having an effective tax policy. Any government that is planning to reform its tax policy must first understand the fundamental components of a good tax policy, as well as consider the strengths and weaknesses of different approaches to taxation, how they impact price and the requirements for tax administration. Understanding how the tobacco market operates in a country is equally important for policy-makers because of the inevitable interaction between the market and tax structures. Beyond the political considerations that strongly influence tobacco tax policy development, this chapter focuses on the technical aspects of tobacco taxation – excise tax in particular. Section 2.1 provides an overview of tobacco tax practices at the global level, focusing on the different ways countries structure excise tax. Section 2.2 emphasizes the importance of carefully designing excise tax policy, highlighting not only the significance of tax increases but also excise tax structure and its impact on prices, taking into account how market structure influences trends. This section also discusses the importance of measuring impact as another aspect of tax policy development, and it presents the crucial elements for performing measurement, as well as the relevant indicators available to monitor progress. Section 2.3 describes external policy considerations in the design phase to ensure that the goals of tobacco control and taxation are achieved. Intersectoral policy integration and coherence at the domestic level is discussed as a strategy to ensure that policies of other sectors do not inhibit or obstruct public health policy objectives. This section also reviews the current state of regional tax harmonization based on the experience of existing regional blocs and draws conclusions on the best policy approaches to preserve the public health interests of individual countries. Section 2.4 discusses new and emerging nicotine and tobacco products, in par- ticular HTPs and ENDS/ENNDS. It reviews the latest evidence on the health impacts of these products and current approaches to regulation. Key policy considerations are identified, and recommendations are provided for adopting an appropriate excise tax policy for these products. Section 2.5 summarizes the issues covered in the chapter and the key takeaways. 12 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.1.1 TAXES APPLIED ON TOBACCO PRODUCTS Taxes are classified as either direct or indirect. Direct taxes are imposed on the profit, income, property or wealth of persons or companies, whereas indirect taxes are imposed on the price of goods and services. Indirect taxes are most relevant to tobacco products taxation, because they directly influence price. A variety of types of indirect taxes can be applied to tobacco products. These include: • excise taxes – taxes that apply to a few selected commodities (they can also be applied to alcohol, fuel, sugar-sweetened beverages, etc.). • VAT or sales taxes – VAT is a multistage tax on all consumer goods and services that is applied proportionally to the price the consumer pays for a product. It is a tax on the amount by which the value of an article has been increased at each stage of its production or distribution. Some countries impose sales taxes instead of VAT. Unlike VAT, which is collected at every stage of the supply chain, sales taxes are generally levied at the point of retail on the total value of goods and services purchased. Ultimately, the consumer ends up paying the tax, whether it is a VAT or a sales tax. • import duties – taxes on selected goods imported into a country to be consumed in that country (i.e. goods that are not in transit to another country). In general, import duties are collected from the importer at the point of entry into the country. • other taxes – other indirect taxes, such as environmental taxes, that do not fall into any of the categories listed above. One of the most well-established and widely understood points in tax policy is that tobacco products should be subject to excise taxation. The focus of this chapter – and of this manual overall – is on excise taxes. They are the most important type of indirect taxes for tobacco control because they are applied directly to tobacco products and contribute the most to increasing the price of tobacco products relative to other goods and, subsequently, to reducing consumption. There are two basic types of excise taxes: • specific – levied as a monetary value per quantity of the product being taxed (e.g. 1 000 cigarettes, pack of 20 sticks, kilogram of tobacco); and • ad valorem – levied as a percentage of the value (e.g. retail price, or the producer/ex-factory price or the cost, insurance and freight [CIF] value1) of the product being taxed. These types of excise tax can be applied at a uniform or a differential (tiered) rate and on their own or in combination (i.e. a mixed system). 1 CIF is the value of an imported product as declared to customs upon entry into a territory. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 13 Given the widespread use of cigarettes – almost the only tobacco product used in some parts of the world – and the scarce availability of data for other tobacco products, this chapter focuses mainly on cigarettes. But there are a few examples and recommendations for other tobacco products, including those that are more prevalent in specific parts of the world (e.g. bidis or smokeless tobacco in South-East Asia and waterpipe tobacco in the Eastern Mediterranean region). 2.1.2 CIGARETTE TAXES AND RECENT TRENDS WORLDWIDE Tax and national income levels: the higher the income level, the higher the taxes and prices At the global level, cigarette price and tax levels correlate positively with a country’s income level: prices and taxes are higher in higher-income countries and lower as income level decreases. This trend has not changed over the years since 2008. Figure 2.1 presents the levels of price and tax by income groups for 2018, using the World Bank classification of income groups. Fig. 2.1 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by income group, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in purchasing power parity (PPP) adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Excise tax per pack Other taxes Retail price minus taxes Pr ic es a nd ta xa tio n pe r p ac k (P PP $ ) High-income Low-income 4.25 0.68 Middle-income 2.06 Total taxes: 5.30 (67.9% of pack price) Total taxes: 2.91 (58.3% of pack price) Total taxes: 1.18 (38.1% of pack price) PPP $ 7.80 PPP $ 4.99 PPP $ 3.09 Global 2.48 PPP $ 5.53 Total taxes: 3.36 (60.8% of pack price) 0 1 2 3 4 5 6 7 8 14 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tax and price levels: the higher the tax share, the higher the price Globally, cigarette prices correlate positively with tax percentage levels: as the total tax share (of which excise represents the largest part) as a percentage of retail price increases, the price of cigarettes generally also increases (see Fig. 2.2 below).2 This indicates that taxes do influence prices. Fig. 2.2 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by total tax levels, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Tax level and WHO regional classification: taxes and prices are highest in the European region, followed by South-East Asia, the Americas and the Western Pacific, with the lowest levels in the Eastern Mediterranean and African regions At the regional level (WHO regional classification), average levels of prices and taxes vary greatly. The highest level can be seen in the European region, which includes the European Union (EU) countries. The EU’s unified tax structure includes high levels of minimum taxes – which lead to high prices – and encourage member 2 This is a general trend and does not apply for every country; there are countries that have a large tax share but low prices for cigarettes. Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax ≥ 75% Total tax ≤ 25% 0 7 4 1 2 3 6 5 50% ≤ Total tax < 75% 25% ≤ Total tax < 50% Excise tax Other taxes Retail price minus taxes 4.44 2.24 1.57 PPP $ 5.07 PPP $ 5.33 PPP $ 7.07 PPP $ 2.60 0.17 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 15 countries to regularly increase their taxes to meet their obligations. However, with the current minimum level now being reached by all EU member countries, the motivation to increase excise taxes may wane. Indeed, the minimum cigarette excise amount in the EU has not been adjusted since it went into effect on 1 January 2014, and it is suffering from inflation erosion. Member States of the EU acknowledged this in June 2020 by stating that action at the EU level is required to ensure that minimum excise duty rates regain traction to effectively reduce the consumption of tobacco products and that the minimum rates of excise duties on a number of tobacco products would be increased (2). Excise taxes are lowest in the African and Eastern Mediterranean regions. And China – reported separately due to its size – has lower tax rates than the Western Pacific region (see Fig. 2.3). Fig. 2.3 Weighted average retail prices and taxation (excise and total) of most-sold brand of cigarettes, by region, 2018 Notes: China is represented separately from the Western Pacific Regional Office (WPRO) average because of its exceptionally large number of smokers compared with the number in other countries in the region. AFRO is the African Region, AMRO is the Region of the Americas, EMRO is the Eastern Mediterranean Region, EURO is the European Region, SEARO is the South-East Asia Region. Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Retail price Excise tax Total tax share % AFRO AMRO EMRO EURO SEARO WPRO w/o China China All 3. 80 4. 02 4. 24 4. 89 4. 02 1. 45 2. 89 2. 65 2. 48 7. 53 7. 27 5. 61 2. 51 1. 95 1. 05 42.2% 55.8% 61.2% 72.9% 63.2% 55.7% 60.8% 56.9% Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax share % 5. 53 0 1 2 3 4 5 6 7 8 9 10 40 30 50 60 70 80 20 16 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Global tax structures trend: more countries are adopting specific excise taxes or mixed excise systems that rely more on the specific component Since 2008, the number of countries that rely solely on ad valorem taxes or apply no excise taxes at all has decreased as more countries have adopted specific or mixed systems. More of the countries that have implemented a mixed system have increased the specific component of the tax structure relative to the ad valorem component (see Figs. 2.4 and 2.5).3 Fig. 2.4 Changes in excise tax structure, 2008–2018 Fig. 2.5 Changes in reliance on specific versus ad valorem component in mixed systems, 2008–2018 Source: (1). 3 For information about countries that applied each type of excise tax structure in 2018, see Annex 2.1. N um be r o f c ou nt ri es Specic excise Ad valorem Mixed excise No excise 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 54 56 57 63 50 45 24 23 24 21 19 15 49 47 44 42 41 55 57 59 56 60 63 62 Mixed excise Relying more on specic Relying more on ad valorem N um be r o f c ou nt ri es 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 45 22 23 23 24 22 26 27 27 27 32 35 37 50 54 56 57 63 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 17 The imposition of a minimum specific excise tax: half of the countries that have a mixed or ad valorem structure impose an excise tax floor A minimum specific excise tax ensures that at least a certain minimum amount of tax is paid, irrespective of price level. Almost half of the 101 countries that impose either ad valorem or a mixed excise for which data on minimum excise are available (47 countries) set a minimum specific excise tax.4 Nearly two thirds of those that set a minimum specific excise tax (29 countries) are high-income countries; most of them are in the EU, which requires its members to impose a minimum specific excise tax. The choice of tax base worldwide: almost half of the countries that apply a mixed or ad valorem excise system use retail price as the base Setting the base for applying a specific excise is relatively easy: most countries use a defined quantity of sticks for cigarettes, the weight in kilograms for tobacco and the weight in grams for other tobacco products (1). Different bases for ad valorem excises are applied in different countries. Nearly half of the 105 countries that implement either ad valorem or a mixed excise for which data are available (47 countries) use the retail price5 as the tax base for the ad valorem part, and most of those (28 countries) are high-income countries. Using the retail price as the base for the excise ad valorem tax is more effective than using the producer price or the CIF value. Unlike retail prices, which are easy for tax administrators to ascertain by monitoring the market, the producer price or CIF value is prone to undervaluation by producers or importers, who may pass on their margins to related parties further down the supply chain and successfully reduce their tax burden. This tactic is also known as transfer pricing. Additionally, global-level data show that the excise ad valorem on the retail price seems to lead to higher retail prices on average compared with an excise ad valorem applied on other bases, such as the producer price or CIF value (see Fig. 2.8 below). On complex tiered structures: 31 countries still apply complex, multitiered excise taxes on tobacco products As of 2018, 31 countries imposed excise taxes that varied according to defined char- acteristics of cigarettes, including price level, type of production, type of package and length of cigarette (Table 2.1). Some countries use more than one criterion to differen- tiate the tax rates. Indonesia, for example, imposes differential rates based on volume 4 This means that countries with a mixed system impose an overall minimum specific excise tax (where the yield of the specific plus the ad valorem excise cannot be below the set minimum specific excise tax), in addition to the excise on a specific component. 5 Countries that impose ad valorem on retail price exclusive of VAT are also included, since retail prices are easy to determine and VAT rates are known variables. 18 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N produced, type of cigarette and price level. In other countries, such as Member States of the EU, differential rates for cigarettes are prohibited by law, and the rate of the ad va- lorem tax and the amount of specific excise duty must be the same for all cigarettes (3). Table 2.1 Criteria used by countries for tiered excise taxes, 2018 BASE OF TIERS COUNTRY Retail price Bangladesh, Belarus, Indonesia, Jordan, Mozambique, Myanmar, Pakistan, Thailand Cigarette grade (e.g. premium, mid-grade, economy) Egypt, Japana, Mali Producer price China, Lao People’s Democratic Republic Production volume Indonesia Type filter/non-filter Belarus, Georgiaa, India, Kenya, Republic of Moldova, Nepal, Papua New Guinea hand/machine made India, Indonesia kretek/white cigarette Indonesia tobacco content (dark/ blonde or dark/light) Algeria, Bolivia (Plurinational State of ) Packaging soft/hard Mozambique, Uganda Cigarette length India, Nepal, Sri Lanka Trade (domestic/imported) Iran (Islamic Republic of ), Lebanon, Myanmar, Solomon Islands, Tonga, Uzbekistan Leaf content (domestic/imported) Fiji, United Republic of Tanzania a Japan and Georgia were using a tiered excise tax structure when these data were collected in 2018, but as of 2020, that is no longer the case. Source: (1). 2.2 DESIGNING EXCISE TAX POLICY Significantly increasing the taxes on and prices of tobacco products is the most effective and most cost-effective policy to control tobacco use (4). Increased taxes that are passed on to tobacco users as higher prices reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face challenges ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues such as addressing the SCARE6 tactics of the tobacco industry. This section provides guidance for policy-makers regarding the best tax structure to use from a health perspective, taking into consideration all the appropriate tax designs. It also proposes recommended indicators to consider when formulating policy change. 6 SCARE tactics are the tactics most commonly used by the tobacco industry when countries plan to increase tobacco taxes. They are described, and refuted, in detail in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 19 2.2.1 THE IMPORTANCE OF THE TYPE OF EXCISE TAX STRUCTURE The existing theoretical and empirical evidence on approaches to the choice of (uniform) specific and ad valorem excises is reviewed below, along with their effects on price, consumption, perceived quality and variety of tobacco products, govern- ment revenue and tax administration. The use of the word “quality” in this chapter does not refer in any way to the health impact of a tobacco product. It refers rather to the consumers’ perceptions of quality and their decision to buy a product, which they may evaluate based on the packaging, the blend used for the cigarette or anything that makes the product more appealing to them. Just to be clear, from a public health perspective, all cigarettes are equally harmful even if perceived by consumers as having higher or lower quality. The choice between ad valorem and specific taxation is influenced by the market structure, i.e. the nature and degree of competition in the market for goods and services. Although each country has its own specific characteristics, the tobacco market structure is typically a monopoly or an oligopoly where firms have the power to control prices – and hence exploit the tax structure – to their benefit. For example, China, the largest producer and consumer of tobacco products in the world (5), has a state monopoly. In Viet Nam, foreign brands are produced under licence by the state monopoly. In Thailand and Egypt, despite the presence of foreign companies, the market is dominated by the state-owned company. In Uruguay, the oligopoly is led by a domestically owned company. In Bangladesh, the oligopoly consists of domestically owned companies competing with foreign companies (6). In most of Africa, the market consists of transnational tobacco companies (7). The impact of tax structure on final price: uniform specific versus uniform ad valorem The choice between specific and ad valorem taxes is a long-standing issue in tax policy, as the level and structure of excises have different implications for the interests and goals of various groups. Given the market structure of the tobacco industry – typically a monopoly or oligopoly for most products in most countries – different excises may have different effects on government revenue, manufacturer profit, consumer price, perceived product quality and variety and tax administration (8–16). Consequently, the two types of excise taxes – specific and ad valorem – may have different implications for public health to the extent that they affect individual consumption via their impact on perceived product quality, variety and prices. Moreover, governments have the potential to influence tobacco excises to manage demand, raise revenue and promote public health. 20 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax on prices and in turn on consumption is also influenced by price and income elasticities, as well as consumer perceptions of quality (“perceived quality”) and the variety of available products, which, in turn, are closely linked to the type of tax structure adopted. Tax structure is affected by both the price elasticity of demand and the price elasticity of supply. The price elasticity of demand measures the responsiveness of consumer demand to changes in prices. The price elasticity of supply measures how sensitive producers are to changes in prices. Tobacco tax structure is also influenced by market structure. In a monopoly, the profit-maximizing firm sets the price, considering the price elasticity of demand: the lower the price elasticity (in absolute value) – i.e. the less sensitive the consumer is to price changes – the higher the price the monopolist can set. Profits are typically abnormal in a monopolist market structure, meaning total sales revenue is higher than total cost (where total cost includes a normal profit). A monopolist producer therefore receives more than the minimum reward required to invest its (physical and human) capital and undertake business risks. Economic theory predicts that in a private monopoly, prices are higher than in an oligopolistic market. This is not, however, necessarily true when the monopoly is owned by the state and the government’s objective is not straightforward profit maximization: the government might have other considerations, such as preserving jobs (e.g. in China) or keeping prices low for low-income consumers (e.g. in Egypt). Under a monopoly, an ad valorem taxation structure enables the monopolist producer to set prices lower than would be possible under a specific tax structure. This is feasible because under ad valorem taxation, when supply increases and price falls, the price reduction is not fully borne by the producer. Rather, the price reduc- tion is partly shared by the government since, as supply increases, the tax per unit of product sold falls. In other words, ad valorem taxation leads to lower prices and higher consumption relative to revenue-equivalent specific taxation. Technically, this means that the supply function is less elastic under ad valorem taxation. In contrast, under a specific taxation structure, any increase in the monopolist producer price will go to the producers as revenue, which incentivizes them to increase prices. The same logic also applies to an oligopolistic market structure, where profits again are, in general, abnormal. KEY TAKEAWAY 1 In a monopoly or an oligopoly, specific taxation incentivizes industry to set prices higher than it would with ad valorem taxation. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 21 Understanding oligopolies, however, is more complicated, since they are characterized by strategic interdependence among a few firms. This strategic interdependence extends to the relationship between industry and regulators. Industry anticipates the government’s regulatory policy – whether through tax or other interventions – and acts accordingly. For example, competitors may coordinate and lobby against a certain tax structure reform or tax rate increase. Under an oligopoly market structure, ad valorem taxation is a relatively more efficient tool for transferring part of the profits to the government as tax revenue, since it acts like both an excise and a profit tax. In contrast, a specific tax has a smaller (negative) effect on profits. This explains why we observe multinationals that are leaders in high-priced brands (e.g. Philip Morris International [PMI]) lobbying in favour of specific taxation (17). As an example, in the countries of the Cooperation Council for the Arab States of the Gulf (GCC), the tobacco industry has been trying for a long time to lobby governments to introduce a specific excise (18–21). After years of consideration and discussions on the possible introduction of excise taxes, the GCC adopted the Common Excise Tax Agreement of the States of the Gulf Cooperation Council in November 2016 (21), which introduced an ad valorem excise on tobacco products. Tobacco companies’ support for excise tax structures ultimately depends on the market segments they control in a particular country. A company selling mainly premium brands will favour specific excises, whereas a company that sells mid-priced or economy brands would favour ad valorem excise (17). When oligopolistic firms produce identical products, a specific tax has a stronger positive effect on price and is more likely to be overshifted to consumer prices than an ad valorem tax (13). Overshifting means that the price increases by more than the tax increase itself. Empirical evidence supports this (22–26). KEY TAKEAWAY 2 In an oligopoly, prices are likely to increase by more than the amount of the specific tax increase when demand is relatively inelastic. In general, demand for a product depends not only on prices but also on consumer perceptions of quality and preferences for variety. For example, the most popular brand in GCC countries is Marlboro, a premium brand (1). Consumers differ in their willingness to pay, depending on their respective perceptions of quality, which influence whether they ultimately purchase high- or low-priced brands. A tax-induced price increase can cause the following plausible responses from consumers or users of tobacco products: (1) a group of consum- ers will quit; (2) a group of consumers will reduce their overall consumption; 22 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N (3) another group, most likely high-income users, will switch to an upgraded version of the tobacco product if its relative price (compared to the cheaper brand) has been reduced, which is the case under a specific tax increase; there might also be a group of consumers in the lower income range who switch to lower-priced variants of the tobacco product if the price gap increases, as is the case under an ad valorem tax increase; and (4) another group might switch to the illegal market or buy products in a neighbouring country with a lower tax where possible. Consumers’ decisions to purchase are also affected by their preferences for variety – meaning preferences among products that consumers perceive as equal in quality but are given different characteristics by the producers to account for consumer taste preferences. Thus, it is possible that a tax increase that leads to an increase in average prices will lead to an increase in the total quantity demanded in the market because of an increase in the variety of product choices available to consumers. Variety enables new consumers to be captured, especially in an environment lacking certain regulations (e.g. without plain packaging and flavour bans). The tobacco industry was able to capture a new group of consumers when it introduced menthol cigarettes into the tobacco market. There is more than sufficient evidence that menthol cigarettes increased youth smoking initiation, increased nicotine dependence and reduced adult smoking cessation (27). To prevent this from happening in their countries, Member States of the EU have prohibited characterizing flavours other than tobacco in tobacco products (28). It is therefore important to consider the broader effects that the structure and level of an excise tax can have on average price, perceived quality and the variety of cigarette brands and other emerging substitutes. When consumers make choices based on dimensions other than quantity, the two types of tax structures are not equivalent, even in a perfectly competitive market where firms have no market power (29–30). To illustrate this point, consider a US$ 1 cost to improve consumer perceptions of quality for a tobacco product. This will lead to an equivalent price increase under specific taxation but not under ad valorem taxation. At an ad valorem rate of 20%, the price must increase by more than US$ 1, or by 1/(1 – 0.2) to cover the US$ 1 cost of improvement, due to the multiplier effect. A specific tax induces consumers to reduce the quantity demanded, but they might still choose to pay a higher price in exchange for a product that they perceive to be of better quality. An ad valorem tax, on the other hand, leads to a reduction in both quantity and perceived quality, not a substitution between them. An ad valorem tax has only an income effect and – unlike specific taxation – does not lead to substitution between perceived quality and quantity. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 23 KEY TAKEAWAY 3 Under specific taxation, the industry has incentives to create upgraded variants of tobacco products that attract new consumers and encourage consumption. When firms produce differentiated products, as the tobacco industry does, economic theory provides ambiguous results regarding specific versus ad valorem taxation. The relative effects of the two types of tax are not as straightforward as in the case of oligopolistic firms producing a homogeneous product. With differentiated products, the relative effects of the tax types depend on various assumptions: whether or not firms face symmetric costs, whether the number of firms is fixed or new firms can enter the market and the level of the tax revenue requirement. When firms face different costs, ad valorem taxes exacerbate the absolute differences in marginal costs between them. The high-cost brand is not considered a perfect substitute for the low-cost brand. A sufficiently high ad valorem tax rate may lead to a relative underproduction of the high-cost products (31). The effect of specific and ad va- lorem taxes on consumer perceptions of quality depends on market structure and the price and income elasticities of demand across various qualities. The relative price of the cheapest product does not necessarily remain unchanged or increase; it might fall (32). Empirical evidence showing that increases in the specific tax lead to a lower market share for the cheaper generic brands and an upward shift to premium brands (33–34) usually considers gradual tax increases and ignores income effects. Chaloupka et al. (35) found that in 21 EU countries that impose a mixed tax system, the price gap between premium and low-priced brands – while not reflecting the full distribution of cigarette prices – is smaller when the specific component of the mixed structure dominates.7 Although the price gaps are narrower under specific taxation, there is evidence that firms sometimes respond by introducing new, very cheap (subvalue) brands, or they exercise differential tax shifting. This practice has been evident in India for quite some time, with the Indian Tobacco Company launching a number of cheaper variants of its flagship cigarette brand, Gold Flake, to take advantage of a lower excise tax rate in the so-called microcigarette (< 60 mm length) market (36). Consequently, the cheapest end of India’s cigarette market has expanded significantly in recent years due in part to the marketing of new brand variants like Gold Flake Century. 7 The EU countries impose a mixed tax structure with a minimum tax floor. Some countries rely on the specific component more than others, but they remain within a given range (the specific component must be between 5% and 76.5% of total tax share of the weighted average price). 24 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As another example, transnational tobacco companies, which have sold ultra- low-priced brands in the United Kingdom since 2006, have managed to double their market share in a few years: their real price did not increase, since they absorbed part of the tax increases (37). The share of ultra-low-priced brands increased between 2001 and 2009 from 5% to 10%, while the market share of economy brands increased from 40% to 50% and the market share of premium brands and mid-priced brands decreased during the same period (from 35% to less than 25% for the former and from 15% to 5% for the latter). In order to keep the price of discount brands low and certain consumers in the market, firms may overshift the tax for premium products and undershift it for the lower-priced products (37–40). KEY TAKEAWAY 4 Evidence suggests that the price gap between brands is narrower under a specific tax structure. As the tobacco industry simultaneously consolidates producers and widens its portfolio of products, evidence is emerging that it is introducing cheaper brands while increasing the price of its expensive brands, therefore paradoxically widening the price gap within its products. The extent of the impact is still unclear, however, and this evidence does not negate the overall conclusion that a specific tax structure reduces price gaps. The impact of tax structure on final price: uniform specific, ad valorem and mixed systems Evidence from the 2019 RGTE (1) data suggests that the average price of the most- sold brand of cigarettes – weighted by the number of smokers – is the highest in countries implementing a mixed system that relies more on specific excise, followed by countries applying specific excise taxes only, followed by countries applying a mixed system that relies more on ad valorem and then by countries that apply ad valorem excise only (Fig. 2.6). The price is lowest in countries that have no excise at all. In past WHO reports on the global tobacco epidemic, countries that applied specific excise only had the highest price, on average. The trend may have changed partly because more countries are adopting mixed excise systems. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 25 Fig. 2.6 Weighted average price and excise for a pack of the most-sold brand of cigarettes, by excise tax structure, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Shang et al. (41) describe and compare price distributions, using data from 16 countries of the International Tobacco Control (ITC) Project that impose different cigarette tax structures. Specific uniform taxation tends to result in less variability in prices than all other structures (ad valorem tax, mixed tax, tiered tax). In general, structures other than uniform specific tax give rise to more opportunities for brand switching and tax avoidance. Reliance on complicated systems is likely to be as- sociated with wider price distribution, leading to greater tax avoidance, as there are more opportunities for substitution with cheaper brands when taxes rise. KEY TAKEAWAY 5 Evidence suggests that the tax structures most likely to lead to higher prices are uniform specific excise tax structures or mixed systems that rely more on specific excises. Mixed system relying more on specic excise Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Mixed system relying more on ad valorem excise Ad valorem excise No excise 2.21 Specic excise 2.62 3.26 0 7 6 5 4 3 2 1 1.66 Excise tax Other taxes Retail price minus taxes PPP $ 7.56 PPP $ 5.66 PPP $ 4.84 PPP $ 3.73 PPP $ 2.52 26 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax structure on final price: uniform excise versus tiered tax systems Another aspect of tax structure that impacts final price is the use of tiered taxation, i.e. tax rates that vary according to product characteristics. The characteristics can vary, from price level to the type of tobacco leaf contained in the cigarette, the size of production volume, the packaging, etc. Table 2.1 (earlier) lists the criteria used by 31 countries as the basis for different tax rates. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax (see Fig. 2.7). Fig. 2.7 Weighted average price and excise for a pack of the most-sold brand of cigarettes for countries with and without tiered taxation, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Uniform specific tax structures are likely to lead to relatively higher prices with less variability in price distribution. Compared with tiered tax structures that have differential rates based on brand characteristics, uniform taxation may reduce consumers’ incentive to switch to cheaper brands (leading to higher quit rates and lower prevalence), as well as decreasing manufacturers’ incentive to reduce their tax liabilities by changing their pricing strategies, production process or size (42–45). In Indonesia, for example, where small producers were taxed more fa- vourably, manufacturers had an incentive to reduce their scale of production but increase the number of affiliated small companies. The issue was resolved when tax authorities considered the aggregate production of all affiliated companies in the application of differential tax rates. By 2017, there were 786 active factories, while Excise uniform Price and taxation per pack of 20 sticks (PPP $) Excise tiers 3.28 1.94 Excise tax Other taxes Retail price minus taxes PPP $ 4.99 PPP $ 6.30 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 27 there had been 4 198 factories in 2006.8 Abolishing the differential tax rates would have been even more effective in removing the incentive for tax avoidance, as well as helping improve both public health and government finances. KEY TAKEAWAY 6 Evidence suggests that applying a uniform excise tax on cigarettes is not only easier to administer than tiered systems but also more likely to lead to higher cigarette prices. The impact of tax structure on final price: the significance of the choice of the tax base It is important for the excise tax to be applied to the base that leads to the greatest possible effect on price and revenue. For specific taxation, the tax base is the quantity of tobacco products. The quantity of cigarettes, cigars and bidis is measured in number of sticks; for other tobacco products, such as smokeless tobacco or roll-your-own (RYO), it is measured in the weight of the tobacco. When the tax is ad valorem, the choice of the tax base is important not only for health considerations – due to its effect on consumption – but also for tax revenue generation and industry profits. An ad valorem tax that is based on the ex-factory price (or CIF value) provides tobacco manufacturers with opportunities to reduce their tax liability, especially when they control the distribution system. Tobacco producers may sell cigarettes to distributors who are related parties at a reduced price, which then serves as the basis for calculating their ad valorem tax liability. Distributors, however, can then set high prices and share the extra profit with the producers (46). Because of the potential for such trade mispricing, the best practice is to use the retail price as the tax base and introduce a minimum excise tax per pack. Data in the 2019 WHO RGTE (1) show that, on average, the price level of a pack of cigarettes and the excise level are both much higher in countries that use retail price as the base for their ad valorem excise (Fig. 2.8). The maximum retail sales price, which includes all taxes, is used as the ad valorem tax base in the EU. That price also forms the tax base for ad valorem taxes in a growing number of LMICs, including Brazil, Egypt, Thailand, Turkey and Rwanda. 8 Indonesian Ministry of Finance, personal communication, 2017. 28 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.8 Weighted average price of the most-sold brand of cigarettes in countries that use retail price as the base for their ad valorem excise, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 7 The base on which the excise is applied is important. For specific excise, the base needs to be clearly defined (for cigarettes, cigars and bidis, it is the number of sticks; for other tobacco products, such as smokeless tobacco or RYO, it is the weight of tobacco). For ad valorem excise – where the base is typically either retail price, CIF value or producer price – evidence suggests that countries that apply the excise tax on the retail price of cigarettes tend to have higher prices than those that apply the tax on other bases. CIF and producer prices are difficult for government authorities to ascertain and are prone to undervaluation. The tax impact on final price: the significance of the minimum excise tax The use of a minimum excise tax in countries with ad valorem or mixed systems is another important factor in determining final price. On average, the price of a pack of cigarettes – as well as the excise level – is much higher in countries that impose a minimum specific excise than in those that do not (see Fig. 2.9). While more than half of the 47 countries that apply a minimum excise are members of the EU, removing EU countries from the average calculations produces the same conclusions. Ad valorem/mixed with retail price as base 3.72 PPP $ 5.01 PPP $ 6.41 Ad valorem/mixed with other base 1.79 Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 29 Fig. 2.9 Weighted average price of the most-sold brand of cigarettes in countries with and without a minimum specific excise tax, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 8 Among countries that apply an ad valorem or mixed excise tax on cigarettes, evidence suggests that those that impose a minimum specific excise tax tend to have higher prices than those that do not. The minimum excise tax also helps guarantee minimum excise revenues. Summarizing the advantages, disadvantages and impacts of the choice of excise tax structure for tobacco products Table 2.2 summarizes the characteristics of different types of tobacco excise taxes and the advantages and disadvantages of each type in relation to its impact on quantity demanded, perceived quality of brands offered, price, certainty and stability of revenue, administration and enforcement and opportunities for tax avoidance and tax evasion as they are predicted by the economic theory of imperfect competition and observed in real life. Ad valorem/mixed with minimum speci c 1.80 4.23 Ad valorem/mixed without minimum speci c Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes PPP $ 4.82 PPP $ 7.41 30 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 2.2 Characteristics of different types of tobacco excise taxes Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) TA X B A SE The unit of product (e.g. 1 000 cigarettes) The value of the product (e.g. retail, wholesale or manufacturer price) Excise is calculated on an ad valorem basis; however, if the calculated tax falls below a specified minimum amount, a specific tax rate applies Unit and value of product Both unit and value, unless the calculated tax falls below a specified minimum, in which case the tax base is the unit A D M IN IS TR AT IV E R EQ U IR EM EN TS The tax should be collected at the point of manufacturing or at the time of importation Low, as only the volume of the products needs to be ascertained Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as with a pure ad valorem regime Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises, as well as minimum specific excise tax compliance U N D ER VA LU AT IO N Not an issue Susceptible to undervaluation Provides an easy tool to prevent undervaluation of low-priced brands subject to the minimum specific excise The ad valorem part of the excise collection may be susceptible to undervalua- tion, depending on the choice of tax base The minimum specific excise prevents possible ad valorem tax base undervaluation of low-priced brands CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 31 Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IM PA C T O N P ER C EI V ED PR O D U C T Q U A LI TY Upgrading effect tends to reduce the relative tax on higher-priced brands Multiplier effect provides a disincentive to costly so-called quality improvement No incentive to upgrade higher- priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands IM PA C T O N P R IC E Tends to lead to relatively higher prices, particularly for low-priced cigarettes Tends to lead to relatively lower prices; price reductions will be subsidized if the multiplier effect is strong Tends to lead to relatively higher price increases for low-priced cigarettes An increase in the specific tax will to lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem payment if the base of the ad valorem includes excise An increase in the specific tax will lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem tax amount if the base of the ad valorem includes excise. Increases in the ad valorem and /or specific tax will raise the minimum tax paid if the minimum is a percentage of the total tax on, for example, weighted average price; they will reduce price gaps, given impact on perceived quality 32 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IN FL AT IO N The real value of the excise tax will be eroded unless the tax is adjusted in line with inflation The real value of the excise tax will be preserved as prices increase, at least to the extent that tobacco product prices follow inflation The real value of the minimum specific excise will be eroded over time unless the excise is adjusted in line with inflation The real value of the specific excise will be eroded unless the excise is adjusted in line with inflation The real value of the specific excise and the minimum specific excise will be eroded unless the excises are adjusted in line with inflation H EA LT H B EN EF IT S Will discourage consumption of tobacco products irrespective of the price band May encourage more trading down in favour of cheaper cigarettes, reducing the health benefit The minimum specific excise reduces incentives for trading down May reduce trading down Reduces trading down Source: (47). 2.2.2 OTHER TAX DESIGN CONSIDERATIONS The significance of automatic adjustments and indexation of specific tax to inflation Specific taxation does not depend on price and therefore, unlike the ad valorem tax, is not automatically adjusted for inflation. The real value of a specific tax is eroded over time as the price of the taxed product increases. Therefore, especially in countries with rapid growth in inflation, the nominal value of the specific tax must be increased regularly in order for the tax to maintain its real value. This is of great importance for both public health and public revenues, especially in countries where manufacturers do not increase prices regularly and/or low-priced tobacco products are the dominant products in the market. Table 2.3 lists countries that include automatic adjustments to their excise in order to avoid the erosion of the specific excise over time, using different units of adjustment and based on different frequencies. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 33 Table 2.3 Countries that include automatic adjustments to the specific excise COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Argentina Inflation (consumer price index [ CPI]), on a quarterly basis Armenia Minimum specific excise set to increase in the Tax Code by 15% on average between 2019 and 2021 Australia Wages – excise rates on tobacco and tobacco products increase in March and September each year, based on average weekly ordinary time earnings Bosnia and Herzegovina Specific excise rate is increased annually by at least 7.50 convertible marks per 1 000 cigarettes; minimum excise tax is increased annually to be at least 60% of the weighted average price Canada Inflation – federal tobacco tax rates are to be increased every five years, indexed to Canada’s CPI starting in 2019 Chile Inflation Colombia Specific tax set to 1 400 pesos, increased to 2100 pesos in 2018; starting in 2019, it will increase yearly by the CPI plus 4 points Costa Rica Inflation Dominican Republic Inflation, on a quarterly basis France Increase from 2017 to reach an average price for cigarettes of €10 per pack by 2020 Honduras Inflation, annually to December of the previous year Italy Minimum tax burden calculated every year in March on the basis of the weighted average price of cigarettes sold in the previous year New Zealand Inflation annually plus 10% annually from 2017 to 2020 Nicaragua Updated annually as of 1 January 2017, taking the highest among the annual devaluation of the official exchange rate of Cordoba with respect to the US dollar, published by the Central Bank of Nicaragua, and the annual inflation rate of the CPI published by the National Development Information Institute, observed in the last 12 months available North Macedonia Specific and minimum specific rate increase by 0.2 denars per cigarette on 1 July each year until 2023 Philippines Agreed tax increases and rates for specific excise tax between 2020 and 2023, with a 5% indexation thereafter Romania Inflation, annual (1 January) adjustment of the total excise according to inflation calculated on 1 October of the previous year Serbia Inflation, every six months Southern African Customs Union (SACU) – Botswana, Eswatini, Lesotho, Namibia and South Africa Inflation, on an annual basis9 Sweden Inflation 9 While the adjustment is not strictly automatic in the SACU, it is greatly informed by the inflation rate. The Treasury has some discretion. In recent years, the increases have typically been slightly above inflation. 34 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Turkey Producer Price Index Ukraine Agreement to increase the specific component 20% annually between 2019 and 2025 United Kingdom Increase by 2% above the retail price index (measure of inflation) for the 2015–2020 Parliament Source: (1). KEY TAKEAWAY 9 To avoid erosion of their specific excise tax, countries need to regularly – and, ideally, automatically – adjust the excise to inflation. The significance of automatic adjustments and indexation of specific tax to income growth In addition to the risk of erosion due to inflation, the effect of a (specific) tax can be significantly reduced if the tax is not adjusted for increases in consumer income. Income growth makes products more affordable – thereby encouraging consump- tion – especially in countries with rapid income growth. Australia is one of the rare countries that explicitly adjusts its specific excise rates according to wage growth (see Table 2.3). However, a number of countries have adopted automatic adjustments that are higher than inflation and sometimes largely cover income growth as well (see also Table 2.3). Adjusting tax for income growth contributes to increases in prices that make tobacco products less affordable (see section 2.2.3). KEY TAKEAWAY 10 The specific excise tax needs to be adjusted to reflect income growth so that tobacco products do not become more affordable over time. Measures for specific contexts: the role of pricing and other non-tax regulation Emerging evidence indicates that the tobacco industry finds ways to mitigate the impact of higher taxes on prices. For example, despite the heavy reliance on specific taxation in the United Kingdom, a price differential between premium and cheap cigarettes still exists. There is evidence that the tobacco industry does not always pass tax increases on to cheaper products (37, 48). Differential shifting among price categories is also observed in the EU (49–50), New Zealand (38) and the United States (51). Therefore, the public health community has suggested that pricing regulation could be considered as a method of eliminating inexpensive tobacco products that are often used by the young and the poor (52). Three types of pricing regulation are described below: minimum mark-up, price floor and price ceiling. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 35 Pricing regulation • Minimum mark-up – It has been argued that a minimum mark-up of wholesale tobacco prices could be a better strategy to raise prices of tobacco products than excise tax increases. Minimum mark-up laws aim to discourage the sale of products below an assessed cost by imposing a mark-up to the cost declared at different levels of the supply chain. Some studies in the United States have shown that minimum mark-ups do not increase average cigarette prices (53–54). However, a recent study of the impact of minimum mark-up/ price laws has shown that these laws are linked with higher prices, especially for the cheapest brands, and could be used as an effective tool to mitigate the impact of the industry’s price-reducing promotions (55). Another concern related to minimum mark-ups is that they can be manipulated by manufac- turers and are likely to lead to higher profits for the industry, as well as extra administrative costs for the government (56). • Price floor – A few studies suggest that setting a price floor, or a minimum price, is an alternative strategy for increasing tobacco taxes, particularly with respect to reducing health inequities (57–60). A price floor, imposed by the government or as a vertical restraint imposed by the supplier upon retailers, is a price that firms cannot legally undercut. Governments impose price floors to restrain unfair competition or, in the case of services, to increase quality. It is difficult, however, to find the right floor or to anticipate unintended conse- quences or an industry’s adjustments. A study in Malaysia, where a minimum price for cigarettes was imposed in 2010, found that the policy did not seem to have a meaningful impact on prices: licit brand prices remained well above the minimum price, while illicit brands remained well below it. This outcome may be a result of the floor being set too low or the proportion of illicit trade being high, either of which would reduce the effectiveness of the policy (52). In the EU, imposing minimum retail sale prices for cigarettes could be a breach of harmonized legislation concerning the internal market, as minimum prices would distort competition. Therefore, increasing minimum excise duties is recommended instead, to discourage consumption (61). Increasing the minimum excise duties would also result in the additional revenue going to the governments instead of contributing to industry profits. A price floor would probably lead to increased industry profits – giving the industry greater funds for its marketing strategies – and lower tax revenue for governments, reducing their ability to cover costs associated with tobacco use. By reducing price competition, the price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifica- tions). Competition among firms may prevent them from raising their prices, 36 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N but a government that imposes a price floor does this for them. Minimum pricing is likely to create windfall profits for manufacturers and retailers. It can even help manufacturers sustain a cartel. If the industry uses the money to increase promotions, advertising or grant proposals for related research, this could undo some of the potential benefits of the policy.10 Some recent evidence shows that, at least in the case of the United Kingdom, increased concentration of power among a handful of multinational corpora- tions is enabling them to undermine tax increases through increased price segmentation, and that requiring minimum prices might be a good way to address the problem. A longitudinal analysis of price data from the United Kingdom (48) has shown that despite regular excise tax increases over time, average real prices for cheaper segments of the tobacco market (in this case, cigarettes and RYO) did not increase – indicating an undershifting of the tax increases in those segments that resulted in increased sales volume. At the same time, average prices for more-expensive market segments increased, indicating overshifting of the tax increases that resulted in decreased sales volume. This industry strategy ensures that the most price-sensitive consumers remain addicted, while encouraging initiation and discouraging cessation. Furthermore, segmenting the market further by overshifting the tax increase on premium brands while undershifting it for cheaper brands mitigates the impact of declining consumption resulting from higher taxes while increasing overall industry margins and profitability. Another situation where setting minimum prices can be a useful policy is specific to the United States. Banning marketing and promotions11 is not possible under the freedom of expression protections of the Constitution of the United States (Amendment I), and it was estimated in 2008 that more than 82% of all advertising and promotional spending by the tobacco industry was focused on reducing the price of their products at the point of sale (62). This limitation on how government can set policy has paved the way for the implementation of minimum price policies in many states and cities to counter the detrimental impact of price promotions on consumption and on the tax policy itself. Huang et al. (55) found that the presence of minimum price laws was associated with higher cigarette prices. They also noted that cigarette prices were even higher than prices resulting from minimum price laws in states that also prohibit industry from engaging in other price-reduction strategies, 10 See, for example, the PMI strategy of setting up the Foundation for a Smoke-Free World and grant proposals for related research. 11 See section below on banning promotional discounts for tobacco products for further discussion about marketing and promotions. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 37 such as below-cost combination sales, using trade discounts to reduce the base cost of cigarettes and distributing below-cost coupons to consumers. In the contexts of both price segmentation and price promotions, the imposi- tion of minimum prices as a complementary policy to excise tax increases – not as a policy alternative – may help guarantee that taxes do indeed lead to the intended reduction in consumption. Nonetheless, more evidence is needed to support the effectiveness of this policy. • Price ceiling – Concerns about differential tax shifting have led to suggestions that a price cap may benefit public health by limiting the tobacco industry’s ability to reduce average prices by differentially shifting tax increases among various price segments (63–64). Because tobacco manufacturers operate across international markets, however, they could maintain low prices in one coun- try but maintain overall profitability by selling more premium products in another country. Additionally, limiting price increases does not fit the public health purpose of reducing consumption. It is worth noting that maximum retail prices are sometimes used as a base for calculating the ad valorem tax payments in countries with ad valorem or mixed tax systems. In a systematic review of the literature on non-tax policy approaches to raising prices, Golden et al. (63) hypothesized how such policies would influence price dispersion and average prices. Their study found that minimum price policies combined with promotion bans have the potential to increase average prices. This is, of course, relevant in a context where price promotions are present. From either a theoretical or a practical standpoint, however, it is clear that price policies cannot be used alone and should always be considered as complements to excise tax increases. Significantly increasing taxes is the most effective way to dissuade consumption, correcting whatever bias may exist. Significant tax increases also provide the added benefit of raising money for the government rather than profits for the tobacco industry. Nonetheless, a minimum price might help narrow the gap between cheap and pre- mium cigarettes when applied to all tobacco products to avoid product substitution. Other non-tax regulation • Banning promotional discounts for tobacco products – The sale of tobacco products at a discount rate – such as through reduced-price coupons or buy-one-get-one-free offers – encourages consumption and undermines tax increases. Such practices should be completely banned. They often exist outside the realm of the finance sector because they are considered a type of marketing – promotional discounts are usually addressed in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. 38 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N According to the 2019 RGTE (1), 118 countries out of 195 with all levels of income had such a provision implemented as of 31 December 2018. • Banning the sale of single sticks of cigarettes – Article 16 of the WHO FCTC, “Sales to and by minors,” paragraph 3, requires Parties to “prohibit the sale of cigarettes individually or in small packets which increase the affordability of such products to minors”. Some smokers opt for buying single sticks partly because of the lower im- mediate costs of buying cigarettes individually (65). Internal (unpublished) analysis of single-stick prices collected by WHO for the 2012, 2014 and 2016 editions of the WHO RGTE shows that, in fact, the aggregate price of 20 single sticks of cigarettes sold separately is generally higher than the price of a 20-cigarette pack sold in the market of a specific country. Despite this fact, single-stick sales – and sales of small-sized packs – make cigarettes accessible to consumers with limited disposable income. De Ojeda (66) found in a study conducted in Guatemala that single-cigarette sales are associated with increased cigarette accessibility for less-educated, lower-income populations and minors. Single-stick sales are also a feature of many markets in South-East Asia, including most notably Bangladesh and India, but also in other parts of the world, e.g. South Africa. Single-stick sales also reduce the impact of a tax increase, since the in- crease per stick is much smaller than the increase per pack (67). In a study investigating how smokers in New York City responded to a tax increase of US$ 1.25 per pack in 2008, Coady et al. (68) found that 15% of smokers bought more single cigarettes than they had previously.12 By allowing single stick sales, governments risk losing part of the ad valorem taxes if the tax base is the retail selling price; the retail price of single sticks is much more difficult to monitor than the retail price of packs of cigarettes, on which, for example, tax stamps with prices can be applied. An internal WHO analysis of the most recent tobacco control laws in 2018 in 195 countries found that 86 countries impose by law a ban on the sale of single sticks of cigarettes (36% of the countries are high-income, and 64% are LMICs). In addition to banning the sale of single sticks of cigarettes, 67 of the 86 countries specify a minimum size for packs of cigarettes. Most (52 countries) use the 20 cigarettes per pack standard, but minimum sizes 12 Before 2018, the use of single sticks was possible, but it has since been banned. See New York City Administrative Code. chapter 7: regulation of tobacco products, subchapter 1: Tobacco Product Regula- tion Act, §17-704.a-1. New York: New York Legal Publishing Corporation; 2020 (http://library.amlegal. com/nxt/gateway.dll/New%20York/admin/title17health/chapter7regulationoftobaccoproducts?f=tem plates$fn=default.htm$3.0$vid=amlegal:newyork_ny$anc=JD_T17C007, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 39 range from five sticks in Ghana up to 25 sticks in Papua New Guinea. Another 15 countries do not specifically ban the sale of single sticks but do specify the minimum size for packs of cigarettes. According to the WHO FCTC, in order to reduce affordability, single stick sales should be banned and a minimum number of cigarettes contained per pack should also be defined. KEY TAKEAWAY 11 A number of non-tax measures are closely connected to tax policies, including price regulations, bans on promotional discounts for tobacco products and bans on the sale of single sticks of cigarettes. The price policies discussed are (1) minimum mark-up, (2) price floors and (3) price ceilings. Current evidence does not yet demonstrate that minimum mark-ups and price floors lead to increases in average price. Nonetheless, they may be relevant in some specific contexts as complementary policies to excise tax increases. Price ceilings limit price increases, which can mitigate their impact on consumption. Price marketing strategies such as promotional discounts and the sale of single sticks undermine the effect of tax policies and should be banned. A minimum pack size should also be required by regulators. Tax increases and their possible impact on inflation At times, the inflationary impact of tax increases on cigarettes and other tobacco products is raised as an argument for not increasing these taxes. This may be a concern in countries where wages and/or a significant share of government spend- ing is indexed to inflation (e.g. for public pension payments) or where government policy is to keep inflation low. The extent to which tobacco product tax increases lead to increases in inflation depends on several factors, most notably the share of these taxes in prices and the weight tobacco prices are given in computing a price index. For example, if taxes account for 25% of tobacco product prices, a doubling of the tax (100% increase) will increase prices by 25%. If the weight given to tobacco products in the price index is 3%, the index will rise by 0.75% in response to the tax increase. As tobacco taxes account for a larger share of tobacco product prices, the inflationary impact of a tax increase will be greater. Similarly, as tobacco products are given more weight in computing a price index, a given tax increase will have a greater inflationary effect. In general, for most countries, the inflationary impact of tobacco product tax increases would be relatively small (47). 40 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Consumer price indexes have multiple purposes. They are an important economic indicator for most countries and are often a key determinant of monetary policy. Inflation rates have a direct impact on interest rates and exchange rates. In many countries, changes in wages, social security benefits and other payments are tied to inflation as measured by a price index. Price indexes are used to provide more accurate comparisons of changes in expenditures, incomes and prices for specific goods over time, as well as to allow comparisons across countries. Given the many uses of consumer price indexes and the potential inflationary impact of tobacco tax increases, some governments have developed alternatives that exclude tobacco (and sometimes other goods) for some uses. For example, since 1992, France has excluded tobacco products from the price index used for adjust- ing minimum wages (47). However, many countries continue to include tobacco product prices in their consumer price indexes. Excluding tobacco products from the basket of goods used in developing key price indexes would greatly reduce concerns about their impact on inflation. In addition, with declining consumption of tobacco products, the inclusion of their prices in key price indexes results in a distorted measure of price for many consumers. KEY TAKEAWAY 12 If governments are concerned about the potential inflationary impact of a tobacco tax increase because wages or some government spending may be tied to a price index, they can use a price index that excludes tobacco products. The importance of taxing cigarettes and other tobacco products in a comparable way While cigarettes are the most commonly used tobacco product globally, other tobacco products are as prevalent and sometimes more prevalent than cigarettes in some parts of the world. Bidis and smokeless tobacco are the main products consumed in some countries in South-East Asia – Bangladesh and India in particular – and waterpipes are widely used for smoking tobacco in the Eastern Mediterranean region (4). These products, as well as RYO, have historically been taxed much less than cigarettes (see, for example, Fig. 2.10 for Bangladesh and India, where the excise tax and prices of bidis and smokeless tobacco are much lower than those for cigarettes). This differential taxation undermines the health impact of excise taxes on tobacco products because (1) it encourages users to switch from cigarettes to the lower-taxed product (see the case of Thailand below); (2) it is not effective in reducing tobacco use in general, especially if the most widely used product in the country is not cigarettes; (3) it can encourage tax avoidance by companies that may CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 41 redefine products that are similar to cigarettes so that they fall within the lower- taxed product category (see the EU example below) and (4) it reduces the size of government revenues since those products could have been taxed at higher levels. Fig. 2.10 Price and tax of the most-sold brand of cigarettes, bidis and smokeless tobacco in Bangladesh and India, 2018 Source: (1). In Thailand, for example, the price of cigarettes has been raised quite successfully through taxation over a number of years, while taxes and prices of loose or RYO tobacco have until very recently remained unchanged. Indigenous tobacco used for RYO cigarettes has historically been exempt from excise, while foreign tobacco was taxed at a low level relative to that of manufactured cigarettes. Consequently, Thailand experienced growth in the RYO market even though cigarette consumption had been falling.13 The Thai government eventually took strong action to address this issue. First, the exemption for indigenous tobacco was removed in 2018. The Cabinet then approved an increase in the excise rate on small producers (of indigenous tobacco) from 0.005 baht per gram to 0.025 baht per gram in 2020, with another increase to 0.1 baht per gram scheduled for 2021 (69). In the EU, the minimum excise duty levels for cigars and cigarillos is significantly lower than that for cigarettes. The Member States of the EU are required to levy an 13 WHO Country Office for Thailand, personal communication, 2019. 12.8 0 190 54.9 64 24 Price Excise tax amount 80 12.8 3.8 44 Lo ca l C ur re nc y 0 50 100 150 200 Cigarettes (20 sticks) Bangladesh, taka India, rupees Bidis (20 sticks) Bidis (20 sticks) Cigarettes (20 sticks) Smokeless (20 g) 42 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N excise duty of at least €90 per 1 000 cigarettes, which should be 60% of the weighted average retail selling price of cigarettes released for consumption. For cigars and cigarillos, only €12 per 1 000 items, or an excise duty of 5% of the retail selling price, is required. As a result, the excise tax share on cigarettes is much higher in many EU countries than the share for cigars and cigarillos. In response, some companies started to market so-called borderline cigarillos. These products have characteristics similar to cigarettes but can be sold at a lower price because for excise purposes, they are considered as cigarillos. Although this issue seems to be largely solved by amendments to the definitions of these products at the EU level and a change in tax structures in some countries, it is important to be aware of the unintended incentives that can be created by large gaps in excise tax levels between product categories (70–71). For more details on industry tactics to undermine tax increases, see Box 2.1. KEY TAKEAWAY 13 To make excise tax on tobacco products more effective in reducing overall tobacco use and to avoid substitution between products, all tobacco products need to be taxed in a comparable way. The Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce the demand for tobacco) (73) recommend that all tobacco products should be taxed in a comparable way. Box 2.1 Industry tactics used to undermine tax increases Tax increases reduce the demand for tobacco products and present a threat to the tobacco industry’s high profits. The industry responds by using various strategies (17, 46, 48), including the following (46): Stockpiling (forestalling/front-loading) – Before the implementation of an an- nounced tax increase, manufacturers overproduce tobacco products, paying the pre-tax-increase rate. As a consequence, sales and tax revenue decline immediately but temporarily after the tax increase (while sales and revenues had increased sub- stantially just before the tax increase) and the industry attributes this drop in revenue to the emergence or increase of illicit trade. This practice results in tax avoidance if there is no law prohibiting it (see also the discussion on anti-forestalling in Chapter 3). Changing certain product characteristics (for example, weight or length) and/ or adjusting the production process – When tobacco products are taxed at different rates or are subject to different tax increases, the industry can, for example, re-label one type of tobacco product as another product that has a lower tax burden (as in the example of the EU above). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 43 Choosing the time of a price increase announcement strategically – The industry may raise prices in anticipation of a tax rate increase, generating extra profits in the period until the tax is actually implemented. When the tax increase is implemented, consumption and tax revenue will fall, but prices will not change, so the industry can claim that the tax policy was ineffective in reducing demand. Adopting price-discriminating strategies or price-related promotions – The industry may offer discounts, retailer rebates or added value (gifts) to tobacco purchases to minimize the loss of price-sensitive consumers. This, however, is not possible in countries where strict bans on tobacco advertising, promotions and sponsorship are implemented. Using brand proliferation (for example, launching a low-priced brand) and price segmentation – Manufacturers can choose to reduce prices of certain brands or introduce new, even cheaper ones to keep price-sensitive consumers in the market. There is evidence that firms introduce new cheaper products and use price-marking – printing the price directly on packs of tobacco products – to lock in their price (48). Such practices compromise both public health and revenue objectives. Differential shifting of tax increases across different price segments, depending on the market circumstances – The industry may increase the price of a product by more than the amount of the tax increase (tax overshifting) and blame the govern- ment for the total increase. Tax overshifting is profitable when demand is inelastic, that is, when the price increase more than offsets the reduction in sales. The industry may overshift the tax increase for higher-priced brands, which are expected to be more price inelastic than lower-priced brands. Additionally, to keep price-sensitive consumers in the market, the industry may temporarily absorb part (or all) of the tax increase on lower-priced brands. The differential tax shifting will lead to different responses in the demand for the different brands (37, 48). Lobbying government to distort interventions – Government policy might be influenced by tobacco industry lobbying, directly or indirectly. Policy-makers are not simply welfare or revenue maximizers; they also value political support. Industry lobbying might lead to adopting a favourable type of taxation, postponing tobacco tax increases or distorting the tax rate downwards (17). Article 5.3 of the WHO FCTC, “On the protection of public health policies with respect to tobacco control from commercial and other vested interests of the tobacco industry”, and its guidelines provide useful guidance on how to address tobacco industry interference. In fact, all 181 countries that are Parties to the WHO FCTC have a legal obligation to implement the requirements of Article 5.3. Having correct expectations about industry responses is important for estimating the impact of a tax increase on consumption and tax revenue. 4 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.2.3 MEASURING IMPACT AND RECOMMENDED INDICATORS Governments need to consider a number of indicators when formulating policy changes. Inappropriate assumptions about consumer behaviour, market structure and industry behaviour can lead to faulty policy analysis. Measuring impact on price and demand Consumption habits, local traditions and industry characteristics – such as the number of different brands offered, the possibilities of cross-border shopping and the presence and level of illicit trade – all affect the shape of the demand and supply of tobacco products, thereby determining the value of the price elasticities. Price elasticity, together with the industry’s pricing strategies – for example, the degree of tax shifting – and the tax share in the retail price, determine the elasticity of the tax base, regardless of whether the base is determined by quantity (for specific taxation) or transaction value (for ad valorem taxation). The importance of elasticity estimates Different types of elasticity should be considered: • price elasticity – own-price elasticity – measures the response of consumers’ demand for a product following a change in the price of the product. – cross-price elasticity – measures the response of consumers’ demand for a product when the price of another product changes. Cross-price elasticity can also occur between different brands or price segments for the same product. • income elasticity – the response of consumers’ demand for a product when their income level changes. Correct estimates of price and income elasticities are important for policy-makers who need to anticipate the impact of a tax increase on consumption and tax rev- enue. Estimates will vary depending on a number of factors, including whether responses are considered in the short run versus the long run, the functional form of the demand function used, whether factors such as addiction or tax evasion are accounted for and the way data are constructed. For example, details such as the degree of aggregation of data, whether gender- or age-specific data are used, the time span covered and which estimation procedures are used (e.g. ordinary least squares, two-stage least squares or generalized method of moments) will all affect the results of the estimate (72). Price elasticities may change over time, as well because of changes in any of the other factors affecting demand, such as income or tobacco control measures, and also because of changes in estimation techniques and the types or sources of data used. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 45 Moreover, what is of most interest is the price elasticity of total demand. A tax increase may reduce tax-paid retail sales but not necessarily total consumption. For example, smuggling can significantly bias price elasticities when the elasticities are estimated using legal sales data; not accounting for possible illicit trade might lead to overestimation. Similarly, when cross-border shopping is included, the price elasticity of demand is lower (in absolute value) (74). Estimating the total price elasticity of demand for legal and illegal consumption can be done by using cross- sectional data from nationally representative household surveys. However, this approach also has its weaknesses. For example, respondents tend to underreport their consumption of tobacco, which leads to bias in the size of demand. Price endogeneity14 is another technical problem that can be challenging to address. To comprehensively estimate the total effect of a tax increase on demand for all tobacco products as well as on tax revenue, the degree of substitutability between them needs to be estimated (55). Cross-price elasticity measures how the quantity demanded of a particular tobacco product changes when the price of another tobacco product increases. When this elasticity is positive, the products are substitutes; the higher the value of the elasticity, the closer substitutes the products are to one another. For example, positive cross-price elasticity between RYO and manufactured cigarettes implies that the demand for RYO increases as the price for cigarettes increases. Substitutability may also arise between different cigarette brands – when the relative price of economy brands increases, demand for premium brands may increase. This effect can be exacerbated when differential (tiered) taxation is ap- plied on different types of cigarettes, further widening the gap in prices between brands and segments and encouraging substitution. The substitutability between traditional and new and emerging tobacco and nicotine products is currently of great interest (see section 2.4 below). In some countries, different tobacco products can also be complementary rather than substitute goods. This means that when the price of a tobacco product increases, the demand for its complement drops because users are unlikely to use the complementary tobacco product alone. For example, a number of studies have found manufactured and indigenous bidi cigarettes to be complementary goods in India (75–76). The sign and magnitude of income elasticity vary across time, countries and demographic groups. For example, in the United States, a high-income country, income elasticity over time has changed from positive to negative, and cigarettes have switched from being a normal good to an inferior good (77–78). On the other hand, among LMICs, where prevalence of smoking tends to be relatively higher, 14 Price is endogenous because it is not an independent variable: it is estimated by dividing expenditure on tobacco by consumption of tobacco, with consumption being a dependent variable in the estimation of price elasticity. 46 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarettes might still be a normal good, with consumption increasing as income levels increase (positive income elasticity) (79–80). There are unobservable characteristics that differentiate higher-income smok- ers from lower-income smokers, such as differences in time and risk preferences, differences in associating a social stigma with smoking and differences in taste for smoking as a pleasurable activity. When these characteristics are ignored, estimates of the correlation between income and smoking-related outcomes are biased. Kenkel et al. (81), using techniques that estimate the causal effect of income on smoking among low-income adults, found that tobacco is a normal (even a luxury) good: higher income is associated with a higher probability of smoking participation and a lower probability of smoking cessation. These results are consistent with those regarding the impact of the business cycle – periods of expansion or recession in economic activity – on health be- haviour and outcomes. Ruhm (82–83), for example, found that smoking declines during temporary economic downturns and increases during economic expansions. Tarantilis et al. (84) found that estimates of income elasticities of demand in Greece were higher after the economic crisis of 2010 than before it. The financial crisis and the austerity measures shifted the demand for cigarettes downwards and turned cigarettes into a more income-elastic good. Interestingly, evidence from Germany suggests that the propensity to become a smoker significantly increases during an economic downturn. However, among those who are already smokers, cigarette consumption actually decreases (85). Ideally, when estimating price and income elasticities, the effect of non-price policies should also be accounted for. A recent study from South Africa shows that failing to take non-price policies into account will overstate the price effect (86). The NCI/WHO Monograph (4) suggests that price elasticity of demand for tobacco is on average -0.4 in high-income countries (ranging from -0.2 to -0.6). Estimates for LMICs are more variable, clustering around -0.5 (ranging from -0.2 to -0.8). A price elasticity of -0.5 means that a 10% increase in price would lead to a 5% reduction in consumption. KEY TAKEAWAY 14 Policy-makers need to know the elasticity of demand – including price elasticity (own-price and cross-price) and income elasticity – for tobacco products in their country in order to correctly assess the impacts of potential policy changes on consumption and subsequent revenues. These estimates need to be made on a regular basis to capture changes in demand over time. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 47 The importance of the tax base elasticity The tax base elasticity measures the sensitivity to a change in the tax rate of the base on which the tax is imposed – the base being tobacco consumption in the case of specific taxation and tobacco expenditure in the case of ad valorem taxation. The magnitude of the elasticity of the tax base depends on price elasticity of demand, the tax structure, the level of the tax rate and its share in price, along with the industry response through its decision to absorb, pass through or overshift the tax on to the retail price. Consumers’ preferences and income, the availability of substitutes and other non-price tobacco control measures also influence the tax base elasticity, essentially through the price elasticity of demand. The magnitude of the elasticity of the tax base also depends on social motivations, including price and tax expectations, which are ultimately impacted by successful tobacco control measures that affect consumers’ willingness to pay taxes or prices. In addition, the tax-base elasticity depends on smokers’ perceptions of the prob- ability of detection and tax enforcement when using illegal products, as well as the availability and accessibility of opportunities for tax evasion and avoidance. Finally, consumers’ willingness to pay taxes depends on their perceptions regarding the use of the tax revenue (87). Therefore, the tax base elasticity is largely influenced by government policy choices. Increasing the tax share in prices is recommended by WHO as a tool to achieve the public health objective of reduced tobacco use: a higher tax share in prices increases the tax base elasticity, all else remaining constant, and therefore increases the reduction in the tax base through the resulting reduction in smoking. However, manufacturers can be expected to attempt to manipulate the tax base elasticity through their pricing policies, such as tax shifting. As discussed earlier in this chapter, industry behaviour is itself affected by government tax policy and regulations. A number of factors need to be taken into account when considering tax pass- through. As discussed earlier, tax is more likely to be overshifted within a specific tax structure than within an ad valorem structure. There is also evidence of industry overshifting the tax for premium or expensive cigarette brands while undershifting the tax for cheaper brands. This indicates that within a given market, the industry’s decision on the extent of tax pass-through will vary based not only on the tax structure but also on the structure of the market. It will also vary by brand. But this does not give an indication about the impact of the tax increase on the average price of a tobacco product. In the context of the tax base elasticity and the impact of tax increases on revenues, it is important to assess how tax increases affect average prices. The example of South Africa is very useful here. Over the past two decades, South Africa has been consistently increasing its specific excise tax on cigarettes, which has led to large price increases. An analysis of the effect of excise tax increases on 48 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarette prices between 2001 and 2015 (26) shows that while there is evidence of tax overshifting, at least within a specified period of time, overall, the tax has been undershifted in real terms. This undershifting is due partly to increased competitive- ness in the market and partly to the introduction of low-priced brands. Of course, this encouraged some consumers to downshift their consumption to cheaper products, but it also pushed manufacturers of more expensive brands to absorb part of the tax increase to reduce the impact on price. A change in the level of the tax rate – with all other factors that influence con- sumption held constant – will result in a change in the tax revenue.15 Estimates of tax base elasticity help governments predict changes in tobacco tax revenues following a tax increase (see details in Annex 2.2). Under specific taxation: • tobacco consumption – the tax base – is expected to be price inelastic (17, 47); • prices increase by less than the tax increase, on average (there is no tax overshifting overall); and • consumption – the tax base – is also expected to be tax inelastic: the quantity of consumption falls less than proportionately to the tax increase, and the tax revenue increases. Under ad valorem taxation: • the tax base is the total consumer expenditure (or, equivalently, the industry sales revenue) on (legal) tobacco consumption – that is, the tax base under ad valorem taxation is determined by both price and quantity, which is itself a function of price; • the sign of the tax base elasticity – which can be either negative or positive – depends on the magnitude of the price elasticity of demand; • since evidence suggests that tobacco demand is price inelastic, the tax base elasticity is positive; • when the ad valorem tax rate increases, both price and quantity adjust, but quantity falls less than proportionately to the price increase, and tax revenue increases; and • a tax rate increase leads to both a higher level of revenue and a lower level of consumption; the value of the elasticity – and hence the tax revenue – increases with the degree of tax shifting. 15 This concept has been used by Laffer to argue that tax increases that are too high will reduce excise tax revenues (the so-called Laffer curve). For a detailed discussion on the Laffer curve, see section 4.4 in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 49 Taxation serves as an instrument for achieving both fiscal and public health objectives. If, after successful tobacco control interventions, prices reach levels where demand becomes elastic, the tax base is still most likely to be inelastic due to tax undershift- ing, since overshifting is not a good pricing policy when demand is elastic. In other words, a tax rate increase – in combination with non-price tobacco control measures that make consumers more sensitive to price (tax) increases – leads to decelerating but still positive marginal revenues. For an example of a tobacco taxation success story, see Box 2.2. For further details of countries’ experiences with tax increases and their impact on revenues, see section 4.4 in Chapter 4. KEY TAKEAWAY 15 Policy-makers’ key policy tool to control demand is tax. Therefore, it is essential they assess not only the impact of price on demand but, more appropriately, the impact of tax on demand: this is the tax base elasticity. The tax base elasticity is essentially determined by (1) the price elasticity of demand, (2) the degree to which the industry will pass the tax on to the retail price and (3) the tax as a share of the retail price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. Box 2.2 A tobacco taxation success story: Turkey Turkey is an example of a country that has been increasing taxes regularly and sig- nificantly over a relatively short period of time and has reaped the benefits of this policy. As shown in Fig. 2.11, the excise tax per pack of cigarettes more than doubled in real terms over 10 years, with the real price almost doubling as well. In parallel, tobacco excise revenues increased by 67% and cigarette sales decreased by 20%. Since the beginning of the country’s Health Transformation Program in 2003, Turkey has successfully increased public health spending and collected more tobacco tax revenue. According to the latest available figures, in 2015, tobacco tax revenue was equivalent to 42% of the country’s public health expenditure and 1.5% of GDP (88). 50 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.11 Tobacco excise revenue and consumption in Turkey (base year 2008), 2008–2018 Sources: Reference 1 for the price of the most-sold brand, Ministry of Finance for the sales and revenue data and IMF world economic outlook, April 2020. See https://www.imf.org/en/Publications/WEO/weo- database/2020/April for the adjustment for inflation. Impact on affordability While price increases clearly have an impact on consumption, when the effects of increasing per capita income of a population are not considered, the price impact may not be as strong as expected. Increases in a population’s income also increase its purchasing power. And, as indicated earlier, tobacco products generally behave like a normal good. Consequently, as income increases, it is expected that tobacco consumption will increase as well. To mitigate this effect, price increases (following tax increases) need to be greater than increases in income. This is where the concept of affordability comes in. Affordability examines the effects of both increasing prices and increasing incomes on consumer behaviour. A common and easy way to calculate affordability, made popular by Blecher and van Walbeek (89), is to use the percentage of GDP per capita required to buy 2000 cigarettes (or 100 packs of 20 cigarettes) in a given year. An increase in this proportion over time will indicate that cigarettes are becoming less affordable and should lead to reductions in consumption. Changes in trends in affordability of cigarettes over time help policy-makers understand how prices are evolving 0 1 2 3 4 5 6 Tu rk is h lir as p er p ac k N um ber of sticks, 10 000 000 Turkish liras 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real excise tax amount Real price Real tobacco excise revenues Cigarettes, per capita sales 1 100 500 1 700 2 300 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 51 relative to a population’s ability to purchase cigarettes and enable them to revise their policies accordingly. Recent studies in India, for example, highlight the wide price differential between manufactured cigarettes and indigenous tobacco products such as bidis and chewing tobacco, as well as the propensity for these indigenous products to become more affordable over time due to favourable or more lenient tax policies towards them (90–91). Figure 2.12 shows the change in affordability of a pack of the most-sold brand of cigarettes by country income group between 2008 and 2018. During this time period, affordability declined in almost 70% of high-income countries, while it declined in slightly more than 35% of middle-income countries and only 26% of low-income countries. Fig. 2.12 Number of countries that have experienced a change in affordability of cigarettes between 2008–2018, by income level Note: Change in affordability was computed as the least squares rate of change in the per capita GDP required to purchase 2000 cigarettes of the most-sold brand in local currency in a given year. The trend rate of growth was computed for countries with four or more years of data, including 2018. Affordability was assessed as not having changed if the least squares trend in the per capita GDP required to purchase 2000 cigarettes over the period 2008–2018 was not statistically significant at the 5% level. Source: (1). KEY TAKEAWAY 16 From a health perspective, in addition to examining the impact of a tax increase on the levels of price, demand and revenues, policy-makers should consider a tax hike that will lead to prices rising more than increases in their population’s income; a tax increase should make tobacco products less affordable to consumers so that demand will be effectively reduced. Cigarettes became less aordable Cigarettes became more aordable Aordability did not change Could not be assessed due to insucient data High income Middle income Low income 39 36 9 13 5 7 37 23 6 13 2 5 52 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Projecting impact on consumption, smoking prevalence and lives saved The WHO interactive smoking projection and target-setting tool (WHO ISPT) The WHO ISPT enables national policy-makers and tobacco control experts to explore the potential impact of proposed tobacco control policies. It uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. The WHO ISPT provides projections of (1) tobacco smoking rates and (2) tobacco-smoking-related deaths in a country under different policy settings and for different time periods. It was designed to promote multisec- toral collaboration within countries by enabling experts from various ministries (for example, health, education, finance, national statistics), civil society, academia and media to explore options for medium- and long-term tobacco control planning together with WHO experts. Use of the WHO ISPT enables strong partnerships for policy change advocacy, program development and evaluation.16 In particular, it can help policy-makers in the Ministry of Finance assess the specific contribution of tax policies – within overall tobacco control policies – towards achieving specific targets in tobacco prevalence reduction. Projecting impact on excise revenue The WHO tobacco tax simulation model (WHO TaXSiM) The WHO TaXSiM is a simple but data-intensive Excel-based tool that helps policy- makers analyse their tobacco tax policy and assess the impact of any excise tax increase or change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Using detailed data about the market – including the majority of brands found in the market, their market share and price levels and the applicable tax – and assumptions about price elasticity of demand, the WHO TaXSiM predicts the impact of tax changes on consumer prices, consumption volume and tax revenues generated by each brand and market segment for the following year. The exercise can be done for multiple years.17 By exploring market data in detail, in addition to assessing the potential rev- enue impact of changes in excise tax, the WHO TaXSiM is a useful instrument for highlighting weaknesses and opportunities in an existing tax system and market. It can also encourage policy-makers to create administrative databases that can be periodically updated to monitor the dynamics of the cigarette market. 16 The WHO ISPT is not available publicly, but WHO will work directly with interested countries upon request to use it to produce data-to-action-type plans. 17 For more information about the methodology, see https://www.who.int/tobacco/economics/tax- sim_background.pdf, accessed 29 September 2020. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 53 KEY TAKEAWAY 17 Policy-makers can use available tools to simulate the impact of tax increases on prices, consumption and revenues, as well as smoking prevalence and lives saved. Recommended indicators to monitor tobacco taxation progress MPOWER WHO publishes a biannual RGTE, which monitors global progress in tobacco con- trol. In particular, the report focuses on the implementation of the policy package MPOWER, a set of proven demand-reduction measures in line with the key provi- sions of the WHO FCTC (1). While raising taxes on tobacco (component R) is proven to be the most effective and cost-effective policy to reduce tobacco use (4), implementing the entire MPOWER package at the best practice level will reinforce the impact of R. For example, as mentioned earlier in this chapter, banning promotional discounts as part of the E measure (enforce bans on tobacco advertising, promotion and sponsorship) will favour price increases following a tax increase. If all the MPOWER tobacco control measures except R were implemented at the best practice level, all else remaining constant, revenues would be expected to decline. Thus, in order to maintain revenue levels, it is important to raise excise taxes on tobacco products regularly to compen- sate for the decline in tobacco use from the other four tobacco control measures. Tax share The main indicator in the R policy in the RGTE (1) is the total share of indirect taxes in the retail price of the most-sold brand of cigarettes.18 Countries whose most-sold brand of cigarettes has a total tax that is equal to or greater than 75% of the retail price are considered to be at the highest level of achievement. While total taxes include excise taxes, VAT (or sales taxes), import duties (when applicable) and other indirect taxes (where applicable), it is preferable to focus on excise taxes, since they are the component that most influences the relative price of tobacco. The share of excise tax in the retail price can be extracted from the RGTE database.19 The 2010 WHO technical manual on tobacco tax administration recommended making excise taxes account for at least a 70% share of excise taxes in the retail price of tobacco products (47). 18 For more details about how this indicator was compiled, see Technical Note III of the RGTE 2019 (https://www.who.int/tobacco/global_report/Technical-Note-III.pdf?ua=1). 19 See taxes and retail price for a pack of 20 cigarettes, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.1-Taxes-and-retail-price-for-a-pack- of-20-cigarette-most-sold-brand.xls?ua=1, accessed 29 September 2020). 54 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Affordability As discussed previously, the share of tax in the retail price is not enough to ensure that a policy will be successful in reducing demand. Any tax increase should lead to an increase in price that will effectively discourage consumption. While global trends indicate that a high tax share is positively correlated with a high price level (see Fig. 2.2 in section 2.1.2), this may not necessarily apply to a particular country; a tax share can be high, while at the same time tobacco products remain afford- able. For this reason, it is important to monitor not only tax increases but also whether those increases led to a price increase that is greater than income increases. As described in section 2.2.3 of this chapter, a common indicator is the percentage of GDP per capita required to buy 100 packs of 20 cigarettes in a given year.20 Other indicators As discussed in detail in sections 2.2.1 and 2.2.2, a good tax structure can make a tax policy more effective in increasing prices and decreasing affordability of tobacco products. Indicators can include whether a uniform excise is applied, whether it is a specific excise and whether it is adjusted regularly for inflation. A number of such indicators are also monitored through the RGTE and can be downloaded online.21 A tobacco tax indicator compiled in 2020 combines the various elements that form a good tobacco tax policy. The Tobacconomics Cigarette Tax Scorecard (92) rates a country’s tobacco tax policy performance based on best practices. The four components that determine the level of performance are (1) cigarette price (in PPP), (2) changes in the affordability of cigarettes over time, (3) the share of taxes (total and excise) in retail cigarette prices and (4) the structure of cigarette taxes (i.e. whether excise is applied; whether it is uniform or tiered; whether excise is specific, ad valorem or mixed; and, for the ad valorem component, if the tax is applied on the retail price and if there is a minimum specific excise and, for the specific component, if tax is automatically adjusted upwards). Each of the four components is given a score, using a five-point index, with the total score reflecting an average of the four component scores. The closer the total score is to 5, the bet- ter the tobacco tax policy performance is in a given country. While this published scorecard is currently applied only on cigarettes, it can be easily applied on other tobacco products, provided the needed data are available. 20 This indicator has also been compiled in the RGTE; see (https://www.who.int/tobacco/global_report/ Table-9.6-Affordability.xls?ua=1, accessed 29 September 2020). 21 See supplementary information on taxation, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.5-Supplementary-information-on-taxation. xls?ua=1, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 55 KEY TAKEAWAY 18 Tobacco taxation works best if implemented as part of a comprehensive MPOWER package. MPOWER is an overall indicator that incorporates all the key demand-side tobacco control measures. KEY TAKEAWAY 19 In addition to assessing the potential impact of a tax increase, policy-makers need to monitor progress over time. The share of the tax in the retail price is an indicator of progress. However, it is important to remember that an effective tax increase must translate into higher prices in order to make tobacco products less affordable. Combining all the components of a good tax policy into one scorecard can also be useful for assessing tobacco tax policy as a whole. 2.3 DOMESTIC AND REGIONAL POLICY INTEGRATION While it is essential to design tobacco tax policies with the utmost consideration of all the aforementioned factors, it is also important to consider how external factors can impact or even impede public health policy objectives. As Chapter 3 explains, cooperation among the various agencies that are directly involved in tax administration, collection and enforcement is important for effective and efficient tax policy implementation. But in the design phase, it is also essential to engage with agencies and other policy-makers that are not directly involved with taxation. Domestically, coordination is required to ensure that policies in non-health sectors do not negatively impact or even counteract tobacco control initiatives. For countries that are part of a regional bloc, harmonization of tobacco taxation is essential to protect the single market – as well as the health of the population – and to prevent tax revenue erosion, tax avoidance and tax evasion. 2.3.1 INTERSECTORAL COOPERATION ON DOMESTIC POLICY Domestic policies in agriculture, industry, trade, finance and labour have the po- tential to create or support incentives at different stages of tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. For example, subsidies provided to farmers or manufacturers involved in growing or processing tobacco can reduce prices and incentivize continued participation or even increase development in these areas, which is counterproductive to the goals of making tobacco products less affordable and reducing tobacco consumption. Multisectoral integration and policy coherence are needed at the country level to ensure that public policies and interventions in non-health sectors do not act against the intended public health impact of tobacco control and taxation. 56 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 20 Greater policy coherence in agriculture, industry, trade, finance and labour should be promoted to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. 2.3.2 REGIONAL TOBACCO TAX HARMONIZATION Policy integration is driven by the recognition that cooperation on domestic policies can substantially increase the gains from forming a regional bloc. Harmonization is desirable and may be necessary in certain areas with spillover effects, such as tax policy, the possibility of a so-called “race to the bottom” or threats to public health. Harmonization could be as simple as setting minimum standards and requirements based on global norms and best practices. Harmonization of tobacco taxation is required to ensure the establishment and proper functioning of a single market, prevent tax revenue erosion, prevent tax avoidance and tax evasion and protect people. When barriers to trade between countries are removed, harmonized tax rates support the single market because they improve the ability of consumers, producers and investors to make decisions that are not distorted by taxation but reflect real opportunity costs. Tax competition – where countries simply undercut each other’s tax rate – could prevent governments from raising sufficient funds to pursue social policy. To avoid such a race to the bottom, countries can establish minimum tax rates within the customs union (93). Even if tax competition is not present, when substantial tax differences exist in neighbouring countries, there is a clear incentive to trade across borders in order to reduce tax payments legally or illegally. The experiences of established regional economic communities offer important policy lessons, not only in terms of the general integration process but also for the process and extent of tax policy coordination. The EU implemented a successful regional tax harmonization scheme. Over the years, the focus in harmonization of tobacco taxes has broadened from the elimination of tax obstacles to the fight against harmful tax competition, tax avoid- ance and tax evasion and, more recently, to public health protection. Naturally, addressing these issues requires increasing convergence in fiscal policy and tax administration. Although price differentials still exist, setting a minimum on the share of taxes in the final price of tobacco products as well as a minimum excise tax has helped countries reach some level of harmonization. The EU experience confirms that both a declining tobacco consumption trend and stable revenues can be achieved with harmonized minimum excise rates (94). Moreover, the harmonization process has CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 57 offered opportunities for the enforcement authorities (tax or customs) to obtain information that can be used in the fight against fraud and tax evasion. On the other hand, the experience of harmonization efforts in the West African Economic and Monetary Union (WAEMU) shows how the absence of a supranational body (like the EU) or a hegemonic member state (see the SACU example below) can slow down policy integration that would benefit all member countries (95). The eight countries of the WAEMU are bound by a Tax Directive22 that requires them to impose an ad valorem excise on the CIF value or producer price of tobacco products, which is subject to under-declaration and is difficult to ascertain. Additionally, a maximum excise rate is imposed, and some members apply additional taxes to deal with this constraint. The Directive was revised in 2017 (96), but unfortunately the tax structure remains the same, and the maximum rate was not removed but rather has been increased. The Southern African Customs Union (SACU), which has five member countries, is the oldest existing customs union, established in 1910. Thanks to the hegemonic lead of South Africa, a country with a sophisticated administration system and an aggressive tobacco tax policy, SACU adopted a well-integrated tax policy that has benefited all its members (95). The GCC, established in 1981, is a regional intergovernmental political and economic union consisting of six states of the Persian Gulf. Home to one fifth of the global oil supply (97), the GCC has never relied on taxation as a source of revenue; no direct or indirect taxes were applicable in the region. Although there was no excise on tobacco products, as a customs union, the GCC countries have a common external tariff. This common tariff includes harmonized rates but also a harmonized structure. The import duty is 100% of the CIF value of tobacco products imported in the region, with a minimum tax per quantity imported. However, in recent years, to reduce their dependence on income from oil, GCC countries have considered diversifying their sources of income, including by de- veloping reliance on indirect taxes such as excise and VAT. In 2015, a decision was adopted at the 36th GCC summit meeting to implement selective taxes on all imported tobacco products and cultivated raw tobacco grown domestically (GCC Decision number 963/1). A follow-up decision in December 2016 formally agreed to the introduction of an excise tax on tobacco and other products such as sugary and energy drinks, as well as special goods (alcohol and pork meat), in all GCC countries. The decisions at the national level to implement this subregional decision came into force gradually in all GCC countries, starting with Saudi Arabia, which began 22 Directive No. 03/98/CM/WAEMU on the harmonization of Member States’ legislation of excise duties was adopted 22 December 1998. It was amended by Directive No. 03/2009/CM/WAEMU of 27 March 2009 with the objective of harmonizing excise duties within WAEMU. 58 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation in June 2017, followed by Bahrain and the United Arab Emirates (UAE) later that year, Qatar in 2018 and Oman in 2019 (1). Only Kuwait has yet to adapt its national laws accordingly. The excise introduced by the GCC countries has a structure somewhat similar to the import duty on tobacco products: the rate is 100%, but the base was changed from the CIF value to the retail price excluding taxes. The introduction of the excise led to large increases in the price of the most- sold brand of cigarettes in member countries between 2016 and 2018 – by 33% in Bahrain, more than 80% in the UAE and more than 100% in Saudi Arabia (1). In federations such as Canada and the United States – where the central govern- ment has real taxing power and some financial and regulatory control over the states or provinces – tobacco taxes are not harmonized (98–99). Even though there are significant interjurisdictional differences in taxes and prices, and tax harmonization holds great potential to reduce the scope of illicit transactions in the tobacco market, there is little evidence that Canadian provinces or individual states in the United States are interested in tobacco tax harmonization. Tax harmonization is most relevant in the context of further economic integration within a group of countries that are already part of a customs union, but it needs to be planned well to be effective. Discrepancies in law interpretation and a lack of standardization of tobacco product definitions and tax base lead to suboptimal situations. Tax rate alignment, or setting minimum rates, should come after tax structure alignment. It is important that governments support the move towards harmonization and are committed to dedicating enough financial resources and skilled personnel to oversee the entire process. KEY TAKEAWAY 21 In the context of regional economic integration and ongoing discussions regarding the possibility of harmonizing tobacco excise taxation among member countries, the experiences of existing groups can be instructive. So far, only the EU, SACU, WAEMU and, more recently, the GCC have effectively implemented a harmonized approach to excise taxation of tobacco products. Lessons learned indicate that harmonization should be planned well and should not come at the expense of tobacco control. Setting a common minimum specific excise tax, adjusted over time, is the best approach. This ensures that taxes and prices are above a minimum level, encouraging equalization of price levels and at the same time reducing affordability across countries. On the other hand, agreeing on maximum tax rates is a bad policy. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 59 Countries that wish to raise their taxes further for revenue purposes, health concerns or both should be given the space to do so. Concerns about illicit trade provoked by higher tax rates are best dealt with by strong cooperation in administration and enforcement, information sharing and adoption of new technology with common or interoperable information systems. 2.4 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In recent years, awareness of tobacco risks and harms, implementation of tobacco control provisions – especially under the WHO FCTC – and tightening of regulations have resulted in declining sales of cigarettes, primarily in high-income economies. This has changed the dynamics of the tobacco market. In response to these effective tobacco control measures, the tobacco industry has diversified its business by promot- ing a new portfolio of products, which they claim to be technological innovations that supposedly reduce the harms and risks associated with conventional tobacco products, particularly cigarettes. So-called novel tobacco products have been promoted by the tobacco industry as “cleaner alternatives,” “safer alternatives” and “reduced harm/risk products” with no smoke and no ash. On the basis of these claims, they negotiate for less-restrictive regulatory environments within countries. Some of the new products are also mar- keted or promoted for smoking cessation, despite the evidence of this outcome being inconclusive. Where these products are not banned, one of the debates in the global health community concerns the issue of their regulation and taxation. 2.4.1 HEATED TOBACCO PRODUCTS (HTPs) HTPs are tobacco products that produce aerosols containing nicotine and toxic chemicals upon heating of the tobacco or activation of a device containing the tobacco. These aerosols are inhaled by users sucking on or smoking the device. They contain the highly addictive substance nicotine (found in tobacco) as well as non-tobacco additives and are often flavoured. The tobacco in HTPs may be in the form of specially designed cigarettes (e.g. so-called heat sticks or Neo sticks) or pods or plugs. These products include IQOS from PMI, Ploom TECH from Japan Tobacco International (JTI), glo from British American Tobacco (BAT) and PAX from PAX Labs. HTPs differ not only from con- ventional cigarettes but also from ENDS – some of which are called e-cigarettes – as ENDS do not contain tobacco but rather a nicotine solution (see next subsection). However, the boundaries between the different products are becoming increasingly 60 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N difficult to define, given the emergence of so-called hybrid tobacco products that contain both nicotine solution and tobacco. HTPs are currently available in more than 40 countries and are banned in fewer than 10 countries. Even in countries where they are regulated, there is significant variation in the approaches taken to regulation. A variety of factors affect a country’s ability to control and regulate the use of HTPs, including national regulatory pow- ers, enforcement capacity regulatory frameworks, country capacity and tobacco industry interference (1). Most countries tax HTPs at a lower rate than cigarettes and on the kilogram of tobacco as a base when applying a specific or mixed excise (see Table 2.4). The use of such a base may be quite challenging for tax collection, especially because of the difficulty of checking the tobacco content in each stick. In the past, some countries taxed cigarettes per kilogram of tobacco, but today it is common practice to tax them per stick regardless of tobacco content. Table 2.4 Excise taxation of HTPs, first collected for July 2018 – updated for July 2020 OVERALL COMPARISON WITH CIGARETTES Type of excise Base unit is kg, overall rate lower than cigarettes Base unit is sticks, rate is the same as cigarettes Base unit is sticks, rate is lower than cigarettes Other Specific excise Albania, Austria, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Greece, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Montenegro, Netherlands, New Zealand, North Macedonia, Romania, Russian Federation, Slovakia, Slovenia, Sweden, United Kingdom Azerbaijan a, Japan, Ukraine b Armenia, Hungary, Jordan, Italy c, Philippines, Republic of Korea d Montenegro e, Republic of Moldova f , Serbia g Ad valorem excise (base is retail price unless specified other- wise between brackets) Spain, Switzerland Saudi Arabia and United Arab Emirates (base is retail price exclusive of excise and VAT) Indonesia h CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 61 Mixed system (ad valorem compo- nent based on retail price unless specified other- wise between brackets) France, Germany, Poland, Portugal Colombia, Georgia, Israel and West Bank and Gaza Strip (ad valorem excise base is wholesale price) a The specific excise rate applied is the same as for imported cigarettes, higher than the rate applied to domestically produced cigarettes. b The rate is the same as the minimum excise on cigarettes per 1 000 pieces. Rate and structure were effective as of 1 January 2021. c The specific excise rate is defined as 25% of the excise tax on cigarettes based on an equivalency used between cigarettes and HTPs. There are planned increases of this proportion to 40% by 2023. d In 2020 the specific excise rate was only 11% lower than cigarettes. e The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. f Specific excise rate is higher than for cigarettes but, unlike HTPs, cigarettes also face an ad valorem excise. Overall effect of excise is a slightly lower for HTPs. g The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. There is a planned phased increase of this proportion aiming equalization with cigarettes by 2025. h While cigarettes face a specific excise tax rate, HTPs face an ad valorem rate, the highest rate as defined by law, on the basis of a pre-defined minimum price. Sources: (1, 100, 101, WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021 and the Campaign for Tobacco-Free Kids website on Taxation and Price for Heated Tobacco Products https://www.tobaccofreekids.org/what-we-do/global/taxation-price/staging-tax-gap). A study by Liber (102) compared prices of HTPs and cigarettes in 34 countries and showed that while taxes have been systematically lower for HTPs than for cigarettes, prices were higher in half of the countries surveyed. KEY TAKEAWAY 22 HTPs, when taxed, are usually taxed lower than cigarettes, although they generally seem to be priced higher than cigarettes. It is important to remember that HTPs are tobacco products, and the same provisions that apply to tobacco products should apply to them as well. This is articulated in WHO’s information sheet on HTPs (103), which provides guidance on how these products should be regulated, as well as Decision FCTC/COP8(22) for novel and emerging tobacco products. Moreover, MPOWER measures, which help WHO Member States to implement the demand-reduction articles of the WHO FCTC, are applicable to HTPs, in particular, Article 6 for taxation. Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. Furthermore, HTPs contain chemicals not found in cigarette smoke, the health effects of which are not yet known. Independent assessment of industry data demonstrates that more than 20 harmful and potentially harmful chemicals 62 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N are significantly higher in HTP emissions than in cigarette smoke (104). Therefore, there is a need to learn more about these products and the health impacts of their emissions, as well as the impacts of exposure to these emissions. KEY TAKEAWAY 23 Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. From both public health and tax administration perspectives, HTPs should be taxed at the same level and in the same way as tobacco cigarettes. Some countries have already adopted this approach and are taxing HTPs at the same rate per stick as cigarettes (Azerbaijan, Colombia, Georgia, Israel, Japan, Ukraine and West Bank and Gaza Strip). Saudi Arabia and the UAE, which have recently introduced an excise tax on tobacco products as part of the GCC, are now applying the same import duty rate and excise tax structure for cigarettes and HTPs. Continuing developments in technology and changes in products have led to a recommendation to tax HTPs per unit. The definition of unit may vary by product within the HTP category. For example, one unit of IQOS is one heat stick, for Glo it is one Neo Stick and for Ploom TECH it is one tobacco pod. Governments will need to determine the exact definition of a unit for each product allowed on the market. The potential complexity of the market strongly supports limiting the types of HTPs allowed in a country and setting strict regulations to standardize the products as much as possible. Countries can also consider taxing the devices used to consume HTPs, i.e. the holder and the charger (see product description in Annex 3.1). KEY TAKEAWAY 24 HTPs are tobacco products, and they need to be treated as such. Where they are not banned, HTPs need to be strictly regulated and taxed. The recommendation is to tax them at the same level as cigarettes on a per-unit basis. Countries can also consider taxing the devices used for HTP consumption. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 63 2.4.2 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS)23 Products like ENDS and ENNDS have evolved rapidly over the past decade. ENDS heat a solution (e-liquid) containing nicotine, but not tobacco, and other chemicals that may be toxic to people’s health to create an aerosol, which is inhaled by the user. Examples of ENDS include Juul from Juul Labs, Vype from BAT and blu from Imperial Brands (1). Electronic non-nicotine delivery systems (ENNDS) are essentially the same as ENDS, but the e-liquid used generally does not contain nicotine. Upon testing, however, many so-called zero-nicotine solutions are found to contain nicotine (105–107). While generally considered as a single product class, ENDS products constitute a diverse group with potentially significant differences in the production of toxicants and delivery of nicotine. There are several coexisting types of devices for ENDS/ ENNDS on the market, including first-generation or so-called cigalikes, second- generation tank systems and even-larger third-generation or personal vaporizers. Collectively, they are also often referred to as e-cigarettes, vapes or vape pens. Other categories of ENDS include e-hookahs, e-pipes and e-cigars – hence, ENDS is an all-encompassing term for multiple product categories. Some of the products resemble their conventional tobacco counterparts – cigarettes, cigars, cigarillos, pipes or hookahs – while others are shaped more generically like pens, USB memory sticks or basic cylinders. Different forms of nicotine are also used in these ENDS, the most recent one being nicotine salts, which deliver high levels of nicotine (1). There are two types of ENDS/ENNDs products: open systems and closed systems. Open systems are devices that allow the user to buy e-liquids and fill their device with the mixtures they want (with no nicotine, different nicotine concentrations and/or flavours). Closed systems are products that come with a prefilled container (called a cartridge, pod or tank). For the past decade, divisive debates have been waged over the effectiveness of ENDS as smoking cessation aids – especially for tobacco users who are unable to give up the habit – as well as the possibility of ENDS playing a role in public health. However, the evidence remains inconclusive. Despite the tobacco industry and other related industries promoting these products as tools for quitting smoking, current evidence does not support their use as part of a population-based cessation strategy (108). Accordingly, the United States Surgeon General, in January 2020, concluded that 23 It is worth noting that ENDS are not tobacco products and not exactly new products – the technology has been around since the late 1980s (e.g. Premier, Eclipse and Accord). However, the recent generation of these products is new and has more or less piggybacked on the success of e-cigarettes. 64 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N E-cigarettes, a continually changing and heterogeneous group of products, are used in a variety of ways. Consequently, it is difficult to make generalizations about efficacy for cessation based on clinical trials involving a particular e- cigarette, and there is presently inadequate evidence to conclude that e-cigarettes, in general, increase smoking cessation (109). The evidence on the adverse health effects associated with use of ENDS is mounting, and when ENDS are used in combination with smoking – which is the practice of the majority of ENDS users (110) – the adverse health effects of two or more products are combined. However, there are insufficient data to understand the full breadth of these effects, as ENDS have not been on the market long enough for their long-term effects to be established. Nevertheless, the evidence is clear that the aerosols of the majority of ENDS and ENNDS, some of which are cancer causing chemicals. ENDS also contain nicotine, which is highly addictive. In addition, ENDS are associated with increased risk of cardiovascular diseases and lung disorders, as well as adverse effects on the developing fetus during pregnancy (108, 110). For adolescents, the use of nicotine can lead to dependence and may harm brain development. Use of ENDS could also lead to a new generation of nicotine and tobacco users, as seen in some countries, especially since these products are designed to appeal to young people. Although the specific level of risk associated with ENDS has not yet been determined conclusively, these products are undoubtedly harmful. Therefore they should be strictly regulated if allowed to be sold in domestic markets, and must be kept away from children. Taxation will be a key component of regulation, since it is an effective tool for influencing consumer behaviour. Some countries have taken the bold decision to completely ban these products. Approaches that have been taken range from partial to comprehensive bans, and ENDS/ENNDS products were banned in more than 30 countries in 2018.24 In other countries, they are regulated as, for example, consumer products, pharmaceutical products or tobacco products, or they are completely unregulated. WHO recom- mends that where ENDS/ENNDS are not banned, they should be regulated to achieve the following objectives: 1. prevent the initiation of ENDS/ENNDS by non-smokers and youth, with special attention to vulnerable groups; 2. minimize as much as possible potential health risks for ENDS/ENNDS users and protect non-users from exposure to their emissions; 3. prevent unproven health claims being made about ENDS/ENNDS; and 24 Data collected for the WHO RGTE 2019. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 65 4. protect tobacco control activities from all commercial and other vested interests related to ENDS/ENNDS, including the interests of the tobacco industry. ENDS/ENNDS present a risk to youth, who have taken up their use in high numbers in some countries, including Canada and the United States (1, 111). The Juul brand, for example, has quickly gained a significant e-cigarette market share in the United States (112–113). Its marketing and popularity have led the United States Food and Drug Administration (FDA) to raise serious concerns and to seek solutions to effectively prevent youth from taking up the use of ENDS/ENNDS (114). The city of San Francisco banned the sale of e-cigarettes in June 2019 (115). In addition to posing a risk for initiation by youth, ENDS can attract non- tobacco users or prevent current smokers from quitting. Taxation could play a role in preventing the uptake of these products, specifically among non-smokers, vulnerable groups, children and adolescents. KEY TAKEAWAY 25 The long-term health effects of ENDS/ENNDS products are still unknown, but they are clearly harmful to health. Furthermore, evidence on the effectiveness of ENDS products as a smoking- cessation aid remains inconclusive. Taxing these products could play a role in preventing their uptake, specifically among non-smokers, vulnerable groups, children and adolescents. Price elasticity of demand for ENDS products In the context of taxation, it is important to ask whether demand for ENDS is price-responsive. Preliminary evidence, although almost exclusively focused on e-cigarette data from the United States, indicates that this is the case: demand for e-cigarettes may be even more price-responsive than the demand for conventional cigarettes, so taxes can be used to deter initiation by never-users (116–123). Most of the studies of price elasticity of demand for ENDS products also demonstrate that e-cigarettes and conventional cigarettes are partial substitutes – that is, they show positive cross-price elasticity. The magnitude of the elasticity indicates the degree of substitutability between products: the higher its value, the closer the products are to being substitutes, with higher cigarette prices being associated with increased e-cigarette sales. Some of the studies also show a substitutability effect in the other direction, with increased prices for e-cigarettes leading to an increase in conven- tional cigarette use (117, 120). All of the studies show evidence of substitutability except for one (124), which differentiates between exclusive and dual users and shows no evidence of substitution between e-cigarettes and conventional cigarettes. 66 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The presence of concurrent (or dual) use – consumers using both conventional cigarettes and ENDS products – complicates results and highlights the need for more research in this area to better disentangle the different effects. Liber et al. (125) analysed sales prices in a sample of 45 countries and concluded that comparable units of conventional cigarettes cost less than disposable e-cigarettes. The units considered for pricing e-cigarettes included both the e-liquid and the rechargeable device. Taken alone, the price of e-liquids is on average much lower than that of cigarettes in high-income countries and the same in LMICs. The time needed to buy back a rechargeable device is estimated to be less than two weeks in most countries. One can argue that increasing price differentials by further increasing taxes on regular cigarettes could be effective in driving current smokers of regular cigarettes to e-cigarettes (126) as a potentially lower-risk alternative (127). However, the ef- fectiveness of ENDS as smoking cessation devices is still being debated; a study by Sweet et al. (128) shows that dual use of e-cigarettes as a potential tool for cessation was effective only in the short term. Moreover, significantly more smokers said they would quit if cigarette prices doubled and e-cigarettes were not available (122) or that they would never have become addicted to nicotine if e-cigarettes had not been so readily available (129). Once an e-cigarette user is addicted to nicotine, there is a risk of initiating traditional tobacco products use (130). In general, cessation can be better facilitated by governments via stronger implementation of the other tobacco control policies that have been proven effective at reducing use. KEY TAKEAWAY 26 Few studies are available on the price elasticity of ENDS products, and the available data come almost exclusively from the United States. These early studies indicate that demand for e-cigarettes will go down as the price of e-cigarettes increases. Generally, the results also show that cigarettes and e-cigarettes are partial substitutes, where an increase in cigarette price would increase the demand for e-cigarettes while reducing demand for cigarettes. But these results do not differentiate between people who are exclusive cigarettes or e-cigarette users and those who are users of both products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 67 Tax structure Different countries impose different tax structures on ENDS/ENNDS products25 (see details in Table 2.5). The Republic of Korea, for example, imposes a specific tax per millilitre of ENDS/ENNDS e-liquid (131), while Indonesia imposes an ad valorem tax on the retail price of the e-liquid; the maximum rate allowed by law for tobacco products (132). In the United States, there is no common way to tax e-cigarettes among the states that do tax them (133–134). The situation is similar in the EU, where new and emerging nicotine and tobacco products are not currently covered by the tobacco tax directive, and Member States may apply a national tax as they see fit under their own rules. All the EU countries that tax ENDS products apply a specific excise per millilitre of e-liquid. These different tax treatments have the potential to distort the functioning of the internal market. In February 2020, the European Commission concluded that the current provisions of the harmonized directive are no longer relevant for the taxation of ENDS and HTPs, and this is a source of concern from the internal market perspective (135). In June 2020, the Member States of the EU reiterated that it is urgent and necessary to upgrade the EU regulatory framework by harmonizing defini- tions and the tax treatment of novel products such as ENDS/ENNDS and HTPs (2). Table 2.5 Types of excises applied on ENDS/ENNDS products e-liquids globally and in individual states in the United States, as of July 2019 (updated as of July 2020 for all countries except the United States) TYPE OF EXCISE COUNTRIES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Albania, Kazakhstan, Kyrgyzstan, Pakistan, Portugal, Republic of Korea, Romania, Russian Federation, Slovenia, Sweden Azerbaijan, Cyprus, Estonia, Finland, Georgia, Greece, Hungary, Italyb, Latvia, Lithuania, Montenegro, Morocco, North Macedonia, Philippines, Serbia Ad valorem (% of retail price or import value) Bahraina Indonesia, Jordan, United Arab Emirates, Yemen TYPE OF EXCISE INDIVIDUAL STATES IN THE UNITED STATES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Delaware, Illinois (Chicago), Cook County, Louisiana, Ohio, Puerto Ricoc, Connecticutc Kansas, North Carolina, Washington, West Virginia, Wisconsin 25 The focus is on the e-liquid used for ENDS/ENNDS products. 68 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ad valorem (% of wholesale/ distributor price) Alaska (Juneau, Matanuska- Susitna Borough), California, Illinois, Maine, Maryland (Montgomery County), Minnesota, Nevada, Pennsylvania, Vermont, Washington DC, Virgin Islandsc New York Mixed New Jersey, New Mexico a Tax applied to e-shisha (or e-hookah) because e-cigarettes are banned in Bahrain. b Italy imposes differential rates for nicotine and non-nicotine containing liquids. c States in which it is unclear if only ENDS or both ENDS and ENNDS products are taxed with an excise. Sources: (1, 135, complementary data from Frank Chaloupka and WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021). Table 2.6 provides reference material on the pros and cons of different considerations for determining the tax structure and base of ENDS/ENNDS products e-liquids. Table 2.6 Excise tax options for ENDS/ENNDS products e-liquids TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Specific Volume of nicotine- containing e-liquid (regardless of concentration) 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only the volume needs to be determined 1. Difficult to compare if tax equivalencya with cigarettes is sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarettes and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed 5. May encourage more do-it- yourself (DIY) products where e-liquids are mixed by the users themselves, which increases the risk of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 69 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of e-liquid regardless of nicotine presence 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only volume needs to be determined 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Does not require laboratory capacity to detect the presence of nicotine in liquids 1. Difficult to compare if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed Volume of all e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL are affected the most; also reduces the price gap between different products 2. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 3. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 4. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 5. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to user 1. Difficult to compare with cigarettes if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed 70 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of nicotine- containing e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL affected the most; also reduces the price gap between different products 2. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as the nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 3. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 4. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to the user 1. Difficult to compare if tax equivalency with cigarettes is being sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids as there is no simple way currently available to determine whether the e-liquid contains nicotine; self-declarations by industry are not sufficient 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed Ad valorem Producer price/ CIF value of nicotine- containing e-liquid (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 71 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Producer price/ CIF value of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Retail price of nicotine- containing e-liquids (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 2. Requires capacity to monitor the market to assess market prices 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled in an e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risk of accidents, illness and death 72 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Retail price of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires capacity to monitor the market to assess retail prices 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Ad valorem with minimum specific or mixed Ad valorem + min specific 1. Volume of nicotine- containing e-liquids will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of nicotine- containing e-liquids will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up. b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products 1. Requires capacity to monitor the market to assess retail prices 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotin; (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Difficult to set a minimum specific excise amount/specific excise amount, especially if tax equivalency with cigarettes is sought 5. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 73 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Ad valorem + min specific 1. Volume of all e-liquids regardless of nicotine presence will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of all e-liquids regardless of nicotine presence will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 1. Requires capacity to monitor the market to assess retail prices 2. Difficult to set a minimum specific excise amount/specific excise amount, especially if equivalency with cigarettes is sought 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels a Tax equivalency can be measured in different ways: (1) in terms of tax burden (as % of the retail price) or (2) as the exact amount of tax for equivalent quantities (assuming an equivalency between a certain volume of e-liquid and a pack of cigarettes). b There is also the option to use the producer price/CIF value as a base for the ad valorem component, but it is a weaker option because the base is difficult to ascertain and therefore prone to undervaluation. Note: Table compiled following a WHO Expert Meeting on Taxation of Electronic Nicotine and Non-Nicotine Delivery Systems (ENDS/ENNDS), Geneva, Switzerland, 2–4 September 2019. There is currently a lack of evidence on the practical challenges being faced by countries favouring one approach over the other. Furthermore, such data are difficult to obtain because the nature of the market is constantly changing. However, a clear recommendation can be made with regard to which e-liquids to tax. As indicated in Table 2.5, some countries tax all e-liquids – whether or not they contain nicotine (ENDS and ENNDS products) – while some tax only nicotine- containing e-liquids (ENDS products). As shown in Table 2.6, there is evidence that in a number of instances, ENNDS products do contain some nicotine. Additionally, ENNDS products are not harmless (136–137). It is therefore recommended that all e-liquids be taxed for both ENDS and ENNDs products. The question of whether to employ differential taxation based on nicotine content seems reasonable from a health perspective, since nicotine is a toxic substance. However, this would likely create an additional burden for tax administrators as 74 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N they would need to determine the nicotine concentration of e-liquids on the market. Additionally, this may no longer be relevant, as advancements in technology indicate that other features of the product can influence nicotine delivery beyond the actual concentration of the e-liquid. It is now possible to increase nicotine delivery at low nicotine concentrations by increasing battery power (by reducing resistance or increasing voltage) (138). In terms of implementation, while most countries seem to have adopted a specific excise tax on ENDS/ENNDS e-liquids per millilitre, one benefit of implementing ad valorem taxation is that it seems relatively easier to regulate in the context of a large heterogeneity of products. However, it is essential that the tax be applied on the retail price value of the products, as this base is easier to ascertain than any other value that could be declared by the manufacturer. It is also important to add that regulation of the characteristics of ENDS/ENNDS products is essential, and it should be implemented along with any tax policy adopted. Regulations should include: 1. setting a maximum nicotine concentration per millilitre to safeguard public health, including reducing the risk of dependence, especially among youth; 2. setting a maximum volume for cartridges to reduce toxicants exposure and possibly limit use; 3. setting a maximum capacity for refill containers to reduce toxicants exposure and possibly limit use; 4. setting a maximum battery power to reduce the possibility of influencing nicotine and toxicant delivery; and 5. taxing nicotine regardless of its source (e.g. tobacco, eggplant, synthetic). Countries may choose to impose an excise tax on ENDS and ENNDS devices26 as well. The easiest type of tax would be an ad valorem tax based on the declared retail price. If countries choose not to impose an excise tax on these products, they should at least impose the regular VAT or sales tax rate. Imposing an excise tax on devices can be challenging from an administrative perspective, as all components need to be clearly defined and classified as devices for ENDS/ENNDS consumption. For example, if the device is assembled after importation and some parts may be used for other purposes than ENDS/ENNDS consumption, authorities may face a challenge in detecting and differentiating which component parts would be subject to excise tax and which would not. 26 See Annex 2.3 for an overview of elements of devices used in ENDS/ENNDS products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 75 KEY TAKEAWAY 27 There is currently not enough evidence to recommend one tax structure over another for ENDS/ENNDS products. However, it is clear that taxing the e-liquids used for consumption is key. The excise tax should be applied on all e-liquids, whether or not they contain nicotine. If the preferred type of excise tax is ad valorem, it should be applied to the retail price. Countries can consider taxing devices as well, but they need to adequately assess their administrative capacity to do so. Policy-makers need to be mindful of the diversity and rapid evolution of ENDS/ ENNDS products and adjust accordingly. Regulation must reflect this reality so that loopholes will not be exploited by the industry. For example, ENDS/ENNDs products include not only e-cigarettes, vapes and vape pens but also other categories such as e-hookahs, e-pipes and e-cigars. Lawmakers need to be clear about how ENDS/ENNDS products are defined so that subcategories do not fall under the radar when regulation comes into effect. Definitions will also be relevant when it comes to taxation. An unclear definition can lead to a seemingly contradictory situation, such as in Bahrain, where e-cigarettes are banned but e-hookahs are not.27 Finally, while policy-makers need to be mindful of the emergence of new products and must take appropriate actions to protect the health of their citizens, it is important to remember that the overwhelming share of nicotine consumption remains that of tobacco products, especially cigarettes. The total market value of ENDS/ENNDS and HTPs sales in 2018 was less than 2.2% of the total market value, while cigarette sales alone accounted for 91% of the same total market value (139–140). 2.5 CONCLUSIONS An overview of excise tax application globally reveals a broad variety of price and tax levels, as well as structures used for taxing tobacco products, in particular, cigarettes. Some trends, however, indicate that tax and price levels are higher among higher- income countries. The rate of taxes also matters: higher tax rates are correlated with higher prices, and higher prices change behaviour, which leads to a reduction in consumption. More countries are moving away from ad valorem taxes and towards either mixed or specific excise systems, and there are few countries that do not impose any excise tax on cigarettes. 27 In Bahrain, the Ministry of Production and Trade Decision 38 of 2013 banned e-cigarettes, while the official list of excisable products from the Ministry of Finance includes e-shishas (or e-hookahs), making them apt to be taxed and therefore considered legal. 76 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Significantly increasing the taxes and prices of tobacco products is the most effective and cost-effective policy to control tobacco use. Increased taxes – which are passed on to smokers as higher prices – reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face several challenges, ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues, as well as the perceived contribution of the tobacco sector to economic development. In designing tax policy, the tax structure adopted not only affects consumption overall, it also shapes the market structure. Ad valorem taxation incentivizes industry to set prices lower than specific taxation does. Evidence suggests that under a specific tax, the price gap between premium and lower-priced products is narrower, therefore reducing incentives for substitution to lower-priced products following a tax increase. However, as industry consolidates producers and widens its portfolio of products, new evidence indicates that the industry is introducing cheaper brands while increasing the price of its expensive brands, therefore, paradoxically, widening the price gap between its products. Evidence also suggests that prices are higher under a specific excise tax structure. Additionally, from a tax administration perspective, a specific tax is easier to imple- ment, since only the quantity produced needs to be ascertained rather than the value of the product. Another aspect of tax structure is the use of tiered taxation – that is, tax rates that vary on the basis of different product characteristics. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax. Tiered taxation encourages substitution from premium to cheaper brands, maintaining smoking prevalence and reducing the health im- pact of tax rate increases. In addition to leading to lower prices, tiered taxation is difficult to administer and creates opportunities for the tobacco industry to avoid and evade taxes. The design of a tax structure must also consider the base on which tax is applied. The choice of base should lead to the highest possible effect on price and revenue. For specific taxation, the tax base is the quantity. When the tax is ad valorem, the CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 77 choice of the tax base is important not only for health considerations, through its effect on consumption, but also for tax revenue generation, as well as industry profits. An ad valorem tax based on the producer price, or CIF value, gives tobacco manufacturers opportunities to reduce their tax liability, especially when they control the distribution system through related parties. The best practice in an ad valorem (or mixed) excise structure is to use the retail price as the tax base and introduce a minimum excise tax per pack. Other tax design considerations include the importance of using automatic adjust- ments and indexation to inflation and income growth for the specific excise tax in order to avoid erosion of the tax over time. Emerging evidence indicates that tobacco taxation does not always achieve the intended results, because the tobacco industry finds ways to circumvent it. Non- tax policies such as pricing regulation (in particular, minimum mark-ups or price floors/minimum prices) may be seen as a complementary approach to ensuring a high price level and discouraging consumption of tobacco products. So far, these policies have not proven to increase average prices. A price floor is likely to lead to increased industry profits, giving the industry greater funds for its marketing strategies (such as the introduction of new products), and lower tax revenues for governments. By reducing price competition, a price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifications). However, where powerful multinationals are operating in certain markets with presence in all market segments and with the capability to overshift a tax on some brands while undershifting the tax on others, or where price promotions cannot be banned, minimum price policies may help increase the effectiveness of tax increases. Other non-tax policies affecting price levels are those relating to promotional dis- counts for tobacco products and the sale of single sticks of cigarettes. Both should be completely banned. The ban of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Higher taxes are the most effective way to dissuade consumption, with the added benefit of raising money for the government – money that can be earmarked for health and education programs, rather than going as profits to the tobacco industry. Additionally, in order to make excise tax on tobacco products more effective in reduc- ing overall tobacco use and in line with the recommendation of the Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce 78 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N the demand for tobacco), all tobacco products need to be taxed in a comparable way; the focus should not be on cigarettes only. Tax choices and reforms have various and sometimes conflicting consequences for the market. For example, there might be a trade-off between quantity and variety or perceived quality implications. It is important for the government to recognize that firms respond strategically to changes in tax policy. Close monitoring of the market is necessary to form correct expectations about industry responses and enable estimates of the impact of a tax increase on consumption and tax revenue. To estimate the total effect of a tax increase on demand for tobacco products and tax revenue, it is important to use correct estimates of the own-price elasticity of demand, the cross-price elasticity and the income elasticity of demand. It is also important to use updated estimates of demand elasticities, as the environment within which consumers make decisions continues to change. For example, financial crises or successful tobacco control interventions can be expected to shift demand and change elasticity. Another key measure of the impact of tax policy is the tax base elasticity. Policy- makers need to be mindful of the three key components of tax base elasticity: (1) the price elasticity of demand of tobacco, (2) the share of the tax in the consumer price and (3) the degree of pass-through of the excise tax rate increase to consumer price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. It is important to acknowledge that if tax increases lead to increases in prices be- low concurrent increases in income levels, they will not be effective in reducing consumption, as tobacco remains a normal good in most countries. Policy-makers need to account for affordability when considering tax increases. They should ensure that tax increases are high enough to increase prices above income growth so that consumption goes down effectively. When designing tax policy and deciding on the right level to impose, policy-makers need to assess and project the impact of their policy decisions. Monitoring and evaluation are important. Tools for measuring impact can be very helpful, and several such tools exist. The WHO ISPT, for example, looks not only at the impact of tax policy but also at a set of tobacco control policies, and this enables national policy-makers and other tobacco control experts to explore the potential impact of CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 79 future tobacco control policies. The tool uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. More specific to tobacco tax policy, the WHO TaXSiM assesses the impact of any excise tax increase and change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the effectiveness of their policies and whether those policies have an impact on tobacco use over time. The implementation of the MPOWER package is one useful indicator for assessing tobacco control overall. Tobacco taxation works best if it is implemented as part of a comprehensive MPOWER package. The tax share in the retail price of a selected tobacco product is one indicator of the effectiveness of tax policy, but a more important one is affordability, that is, whether tax increases do lead to price increases that are above income and general price increases. A useful indicator to assess the performance of the tax policy overall is the Tobacconomics Cigarette Tax Scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). Domestic policies in agriculture, industry, trade, finance and labour all have the potential to create or support incentives at different stages in tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. Greater domestic policy coherence should be pro- moted across different sectors of the government to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. Differential tax structures and rates have the potential to distort the functioning of the internal market. Harmonization of tobacco taxation ensures the establishment and proper functioning of a single market; prevents tax revenue erosion, tax avoidance and tax evasion; and protects people’s health. In this context, tax competition, where countries simply undercut each other’s tax rate, might prevent governments from achieving their tobacco control objectives and raising sufficient funds to pursue public health policies. To avoid such a race to the bottom, countries can establish minimum tax rates on all tobacco products. A common high minimum specific excise tax is the best approach to ensure that taxes and prices are above a minimal level. 80 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N In recent years, the world has experienced the rise of new and emerging tobacco and nicotine products, including ENDS/ENNDS and HTPs, which the industry claims are safer than traditional tobacco products. The evidence so far suggests that these products could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of these products as well as initiation, cessation and switching of tobacco use behaviours among different socioeconomic groups, are not yet clear. Until more evidence for the claimed benefits of these tobacco products is available, caution should be taken in developing tax policy. Therefore, the current recommendation is for HTPs to be taxed at the same level as cigarettes on a per-unit basis regardless of tobacco content. In countries where they are not banned, ENDS/ENNDS products must be regulated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS products taxation. Nicotine- and non-nicotine-containing e-liquids should be taxed equally. If ad valorem excise is chosen as the structure, the base should be applied on the retail price. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While the evolution of new and emerging tobacco and nicotine delivery systems merits the attention of administrators and regulators, it is worth remembering that the bulk of the world’s tobacco consumption is still in conventional cigarettes. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 81 REFERENCES 1. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 2. Outcome of proceedings: Council conclusions concerning the structure and rates of excise duty applied to manufactured tobacco. Brussels: Council of the European Union; 2 June 2020 (8483/20; https://www.consilium.europa.eu/media/44235/st08483-en20.pdf, accessed 28 September 2020). 3. Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco: Article 7, paragraph 2. Brussels: OJEU. 2011;176:24–36 (https://eur-lex.europa. eu/LexUriServ/LexUriServ.do?uri=OJ:L:2011:176:0024:0036:EN:PDF, accessed 28 September 2020). 4. U.S. National Cancer Institute and World Health Organization.The economics of tobacco and tobacco control. NIH Publication No. 16-CA-8029A. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute; 2016 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/brp/tcrb/monographs/monograph-21, accessed 28 September 2020). 5. Drope J, Schluger N, Cahn Z, Drope J, Hamill S, Islami F, et al. The tobacco atlas. Atlanta: American Cancer Society and Vital Strategies; 2018 (https://tobaccoatlas.org/wp-content/uploads/2018/03/ TobaccoAtlas_6thEdition_LoRes_Rev0318.pdf, accessed 24 January 2021). 6. Sector reports, cigarettes. In: GlobalData [website]. London: John Carpenter House; 2019 (restricted access). 7. Vellios N, Ross H, Perucic AM. 2018 trends in cigarette demand and supply in Africa. PLOS ONE. 2018;13(8):e0202467 (https://doi.org/10.1371/journal.pone.0202467, accessed 28 September 2020). 8. Keen MJ. The balance between specific and ad valorem taxation. Fiscal Studies. 1998;19(1):1–37 (https://www.ifs.org.uk/fs/articles/fsmick1.pdf, accessed 24 January 2021). 9. Kay JA, Keen MJ. The structure of tobacco taxes in the European community. IFS report series. Institute for Fiscal Studies; 1982. 10. Kay JA, Keen MJ. How should commodities be taxed?: Market structure, product heterogeneity and the optimal structure of commodity taxes. Euro Econ Rev. 1983;23(3):339–58. 11. Kay JA, Keen MJ. Commodity taxation for maximum revenue. Public Finan Q. 1987;15(4):371–85. 12. Kay JA, Keen MJ. Product quality under specific and ad valorem taxation. Public Finan Q. 1991;19(2):238–47. 13. Delipalla S, Keen MJ. The comparison between ad valorem and specific taxation under imperfect competition. J Public Econ. 1992;49(3):351–67. 14. Suits DB, Musgrave RA. Ad valorem and unit taxes compared. Q J Econ. 1953;67(4):598–604. 15. Skeath SE, Trandel GA. A pareto comparison of ad valorem and unit taxes in noncompetitive environments. J Public Econ. 1994;53(1):53–71. 16. Myles G. Imperfect competition and the optimal combination of ad valorem and specific taxation. Int Tax Public Finance. 1994;3:29–44. 17. Effectiveness of tax and price policies for tobacco control. Geneva: WHO Press, World Health Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications. iarc.fr/Book-And-Report-Series/Iarc-Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And- Price-Policies-For-Tobacco-Control-2011, accessed 28 January 2021) 18. Bevan J. GCC taxation. Aylesbury (UK): Rothmans Exports Limited; 1992 (https://www.industrydocuments. ucsf.edu/docs/yxpy0198, accessed 29 September 2020). 19. Morton IT. Note from IT Morton to I Birks regarding agreement on specific taxation; 8 February 1992 (https://www.industrydocuments.ucsf.edu/docs/#id=nxpy0198, accessed 29 September 2020). 20. Galgut R. Oman. BAT (UK and Export) Limited; 16 April 1992 (https://www.industrydocuments.ucsf. edu/docs/xxpy0198, accessed 29 September 2020). 21. Common excise tax agreement of the states of the Gulf Cooperation Council [website]. Abu Dhabi: UAE Ministry of Finance; 2016 (https://www.mof.gov.ae/En/lawsAndPolitics/govLaws/Documents/ Common%20Excise%20Tax%20Agreement.pdf, accessed 29 September 2020). 82 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 22. Brock B, Choi K, Boyle RG, Moilanen M, Schillo B. Tobacco product prices before and after a statewide tobacco tax increase. Tob Control. 2016;25(2):166–73. 23. Hanson A, Sullivan R. The incidence of tobacco taxation: evidence from geographic micro-level data. Natl Tax J. 2009;62(4):677–98. 24. Delipalla S, O’Donnell O. Estimating tax incidence, market power and market conduct: the European cigarette industry. Int J Ind Organ. 2001;19(6):885–908. 25. Keeler TE, Hu TW, Barnett PG, Manning WG, Hai-Yen S. Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxation. J Health Econ. 1996;15:499–512. 26. Linegar DJ, van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2017;0:1–7 (https://tobaccocontrol.bmj.com/content/27/1/65, accessed17 February 2021). 27. Villanti AC, Collins LK, Niaura RS, Gagosian SY, Abrams DB. Menthol cigarettes and the public health standard: a systematic review. Public Health. 2017;17:983 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5747135/, accessed 24 January 2021). 28. Directive 2014/40/EU of the European Parliament and of the Council of 3 April 2014 on the approximation of the laws, regulations and administrative provisions of the member states concerning the manufacture, presentation and sale of tobacco and related products and repealing Directive 2001/37/EC: Article 7. Brussels: OJEU. 2014;57:1–38 (https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=OJ:L:2014:127:FULL&from=EN, accessed 24 January 2021). 29. Delipalla S, Keen MJ. Product quality and the optimal structure of commodity taxes. J Public Econ Theory. 2006;8:547–54. 30. Obara T, Tsugawa S. The welfare comparison of ad-valorem tax and specific tax with both quality and quantity choice of a consumer. Theor Econ Lett. 2017;7:1797–1813 (https://www.scirp.org/journal/ paperinformation.aspx?paperid=79862, accessed 24 January 2021). 31. Anderson S, de Palma A, Kreider B. Tax incidence in differentiated product oligopoly. Journal of Public Econ. 2001;81(2):173–92. 32. Razzolini L, Shughart WF, Tollison RD. On the third law of demand. Econ Inq. 2003;41:2. 33. Sobel RS, Garrett TA. Taxation and product quality: new evidence from generic cigarettes. J Polit Econ. 1997;105:880–7 (http://faculty.citadel.edu/sobel/All%20Pubs%20PDF/Generic%20Cigarettes. pdf, accessed 24 January 2021). 34. Nargis N, Fong GT, Chaloupka FJ, Li Q. The choice of discount brand cigarettes: a comparative analysis of International Tobacco Control (ITC) surveys in Canada and the United States (2002-2005). Tob Control. 2014;23:i86–i96 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/suppl_1/i86. full.pdf, accessed 24 January 2021). 35. Chaloupka FJ, Peck R, Tauras JA, Xu X, Yurekli A. Cigarette excise taxation: the impact of tax structure on prices, revenues and cigarette smoking. Cambridge: National Bureau of Economic Research; 2010 (NBER Working Paper 16287; https://www.nber.org/system/files/working_papers/w16287/w16287. pdf, accessed 24 January 2021). 36. Cigarettes in India, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 37. Gilmore AB, Tavakoly B, Taylor G, Reed H. Understanding tobacco industry pricing strategy and whether it undermines tobacco tax policy: the example of the UK cigarette market. Addiction. 2013;108(7):1317–26 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3746125/, accessed 24 January 2021). 38. Marsh L, Cameron C, Quigg R, Hoek J, Doscher C, McGree R, et al. The impact of an increase in excise tax on the retail price of tobacco in New Zealand. Tob Control. 2016;25(4):458–634. 39. Brock B, Schillo BA, Moilanen M. Tobacco industry marketing: an analysis of direct mail coupons and giveaways. Tob Control. 2015;24:505–8. 40. Chaloupka FJ, Warner KE. The economics of smoking. Handbook Health Econ. 2000;1:1539–1627. 41. Shang C, Chaloupka FJ, Fong G, Thompson M, O’Connor RJ. The association between tax structure and cigarette price variability: findings from the ITC Project. Tob Control. 2015;24:iii88–iii93 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/24/Suppl_3/iii88.full.pdf, accessed 24 January 2021) 42. Li Q, Hyland A, Fong GT, Jiang Y, Elton-Marshall T. Use of less expensive cigarettes in six cities in China: findings from the International Tobacco Control (ITC) China Survey. Tob Control. 2010;19(suppl CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 83 2):i63–i68 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/19/Suppl_2/i63.full.pdf, accessed 24 January 2021). 43. Hyland A, Laux FL, Higbee C, Hastings G, Ross H, Chaloupka FJ, et al. Cigarette purchase patterns in four countries and the relationship with cessation: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15(suppl III):iii59–iii64 (https://tobaccocontrol.bmj.com/ content/tobaccocontrol/15/suppl_3/iii59.full.pdf, accessed 24 January 2021). 44. Tsai YW, Yang CL, Chen CS, Liu, TC, Chen PF. The effect of Taiwan’s tax induced increases in cigarette prices and brand switching and the consumption of cigarettes. Health Econ. 2005;14:627–41. 45. Cummings KM, Hyland A, Lewit E, Shopland D. Use of discount cigarettes by smokers in 20 communities in the United States, 1988-1993. Tob Control. 1997;6(suppl 2):S25–S30 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/6/suppl_2/S25.full.pdf, accessed 24 January 2021). 46. Ross H, Tesche J. Undermining government tax policies: common strategies employed by the tobacco industry in response to increases in tobacco taxes. Economics of Tobacco Control Project, School of Economics, University of Cape Town and Tobacconomics, Health Policy Center, Institute for Health Research and Policy. Chicago: University of Illinois at Chicago; 2016 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC5747349/, accessed 26 January 2021). 47. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 48. Hiscock R, Branston JR, McNeill A, Hitchman SC, Partos TR, Gilmore AB. Tobacco industry strategies undermine government tax policy: evidence from commercial data. Tob Control. 2017;0:1–10 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 49. Filippidis F, Laverty A, Hone T, Been JV, Millett C. Association of cigarette price differentials with infant mortality in 23 European Union countries. JAMA Pediatr. 2017;171(11):1100–1106 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5710372/#:~:text=Cigarette%20price%20increases%20across%20 23,priced%20cigarettes%20during%20this%20period, accessed 26 January 2021). 50. López-Nicolás A, Belén Cobacho M, Fernández E. The Spanish tobacco tax loopholes and their consequences. Tob Control. 2013;22(E1):e21. 51. Ballester L, Auchincloss A, Robinson L, Mayne S. Exploring impacts of taxes and hospitality bans on cigarette prices and smoking prevalence using a large dataset of cigarette prices at stores 2001–2011, USA. Int J Environ Res Public Health. 2017;14(3):e318 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5369154/, accessed 24 January 2021). 52. Liber A, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia survey. Tob Control. 2015;24:iii83-iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed on 26 January 2021). 53. Feighery E, Ribisl K, Schleicher N, Zellers L, Wellington N. How do minimum cigarette price laws affect cigarette prices at the retail level? Tob Control. 2005;14:80–85 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/14/2/80.full.pdf, accessed 26 January 2021). 54. Tynan MA, Ribisi KM, Loomis BR. Impact of cigarette minimum price laws on the retail price of cigarettes in the USA. Tob Control. 2012;22:e1. 55. Huang J, Chriqui JF, DeLong H, Mirza M, Diaz MC, Chaloupka FJ. Do state minimum markup/price laws work? Evidence from retail scanner data and TUS-CPS. Tob Control. 2016;25:i52–i59 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i52.full.pdf, accessed 26 Jan 2021). 56. Whitehead R, Brown L, Riches E, Rennick L, Armour G, McAteer J, et al. Strengths and limitations of tobacco taxation and pricing strategies. Edinburgh: NHS Health Scotland; 2018 (http://www. healthscotland.scot/media/1829/rapid-evidence-review-strengths-and-limitations-of-tobacco- taxation-and-pricing-strategies.pdf, accessed 27 January 2021). 57. Golden S, Farrelly M, Luke D, Ribisl K. Comparing projected impacts of cigarette floor price and excise tax policies on socioeconomic disparities in smoking. Tob Control. 2016;25:i60-i66 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i60.full.pdf, accessed 27 January 2021). 58. Doogan NJ, Wewers ME, Berman M. The impact of a federal cigarette minimum pack price policy on cigarette use in the USA. Tob Control. 2018;27(2):203–08. 84 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 59. Marynak K, Xu X, Wang X, Holmes CB, Tynan MA, Pechacek T. Estimating the impact of raising prices and eliminating discounts on cigarette smoking prevalence in the United States. Public Health Rep. 2016;131:536–43 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4937114/#:~:text=We%20found%20 that%20each%20price,smokers%20aged%20%E2%89%A526%20years., accessed 27 January 2021). 60. McLaughlin I, Pearson A, Laird-Metke E, Ribisl K. Reducing tobacco use and access through strengthened minimum price laws. Am J Public Health. 2014;104(10):1844–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4937114/#:~:text=We%20found%20that%20each%20price,smokers%20aged%20 %E2%89%A526%20years., accessed 27 January 2021). 61. European Commission v. French Republic, C-197/08, European Court of Justice 4 March 2019 (https:// eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A62008CC0197, accessed on 27 January 2021). 62. Tobacco price promotion: policy responses to industry price manipulation. Boston: Public Health and Tobacco Policy Center; October 2011 (http://tobaccopolicycenter.org/documents/Price%20 Promotion%20Overview%20FINAL.pdf, accessed 28 September 2020). 63. Golden SD, Holt Smith M, Feighery EC, Roeseler A, Rogers T, Ribisl KM. Beyond excise taxes: a systematic review of literature on non-tax policy approaches to raising tobacco product prices. Tob Control. 2016;25:377–85. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/4/377.full.pdf, accessed 27 January 2021). 64. Branston J, Gilmore S. The case for Ofsmoke: the potential for price cap regulation of tobacco to raise ₤500 million per year in the UK. Tob Control. 2014;23(1):45–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3812939/, accesed 27 January 2021). 65. von Lampe K, Kurti M, Johnson J. “I’m gonna get me a loosie” understanding single cigarette purchases by adult smokers in a disadvantaged section of New York City. Prev Med Rep. 2018;12:182–5. (https:// www.sciencedirect.com/science/article/pii/S2211335518302195, accessed 27 January 2021). 66. de Ojeda A, Barnoya J, Thrasher JF. Availability and costs of single cigarettes in Guatemala. Nicotine Tob Res. 2013;15(1):83–7 . 67. Lal P, Kumar R, Ray S, Sharma N, Bhattarcharya B, Mishra D, et al. The single cigarette economy in India - a back of the envelope survey to estimate its magnitude. Asian Pac J Cancer Prev. 2015;16(13):5579-82 (http://journal.waocp.org/article_31295_9ebd3981652934bbf5d78ec066b6a39d.pdf, accessed 27 January 2021). 68. Coady MH, Chan CA, Sacks R, Mbamalu IG, Kansagra SM. The impact of tax on purchasing behaviors among New York City smokers. Am J Public Health. 2013;103(6):e54-e60 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC3698745/, accessed 27 January 2021). 69. Cigarettes in Thailand, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 70. Cigars in Germany, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 71. Report from the Commission to the Council on Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2018 (COM(2018) 17 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/report_excise_duty_manufactured_ tobacco_12012018_en.pdf, accessed 28 September 2020). 72. Gallet CA, List JA. Cigarette demand: a meta-analysis of elasticities. Health Econ. 2003;12(10):821–35. 73. Guidelines for implementation of article 6 of the WHO FCTC. Geneva: World Health Organization; 2016 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 29 September 2020). 74. DeCicca P, Kenkel D, Liu F. Excise tax avoidance: the case of state cigarette taxes. J Health Econ. 2013;32(6):1130–41 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3855906/, accessed 27 January 2021). 75. Jawad M, Lee JT, Glantz S, Millett C. Price elasticity of demand of non-cigarette tobacco products: a systematic review and meta-analysis. Tob Control; 2018;27(6):689–95 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6056338/, accessed 27 January 2021). 76. John RM. Price elasticity estimates for tobacco products in India. Health Policy Plan. 2008;23(3):200–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2800993/, accessed on 27 January 2021). 77. Wasserman J, Manning WG, Newhouse JP, Winkler JD. The effects of excise taxes and regulations on CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 85 cigarette smoking. Journal of Health Econ, Elsevier. 1991;10(1):43–64. 78. Cheng K, Kenkel D. U.S. cigarette demand: 1944-2004. B.E. Journal of Econ Analysis & Policy. 2010;10(1):10.2202/1935-1682.2438 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3820165/, accessed 28 January 2021). 79. Bobak M, Jha P, Nguyen S, Jarvis M. Poverty and smoking. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000. 80. Peck, RM. Equity issues, tobacco, and the poor. Yurekli A, de Beyer J, editors. Washington (DC): World Bank Group; 2013. (Economics of Tobacco Toolkit, Tool 6; https://openknowledge.worldbank.org/ handle/10986/16265, accessed 28 January 2021). 81. Kenkel DS, Schmeiser MD, Urban C. Is smoking inferior?: evidence from variation in the earned income tax credit. J Hum Res. 2014;49:1094–1120. 82. Ruhm, CJ. Healthy living in hard times. Journal of Health Econ. 2005;24(2):341–63. 83. Ruhm, CJ. Are recessions good for health? The Quarterly Journal of Econ. 2000;115(2):617-50. 84. Tarantilis F, Athanasakis K, Zavras D, Vozikis A, Kyriopoulos I. Estimates of price and income elasticity in Greece. Greek debt crisis transforming cigarettes into a luxury good: an econometric approach. BMJ Open. 2015;5(1):e004748. (https://bmjopen.bmj.com/content/bmjopen/5/1/e004748.full.pdf, accessed 28 January 2021). 85. Kaiser M, Reutter M, Sousa-Poza A, Strohmayer K. Smoking and the business cycle: evidence from Germany. Discussion Paper Series. Bonn, Germany: IZA Institute of Labour Economics; 2017 (http:// ftp.iza.org/dp10953.pdf, accessed 28 September 2020). 86. Tingum EN, Mukong AK, Mdege N. The effects of price and non-price policies on cigarette consumption in South Africa. Tob Induc Dis. 2020;18(July):62 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC7398599/, accessed 28 January 2021). 87. Vardavas CI, Filippidis F, Agak, I, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the importance of earmarking revenue to health care and tobacco control. Tob Induc Dis. 2012;10(1):21 (http://www. tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the-revenue-to- health-care-and-tobacco,66110,0,2.html, accessed 28 January 2021). 88. Marquez P, Catinkaya V. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 89. Blecher EH, van Walbeek CP. An international analysis of cigarette affordability. Tob Control. 2004;13:339– 46 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/13/4/339.full.pdf, accessed 28 January 2021). 90. Goodchild M, Sinha P, Munish VG, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/handle/10665/331836, accessed 28 January 2021). 91. John RM, Dauchy E. Trends in affordability of tobacco products before and after the transition to GST in India. Tob Control Online First. 20 March 2020. 92. Chaloupka FJ, Drope J, Siu E, Vulovic V, Stoklosa M, Mirza M, et al. Tobacconomics cigarette tax scorecard. Chicago: Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/636/uic-tobacco-scorecard-report-final.pdf, accessed 28 January 2021). 93. Kanbur R, Keen MJ. Jeux sans frontieres: tax competition and tax coordination when countries differ in size. Am Econ Rev. 1993;83(4):877–92 (https://www.researchgate.net/publication/4901220_Jeux_ Sans_Frontieres_Tax_Competition_and_Tax_Coordination_When_Countries_Differ_in_Size, accessed 28 January 2021). 94. Bouw A. Tobacco taxation in the European Union: an overview (English). Washington (DC): World Bank Group; 2017 (http://documents.worldbank.org/curated/en/493581492415549898/Tobacco- taxation-in-the-European-Union-an-overview, accessed on 28 September 2020). 95. Blecher E, Drope J. The rewards, risks and challenges of regional tobacco tax harmonization. Tob Control. 2014;0:1–5. 96. Directive No.n° 01/2017/CM/UEMOA of 22 December 2017. Paris: Droit-Afrique; 2017 (https://www. 86 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N droit-afrique.com/uploads/UEMOA-Directive-2017-01-droits-accise-tabacs.pdf, accessed 29 September 2020). 97. Mirzoev T, Zhu L, YangY, Pescatori A, Matsumoto A, Callen T. The future of oil and fiscal sustainability in the GCC region. Middle East and Central Asia Research Department. Washington (DC): International Monetary Fund; 2020 (https://www.imf.org/-/media/Files/Publications/DP/2020/English/FOFSGCCEA. ashx, accessed 28 January 2021). 98. State cigarette excise tax rates & rankings. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/factsheets/0097.pdf, accessed 28 September 2020). 99. A map and table comparing cigarette prices in Canada (March 2018). Toronto: Non-Smokers’ Rights Association/Smoking and Health Action Foundation; 2018 (https://nsra-adnf.ca/wp-content/ uploads/2018/03/180302_map_and_table.pdf, accessed 4 February 2021). 100. European Commission excise duty rates on manufactured tobacco. Brussels: European Commission; 31 December 2017 (https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco- energy/excise-duties-tobacco_en, accessed 1 July 2018). 101. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2019 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study-on-the-tobacco-taxation-directive-2019_en.pdf, accessed 28 September 2020). 102. Liber AC. Heated tobacco products and combusted cigarettes: comparing global prices and taxes. Tob Control; 2018;28(6). 103. Heated tobacco products: information sheet - 2nd edition. Geneva: World Health Organization; 2020 (https://www.who.int/publications/i/item/WHO-HEP-HPR-2020.2, accessed 28 January 2021). 104. Kass Lempert L, Glantz S. Analysis of FDA’s IQOS marketing authorisation and its policy impacts. Tob Control 2020;0:1–9. 105. Chivers E, Janka M, Franklin P, Mullins M, Larcombe A. Nicotine and other potentially harmful compounds in “nicotine-free” e-cigarette liquids in Australia. Med J Aust. 2019;210(3):118–9 (https://onlinelibrary. wiley.com/doi/epdf/10.5694/mja2.12059, accessed 28 September 2020). 106. Wenzl T, Zelinkova Z, Regueiro J, Giri A. Scientific-technical support activities to DG TAXUD-C-2 on the option to include e-cigarettes within the scope of excisable goods for the Impact Assessment on a possible revision of Directive 2011/64/EU. Administrative Arrangement TAXUD/2015/DE/329. Luxembourg: Publications Office of the European Union; 2017 (https://publications.jrc.ec.europa.eu/ repository/bitstream/JRC105828/kjna29062enn.pdf, accessed 28 September 2020). 107. Lisko JG, Tran H, Stanfill SB, Blount BC, Watson CH. Chemical composition and evaluation of nicotine, tobacco alkaloids, pH, and selected flavors in e-cigarette cartridges and refill solutions. Nicotine Tob Res. 2015;17(10):1270–78. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573955/, accessed 28 January 2021). 108. Ghebreyesus TA. Progress in beating the tobacco epidemic. Lancet. 17 August 2019;394:548–49 (https:// www.thelancet.com/pb-assets/Lancet/pdfs/S0140673619317301.pdf?hss_channel=tw-27013292, accessed 29 September 2020). 109. Smoking cessation, a report of the Surgeon General. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2020. (https://www.hhs.gov/sites/default/ files/2020-cessation-sgr-executive-summary.pdf, accessed 8 October 2020). 110. E-cigarettes are harmful to health. Geneva: World Health Organization. 5 February 2020 (https://www. who.int/news-room/detail/05-02-2020-e-cigarettes-are-harmful-to-health, accessed 29 September 2020). 111. Shan L, Manzione LC, Azagba S. Psychological well-being and dual-use of cigarettes and e-cigarettes among high school students in Canada. J Affect Disord. 2020;265:357–63 112. King BA, Gammon DG, Marynak KL, Rogers T. Electronic cigarettes sales in the United States, 2013- 2017. JAMA. 2018;320(13):1379-80 (https://jamanetwork.com/journals/jama/fullarticle/2705175, accessed 28 January 2021). 113. LaVito A. Popular e-cigarette Juul’s sales have surged almost 800 percent over the past year [website]. CNBC. 2 July 2018 (https://www.cnbc.com/2018/07/02/juul-e-cigarette-sales-have-surged-over-the- past-year.html, accessed 28 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 87 114. Statement from FDA Commissioner Scott Gottlieb, M.D., on new enforcement actions and a youth tobacco prevention plan to stop youth use of, and access to, JUUL and other e-cigarettes. Washington (DC): US Food and Drug Administration; 23 April 2018 (www.fda.gov/NewsEvents/Newsroom/ PressAnnouncements/ucm605432.htm, accessed 29 September 2020). 115. Paul K. San Francisco becomes first US city to ban sale of e-cigarettes. The Guardian. 25 June 2019 (https://www.theguardian.com/us-news/2019/jun/25/san-francisco-e-cigarettes-vaping-juul-ban, accessed 29 September 2020). 116. Pesko MF, Courtemanche CJ, Maclean JC. The effects of traditional cigarette and e-cigarette tax rates on adult tobacco product use. J Risk Uncertain. 2020;60:229–258. 117. Cotti CD, Courtemanche CJ, Maclean JC, Nesson ET, Pesko MJ, Tefft N. The effects of e-cigarettes taxes on e-cigarettes prices and tobacco product sales: evidence from retail panel data. Cambridge: National Bureau of Economic Research; 2020 (NBER Working Paper 26724; https://www.nber.org/ system/files/working_papers/w26724/w26724.pdf, accessed 28 January 2021). 118. Saffer H, Dench DL, Grossman M, Dave DM. E-cigarettes and adult smoking: evidence from Minnesota. Cambridge: National Bureau of Economic Research; 2019 (NBER Working Paper 26589; https://www. nber.org/system/files/working_papers/w26589/w26589.pdf, accessed 28 January 2021). 119. Huang J, Gwarnicki C, Xu X, Caraballo RS, Wada R, Chaloupka FJ. A comprehensive examination of own- and cross-price elasticities of tobacco and nicotine replacement products in the U.S. Prev Med. 2018;117:107–14 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6195827/, accessed 28 September 2020). 120. Zheng Y, Zhen C, Dench D, Nonnemaker JM. U.S. demand for tobacco products in a system framework. Health Econ. 2017;26:1067–86. 121. Stoklosa M, Drope JM, Chaloupka FJ. Prices and e-cigarette demand: evidence from the European Union. Nicotine Tob Res. 2016;18(10):1973–80. 122. Grace RC, Kivell BM, Laugesen M. Estimating cross-price elasticity of e-cigarettes using a simulated demand procedure. Nicotine Tob Res. 2015;17(5):592–8. (https://www.researchgate.net/ publication/270291421_Estimating_Cross-Price_Elasticity_of_E-Cigarettes_Using_a_Simulated_ Demand_Procedure, accessed 28 January 2021). 123. Huang J, Tauras J, Chaloupka FJ. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob Control. 2014;23:iii41–iii47 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/23/suppl_3/iii41.full.pdf, accessed 17 February 2021). 124. Cheng KW, Chaloupka FJ, Shang C, Ngo A, Fong GF, Borland R, et al. Prices, use restrictions and electronic cigarette use—evidence from wave 1 (2016) US data of the ITC Four Country Smoking and Vaping Survey. Addiction. 2019;114(suppl 1):115–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC6661221/, accessed 28 January 2021). 125. Liber AC, Drope JM, Stoklosa M. Combustible cigarettes cost less to use than e-cigarettes: global evidence and tax policy implications. Tob Control. 2016;0:1–6 (https://www.researchgate.net/ publication/299473181_Combustible_cigarettes_cost_less_to_use_than_e-cigarettes_Global_ evidence_and_tax_policy_implications, accessed 28 January 2021). 126. Chaloupka FJ, Sweanor D, Warner KE. Differential taxes for differential risks — toward reduced harm from nicotine yielding products. N Engl Journal Med. 2016;373(7):594-7. 127. Goniewicz ML, Knysak J, Gawron M, Kosmider L, Sobczak A, Kurek J, et al. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob Control. 2014;23:133–9. 128. Sweet L, Brasky TM, Cooper S, Doogan N, Hinton A, Klein EG, et al. Quitting behaviors among dual cigarette and e-cigarette users and cigarette smokers enrolled in the tobacco user adult cohort. Nicotine Tob Res. 2019;21(3):278–84 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6379027/, accessed 28 September 2020). 129. Wills TA, Knight R, Williams RJ, Pagano I, Sargent JD. Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics. 2014;135(1):e43–e51 (https://www. ncbi.nlm.nih.gov/pmc/articles/PMC4279062/, accessed 28 January 2021). 130. Electronic Nicotine Delivery Systems: report by WHO. Geneva: World Health Organization; 2014. (FCTC/ COP/6/10 Rev.1; https://apps.who.int/gb/fctc/PDF/cop6/FCTC_COP6_10Rev1-en.pdf, accessed 28 January 2021) 88 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 131. WHO report on the global tobacco epidemic 2017: monitoring tobacco use and prevention policies. Geneva: World Health Organization; 2017 (https://www.who.int/tobacco/global_report/2017/en/, accessed 28 January 2021). 132. Regulation of the Minister of Finance of the Republic of Indonesia, number 146, PMK.010/2017 on excise tax rates of tobacco products. Jakarta: Ministry of Finance Indonesia; 2017. 133. Gourdet CK, Chriqui JF, Chaloupka FJ. A baseline understanding of state laws governing e-cigarettes. Tob Control. 2014;23:iii37-iii40 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4078672/, accessed 28 January 2021). 134. An act to ensure that nicotine products are taxed equally. 127th Maine Legislature: first regular session-2015. H.P. 670, Legislative Document No. 973. 2015 (https://www.mainelegislature.org/legis/ bills/getPDF.asp?paper=HP0670&item=1&snum=127&PID=, accessed 28 January 2021). 135. Evaluation of the Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco. Commission Staff Working Document. Brussels: European Commission; 2020 (SWD(2020) 32 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/10- 02-2020-tobacco-taxation-report.pdf, accessed 28 September 2020). 136. Electronic nicotine delivery systems and electronic non-nicotine delivery systems (ENDS/ENNDS): report by WHO. Geneva: World Health Organization; 2016. (FCTC/COP/7/11; https://www.who.int/ fctc/cop/cop7/FCTC_COP_7_11_EN.pdf, accessed 28 September 2020). 137. E-cigarette use among youth and young adults: a report of the Surgeon General. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2016 (https://e-cigarettes. surgeongeneral.gov/documents/2016_SGR_Full_Report_non-508.pdf, accessed 28 September 2020). 138. Wagener TL, Floyd EL, Stepanov I, Driskill LM, Frank SG, Meier E, et al. Have combustible cigarettes met their match? The nicotine delivery profiles and harmful constituent exposures of second-generation and third-generation electronic cigarette user. Tob Control 2017;26:e23–e28 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/26/e1/e23.full.pdf, accessed 28 September 2020). 139. Vapour products data. In: Euromonitor International [website]. London: Euromonitor International; 2019 (restricted access). 140. Global market data: segment insights. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 89 ANNEX 2.1 Table A2.1 Countries that apply different types of cigarette excise tax structures, 2018 SPECIFIC EXCISE (65 COUNTRIES) AD VALOREM EXCISE (42 COUNTRIES) MIXED EXCISE (63 COUNTRIES) NO EXCISE (15 COUNTRIES) Albania, Andorra, Australia, Azerbaijan, Barbados, Belarus, Belize, Bolivia (Plurinational State of ), Burundi, Canada, Cook Islands, Dominica, Ecuador, Eswatini, Fiji, Gambia, Honduras, Iceland, India, Indonesia, Jamaica, Japan, Jordan, Kazakhstan, Kenya, Kiribati, Kyrgyzstan, Lesotho, Malaysia, Mauritius, Mongolia, Mozambique, Myanmar, Namibia, Nepal, New Zealand, Nicaragua, Norway, Pakistan, Palau, Papua New Guinea, Peru, Philippines, Republic of Korea, Saint Lucia, Saint Vincent and the Grenadines, Samoa, Seychelles, Singapore, Solomon Islands, South Africa, Sri Lanka, Suriname, Tajikistan, Timor-Leste, Tonga, Trinidad and Tobago, Uganda, United Republic of Tanzania, USA, Uruguay, Uzbekistan, Vanuatu, Yemen, Zimbabwe Argentina, Armenia, Bahrain, Bangladesh, Benin, Burkina Faso, Cabo Verde, Cambodia, Cameroon, Chad, Comoros, Côte d’Ivoire, Cuba, Democratic Republic of the Congo, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Ghana, Grenada, Guatemala, Guinea-Bissau, Liberia, Madagascar, Mali, Mauritania, Niger, Panama, Paraguay, Saint Kitts and Nevis, Saudi Arabia, Senegal, Sierra Leone, Sudan, Syrian Arab Republic, Togo, Turkmenistan, Tuvalu, United Arab Emirates, Venezuela (Bolivarian Republic of ), Viet Nam, Zambia Algeria, Austria, Belgium, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Central African Republic, Chile, China, Colombia, Congo, Costa Rica, Croatia, Cyprus, Czechia, Denmark, Dominican Republic, Egypt, El Salvador, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iran (Islamic Republic of ), Ireland, Israel, Italy, Lao People’s Democratic Republic, Latvia, Lebanon, Lithuania, Luxembourg, Malta, Mexico, Montenegro, Morocco, Netherlands, Nigeria, North Macedonia, Poland, Portugal, Republic of Moldova, Romania, Russian Federation, Rwanda, Sao Tome and Principe, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Thailand, Tunisia, Turkey, Ukraine, United Kingdom, West Bank and Gaza Strip Afghanistan, Angola, Antigua and Barbuda, Democratic People’s Republic of Korea, Iraq, Kuwait, Libya, Maldives, Marshall Islands, Micronesia (Federated States of ), Nauru, Niue, Oman,a Qatar,a Somalia a This table shows the status of cigarette excise tax structures as of July 2018 and does not account for changes occurring after that date, in particular for the cases of Qatar and Oman, which introduced excise on tobacco in January 2019 and June 2019, respectively. Source: WHO RGTE. 90 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 2.2 THE ANALYTICS OF THE TAX BASE ELASTICITY Assume tax revenue R = tsQ or R = tvPQ , where Q is the quantity consumed, ts is the specific tax, tv is the ad valorem tax and P is the consumer price. The following equations can help to illustrate the different components of the tax base elasticity. Under a specific excise regime, change in revenue depends essentially on the change in consumption: where R is the tobacco tax revenue, is the specific excise tax and is the tobacco tax base elasticity. The tax base elasticity is made of: where ε, the price elasticity = , is the degree of pass-through of the specific excise tax rate increase on consumer price and is the tax-price ratio. Under an ad valorem excise regime, change in revenue depends essentially on the change in tobacco expenditure: where R is the tobacco tax revenue, tav is ad valorem excise tax and ηav is the tobacco tax base elasticity. The tax base elasticity here is made of: where is the degree of pass-through of the ad valorem excise tax rate increase on consumer price, is the tax-price ratio and ε the price elasticity = . ∂R = Q (1+ηs)∂ ts ηs = ε ts P ∂P ∂ts tav P ηav = (1 + ε) tav P ∂P ∂tav∂P ∂tav ∂P ∂ts ΔQ ΔP P Q ΔQ ΔP P Q ts P ∂R = P Q (1+ηav)∂tav CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 91 ANNEX 2.3 ELEMENTS OF THE DEVICES THAT MAKE UP ENDS/ENNDS PRODUCTS The main components of any ENDS/ENNDS kit include essentially: • USB charger (not a car charger) • Inbuilt battery Additionally, For open systems • Tanks (refillable containers) with removable atomizer (often sold bundled with atomizers) • Clearomizers/refillable pods (no removable atomizer) • E-liquid For closed systems • Disposable e-cigarettes: not rechargeable, thrown away after e-liquid is finished • Nondisposable e-cigarettes: – Pre-filled cartomizers (cartridges designed to go with the cigalike kit) – Pre-filled tank refills/pods (pods or cartridges designed to go with the prefilled tank/pod kits) Some definitions: • Atomizer: uses a heating element to vaporize the e-liquid • Cartomizer: combines the cartridge/tank and the atomizer • Clearomizer: same as cartomizer, uses different technology • Cartridge/tank/pod: container that includes the e-liquid In summary, ENDS/ENNDS product devices include the following: • USB charger (not a car charger) • Inbuilt battery • Disposable e-cigarettes • Atomizer • Cartomizer/clearomizer • Cartridge/tank/pod with or without atomizer • Pre-filled cartridge/tank/pod (for closed systems, includes e-liquid) • E-liquid (added in the cartridge/tank/pod in open systems) Source: ECigIntelligence, 2020. Information also obtained from vaping websites, including https://www. misthub.com/blogs/vape-tutorials/76788357-tutorial-atomizer-vs-cartomizer-vs-clearomizer, http:// www.bestclearomizer.com/clearomizer-vs-cartomizer-vs-atomizer/, https://wayofleaf.com/accessories/ vapes/atomizer-vs-clearomizer-vs-cartomizer, https://wayofleaf.com/accessories/vapes/atomizer-vs- clearomizer-vs-cartomizer, accessed 15 July 2020. 92 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 93 CHAPTER 3. Tobacco tax administration 3.1 INTRODUCTION Imposing excise taxes on tobacco products usually serves more than one purpose. Governments often find themselves balancing interests between financial and public health objectives. Both objectives can best be achieved by an efficient and effective competent authority with strong technical capacity to enforce and collect taxes. A competent authority is the agency, organization or department that is legally as- signed to complete a particular activity; in the case of administering tobacco taxes, the competent authority is often a tax administration, revenue authority, customs department or ministry of finance. Article 6 of the WHO FCTC (1), along with its guidelines (2), provides a solid foundation for sound tax administration. As stated under section 1.5 in the guidelines, tobacco tax systems should be efficient and effective. They should be structured to minimize the costs of compliance and administration, while ensuring that the desired level of tax revenue is raised and health objec- tives are achieved. Efficient and effective administration of tobacco tax systems enhances tax compliance and collection of tax revenue while reducing tax evasion and the risk of illicit trade. Efficiency in tax administration refers to minimizing the costs per unit of tax revenue collected. It is measured by comparing the resources used with the revenues gener- ated. Effectiveness in tax administration refers to a high level of compliance – also described as taxpayers meeting their obligations. Thus, an efficient and effective competent authority collects the tax at a minimum cost while ensuring conformity to the rules. Tobacco taxation is the single most effective tobacco control measure for re- ducing tobacco use and is best implemented as part of a comprehensive tobacco control plan (3). Illicit trade – including smuggling and illicit manufacturing – and tax avoidance undermine the effectiveness of tax policies and their objectives (4). The impact on illicit trade is often cited by opponents of tax increases, who argue that increasing taxes increases illicit trade. They contend that illicit trade can lead 94 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N to lower revenues for governments and lower prices. The challenge faced by the competent authorities is to ensure that due taxes are declared and collected on all tobacco products that are manufactured in and/or imported into its jurisdiction, while at the same time detecting tobacco products that are illegally manufactured in and/or imported into its jurisdiction, stopping such activity and prosecuting the responsible parties. This chapter describes the shared characteristics of good tax administrations, including best practices based on country experiences. It regularly refers to the WHO FCTC, and – given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade – draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (5). Any practice or conduct prohibited by law and related to production, shipment, receipt, possession, distribution, sale or purchase of tobacco products – including any practice or conduct intended to facilitate such activity – is considered as illicit trade (Article 1). The objective of the Protocol is to eliminate and prevent all forms of illicit trade in tobacco products. At the same time, the Protocol includes measures for tobacco tax administration based on international best practices, which makes it relevant for all countries, even those that are not Parties to it. The Protocol was adopted at the fifth session of the COP to the WHO FCTC in 2012 and entered into force on 25 September 2018. As indicated in the Preamble, it was developed in response to the increasing international illicit trade in tobacco products (5). The Protocol covers three main areas: (1) measures to control the supply chain (Part III); (2) measures dealing with offences, including sanctions (Part IV) and (3) international cooperation (Part V). Different provisions of the Protocol are discussed in detail throughout this chapter, and section 3.4 is devoted specifically to control and enforcement. 3.2 INSTITUTIONAL ARRANGEMENTS Competent authorities that collect taxes effectively in an efficient way share a number of attributes. The organizational structures of these authorities contain clearly defined roles, responsibilities and rules for coordination among relevant bodies. Moreover, competent authorities collect data regularly and manage information needed for assessing risks. The key to successful risk management is to share this information among relevant authorities both within a country and between countries. Effective and efficient competent authorities also regularly evaluate their performance and accountability according to key performance indicators to identify areas for improve- ment. These characteristics are discussed in greater detail in the following sections. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 95 3.2.1. CLEARLY DEFINED ROLES AND RESPONSIBILITIES OF COMPETENT AUTHORITIES The designation of competent authorities for the implementation and enforcement of tax laws – including clear definitions of the boundaries of authority among numerous agencies within a country – is essential for efficient collection of taxes. Areas where different agencies need to cooperate and share data must also be defined. Overlap of activities by different authorities leads to inefficient use of resources, whereas gaps create opportunities for fraud, leading to ineffective tax laws. The importance of clearly defined roles and responsibilities applies not only to tax authorities and customs but also to law enforcement agencies, including police and border control forces. The implementation and enforcement of taxation is organized differently in various countries. The most common structure separates customs and tax administration. The trend since the 1990s, however, has been to combine these functions into one agency, such as Her Majesty’s Revenue and Customs (HMRC) in the United King- dom, SUNAT in Peru1 and AFIP in Argentina.2 Several countries have increased coordination between tax and customs by creating a revenue secretariat and also implementing systems to share tax records as a single taxpayer account. Coordina- tion between tax policy and tax administration authorities has also increased. One can think of combining both into one department within the ministry of finance or ensure that tax administration authorities are consulted during the tax policy process. Some tasks, such as licensing, may be handled by other ministries such as the ministries of health, agriculture or trade. For example, the Ministry of Health of Brunei and the Health Science Authority of Singapore are responsible for the licensing of importers of tobacco products (6). In some federal countries, including Colombia and the United States, excise taxes – including tobacco taxes – are collected and enforced by local or state tax administrations. Other countries have organized the administration of national taxes by establishing a single unified revenue body. Particularly in larger economies, that body is often responsible for both direct and indirect taxes, including excise taxes, and reports to the ministry of finance. All the functions needed for effective and efficient tax administration are established within these bodies (7). No matter what the institutional arrangements may be, it is vital that the agencies cooperate and exchange information and that their competencies find their basis in law. More information on this topic is provided in section 3.2.2. 1 Law Decreto Supremo 061-2002-PCM - Disponen fusión por absorción de la Superintendencia Na- cional de Administración Tributaria – SUNAT con la Superintendencia Nacional de Aduanas - Aduanas [Supreme decree year 2002 about the merger between Tax and Customs Administration]. Lima: El Peruano, 12, July 2002 (in Spanish) (http://www.sunat.gob.pe/legislacion/sunat/ds061-2002-PCM.pdf, accessed 13 November 2020). 2 Administracion Federal de Ingresos Publicos, Decreto 618/1997 [Federal Administration of Public Revenue, Decree 618] (in Spanish) (http://servicios.infoleg.gob.ar/infolegInternet/an- exos/40000-44999/44432/norma.htm, accessed 13 November 2020). 96 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Many countries, however, have separate bodies for the collection of taxes and customs duties. A 2015 survey of 135 tax administrations worldwide found that only 36% of them were responsible for both tax administration and customs ad- ministration (8). In most countries, customs authorities are more likely to collect excise duties on imports, and in many countries, VAT or sales tax is collected jointly with tobacco tax, particularly for imported products. This simplifies controls and creates synergy by unifying common processes and procedures, resulting in cost savings for tax administrations and taxpayers. The involvement of multiple bodies in tax collection requires especially good collaboration and information-sharing to ensure efficient and effective collection of taxes and duties. KEY TAKEAWAY 1 Institutional arrangements with clearly defined roles and responsibilities – designed to prevent overlaps and voids – contribute to effective and efficient tax administration. 3.2.2. EFFECTIVE COORDINATION AMONG RELEVANT BODIES Coordination at the national level Coordination among relevant bodies is key to effective tobacco tax administration. This means not only clearly defined roles and responsibilities, as described in the previous section, but also coordination among the competent authority, customs and those responsible for formulating, analysing and implementing tax policy. Regardless of the institutional arrangements – whether the responsible parties are all within the ministry of finance or in separate government agencies – all parties need to cooperate and exchange information to optimize tax collection and enforcement of tax policy. In practice, this means that information should be shared among, for example, customs, local government units that issue licences and health authori- ties – particularly those that regulate the sale of tobacco products. For tax authorities, the most relevant information concerning excise taxes in- cludes the identity of taxpayers and those involved in the trade of tobacco (import and export data, licences, criminal records, tax returns, bank statements, etc.); the category, quantity, value and location of manufactured goods; and the movement of those goods until all taxes are paid. Legal impediments to obtaining this informa- tion – such as bank secrecy or privacy regulations – should be kept in mind, and where needed, exceptions for fiscal procedures should be incorporated into law. Seizure data are also a valuable source of information; more details on this are provided in section 3.4. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 97 Tax authorities should regularly coordinate with law enforcement agencies – such as the police and border control forces, depending on a country’s laws – to properly monitor tobacco-related activities and enforce the tax laws. Often, the competent tax authority and customs authorities work in close cooperation with anti-fraud teams.3 Coordination and sharing of information can be required in legislation or regulations to ensure a streamlined process and avoid confusion. This can be done on an ad hoc basis as needed or with formal planned exchanges of information and regular meetings. It is recommended that at least a legal basis for exchange or access to information among government bodies be established to prevent claims during legal procedures that evidence was obtained unlawfully. Some countries go beyond exchanging information and cooperation. In the Neth- erlands, for example, customs authorities not only carry out work for the Ministry of Finance, they also carry out non-fiscal tasks for seven other departments, including the Ministry of Agriculture, Nature and Food Quality; the Ministry of Justice and Security; and the Ministry of Foreign Affairs (9). These activities are often based on bilateral agreements between the Ministry of Finance and the other departments. In other countries, such as the United States and Canada, Customs and Border Protection are not part of the Ministry of Finance; they are part of the Department of Homeland Security in the United States and the Ministry of Public Safety and Emer- gency Preparedness in Canada. These agencies also carry out many non-fiscal tasks. Along with the implementation of new tobacco control and tax laws, several countries have also created high-level committees to ensure good coordination and implementation of the laws. Led by health and finance ministries, committees ensure coordination and fine-tuning to achieve desired results. Botswana, Chile, Colombia, Indonesia and Senegal, among other countries, have successfully started with coordina- tion, planning and monitoring of tobacco laws’ implementation through periodic com- mittee meetings. The committees usually include representatives from the ministries of health, finance, tax and customs, police, transport and, in some cases, education. Coordination across borders Effective approaches to control smuggling in tobacco products require interventions at the borders of jurisdictions and therefore must involve the border agencies. However, with the globalization of trade, there is a need for close coordination not only between tax and border control authorities but also between different jurisdictions. Recent cases have demonstrated that an absence of formal cooperation frameworks may expose a market to financial crime, including money-laundering and financing of terrorism (10). 3 See, for example, Focus on tax fraud. Customs administration of the Netherlands, tax and customs administration. 2017;2 (https://customsnl-insight.nl/article/309563676, accessed 3 October 2020). 98 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Accession to international cooperation agreements such as the Protocol, the Organisation for Economic Co-operation and Development (OECD) Multilateral Convention on Mutual Administrative Assistance in Tax Matters and other regional arrangements will contribute greatly to the effective exchange of information and cooperation among enforcement agencies. An effective exchange of market data and information from participating jurisdictions can prevent potential cross-border crimes and loss of domestic revenue. International cooperation reinforces domestic measures to stop illicit trade and raise much-needed revenues. Parties to the Protocol have a commitment to cooperate with one another and to share information to meet their obligations under the Protocol (Article 20). The Protocol itself is the legal instrument that allows Parties to cooperate and share information across borders. Authorities of governments that are not Parties to the Protocol or another coopera- tion agreement that represents a legal instrument to exchange information could conclude a mutual assistance agreement or exchange of information agreement to guide the procedures under which information exchange can take place effectively. The Revised Kyoto Convention of 2010 promulgated by the World Customs Organization (WCO) recommends that jurisdictions that enter into bilateral agree- ments require the other jurisdiction to provide pre-arrival information on goods bound for their customs territory. A survey of 87 WCO members in 2013 found that the vast majority of customs administrations had the legal authority to share information related to the supply chain of tobacco products with other administrations (11). Some economic blocs have also established harmonized legislation applying to all of their Member States to provide administrative cooperation to efficiently cooperate on tax matters (12). Coordination can include the establishment of a special agency to ensure the safety and proper functioning of external borders, such as the European Border and Coast Guard Agency, also known as Frontex (from the French frontières extérieures, “external borders”). In some of the Frontex-led operations, EU and non-EU countries cooperate together with international organizations to target cross-border crime, including the smuggling of cigarettes and raw tobacco (13). Criminals who engage in illicit trade of tobacco products are usually also en- gaged in related criminal activities such as bribery, money laundering, corruption, obstruction of justice and even financing of terrorist organizations (14). A number of international treaties provide the legal framework for addressing such conduct through mechanisms that tackle illicit trade from a criminal justice perspective, such as the United Nations Convention against Transnational Organized Crime, the United Nations Convention against Corruption and the International Convention for the Suppression of the Financing of Terrorism. Table 3.1 summarizes the types of structures available for such coordination. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 99 Table 3.1 Structures for coordinating mechanisms TYPE OF COORDINATION BASIS INVOLVED ACTORS National coordination Agreements with a basis in law between national agencies Customs authorities, ministries of finance and those responsible for formulating, analysing and implementing tax policy; law enforcement agencies, such as police and border control forces; and anti-fraud teams Agreements between ministries or a basis in law or regulation on the establishment of high- level committees Ministry of health, finance, revenue, justice, transport and sometimes education and enforcement entities such as customs and police Bilateral coordination Bilateral cooperation agreements National governments Regional coordination Regional arrangements such as • Harmonized legislation applying to all Member States of an economic bloc to provide administrative cooperation in taxation to efficiently cooperate on tax matters • Regulation to jointly establish a special agency to ensure the safety and functioning of external borders EU Member States, the European Border and Coast Guard Agency (Frontex), customs, law and border enforcement agencies International coordination International treaties or conventions such as • The Protocol • OECD multilateral Convention on Mutual Administrative Assistance in Tax Matters • United Nations Convention against Transnational Organized Crime • United Nations Convention against Corruption • International Convention for the Suppression of the Financing of Terrorism Parties to international treaties and conventions, law and border enforcement agencies KEY TAKEAWAY 2 Regardless of differing institutional arrangements, coordination and cooperation within a country and across jurisdictions are essential to optimize tax collection and enforcement of tax policy. 100 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 3.2.3. EVALUATION OF PERFORMANCE AND ACCOUNTABILITY Key strategic indicators are useful for assessing the performance of a competent authority. Performance indicators can include measures such as net revenue col- lected, total expenditures compared with budgeted amounts, the ratio of costs to collection, measures of filing and payment compliance and taxpayer satisfaction (15). Several international organizations, including the IMF, the World Bank, the Inter-American Development Bank and OECD have developed tools to evaluate tax and customs with key performance indicators. This section provides information on some of the indicators that are particularly useful for measuring performance related to tobacco taxes, including the cost of collection ratio, tax gap analysis and tax revenue targets. Cost of collection ratio Collection costs vary among countries. The cost of collection ratio is the total ex- penditure as a percentage of the total net taxes collected. This ratio is often used as a measure of efficiency and effectiveness of competent authorities. In Table 3.2, the cost of collection ratio is calculated for country groups by income level, based on an annual IMF survey. The numbers in the table give an indication of resources used and revenues collected for taxes in general. The same definition of cost of collection was used for all countries. The tax revenue excludes VAT and excise taxes on imported products, so it reflects internal taxes only: personal and corporate income taxes, VAT and excise on domestic production. Customs duties are also not included. The results show the differences among countries at various income levels. Other contributing factors include differences between tax systems, economic situations and compliance levels. Table 3.2 Cost of collection ratio in 2015 per 100 units (ratio of average recurrent budget to revenue collecteda) GROUP (SAMPLE SIZE) 2015 Low-income countries (6) 1.3 Lower-middle-income countries (15) 1.6 Upper-middle income countries (18) 0.9 High-income countries (36) 0.9 All (76) 1.1 a Does not include VAT or excise on imports Source: (Reference 8, Appendix Table 12). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 101 As one would expect, given lower levels of automation and resources, the ratio is higher for low- and lower-middle-income countries, greater than 1.0 (more than 1.0 currency unit needed to collect 100 currency units). The ratios for upper-middle income and high-income countries are below 1.0, indicating more efficient and/or effective collection systems. The cost of collection might be less relevant for taxes that are introduced with other than solely financial objectives, such as influencing a change in behaviour. In particular, in the case of excise taxes applied on tobacco products, the cost of tax collection does not reveal the full picture. If excise tax rates are increased substan- tially – or at least increased above inflation and income growth – consumption will be reduced. As a result, health care costs will be reduced due to reduced tobacco- related mortality and morbidity and increased productivity. These savings are not factored into the ratio of cost of collection to revenue, but governments do benefit from these lower expenses overall. Nevertheless, the cost of collection can be used as an indicator of the efficiency of a competent authority. Tax gap analysis Tax gap analysis is another method of determining how effectively taxes on tobacco products are collected. The tax gap is the difference between the tax due and the tax that is collected. For example, the theoretical tax due under an ad valorem tax on the retail price of cigarettes would be the average price of a pack of cigarettes multiplied by the number of packs sold (estimated from household expenditure surveys, for example) multiplied by the tax rate. This outcome can then be compared to the actual revenues collected (16). The effectiveness of tax collection can also be determined by using the macro- economic input-output matrix, measuring the added value of the economic sector – tobacco in this case – and the theoretical VAT due and then comparing the result with the real VAT collection. This methodology is valid for measuring domestic tax evasion (more information on the use of this method to estimate illicit trade is provided in Chapter 4, section 4.1). Tax revenue target The performance of a competent authority can also be evaluated by determining whether the tax revenue target has been met, if mandated, for a given tax period. Although revenue forecasts are often used as targets, caution is advised. Forecast revenues could include assumptions such as economic growth, inflation and amount collected. Forecasting is a good practice, however, and competent authorities should provide input to the government for the forecasting of revenues to improve the quality of the estimates. Competent authorities should monitor the actual collections 102 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N in comparison with the forecasted revenues, but the theoretical base may not be attainable for a variety of reasons. In addition, a revenue target could provide an incentive for some customs and competent authorities to simply aim to reach the target amount, rather than making efforts to collect the maximum amount possible with the available resources. 3.3 THE TAX COMPLIANCE CYCLE For any tax, there are associated compliance, control and enforcement processes. The compliance cycle usually includes registration and licensing, tax declarations, recordkeeping, storage in warehouses, duty suspension, collection of tax and tax refunds. Figure 3.1 illustrates the typical stages of the tax compliance cycle. Fig. 3.1 Tax compliance cycle 3.3.1 REGISTRATION AND LICENSING Along with regulating and ensuring the integrity of those who deal with controlled substances or goods, the main objective of licensing is to regulate the supply chain. Licensing is a powerful tool for obtaining more information and securing the supply chain of tobacco products. Parties to the Protocol are committed to licensing the manufacturing, import and export of tobacco products and manufacturing equipment (Article 6). In addition, Parties are committed to endeavouring to license – as considered appropriate – the persons involved in the growing of tobacco and the retailing, transporting, wholesal- ing, brokering, warehousing and distribution of tobacco products or manufacturing Registration and licensing (renewal) Tax declaration Authorities: audit and control Taxpayers: recordkeeping Audit and control of information provided in tax declaration Payment and collection of tax Refund Licensed activities, for example: import, production, transport, storage, export, etc. 1 2 3 456 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 103 equipment (Article 6). To ensure an effective licensing system, Parties shall monitor and collect, where applicable, any licence fees that may be levied and consider using them in effective administration and enforcement of the licensing system, for public health or for any other related activity in accordance with national law. If feasible, each Party shall require that retailers and tobacco growers – except for traditional growers working on a noncommercial basis – maintain complete and accurate records of all relevant transactions in which they engage, in accordance with its national law (Article 9.4). Article 6.3(b) of the Protocol provides a list of information to be requested from the applicant of the licence, including: • relevant identity information on the applicant • business location of the manufacturing unit or warehouse and production capacity • detailed list of tobacco products and equipment used • description of where the manufacturing equipment will be installed and used • documentation or declaration of any criminal records • information on bank accounts to be used for transactions and payments • description of intended use and intended market of sale of the tobacco products. To make it easier for authorities to collect all the information they need, rules of confidentiality could be exempted in the licensing process. Licences can be general – covering all activities requiring a licence – or issued for each activity separately, such as different licences for manufacturing, importing and retail. A general licence is less burdensome for the licensing authority, whereas licences for each type of activity offer greater control but at the cost of more adminis- tration (17). The cost of implementing the licensing system should be proportionate to the potential impact of the system. Not only should the type of licences be taken into consideration, the process and information needed to obtain a licence should be carefully considered to ensure proportionality. The more stringent the process is – in terms of the information required and the obligations the system imposes on licensees – the more burdensome the regime will be on both businesses and the authorities who must administer and enforce it. The more information is collected, the higher the compliance and administrative burden will be. It is recommended that the added value of the information be balanced with the additional compliance, administrative and/or enforcement burden. The level of stringency should be decided with consideration of factors such as the level of risk of the activity and the availability of enforcement capacity. A more stringent regime might be justifiable for activities that pose a higher risk for the 104 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government in terms of potential loss of tax revenues – such as the import, production and handling of excisable products on which the excise taxes have not yet been paid. Authorities could consider setting licence fees at a high enough level to cover the costs of administering and enforcing the system. For an example of a system that relies mainly on licensing and permissions, see the case study of Australia in Box 3.1. Wholesalers, distributors and retailers of tobacco products could also be required to obtain a licence before they can engage in the trade of those products. This would enable the competent authority to require reports on, for example, transactions relating to the purchase and sale of tobacco products. Moreover, it would allow the authorities to complete the audit trail of the entire supply chain and to obtain data that will help tax and health policy-makers properly and effectively monitor tobacco products. Governments could also require a licence for entities dealing with raw materials or growing tobacco, including farmers. If licensing of tobacco farmers is deemed appropriate and subsequently required in a country, the farmers have to identify and register their farm areas and location to obtain a licence. The benefit of requiring licences for farmers is that the control of the legitimate supply chain is extended to the identification of the source of the raw material for tobacco products. It also makes it more difficult to divert raw tobacco from the licit to the illicit supply chain. Licences are issued by different agencies across the world. In Brazil, for example, the Health Surveillance Agency is responsible for providing licences. Operators need to obtain approval of the layout of manufacturing and warehousing facilities before they can operate. In addition, they must demonstrate how they will comply with other laws and regulations – for example, by showing the design of product packaging, including the pack, carton and master case. The factory location must be identified before manufacturers can obtain a licence. Finally, a licence is required for the importation of machinery to produce tobacco products (18). Licences can be a source of useful information if authorities establish the informa- tion that applicants must supply in order to obtain the licence. Such information could include the quantity, price and how the tobacco harvests are disposed, as well as the identity of the buyers. It is recommended that an effective licensing regime collect information to establish both the identity and characteristics of applicants by requiring criminal records on relevant offences, such as previous noncompliance with tobacco licences or fraud. To avoid loopholes for monitoring raw tobacco, importers of tobacco leaf could also be licensed or at least required to register and report information on quanti- ties, sources and sales. In some countries, this information is already collected by a government agency other than tax authorities, for example, by the ministry of agriculture. Duplication of requirements and reporting should be avoided through CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 105 legislation and coordination among agencies. If licensing of (small-scale) farmers is difficult to implement, subsequent purchasers (first processors) in the supply chain could be licensed and regulated instead. Licensing first processors is often less burdensome to enforce for competent authorities because, in general, there are far fewer first processors than there are growers. For example, in the EU, between 50 and 100 first processors have been identified, compared with 55 000 farmers (19). Countries could also consider requiring registration of persons or entities engaged in the manufacture and import or sale of materials used for the manufacturing of tobacco products, such as cigarette papers, tobacco leaves, additives, adhesives, acetate or any other type of filters used for cigarettes, tipping paper and cellophane or plastic wraps, as well as materials for packing the cigarettes into packs, reams and master cases. In addition, tobacco manufacturers could be required to obtain a licence before they can purchase these materials. The Parties to the Protocol should decide on appropriate measures, depending on research as to whether key inputs that are essential for manufacturing of tobacco products exist and can be identified and subject to effective controls. KEY TAKEAWAY 3 The objective of licensing is to regulate and secure the supply chain. It is a powerful tool for obtaining information for verification, further investigation and audits. Ideally, all persons involved in the growing of tobacco and retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. Licensing requisites Based on case studies and best practices – including experiences from managing bonded warehouses where the value of merchandise or suspended duties or taxes is high – the following kinds of information could be required to obtain a licence, in particular, for producers, warehouses and distributors of tobacco products: • certification of safety of installations, perimeter security for production and storage (may include CCTV [closed-circuit television]4 access for tax administration) • certification of financial solvency • detailed online, real-time inventory of tobacco products and main raw materi- als, accessible by tax administration • electronic accounting systems 4 The term “closed-circuit television” is used generically to describe surveillance camera systems. 106 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • detailed lists of owners and managers • banking and other financial records • periodic electronic reports of transactions for tobacco products • anytime tax administration right of entry for inventories • mandatory electronic tax returns and payments • mandatory prior-to-arrival customs declarations for tobacco products • declarations of compliance with the tax stamp system (if applicable) • for those involved in import or export, authorized economic operator (AEO) certification • proof of compliance with the bond or guarantee regime • agreement to finance reasonable cost of inspections and tracking and tracing. Box 3.1 Case study of licensing in Australia Australia has taken an approach to controlling tobacco taxes that differs from that in many other countries. It has not used fiscal marks or tracking and tracing.5 Instead, it administers tobacco taxes through licensing and permission-based systems aimed at facilitating operations by lower-risk entities while preventing or tightly controlling commerce involving higher-risk entities. The domestic tax agency, the Australian Taxation Office (ATO), is responsible for most of the controls. These controls cover tobacco that is grown or manufactured and imported as finished goods or as leaf for manufacturing in Australia. In fact, the legal tobacco market in Australia consists only of imported finished tobacco products. In 2006, all tobacco-growing licences were cancelled by the ATO because manufacturers switched to cheaper leaf from external suppliers. The last domestic cigarette manufacturers closed in 2015 and 2016, and there has been no legal domestic tobacco growing or manufacture since then. The ATO administers all other functions relating to the import of tobacco and tobacco products, including licensing of bonded warehouses used to store imported products and issuance of permissions to undertake movement of bonded tobacco products between licensed bonded warehouses or to places of export. Importers must apply for a licence for a bonded warehouse to store imported tobac- co. The applicant must meet general criteria such as fitness, recordkeeping and security.6 These criteria are designed to ensure that only low-risk entities are able to enter the ex- cise tax system. Risk levels are also kept at an acceptable level through provisions allow- ing the suspension or cancellation of licences, subject to appeal. Licences are valid for a 5 See sections 3.4.4 and 3.4.5 for detailed discussions on fiscal marks and tracking and tracing. 6 The entity must not have been charged with an offence under the Excise Act or any Commonwealth, State or Territory Act that carries a penalty in excess of US$ 105 000 in the previous 12 months (or convicted in the previous 10 years), has shown a history of compliance with tax law in the previous four years, has had no previous cancellation of a licence, has adequate financial resources and is not in receivership. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 107 three-year period, with automatic renewal for licensees with demonstrated compliance. The permission system relies on post-transaction audits of commercial records. Criteria used to assess risk include the size of the duty liability, the compliance record of both parties and the possibility of diversion into the market. When there is a perceived risk of revenue loss, the application can be denied or a financial security deposit can be required. Exports of tobacco products are also subject to an export declaration process with the Australian Border Force. An approved export declaration is required for the products to be able to leave the country. Following recommendations from a government task force in 2017, the status of tax-suspended, bonded tobacco was eliminated as of 1 July 2019. In addition, an import licensing regime was introduced, and commercial tobacco imports without a licence are banned. Importers are required to identify their duty liabilities at import and make immediate payment; there are no credit terms available. Full payment of duties and taxes to the Australian Border Force are required prior to a release of tobacco products into the country. Sources: (20–21). As mentioned above, certification as an AEO could be requested as part of the licensing process. Most customs authorities are familiar with the concept of AEOs. Created by the WCO, AEO principles were initially focused on security concerns (22). Having a special licensing regime for operators of the tobacco supply chain is recom- mended due to the special nature of the product. For countries that have no system in place, AEO certification could be a starting point for setting up such a regime. An AEO is defined by the WCO SAFE Framework of Standards (22) as a party involved in the international movement of goods – in whatever function – that has been approved by, or on behalf of, a national customs administration as complying with WCO or equivalent supply chain security standards. AEOs include, inter alia, manufacturers, importers, exporters, brokers, carriers, consolidators, intermediaries, ports, airports, terminal operators, integrated operators, warehousers and distributors. For many years – in some cases, even since the 1970s – customs administrations have been increasingly involved in the security of the international trade supply chain. More recently, customs administrations have developed security programmes in a global context. The AEO is part of these programmes, and in 2005, the WCO adopted the SAFE Framework of Standards. Since then, a number of traders have been required to make substantial investments in order to obtain AEO status and must continue to invest to maintain that status. The AEO program is also recognized by the Trade Facilitation Agreement, a multilateral agreement signed by 174 countries (23). 108 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Some regional blocs have further specified the standards for AEOs and provide clear and well-structured information on their websites to guide and encourage operators to apply for AEO status. A good example of this practice is the website of the Revenue Commissioners of the Republic of Ireland, which contains the in- formation shown in Box 3.2. Box 3.2 AEO: Republic of Ireland Tax and Customs What are AEOs? AEO status is a certified standard authorization issued by customs administrations in the European Union (EU). It certifies that an economic operator has met certain standards in relation to: • safety and security • systems to manage commercial records • compliance with customs rules • financial solvency • practical standards of competence or professional qualifications. This is primarily a trade facilitation measure that recognizes reliable operators and encourages best practices in the international supply chain. As an AEO, an operator could benefit from: • recognition worldwide as a safe, secure and compliant business partner in international trade; • lower risk scores in risk analysis systems when profiling; • priority treatment if physical controls are conducted; • mutual recognition of AEO programmes under Joint Customs Cooperation Agreements, which could result in faster movement of goods through third- country borders; • reduced data sets for entry and exit summary declarations (this applies only to AEO safety and security); • easier access to simplified procedures; • reduction or waiver of comprehensive guarantees. The conditions for AEO status apply to all businesses regardless of size. Manufacturers, exporters, freight forwarders, warehouse keepers, clearance agents, carriers and importers may all apply for AEO status. Source: (24). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 109 3.3.2 DATA COLLECTION, DECLARATIONS AND ACCOUNTING The effectiveness of risk analysis depends on the quality and reliability of the available data. This is also the case for risk analysis in relation to tobacco taxes. Obtaining reliable data can be a challenge in many countries, but the use of electronic sys- tems to collect and manage data is increasing in most competent authorities. The introduction of VAT in many countries around the world has greatly improved the availability of data that can be used for tobacco tax analysis, since reporting is done along the supply chain on, for example, the value, quantity of goods and transaction date. Most countries applying excise duties also have a VAT system in place. In addition, more countries are becoming Parties to the Protocol. With the implementation of the Protocol, more data will become available because countries will be obliged to implement, among other measures, licensing systems with report- ing requirements and tracking and tracing systems. More information on tracking and tracing systems is provided in section 3.4. The obligations of the Protocol will also assist in monitoring the stock of tobacco products. Ideally, all entities involved in the tobacco product supply and distribution chains should be licensed and required to record every transaction that occurs. As this might be burdensome for both tax authorities and taxpayers, the use of automated and electronic systems is recommended in order to decrease the costs of compliance. An accurate inventory system for all raw materials, machinery, goods in process and finished products can be required. It is even more important to have good recordkeeping of the required data. As the volume of reported data increases, a good information technology (IT) system will be needed. The use of IT for periodic tax declarations, accounting, inventory and financial data is critical for obtaining accurate information and decreasing costs for the entire reporting system. Most countries now have some level of automation that can facilitate data analysis. An emerging trend is the use electronic invoices, issued by traders, as part of online real-time information for tax administration. Countries generally start by using electronic invoices at public utility companies and then later expand the use to large companies. Electronic invoices minimize the use of paper, contribute to automated recordkeeping and give accurate and timely information about transactions for tax administration. Several countries began using electronic invoices for companies on a voluntary basis and later made their use mandatory, especially for large companies with a high number of transactions – including the tobacco industry. Electronic invoices have been implemented successfully in EU countries and almost all Latin American countries, as well as several Asian countries.7 7 Electronic Invoicing in Latin America: English Summary of the Spanish Document; Inter-American Development Bank, Inter-American Center of Tax Administrations, 2018 (https://publications.iadb.org/ publications/english/document/Electronic-Invoicing-in-Latin-America.pdf ). 110 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N To verify that information is accurate, competent authorities could systematically cross-check declared information against third-party information (e.g. from banks, financial institutions, employers) or match the data with the information in registers of other government agencies. Processes of cross-checking and data matching could also be automated to minimize the administrative burden (25). KEY TAKEAWAY 4 Reliable data are essential for effective risk analysis. While obtaining these data can be challenging, electronic systems can help reduce the burden by automating procedures of data collection and cross-checking of information with different sources. 3.3.3 RECORDKEEPING Parties to the Protocol are committed to requiring, as appropriate, that all persons or entities engaged in the supply chain of tobacco, tobacco products and manu- facturing equipment keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of tobacco products (Article 9). Relevant information includes market volumes, trends, forecasts of tobacco products and quantities of tobacco products and manufacturing equipment kept in stock in tax and customs warehouses in transit, transhipment and under duty suspension. This information should be required from the persons and entities engaged in the supply chain and submitted to the competent authority on a regular basis, as provided for in the law. The competent authority can use the submitted information to monitor compliance with tobacco regulations and payment of taxes. A registry with this level of detail can realistically be kept only in electronic form. Records must provide full accountability for materials used in the production of tobacco products. The intention is that tax authorities and manufacturers should be able to reconcile the production quantities with the inputs used in production – thereby providing confidence that no unrecorded or illicit production has occurred. Obligations should also be imposed on suppliers of key inputs to show that supply is commensurate with demand (17). KEY TAKEAWAY 5 To monitor compliance and payment of taxes, all persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions, as well as details of materials used in the production of tobacco products. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 111 3.3.4 WAREHOUSING, STORAGE AND DISTRIBUTION According to Article 6.2 of the Protocol, all Parties shall endeavour to license persons involved in any wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Maintaining a system of authorization allows the authorities to carry out controls in production and storage facilities to ensure that taxes are paid (2). The approval process to obtain an authorization could include an evaluation of the layout of the plant or warehouse, the machinery that will be used and the flow of production, warehousing and shipping, including the points of entry and exit of raw materials and finished products. The basic method of monitoring production and ensuring that only tax-paid products are released to the market from the premises is to identify the production facilities and to control the entry and exit points. From time to time, the competent authority should conduct a physical inventory of the goods contained therein to check whether all documentation was duly prepared and approved and to determine the accuracy and completeness of the records kept. If the jurisdiction requires tax stamps to be placed on the tobacco products, only products with the proper stamps affixed can be withdrawn. Generally, tobacco products for which the required taxes have not been paid and, if required, fiscal marks have not been affixed should not be allowed into warehouses. For practical reasons, many countries allow suspension of excise duties, meaning that prior authorized persons can produce, send, receive and store tobacco products on which the excise duty has not yet been paid. The relevant authorities could also require that products on which the taxes have been paid should not be stored in the same areas as the products under duty suspension. Obviously, products under suspension of payment of excise duties are at high risk, which could justify stricter requirements for production, trade, storage and handling. Australia, which has a strict system of licensing and requirements for permission to move tobacco products, has migrated to a new system that eliminates bonded warehouses from the supply chain as of 1 July 2019. Importers are required to have an import licence and to pay excise taxes on cigarettes immediately upon import (see Box 3.1 in section 3.3.1). 3.3.5 DUTY SUSPENSION Many countries require authorization of natural or legal persons (as authorized warehouse keepers) to produce, process, hold, receive and dispatch products sub- ject to excise duty during their business. Producing, processing, holding, receiving and dispatching excise goods often take place under suspension of the excise duty. Guarantees can be requested from authorized persons to secure the payment of taxes. Features of such a system may include strict criteria for granting authorization, 112 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N warehouse pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. The use of a computerized system for monitoring movements of excise goods under suspension of excise duty can be a control as well. Different licences for products under duty suspension could also be considered. This would make enforcement easier and less burdensome for both authorities and operators. In general, it is recommended to allow the handling of excise goods under suspension of duties only if strict criteria are met. Such criteria could include pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. In principle, the movements of tobacco products should also be covered by the tracking and tracing system. Considering the high risk related to these products, additional monitoring could be considered appropriate, such as a computerized system monitoring the movements of excise goods under suspension of excise duty. In the design of such a system, it is recommended that close attention be paid to customs procedures for import and export to ensure alignment and avoid a vacuum in monitoring. An example of a computerized system is the EU’s Excise Movement and Control System, which follows the movement of all excise products – including manufactured tobacco products – for which excise taxes have not been paid. The system records the movement in real time and is thereby an important tool for combatting fraud. In addition, this system is indispensable for the exchange of information and co- operation between the relevant authorities of Member States of the EU (26). Finally, authorization is required before tobacco products can be produced, imported or stored under suspension of excise duties (27). KEY TAKEAWAY 6 Products under duty suspension of excise taxes are at a higher risk of tax evasion, which can justify stringent measures such as requesting guarantees to ensure the payment of taxes, additional licensing requirements, compliance with computerized systems to monitor the movement of excise goods under suspension and on-site authorization and audits. 3.3.6 COLLECTION OF TAXES To reduce the complexity of tax collection systems, it is recommended that excise taxes be imposed at the point of manufacture, import or release from storage or production warehouses for consumption. This is common practice in the majority of countries that impose excise taxes. Collecting taxes at this level of the supply chain CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 113 greatly limits the number of taxpayers and thus the resources needed to control them. Encouraging taxpayers to use electronic payment methods can also increase the chances of collecting all taxes. The same applies to requiring guarantees for certain high-risk activities, such as the handling of goods under duty suspension. Many countries decide on a case- by-case basis the level of the guarantee, depending on the situation of the requestor and the level of risk (quantity or value and potentially due excise taxes) that the regular business activities represent in a given time frame. Some countries allow a reduction of guarantees for operators with a track record of good compliance. It should be noted that a guarantee is not a limitation of the liability; taxpayers can still be requested and liable to pay an amount far above the level of the guarantee. Tax payments should be required by law to be remitted at fixed intervals after sales or on a fixed date each month (2). Many countries have a specialized collection enforcement unit that works full-time on the collection of taxes. It is important to have a stop-filer or payment control that can act immediately when noncompliance occurs, by sending a message and phone call of late declaration or late payment to the taxpayer. This increases the likelihood of keeping taxpayers compliant. If nondeclaration or nonpayment persists, the bond or guarantee could be executed. Another reason for collecting excise taxes around the time of production or import is that quantities can be monitored more effectively at these points. There are different options for monitoring the supply chain of tobacco products. The decision about what kind of monitoring system to use depends on the country’s financial, technical and human resources. The weakest form of monitoring is in- dustry self-declaration. Activities to verify compliance and ensure the collection of the full amount of taxes due can include, for example, physical checks, audits, cross-checking of declared information with third-party data and inspection of administration and recordkeeping. In general, in countries with poor administration systems, enforced compliance is carried out by imposing physical control over the production or manufacturing process. The cost of physical control increases when there is a potential for fraud by excise officers. However, fraud can be diminished significantly when excise officers are rotated frequently among different locations and supervisors make surprise visits. Historically, some countries (e.g. India) have posted tax administration staff at production facilities to monitor production and removals. In India, a staff member of the competent authority is placed in cigarette and large bidi manufacturing facili- ties around the clock. Each officer records the daily production and the quantity of cigarettes/bidis that leaves the factory and reports to the next officer.8 8 Ministry of Finance India, personal communication, 2009. 114 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A better option is to monitor production remotely. The competent authority can require the installation of CCTV cameras in strategic places throughout the manufacturing and warehousing facilities. With these, the authority can establish a central command post from which the facilities and activities can be continu- ously monitored and documented. In addition, the competent authority can carry out physical inventory controls from time to time and – if electronic invoices are implemented – cross-checking between invoices and declared inventory. This is also an effective way to prevent collusion between staff of a competent authority and manufacturers or importers. For example, in 2015, the Bureau of Internal Revenue of the Philippines required all tobacco companies to install CCTV cameras in their production lines and warehouses. This decision was taken in response to large seizures of untaxed cigarettes, with the objective of monitoring production to ensure the payment of all taxes. The collection process must also be supported by IT systems. These systems must provide for transparency and accuracy to ensure a safe process for the flow of payments from taxpayers to the tax treasury. Most countries have implemented automated electronic systems for tax payments linked to each declaration, for both domestic and import. It is key for tax administrations to have a comprehensive agreement with the banking system in order to obtain lower transaction costs, if applicable. Some countries have implemented a state payment web portal that allows citizens to pay their taxes and other fees such as county fees, fees for car permits and licences and agricultural, health and environmental fees online. KEY TAKEAWAY 7 Excise taxes should be imposed at the point of manufacture, import or release from storage or production warehouses for consumption, to ensure that quantities can be monitored effectively. This also reduces the complexity of tax collection systems by limiting the number of taxpayers and thus the resources needed to control them. 3.3.7 TAX REFUNDS Refunds for VAT, excise taxes and customs duties are a common process in most countries, under the principle that consumption taxes are not exported. Frequency and methods of refund vary by country. It is common to have monthly refunds (if there are exports during the period), and the reimbursements may be sent directly to the exporter or reserved as a credit to pay other taxes. An alternative used by some countries that have a high volume of exports is a so-called zero rate, or suspension, meaning that indirect taxes (VAT, excise taxes and customs duties) are suspended for the whole chain – from import of raw materials to production and packing until CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 115 export. This regime requires a special licensing process. Since the tobacco industry has an export component, the refund process for this sector requires special atten- tion for tax administration. 3.4 CONTROL AND ENFORCEMENT Control and enforcement are the main functions of tax administration. In fact, most tax laws include the objective “to control and enforce tax compliance” and, for customs, “to control and enforce tax and duty payments at the border” or similar phrases. The Protocol provides guidance for control and enforcement of tobacco taxes. Efficient and effective competent authorities often have a strategic plan to ensure compliance, a risk-based approach to identify the problematic points in the chain and the ability to direct resources accordingly to high-risk or high-value areas. Tasks that can play a role in control and enforcement include controlling the registration and licensing process, due diligence, verifying declarations and collec- tion of taxes. Production and distribution controls including tracking and tracing, fiscal markings, audits and import and export controls all play a role in control and enforcement. This section describes the main activities for improving control and enforcement, focusing on the tobacco supply chain. The procedures and penalties that can be enacted once illicit trade in tobacco has been detected are also discussed. 3.4.1 CONTROL AND ENFORCEMENT PLANNING Strategic plan In modern tax administrations, it is common to have a strategic plan, with control and enforcement as pillars. Appropriate control of the compliance cycle is key to keeping taxpayers in compliance and preventing illicit trade and tax avoidance. For this reason, most tax administrations focus a majority of their resources on preventive policy. Some examples of this can be found in the strategic plans of the United Kingdom and the United States’ Internal Revenue Service (IRS). In the United Kingdom, HMRC has had a well-developed strategic plan for years. A key pillar of the plan focuses on keeping taxpayers compliant. This is the concept of prevention: controlling initial minor noncompliant behaviour for the majority of taxpayers, while using strong enforcement for the minority on the noncompliant side (28). The strategic plan of the IRS has a similar approach, with a focus on control. If noncompliance is detected, data analysis and behavioural insights are used to identify the best way to address noncompliance. Early intervention or self-correction are examples of ways to address detected noncompliance. The IRS also highlights the importance of resolving noncompliance to ensure taxpayer confidence in the tax system and protecting the integrity of the system (29). 116 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Risk-based approach Following the establishment of a strategic plan, an enforcement and control plan must be drafted. This plan should include definitions of the activities that will be enforced, the taxpayers upon whom they will be enforced and the circumstances under which they will be enforced, as well as allocating resources for staffing, audit- ing, infrastructure and IT. Targets must also be defined, including the number of interventions and the amount of additional collected revenue or reduction of tax evasion. Several tax administrations elaborate annual plans with periodic perfor- mance reviews aimed at improving results, and they correct allocations and targets as needed. Clear targeting of interventions is needed for better results, more efficient use of resources, lower costs for taxpayers and more effective collection. In other words, the point is to focus interventions on those who have a higher probability of noncompliance. Using a risk-based approach can be particularly beneficial. Tax risk management is a key element of control strategy in modern tax admin- istration. A risk is a possible threat to reaching objectives such as collecting taxes in an effective and efficient way for competent authorities. Risk assessment is the process of analysing risks and deciding on the best way to manage an identified risk. The responses can vary from acceptance to mitigation to avoidance. Proper risk assessment allows competent authorities to use their available resources most efficiently and to become more effective in dealing with risks. It can be used to improve compliance by identifying taxpayers or types of activities with a high risk of noncom- pliance. Groups of taxpayers with the same characteristics often have similar risks. Groups with a high risk of noncompliance could then be subject to greater review. Areas of potentially greater risk of noncompliance in the tobacco supply chain include import, export and transfers to and from warehouses, particularly when they take place under duty suspension. Gathering risk-related information from internal and external sources is a best practice in compliance risk management. Such sources could include third-party information (e.g. from banks, credit card companies, transport companies), studies on taxpayer behaviour and research on compliance issues, tax gap analysis, tax audits and declarations (30). Risk management uses these different sources of data along with algorithms to find patterns of high noncompliance. Risk analysis can indicate reduced risk as well. Lower-risk areas are likely to need less governance to ensure compliance, which allows for resources to be directed elsewhere. Risk assessment can therefore help with strategic allocation of limited resources to the areas of greatest risk while at the same time reducing the burden on lower-risk taxpayers. Risk management has always been done by competent authorities, but data availability and statistical methods to identify patterns have changed the way risks CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 117 can be assessed. Although many risk assessment systems are still done manually or include manual elements, the use of intensive data techniques allows systematic, deeper and more targeted analysis (31). Modern risk assessment makes use of electronic data on taxpayers, tax payments, declarations from other taxes, such as VAT, and third-party information. With these data, tax authorities can identify indicators that suggest where further activities might be required to ensure compliance. For example, VAT invoices can be used to match reported purchases of inputs of tobacco leaf to sales invoices of tobacco leaf wholesalers. For taxpayers (i.e. those who are licensed and provide required reports), competent authorities can create a business analytics program to determine whether the data reported are consistent on each side of the transactions. Moreover, in countries with a VAT system, competent authorities can compare the data reported by taxpayers under the VAT system with data reported under the tobacco excise tax system to detect any inconsistency. VAT invoices can also be used to verify inputs and sales data. If VAT is collected at all levels of the supply chain, it is easier for govern- ments to monitor the supply chain for the enforcement of excise duty obligations. Regular surveys on tobacco consumption that use the same methodology can also provide indications about the level of compliance with excise tax policy. A sudden drop in revenue that is not reflected in consumption data could be an indication of illicit manufacturing, illegal imports, cross-border shopping or forestalling. In addition, seizure data can provide valuable information on areas and activities at high risk of noncompliance. The structure of tobacco tax policies should also be taken into account when con- ducting risk analyses. If excise tax rates are increased, there might be a greater risk of forestalling or front-loading (see the discussion on anti-forestalling later in this section). Differentiated excise tax rates based on product or packaging characteristics – such as distinctions between soft and hard packs or filter and nonfilter cigarettes – are also prone to manipulation by operators, which could affect tax revenues. One of the options to mitigate these risks is to amend the excise tax policy and apply a uniform tax rate. For customs transactions, the use of risk management is a key element in target- ing merchandise and support declarations to be inspected. Before the 1990s, most customs agencies used random criteria for selecting targets for inspection. Since that time, many countries have implemented risk-based approaches for selecting inspections. Historical data on importers and trade communities, complemented by artificial intelligence technology, show that risk management tools dramatically increased the effectiveness of physical inspections. Most modern customs agencies have implemented such techniques, allowing for more effective control processes while facilitating smoother processes for those transactions that are in compliance. Box 3.3 details some of the recent changes in risk management processes. 118 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.3 Changes in risk management The OECD developed Fig. 3.2 to show the framework and key steps for understanding compliance risks in 2004 (32). The same approach is still used to identify, assess and prioritize risk. However, many competent authorities now use new technologies and advanced data analytics, along with more information sources, including external data from banks, employers and sales invoices for VAT, for example. Fig. 3.2 Compliance risk management process Source: (32). The methods of identifying risks and the analysis of compliance behaviour have also changed. Traditionally, competent authorities used audits to identify high-risk cases. With more diverse and better data, competent authorities can now use more evidence-based approaches to examine risk patterns. Success of compliance activi- ties is now more often measured in terms of their impact on the overall compliance environment, rather than only on increased revenues.9 9 For more information on effective risk management with several indicators and a checklist of questions, see the Tax Administration Diagnostic Assessment Tool (68). OPERATING CONTEXT Assess and prioritize risks Evaluate compliance outcomes: - Registration - Filing - Reporting - Payment Monitor performance against plan Analyse compliance behaviour (causes, options for treatment) Determine treatment strategies Identify risks Plan and implement strategies CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 119 In Indonesia, use of the compliance risk management process reduced the share of illicit trade in total consumption of cigarettes from 12% to 3%. More information on this can be found in the case study of Indonesia in Box 3.13 later in this chapter. Understanding the products – as well as the supply and distribution chains – al- lows competent authorities to identify which areas along a chain pose the greatest risk and therefore require more resources. Detailed information on the composition of selected tobacco products is given in Annex 3.1. KEY TAKEAWAY 8 Risk analysis helps identify the points of intervention that have higher probabilities of noncompliance. A risk-based approach with targeted interventions allows for better results and more efficient use of resources to ensure effectiveness of tax collection. 3.4.2 CONTROLS OVER THE TOBACCO SUPPLY CHAIN As defined in Article 1 of the Protocol, the supply chain covers the manufacture of tobacco products and manufacturing equipment – as well as their import or export – and may be extended, where relevant, to one or more of the following activities when so decided by a Party: 1. retailing of tobacco products 2. growing of tobacco, with the exception of traditional small-scale growers, farmers and producers 3. transporting of commercial quantities of tobacco products or manufacturing equipment 4. wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Article 4.1 of the Protocol requires parties to “adopt and implement effective mea- sures to control or regulate the supply chain of tobacco products to prevent, deter, detect, investigate and prosecute illicit trade in such goods and to cooperate with one another to this end”. Concrete measures to regulate the supply chain, as well as best practices in this regard, are discussed further below. Figure 3.3 shows the main places for reporting and monitoring along the supply chain: import, ex-factory and removals from warehouses. Manufacturers could be required to report imported inputs at the border, as importers of finished products do. If components are subject to licensing, information can be required as part of the licensing process. The arrows in Fig. 3.3 represent transporting, which is also part of the supply chain. 120 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 3.3 Cigarette supply chain from manufacture or import to retail sale 3.4.3 LICENSING10 AND DUE DILIGENCE A licensing system is effective only if it is properly controlled. Most tax administra- tions have experience with licensing processes for excise taxes on products such as alcoholic beverages and energy products. It is strongly recommended that lessons learned with the licensing process of such products be applied when implementing and enforcing tobacco-related licensing. Licensing provides timely and accurate data that can serve as the basis for audits because it identifies and controls legitimate operators. For new operators, the process to obtain a licence could include visits and verification of production factories, storage facilities and distribution premises. Countries that have no licensing system in place and would like to start applying licences could allow a transitional period for existing operators to comply with the new licensing requirements. The process of licensing control must be carried out and updated periodically, in particular by controlling the validity of bonds or guarantees, the proper functioning of the required systems (CCTV, for example) and recordkeeping. 10 Licensing is discussed here in the context of due diligence and enforcement. Details about how licensing can be set up and what information could be requested are presented in section 3.3.1. Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blend d. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: wher he tobac o will be roll d into st cks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 121 Where licences are required, the law should include a provision specifying that purchases from unlicensed suppliers – or sales to unlicensed purchasers – are not allowed. This means that both suppliers and purchasers would need to verify those with whom they are doing business. This requirement provides enforcement au- thorities with an entry point to enforce the licensing system at both ends. Also, a licensing requirement for manufacturing equipment assists authorities in identifying and prosecuting illegal manufacturing of tobacco products, reducing the burden of proof substantially. In many countries, the presence of manufacturing equipment is not sufficient proof that illegal manufacturing is taking place; the machinery has to be in operation and producing illegal tobacco products when authorities inspect the location. With a licensing requirement, however, the presence of machinery without a licence is sufficient for authorities to act. The validity of licences should be time-limited, requiring renewals or reapplica- tion, to maintain a high level of control. Adherence to the conditions required for a licence should be controlled by the authorities, and penalties for noncompliance – for example, suspension or withdrawal of a licence – should be severe enough to act as a deterrent (33). Regulations for licensing should provide for inspection of the licensee’s products and premises, with penalties for noncompliance, which could include criminal and civil prosecution for serious or repeated offences. As stated in Article 6.3(a) of the Protocol, Parties need to establish or designate a single authority or multiple authorities to issue, renew, suspend, revoke and/or cancel licences. In accordance with Article 7 of the Protocol, persons engaged in the supply chain are required by law to conduct due diligence before and during business relationships. They also must report to the competent authorities any evidence that a customer is engaged in activities in contravention of its obligations arising from the Protocol. This requirement includes customer identification, monitoring of sales to ensure that the quantities are commensurate with demand for such products within the intended market and taking measures to ensure compliance. Knowledge of the demand of a market is indispensable for determining if there is a case of oversupplying. If the supply of tobacco products to a lower-taxing foreign market exceeds the demand, it creates a higher risk that these products will be smuggled back into a higher-taxing country, undermining the objectives and effectiveness of the higher-taxing jurisdiction. In the past, some governments decided to impose a fine on tobacco companies if the quantities supplied were significantly higher than the demand and the risk of being smuggled back into their jurisdiction was judged to be high (34,35).11 11 Excise duty rates applied in all the EU countries can be found on the European Commission’s webpage: https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco-energy/ excise-duties-tobacco_en. 122 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 9 Licensing helps to identify and control legitimate operators. The data obtained from licensing can serve as a basis for audits. Licences should be controlled on a regular basis and updated periodically to ensure their validity. 3.4.4 FISCAL MARKINGS (E.G. TAX STAMPS) Fiscal markings are another important tool for controlling and monitoring pro- duction and import of tobacco products. Their use is generally considered to be appropriate for increasing compliance with tax laws. Fiscal markings can also be of help in distinguishing between genuine and illicit tobacco products. Tax stamps or other fiscal markings affixed to packs of cigarettes or tobacco products facilitate the collection of excise taxes, as well as audits and enforcement actions. The pres- ence of fiscal markings enables both the competent authority and the public to monitor whether the taxes on tobacco products were properly paid. It thus assists the competent authority in investigating illicit trade and prosecuting violations. Fiscal markings include tax stamps, enhanced tax stamps (banderols) and digital tax stamps. Examples of fiscal marks are tobacco stamps, tax stamps, excise stamps, tax stickers and banderols. Box 3.4 presents details on the different types and features of tax stamps. Tobacco products for export are often required to be marked that they are for export. Box 3.5 provides useful information regarding the International Organization for Standardization (ISO) standard for excise tax stamps. The terminology “fiscal mark” holds no indication of the characteristics of the mark. A fiscal marking is affixed to each pack of tobacco product. Requiring a standard package size can facilitate the application of the markings (2). Fraudsters can be deterred from attempting to re-use fiscal markings (in particular stamps) by having the marking affixed to each pack of cigarettes (or other tobacco product) before the pack is wrapped with cellophane (36). In most cases, tax stamps are purchased by the producer or importer and applied to each product sold as proof of excise tax payment (33). Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Box 3.4 Types and features of tax stamps Over time, tax stamps and markings have become more sophisticated. In the past, tax stamps were often paper-based and easy to counterfeit. New tax stamps use additional security features to make them more difficult to counterfeit. Authentica- tion solutions against counterfeiting can utilize various security features, including: • overt features – features that can be verified by the naked eye; CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 123 • covert features – features that can be authenticated only by using dedicated and specialized electronic readers; • semi-covert features – features requiring a simple tool that does not involve extensive training; and • forensic features – features that can be identified through laboratory analysis. Tax markings can be either physical or digital: • physical markings – the information is contained in the document or device attached to the package. • digital markings – information is obtained through a link with a database and by decrypting with the tools and keys used for creation of the data. The term “digital tax stamp” sometimes leads to confusion, as some paper-based stamps with digital components are also described as digital tax stamps. Tax markings that are fully digital do not contain information in the document or device attached to the package. It is probably too simplistic to say that digital tax stamps are more secure than paper- based tax stamps. Both types have advantages and weaknesses. For instance, both physical and digital tax stamps can be weak or strong on security features. Neverthe- less, the management, production, sales, transport and monitoring of physical tax stamps require increased attention. Tax stamps have the same value as banknotes and are a possible attraction for theft, loss and fraud. In Belgium, for example, the Court of Audit severely criticized the lack of control of the production and stock management of tax stamps in 2015. The Court of Au- dit concluded that tax stamps issuance should operate under recognized security practices and procedures relative to the security risk associated with the various production, distribution and issuance processes. Moreover, it was noted that new printing technology of digital tax stamps on packs may facilitate stock management and lead to less fraud. In March 2016, Belgium changed its stamps. The printing became an in-house process by the financial federal government department. The new stamps are still printed with a watermark, but they also have a digital component. The change resulted in a cost reduction by standardizing the sizes and optimizing the production process. Some of the more advanced fiscal marking technologies include embedded threads and watermarks; special inks and coatings, such as so-called invisible inks, holograms and foils; and calculated or changeable content. Because of their enhanced security features, these stamps can be more expensive than traditional stamps. In the state of California in the United States, the traditional stamps cost US$ 0.42 per 1 000 stamps. The cost of the first generation of high-tech stamps was 10 times higher, at US$ 4.77 per 1 000 stamps. This price nearly doubled for the second-generation encrypted stamp, to US$ 8.20 per 1 000 stamps. Nevertheless, California collected 124 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N about US$ 450 million of additional tax revenue in the first decade following the implementation of encrypted tax stamps. This additional revenue was far greater than the costs of implementation and enforcement. Other jurisdictions have also revised their tax stamps to incorporate new technolo- gies. The state of Michigan, for instance, replaced heat-applied cigarette tax stamps with digital pressure-applied stamps in 2015. Michigan deployed a tax stamp with several overt and covert security features and a unique quick response (QR) code and serial number. QR codes (machine-readable codes consisting of an array of black and white squares, typically used for storing URLs or other information) can have purposes beyond tracking and tracing. The QR code can be read by consumers with a smartphone or tablet application to access information on smoking-cessation programs, report violations of the state’s youth access policies, connect to a tip line to report noncompliant packs and learn about the harms from illicit tobacco sales and purchases. Enforcement authorities can validate stamps using the smartphone- based eTRACS (Electronic Tax Reporting and Audit Compliance System). As part of the system’s implementation, the Michigan State Police department created teams of enforcement officers in each of the state’s seven districts and the state Department of Treasury created its own enforcement team. Sources: (37–42). Box 3.5 The ISO standard on excise tax stamps The ISO published its excise tax stamp standard (ISO/TC 292/SC) in October 2018. The purpose of the ISO standard is to assist tax and finance authorities in enhancing compliance with excise tax regulations. A tax stamp is defined as a visible tax stamp, label or mark placed on certain types of consumer goods to show that the applicable excise tax has been paid. The ISO standard applies to tax stamps that are physical in nature – not to digital markings, which are directly printed on to packs without a physical component. “Authentication” in this standard refers to the authentication of the tax stamp, not the product on which the tax stamp is affixed. In other words, authentication of a tax stamp on a cigarette pack means that the tax stamp is authentic but does not guarantee that the pack is authentic. In addition, control measures are needed at the time of the application of the stamp to verify the conformity of the tax stamp with the corresponding product. The standard provides guidance on the content, security, issuance and examination of physical tax stamps used to indicate that the required taxes have been paid and that the tax stamp is authentic. The use of stamps to facilitate tracking and tracing within the supply chain is not described. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 125 Specifically, the ISO standard deals with the following issues: • defining the functions of a tax stamp • identifying and consulting with stakeholders • planning the procurement process and selection of suppliers • the design and construction of tax stamps • the overt and covert security features that provide protection of the tax stamp • the finishing and application processes for the tax stamp • security of the tax stamp supply chain • serialization and unique identifier codes for tax stamps • examination of tax stamps • monitoring and assessing tax stamp performance. A stamp may fulfil many functions, but the core business of tax stamps is to ensure and facilitate the collection of revenue. The tax stamp must use a combination of security features. The tax authority should ensure that the tax stamp can be authen- ticated and that counterfeit, altered, tampered or otherwise fraudulent tax stamps can be detected. The standard provides detailed information on the different components of the tax stamp such as the substrate, inks, adhesives, laminate, authentication or security features and the unique identifier that should enable checks on the payments of the required tax. The process of procurement is discussed in detail in the standard. The tax author- ity should ensure that the procurement process is open, transparent and meets the sustainability objectives. The tax authority should set out the goals and requirements to give tendering organizations more leeway in proposing optimum solutions that might be different from those the authority would specify. The standard is not prescriptive; rather, it provides a catalogue of options. It does not, for instance, recommend specific security features, but it does describe the different types of features that are necessary for a tax stamp to be secure. Tax officials still need to make decisions and choose the option that suits them best, but the standard remains recommended reading for those who would like to introduce tax stamp programmes in their jurisdiction. ISO standards are not freely available but can be purchased at the ISO Store (www.ISO.org) or from an ISO national member body. Source: (43). According to Article 8 of the Protocol, each Party shall require that unique, secure and nonremovable identification markings – such as codes or stamps – are affixed to or form part of all unit packets, packages and any outside packaging of cigarettes for the implementation of the tracking and tracing system within a period of five 126 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N years, and of other tobacco products within a period of 10 years, of entry into force of the Protocol for that Party. The Protocol specifies that at least the following information shall form part of the unique marking: • date and location of manufacture • manufacturing facility • product description • where available, the intended market of retail sale. In several countries, QR codes are used as fiscal markings for tobacco and alcohol tax control. Each stamp has a unique identifier code and a QR code. The data stored in the QR code provide the following product information: • manufacturer • production location • stamp order date • tax status and class • brand • intended market • unique identifier (serial number). KEY TAKEAWAY 10 The use of fiscal markings is generally considered to be an appropriate tool for increasing compliance with tax laws. Fiscal markings can also be helpful for distinguishing between genuine and illicit tobacco products. 3.4.5 TRACKING AND TRACING A tracking and tracing system assists authorities in determining the origin of tobacco products – and the point of diversion, if applicable – as well as monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to enable authorities to have information on all transactions through the entire tobacco product supply chain until duties are paid or other obligations are discharged. Traceability is not used only for tobacco products. It is also used to improve the supply chain function, as in the case of parcel services, as well as for product safety reasons, to manage potential product recalls and for regulatory reasons. Tracking is the process that monitors where a product is at all times while also creating a time and location record for all movements. Tracing is the ability to identify the past locations of a product, so that the product’s route can be followed CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 127 back to its origin (44). In other words, traceability is “the ability to trace the history, application or location of an object” (45). A tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it is possible to unambiguously register that product’s movements. Other necessary characteristics include the ability to share the registered move- ment information and to authenticate products. This enables a product’s status to be captured through the supply chain and its history to be identified and verified retrospectively. According to Article 8.4.1 of the Protocol, Parties should require the following information to be available: • the date and location of manufacture • the manufacturing facility • the machine used • the production shift or time of manufacture • the name, invoice, order number and payment records of the first customer not affiliated with the manufacturer • the product description and intended market of retail sale • any warehousing and shipping • the identity of any known subsequent purchaser • the intended shipment route, date, destination, point of departure and consignee. A good tracking and tracing system enables the government to properly monitor the supply chain, improves its ability to ensure collection of the proper duties and taxes, provides it with the ability to authenticate whether the identification marking is genuine and matches the product and improves its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. The following elements are required for an effective tracking and tracing system (46): • A serialized unique identification marking for each package of product. These identifiers are a distinctive combination of numbers, letters or both. They cannot be predictable or used more than once. The representation of the identifier on the package can be human-readable (letters or numbers) or machine-readable (barcodes). Generation of codes and encryption that are part of a tobacco industry patent should be excluded. • A data carrier with the serialized unique identifier and other information such as date and location of manufacture, manufacturing facility, product description and, where available, the intended retail market. This informa- tion should be readable by authorized agencies of any Party to the Protocol. The data carrier should comply with quality standards and be suitable for 128 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N high-speed production lines. Two-dimensional barcodes, for example, meet these standards and are readable with inexpensive equipment. • A link and parent-child relationship (called aggregation) between different packaging units that offers the option to trace a pallet without the need to scan all the packs and master cases of that pallet. • Recordkeeping of all shipping and receiving events along the supply chain. This includes, for example, the departure location and the arrival location, as well as the involved operators. International standards from the ISO are recommended for the capture and exchange of data and events. • The use of international standards for key information that is encoded in the data carrier (5). An example of a unique and internationally recognized identifier for products is a Global Trade Item Number. The following details on information storage and sharing are drawn from various sections of the Protocol. Data and events along the supply chain must be stored in an independent database that is controlled by competent government authorities. At the global level, national and/or regional databases can be interconnected to facilitate international inquiries by competent authorities. Parties to the Protocol agree to establish a global information-sharing focal point located at the Conven- tion Secretariat of the WHO FCTC, accessible to all Parties, enabling them to make enquiries and receive relevant information. Each Party shall ensure that the information recorded under paragraph 5 of Article 8 of the Protocol is accessible to the global information-sharing focal point on request, subject to paragraph 9, through a standard electronic secure interface with its national and/or regional central point. The global information-sharing focal point shall compile a list of the competent authorities of Parties and make the list available to all Parties. The cost of tracking and tracing systems is a concern for many countries, but as indicated in paragraph 14 of Article 8 of the Protocol, jurisdictions may require the tobacco industry to bear any costs associated with putting in place the tracking and tracing system in a country (46). In Brazil, the cost for cigarette manufacturers was US$ 0.0185 per pack (42). In Kenya, the cost for manufacturers was US$ 0.024 per pack (42). Along with considering the characteristics of a tracking and tracing system in selecting a particular one, it is important to avoid conflicts of interest, ensure fair and transparent dealing with suppliers, implement a zero-tolerance policy for corruption or anti-competitive behaviour and ensure compliance. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, which deals with industry interference, and Article 8 of the Protocol. Article 8.13, which states that “each Party shall ensure that its competent authorities, in participating CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 129 in the tracking and tracing regime, interact with the tobacco industry and those representing the interests of the tobacco industry only to the extent strictly necessary in the implementation of this Article”. Box 3.8 provides a cautionary example of a tracking and tracing system that is not compliant with Article 5.3 of the WHO FCTC. While the Protocol contains a great deal of information on the requirements that a tracking and tracing system should meet, questions come up in relation to the implementation of such systems. To achieve the objectives of the Protocol, the Meeting of the Parties (MOP) to the Protocol, as the governing body of the treaty, has the prerogative to establish subsidiary bodies, such as expert groups and working groups. In decision FCTC/MOP1(6), the MOP established a working group for the development and implementation of tracking and tracing systems in accordance with Article 8 of the Protocol, including the global information-sharing focal point (Article 8.1) and unique identification markings for cigarette packets and pack- ages (Article 8.3), to further elaborate on the next steps. The working group will produce a comprehensive report compiling good practices and experiences on the implementation of tracking and tracing systems, as well as unique identification markings for cigarette packets and packages at national or regional levels. The working group was also given a mandate to prepare a conceptual analysis of how a global information-sharing focal point could be set up. Implementing a complete tracking and tracing system with fiscal markings takes time. In most of the countries that have already implemented tracking and tracing, it took several years from starting with the legal framework to final implementation. Several hurdles need to be overcome: • Legal framework approval is usually delayed by the tobacco industry. • Knowledge of tracking and tracing and associated technologies is scarce at tax administrations. • Tender and bidding processes are complex. • Coordination between domestic tax authorities and customs is weak. Even though the process might be lengthy, the investment in a tracking and tracing system will be repaid with the amount of tobacco taxes that are not lost due to evasion. When implementing a new tracking and tracing system, tax administration should ask for collaboration and technical assistance from intergovernmental organizations and countries that have successfully implemented such systems, in order to speed up and ensure success of the process. Examples of tracking and tracing systems implementation in Chile, Kenya and the EU are detailed in Boxes 3.8, 3.9 and 3.10. For countries that already have fiscal markings in place, the potential interaction between the markings and the implementation of a tracking and tracing system should be taken into account. Further information on this interaction is presented in Box 3.6. 130 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.6 Tracking and tracing and fiscal markings It is becoming increasingly common for stamps to contain some tracking and trac- ing features, such as unique identification markings and basic information on the product that describes the company, tax status or the product itself. The intention is to mark each pack with a unique identification marking so it can be monitored from the point of production to the retailer, including each step in between, thereby creating a complete time and location history. Although a tax stamp could meet the requirements of Article 8 of the Protocol and have tracking and tracing features, in general, the focus of tax stamp systems differs from that of tracking and tracing systems. Tracking and tracing is more than the unique, secure and nonremovable identification markings on the packages of tobacco products. It implies reading or scanning the codes; linking the codes between packs, cartons, master cases and pallets; uploading the information to a database; recording of any shipping and receiving events along the supply chain; and interconnecting the different databases. While new tax stamp programs contain tracking and tracing features, they are primarily intended to facilitate tax collection on the domestic market and not to track duty-suspended cross-border trade or the export of products. The focus of tax stamp systems is on authenticity and the proof that taxes are paid. The focus of tracking and tracing systems is on unique identification and on control of the movements in the supply chain by monitoring and investigating the past and future location of products. Tax stamp programs focus on stock management, verification (that the stamps correspond to the product) and authentication (that the stamps are genuine), while the focus of tracking and tracing systems is on the origin, intended route, first customer and final destination. The focus of tax stamps is primarily on individual packs intended for the duty-paid domestic market, while the focus of tracking and tracing systems is on all packaging (packs, cartons, master cases, pallets) and certainly – but not exclusively – for the duty-suspended export market. Nevertheless, sometimes there can be synergies. For example, the EU countries that require a tax stamp or national identification mark for fiscal purposes have the option to use it as the security feature for tracking and tracing purposes, provided that the requirements are met. In summary, tax stamps can be converted to or be part of a tracking and tracing system when the converted system provides aggregation between packs, cartons and master cases and records all movement along the sup- ply chain. For the export market, a unique identification marking should be added. Sources: (39–42). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 131 Box 3.7 What not to do: use the industry solution to tracking and tracing Codentify is a serialization system used to produce cigarette pack markers for the purpose of verifying whether cigarette packs are legal. It was patented by PMI but subsequently licensed at no cost to other major cigarette manufacturers. In 2016, Codentify was transferred to Inexto, which is an affiliate of the French group Impala. What is the problem with Codentify/Inexto? The main issue is that Codentify/Inexto’s links to the tobacco industry make it incom- patible with the Protocol, which came into force in September 2018. The Protocol specifies that obligations assigned to a Party shall not be performed by or delegated to the tobacco industry. Additionally, many elements indicate it is an ineffective means of authentica- tion. For example, the 12-character digital codes generated by Codentify can be easily duplicated or cloned and used as originals on either a counterfeit or genuine pack, which can then pass the system’s basic verification test. The codes are also produced by relatively unsecured, commercially available equipment and do not include high-security features capable of protecting the authenticity of identifier numbers. Systems that use multilayered, advanced security solutions that enable distributors, retailers, customers and authorities to identify noncompliant products are more secure. Another problem is that Codentify/Inexto cannot track products as efficiently as other available systems. It requires a much larger enforcement capacity to achieve the same detection rates as other systems that are not linked to the tobacco indus- try. Authorities would have to inspect significantly more packs marked under the Codentify system than is necessary under some other systems to achieve the same certainty of not missing a fraudulent pack. In addition, not all stakeholders will be able to verify that a pack marked under the Codentify system is genuine, while other available systems do offer this possibility. Source: (47). 132 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.8 A successful tobacco traceability system: SITRAF, Chile The Servicio de Impuestos Internos (SII, Internal Revenue Service of Chile) has success- fully coordinated and implemented a tracking and tracing system for tobacco products. Application of a compliance management model In the framework of a compliance management model, a traceability system for tobacco products utilizes a structural measure to reduce tax evasion. It is estimated that evasion of taxes (VAT and excise taxes) in the cigarette market in Chile amounts to 16.6% of the country’s total market annually – approximately US$ 300 million. The traceability system implemented in Chile (SITRAF (TAB2)) allows authorities to know in a certain and timely manner the quantities of cigarettes produced or imported into the country. Moreover, it helps authorities to distinguish between counterfeit products and original products that did not comply with payment of the tax. In 2018, the implementation of the traceability system was awarded to a company through public bidding. The company is in charge of the implementation and operation of the system for five years, according to a contract signed with the SII, and must maintain a team of 20 people available for the project. Direct markings are applied to items produced in Chile for national consump- tion, and stamps are used for imported products. For both types of product, the marking is based on a data matrix code, which is printed using security ink that is distinguishable from any other type of ink with specific devices that are provided by the awarded company. Although products for export are not subject to marking, they are controlled and accounted for by the traceability system. For national production, devices are installed on each production line that rec- ognize the type of pack being produced, print a unique code on each pack and then read it (activation) to save all the information on servers located in the production plant. This information is transmitted to the central servers of the system and then to the SII. It is also available for on-site inspection. In the case of imported products, the stamps must be acquired in Chile by each importer and then sent to its producer abroad, which is responsible for adhering them to each pack of cigarettes prior to wrapping the packs with cellophane, us- ing applicators on the production lines. Once the cigarettes enter Chile, the tax determination process has been completed in the service (Provisional Free Transit Guide) and the corresponding taxes have been paid to customs (Import Declara- tion), the importer must enter the data on the stamps used by the importer on the platform of the traceability system. After validation, the stamps can be activated in the system – that is, they are recognized as valid for commercialization. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 133 In addition to the devices provided by the company, SII has developed a smart- phone application for verification by citizens. Although the application cannot verify the authenticity of the ink used, it is able to verify whether a code is correctly gener- ated and display the information contained in the traceability system for the code (brand, variety, quantity of cigarettes, products/importer) so that the taxpayer can verify its consistency. Progress of the compliance management model Implementation of the traceability system has required a coordinated effort both within the service – for the generation of instructions, procedures and computer developments – and with other institutions, such as the National Customs Service and the Ministry of Health. It is an unprecedented project in terms of coordinating the implementation of the system in the production lines of the different tobacco companies in the country and the provider company. Some of the main milestones of the project are: September 2014: Law 20,780 on Tax Reform establishes an obligation to implement the system within a term of six months, after the publication of the resolution determining the obligated tax payers. May 2015: Resolution No. 47 determines obligated taxpayers. June 2015: Circular No. 47 describes obligation to incorporate stamps or distinctive marks as a traceability mechanism. February 2016: Law 20,899 on Tax Reform simplifies the definition of the system, allowing the system to be outsourced or provided by the SII, in addition to making the type of traceability more flexible. August 2016: Traceability system regulation D.S. 1,027 is issued (published on 28 December 2016). March 2017: Exempt Resolution No. 49 of the Ministry of Finance authorizes the SII to outsource all or part of the traceability system. June 2017: Bidding bases in public market are published. February 2018: Tender is awarded to selected company. June 2018: Decision of contract is made by General Comptroller of the Republic. August 2018: Resolution No. 61 determines taxpayers obliged to apply the trace- ability system. August 2018: Holding of first workshops for detailed definitions of the project, with the participation of Customs, Ministry of Health, provider and SII. September – October 2018: Visits of plants to coordinate with producers and define adaptations to production lines for system implementation. January 2019: Resolutions No. 6, 7 and 8 passed, with response to a request for an 134 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N extension of producers; start of system implementation in all production lines in the country. February 2019: Resolution No. 16 establishes a term to commercialize the remaining stock without marking. 2019: Resolution No. 24, with request for extension to importer, includes training of customs staff. March 2019: The traceability system is started up. June 2019: Stamp process begins for all imported cigarettes. Currently, the system is installed and operating in all production lines in the coun- try, placing traceability markings on virtually 100% of the cigarettes produced and imported. During 2019, the traceability system enabled controlling approximately 1 175 million packs: 744 million produced for national consumption, 409 million produced for export and 22 million imported yearly. Source: (48). Box 3.9 Case study of Kenya’s implementation of a tracking and tracing system Kenya’s current tracking and tracing system was preceded by a series of reforms in both tax structure and administration of excise taxes. The reforms included electronic cargo monitoring of exports, which allowed for automatic monitoring and reporting. The system appears to be highly effective because it requires less capacity and is less prone to manipulation than earlier systems. The experience of Kenya shows that a lower-middle-income country can successfully implement a sophisticated system capable of decreasing illicit trade. It also shows the importance of other measures such as strengthening enforcement, increasing cooperation and communication among different agencies and increasing penalties for noncompliance. Illicit trade, as measured by the Kenya Revenue Authority (KRA), was estimated to be around 15% of total consumption in the market during the initial reform period. After the introduction of the new system in 2015, it dropped to 5%. Timeline of the major reforms: 2003: In this period, the paper tax stamps used had a unique identifier and were colour-coded to indicate the type of product. Regular compliance checks were in- troduced. In 2007, the cost of a stamp was 2.124 Kenyan shillings or US$ 0.023 per pack. However, the stamps were found to be easily counterfeited and could not be linked to specific brands. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 135 2008: The KRA proposed a tracking and tracing system and increasing tax rates. The new system was introduced gradually. 2010: Enhanced security features, including ultraviolet markings, were added to the paper stamps. The stamps were to be clearly visible when packs were displayed for sale and placed so that opening a pack would destroy the stamp. The stamps were verified at four different points along the supply chain. The costs were just slightly higher than those of the previous stamps at US$ 0.024 per pack. Licensing was introduced for domestic manufacturers, subject to annual renewal. Importers were required to register with the KRA. Licences required submission of details on the company directors, inventories and equipment, accounting systems, input-to-production ratios and brands produced. Penalties for noncompliance were increased and included up to three years in prison. An electronic cargo tracking system was launched. Electronic seals were affixed on containers or trucks, and GPS technology was used for tracking. A bond was payable on exports to cover excise and VAT taxes. The bond was released only when the goods reached the final destination and taxes were paid. Verification involving both countries of the business deal takes place at the bor- ders. The electronic system provides information about the departure and arrival of the goods and the disarming of the seals. Authorities in the importing country are notified before the shipment leaves the domestic production facility. The system reduces the number of checkpoints and staff needed and generates arrival reports that can be verified with VAT refund requests. As a result of these changes, three factories and seven of the 10 importers were shut down due to noncompliance. Exports to Côte d’Ivoire, Eritrea, Mali and Sudan stopped because companies could not provide evidence that the goods reached the final destination and taxes were paid. More than US$ 11 million in excise tax losses was recovered in 2011. The KRA estimated that illicit trade dropped to 8%. 2013: A contract was signed to introduce a tracking and tracing system for tobacco and alcohol, the Excisable Goods Management System, in April. The system added production counting, tracking and tracing, stock control, processing and other data collection to the existing system. Infrastructure requirements included high-speed broadband internet at production facilities, warehouses, the KRA and ports, along with reliable power or backup generators at those points. Implementation was planned in three stages: • Stage 1 – A new electronic digital stamp with a unique identifier was introduced. It included a data matrix code plus overt markings (holograms, fluorescent fibres, a security link for KRA authentication and visible two-dimensional codes for verification and activation), semi-covert markings (UV features, fluorescent prints detectable by specialized devices, mini text printing for retailers and 136 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N distributors) and forensic taggants for use in prosecutions. The stamp also included human-readable codes for verification by short message service using the KRA web portal. • Stage 2 – Control and monitoring systems were automated in February 2014. Manufacturers had to install photosensitive readers on production lines, with data automatically sent to the KRA in real time. Each stamp was activated and associated with a brand and package size on the line. The KRA database is automatically updated every 15 minutes. • Stage 3 – Market surveillance began, with 83 officers given powers to seize illicit cigarettes and make arrests. The officers were equipped with hand-held devices that transmitted data to the KRA for authentication. Distributors and retailers became liable for selling products without an excise stamp and were subject to fines plus prison sentences of up to three years for noncompliance. In 2016, a smartphone application became available with which the public could authenticate cigarette packs. Importers must now buy digital stamps and send them to export facilities in other countries to be affixed. Tax liability is due at removal from a factory or at import. The electronic cargo monitoring system is still in effect. 2016: The Excise Duty and Tax Procedures Acts clarified new obligations and penalties. 2017: A new integrated customs management system was launched. The KRA estimates that illicit trade levels are now around 5%. The current, more comprehensive digital system is cheaper than the previous paper tax stamp system. Manufacturers pay for the production monitoring system, but it counts as a busi- ness expense on corporate tax returns. In 2018, two manufacturers and 10 licensed importers were operating in Kenya. In 2018, aggregation between the markings of packs, cartons and master cases had not yet been implemented but was expected to be forthcoming. Source: (36) Box 3.10 Case study of the new EU tracking and tracing system, May 2019 Cigarette smuggling and other forms of illicit trade in the EU is estimated to cause a loss of €10 billion in revenue annually. In 2018, 4.2 million packs (20 sticks per pack) of illegal cigarettes were seized by customs in the EU. Illicit tobacco production was also increasing: an illegal factory in Ireland, dismantled in 2018, was capable of producing 250 000 cigarettes per hour. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 137 The EU tobacco control policy is described in the Tobacco Products Directive and is influenced by the Protocol. Article 15 of the Directive calls for the traceability of cigarettes and RYO tobacco products by May 2019 and of other tobacco products by May 2024. The EU tracking and tracing system is sufficiently flexible to be imple- mented at both the regional and the single-country level. Countries can choose among providers as long as the basic requirements are met. The policy provides a high level of protection against any attempts at manipulating the data. The report- ing obligations cover all the economic operators involved in the manufacture and distribution of tobacco products. The EU system requires all unit packets of tobacco products to be marked with a unique identifier Information on the movements of those products is to be stored by third-party data storage providers. The suppliers of the unique identifiers and data stor- age are to be financially and legally independent from the tobacco industry. The data are to be fully accessible to authorities of EU Member States for enforcement purposes. The generation of unique identifiers, as well as all other codes required for pre- registration of economic operators, facilities and machines, will be done at the Member State level by designated identifier issuers. Manufacturers and importers are required to supply information relating to the product and production lines when requesting unique identifiers from the issuers. The issuers will then generate and deliver batches of unique identifiers. On the production line, manufacturers of tobacco products will complete each unique identifier with a marking indicating the date and time. The unique identifier will be a machine-readable, optical, one- or two-dimensional barcode. An anti-tampering device, capable of creating an unalterable independent record of the verification process, must have been installed previously. This additional record will be accessible to public authorities for potential investigation and inspection. Unit packets, as well as aggregated packages such as cartons, master cases or pallets, can be tracked and traced throughout the supply chain. Tracking is also allowed at an aggregated packaging level as long as unit packets remain traceable. During transport, each dispatch and arrival up to the final dispatch to the first retail outlet must be recorded and reported. All recorded information must be submitted to the independent third-party data storage facility, generally within three hours, and 24 hours before dispatch and transloading. Costs, including operational costs, are shifted to the tobacco industry, in line with Article 8 of the Protocol. The EU system of tobacco traceability and security features became operational on 20 May 2019. Sources: (36, 49–50). 138 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 11 A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as monitoring and controlling the movement of tobacco products and their legal status. 3.4.6 ANTI-FORESTALLING “Forestalling” is a term that describes increases in the production or stock of products in anticipation of a tax increase (2). Other terms referring to this practice include “stockpiling” and “front-loading”. Forestalling occurs when manufacturers or importers increase their tax-paid stock or oversupply the market by increasing production or imports in order to pay the previous lower rate. It reduces and delays the effective- ness of tax measures. The effective starting date of the new tax rate will be delayed, revenues will be lower and the possible effect on prices and thus consumer behaviour will also be postponed. To illustrate how anti-forestalling measures function, an example is presented in Annex 3.2. A legal basis must exist for anti-forestalling measures; otherwise, the govern- ment cannot prevent the industry from forestalling. Legal measures to deal with forestalling include (51–52): 1. Limiting the amount of tobacco products that can be released subject to the old tax rate and levying the new tax on the products exceeding that limit. 2. Levying the new tax rate on all goods that are still in stock and not yet sup- plied to the final consumer. 3. Limiting the number of tax stamps issued at the rate that was in effect before the increase or limiting the time that products with a tax stamp with the old rate can be sold. 4. Requiring producers and importers to buy new tax stamps annually or after a tax increase. Under the first three measures, the competent authority determines the limit for taxation at the previous (lower) rate. The quantity allowed may be based on the shelf life of tobacco products – around six months for cigarettes – or normal inventory levels, such as an average over the previous three years. The first measure, limiting the amount of tobacco products that can be released, requires resources from the competent authority for enforcement. Authorities may decide to post inspectors in each production facility, but even without posting inspectors, procedures are necessary for determining when the allowed quantity has been exceeded and what subsequent actions to take. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 139 The second measure, levying the new tax on goods in stock and not yet supplied to the final consumer, might be difficult to implement. The competent authority is required to monitor the manufacturing process and, at the very least, to conduct a stock-taking of all players in the supply chain – including the various manufacturers, importers, wholesalers and retailers – in a very short window of time. If monitoring covers only the stock of manufacturers and importers, this measure could easily be circumvented by ensuring the stock is sold to others in the supply chain or by setting up separate distribution companies to purchase the stock. Controlling stock at the retail level is burdensome and might not be administratively feasible given the large number of cigarette retailers. It becomes even more burdensome if there is no licensing requirement for retailers because they will first have to be identified. The third measure accomplishes the same thing as the first if a country uses tax stamps. The fourth also requires tax stamps and is simpler for the competent authority but somewhat more burdensome for the tobacco companies, since stamps must be purchased every year. Box 3.11 provides examples of anti-forestalling measures in EU countries. Box 3.11 Examples of anti-forestalling measures in EU countries Several EU Member States have taken measures to limit forestalling. A cautious ap- proach seems to be required for designing such measures to ensure that they comply with EU legislation and the general principles of EU law – in particular, the principle of proportionality. No disputes have occurred concerning the right of initiative of EU Member States to implement anti-forestalling measures. Nevertheless, several EU Member States had to defend their measures in front of the Court of Justice of the European Union, the institution that ensures all national legislation is in line with EU law and a consistent application of that law (53). The Court acknowledged that anti-forestalling measures are appropriate to combat tax evasion and tax avoidance. Moreover, the Court emphasized that fiscal legislation is an important and effective instrument for discouraging consumption of tobacco products and therefore for protecting public health (54–55). However, the measures taken should be proportion- ate to the objectives. The principle of proportionality means that only the action needed to achieve the objective should be taken, and it should not exceed what is necessary. This principle regulates the measures taken within the EU and is included in the Treaty on the European Union. The Court demanded that Portugal amend its legislation to ensure compliance with the principle of proportionality. Belgium, Estonia and Hungary were also urged to change their anti-forestalling measures to bring them in line with EU legislation (56–57). 140 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The majority of the EU Member States have anti-forestalling measures in place, but there is no harmonization of these measures. The following are some examples:12 • Portugal limits the quantity of cigarettes that can be released in the last four months of a year to the average of the previous 12 months plus 10%. In addition, manufacturers and importers must sell cigarette packages with a tax marking of the preceding calendar year within three months. For other tobacco products, longer limits apply. • Denmark limits the number of tax stamps issued before a tax increase at the old rate to 20% more than are usually purchased in the two months before the end of the year. • In Poland, tax stamps are valid only for the current calendar year, and cigarettes with the old stamp can be sold only through February of the following year. • In Romania, companies must apply for approval to release for consumption from the customs office. Source: (58). KEY TAKEAWAY 12 Forestalling reduces and delays the effectiveness of tax measures. Implementing anti-forestalling measures can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. 3.4.7 ADDITIONAL NATIONAL AUDITS AND CONTROLS In addition to the previously described measures, several periodic audits and controls could be implemented to increase compliance with tax laws. The most common audits and controls are the following: • Cost audit – The cost audit method provides expected VAT and tobacco tax collection by simulating the intermediate and final cost of cigarettes. It starts with the inventories of raw materials and estimates added values and final cost, then matches the results with real collection from the tobacco supply chain. Annex 3.1 provides more information about the components that make up some selected (tobacco) products. • Transfer pricing audit – To ensure companies pay their fair share of tax, prices of transactions between related companies should be assessed, and when prices are not in line with the market conditions, they should be corrected. Companies that operate at the international level (transnational companies), including many tobacco companies, can manipulate import or export prices 12 Considering the frequency of court cases concerning anti-forestalling incidents, these measures might have been replaced or amended. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 141 of merchandise or raw material to related companies or branches in other countries, with the objective of lowering profits in countries with higher tax rates, and can transfer those profits to countries with lower taxes. • Price and market monitoring – Retail price surveys can provide information about variance from the market price in certain locations, highlighting areas of potential tax avoidance or illicit trade. Physical control of such locations requires rapid response teams, as implemented in the Philippines. To monitor the tax compliance of its taxpayers, the competent authority in the Philippines needs to understand the tobacco market; it must have information on brands, market segments and prices of products. This information enables authori- ties to estimate the impact of tax and price changes on consumer behaviour and revenues. Market data can be analysed as part of risk management and anti-fraud analysis to determine whom to investigate for noncompliance and when to do so. Sales data can be triangulated to validate other data sources, such as household surveys on prevalence. Market data and trends are also useful indicators for determining whether there is a case of oversupplying. See also section 3.4.3. • Consumer control – Involving the public via awareness campaigns has also been shown to be effective. Consumers have the right to be assured that the products available in the market are authentic and come from legitimate sources. Thus, it is in the consumers’ interest to understand and be able to verify that they are buying genuine products. The features of the fiscal marks on tobacco products should help consumers distinguish between genuine and illicit products. Some countries – Kenya, for example (see the case study in Box 3.9) – use a smartphone application to allow anyone to check both covert and overt features and to report any cigarettes with incorrect markings. Other countries, such as the Netherlands, have developed a smartphone application that allows anyone to report a suspected case of excise tax fraud. • Cross-check controls – Competent authorities should consider using multiple sources to obtain market data and determine if these data are consistent with tax declarations. VAT declarations can be used to verify that suppliers and purchasers of raw materials and final products are reporting the same amounts. Bank information can be used to verify both sides of transactions along the supply chain. Any discrepancy can alert the competent authority to conduct further investigation for possible illicit trade of tobacco or tax evasion. 142 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 13 Several different types of periodic audits and controls that can be carried out to increase compliance, including cost audits, transfer pricing audits, price and market monitoring, consumer controls and cross-check controls. 3.4.8 IMPORT AND EXPORT CONTROLS Parties to the Protocol should allow import and export of tobacco products and manufacturing equipment only by duly licensed natural persons or legal entities (Article 6.1). A well-known strategy used by fraudsters is to declare products for export so that no duties are due to the country of export. These products are subse- quently transported through other countries, using the in-transit regime that allows temporary suspension of duties until the goods arrive at their final destination. Before arriving at their final destination – where the excise duties would be due – the goods disappear or are lost while being diverted to the illegal supply chain. The goods may never leave the country, or they may be smuggled back into the country from which they were exported without declaring or paying duties. This risk of loss of revenue can be mitigated by requiring a guarantee or bond, which will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless the required fiscal marking (such as a tax stamp or export label) is affixed on the pack, according to the national law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Illicit tobacco trade could be decreased significantly if the various competent authorities that have jurisdiction over manufacturers and exporters of tobacco prod- ucts and manufacturing equipment would provide the competent authority at the destination with prior information when a shipment has been authorized and is about to take place. The information could include the name of the consignee, a description of the item shipped and the quantity. Also, the competent authority at the destination should inform the competent authority having jurisdiction over the shipper that a shipment was received along with the pertinent information relating to the shipment. A good IT system is also required for import and export. Electronic processing of prior-to-arrival manifest and import declarations is recommended. Most countries have implemented an online customs system to process import and export declara- tions, including all required data such as country of origin or export, description of merchandise, value, weight, cargo insurance, carrier, importer or exporter and broker identification, detailed tax duties to pay and final destination. The World Trade Organization (WTO) Trade Facilitation Agreement provides sev- eral tools for better controls, including collaboration between customs administrations, CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 143 exchange of information, use of non-invasive devices and prior-to-arrival import declaration (23). Non-invasive detection equipment at customs posts is highly capable of detect- ing contraband merchandise. The most common tools are X-ray scanners that are used for small parcels, containers, trucks and trains. Most modern ports have also implemented the use of X-ray scanners, and such technology is improving the speed of controls to as little as two minutes per container. Although the cost of scan- ners is declining, it remains inaccessibly high for countries with limited resources. Fortunately, scanners are often available for lease, making them accessible for tax administrations in those countries. Less sophisticated and less costly detection equipment includes endoscopes, mirrors, night vision equipment, cameras and automatic licence plate readers (33). A still less expensive alternative is the use of dogs, which can be trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied in order to reduce contra- band. These include the separation of processing operations from the sealed storage of taxed and untaxed products. Physical and direct control by officials of the excise authority during a part or the whole of an operation can be applied (for example, physical escort of the transit consignment from border to border by individual trucks or in a convoy, or application of radio or satellite tracking systems such as GPS-enabled devices to goods, conveyances, vehicles or containers). Control at borders is essential and should include integrated technology and cooperation with agencies at the border station. Frontline officers should be sup- ported by appropriate intelligence, guidance and supervision from management, as well as technical aids to enforcement. Within a country, mobile excise control units are helpful for verifying excisable goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. These operations require close coordination between police, border guards and other public services. Exports also require special attention, in particular if VAT and tax refunds are granted to the export of tobacco products. Validating the real exit in the declared amount is essential to avoid illegal re-entry to a territory and the improper refund of taxes. For any tax refund, an audit including tax credit information must be carried out. The audit may include the invoices for the whole chain involved in the export, including tobacco farmers, first processors, manufacturers, wholesalers, storage and transport. 14 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 14 To ensure control of import and export, it is recommended that only duly licensed persons and entities be allowed to import and export tobacco products and manufacturing equipment. 3.4.9 FREE ZONES AND TRANSHIPMENT POINTS The term “free zone” is very broad and can refer to a number of different types of areas. The Financial Action Task Force listed the following types in its 2010 report (59): free trade, export processing, enterprise, free ports, foreign trade, special economic zones and bonded warehouses. A number of these areas can include tobacco manufacturing and trade. By definition, controls such as regulation and oversight within free zones are less strict than in other areas. This can make them appealing to persons involved in illegal cigarette manufacturing or trade (17). In fact, illicit activities related to free zones (not limited to tobacco) are regularly documented by organizations that recognize the linkage. These activities include money laundering, tax evasion and trade in counterfeit goods or other illicit goods (60). A report from the European Parliament (61) referring to free ports in particular, mentioned that the motivation for using them included a “high degree of secrecy and deferral of import duty and indirect taxes”. The report even proposed the “urgent phasing out of free ports”. In the European Parliament report, free ports are free zones that function as (semi-) permanent storage areas for high-value goods. The Protocol includes a time-bound provision of effective controls on all manu- facturing and transactions of tobacco products in free zones (Article 12). Free zones are defined as a part of the territory of a Party where goods are considered to be outside the customs territory for import duties and taxes (Article 1.5). This is the same definition used in the International Convention on the Simplification and Harmonization of Customs Procedures (Revised Kyoto Convention) (62). Parties to the Protocol must implement effective controls in free zones within three years of entry into force of the Protocol. For countries not yet Parties to the Protocol, stringent controls of manufacturing and transactions involving tobacco products in free zones are an important component of an effective and efficient tax administration. One of the measures for dealing with free zones within the Protocol includes implementing “effective controls on all manufacturing of, and transactions in, tobacco and tobacco products, in free zones, by use of all relevant measures as provided in this Protocol”. As indicated in an Interpol report (17), a significant vulnerability of free zones is the fact that different economic operations (e.g. manufacturing, assembly, re-packaging and warehousing) take place outside the control of authorities. It is therefore essential for customs administrations to exercise their authority in free zones CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 145 to effectively identify and fight illicit trade in tobacco products. Relevant measures listed in the Protocol should be applied. These include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a track- ing and tracing regime. Removing exemptions on excise taxes is an additional way to increase control and remove incentives for using free zones as a means for tax evasion. Indonesia, for example, imposes excise taxes on cigarette manufacturing in its free trade zone.13 Parties to the Protocol shall also prohibit the intermingling of tobacco products with non-tobacco products in a single container or any other such similar transporta- tion unit when removed from free zones. Finally, each Party is to “adopt and apply control and verification measures to the international transit or transhipment of tobacco products and manufacturing equipment in conformity with the provisions of the Protocol”. Article 13 of the Protocol, which covers all duty-free sales of tobacco products, requires Parties to the WHO FCTC to consider prohibiting or restricting the sale to or import by international travellers of tax-free or duty-free tobacco products, as mentioned in Article 6 of the WHO FCTC. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect government revenues by creating a loophole in the tax structure (2). KEY TAKEAWAY 15 Customs administrations should exercise their authority in free zones to prevent different economic operations from taking place outside the control of authorities. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a tracking and tracing regime. 3.4.10 PROCEDURES AFTER DETECTING ILLICIT TRADE OF TOBACCO The procedures described in previous sections are intended to increase compliance and to prevent illicit trade. When smuggling or illicit trade is detected – through, for example, audits, tracking and tracing systems, verification of declarations or border control – actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. To deter further illegal behaviour, a comprehensive audit of everyone and everything involved in the illicit acts must also be carried out. Assets and vessels involved in the illicit activity can be seized, and financial accounts can be frozen. Some countries, including the United Kingdom, Canada and Chile, have also adopted a strategy known as “follow the 13 Indonesian Ministry of Finance, personal communication, January 2020. 146 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N money” to obtain more information on those who finance illicit trade. This strategy is intended to have a further-reaching effect by targeting those who finance the transport, production and storage of illicit products. The United Kingdom, Canada and Chile all created special teams tasked with identifying and targeting the offenders. Furthermore, Article 18 of the Protocol provides for the confiscation and destruction of tobacco, tobacco products and manufacturing equipment. One of the difficulties faced by competent authorities in exercising the authority to seize and forfeit products and/or equipment used in the manufacture or distribution of tobacco products is the cost of keeping or storing the goods and/or machinery before destruction. Thus, the law should also provide for a mechanism and timetable for the disposal and/or destruction of seized and forfeited goods or machinery, while prescribing a mechanism by which these properties can still be presented as admissible evidence in a judicial proceeding. Boxes 3.12 and 3.13 provide examples of successful efforts to combat illicit trade in the United Kingdom and Indonesia. KEY TAKEAWAY 16 As soon as smuggling or illicit trade in tobacco products is detected, actions such as collecting taxes and seizing and destroying smuggled and/or illicit tobacco must be taken. Box 3.12 The United Kingdom’s experience in fighting illicit trade in tobacco products In 2000, illicit cigarettes accounted for 22% of the cigarette market in the Unit- ed Kingdom. To deal with the problem, Her Majesty’s Customs and Excise14 implemented a major anti-smuggling effort. The strategy was refreshed with ad- ditional resources and measures in 2011 and reviewed in 2015. The result was a steady decline in the illicit cigarette market to 10% by 2013/2014. The measures taken were comprehensive and included hiring 1 000 new customs officers and investigators. In addition, tobacco supply chain legislation was introduced, aimed at discouraging tobacco manufacturers from facilitating smuggling. Tougher sanctions included increased fines of up to £5 million levied on a manu- facturer, criminal prosecution with sentences up to seven years, confiscation of assets as part of the proceeds of the crime, payment of duty on the confiscated goods plus penalties up to 100% of the duty, prohibition of the sale of tobacco products for 14 By the time of the renewed strategy, the respective bodies were the HMRC and the United Kingdom Border Agency. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 147 up to six months, unlimited fines for selling tobacco without the United Kingdom duty-paid fiscal mark after 13 March 2015, referral for withdrawal of the transporter’s licence, use of immigration sanctions to refuse entry to the United Kingdom for tobacco smugglers and civil action, including bankruptcy. The cost of these measures was £209 million over the first three years of the program and around £100 million annually by 2008/2009. This figure covers only HMRC and excludes any costing of the United Kingdom Border Agency. In 2013/2014, tobacco tax revenues were £9.5 billion. After a review of the strategy in 2015, the controls on raw tobacco were strength- ened by the introduction of an approval system in 2017. Anyone who manufactures, purchases, acquires, owns or is in the possession of a tobacco products manufacturing machine must be licensed with customs as of 1 August 2018 (63). The United Kingdom ratified the Protocol on 27 June 2018. It was the 40th country to ratify, which was the trigger point for the Protocol to enter into force. Fig. 3.4 Estimate of the illicit cigarette market and United Kingdom tax-paid consumption Sources: (16, 64). Bi lli on c ig ar et te s UK tax paid consumption Illicit market 20 00 -01 20 01 -02 20 02 -03 20 03 -04 20 04 -05 20 05 -06 20 06 -07 20 07 -08 20 08 -09 20 09 -10 20 10 -11 20 11 -12 20 12 -13 20 13 -14 20 14 -15 20 15 -16 20 16 -17 20 17 -18 0 10 20 30 40 50 60 148 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.13 A success story: Indonesia reduced illegal cigarettes from 12% to 3% of the market In some countries, most of the illegal cigarettes are imported, but in Indonesia most of them are produced within the country by unregistered manufacturers that are usually home-based and relatively small. In several territories, specifically on the Island of Java, illegal cigarettes have been produced for generations. This practice is supported by the availability of raw tobacco materials and cloves, as well as cheap labour costs, especially for female workers. Indonesia also produces cigarette products that are not available in most other countries. These products – known as handmade clove cigarettes (sigaret kretek tangan) – contain cloves, and the production process covers blending, rolling and packing. Parts of the process are done by hand, and 99% of the labourers are women. The strategy for combatting the illegal cigarette trade in Indonesia is divided into two main parts: preventive actions and responsive actions. Indonesia’s success in effectively tackling illicit trade is attributed to the following key factors. Monitoring and surveillance Preventive actions consist of administrative measures – such as issuance of permits and the excise stamp purchasing mechanism – that use risk management by optimizing the Excise Service Information System (ExSis). With this IT system, the Directorate General of Customs and Excise (DGCE) can oversee both daily transactions and daily production from factories. When information of suspicious activities is obtained, DGCE can suspend the purchase of excise stamps. The efforts to fight the illegal cigarette trade also invite stakeholders to be involved by supplying information regarding high-risk areas and regional governments. Strategic communications and community involvement DGCE continuously disseminates information and conducts public education to fight illegal cigarettes. These efforts are conducted every year, using a special campaign slogan. In 2019, the slogan was “Gempur Rokok Ilegal” (“Fight Illegal Cigarettes”). Key performance indicators for DGCE units and offices Parallel with the above-mentioned preventive actions, DGCE also continuously conducts responsive actions: enforcement, investigation and audit activities in cigarette factories. Enforcement activities are planned and measured by consider- ing the limited human resources and the large scale of the monitored territories. To demonstrate the effectiveness of administrative and enforcement measures in curbing the trade of illegal cigarettes, both types of activities are translated into key CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 149 performance indicators for all DGCE working units and offices, including regional offices and personnel. Use of technology and intelligence Coordination between different DGCE offices responsible for monitoring the produc- tion and marketing of cigarettes is maintained by sophisticated IT applications that enable efficient distribution of information and investigation activities. The applica- tions are the Customs Intelligence and Tactical Centre for data and analysis and the Centre for Command and Control (Pusat Komando dan Pengendalian/Puskodal) for ensuring that sea patrols work effectively and efficiently. Independent evaluation To evaluate the efforts and activities to reduce the circulation of illegal cigarettes in Indonesia (e.g. cigarettes without stamps, with fake stamps or with used stamps), in 2016, the government commissioned the University of Gadjah Mada in Yogyakarta to conduct a survey using a stratified random sampling method. To maintain objectiv- ity and independence, an independent body from this well-known university was appointed to conduct the survey. The survey results showed that the level of illegal cigarette circulation in Indonesia was 12.1% of total consumption. In 2018, the DGCE commissioned the University of Gadjah Mada to conduct another survey. Results showed that circulation of illegal cigarettes had been reduced to 7.0%. In 2019, using the same method the university used, DGCE conducted a survey that showed a reduction to 3.0%. Fig. 3.5 shows the results of the surveys. Fig. 3.5 Share of illicit trade in total cigarette consumption in Indonesia, 2016–2019 Source: Customs and excise department, Ministry of Finance, Indonesia, personal communication, 2020. 2016 2017 2018 2019 12.1% 10. 9% 7% 3% 150 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The results of the actions taken can be used as feedback for DGCE in developing excise policies for both service and enforcement, including policies for excise tariffs. DGCE will provide recommendations for an optimum policy format for controlling consumption, maintaining labour protections, optimizing revenue and, most impor- tantly, constantly reducing the consumption of illegal cigarettes on a national level. 3.4.11 PENALTIES Penalties and sanctions must be sufficient to deter illegal activities. Otherwise, finan- cial penalties may simply be paid as a cost of doing business while the illegal activity continues. The Protocol specifies commitments for Parties and provides information on best practices for non-Parties. Article 14.1 in Part IV of the Protocol requires each Party to establish unlawful activities, including manufacturing, wholesaling, broker- ing, selling, transporting, distributing, storing, shipping and importing or exporting tobacco products or manufacturing equipment without the payment of applicable duties or taxes or without using fiscal stamps or other required markings or labels. Articles 14.2 and 15 of the Protocol mandate Parties to determine which of the types of unlawful conduct set out in Article 14.1 shall be criminal offences. Parties must adopt legislative and other measures to give effect to such determinations, as well as to define whether the liability for committing illicit trade in tobacco is a criminal, civil or administrative offence. Article 17 further provides that the Parties shall consider adopting measures as needed to authorize competent authorities to levy penalties in an amount proportionate to lost taxes and duties resulting from the commission of illicit trade. Box 3.14 provides a case study of how Colombia used penalties to fight illicit trade. Box 3.14 The use of penalties to combat illicit trade in Colombia In 2017, the specific tax on cigarettes in Colombia was doubled, increasing from COL$ 700 per pack in 2016 to COL$ 1 400 in 2017. The tax rate was tripled from 2016 levels in 2018, reaching COL$ 2 100 per pack. A provision was added to increase taxes annually after 2018 at the rate of inflation plus 4%. In 2015, before the tax increase, Law 1762 introduced a number of measures to fight illicit trade more effectively. The length of imprisonment for dealing in contraband cigarettes was increased from 3–5 years to 4–12 years. Moreover, government officials who facilitate illicit trade – or anyone involved in transporting or retail sales of illicit tobacco – face similar prison terms. The law allows vehicles used for smuggling to be confiscated, and penalties were increased for illicit trade that is conducted through areas such as special economic zones. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 151 Under the law, illicit trade is considered to be a source of money laundering, which means that the Financial Intelligence Unit can use the same methods it uses to investigate other illegal financial activities. This practice is not common. The law specifically created new sanctions related to alcohol and tobacco excise tax evasion, including the seizure of goods, fines, closure of retail outlets and the suspension or cancellation of licences, authorizations or registries. Arrests and seizures have increased under the new law. Indeed, since its enactment, law authorities reported that between 2016 and 2018, five criminal organizations were dismantled, 53 individuals were apprehended and 72 assets were confiscated. In addition, 2 236 individuals were apprehended and 503 vehicles transporting smuggled goods were confiscated, as transport of such goods is now also considered a crime under the law. More importantly, thanks to the large tax increases, consumption decreased while revenues increased substantially in 2017 and 2018 (see Fig. 3.6). It is estimated that illicit trade in cigarettes in five Colombian cities in 2016 constituted 3.5% of total consumption, a much lower estimate than the industry data suggest. In 2017, after nine months of the tax increase implementation, a similar study found that illicit cigarettes remained low, at 6.4% of total consumption. Fig. 3.6 Packs sold and tobacco tax revenue before and after the tax increase in Colombia, 2016–2018 Sources: (65–67 and Ministry of Finance, Colombia (Direccion de Apoyo Fiscal), personal communication, 2020). 2016 2017 2018 Cigarette sales, million packs Cigarette excise revenue, million US$ current 0 100 200 300 400 500 600 700 800 673.7 521 446.4 195.1 301.2 386.2 33.7% reduction 97.9% increase 152 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N For consumers in possession of illicit tobacco, the minimum penalty should be confiscation and destruction of the illicit tobacco products found in their possession, with payment required for the unpaid tax and duties on those products. In the state of California in the United States, it is illegal to possess a tobacco product on which taxes are due and not yet paid. The burden of proving that taxes have been paid is on those who have the products in their possession. The provi- sion is enforced by the California Department of Tax and Fee Administration and local law enforcement agencies. A violation is a misdemeanour, with a maximum fine of US$ 5 000 and/or up to one year in prison. Illegal packages are subject to seizure and forfeiture. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are using this type of legislation to address illicit trade of tobacco. Asset confiscation and increased penalties for involvement in illicit trade are becoming more common as well. Withholding or even confiscation of trucks involved in smuggling is also common in several countries. KEY TAKEAWAY 17 Penalties and sanctions imposed should be sufficient to deter illegal tobacco trade activities. Penalties should be levied in amounts proportionate to lost taxes and duties resulting from illicit trade. 3.5 TAX ADMINISTRATION OF OTHER TOBACCO PRODUCTS In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. However, there is a lack of standard- ization of other products and sometimes large informal markets. For example, it is estimated that two thirds of waterpipe tobacco in the EU is non-duty-paid (19). Other tobacco products – such as bidis in South-East Asia, waterpipe tobacco in the Eastern Mediterranean region and snus in Sweden – are considered part of a country’s traditions. This sometimes leads to situations where governments are hesitant to strongly regulate and tax these products. Some products, such as kreteks (clove cigarettes) in Indonesia and bidis in India and Bangladesh, are mainly sold in one market. Other tobacco products are more likely to be produced by hand on a small scale, making it difficult to detect and collect taxes on them. The same applies to RYO tobacco, which can be produced on a small scale by hand or with the use of small machinery. The trade in raw tobacco and small-scale home production of tobacco often take place outside of monitoring and control systems (19). As mentioned in section 3.3.1, countries have found various solutions to address this problem, including prior approval for purchase or sale of raw materials and reg- istering, authorizing or licensing of all operators and growers that handle raw tobacco. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 153 3.5.1 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In principle, adding a product to an existing tax framework is not likely to impose significant costs. It is reasonable to expect that challenges similar to those faced in dealing with conventional tobacco products will be faced in the collection of taxes on new products, as market players will attempt to use loopholes in tax regulation to avoid or evade taxes whenever possible. However, new challenges are expected to arise when those new products involve rapidly changing technology and where their market dynamics are widely unknown. Furthermore, taxation of new tobacco products may require additional capacity, as a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. 3.5.2 HEATED TOBACCO PRODUCTS (HTPs) Many countries apply a specific excise on tobacco products according to tobacco weight (see Table 2.4 in Chapter 2). With HTPs, this is likely to impose a challenge, since assessing the content of tobacco in a heated tobacco stick will be an additional burden. From a tax administration perspective, it will be easier for authorities to apply taxes per stick or per unit, as is done for cigarettes. 3.5.3 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS) PRODUCTS Some countries tax only nicotine-containing e-liquids while others tax both nicotine- and non-nicotine-containing e-liquids. Taxing only nicotine-containing e-liquids re- quires laboratory capacity to detect the presence of nicotine (see Table 2.5 of Chapter 2). Self-declarations by industry are not sufficient, since some e-liquids labelled as nicotine-free have been found to contain nicotine (see section 2.4.2). Therefore, it is simpler to tax both nicotine- and non-nicotine-containing e-liquids. One likely challenge of taxing all e-liquids will be the capacity to detect and differentiate whether the e-liquids used in ENDS/ENNDS are falsely declared as being for other purposes at the import and manufacturing levels. More information on advantages and disadvantages of different excise tax policies is given in Table 2.6 of Chapter 2. The challenge in taxing the other components of ENDS/ENNDS products is their diversity (see section 2.4.2 of Chapter 2) and the possibility that some parts may be used for other purposes (e.g. in batteries). As indicated earlier, rapidly changing technology and the lack of control and knowledge of the market make taxation of ENDS/ENNDS devices challenging. It may be for this reason that the majority of countries that tax those products address only the e-liquids. When applying a tax on these newer products, countries should be aware that many customers buy their products online. It is therefore recommended that countries 154 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N deciding to tax these products draw up a proper implementation plan, including how taxes will be collected on imported products and online sales. Online cross- border sales of tobacco products are not permitted in all countries. Several EU countries have banned such sales, which include online sales from retail outlets to consumers in another country. Of course, such bans makes sense only if there is also capacity to enforce them. As in the case of taxing tobacco products, the following actions will be important to more effectively impose taxes on these products: 1. implementing strong enforcement mechanisms such as licensing, recordkeep- ing and control of the supply chain, which can include but is not limited to: a) imposing strict licensing of retailers, importers and manufacturers; ideally, licensing of all those involved in the supply chain and developing a tracking and tracing regime for ENDS/ENNDS products and HTPs (to share cost, this can be done in tandem with the system developed for cigarettes); b) exercising the right to set the frequency and type of audits or controls; c) exercising the right to confiscate goods; and d) imposing sanctions such as penalties, fines and/or withdrawal of licences (if applicable) if legislation is not respected. Specific to ENDS/ENNDS products: 2. implementing highly consequential sanctions for producers who declare nicotine-containing e-liquids as “non-nicotine-containing”; and 3. requiring a fee (contribution to the costs) for laboratory tests when a new product is brought on the market or when there are significant modifications to an existing one. More information on the policy options to apply excise taxes on ENDS/ENNDS is given in Chapter 2, section 2.4.2, Table 2.6. KEY TAKEAWAY 18 In principle, the administration of taxes on new and emerging nicotine products and tobacco products should be similar to that for cigarettes. Due to the lack of standardization of these products, however, a rapid and constantly evolving understanding of them and their supply chain will be required to achieve effective and efficient administration of taxes. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 155 3.6 THE BROADER ELEMENTS OF A GOOD TAX SYSTEM 3.6.1 PROPER RESOURCING OF COMPETENT AUTHORITIES In addition to legal tools and a legal basis on which to act and enforce, the authori- ties in charge of implementing excise tax laws should be provided with sufficient resources to hire the necessary staff to properly implement and enforce them. The necessary staffing could encompass multiple agencies and will often require coop- eration between agencies, since some aspects – such as regulation, licensing and border control – may be performed by agencies other than the competent authority. Staff of competent authorities need the necessary tools, equipment, training and supplies to carry out their functions. This requirement includes the means to build or purchase and maintain a software system that will allow taxpayers to submit required information electronically. Electronic filing has benefits for both taxpayers and authorities. It minimizes the compliance cost for taxpayers and can therefore support voluntary compliance (68). To identify risks of noncompliance, the software system should offer competent authorities the ability to analyse the data submitted by taxpayers and cross-check it with data from other taxes – such as VAT – and third-party sources, such as banks and household surveys. Another option for authorities is to make more efficient use of existing resources. For example, an authority could optimize the risk management system by switching to a risk-based approach: resources could be saved by auditing taxpayers who are more likely to be noncompliant based on risk analysis rather than auditing all of them. Other problems that challenge the effective functioning of a competent authority are lack of a coherent strategy and problems with professionalism related to lack of training or corruption (69). Having a strategy avoids directing resources towards less- important areas. The strategy should always be aligned with the objectives, so that com- petent authorities can identify which steps they should take and in which order they should take them to reach these objectives. A strategy is indispensable to prioritizing and organizing resources so that identified issues or risks can be addressed efficiently. 3.6.2 CORRUPTION Competent authorities should implement tax laws with integrity and have strict rules and regulations for detecting corruption. Strict rules and regulations should also be in place for the punishment of both agency personnel and taxpayers who engage in corrupt practices. Corruption within a competent authority results in the improper monitoring of tax compliance and is one of the causes of the proliferation of illicit trade in tobacco products. It also erodes confidence in competent authorities and ultimately in governments overall. In addition to effective laws and regulation, strong internal audits covering prevention, investigation and sanctions should be 156 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implemented. To improve prevention, a risk map should be created that highlights areas of misconduct and possible leakages. An action plan to update controls should be established to improve areas of weakness detected in procedures and systems. The audits should also be scheduled regularly. Prepared internal auditors with pow- ers to conduct investigations are necessary. Sanctions for corruption, including administrative sanctions and criminal prosecution, must be strong. 3.6.3 A STRONG JUDICIARY The judicial system should be honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. The appeals process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially in the context of illicit trade. KEY TAKEAWAY 19 Broader elements of a good tax system include (1) proper resourcing of competent authorities to hire staff and obtain necessary equipment and systems, (2) strict rules and regulations to detect and punish corruption among both agency personnel and taxpayers and (3) ensuring that the judicial system is honest and independent, with disputes being solved as quickly as possible. 3.7 CONCLUSIONS Policies are more effective if they are properly implemented and enforced. Com- petent authorities have a key role in the achievement of financial and public health objectives of excise taxes. Given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade, this chapter draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (i.e. the Protocol). The Protocol provides a blueprint of measures to address the problem of illicit trade and can be used as a model even by countries that are not Parties to it. Qualities of an effective and efficient tax administration include institutional ar- rangements where roles and responsibilities of competent authorities are clearly defined to avoid overlap and voids. Additionally, effective collaboration among relevant bodies must be facilitated. At the national level, within any organizational arrangement, it is vital that agencies cooperate and exchange information and that their competences find their basis in law. A legal basis for exchange or access to information between government bodies should be ensured. At the international level, especially for border control, the role of customs is key, and access to international CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 157 cooperation agreements such as the Protocol is very useful. An organizational tax administration structure must include a system of performance evaluation and accountability through pre-defined key indicators. To ensure compliance, the accuracy of information for the tax compliance cycle is key, including clear and straightforward taxpayer registration and licensing, declara- tion, recordkeeping, warehousing, distribution, collection and tax refund processes. • Licensing is a powerful tool for obtaining information and securing the supply chain of tobacco products. Ideally, all persons involved in the growing of tobacco and the retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. • Collecting as much information as possible on the business of tobacco and recording all transactions are key to reducing tax evasion, but this may be burdensome for authorities. The use of IT for periodic tax declarations, ac- counting, inventory and financial information is critical for obtaining accurate information and can help decrease the cost of the whole reporting system. • Recordkeeping should be ensured. All persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of those products. • Maintaining a system of authorization for warehousing allows the authori- ties to carry out controls in production and storage facilities to ensure that taxes are paid. Ideally, bonded warehouses should be eliminated from the supply chain. • Duty suspension – which is often applied during the producing, processing, holding, receiving and dispatching of excise goods – should be granted only if strict criteria are met (e.g. for granting authorization, warehouse pre-authoriza- tion visits, adequate stock control measures, checking the origin of excise prod- ucts and the entire production process and coding and marking of products). • To limit the number of taxpayers a competent authority has to manage, tax collection should take place close to the point of production and import. • Refunds for VAT, excise taxes and customs duties are common in most coun- tries, under the principle that taxes are not exported. The refund process must be closely monitored to avoid opportunities for tax evasion. Control and enforcement – key components of tax administration – include a number of measures to secure the supply chain: licensing and due diligence, fiscal markings, 158 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N track and trace, anti-forestalling measures, audits and controls, import and export control and attention to free zones and transhipment points. Control and enforce- ment need to be included as pillars in the strategic plan of the tax administration. Enforcement and control plans must be designed to define the activities and taxpayers that are subject to enforcement and to allocate staffing, auditing, infrastructure and IT resources. Targets must be defined, including the number of interventions and any additional collection or reduction of tax evasion. This includes choosing interventions for those who have a higher probability of noncompliance (the risk- based approach). In the tobacco supply chain, import, export and transfers to and from warehouses may be areas at greater risk of noncompliance. • Licensing provides timely and accurate data that can serve as the basis for audits, since it identifies and controls legitimate operators. The process of licensing control must be carried out and updated periodically – in particular, by controlling the validity of bonds or guarantees and the proper functioning of the required systems and recordkeeping. Where licences are required, the law should include a provision that disallows purchases from unlicensed suppliers or sales to unlicensed purchasers. This means that both suppliers and purchasers will need to verify those with whom they are doing business. This will substantially help to reduce the burden of proof for authorities. In addition, to maintain a high level of control, the validity of licences should be limited in time, making renewals or reapplication required. • Another important measure for controlling and monitoring production and import of tobacco products is the use of fiscal markings (e.g. tax stamps). In addition to increasing compliance with tax laws, fiscal markings can help distinguish between genuine and illicit tobacco products. The use of fis- cal marks enables both the competent authority and the public to monitor whether the taxes on tobacco products have been properly paid. In addition to locally produced and imported products, tobacco products for export should also be required to be marked, but with an indication that they are for export. Requiring a standard package size can facilitate the application of fiscal markings. To lower the chance that fraudsters attempt to re-use fis- cal markings (in particular, stamps) the marking should be applied to each pack of cigarettes (and other tobacco products) before the pack is wrapped with cellophane. Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Fiscal markings should include several security features to make them more difficult to counterfeit. These can include overt, covert, semi-covert and/or forensic features. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 159 • Tracking and tracing systems assist authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as in monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to provide authorities with information on all transactions throughout the entire tobacco product supply chain until duties are paid or other obligations are discharged. Any tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it becomes possible to unambiguously register that product’s movements. A good tracking and tracing system enables the government to properly monitor the supply chain, improve its ability to ensure collection of the proper duties and taxes, authenticate whether the identification marking is genuine and matches the product, improve its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. To reduce the financial burden of implementing such a system, jurisdictions could require the tobacco industry to bear the cost. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, and governments should ensure that the system is independent from the tobacco industry. While the objectives of fiscal markings and tracking and tracing systems are different, stamps increasingly contain tracking and tracing features. • Implementing legal measures to prevent forestalling can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. Forestalling, stockpiling or front-loading occur when manufacturers or im- porters increase their tax-paid stock or oversupply the market by increasing production or imports before a tax increase in order to pay the previous lower rate. • Periodic audits and controls can be implemented to increase compliance. These include cost audits, transfer price audits, price and market monitoring, consumer controls and cross-check controls. • Import and export of tobacco products and manufacturing equipment should be allowed only for duly licensed natural persons or legal entities. The risk of loss of revenue can be mitigated by requiring a guarantee or bond that will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless required fiscal markings (such as tax stamps or export labels) are affixed on the pack, according to the law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Exchange of information between jurisdictions on the movement of goods can also reduce the risk of evasion. 160 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Non-invasive detection equipment (such as X-ray scanners) can be used at customs posts to detect contraband merchandise. A cheaper alternative is the use of dogs that are trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied to reduce contraband. Such measures include the separation of processing operations from the sealed storage of taxed and untaxed products. Within a country, mobile excise control units are helpful in verifying excis- able goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. Physical control operations require close coordination between police, border guards and other public services. • Controls such as regulation and oversight are usually less strict in free zones and transhipment points. This can make free zones appealing to persons involved in illegal cigarette manufacturing or trade. Customs administrations should exercise their authority in free zones to effectively identify and fight illicit trade in tobacco products. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing tracking and tracing regimes and removing exemptions from excise taxes. Other actions include the prohibition of intermingling of tobacco products with non-tobacco products in a single container or other similar transportation unit when the products are removed from free zones. Sale of tax-free or duty-free tobacco products to international travellers should be prohibited, as these sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect govern- ment revenues by creating a loophole in the tax structure. Procedures after detection of illicit trade of tobacco products should be clearly defined. If smuggling or illicit trade is detected through audits, tracking and tracing systems, verification of declarations or border control, actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. It is also important that penalties and sanctions be sufficient to deter illegal activities. Low financial penalties may simply be paid as a cost of doing business while the illegal activity continues. The minimum penalty for consumers in possession of illicit tobacco products should be confiscation and destruction of the products found in their possession and required payment for the unpaid tax and duties on those products. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are taking advantage of this type of legislation and using it to address illicit trade of tobacco as well. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 161 In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. The challenge for taxation of other products includes the lack of standardization of those products and sometimes large informal markets. Knowledge of the product and the supply chain greatly help to facilitate effective tax administration. The trade in raw tobacco and small-scale home production of RYO and other products such as bidis often takes place outside of monitoring and control systems. The best way to address this challenge is to enforce prior approval for purchase or sale of raw materials and a requirement to register, obtain an authorization or license all operators and growers handling raw tobacco. In principle, adding new and emerging nicotine and tobacco products to an exist- ing tax framework is not expected to impose significant costs. It is reasonable to expect that similar challenges will be faced in the collection of taxes on these newer products, as market players will attempt to use the current loopholes in tax regulation to avoid or evade taxes on these products whenever possible. However, challenges are expected to arise, as newer products involve rapidly changing technology, and their market dynamics are widely unknown. Furthermore, additional capacity may be required, since a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. Because the newer nicotine and tobacco products are widely purchased online, countries deciding to tax these products should draw up a proper implementation plan that includes rules on how taxes will be collected on imported products and online sales. Online cross-border sales are not permitted in some countries. The elements of a good tax system include (1) proper resourcing of competent authorities sufficient for hiring the necessary staff to properly implement and en- force excise tax laws; (2) implementation of tax laws with integrity and with strict rules and regulations to detect corruption and for the punishment of both agency personnel and taxpayers who are engaged in corrupt practices; and (3) ensuring that the judicial system is honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. 162 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N REFERENCES 1. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 2. WHO FCTC guidelines for implementation of Article 6 Price and tax measures to reduce the demand for tobacco. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/treaty_instruments/ Guidelines_Article_6_English.pdf?ua=1, accessed 5 October 2020). 3. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 4. Joossens L, Raw M. From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–234 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 5 October 2020). 5. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 6. Reyes IP. Measures to control the tobacco supply chain in the ASEAN: a review of excise tax stamps, track and trace mechanisms, and licensing systems in the ASEAN region. Bangkok: Southeast Asia Tobacco Control Alliance; 2014 (https://seatca.org/dmdocuments/SEATCA%20MEASURE%20BOOK%20. pdf, accessed 5 October 2020). 7. Italy’s tax administration: a review of institutional and governance aspects. Paris: Organisation for Economic Co-operation and Development; 2016 (https://www.oecd.org/tax/administration/italy- tax-administration-a-review-of-institutional-and-governance-aspects.pdf, accessed 5 October 2020). 8. Crandall W, Gavin E, Masters A. ISORA 2016: understanding revenue administration. Washington (DC): International Monetary Fund; 2019 (Departmental Paper Series No. 19/05; (https://www.imf.org/en/ Publications/Departmental-Papers-Policy-Papers/Issues/2019/03/07/ISORA-2016-Understanding- Revenue-Administration-46337, accessed 5 October 2020). 9. Dutch customs in 2018. Heerlen: Customs Administration of the Netherlands; 2019 (https://over-ons. belastingdienst.nl/wp-content/uploads/2019/04/Dutch-Customs-in-2018.pdf, accessed 8 October 2020). 10. ESAs transform the way competent authorities cooperate with each other on AML/CFT matters [press statement]. European Banking Authority; 16 December 2019 (https://eba.europa.eu/esas-transform- way-competent-authorities-cooperate-each-other-amlcft-matters, accessed 5 October 2020). 11. Illicit trade report 2013. Brussels: World Customs Organization; 2014 (http://www.wcoomd.org/-/ media/wco/public/global/pdf/topics/enforcement-and-compliance/activities-and-programmes/ illicit-trade-report/itr_2013_en.pdf?la=en, accessed 5 October 2020). 12. Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation and repealing Directive 77/799/EEC. Brussels: OJEU; 2011. (L 64/1; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:32011L0016&from=en, accessed 29 January 2021). 13. 439 stolen cars seized in 12 days as part of Frontex-led operation. Frontex News Release; 14 October 2019 (https://frontex.europa.eu/media-centre/news-release/439-stolen-cars-seized-in-12-days-as- part-of-frontex-led-operation-vvvirI, accessed 5 October 2020). 14. Focus on: the illicit trafficking of counterfeit goods and transnational organized crime. Vienna: United Nations Office on Drugs and Crime; 2014 (https://www.unodc.org/documents/counterfeit/FocusSheet/ Counterfeit_focussheet_EN_HIRES.pdf, accessed 5 October 2020). 15. Crandall W. Revenue administration: performance measurement in tax administration. Washington (DC): International Monetary Fund; 2010 (Technical Notes and Manuals 10/11; https://www.imf.org/ external/pubs/ft/tnm/2010/tnm1011.pdf, accessed 7 October 2020). 16. Measuring tax gaps 2019 edition – tax gap estimates for 2017-18. London: HM Revenue and Customs; 2019 (https://webarchive.nationalarchives.gov.uk/20200701215139/https://www.gov.uk/government/ statistics/measuring-tax-gaps, accessed 29 January 2021). 17. Countering illicit trade in tobacco products: a guide for policy-makers. Lyon: INTERPOL; 2014 (Legal handbook series; https://www.aph.gov.au/DocumentStore.ashx?id=704d9168-2aa7-4df4-bb81- CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 163 0fe1f357dfa6&subId=408832, accessed 29 January 2021). 18. Resolução- RDC No. 226 de 30 de Abril de 2018 [Resolution of April 30, 2018]. Ministério da Saúde/ Agência Nacional de Vigilância Sanitária/Diretoria Colegiada; 2018 (in Portuguese) (https://www.in.gov. br/materia/-/asset_publisher/Kujrw0TZC2Mb/content/id/12527411/do1-2018-05-02-resolucao-rdc- n-226-de-30-de-abril-de-2018-12527407, accessed 17 February 2021). 19. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 20. Preece R. Addressing the illicit flow of tobacco products in Australia. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 21. Excise on tobacco [website]. Canberra: Australian taxation office; 2019 (https://www.ato.gov.au/ business/excise-on-tobacco/, accessed 5 February 2021). 22. Safe Framework of Standards. Brussels: World Customs Organization; 2018 (http://www.wcoomd. org/-/media/wco/public/global/pdf/topics/facilitation/instruments-and-tools/tools/safe-package/ safe-framework-of-standards.pdf?la=en, accessed 29 January 2021). 23. The Trade Facilitation Agreement. Geneva: World Trade Organization; 2017 (https://www.wto.org/ english/docs_e/legal_e/tfa-nov14_e.htm, accessed 29 January 2021). 24. Authorised economic operator (AEO) [website]. Kildare: Irish Tax and Customs; 2020 (https://www. revenue.ie/en/customs-traders-and-agents/authorised-economic-operators/index.aspx, accessed 7 October 2020). 25. Performance outcome area 1: integrity of the registered taxpayer base. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www. tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 26. Commission implementing regulation (EU) 2016/323 of 24 February 2016 laying down detailed rules on cooperation and exchange of information between member states regarding goods under excise duty suspension pursuant to Council Regulation (EU) No 389/2012. Brussels: OJEU; 2018 (http://data. europa.eu/eli/reg_impl/2016/323/2018-02-15, accessed 7 October 2020). 27. Council Directive 2008/118/EC concerning the general arrangement for excise duty repealing Directive 92/12/EEC. Brussels: OJEU; 2009 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:3 2008L0118&from=en, accessed 29 January 2021). 28. Our strategy. London: HM Revenue & Customs, 2017 (https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/629941/HMRC-Strategy.pdf, accessed 31 January 2021). 29. Strategic plan FY2018-2019. Washington (DC): Internal Revenue Service, 2017 (https://www.irs.gov/ pub/irs-pdf/p3744.pdf, accessed 31 January 2021). 30. Performance outcome area 2: effective risk management. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/ assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 31. Tax administration 2017: comparative information on OECD and other advanced and emerging economies. Paris: Organisation for Economic Co-Operation and Development; 2017. 32. Compliance risk management: managing and improving tax compliance. Paris: Organisation for Economic Co-Operation and Development; 2017 (https://www.oecd.org/tax/administration/33818656. pdf, accessed 7 October 2020). 33. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 34. Evans P. U.K. Fines British American Tobacco for Oversupplying Cigarettes in Belgium. The Wall Street Journal. 13 November 2014 (https://www.wsj.com/articles/u-k-fines-british-american-tobacco-for- oversupplying-cigarettes-in-belgium-1415873470, accessed 31 January 2021). 35. Doward J. BAT fined for oversupplying tobacco in low-tax European jurisdictions. The Guardian. 164 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 16 November 2014 (https://www.theguardian.com/business/2014/nov/16/bat-fined-for-oversupplying- tobacco-in-low-tax-european-jurisdictions, accessed 31 January 2021). 36. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021) 37. Cour des comptes. Accises sur les produits du tabac, Bruxelles [Court of Auditors, Excise duties on tobacco products, Brussels]. Bruxelles: Cour des comptes; 2015 (in French) (https://www.ccrek.be/ docs/2015_27_AccisesProduitsDuTabac.pdf, accessed 7 October 2020). 38. Chaloupka FJ, Edwards SM, Ross H, Diaz M, Kurti M, Xu X, et al. Preventing and reducing illicit tobacco trade in the United States. Atlanta: Centers for Disease Control and Prevention; 2015 (https://www. cdc.gov/tobacco/stateandcommunity/pdfs/illicit-trade-report-121815-508tagged.pdf, accessed 29 January 2021). 39. The Tax Stamp Forum 2015. Hexham: Reconnaissance International (https://10times.com/tax-stamp- forum-miami, accessed 10 September 2020). 40. Tax stamps: a technical study and market report. London: Reconnaissance international; 2012. 41. ЕАНС/2013/Health/ll concerning the provision of an analysis and feasibility assessment regarding EU systems for tracking and tracing of tobacco products and for security features. Brussels: European Commission, 2015 (https://ec.europa.eu/health/sites/health/files/tobacco/docs/2015_tpd_tracking_ tracing_frep_en.pdf, accessed 29 January 2021). 42. Ross H. Controlling illicit tobacco trade: international experience. Economic research informing tobacco control policy. Cape Town: University of Cape Town; 2015 (https://tobacconomics.org/ uploads/misc/2015/05/Ross_International_experience_05.28.15.pdf, accessed 18 December 2020). 43. Security and resilience — authenticity, integrity and trust for products and documents — guidelines for the content, security, issuance and examination of excise tax stamps. Geneva: International Organization for Standardization; 2018. 44. Borkowski F, Twomey C. European Union: an update on EU policies. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 45. Quality management systems — Fundamentals and vocabulary. ISO 9001:2015(en). Geneva: International Organization for Standardization; 2015 (https://www.iso.org/obp/ui/#iso:std:iso:9001:ed- 5:v1:en, accessed 16 February 2021). 46. Protocol to Eliminate Illicit Trade in Tobacco Products: questions & answers. Geneva: World Health Organization; 2019 (https://www.who.int/fctc/protocol/faq/en/, accessed 7 October 2020). 47. Ross H, Eads M, Yates M. Why governments cannot afford Codentify to support their tracking and tracing solutions. Tob Control. 2018;27(6):706–08. 48. Ser vicio de Impuestos Internos, Chi le, Noticias, 28 May 2019. (http://www.si i .c l/ noticias/2019/280519noti01er.htm, accessed 20 October 2020). 49. Directive 2014/40/EU of the European Parliament and of the Council. OJEU. 2014;127:1–38 (https:// ec.europa.eu/health/sites/health/files/tobacco/docs/dir_201440_en.pdf, accessed 7 October 2020). 50. Report on the EU customs enforcement of intellectual property rights: results at the EU border. Luxembourg: Publications Office of the European Union, 2019 (https://ec.europa.eu/taxation_customs/ sites/taxation/files/2019-ipr-report.pdf, accessed 7 October 2020). 51. Matthews P. Forestalling ahead of property tax changes. London: Office for Budget Responsibility; 2016 (OBR Working Paper No. 10; https://obr.uk/docs/dlm_uploads/Working-paper-No.10-1.pdf, accessed 29 January 2021). 52. Ross H, Tesche J, Vellios N. Undermining government tax policies: common strategies employed by the tobacco industry in response to tobacco tax increases. Prev Med. 2017;105S:S19–S22 (https://www. sciencedirect.com/science/article/pii/S0091743517302165?via%3Dihub, accessed 17 February 2021). 53. European Commission v Portugal supported by Belgium, Estonia and Poland, Case C-126/15. Judgement of the Court (Fifth Chamber); 29 June 2017 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?u ri=CELEX:62015CJ0126&from=EN, accessed 31 January 2021). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 165 54. Amalia Valesko v Zollamt Klagenfurt, Case C-140/05. Judgement of the Court (Second Chamber); 5 October 2006 (http://curia.europa.eu/juris/showPdf.jsf;jsessionid=02DEE17CC11845BAE0F42FC7 CFD24C85?text=&docid=65599&pageIndex=0&doclang=en&mode=lst&dir=&occ=first&part=1&c id=2591459, accessed 29 January 2021). 55. European Commission v French Republic, Case C-197/08. Judgement of the Court (Third Chamber); 4 March 2010 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:62008CJ0197&qid =1612223322830&from=EN, accessed 31 January 2021). 56. Commission requests that Belgium amend its legislation on tobacco: letter of formal notice. Brussels: European Commission; 8 November 2018. (Memo 18/6247; https://ec.europa.eu/commission/ presscorner/detail/EN/MEMO_18_6247, accessed 29 January 2021). 57. Commission requests that Belgium amend its legislation on tobacco, MEMO/18/6247 of 8 November 2018. 58. Pederson H, Floristean A, Iseppi L, Dawkins R, Smith C, Mørup C, et al. Study on the measuring and reducing of administrative cost for economic operators and tax authorities and obtaining in parallel a higher level of compliance and security in imposing excise duties on tobacco products. Brussels: European Commission; 2014 (https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/ ramboll-tobacco-study.pdf, accessed 3 October 2020). 59. Money laundering vulnerabilities of free trade zones. Paris: Financial Action Task Force; 2010. https:// www.fatf-gafi.org/media/fatf/documents/reports/ML%20vulnerabilities%20of%20Free%20Trade%20 Zones.pdf, accessed 7 October 2020). 60. Omi K. ‘Extraterritoriality’ of free zones: the necessity for enhanced customs involvement. World Customs Organization; 2019 (WCO research paper no. 47; http://www.wcoomd.org/-/media/wco/public/global/ pdf/topics/facilitation/ressources/permanent-technical-committee/225-226/itemixb_fzs_wco_e. pdf?la=en, accessed 29 January 2021). 61. European Parliament resolution of 26 March 2019 on financial crimes, tax evasion and tax avoidance. European Parliament; 2019 (2018/2121(INI)); http://www.europarl.europa.eu/doceo/document/TA- 8-2019-0240_EN.html, accessed 7 October 2020). 62. International Convention on the Simplification and Harmonization of Customs Procedures. Brussels: World Customs Organization; 2008 (http://www.wcoomd.org/Topics/Facilitation/Instrument%20 and%20Tools/Conventions/pf_revised_kyoto_conv/Kyoto_New, accessed 7 October 2020). 63. Evidence Notice 2004: tobacco duty – tobacco products manufacturing machine licensing scheme. London: HM Revenue & Customs, 2018 (https://www.gov.uk/government/publications/excise-notice- 2004-tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme/excise-notice-2004- tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme, accessed 31 January 2021). 64. Tackling illicit tobacco: from leaf to light: the HM Revenue & Customs and Border Force strategy to tackle tobacco smuggling. London: HM Revenue and Customs; 2015 (https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/418732/Tackling_illicit_ tobacco_-_From_leaf_to_light__2015_.pdf, accessed 7 October 2020). 65. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 66. Cardenas M. Case studies in illicit tobacco trade: Colombia. Chicago: University of Illinois at Chicago; 2020 (Tobacconomics Fact Sheet; https://tobacconomics.org/files/research/606/UIC_Colombia-Illicit- Trade-Fact-Sheet_v1.4.pdf, accessed 29 January 2021). 67. Maldonado N, Llorente BA, Escobar D, Iglesias RM. Smoke signals: monitoring illicit cigarettes and smoking behaviour in Colombia to support tobacco taxes. Tob Control. 2019;29:s243-s248 ( https:// tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/s243.full.pdf, accessed 29 January 2021). 68. Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 69. Pellechio A, Tanzi V. The reform of tax administration. Washington (DC): International Monetary Fund; 1995 (Working Paper no. 95/22; https://www.elibrary.imf.org/view/IMF001/07242- 9781451843941/07242-9781451843941/07242-9781451843941_A001.xml?language=en&redirect=true, accessed 29 January 2021). 166 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.1 COMPOSITION OF TOBACCO PRODUCTS To implement and enforce tobacco taxes in the most efficient way, competent authori- ties should be familiar with all of the components of tobacco products, including each of the raw materials used in their manufacture, production inputs and tobacco manufacturing machinery. Knowledge of the components of excisable products and machinery provides valuable information to identify activities at high risk for non- compliance, implement measures to ensure all taxes are paid and prevent illicit trade. Nicotine, non-nicotine and tobacco products and their component parts In most countries, the ministry of finance determines tax policy, including which tobacco products are taxed, while the ministry of health is responsible for product and use regulation. This could lead to different definitions of the same product, depending on which ministry is responsible for a given law or regulation. Wherever possible, a clear and common definition should be developed to simplify procedures and avoid confusion. Tobacco products take various forms, and not all may be regulated or subject to excise tax in a specific jurisdiction. In addition to cigarettes, other traditional tobacco products include smokeless tobacco – such as chewing tobacco, snuff and snus – as well as bidis and kreteks (clove cigarettes), which can be hand-rolled or manufactured, pipes, hookah or waterpipe and cigars. Cigarettes It is important to understand the materials and component parts of the tobacco products most commonly used in a particular country. Cigarettes are the most common and significant tobacco products in terms of volume and tax revenues in most jurisdictions. A cigarette stick is composed of: • the tobacco blend of various types of tobacco plant (leaves and stem and other plant parts) and additives (including flavours); • the cigarette paper used to wrap the tobacco blend to make up the tobacco rod; • the acetate filter that forms the white portion at the tip of a filtered cigarette, which is in direct contact with the smoker’s mouth; • the tipping paper or wraps around the filter; and • the adhesive that secures the cigarette paper around the tobacco blend and the tipping (1). Each manufacturer follows a specific process to produce cigarettes. Aside from the tobacco blend, manufacturers also vary the size of cigarette paper and tipping paper and the length of acetate filter used per stick (1). In some countries, these elements are standardized. In an ideal regulatory framework, a manufacturer would CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 167 be required to submit information on the specific process for each brand and variant of the tobacco product that it manufactures to the competent authorities as part of the licensing requirement (see section 3.4.3). Authorities could, for example, require manufacturers to submit this information in order to obtain a licence. The minimum requirements of the administration and the information that should be included could be laid down in law or lower regulation to ensure that authorities have the information in their possession for all licensed manufacturers. This information contributes to verifying whether a company is reporting the actual quantity of cigarettes manufactured for sale and sold cigarettes by comparing the amount of materials used for production and the quantities used per cigarette with the total number of manufactured cigarettes. Eventually, this information also contributes to validating whether the taxes are properly paid. Figure A3.1 shows the component parts of a typical machine-made traditional cigarette. Fig. A3.1 Component parts of a machine-made cigarette Source: Author’s compilation. Photo by Walter Klerx. Not all parts of tobacco products are subject to the same level of control. According to Article 6.5 of the Protocol, five years following the entry into force of this Protocol, the MOP shall ensure at its next session that evidence-based research is conducted to ascertain whether any key inputs exist that are essential to the manufacture of tobacco products, are identifiable and can be subject to an effective control mechanism. On the basis of such research, the MOP shall consider appropriate action. In addition to the component parts of tobacco products, materials needed for packaging a specific number of sticks into a pack of cigarettes, usually 20 per pack, can be monitored. These materials include the foil paper, the package paper (which could bear the brand name, design and health warnings), the fiscal marking (if required) and the plastic or cellophane wrap. A fixed number of packs of cigarettes, Filter Tipping paper Cigarette paper Tobacco Tobacco rodFiltration zone 168 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N normally 10 packs, are packed into cartons, also called reams. These cartons are usu- ally made of soft paperboard or cardboard, possibly with branding, and are wrapped in plastic wrap or cellophane. Fifty cartons are packed in master cases, which are made of sturdier and thicker paperboard and stacked on pallets (usually 50 master cases to a pallet). An effective regulatory framework would require manufacturers and importers to provide information to the competent authorities on packaging and design, as well as the number of sticks per pack, carton and master case. Prior approval for purchase or sale of materials used in the cigarette production process can also be required. In the Philippines, suppliers of such raw materials, including those providing tobacco papers and filter components, are required to have a licence (2). In some of the Member States of the EU, raw tobacco is also subject to fiscal and legal requirements. For example, in Slovakia and Poland, raw tobacco can be handled only by authorized operators. While authorized operators do not have to pay excise duties on raw tobacco, if raw tobacco is detected by an unauthorized operator, excise duties will be due. Hungary, Italy and the United Kingdom require registration or authorization for all operators and growers handling raw tobacco (3). In addition to knowing the quantities of inputs required to produce a specific amount of a regulated product (e.g. cigarettes), the competent authority also needs to understand the supply, manufacturing and distribution chains to be able to properly monitor, regulate and determine whether taxes have been paid (see also Fig. 3.3). Novel and emerging nicotine, non-nicotine and tobacco products In recent times, new products have been introduced to several markets, namely, ENDS, ENNDS and HTPs. ENDS usually comprise a nicotine-containing e-liquid but do not contain to- bacco. ENNDS are essentially the same but do not (ostensibly) contain nicotine. The WHO COP requested the Convention Secretariat to invite Parties to monitor and report on scientific, regulatory and market developments such as initiation, cessation, advertising and promotion of ENDS and ENNDS. Furthermore, the COP requested WHO to report on the development of methods by regional and international standards-development organizations for the testing and measuring of contents and emissions of these products (4). There are different types of e-cigarettes – the most common type of ENDS and ENNDS – and currently there are four generations of products. However, they can be divided into two broad categories: open systems and closed systems. Both types of e-cigarette use a wick and a heat source to generate an aerosol. The wick is saturated with e-liquid, and a microprocessor is used to control operations (not all include this). Some e-cigarettes also have an LED light to imitate the burning end of a conventional cigarette (5). Fig. A3.2 presents examples of open and closed systems. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 169 E-liquid pod cover Rechargeable battery Heating element (heats solution, aerosolizing nicotine) Mouthpiece E-liquid pod Fig. A3.2 Examples of open and closed systems of ENDS/ENNDs products Open ENDS/ENNDS system (e-cigarette) Closed ENDS system (e-cigarette) Source: (6). Unlike ENDS/ENNDS, HTPs do contain tobacco. HTPs produce aerosols containing nicotine and toxic chemicals when tobacco is heated or when a device containing tobacco is activated (7). HTPs are composed of two elements: the sticks or pods that contain the tobacco and the device used to heat the tobacco. Both are necessary for the product to be used. Fig. A3.3 shows an example of a heated tobacco product. HTPs are tobacco products and are therefore subject to the regulatory measures contained in the WHO FCTC. Rechargeable battery Power button (to start vaping) Mouthpiece Atomizer / Heating element (heats solution, aerolizing nicotine) E-liquid tank (refillable e-liquid nicotine tank) 170 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. A3.3 Components of an HTP Holder: Tobacco stick: Note: PLA: polyactic acid, MPF: mouthpiece filter. Sources: (8-9). More information on tax administration of other tobacco products is presented in section 3.5. Tobacco stick Holder Charger Casing Control Electronics Battery Heating Blade Heatstick total lenght: 45mm MPF (7 mm) PLA (18 mm) Tobacco plug (12 mm) Hollow acetate tube (8 mm) Outer paper Diameter max 7.42 mm Tipping paper CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 171 REFERENCES 1. Memorandum by Gallaher Group Plc: the tobacco industry and the health risks of smoking. Select Committee on Health Minutes of Evidence. London: UK Parliament, 2000 (https://publications. parliament.uk/pa/cm199900/cmselect/cmhealth/27/0011323.htm, accessed 7 October 2020). 2. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 3. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco.. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 4. Electronic nicotine delivery systems and electronic non-nicotine delivery systems. Geneva: World Health Organization; 2016 (Decision FCTC/COP7(9) of the Conference of the Parties to the WHO Framework Convention on Tobacco Control; https://www.who.int/fctc/cop/cop7/FCTC_COP7_9_EN.pdf?ua=1, accessed 7 October 2020). 5. Brown CJ, Cheng JM. Electronic cigarettes: product characterization and design considerations. Tob Control. 2014;23:ii4-ii10 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3995271/pdf/ tobaccocontrol-2013-051476.pdf, accessed 31 January 2021). 6. E-cigarettes and vapor products [webpage]. King County, Washington (USA); 2019 (https://www. kingcounty.gov/depts/health/tobacco/data/e-cigarettes.aspx, accessed 9 October 2020). 7. Heated tobacco products. Geneva: World Health Organization; 2020 (WHO/HEP/HPR/2020.2 Information sheet; https://apps.who.int/iris/bitstream/handle/10665/331297/WHO-HEP-HPR-2020.2- eng.pdf?sequence=1&isAllowed=y, accessed 7 October 2020). 8. Tobacco heating system (IQOS) briefing document. Silver Spring: US Federal Drug Administration; 2018 (https://www.fda.gov/media/110377/download, accessed 31 January 2021). 9. Premarket tobacco product application: technical project lead review. Silver Spring: US Federal Drug Administration; 2017 (https://www.fda.gov/media/124247/download, accessed 7 October 2020). 172 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.2 EXAMPLE OF FORESTALLING AND COUNTERMEASURES This example is hypothetical but inspired from the situation in the Philippines. The amounts and prices have been altered, however, and it is assumed that the normal inventory kept by a tobacco manufacturer is two months.  The excise tax imposed on a pack of cigarettes in the current year is US$ 3.00; it will be increased to US$ 3.30 at the beginning of the new fiscal year (January in this example). The monthly production of Brand Y cigarettes of company X, which it declares for tax purposes, is as follows: MONTH OF THE CURRENT YEAR PACKS OF CIGARETTES January 10 000 000 February 10 500 000 March 9 900 000 April 11 000 000 May 10 200 000 June 10 600 000 July 9 700 000 August 10 100 000 September 9 900 000 October 10 100 000 November 20 000 000 December 25 000 000 Since the normal inventory is two months, the quantity in the two months prior to the implementation of the new excise tax rate is disregarded. The shelf life of tobacco products is approximately six months. The average of the six months prior to November is computed to obtain the quantity presumed to be produced or imported if there was no tax increase. The quantity from May to October (inclusive) divided by 6 is 10 100 000 packs. Thus, any quantity produced beyond 10 100 000 packs for the months of November and December (the months prior to the implementation of the new tax rate) is assessed using the new tax rate. In this example, 10 100 000 of the packs produced in November will be taxed at the old rate of $3.00, and 9 900 000 packs will be taxed at the new rate of $3.30. For December, 10 100 000 packs will be taxed at $3.00, while 14 900 000 packs will be taxed at $3.30. Without imposing these measures, the government would have been deprived of the excise tax increase on 24 800 000 packs. In addition, the effect of the increase on prices and consumers would have been delayed by approximately two months. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 173 In countries using tax stamps, the withholding of the issuance of stamps is a well-known approach to counter forestalling. Another practical solution is to allow the competent authorities to request advances from the industry to cover revenue shortfalls, provided there is a legal basis for such requests. 174 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 175 CHAPTER 4. Political economy As with any proposed government action, policy-makers need to navigate the political environment of tobacco taxation at every stage of policy development, implementa- tion and administration. While every country’s distinct history, culture, systems and structural forces shape its unique political landscape, there are some universal themes when it comes to tobacco control and particularly tobacco taxation. These themes boil down to the distribution of money, power and resources. The tobacco industry, as both a political and economic player, understands these themes well. The industry has been effective in using principles of the political economy of tobacco taxation in its efforts to block important advancements in tobacco control. Nevertheless, the savvy policy-maker can see through the industry arguments by considering who benefits from industry-favoured policy measures and interven- tions. The industry’s challenges to tobacco tax policies can be organized into the five categories of SCARE tactics. This chapter provides a road map to help policy-makers navigate the political economy of tobacco taxation through each of these themes. The first five sections dissect the tobacco industry framing of each issue, pinpoint- ing the flaws in each argument, identifying the extent to which each concern has merit and suggesting how a responsible government can address each one. These discussions are supported by unbiased evidence from independent, peer-reviewed research, as well as specific examples from country experiences. Sections 4.1 through 4.5 on SCARE tactics will equip policy-makers with the tools they need to proceed with confidence that their tobacco tax policy – developed and implemented fol- lowing the guidelines spelled out in this technical manual – will bring about the greatest health and economic benefits for their constituents, regardless of industry attempts to thwart them. Section 4.6 further buoys policy-makers’ efforts to ensure the beneficial impacts of their policies, as it describes how earmarking can improve the political economy of tobacco taxation by funding programmes and initiatives that promote and support the health and well-being of the population. 176 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.1 SCARE TACTIC S: SMUGGLING AND ILLICIT TRADE 4.1.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases will inevitably result in an increase in the illicit trade in tobacco products (1–2). They claim that higher tax rates and higher prices strengthen the financial incentives for criminal enterprises to supply cigarettes from lower-tax jurisdictions, boost domestic tax evasion and encourage smokers to seek cheaper illegal cigarettes. The industry also challenges the argument that tobacco tax hikes increase government revenue by claiming that the presence of an illicit tobacco market will actually reduce revenue collection following a tax increase. More recent versions of this argument – adapted to address public health concerns about tobacco use – claim that illicit market growth also offsets reductions in smoking prevalence that would otherwise be brought about by tobacco tax increases. In summary, the tobacco industry and its allies claim that raising tobacco taxes are ineffective – and even counterproductive – because they are circumvented by illicit markets, which prevents the government from achieving its public health objective and reduces rather than increases tax revenues. When a country considers a proposal to increase tobacco excises, the tobacco industry and its allies frequently make exaggerated claims about the size and scope of illicit tobacco trade in that country. Opponents of tobacco tax increases argue that price differentials are the exclusive – or at least the dominant – cause of illicit trade. Influenced by this fear-inducing faulty diagnosis, tax authorities frequently find it difficult to make decisions about tobacco taxes. However, the industry diagnosis always contains the same erroneous elements. First, the illicit trade in a country is frequently less than the industry portrays it to be, and the country’s tax enforcement policy towards tobacco products is rarely unique or in any way different from the norm in the country (3). Second, the scale of illicit trade in tobacco is not exclusively or even primarily determined by tax or price differentials. Typically, it results from a set of governance problems characterized by government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, lack of dissuasive sanctions and/or weak norms regarding participation in illegal and informal markets (4–7). This section provides guidance for tax and other relevant authorities on how to respond to the tobacco industry SCARE tactic that increasing tobacco taxes will lead to smuggling and illicit trade in their countries. Tax authorities need to know the nature, causes and extent of illicit trade so that they can define the problem properly and formulate an appropriate response. This chapter addresses the available tools for better defining and understanding specific illicit trade problems, particularly tools that facilitate independent assessment of the magnitude of that trade. Improvements CHAP T ER 4. PO LI T I C AL ECO N OMY 177 to governance within the realm of tax authorities – such as best practices in tobacco tax administration and policies to improve the effectiveness of fiscal regulations or norms regarding participation in informal and illegal markets – are discussed in Chapter 3. This section first describes the nature of the illicit tobacco trade to highlight some of its complexities and identify complementary policies for tackling the problem. Next, evidence that calls into question the link between illicit trade and high prices or tax rate changes is discussed. Finally, to help tax authorities assess their own situation, several different methodologies are presented to estimate the scope of the illicit tobacco trade and to evaluate estimates of that trade for a particular country or tax jurisdiction. 4.1.2 THE NATURE AND EXTENT OF THE ILLICIT TOBACCO TRADE The WHO FCTC defines illicit trade as: any practice or conduct prohibited by law and which relates to production, ship- ment, receipt, possession, distribution, sale or purchase, including any practice or conduct intended to facilitate such activity (8). Non-duty-paid tobacco products found in a jurisdiction (i.e. through littered-pack surveys) could be the result of either of two related but distinct activities: tax evasion and tax avoidance. Tax evasion is a set of unlawful actions seeking the non-payment of tobacco taxes and duties, whereas tax avoidance comprises legal actions with the purpose of avoiding payment of some or all taxes, such as bringing an amount of cigarettes up to the legal allowance from a lower- into a higher-tax jurisdiction. Tax avoidance is not illegal and is therefore not considered part of illicit trade in tobacco products.1 The focus of this section is on tax evasion activities, which can occur in the movement across borders or in domestic production and distribution. When tax evasion happens across borders, it is known as smuggling (9) and can be done on a large scale or a small scale. Tax evasion in the domestic market can be partial, when licensed and authorized producers or distributors comply with only part of their tax obligations, or total, when the whole production and distribution system is illegal and out of sight of tax administrators (5, 10–14).2 Large-scale tax evasion schemes can be run by different types of producers and their associated distributors, such as the transnational tobacco companies (TTCs) and their national subsidiaries, 1 Tax avoidance practices –common among states in the United States and countries in the EU – are not analysed in this section. 2 Tax evasion is normally considered as illicit manufacturing in the literature (14). 178 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N other local tobacco companies producing their own brands and illegal factories that normally counterfeit other brands or produce illicit brands. Large-scale smuggling involves, for example, taking advantage of tax-free zones and mislabelling shipping products prior to or during transit (11) or using so- phisticated clandestine networks. This form of tax evasion is systemic and can be carried out by TTCs (12), by local companies producing in countries with low tax enforcement, e.g. Paraguay (15) that feed neighbouring countries and regional illicit hubs through a network of clandestine distributors (15) or by companies located in tax-free zones like Jebel Ali and Dubai in the UAE (16) or such zones in Russia or Cyprus (17). The origins, routes and quantities of large-scale smuggling frequently change as affected countries or markets react by strengthening enforcement and seeking bilateral cooperation with the jurisdictions of origin. For example, the 2013 comprehensive strategy of the EU (10) enhanced bilateral cooperation with major source and transit countries of illicit cigarettes coming to Europe, including Russia, Belarus and Ukraine. These agreements improved day-to-day cross-border coopera- tion, reduced illegal flows and introduced gradual tobacco excise rate adjustments in those three countries to bring them to European levels. Small-scale smuggling (also known as ant smuggling or bootlegging) is the cross-border trafficking of cigarettes in quantities that are larger than the allowable limits (e.g. two cartons) but smaller than large shipments (e.g. truckloads, cargo containers), normally for the purpose of selling at a profit (11). This type of illicit trade may exist in places where there are opportunities within neighbouring tax jurisdictions. For example, small-scale smuggling is commonly done by individuals living in French and German provinces near lower-taxed countries (e.g. Belgium, Luxembourg, Switzerland, Spain, Poland and Czechia) (18). Counterfeiting is a form of illicit manufacturing that involves the production of tobacco products (including packaging and tobacco filler) without the approval of the trademark holder (13). Another product of illicit manufacturing is so-called cheap or illicit whites. Cheap whites are branded (e.g. Jin Ling) or unbranded cigarettes that are legally or illegally produced3 and knowingly sold in the illicit market (17). Cheap whites are not usually produced by TTCs (17, 19).4 They are produced by small tobacco produc- tion companies in one country and often sold in illegal markets of neighbouring 3 Ross et al. (17) analysed this issue and found that the sale to the first purchaser is usually legal. Their analysis covers the production in free zones (i.e. in the UAE, Russia and Cyprus) and production exported from Viet Nam, Indonesia and China. In those cases, there is no need to make the first sale illegally. However, cheap white production in Paraguay is sold to domestic distributors, and most of those sales are completely illegal. 4 Ross et al. (17) and Gilmore et al. (19) identify some cheap white brands sold by TTCs, such as President (PMI), produced in Ukraine, and Esse (Korea Tobacco & Ginseng Company, KT&G), produced in Indonesia. CHAP T ER 4. PO LI T I C AL ECO N OMY 179 countries. For example, in Paraguay, cheap whites are produced on a large scale by a few companies under the guise that they are marketed domestically, but a large share is smuggled into Uruguay and Brazil (4). Iglesias et al. (20) showed how TTCs’ cheap brands were illicitly shipped through Paraguay to be sold in the Brazilian and Argentine markets in the 1990s. This contributed to increased production of cheap whites in Paraguayan firms, which continued the illicit business even after Brazilian legislation obstructed the illegal activity of the TTCs. Domestic tax evasion is a pervasive phenomenon, particularly in LMICs. Partial tax evasion in tobacco products can be found at any level of tax rates or prices and is generally the result of defective legislation or weak tax enforcement.5 Complete or total tax evasion occurs when producers and distributors are clandestine or when there are serious institutional challenges to tax enforcement between two tax jurisdictions, such as between the United States and Native American Reservations. Evidence of illicit manufacturing has increased in recent years in several places in the world, including the EU (10) and Brazil (21). TTCs were predominant in illicit trade activity until the end of the 20th century, and even with the entrance of new actors into the illicit business, TTCs have not entirely exited. Gilmore et al. (22) analysed industry-funded data and seizure data and concluded that TTCs are still involved in illicit trade in Europe, despite the Anti-Contraband and Anti-Counterfeit Agreements (the “Agreements”) signed between the four TTCs and the EU (23).6 Using industry-funded data, Gilmore et al. show that 58% of illicit EU cigarettes can be attributed to the four main TTCs. When seizure data are used, 69% to 73% of illicit EU cigarettes can be attributed to these firms (22). It is always difficult to assess the extent of the global illicit tobacco trade because of its illegality, its global and changing nature and problems with data collection (24). Before the 21st century, when TTCs were almost unique actors in the large-scale smuggling of well-known cigarette brands, the difference between global exports and imports of cigarettes could provide a good approximation of the size of this problem globally (7). However, with the growth of illicit manufacturing in general, the manufacturing of cheap whites and the illegal movements of those products 5 This occurs when licensed and authorized producers underreport actual quantities and sell the non- duty-paid produced quantities through illegal channels. It can also include instances when producers do not report quantities at all, as in many ad valorem systems of LMICs. 6 From 2000 on, the European Commission and 10 Member States launched court cases regarding smuggling and money laundering against several TTCs. To end the court cases, the Anti-Contraband and Anti-Counterfeit Agreements were signed, which required the TTCs to exercise stringent control over their supply chain (through tracking and tracing, due diligence and anti-money-laundering and reporting obligations), share operational intelligence with Member States and the EU and pay penalties for seizures, as well as annual payments over a period of 12 years. The agreement with PMI has ended, the one with JTI will end in 2022 and the others with Imperial Tobacco and BAT will run until 2030 (23). 180 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N over the past two decades, trade statistics are no longer as useful as they were in the past. Joossens et al. (25) tried to estimate the size of the global illicit cigarette market by adding different types of national estimations prepared around 2007. They found that the estimated size was 657 billion cigarettes per year, or 11.6% of the global cigarette market. According to Joossens et al., illicit trade costs govern- ments US$ 40.5 billion in tax revenue worldwide, and eliminating illicit tobacco trade would recover US$ 13 billion in immediate revenue in high-income countries and US$ 18.3 billion in LMICs.7 Descriptions of the types of illicit trade are useful for developing the first com- ponent of a strategy to fight it: assess the nature and size of the problem. Table 4.1 presents all the main components of a strategy to fight illicit trade. To make progress in this first component – knowing the problem – authorities could use and adapt existing instruments of health surveillance or seek partnerships with academia and independent specialists to investigate the issues involved, using different methodolo- gies (see subsection 4.1.4 and Annex 4.1 on methodologies to assess the nature and size of the problem). Knowing the nature of the problem requires the cooperation of different government actors – for example, to investigate both the financial and criminal operations of organized crime behind the illicit trade. The gathering of qualitative information on the nature of the illicit trade should start simultaneously with the statistical work of measuring the magnitude of the problem. Table 4.1 Components of a strategy to fight the illicit tobacco trade 1. Assess the nature and size of the problem Use and adapt existing health surveillance and other existing national surveys to assess the problem Seek partnerships with academia and independent specialists to find ways to rigorously study illicit trade Use financial and police investigations to identify and fight organized crime operating in illicit trade 2. Start identifying and implementing appropriate country-specific policies and strategies to address illicit trade Improve tax and customs administration to close the legal and administrative loopholes facilitating illicit trade Implement other appropriate policies to deal with country-specific problems 3. Become a Party and/or implement the Protocol to Eliminate Illicit Trade in Tobacco Products Adapt the Protocol supply-chain control obligations Adjust national penalties for illicit trade offences Seek and build international cooperation 7 The WCO publishes an Illicit Trade Report annually, with the main characteristics and trends of illicit flows in key products, including tobacco, using data based on customs seizures. CHAP T ER 4. PO LI T I C AL ECO N OMY 181 Methodologies available to estimate the nature and size of illicit trade are discussed in subsection 4.1.4. This is the first step for dealing with SCARE tactic S. Chapter 3 discusses at length the relevant tax administration measures and best practices to minimize opportunities for illicit trade in tobacco products. Table 4.2 presents examples of appropriate policies and strategies targeted to address specific types of illicit trade in addition to the best practices described in Chapter 3. After completing the first step of this strategy, tax, health and justice authorities should discuss how to face country-specific problems, considering not only tax and customs administration measures but also social, law enforcement and international cooperation policies and strategies. Table 4.2 Suggested policies and strategies to address country-specific illicit trade problems MAIN TYPE OF ILLICIT TRADE IN THE JURISDICTION PROBLEMS POLICIES/STRATEGIES TO USE Bootlegging Neighbouring low-tax jurisdiction Bilateral negotiations to harmonize tobacco tax systems Difficulty of controlling people’s movements in countries with extensive land borders Identify and establish suitable social protection or employment policies for targeted populations in border regions Extensive land border with multiple accesses Bilateral cooperation with law enforcement and border control forces, monitoring of access routes to main consumption markets Large-scale smuggling from neighbouring jurisdiction Neighbouring low-tax jurisdiction and difficulties in controlling borders Bilateral negotiations to harmonize tobacco tax systems and bilateral law enforcement cooperation Producers and distributors in the lower-tax jurisdiction aiming to supply the high-tax jurisdiction Bilateral cooperation to harmonize tax systems and control producers and distributors in the origin country, create conditions for legal exports and taxed imports Large-scale smuggling from a third country or tax-free zones Producers and distributors aiming to supply non-duty-paid tobacco products wherever possible Customs and other forms of international cooperation to control and monitor exports from identified areas Domestic tax evasion Existence of many small informal or semi-formal producers Encourage business concentration through producer associations and cooperatives, create incentives for formalization and establish licensing rules and basic electronic information systems for raw material and production 182 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Underreporting from formal producers Improve tax administration with policies such as basic electronic information systems for inputs and production, establish neutral procedures to verify production, improve audit systems, increase third-party information on inputs and production of tobacco products Clandestine factories Law enforcement investigation of commercial associations with raw- material and machine producers and distributors Governments should try to identify the incentives and governance problems that encourage and allow illicit trade movement inside their country. As seen in Table 4.2, the design and implementation of policies to deal with those problems do not depend exclusively on tax and customs authorities; they also depend on the efforts of the police and law enforcement, the Justice Department and the judicial apparatus. In other words, a great deal of coordination and consultation among different types of government bodies and expertise is needed to produce an adequate response. It is also clear from Table 4.2 that domestic tax evasion by formal producers can be tackled by tax authorities and is mainly related to the supply-chain-control provisions of the Protocol. The Protocol (26) builds upon and complements Article 15 of the WHO FCTC, which addresses means of countering illicit trade in tobacco products as a key aspect of a comprehensive tobacco control strategy. The Protocol is a blueprint of measures to deal with this problem, and its provisions should be part of any strategy for fighting the illicit market. It is a legally binding treaty in its own right that entered into force on 25 September 2018. As described in detail in Chapter 3, the Protocol has three main lines of action: supply-chain controls, recommendations on how to treat unlawful conduct related to the illicit tobacco trade and suggested mechanisms to seek and build international cooperation to fight that trade. Countries can start implementing Article 15 of the WHO FCTC and the appropriate polices or strate- gies recommended by the Protocol even before acceding to it, selecting those most suitable to the nature and extent of their particular problem. Such transitional work will facilitate the eventual implementation of the Protocol, because any plan to correct loopholes in tax and customs practices will bring government authorities closer to the best practices recommended in the Protocol. 4.1.3 DETERMINANTS OF TAX EVASION: THE ROLE OF PRICE LEVELS The argument that price and tax rates are the main determinants of the illicit tobacco trade has persuaded some governments (e.g. Uruguay and Georgia in the past) to avoid policies that may lead to cigarette price increases (e.g. excise tax rate increases) (4). CHAP T ER 4. PO LI T I C AL ECO N OMY 183 Some governments (e.g. Canada in 1994, Brazil in 1999 and Pakistan in 2017) (20, 27) have even reduced tax rates in attempt to reduce the illicit trade. The wider scholarly literature demonstrates that illicit trade is not a monocausal phenomenon (7) but is the result of many factors, most of them related to gover- nance issues. Government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, weak norms regarding participation in illegal and informal markets and conflicts between neighbouring countries (5) all contribute to the existence and growth of the illicit tobacco trade. It is difficult to isolate the role of price from each of the other factors because (1) obtaining prices and quantity measures of illicit trade is inherently challenging; (2) in most countries, there are many cigarette brands, and prices vary between and even among brands; and (3) there is a lack of good measures to deal with nonprice factors affecting illicit trade, such as government corruption and ineffective criminal justice. These constraints make it challenging to develop rigorous empirical evidence about how price and other factors affect illicit trade. Despite these fundamental challenges, the economic literature has produced credible evidence that price is always only one factor – and often not the most important factor – determining the extent of illicit trade. Many econometric studies about the influence of price and other factors have focused on cross-border shopping (or small-scale bootlegging from low- to high-tax jurisdictions), given the availability in the United States and Europe8 of sales data for low- and high-tax jurisdictions, classified in a convenient way by geographical zones – i.e. close to or far from the borders. Those studies attempted to explain the illicit trade flows or the relatively higher sales in low-tax jurisdictions as a function of price and tax differentials between the lower-tax and surrounding higher-tax jurisdictions, after controlling for other important factors affecting cross-border sales such as proximity to borders and levels of corruption (6, 7, 11, 28).9 The main conclusion of the studies is that illicit trade flows are not linked solely to price (29). Some show a significant effect of price differentials together with other factors, but others do not find significant price differential effects. The important policy implication of these analyses is that decreasing tobacco tax rates and real prices in higher-tax jurisdictions could have minimal or no effect on illicit market shares.10 8 This was a traditional strand of the literature in the United States on trade among states, and to a lesser extent in European countries, most of which used conventional but inaccurate illicit trade measurements. 9 Recently, PMI-Altria financed some studies of factors affecting cross-border sales. One of those studies, Prieger and Kulic (28), criticized Merriman et al. (2000) (9) and arrived at the conclusion that in cross-border shopping, price differentials are important for determining the magnitude of illicit trade. 10 Brazil decreased tax rates and real prices at the beginning of the 21st century to fight illicit trade coming from lower-tax jurisdictions. After this action, however, the government lost revenues, and the size and scope of illicit trade remained unaltered, according to industry sources (20). 184 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Observational and case studies provide information that may improve public policy even when they are unable to produce compelling evidence of causal relation- ships. Some observational studies have correlated price levels with illicit market shares, using large samples of countries. Joossens et al. (25) found that countries with high taxes and prices normally have lower shares of illicit trade than countries with lower tax shares and prices. In their sample, high-income countries generally have relatively high cigarette prices and tax shares, but their favourable results (i.e. lower levels of illicit trade) are related to effective tax administration and lower corruption levels. In contrast, LMICs generally have lower prices and tax shares, along with significant illegal market shares. Joossens et al. attribute difficulties in fighting illicit trade to weak tax and customs administrations and, in most cases, institutional and legal challenges (25). Figure 4.1.1 illustrates the relationship between price and illicit trade, using the price (in US$) per pack of the most-sold brand of cigarettes and the estimated level of illicit trade for 94 countries in 2018.11 There is no apparent unique association between the two variables. Running a linear regression with retail price as the explana- tory variable and share of illicit trade as the dependent variable shows an inverse, but not statistically significant, relationship between price and illicit market share.12 Figure 4.1.1 illustrates some particular cases: • Many countries with low prices (i.e. lower than US$ 2 per pack) have the highest levels of illicit trade in the sample, e.g. Brazil (BRA) ($1.33 and 46.3% illicit share), Pakistan (PAK) ($0.39 and 40%), Ethiopia (ETH) ($0.55 and 32.9%), Ghana (GHA) ($1.06 and 29%) and Cameroon (CMR) ($0.89 and 25%). • In contrast, many of the countries with prices between US$ 4 and $8 – which could be considered high enough for financial incentives to operate – have illicit trade shares of less than 10% of total consumption. These countries include the Republic of Korea (KOR) ($4.02 and 0.8%), Czechia (CZE) ($4.31 and 2.9%) and Sri Lanka (LKA) ($6.89 and 1.6%). • All countries that have very high prices – higher than US$ 8 – except for Ireland, register illicit trade shares below 20%. These countries include France (FRA) ($9.39 and 17.8%), Switzerland (CHE) ($8.71 and 5.5%), Singapore (SGP) ($10.35 and 3.7%) and Norway (NOR) ($14.51 and 9.6%). 11 National estimates of the magnitude of illicit trade are controversial. The tobacco industry’s numbers overestimate the problem and are based on questionable methodologies. Estimates with a rigorous and transparent methodology are not available for a large sample of countries for the same year. In order to compare price levels with illicit market shares, Euromonitor’s estimations of illicit market share were selected, for two reasons: they are comparable estimates for a large sample of countries in a given year, and no one could argue that they are biased towards tobacco control’s points of view. The use of Euromonitor data does not imply that WHO fully agrees with all the details and methodologies used to obtain them. 12 Other factors must be taken into account to transform this observational analysis into a rigorous analysis of cause and effect. CHAP T ER 4. PO LI T I C AL ECO N OMY 185 Fig. 4.1.1 Share of illicit trade versus retail price of the most-sold brand of cigarettes in US$, by country, 2018 Note: The extent of illicit trade in cigarettes is measured by Euromonitor as the estimated quantity of illegal cigarettes consumed in a country divided by the estimated total consumption of cigarettes in that country. Sources: (27, 30). 51 2 3 4 6 7 8 9 10 11 12 13 14 1615 0 10 20 25 15 5 30 35 40 45 50 MYS BRA ECU ECUPAK ETH PAN CRI URY GRCIND GTM SLV ARE LVAHND NGA ZAF CAN FRA LBNAGO IRL DOM GHA CMR Retail price, USD per pack Ili ci t t ra de % IRQ VNM COL MMR AUT SWE EST PER ESP ROM POL CHL ISR GBR KEN LAO KHM EGY MKD TZA THA TUN BGD UZB UKR GEO KAZ CHN KWT HRVJOR CZE SVK KOR JPN BLR SRB TUR OMN FIN NLD CHE SGP SAU USA LKA DNK DEU ITA HUN SVN PRT BOL DZAIDN CIV BEL MAR ARG AUS NOR NZL LTU BIH MEX BGRPHLAZE RUS 186 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As indicated in numerous studies and analyses of illicit trade in tobacco products (4–5), the most effective way to tackle the problem is not to forgo tax increases but rather to strengthen the capacity to fight the trade. Therefore, it is important to consider the relationship between good governance and illicit trade. The more capacity a country has to counter illicit trade in general, the lower the level of that trade will be. An index compiled by the Economist Intelligence Unit (EIU), the Global Illicit Trade Environment Index, measures countries’ structural capacity to fight illicit trade overall. The EIU indicator is a combination of four indicators or categories designed to assess countries’ performance in those areas; the closer the overall indicator is to 100, the better the country’s capacity to fight illicit trade. The four categories are:13 1. government policy, which measures the government’s commitment to pro- actively monitoring and preventing illicit trade; 2. supply and demand, which measures the extent to which the domestic en- vironment discourages or encourages supply and demand for illicit goods; 3. transparency and trade, which measures transparency and the degree of governance applicable to free-trade zones and transhipments; and 4. customs environment, which measures how effectively customs services facilitate legitimate trade while at the same time preventing illicit trade. Figure 4.1.2 illustrates the relationship between the EIU indicator and the estimated level of illicit trade in cigarettes in a set of countries. There is an inverse and statisti- cally significant relationship between the indicator and the estimated level of illicit trade in cigarettes. This suggests that as the capacity to fight illicit trade in general increases, the illicit trade in cigarettes falls.14 13 For more details about this indicator, visit http://illicittradeindex.eiu.com/. 14 The association was significantly different from zero at a 90% confidence level, using a linear regression between the two variables. CHAP T ER 4. PO LI T I C AL ECO N OMY 187 Fig. 4.1.2 Share of illicit trade versus the EIU indicator in 70 countries, by country, 2018 Sources: (30–31). % Il ic it tr ad e ci ga re tt es MYS 0 10 15 5 20 25 30 35 40 45 50 55 60 EIU GIT indicator 20 3010 40 50 60 70 80 90 JPN KOR HRV SAU UKR BLR KHM LAO MMR IRQ GTM DOM VNM MAR SRB PER BGR MEX ROM COL ZAF BIH PHL IDN ITA PRT HUN BEL DEU ESP ARG AUS SWECHL TW POL ISR AUTLTU FRA CAN IRL LVA ARE GRC IND URY CRI PAN PAK ECU BRA EST GBR FIN SVN NLD SVK TUN RUS DZA TUR CHNKAZ THA CZE USA SGP NZL DNK 188 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Rigorous independent research has established that despite the challenges of illicit trade, taxation of tobacco products is an effective public health intervention that substantially reduces tobacco use and generates government revenue (5). Further, when cigarette taxes increase, governments generate higher revenue and consumption is reduced (32–33). However, ineffective tax administration can allow illicit trade to grow and can undermine some of the benefits of tobacco taxation by making cheaper cigarettes available. For example, the average street price of smuggled cigarettes in Malaysia is 55% lower than its legal tax-paid equivalent (34). Illicit tobacco trade also reduces government tax revenue and may increase health costs associated with smoking and costs associated with policing. 4.1.4 MEASURING ILLICIT TRADE IN TOBACCO PRODUCTS The magnitude of illicit trade is a powerful argument in tax policy discussions, and for this reason the tobacco industry funds estimation of illicit trade in countries or regions of particular interest to itself (i.e. Project Sun and Project Star in the EU and Oxford Economics in East Asia). However, a recent systematic review of industry data on illicit trade finds substantial methodological weaknesses in industry-commissioned reports (24). Furthermore, Blecher et al. (35) argue that industry-funded studies tend to systematically overestimate the size of illicit trade to persuade authorities to abandon tobacco tax reforms. Independent researchers have also uncovered inconsistencies in tobacco-industry-funded estimates (36). Some examples of inflated industry-linked illicit trade estimates are given in Table 4.3, which compares peer-reviewed and independent studies with estimates funded by the tobacco industry. Because some countries have several industry estimates from different sources or years, Table 4.3 presents the estimate included in the article that published the independent study, because it was considered as representative and adequate to illustrate the overestimation. In all cases, the industry estimates exceed those of the independent studies. Measuring the scale of illicit trade can be a daunting task for governments because different methods are employed by independent researchers, governments and the tobacco industry. Nonetheless, it is worth investing in these studies because they drive policy discussions and can be used to evaluate the impact of policies (e.g. tax increases, plain packaging and health warnings). CHAP T ER 4. PO LI T I C AL ECO N OMY 189 Table 4.3 Illicit market share estimated in independent studies compared with estimates in tobacco-industry-funded studies COUNTRY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDEPENDENT STUDIES SOURCE OF THE INDEPENDENT STUDY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDUSTRY- FUNDED STUDIES INSTITUTION RESPONSIBLE FOR THE INDUSTRY- FUNDED STUDIES Colombia 3.5% of the total market in five cities, 2016 Maldonado et al., 2018 (37) 13% of the total market, 2014 FND and INVAMER, 2015 Chile 16.3% of the total market in the Metropolitan Region of Santiago, 2017 Paraje et al., 2020 (38) 24.3% of the total market, 2017 Observatorio del Comercio Ilícito BATC, 2017 Brazil 28.8% of the total market, 2014 Iglesias et al, 2017 (39) 34%, of the total market, 2014 BAT public statement, 2015 Mexico 8.8% of the total market in eight major cities, 2017 Saenz de Miera Juarez et al., 2020 (40) 16.6% of the total market, 2012 Confederación de Cámaras Industriales, 2012 As shown in Table 4.4, methodologies to measure illicit trade can be grouped into three types: (1) direct measurement; (2) residual methods and (3) expert opinion (12). Direct measurements rely on evidence directly linked to actual illicit behaviour and pack observation; residual methods infer evasion based on theory and evidence about consumption and legal sales; and expert opinion distills information garnered from talking to individuals with the most direct knowledge of the tobacco market. Each method has advantages and disadvantages. No single method is unambigu- ously superior to others, but direct measurement and residual methods are more conducive to determining the size of the illicit market, whereas expert opinion could provide insight into the details of the market’s operations.15 Table 4.4 presents the relative amount of resources and the degree of expertise required to implement each main measurement method, as well as the primary purpose, data collection characteristics, sampling features and unit of analysis. A brief description of each of the methods is presented in Annex 4.1. Merriman (11) and Ross (9) provide more expansive details. There is no simple selection rule for deciding what measurement method to use. The major factors to consider when selecting a method or methods include (1) the nature and characteristics of the illicit trade problem (i.e. where and how the 15 In interviews with experts from the tobacco industry, provisions of Article 5.3 of the FCTC and its Guidelines need to be followed. 190 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N problem manifests and whether domestic tax evasion or illegal inflows of foreign brands or a combination of both predominates), (2) previously collected data, (3) available budget and (4) expertise of available analysts. Available budget and staff skills are often the main restrictions that governments face. Therefore, Table 4.4 orders the measurement methods according to resources needed and available expertise. For example, residual methods and expert opinion can provide crude but useful estimates at low cost and require the lowest levels of technical sophistication. Another low-cost option for countries that employ popula- tion health surveillance surveys is to add questions to measure illicit trade, such as brand name, value and quantities of the last purchase. In contrast, the direct measurement approach often requires sophisticated research designs and expensive (and time-consuming) field research. Table 4.4 Overview of resources and expertise needed and main purpose of measurement methods METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Seizures (D) $ Low Identify trends in types of products, transporta- tion methods, points of entry and brand names Secondary data use Non- probability Shipments Use of existing health sur- veillance surveys – self- reported consump- tion (D) $ Low, only additional questions Size of illicit trade, adding or improving questions on brands, value and quantities of the last purchase Additional primary data collection Probability Individuals Gap analysis (R) $ Medium Provides a measure of changes in illicit trade Secondary data use Universe Nations Econo- metric modelling (R) $ High Estimation of price elasticity of substitution from tax- paid to illicit products Secondary data use Universe Geography Expert interviews (E) $ Low Characteristics of the illicit trade Primary data collection Non- probability Individuals Smoker intercepts and pack observa- tion surveys (D) $$$ Medium Size and characteristics of illicit trade, probability-based sample to be representative of population Primary data collection Probability Individuals CHAP T ER 4. PO LI T I C AL ECO N OMY 191 METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Pack return and swap surveys (D) $$$ Medium Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Littered- pack surveys (D) $$$ Medium Size and characteristics of illicit trade, compa- rability with industry estimation using empty-pack surveys Primary data collection Probability Individuals Covert purchases (D) $$$ Medium Type of products and trade channels of illicit trade Primary data collection Probability Geography Self-report consumer surveys (D) $$$ High Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Notes: Universe includes total population; D = direct measurement, R = residual method, E = expert opinion. Scale for resource costs assuming a moderately sized study (e.g. a representative study of a region of several million): $ (cheapest) – weeks of skilled labour hours; $$ (moderately expensive) – 1 to 2 months of skilled labour hours; and $$$ (most expensive) – 6 to 12 months of skilled and unskilled labour hours. A more detailed description of the different measurement methods is given in Annex 4.1. To further assist responsible authorities in deciding which method to select, Table 4.5 presents the key characteristics of each of the measurement methods, along with the main advantages and disadvantages of each. Countries may begin with methods that require fewer resources and less skills to obtain an overview of the problem. Seizures – which are a by-product of law enforcement efforts – pro- vide a first step, and countries can analyse the information obtained (origin of the products, brands, location, etc.) and report the results to increase public awareness of the problem.16 Alternatively, countries can add questions related to illicit trade to existing and funded health surveillance surveys conducted regularly by health surveillance authorities and statistical authorities. In that way, cooperation in using existing measurement methods between health authorities – the tobacco control office and health surveillance unit – tax and customs authorities and the national 16 Seizures are useful for obtaining qualitative information about the illegal activity, but they have to be treated very cautiously in projecting the size of the problem. Countries may think they have a very large problem because they have competent authorities doing an extraordinary job at finding illicit goods. On the other hand, countries can have less-efficient authorities making few seizures, and in these environments, seizures tell nothing about the size and nature of the problem. 192 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N statistical office could be a starting point for identifying the nature and size of the illicit trade problem in the country. Direct observation of packs has been increasingly implemented in many LMICs, through different types of surveys such as intercepts of smokers or retailers, pack return, littered-pack inspections and covert purchases of cigarettes. These activities have expanded the skills of independent researchers and academia and increased knowledge of these methods. Also, increasingly cheaper digital technologies allow interviewers to take pictures and record pack characteristics in direct observation surveys or in larger national self-report consumer surveys. Table 4.5 Key characteristics, advantages and disadvantages of illicit trade measurement methods METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Seizures (D) Statistics of tobacco products confiscated by local and national authorities Readily available from law enforcement agencies May not provide a representative picture of the size and/or nature of illicit trade Using existing health surveillance surveys to obtain self-reported consumption (D) Adding or improving questions about brand names, quantities, prices, locale of purchase and other factors Produces good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Gap analysis (R) Compare self-reported consumption data with observed (usually administrative) data about tax-paid sales When quality data are available, is simple and easily reproduced (providing for measurements over time) and explainable Data on tax-paid sales and/or consumption are frequently inaccurate and in many cases do not provide information on the size of the illicit market, but only on changes over time Econometric modelling (R) Estimated according to the difference between tax-paid sales and predicted consumption given by the model Because it is consistent with a long tradition of economic theory, empirical estimates can be evaluated Requires high-quality data on a variety of important variables over a period of time and advanced econometric modelling expertise CHAP T ER 4. PO LI T I C AL ECO N OMY 193 METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Expert interviews (E) Experts include researchers (e.g. in economics, criminal justice and public health), journalists, tax and enforcement specialists, product manufacturers and wholesalers Useful for identifying the nature of and trends in the marketplace (e.g. venues where illicit cigarettes are sold, modes of entry), and the interviews can be useful for defining the method to assess the size of the illicit trade Information obtained may not be generalizable, and expert knowledge may be outdated or limited by the experts’ experience; also, experts often have strong biases Smoker/retailer intercepts and pack observation surveys (D) Examining the packs of smokers and cigarette retailers, convenience or probability-based sample Is direct and objective, and smokers do not suffer from any value judgements The difficulty of identifying areas representative of the tobacco use population and sampling important subpopulations such as elderly and immobile smokers, but household surveys could overcome sampling issues Pack return and swap surveys (D) Also a pack observation study using survey sampling techniques to examine smokers’ pack characteristics May decrease the stigma associated with traditional smoking surveys In LMICs, survey distribution may be unreliable because of the mail delivery system Littered pack surveys (D) Also known as empty-discarded-pack surveys; publicly discarded packs bear characteristics (e.g. tax stamps, public health warnings) that indicate whether they are tax compliant Yields estimates that are less likely to be biased from issues of social desirability, recall error and confidentiality Significant budgets could be needed to employ field researchers to collect, code and analyse the data; surveys do not provide information about the smoker and the price paid Covert purchases (D) Uses covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes Directly identifies sources of illicit cigarettes It is difficult to create a sampling frame of retailers for illicit sources or to know what smokers are actually buying and how much Self-report consumer surveys (D) Surveys can be distributed to individuals or households, using various modes of distribution Good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Notes: D = direct measurement, R = residual method, E = expert opinion. 194 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ultimately, when capacity allows, more solid estimates will need to be made using more than one methodology at a given point in time. Ideally, estimates will be made on a regular basis in order to assess the evolution of illicit trade over time and its possible connection to policy changes. 4.1.5 CRITIQUING STUDIES THAT MEASURE THE SIZE OF THE ILLICIT MARKET Measuring illicit trade is a challenge for researchers, industry and governments, because the trade is, by definition, hidden from plain sight. Buyers and dealers sometimes go to great lengths to ensure that their participation in illicit activity is concealed. Over the years, researchers and government agencies have been increas- ingly interested in estimating the size of illicit markets and identifying effective interventions. As consumers of research, governments should critically examine available studies and evaluate them on their scientific rigor and methodological transparency. Measurement issues are particularly acute with respect to the illicit tobacco trade because it is a politicized topic. High estimates may raise questions about the tobacco industry’s ability to control the supply chain, its involvement in illicit diversion, the impact of taxation policies and the effectiveness of enforcement strategies. While the industry has portrayed itself as taking an active stance in measuring and fighting illicit trade (e.g. Project Star, conducted by KPMG LLC but paid for by PMI, later followed by Project Sun), in the past it has used smuggling as a strategy to enter closed markets – for example, in China and Russia (19, 41). Govern- ments should carefully scrutinize evidence about the illicit tobacco trade produced by industry or quasi-industry sources and are advised to seek alternative evidence. Quasi-industry reports are studies commissioned by the industry but published by private research companies (e.g. Ernst and Young, Oxford Economics) (11). Characteristics of good analyses One of the main characteristics of a good analysis is scientific rigor, which involves the use of relevant theoretical frameworks, sound statistical methods and examination of the robustness of findings (e.g. sensitivity analyses). High-quality research reports provide transparent explanations about their methodology and statistical analysis steps undertaken, as well as supplementary analyses that established the robustness of the findings. For example, Joossens et al. (25) clearly describe the data sources used (limitations and advantages and where they can be found) and calculations performed on the number of lives that would be saved if the global market share of illicit cigarettes was eliminated. Explanations should be detailed enough to allow future researchers to scrutinize the analysis and replicate the findings. Replicability is another hallmark of good science. For example, littered-pack studies should detail where and when data collection took place, how many packs CHAP T ER 4. PO LI T I C AL ECO N OMY 195 were collected per geographical unit, the protocol of identifying the illicit packs (e.g. characteristics of the warning labels, brands, tax stamps, etc.) and details of statistical analyses. There should also be explanations of the representativeness of the selected geographical areas. Failure to provide this depth of information may call into question the generalizability of a study and whether there are faults with the chosen method. In the context of policy decisions regarding illicit trade, the most useful data provide information about a representative sample of individuals and geographies. Studies that are limited to, for example, one group of individuals based on specific characteristics or a given geography may yield biased information. Research reports also should be clear about the study’s limitations. For example, studies that measure illicit trade often do not measure product counterfeiting and do not include non- cigarette tobacco products in their estimates (42). When statistical estimates are included, they should provide confidence intervals as well as point estimates to account for uncertainty resulting from simple random chance (11). Characteristics of flawed analyses Flawed analyses can convolute and distort scientific knowledge about illicit trade. Flaws usually manifest in the data, methodology, statistical analysis and/or interpreta- tion of the results (11). Studies may be purposefully designed with methodological flaws to yield high or low estimates of the trade. For example, research showing that illicit trade constitutes a large share of the total market may be used to support arguments that taxes cause sharp increases in illicit trade, whereas lower estimates may be used to support arguments that certain governmental interventions (e.g. increased retail inspections) are effective. Pressures to skew data may also be tied to funding. For example, high estimates can sway governments to provide more resources for law enforcement activities. Analyses can be purposefully skewed by using data sources or data collection methods that will provide biased estimates. Flawed studies sometimes provide incomplete or inaccurate descriptions of their methodology. They may lack detail regarding the quality of the data used or information about how the data were collected and analysed. For example, a common weakness in industry-funded research on discarded packs is that the methods of collection and forensic analysis are not reported, ostensibly because they are “proprietary” information (43). However, these methodological details are key to assessing whether a study’s findings are biased by sampling error, model misspecification, measurement error, non-response or other flaws. It may be impossible to assess measurement error if researchers fail to disclose questions included in a survey instrument. Survey items used to measure the illicit tobacco trade may be imprecise. For example, asking respondents the frequency with 196 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N which they purchase “cheap” cigarettes may yield biased estimates, particularly if consumers can purchase cigarettes at discounted prices by using coupons. To more ac- curately measure tax evasion, surveys must include questions about the location of last purchase, purchase price, presence of public health warning labels and brand names. There are other ways that flawed studies can inadvertently or purposefully distort estimates of illicit trade. For example, data collectors can intentionally oversample areas known to be hot spots of illicit sales or sites that residents from lower-tax jurisdictions visit. Researchers can collect discarded cigarette packs close to the borders of countries with lower taxes to (inadvertently or purposefully) demonstrate the undesired side effects of tax policies. Studies published in non-peer-reviewed or lightly peer-reviewed outlets such as edited book volumes or policy briefs should be viewed with more scepticism than those published in highly regarded peer- reviewed outlets. 4.1.6 CONCLUSIONS Globally, the illicit tobacco trade continues to be a major concern for tax admin- istrators because of the challenges it generates to collecting higher revenues as well as the challenges to accurate and independent measurement. Industry figures provide distorted conclusions regarding the extent of the problem – frequently with a monocausal explanation of the link between illicit trade and tobacco taxation. Illicit trade comprises multisystemic issues and requires multiple strategies. Worldwide, countries at different levels of economic development have implemented a variety of effective measures to combat the illicit trade in tobacco products. The Philippines and the United Kingdom, for example, have addressed illicit trade as part of their overall tobacco tax reform (4). Price (and tax) levels are not a key determinant of illicit trade, the presence of which is exacerbated by the lack of tax administration capacity. Refraining from increasing taxes is not the solution; countries should instead respond with a com- prehensive strategy that includes at least these three main components: 1. It should identify – independently from the industry – the nature and dimen- sions of the problem. It is necessary to assess scientifically and with the best statistical practices the size of the illicit trade to understand the characteristics and scope of the problem. 2. It should identify and implement appropriate policies and strategies targeted at addressing the specific type of illicit trade the country is experiencing. It should address directly the country-specific institutional and/or governance challenges – as well as the lack of multilateral coordination that can exacer- bate illicit trade – and improve tax and customs administration practices as described in Chapter 3. CHAP T ER 4. PO LI T I C AL ECO N OMY 197 3. It should implement best practices contained in the WHO FCTC Protocol to Eliminate the Illicit Trade in Tobacco Products and accede to the Protocol if the country is not yet a Party. There are proper methods and policies with which to address the illicit tobacco trade. If countries start implementing the appropriate policies, they can raise tobacco taxes and reap health and revenue benefits even in the presence of illicit trade. 198 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.2 SCARE TACTIC C: COURT AND LEGAL CHALLENGES 4.2.1 INTRODUCTION The tobacco industry views well-designed and significant tax increases as a threat to the profit, growth and long-term sustainability of its business. As noted by PMI in 1985: Of all the concerns there is one – taxation – that alarms us the most. While [other restrictions] … do depress volume, in our experience taxation depresses it much more severely (44). The industry is, however, less likely to launch direct legal challenges to excise taxes than to other tobacco control measures (see Box 4.2.1 for details), because taxation – and excise tax in particular – is a comparatively well-established regulatory measure; in many jurisdictions, taxes have been levied on tobacco products for more than a century. There is also less unanimity in opposition to taxes among tobacco industry actors, because differences in the market position of different tobacco companies affect their interests in tax policy. This, in turn, decreases the likelihood that they will act collectively on the issue (45). BAT’s stated strategy in the early 1990s was to influence governments with regard to the level and structure of tobacco taxation in order to promote market growth and to secure competitive advantage (46). Nevertheless, tobacco industry actors will still legally challenge, or at least legally threaten, significant tax measures when vulnerabilities in their design, adoption or implementation are apparent. Box 4.2.1 Court and legal challenges to tobacco tax measures Evidence suggests that the tobacco industry and its allies instigate fewer legal actions against tax measures than against other tobacco control measures: 1. The Campaign for Tobacco-Free Kids’ tobacco control laws database contains only a handful of cases concerning tobacco tax measures, but hundreds on other tobacco control topics. This pattern can also be seen in a 2018 review of tobacco control legal challenges that examined this and two other databases to select 96 cases relevant to the question of the WHO FCTC’s usefulness in litigation (47). Only 6 of these 96 cases were challenges related to tax measures. CHAP T ER 4. PO LI T I C AL ECO N OMY 199 2. A 2013 systematic review of empirical studies on tobacco industry interference with tobacco tax policy found that only 9 of 36 relevant articles reported the specific use of litigation as a tobacco industry tactic (1). All 9 concerned constitutional challenges to earmarking provisions for tobacco tax initiatives in the United States (1). 3. A 2015 study on industry interference in LMICs cited legal challenges to tobacco control measures in 15 countries as examples of industry interference, but none of the challenges concerned a tobacco tax measure (48). 4. A 2016 analysis of papers published in systematic reviews of industry inter- ference with tax and marketing measures found that only 5 of 65 papers concerning tobacco tax related to the use of litigation or threats of litigation to interfere with tobacco tax measures (49). The tobacco industry makes extensive use of legal experts (1, 50–52) who study all relevant laws and regulations closely to determine their likely and arguable boundar- ies for the purpose of manipulating regulations and regulators (1, 50–52). Based on this expert advice, tobacco companies know when regulations remain within the bounds of both international and domestic obligations but can still argue that legally permissible tobacco control measures would be defeated in litigation if passed (48, 51–52). As the threat of a legal challenge alone can be used to the industry’s advantage, recourse to litigation is seldom needed or desirable (1, 45, 48, 51, 53–56). Even when litigation is launched, the objective may be to delay or weaken a measure rather than to win on the merits of the case (1, 45, 48, 53). To counter actual and threatened legal challenges, policy-makers need to be aware of relevant legal obliga- tions when preparing and implementing tobacco control measures. Fortunately, the tobacco industry playbook is relatively predictable. Tax and other tobacco control measures can thus be designed to strengthen the regulators’ legal position against genuine threats and enable them to dismiss baseless industry threats. 4.2.2 COUNTRY EXPERIENCES WITH LEGAL CHALLENGES TO TOBACCO TAXATION Legal obligations that are relevant to tobacco taxation include those under do- mestic law and international instruments such as international trade agreements and international investment agreements (IIAs).17 Some of the legal issues that a tax measure may encounter are outlined in Table 4.6. Case studies from various countries illustrate how these legal issues have and have not been avoided in the 17 Relevant international trade agreements include the WTO Agreement and custom unions such as the EU, the East African Customs Union and Mercosur. Relevant IIAs include bilateral investment treaties and the investment chapters in free trade agreements and within custom unions. 200 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N passage, design and implementation of tobacco taxes. These issues are not the norm, however, and should not give rise to undue apprehension. The case studies are rated as positive, mixed or negative based on the extent to which the legal decision upheld the taxation measure in question. Table 4.6 Potential legal issues for tobacco tax measures VULNERABILITIES LEGAL OBLIGATIONS CASE STUDIES Inadequate consultation and other procedural vulnerabilities Domestic procedural law 1, 2 Due process protections for investors under IIAs None Procedural requirements under WTO Agreements and Custom Unions 3 Discrimination against imports or investors Nondiscrimination obligations under WTO Agreements and Customs Unions 8, 9, 10 Nondiscrimination obligations under IIAs 11 Investment incentives or inducements Arbitration mechanisms under investor-state contracts 12 Fair and equitable treatment clauses of IIAs None Other substantive breaches Constitutional rights and restrictions on taxation 4 Statutory restrictions on the imposition of taxation 6 Expropriation clauses of IIAs 5 Ultra vires (the scope of legal authority) 7 Avoiding procedural vulnerabilities in tax laws Procedural defects can be avoided by taking great care in progressing and imple- menting regulatory or legislative provisions. Procedural concerns pose a dilemma for tobacco control regulators. Article 5.3 of the WHO FCTC and the COP guide- lines for its implementation state that policy-makers and regulators should interact with the tobacco industry only when and to the extent strictly necessary (57). For taxation measures, interaction might be necessary because consultative and de- liberative processes could be prescribed under domestic constitutional provisions and procedures for good governance, due process requirements of IIAs and some international trade agreements. The tobacco industry may use these requirements as leverage to delay, distort or hijack the rule-making process in contravention of Article 5.3. Accordingly, interactions with the tobacco industry should be limited to strictly necessary consultation conducted in a transparent or public manner but with care that this does not come at the expense of a measure’s defensibility. The proper balance will depend on the jurisdiction in question, since constitutional, statutory and applicable international legal obligations vary. CHAP T ER 4. PO LI T I C AL ECO N OMY 201 CASE STUDY 1 (MIXED): Industry manipulation of legislative procedures In 2012, a bill stipulating, among other things, the creation of a new specific excise tax on cigarettes passed its final reading in Costa Rica’s Legislative Assembly. Passage of the bill had, however, proceeded under “urgency” and notwithstanding a pending constitutional enquiry (a constitutional query is meant to prevent passage of a bill).18 ISSUE MAJORITY DECISION MINORITY DECISION LESSON Whether the court could consider the enquiry despite passage of the bill and the effect the bill’s passage could have despite the enquiry. The enquiry was taken up by the Supreme Court’s Constitutional Division’s majority (58). The signing and publication of the bill by the executive was suspended by the Constitutional Division pending their decision on the merits of the case – which, in the end, found any question of the bill’s constitutionality baseless (58). The enquiry was inadmissible by reason of having been filed too late and notice of its filing having not been received by the legislature prior to the reading of the bill (58). In disagreement with the majority, the minority held that the court could not consider the enquiry or suspend the bill’s signing by the executive – the final step in becoming law. This challenge demonstrates how the tobacco industry’s defenders may attempt to frustrate and impede a tax measure’s passage. In this case, the challenge seemed to have been a delaying tactic, as it was posted on the same day as the final reading of the bill. Its authors may have either wanted its pending nature to cause the legislature to delay or, as occurred, to create conditions for a procedural and constitutional challenge in the absence of delay. All the grounds of the challenge itself were found to be without merit. Although such frivolous challenges cannot be prevented, they can and should be anticipated to ensure that they do not lead to a tax measure’s defeat. 18 “Urgency” is a procedure under which a bill is progressed through a legislature in an expedited fashion. 202 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N CASE STUDY 2 (POSITIVE): Adhering to domestic procedural requirements Kenya’s tobacco control regulations required the tobacco industry to pay a levy to compensate the state for health care and other negative externalities of smoking. In a 2016 challenge brought against these regulations, the plaintiff, BAT, was unsuccessful on every count (59–60). Even though the levy was not considered a tax measure by the court, the case study is instructive on how regulators may safeguard tax measures against procedural challenges. ISSUES LAWS AND ARGUMENTS DECISIONS LESSONS Whether the government’s consultations on the measure were adequate. Asserting that the Constitution and the Statutory Instruments Act together meant that “appropriate consultations with persons who are likely to be affected” were required because of the measure’s likely substantial effect on business. BAT claimed that this standard was not met. Kenya’s government claimed that it was under no obligation to undertake special or extensive consultation with the tobacco industry. The judge found in favour of Kenya’s government, noting that (1) the requirement to consult does not imply that any particular view needs to prevail; (2) dissatisfaction with the level of consultation is not decisive; (3) on the facts, industry was allowed, and often invited, to send representatives to all relevant public consultative meetings and parliamentary committee hearings; and (4) consultation on the regulations was adequate (59). The tobacco industry carefully scrutinizes legislative and regulatory processes for defects. In this case, Kenyan government officials appropriately distanced themselves from the tobacco industry by not permitting its representatives special consideration but did permit their attendance at public meetings and the ability to submit their views under usual procedures. In this way, both the principles behind WHO FCTC Article 5.3 and the requirement for consultation under Kenyan law were observed. CHAP T ER 4. PO LI T I C AL ECO N OMY 203 Avoiding procedural issues in tax administration CASE STUDY 3 (NEGATIVE): Contravening procedural requirements in international obligations In 2010, a WTO panel held that Thailand violated the Customs Valuation Agreement (CVA) by the process it used to value cigarettes that Phillip Morris (PM) Thailand imported into the country from a related party, PM Philippines. Customs values are important as they are the tax base for tariffs and can feed into the base for other taxes levied against the value of the good, such as ad valorem excise taxes and VAT. Transaction values declared by PM Thailand were rejected by Thai tax authorities as influenced by the relationship between the parties and a customs value determined by deduction was substituted (61). ISSUE LEGAL OBLIGATION DECISION LESSONS Whether Thailand adequately consulted with PM Philippines before rejecting its declared transaction value (61). The CVA requires good faith exchange of reasons and information, with opportunities for response (61). Thailand had failed to properly explain its reasons for rejecting the transaction value, as well as its belief that price was influenced by the relationship between the two parties (61). This was a violation of the CVA. Thailand did not appeal these findings. Thailand’s authorities needed to take greater care in their dealings with the tobacco industry to ensure they met the pertinent procedural obligations. In this instance, a specific and high standard of consultation – the provision of detailed reasons and an opportunity for response – was prescribed by the CVA and Thailand failed to meet it. 204 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ensuring compliance with substantive requirements Rules found in domestic and international law also establish substantive obliga- tions. This subsection focuses on substantive obligations found in domestic law. International obligations concerning discrimination and investment incentives are considered in the next subsections. CASE STUDY 4 (POSITIVE): Tax measure found to be consistent with the Constitution The Chilean government introduced a substantial increase in tobacco and fuel excise, and in 1995, a coalition of taxpayers brought a Constitutional challenge to the measure (62). ISSUE DECISION LESSON Whether the tax was, per article 19 of the Chilean Constitution, “obviously disproportionate or unjust” (62). The excise tax increase did not violate the Constitution, as it was neither confiscatory nor manifestly irrational. Generally applicable excise taxes are not vulnerable to challenges for being excessive, unfair or disproportionate. CASE STUDY 5 (POSITIVE): Failure to grant tax rebates not an expropriation under an IIA This case study is an example of a claim for breach of an expropriation clause in an investment treaty. Such clauses protect foreign investors against measures that can be construed as directly or indirectly seizing an investment or depriving it of its value (63). In the case, an investor was, for more than a decade, denied tax rebates by the Mexican government. This affected the profitability of the business of purchasing and reselling Mexican cigarettes abroad, and the investor brought the claim to an investment agreement arbitral tribunal in 2002 (64). ISSUE DECISION LESSONS Whether Mexico’s failure to grant rebates to the investor exceeded the bounds of valid regulation to constitute indirect expropriation of the investor’s investment (65). There was no expropriation. The arbitral tribunal noted that not all business problems are violations: the investor had no right to participate in the “grey market” export of cigarettes and there were sound reasons to restrict that market (65). Further, the investor was able to participate in other business ventures and actually continued to have business success (65). Claims of indirect expropriation made under IIAs are unlikely to be successful, as generally applicable tax measures are a legitimate form of regulation. A mere loss of profit will not suffice. Claims of expropriation will not succeed unless a substantial or significant deprivation of the investment results. CHAP T ER 4. PO LI T I C AL ECO N OMY 205 CASE STUDY 6 (NEGATIVE): A regulation contrary to superior domestic legislation In 2011, an Indonesian tobacco industry association group, FORMASI, challenged a new excise regulation. Since 2009, the government had been implementing a tiered specific excise tax system based on a set of characteristics (size of production, type of cigarettes and price levels). In 2011, excise rates were increased in nearly all of the 19 tiers, but the reference prices were not accordingly adjusted. This gave rise to a legal issue. ISSUE DECISION LESSONS Whether new excise regulations breached a 57% ceiling for the rate of excise on the retail sale price of tobacco products under the superior Excise Law (66–70). The challenge specified that excise exceeded this ceiling for hand-rolled domestic clove cigarettes (kreteks) (68, 71). The Court found in favour of the tobacco industry association, and the government was required to immediately revoke the 2011 regulation. (69–70). It is advisable to stay within the rules and be aware of legal hierarchies – including superior domestic legislation. The tobacco industry scrutinizes all increases in tobacco taxes. In this case, a breach of a legislative requirement for a single category of tobacco product resulted in Indonesia suffering lost revenue and a setback in its efforts to reduce tobacco consumption. Ensuring a tax measure is within an authority’s legal power A tax measure is ultra vires when it goes beyond the legal power of the enacting body. As with case study 6, this is a legal issue that involves legal hierarchies. In ultra vires cases, however, instead of centring on conflict between inferior and superior law, the issue is whether an authority that enacts a tax measure is authorized to do so. This issue may arise when a tax measure is enacted by a subnational jurisdiction or by an executive acting under a statutory delegation. CASE STUDY 7 (NEGATIVE): Tobacco taxation contrary to the Australian Constitution ISSUE DECISION LESSONS Whether New South Wales’ licensing and penalty fees regime constituted an excise tax by other means contrary to the Australian Constitution’s exclusive grant of that power to the federal government (72). The court found that state licensing fees were excise taxes and that this was contrary to the Australian Constitution (72). Authorities enacting tobacco tax measures must act within the scope of their legal power. 206 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Avoiding discrimination against imports and foreign investments Although inherently discriminatory, customs duties may be used subject to the agreed upper limits in a country’s trade agreements. Excise and other taxes designed with the aim of raising tobacco prices to reduce demand and advance human health should be origin-neutral: they should not seek to tax local products less than foreign products or aim to treat foreign products differently from one another. Tobacco tax measures are pursuing objectives other than health when they aim to raise the price of imports more than that of local products or seek to burden favoured market participants less than others. Solely health-protective tobacco taxes will not ordinarily violate Articles III:2 and I:1 of the GATT (the WTO’s General Agreement on Tariffs and Trade), which prohibit discriminatory taxation (in light of general exceptions). Nor will solely health-protective tobacco taxes directly violate anti-discrimination protections for investors found in the national-treatment (NT), most-favoured-nation (MFN), expropriation and fair-and-equitable-treatment (FET) clauses of IIAs (63, 73–74). It is possible to make claims for breach of international obligations on grounds other than discrimination, but such claims are generally highly unlikely to succeed. CASE STUDY 8 (NEGATIVE): BAT v Uganda (2017 East African Court of Justice) DISCRIMINATION LESSON Uganda established a higher level of excise taxes on imported cigarettes – including those from Partner states of the East African Customs Union (75) – than on local cigarettes. Its implementation was discrimination contrary to Article 15 of the Customs Union Protocol (75). Differential taxation explicitly based on origin can be construed as protectionist discrimination in violation of international obligations. The tobacco industry can also turn to international trade agreements outside of the WTO – in particular, customs union mechanisms. Difficulties arise when ostensibly origin-neutral and health-protective tobacco taxes result in dissimilar taxation of tobacco products (73). Discrimination does not exist simply because there is dissimilar taxation – the taxation must adversely impact imported goods more than local products, the imports of one nation more than another or a particular investor’s products more than comparable products. Where dissimilar taxation between product categories results in discrimination, the tax will ordinarily still be lawful if the dissimilar taxation is based solely on a legitimate regulatory distinction between the product categories in question.19 19 The precise applicable rules vary depending on the nature of the legal obligations in question. Under the GATT, dissimilar taxation of like or directly competitive products can be justified based on scientifically grounded distinctions between products under Article III:2 and, in the alternative, discrimination that is necessary under the explicit carve-out for health-protective measures, Article XX(b) (73). For the MFN and NT clauses of IIAs, differential taxation can be argued as nondiscriminatory on the basis that difference in harm means the products are not “alike” or, in the alternative, discrimination is justified based on scientific evidence of differences in harm and rational reasons for the health-protective role of differential taxation (63). CHAP T ER 4. PO LI T I C AL ECO N OMY 207 Where discrimination is inadvertent, lack of an intention to discriminate is not sufficient as a defence for breach of obligations under IIAs or the GATT (73, 76). Policy-makers should carefully scrutinize measures to determine: 1. whether an aspect of a tax measure’s design or implementation may be more to the detriment of imports or foreign investors than of local products or domestic investors; 2. whether the potentially discriminatory aspect of the tax measure serves any useful purpose in supporting the tax measure (i.e. it is needed to achieve the health goal); 3. whether there is any reasonable alternative that could achieve the same effect without the potential for discrimination; (i.e. it is indispensable) and 4. when it is needed and indispensable there is a good chance that it will be defensible. The case studies below provide examples of discrimination arising in connection with a tobacco tax measure. CASE STUDY 9 (NEGATIVE): Thailand – Customs and Fiscal Measures on Cigarettes from the Philippines (2010 WTO panel) The facts of this case are presented in case study 3. This case study examines claims of discrimination rather than the procedural issues. DISCRIMINATION EXPLANATION LESSONS Thailand implemented its policy for determining the tax base for VAT on cigarettes inconsistently (61). Thailand applied a methodology in fixing the tax base, in particular a marketing cost component, of imported cigarettes that differed from that for local products (61). This resulted in the marketing cost component for the imported cigarettes being higher than it would have been under the general methodology. This difference in treatment was insufficiently justified and therefore considered discriminatory. As there is potential for inadvertent discrimination when the base for an ad valorem tax is fixed, tax base determinations must be consistent and well- reasoned (61). This case study demonstrates how policy- makers need to take care in designing and implementing ad valorem taxes to ensure they are nondiscriminatory and legally defensible. 208 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Thailand’s VAT rebate policy imposed a potentially higher tax burden and also created more burdensome administrative requirements for imported cigarettes (61). Resellers of cigarettes produced by a government entity were granted an exemption from VAT (61). Although resellers of imported cigarettes would be eligible for a tax credit on their VAT, this was not an automatic process (61). The distinct treatment of resellers of imported cigarettes and those of local cigarettes resulted in the risk that there would be a higher VAT burden for the former (61). The distinct treatment also imposed an additional administrative burden on resellers of imported cigarettes and altered conditions of competition (61). Rules for the collection and enforcement of tax obligations should be the same, or as similar as practicable, in both form and effect for domestic and imported tobacco products. CASE STUDY 10 (NEGATIVE): Dominican Republic – Measures Affecting the Importation and Internal Sale of Cigarettes (2004 WTO panel; 2005 WTO Appellate Body) Under article XX(d) of the GATT, discrimination that is necessary to secure com- pliance with a legitimate tax measure will be justified provided there is no less- discriminatory alternative. In this case, this justification was used unsuccessfully. DISCRIMINATION EXPLANATION LESSONS The Dominican Republic’s tax stamp regulations were discriminatory towards imported goods (77). Under the regulations, all cigarette packs had to be affixed with tax stamps, but imported cigarettes were to be affixed with tax stamps under the supervision of local tax authorities following importation, while locally manufactured cigarettes could be affixed with a tax stamp in the course of production. This de facto distinction between local and imported products modified the conditions of competition to the detriment of imported cigarettes by (1) increasing costs for importers and (2) impairing the aesthetics of imported products (77). The panel did not consider this discrimination justified: it was not necessary for the enforcement of tax measures, because less restrictive alternatives were available such as permitting importers to affix tax stamps during the course of production (Dominican Republic – measures affecting) (77). The panel’s findings were upheld on appeal (78). Policies crafted to ensure compliance with tax measures need to also be nondiscriminatory. Discrimination claims can arise when compliance costs are higher for imports than for local products and this de facto distinction is avoidable. It is important to consider whether less burdensome alternatives may achieve the same objective. CHAP T ER 4. PO LI T I C AL ECO N OMY 209 CASE STUDY 11 (NEGATIVE): Feldman Karpa v Mexico (2002 ICSID [International Centre for Settlement of Investment Disputes] Arbitral Tribunal) Arbitral tribunals have accepted differences in treatment accorded to investors protected by IIAs when there is a legitimate connection between the distinctions drawn and public welfare objectives (76). The facts of this case are presented in case study 5. This case study examines aspects of the case involving the investor’s claim of discrimination, rather than the substantive issue of expropriation. Claims of discrimination are made on different grounds than claims for expropriation, which is why the case was decided differently on this claim. DISCRIMINATION LESSONS Denial of foreign investors’ claims for tax rebates. Tax rebate claims were granted to similar local investors (65), which was a violation of an IIA’s national treatment clause (65). Foreign and local investors must be treated similarly, and consistent and well-documented policies must be used to guide administrative decisions. The denial of the rebates may have been justified, but the government was unable to establish this due to a lack of documentation. Avoiding the investment incentives trap Investor-state contracts between the tobacco industry and governments should be avoided. They are not merely “contractual” in the domestic law sense, as even in the absence of an applicable IIA, they can be internationalized to provide inves- tors the right to (1) remove dispute settlement from the state’s court in favour of independent arbitration and (2) remove the dispute from the state’s legal framework in favour of general principles of law (63, 76). Commitments under these clauses cannot, therefore, be legislatively moderated or extinguished, nor can liability be limited within domestic courts that may be more likely to favour the state’s right to regulate in favour of public health (76). Investor-state contracts and other noncontractual inducements can be further internationalized by umbrella clauses within IIAs. Such clauses make reneging on undertakings assumed towards investors a breach of the IIA (76). Moreover, even in the absence of an umbrella clause, contracts and inducement can underpin a claim for legitimate expectation and breach of fair and equitable treatment and can also strengthen an investor’s claim for indirect expropriation (63). Arbitral awards make clear that although taxes can be expected to vary and tobacco will be regulated, investors can have the legitimate expectation that states will abide by formal inducements and written contractual undertakings. A common clause within investor-state contracts, the stabilization clause, is ruinous to evidence-based tobacco control’s most effective measure: excise tax in- creases. Stabilization clauses purport to freeze specific domestic law from the time 210 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of investment (63). Seemingly less onerous, economic equilibrium clauses require contracting states to compensate for regulatory changes that negatively affect an investment’s value (63). There is little difference in effect between these two types of clauses: liability for the cost of breaching an equilibrium economic clause can be onerous enough to make it fiscally challenging and politically unpalatable. CASE STUDY 12: An investor-state contract A state entered into an investment agreement with a TTC in 2001 on the privatiza- tion of its state-owned tobacco enterprise and creation of a joint venture. This investment was to provide economic benefits under the agreement: the joint venture would increase exports and profit using the TTC’s cash and expertise while also ensuring prioritization of local employment, manufacturing and resources. The final investor-state contract included a form of economic equilibrium clause under which any increase in the excise tax rates applied to the company’s tobacco products before a set date would be compensable. While the agreement was not removed from the state’s law, it provided for independent arbitration in case of a dispute over its compensation. In addition, there is a bilateral investment treaty between the host state and another state in which the TTC’s subsidiary has residence that includes a FET clause – this could buttress, if needed, the protection provided by the stand-alone arrangements of the investor-state contract. There were similar less-formal inducements offered to a separate TTC. The extent to which incentives have been granted to the tobacco industry is unknown, but contracts and inducements are likely to be offered in the context of the privatization of state-owned tobacco interests and in dealings between investors and state-owned tobacco enterprises (63). Although countries have been entrapped by their incentives to industry, the investor-state contract provides the clearest example of how undertakings and inducements with the tobacco industry under- mine tobacco control (56, 79–80). States should avoid offering industry incentives and, in particular, entering into contractual undertakings with the industry. More systematically, government should consider avoiding IIAs that elevate incentives and inducements above sensible and reasonable regulation. 4.2.3 CONCLUSIONS Health-protective and origin-neutral tobacco excise taxes are legally defensible, and industry threats are usually baseless. There are, however, certain rules governing procedure, design and consultation that governments may need to consider: 1. Governments should be aware of the standard of consultation required under do- mestic law and any applicable international obligations (case studies 1, 2 and 3). CHAP T ER 4. PO LI T I C AL ECO N OMY 211 It is important to distance the tobacco industry from the policy-making process to the extent that this is permissible. Do not grant the industry special consideration, but do ensure that it is consulted with as required – for example, by providing public meetings, timely information and the ability to submit industry views – while being aware of potential procedural manipulation (case studies 1, 2 and 3). 2. Excise tax is generally safe from challenges that claim it is confiscation or expropriation under domestic or international law (case studies 4 and 5). But express limits on taxation can be found in other laws or a country’s constitution or in the limits of the power to tax granted to an authority (case studies 6 and 7). 3. Explicit and de facto discrimination against foreign tobacco products or investors must be avoided in the design, implementation or enforcement of tax measures (case studies 8, 9, 10 and 11). Legal issues may arise not from the tax measure itself, but rather from ancillary measures that support its implementation (case studies 9 and 10). 4. Explicit differentiation between products based on their effect on health may be challenged as discrimination if it falls heaviest on imported products and has to be justified on the basis of evidence of impact on health and a lack of alternatives. 5. Investment incentives in the form of inducements or contractual undertak- ings should not be offered, as these may be binding (case study 12) or may ground a challenge under an IIA; they are also contrary to the WHO FCTC Article 5.3 Guidelines. 212 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.3 SCARE TACTIC A: ANTI-POOR RHETORIC (REGRESSIVITY) 4.3.1 INTRODUCTION In their efforts to lobby against tax increases, the tobacco industry and its affiliates often claim that increases in tobacco taxation will hurt the poor (81–82). This argu- ment is based on the concept of regressivity in relation to taxation. Conceptually, a tax can be regressive if it means that lower-income people pay a greater proportion of their household income to meet the tax burden than do wealthy people. In other words, the tax burden tends to be relatively higher for lower-income households than for middle- and high-income households. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use. Second, higher tobacco taxes and prices can induce behaviour change among the population, as reflected in the price elasticity of demand (83–84). In combination, these broader considerations effectively make tobacco taxation a progressive – rather than regressive – public health intervention. 4.3.2 REGRESSIVITY AND THE BROADER PERSPECTIVE In a narrow sense, tobacco taxation can be seen as regressive because lower-income people must allocate a relatively greater proportion of their household income than wealthy people to pay for tobacco products when those products become more expensive following a tax increase. In many countries, people from lower-income groups use tobacco more than other people (85). A systematic literature review by WHO found a robust association between lower income and a higher prevalence of current smoking among adults, both men and women (86). This finding was consistent across three decades of studies, across most geographic regions and across countries of different income classifications. For example, in India, high rates of tobacco use – i.e. use by more than 30% of the adult population – are found only in lower-income states such as Assam and Odisha, where net state domestic product is still below 100 000 rupees per capita (see Fig. 4.3.1) (87). CHAP T ER 4. PO LI T I C AL ECO N OMY 213 Fig. 4.3.1 Relationship between adult tobacco use and net state domestic product per capita in states and union territories of India, 2016–2017 Source: (87). However, this finding does not account for broader health and economic factors that determine the full impact on households. Tobacco taxation can in fact be viewed as a progressive – or pro-poor – policy when these wider considerations are properly ac- counted for and explained. In terms of health concerns, the relatively high use of tobac- co among low-income populations translates into a much greater burden of tobacco- attributable diseases for these populations, including higher morbidity and mortality. Low-income groups are also less able to afford medical care to treat tobacco- attributable diseases, and large out-of-pocket medical expenditures can further impoverish many families. Consequently, many poor individuals do not get or even seek the medical care they need. One study found that in Bangladesh, 55% of patients diagnosed with a tobacco-attributable illness did not seek further medical care. This lack of health care utilization was attributed in part to prohibitively high out-of-pocket treatment costs (88). The combination of high rates of tobacco use and lack of access to affordable medical care means that tobacco use measurably contributes to the poverty rate in a number of high-tobacco-burden countries, including China and India (89–90). A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India 214 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The poor are also known to be more price-sensitive than the wealthy; lower-income smokers exhibit higher price elasticities than their higher-income counterparts. This is demonstrated in recent studies by the World Bank, findings of which are shown in Fig. 4.3.2 (91). The poor respond more strongly to higher tobacco taxes and prices by reducing their use of tobacco products more than others, and thus they benefit disproportionately in terms of avoiding tobacco-related deaths, diseases and associated medical costs. A similar conclusion was drawn in a systematic review of the population impact of tobacco control policies on socioeconomic inequities in high-income countries at the late stage of the tobacco epidemic (92). The review found 16 relevant studies relating to taxation, only one of which found a regressive association between tax and the social economic gradient (seven found a progressive impact, while the others produced mixed results). Fig. 4.3.2 Price elasticity of tobacco consumption, medium estimate, by decile Source: (91). This wider economic perspective is explained in the World Bank’s Extended Cost- Benefit Analysis (ECBA) framework, which assesses the distributional impact of tobacco tax increases on health, among other factors (82, 83). That is, the ECBA framework looks beyond the simple or partial definition of regressivity (i.e. impact El as tic it y Deciles -1.2 1 2 3 4 5 6 7 8 9 10 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina CHAP T ER 4. PO LI T I C AL ECO N OMY 215 on household expenditure by income levels) to capture the full distribution of benefits, including improved health and income. The ECBA framework has been applied in studies of various countries, including Bangladesh, Bosnia and Herzegovina, Chile, Indonesia, Republic of Moldova, South Africa, the Russian Federation, Ukraine and Viet Nam. The evidence from these studies supports the view that effective tobacco tax policies can generate pro-poor and welfare-improving outcomes. When reductions in medical expenditures and additional years of working life that result from lower smoking-related mortality are taken into account, the overall policy of tobacco tax increases becomes progressive rather than regressive (see Fig. 4.3.3) (84). A similar conclusion has been reached in studies of high-income countries, such as the United States, where a tobacco tax increase was enacted in 2009 (93). Fig. 4.3.3 Impact of a 100% price increase, with medium elasticities, by deciles Source: (91). Tobacco tax increases will also often lead wealthier smokers to contribute relatively more than poorer smokers to the overall amount of tax revenue collected. This is because poorer smokers reduce their consumption the most, since they are more price- sensitive and wealthier smokers also tend to purchase premium (higher-priced and In co m e ga in s (% ) Deciles -1 2 3 54 6 7 8 9 10 0 1 2 3 4 5 1 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina 216 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N taxed) tobacco products (94). Hence, higher tobacco taxes can be seen as progressive in terms of additional revenue collection and health gains resulting from cessation, as well as from preventing the young from taking up smoking in the first place. One study from China suggests that a 50% tax increase would raise US$ 703 billion over 50 years, with just 14% of this increase being borne by smokers in the lowest income quintile (95). In addition, the tax increase would yield a savings of US$ 24 billion in expenditures on tobacco-related diseases, with about 28% of these savings being enjoyed by smokers in the lowest income quintile. The ECBA framework actually presents a rather conservative profile of the net benefits of raising tobacco taxes, since it does not include other sources of gain, such as reduced harm from exposure to second-hand smoke, increased productivity and the potential for poor households to benefit from social programmes funded through increased tax revenues (96). Assessments of the distributive impact of the 2009 tobacco tax increase in the United States found that the overall progressivity of the increase was enhanced by the tax rev- enue being used to expand health insurance coverage for children of low- and middle- income families (97). Accounting for this expanded coverage added to the progressiv- ity of the overall legislative package, the bottom line being that the impacts are positive for lower-income quintiles and greatest, on average, for low-income households (93). Similarly, a large proportion of the tobacco tax revenues from the Philippines’ so-called Sin Tax Reform was used to subsidize universal health coverage (UHC) for poor and near-poor families. Globally, 37 countries are known to earmark some tobacco tax revenues for health programs, with many of these programs indirectly benefiting the poor and less-advantaged disproportionately more than other groups (27) (for details on earmarking, see section 4.6). 4.3.3 CONCLUSIONS Contrary to the perception of tobacco taxation being regressive, it is a strong pro- poor policy when the broader economic impacts are taken into consideration. The tax burden is not a complete indicator of regressivity, since it does not include the negative health and economic impacts of tobacco-attributable diseases or the positive impacts of behaviour change in response to tax and price increases. The health and economic burdens of tobacco-attributable diseases fall dispropor- tionately on the poor, who tend to have higher tobacco use and are also the least able to afford the necessary medical care. Because the poor tend to be more price-sensitive, they curtail their use and consumption more significantly than wealthier smokers in response to tax increases, which in turn reduces their downstream health and economic costs. Tobacco taxation can be made even more progressive by earmarking or allocating tobacco tax revenues for social goods and services that benefit the poor (see section 4.6). CHAP T ER 4. PO LI T I C AL ECO N OMY 217 4.4 SCARE TACTIC R: REVENUE REDUCTION 4.4.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases result in reduced tax revenues for the government. According to them, the reduction in revenues is caused either by substitution to cheaper, lower-taxed or smuggled tobacco products or by reductions in consumption overall (98–99). The tobacco industry often refers to the Laffer curve to make this argument. According to this curve, revenues increase along with tax rates up to a certain point, after which further increasing tax rates leads to declining revenues. When considering tobacco taxes, the tobacco industry assumes that countries are already approaching or are even beyond the critical tax rate level (98). However, the argument rests on a narrow theoretical and empirically unsubstanti- ated foundation (98–100). The price inelastic demand for tobacco and the relatively low tax share in prices in many countries explain the win-win for public health and finance, i.e. that declines in consumption and increases in revenues can occur simultaneously (98, 101). Furthermore, many country examples (see case studies below) demonstrate that well-designed and well-implemented tobacco tax increases lead to increases in revenue, at least in the short to medium term (98, 100). Although consumption will diminish with a tobacco tax increase, the percentage increase in excise tax per unit is greater than the percentage decrease in tobacco consumption, cancelling out at least some of the effect of reduced consumption on revenue (98–99). A change in the tax rate, with all other factors influencing consumption kept constant, corresponds to a change in the tax revenue and is represented by a move- ment along the Laffer curve. As the tax rate changes, so does the elasticity of the tax base; each point on the Laffer curve corresponds to a different tax base elasticity. When one or more of the other factors changes, this affects the position of the curve, and the tax base elasticity changes at a given tax rate. For example, a successful smoke-free policy or advertising ban that reduces the demand for tobacco shifts the curve down, reducing the tax revenue potential for each tax rate. To demonstrate that few, if any, countries are beyond the revenue-maximizing point on the Laffer curve, Table 4.7 shows the revenue impact of increasing excise taxes under different scenarios, using different price elasticities of demand, different levels of tax increases and different starting tax shares, depending on country income levels. This is the tax base elasticity approach from which the Laffer curve is derived (for more details, see section 2.2.3 and Annex 2.2). The total and excise tax shares shown are weighted averages for each country income group, calculated from the RGTE dataset. The revenue gains were simulated using progressive levels of excise tax increases (25%, 50%, 75% and 100%) and varying price elasticities of demand (-0.4 to -1.2). 218 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.7 Percent increase in excise revenues under different scenarios of tax levels, tax increases and price elasticities20 Income group Total tax as % of retail price Excise tax as % of retail price Increase in excise tax Increase in excise revenue when price elasticity of demand is: -0.4 -0.6 -0.8 -1.0 -1.2 LOW INCOME 38% 22% 25% 22% 20% 19% 17% 16% 50% 43% 39% 36% 33% 29% 75% 63% 57% 52% 46% 41% 100% 82% 74% 66% 59% 51% MIDDLE INCOME 58% 41% 25% 19% 17% 14% 11% 9% 50% 37% 31% 26% 20% 15% 75% 54% 45% 36% 27% 19% 100% 71% 57% 45% 34% 23% HIGH INCOME 68% 55% 25% 18% 15% 11% 8% 5% 50% 35% 27% 21% 14% 8% 75% 50% 39% 29% 19% 10% 100% 65% 50% 36% 23% 11% Source: Authors’ calculations using data from the RGTE (27).21 Substantial revenue increases occurred in all the scenarios that were considered in the simulation. These results reaffirm much of what is already known, i.e. that higher tax increases generate higher revenue gains, and that these gains increase with the increasing inelasticity of demand. Even when demand is relatively price elastic (-1.2), the simulation predicts a gain in revenue. The tax share in price also affects revenue potential. The lower the tax share in price, the larger the revenue potential. This suggests that revenue reductions as a result of an excise tax increase will occur only if the scenario is extreme (i.e. a very elastic demand coupled with a very high current tax share). It is important to note that the vast empirical literature 20 These projections use 2018 data from 185 countries. The countries were classified according to World Bank income group, with the average total tax share, excise tax share and VAT/sales tax share for each country weighted according to the number of current adult cigarette smokers. To calculate the projected revenue for each stated elasticity, it was assumed that there would be full pass-through of the excise tax increase, along with constant percentages of non-excise taxes (VAT/sales tax) as a share of the retail price. The consequent changes in price were multiplied against the respective elasticities to derive the expected change in consumption. The projected revenues could be easily computed by multiplying the new consumption figures against the increased excise tax rates. 21 These calculations do not take into account brand substitution (cross-price elasticities), income ef- fects or illicit trade. The excise tax was assumed to be a specific tax, while the non-excise taxes (VAT and others) were bundled and treated as an ad valorem tax with retail price as the tax base. The difference between retail price minus all taxes was also assumed to be constant, with full pass-through of the tax increase to consumers. CHAP T ER 4. PO LI T I C AL ECO N OMY 219 shows tobacco to be universally inelastic; thus the extreme scenario should not be given credence by policy-makers. Furthermore, as demonstrated by the data in Table 4.7, tax shares in most countries are relatively low and reinforce the revenue potential of tobacco tax increases. The revenue potential of tobacco taxes is indeed quite significant. It is estimated that in 2018, excise taxes on cigarettes generated a total of US$ 361 billion in revenues worldwide, including US$ 162 billion in LMICs. If all countries were to raise excise rates by the equivalent of US$ 1 per pack of cigarettes, the amount of excise revenue would increase by between US$ 178 billion and US$ 219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from these tax increases, with excise revenues in these countries increasing by US$ 133 billion to US$ 167 billion, or by 82–103%.22 Revenue reduction in the countries examined was due to other causes, not the tax increase per se. For example, Tonga significantly increased its excise tax on cigarettes in 2016 and saw a very sharp decrease in its consumption (40% decrease), followed by a revenue decrease. This occurred because 20% of smokers switched to an untaxed, cheap local loose tobacco product called Tapaka Tonga (102). The lesson learned was that Tonga needed to tax all its tobacco products at the same level to avoid substitution to lower-price/untaxed tobacco products. Another example of revenue decrease that was not related to tax increases but rather to tax administration mismanagement is the case of South Africa (see explanation in the case study later in this section). Finally, declines in revenue due to long-term declining trends in tobacco use should not be confused with being beyond the revenue-maximizing point of the Laffer curve. For example, in the United Kingdom, where long-term declines in tobacco use are being experienced, a nominal decline in revenues occurred between 2017 and 2018 even though excise taxes remained unchanged. Conversely, even countries with very high tobacco excise rates experience increases in revenues as a result of tobacco tax increases (see the case study of Australia below) (98, 100). This suggests that few countries, if any, are beyond the revenue-maximizing point on the Laffer curve. Tobacco consumption is expected to be tax inelastic, even if demand becomes effectively price elastic as a result of successful tobacco control interventions. Taxation serves as an instrument for both fiscal and public health objectives. If after successful tobacco control interventions, prices reach levels where demand is elastic, the tax base is still most likely to be inelastic due to tax undershifting, since overshifting is not a good pricing policy when demand is elastic (for a more detailed discussion on the shifting of tax, see section 2.2.2). In other words, a tax rate increase in combination 22 Goodchild M, Perucic AM, Paul J. Tobacco taxation as a strategy to achieve global targets for smoking prevalence. Unpublished manuscript. October 2020. 220 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N with non-price tobacco control measures, which make consumers more sensitive to price (tax) increases, leads to declining but still positive marginal revenues. In the long run, tobacco control policies, including price and tax measures, may be so successful in reducing consumption that revenues will plateau or fall. This is ultimately the long-term policy goal. Reducing the impact of the policies and ending the global tobacco epidemic is the aim of tobacco control and not something to be avoided. However, ending the global tobacco epidemic is unfortunately not foreseen in the short to medium term; therefore, governments can currently rely upon tobacco taxes as a reliable source of revenue (103). 4.4.2 THE REVENUE IMPACT OF EXCISE TAX INCREASES: CASE STUDIES The following case studies illustrate four key points: (1) large and regular tax increases result in large and consistent revenue increases; (2) countries with high taxes and falling prevalence of tobacco use can still increase revenue with tax increases; (3) countries that reduce taxes experience revenue declines; and (4) countries that increase taxes in the face of illicit trade still increase revenue. Large and regular tax increases usually mean large and consistent revenue increases South Africa’s experience shows how successive tax increases, well above inflation and year after year, generate additional revenues even after taxes have been increased substantially. After two decades of declining real revenue in the 1970s and 1980s as real excise per pack declined, South Africa implemented successive excise tax increases from 1994 until 2011 (Fig. 4.4.1) (98, 104). After adjusting for inflation, this resulted in a real excise tax revenue increase of 245% (98). Revenues began to plateau from 2012 as tax increases stalled. They began to decline after 2015 – not due to tax increases, however, but due to a dramatic decline in administrative capacity and enforcement measures exacerbated by large-scale corruption in the government, including the tax administration authority (105). The rapid and catastrophic decline in tax administration and enforcement has been the subject of much attention (106). CHAP T ER 4. PO LI T I C AL ECO N OMY 221 Fig. 4.4.1 Real excise tax per pack of cigarettes and real excise tax revenue in South Africa, 1961–2020 Source: Data shared by University of Cape Town, 2020. Similarly, the Philippines provides a compelling example of how large and regular tax increases alongside reforms to tax structure can lead to large and consistent revenue increases – in this case, also through an accompanying reform to the tax structure (Fig. 4.4.2) (98). The 2012 Sin Tax Law consolidated the country’s four tax tiers into two by 2013 and established a uniform structure by 2017. The same law provided for large, progressive increases across the board, but in particular for the lowest tax categories (98). Not only were the revenue gains substantial, they exceeded all the projections for 2013–2017 made prior to the law’s passage (98). Excise tax per pack Excise tax revenue Ra nd s pe r p ac k (c on st an t 2 02 0 ra nd s) Excise revenue (constant 2020 rands) 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11 20 13 20 15 20 17 20 19 0 0 5 6 10 10 15 14 20 18 222 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.2 Real excise tax per pack of cigarettes (multitier, two-tier, unitary) and total tobacco real excise tax revenue in the Philippines, 2012–2018 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2012 as the base year. Sources: (107, 108 and data shared by the Philippines Department of Finance, September 2020). Ukraine is another example of a country that has regularly increased taxes over the past 10 years and has experienced increased revenues along with decreases in consumption and the number of smokers. Figure 4.4.3 the shows the trends in excise tax, revenues, cigarette sales and number of smokers in 2008–2017. Increases in excise rates were consistently accompanied by increases in revenues. In 2014–2015, excise tax was not increased above inflation (and inflation, especially in 2015, was very high, at 48.7%), so real values of excise and revenues went down. But it is evident from the data that revenues closely follow the path of excise levels even when sales go down. 12 Re al e xc is e ta x pe r p ac k, P hi lip pi ne p es os (2 01 2 ba se ) Billions (Philippine pesos) in tobacco real excise tax revenue (2012 base) 00 20 60 100 140 2012 HIGH PREMIUM SIN TAX LAW RA 10354 TRAIN LAW RA 10963 UNITARY RATE 2013 2014 2015 2016 2017 2018 5 10 15 20 25 30 32 .1 9 68 .6 6 71 .0 9 93 .5 2 87 .2 0 96 .0 1 11 4. 39 24.4 25.4 26.2 26.8 26.9 27.5 11.7 16 23.1 19.6 LOW Real tobacco excise revenues Dierent levels of real tobacco excise tax per pack MEDIUM 7.6 2.7 28.3 CHAP T ER 4. PO LI T I C AL ECO N OMY 223 Fig. 4.4.3 Average real cigarette excise tax rates, real cigarettes excise tax revenues (base year 2008) and cigarette sales and number of cigarette smokers in Ukraine, 2008–2017 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2008 as the base year. Source: Data provided by Konstantin Krasovsky, July 2020. Countries with high tax and falling prevalence of tobacco use can still increase revenue with tax increases Countries with already high tobacco taxes and rapidly diminishing tobacco use can still increase revenue by increasing taxes (98, 109). Australia has implemented comprehensive tobacco control policies and enacted consistent tobacco tax increases on top of what were already some of the highest tax rates in the world (see Fig. 4.4.4). Between 2001 and 2010, revenue increased with increasing tax rates, but in real terms (inflation-adjusted) it remained static (109). Then, in 2010, a 25% excise tax increase was introduced, with large annual increases scheduled from 2013 onward (98, 109). The result of this tax policy has been consistent and large increases in revenue year after year for nearly a decade, even when the increases were being made on already high tax rates.23 23 The apparent reduction in revenues in 2012 and 2013 was due to a change in the source of the data for 2001–2011 and 2012–2016. Data for 2012 and 2013 do not include customs duty, while all other years do. 73 95 125 112 88 82 75 76 67 11.8 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Number of daily smokers, in millions Real tobacco excise revenue, in billions of Ukrainian hryvnia Real average excise per cigarette packs of 20, Ukrainian hryvnia Number of taxed cigarettes (sales), in billions of sticks 0.6 3.6 10.1 9.2 8.7 8.6 8.4 8.1 7.3 6.2 6.5 6.3 7.8 10.3 11.2 12.1 13 1.4 2.2 2.5 2.9 3.5 3.2 2.7 3.3 4 12.7 13.4 74 9.7 224 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.4 Real excise tax and customs duty per stick and real total revenue (all tobacco products) in Australia, 2001–2019 Notes: Rates published by Australian Taxation Office and Australia Department of Immigration and Border Protection, adjusted using Australian Bureau of Statistics Consumer Price Index rates. The 2011–2012 and 2012–2013 figures do not include customs duty, which explains the apparent decline in revenue. Using official disclosures, Scollo and Bayly estimate that duties in these years were $7397.2 and $7687.2 respectively (110). Sources: (109, 110). Countries that reduced taxes and saw revenues decline Prior to 1982, Canada lowered taxes on cigarettes and experienced declining revenues as well as increased smoking – particularly among youth. Subsequent fivefold in- creases in cigarette taxes between 1982 and 1992 resulted in more revenue, increases in retail price and substantial reductions in consumption, with teenage smoking declining by nearly two thirds (5). In the early 1990s, a growing illicit trade in ciga- rettes emerged in which Canadian cigarettes exported to the United States were then smuggled back into Canada (5). The tobacco industry – which was later found to be complicit in and profiting from this illicit trade – sought to frame Canada’s high tax rates as the cause of smuggling (111–112) and succeeded in convincing the federal government, as well as six provincial governments, to make massive reductions in the tobacco tax (111–112). As a result, federal tax revenues fell significantly – more than twice as much as the government had predicted – and smoking rates among both adults and youth began to increase (5, 112). The Canadian government later changed its strategy, and the federal excise tax was restored, resulting in increased Specic excise/ duty per stick (cigarettes/cigars less than 0.8 g) Total revenue (all tobacco products) To ta l c us to m s/ du ty re ve nu e (in a tio n, a dj us te d, 20 19 A us tr al ia n do lla rs , m ill io ns ) Total excise/duty per stick (ination, adjusted, 2019 A ustralian dollars) 0 3 000 6 000 9 000 12 000 $0.8 $0.6 $0.4 $0.2 $0 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 CHAP T ER 4. PO LI T I C AL ECO N OMY 225 revenues and decreased smoking (5). Canada’s focus then shifted to using customs enforcement, rather than tax rates, as the best means of countering illicit trade (5). Countries that increased taxes in the face of illicit trade and still increased revenue As discussed in section 4.1, the tobacco industry exploits illicit trade as a strategy to undermine tobacco tax policy, with the goal of deterring governments from increasing tobacco taxes. The narrative that has been created is that higher tobacco tax rates result in increased illicit trade and undermine the policy goals by resulting in lower (or no) declines in tobacco use or lower (or no) increases or even decreases in revenue. However, as shown in section 4.1, the empirical evidence does not sup- port the industry arguments. Furthermore, the evidence shows that the industry and its allies have consistently overstated and exaggerated the scale and extent of illicit trade (see section 4.1). As was the case in Canada, Brazil’s tobacco tax policy suffered from a fear that the illicit market would expand unless it was undercut by price competition in the legal market, which it was thought could be best encouraged through tax cuts (20). Real excise tax rates declined from 1999 until the mid 2000s, as nominal increases were below the rate of inflation. This resulted in declines in real tax revenues (20). In these years, the tobacco industry used the tax cuts to increase profit margins rather than decrease prices and outcompete the illicit market, while also exaggerating the size and scope of the illicit trade problem (20). This caused the industry’s argument on illicit trade and revenue to lose credibility and resulted in increases in tax rates from 2007 onwards, with a major reform passed in 2011 (20). Tobacco excise rates and minimum prices were scheduled by the law to increase at levels above expected inflation from 2011 until 2015 (20). This resulted in substantial increases in the tobacco excise per pack, as well as overall revenue, which by 2015 had more than doubled from its low point in 2013 – equating to more than 50% in real terms (see Fig. 4.4.5). The success of this reform shows that revenues can be increased by higher rates despite the presence of a sizeable illicit market (113). More recent data show that revenues in Brazil declined in 2015 and 2016, coinciding with an increase in illicit trade, but also with an exceptionally bad economic recession that saw GDP decline by more than 3% in those years. 226 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.5 Average real excise tax per pack, real excise tax revenues and domestic cigarette sales in Brazil, 1999–2014 Notes: Data based on domestic sales and tobacco excise revenues, Federal Revenue Secretariat, indexed to 2013 Brazilian reals, using Consumer Price Index. Revenue collection indexed to 2013 reals, using Consumer Price Index. Source: (20). Improvements in tax administration and enforcement can also generate increases in revenues. In Kenya, several measures, including fiscal markings and, later, an advanced tracking and tracing system, improved collection, resulting in increases in both legal sales and tax revenues and a reduction in illicit sales (114). Moreover, these examples of poor governance indicate that attention should be focused on countries where a significant loss in administrative and enforcement capacity un- dermined revenue collection. 4.4.3 CONCLUSIONS The tobacco industry uses revenue concerns as a SCARE tactic to avoid, dilute and/or delay tobacco tax increases. The argument that higher taxes will decrease revenue is theoretically plausible, but real-world examples have demonstrated that this has not occurred. Furthermore, simulations show that even large tax increases in current average tax shares yield substantial revenue gains. The use of the Laffer curve by the tobacco industry should be challenged and refuted. The relatively price inelastic nature of cigarette demand combined with the Real excise tax amount per pack Domestic sales Real excise tax revenue D om es tic s al es (b ill io n pa ck s) Re al e xc is e ta x re ve nu es (b ill io n 20 13 re ai s) Real excise tax am ount per pack (2013 reais) 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 0 1 2 3 4 5 6 0 1 2 3 CHAP T ER 4. PO LI T I C AL ECO N OMY 227 low tax share and no overshifting of the tax means that most – if not all – countries are still far from the revenue-maximizing point, indicating that increases in taxes will lead to increases in revenues. The case studies in this section refute each of the tobacco industry’s arguments regarding alleged potential revenue loss due to tax increases. The experiences of South Africa, the Philippines and Ukraine demonstrate that large and regular tax increases result in large and consistent revenue increases. Well-designed tax structures have also proven to play an important role in generating revenues. The experience of Australia shows that even countries with already high tax rates and declining prevalence of tobacco use can increase revenues with regular, large tax increases. The experience of Canada warns against following the advice of the tobacco industry to decrease taxes as a way to fight illicit trade. It demonstrates clearly that decreasing tobacco taxes will decrease revenue and encourage consumption, rather than counteract illicit trade. The experience of Brazil shows that countries with substantial illicit trade issues can still increase revenue by increasing taxes. Finally, in the few cases where revenue decreases were seen, the reasons for the decreases were not strictly linked to tax increases. This was the case in Tonga, where the increase in tax was applied only to cigarettes and not to their close substitute, loose tobacco – leading smokers to switch products. In South Africa, a decrease in revenue was the result of the weakening of government institutions. And in Ukraine, real revenues decreased only during the two years when taxes were not increased above inflation. 228 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.5 SCARE TACTIC E: EMPLOYMENT 4.5.1 INTRODUCTION In opposing tax increases, the tobacco industry often seeks to frame tobacco taxes as an economic rather than a public health issue (5, 48, 53, 115). Particular emphasis is placed on the alleged threat that tax increases pose to employment in tobacco farm- ing and manufacturing, as well as other related industries (5). This so-called choice between health and jobs is, however, based on several false premises, including (5): 1. tobacco is a significant source of jobs within the context of broader labour markets, and domestic tobacco tax increases will have a drastic effect on domestic employment (48, 53, 116–117); 2. tobacco consumption is an indispensable engine for job creation (5, 48, 54); and 3. tobacco provides highly prosperous, sustainable and irreplaceable livelihoods (5, 53, 118). In reality, the relationship between tobacco taxation and employment is consider- ably more complex than the industry makes it out to be. In fact, there is ample evidence to show that tobacco taxes are a win-win for public health and the fiscal space, without measurable risks to employment. 4.5.2 THE LINK BETWEEN TOBACCO EMPLOYMENT AND TOBACCO TAX RATES Tobacco farming, production and manufacturing (including hand-rolling in some countries, most of them in South-East Asia) constitute a small proportion of the labour force, even in countries where the industry is most heavily concentrated (5, 48, 103, 116). Employment in tobacco farming and manufacturing has been declining globally due to advances in technology, trade liberalization, market consolidation and the privatization of formerly state-owned tobacco companies (5, 103, 119). These same trends have led to the heavy concentration of tobacco growing and manufacturing in only a handful of countries – and within these countries, often in only a small number of regions (103, 119–121). Even in those countries that lead in tobacco growing and manufacturing, tobacco’s overall share of total agricultural and manufacturing employment is relatively small and is often decreasing as efficiencies in production reduce labour intensity (5, 117–118, 122). Similarly, the industry’s claim that tobacco taxes reduce employment is exag- gerated and typically overlooks wider trends driving tobacco industry employment. Indeed, tobacco industry developments and innovations have played a greater role in the reduction of employment in the tobacco industry than have tobacco control policies (103). Despite industry claims that tobacco taxes can affect employment, CHAP T ER 4. PO LI T I C AL ECO N OMY 229 characteristics of the location of production – such as market size, labour costs, growing conditions and leaf preferences – have much more to do with tobacco industry interests than with the tobacco tax rate (103, 123). Moreover, jobs in countries that produce tobacco primarily for export are not greatly affected by reductions in local consumption resulting from tax increases (5, 103, 116, 123). Finally, it has been demonstrated that tobacco tax increases do not have a significant effect on employment in the retail sector, as most retail businesses sell other goods (103). Estimates of the gross employment impact of tobacco tax hikes demonstrate that job losses that do occur can be more than compensated for by increases in revenue. A 2018 World Bank study estimated that in Indonesia, for example, an ambitious tax reform that would simplify tiers and increase prices by close to 50% would reduce gross employment in the tobacco manufacturing sector by less than 0.5% (a loss of 2 914 jobs). The government could provide income support to the displaced workers (for example, through training, temporary transport/mobility or income support) with less than 2% of the revenue gained from the tax increase (117). Similarly, a 2019 study by Bangladesh’s National Board of Revenue estimated that a substantial increase in tobacco taxation would cause 7 012 lost jobs, but that the total income associated with these job losses in the bidi industry would amount to only 3.5% of the revenue gained (120). Accordingly, increased revenue can more than compensate for the expenditure of supporting those who lose jobs and need to acquire new skills before transitioning to new employment (120). Box 4.5.1 Employment fears deployed to frustrate tobacco tax reform in Indonesia In 2017, the Indonesian Ministry of Finance decided to implement tobacco tax in- creases by 2019 and tier simplification by 2021 (124). This resolution was, however, abandoned within a year, after a concerted campaign by tobacco industry actors and their allies to reframe the increase as an economic issue with a focus on, among other things, the effect the tobacco tax increase would have on employment (124). This defeat for the tobacco tax initiative came despite estimations of how the loss of income associated with lost jobs would be dwarfed by the additional revenue gained by the tax (117). Earlier analysis had estimated an overall large net positive impact on employment from tobacco tax increases (125), which illustrates how evidence that challenges assumptions around the negative socioeconomic impacts of tobacco control tends to be discounted (126). In this case, tobacco industry arguments seem to have resonated strongly with politicians from the electoral districts of West Java, East Java, Central Java and West Nusa Tenggara, where employment in tobacco farming and manufacturing is concentrated (124). Although tobacco manufacturing 230 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N represented only 5.13% of total manufacturing employment, the concentration of the job and economic activity meant that arguments regarding employment were particularly salient (117,119). Accordingly, tobacco industry arguments that tobacco taxation would negatively impact employment and farmer livelihoods prevailed, despite strong opposing evidence (119). Concentration of tobacco industry activity within countries poses particular obstacles to overcoming industry arguments on employment and needs to be given careful attention. 4.5.3 THE EVIDENCE ON THE NET EFFECT OF TOBACCO TAX INCREASES ON EMPLOYMENT A proper analysis of the effect of tobacco tax increases on employment must examine their impact on net or economywide employment. Decreases in expenditures on tobacco associated with tobacco control do not mean that expenditures simply disappear; rather, they are redistributed towards consumption of other goods and services, thereby generating employment elsewhere in the economy (5, 103, 123). Similarly, though the effect of higher tobacco taxes on net consumption is arguably more ambiguous, revenues from this intervention do generate spending, invest- ment and employment in public services such as health and education (5). Tobacco control polices usually have a marginal neutral or positive effect on net employment, particularly in countries that are net importers of raw or manufactured tobacco products, as expenditures on these imported items tend to flow out of the country (5). Export-oriented tobacco producers are less sensitive to local demand and are not significantly affected by domestic tobacco tax measures, which likely have a near-neutral net impact (5). In some cases, the net employment impact is a very small negative number, typically less than 1% (127–128). A recent study estimated that in the United Republic of Tanzania – a large tobacco-producing and exporting country – a 30% reduction in smoking prevalence would result in a net employment decline of just 0.5% across the economy as a whole (129). A similar study of Pakistan found that, with some variance depending on where spending was redistributed from tobacco consumption, the overall net effect on employment from a significant reduction in expenditure on cigarette employment – 1 billion rupees – would be a gain of between 6 651 and 5 803 jobs (122). This increase would occur because expenditure on cigarettes produces much less employment in the broader economy than expenditure on food and education (122). In the United Republic of Tanzania, as elsewhere, increased revenue could be used to assist those who lose employment with transitioning to new livelihoods. CHAP T ER 4. PO LI T I C AL ECO N OMY 231 4.5.4 THE VIABILITY OF BETTER LIVELIHOODS In arguing against tobacco tax increases, the tobacco industry advances the myth that people employed in tobacco production – particularly tobacco farming, but also manufacturing – lack any other prospect for a comparably attractive livelihood. However, studies based on extensive survey data in Indonesia, Kenya, Malawi, the Philippines and Zambia have shown that despite needing to commit significant amounts of labour to their crop, tobacco farmers often suffer losses rather than gain profits (119, 121, 123, 130–132). Furthermore, the Indonesian studies demonstrate that tobacco farming has a negative impact on household income and opportunity compared with the experience of other farming households that have given it up (119). Declines in consumption as a result of tobacco tax increases are gradual and susceptible to the same progressive adaptation that has occurred for decades (5, 103). While there will be a need in some countries for the government to help farmers transition to other crops or industries in the longer term, this process will not be a major short-term shock to employment or the wider economy (123). Because tobacco growing and manufacturing can be concentrated in just a few locations within a country, job losses within the tobacco industry might have a disproportionate effect in one location, while employment gains from reduced consumption may be spread across the whole country (120, 123). A study of the employment effects of tobacco tax increases in Bangladesh estimated that up to 60% of all job losses would occur in only two districts – among the poorest in the country – due to the high level of industry concentration (120). Studying the need for support, as well as the means of delivery and funding of support, is particularly necessary in these circumstances. Beyond the need to ensure equity and support employment, a failure to provide for targeted relief can exacerbate fear of job losses and may prove fatal to a tobacco tax proposal (120). Box 4.5.2: Supporting alternative livelihoods in the Philippines The Philippines earmarked 15% of the revenue from a 2012 increase in tobacco taxes to supporting economically viable alternative livelihoods for tobacco farmers and workers (5). Tobacco farming in the Philippines is regionally concentrated, and the tobacco industry had previously been successful in deploying concern for smallholder tobacco farmers to undermine tobacco control measures (136). The provision of economic support was a politically effective countermeasure to tobacco industry SCARE tactics and eased the tax increase’s passage. Given the Philippines’ integration with global tobacco markets and demand, tobacco farmers have not been seriously affected by the tobacco tax increase and 232 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N reduced domestic demand (108, 132). Nevertheless, transfers to tobacco-growing regions from the earmarked tax have been substantial (108). These funds are support- ing gradual transitions to alternative livelihoods, with farmers being encouraged to take up alternative crops, as well as establishing infrastructure, such as market-to-farm roads, that will make these alternative crops more economically viable (108, 133–134). Supporting alternative livelihoods for farmers and other tobacco workers is important because it can offset the political effect of industry arguments, even though domestic tax increases usually have only a modest and gradual effect on employment. There are various models for supporting alternative livelihoods when employment in the tobacco sector gradually diminishes due to decreases in either global or national demand. The Philippines is exemplary, but many other countries have either implemented or experimented with supporting crop transitions. Turkey’s alternative crop pro- gramme, implemented in anticipation of the privatization of the country’s cigarette monopoly, has proven effective in supporting many tobacco farmers’ move to other crops (135). Smaller-scale crop substitution projects in Kenya and Yunnan Province in China have shown how financial, regulatory and infrastructure support from government can contribute to crop transitions (5, 53). Argentina, Bangladesh, Mexico and the state of Maryland in the United States provide additional case studies of how governments can support these transitions (5, 136). 4.5.5 CONCLUSIONS The tobacco industry exaggerates the importance of tobacco employment and over- states the impact that domestic demand reduction due to local taxes will have on tobacco farmers serving a global market. The industry also simplifies employment’s relationship with taxation by focusing only on gross employment in tobacco, which ignores the reality that expenditures on tobacco do not disappear but rather are redistributed for other consumption that can produce a similar or higher number of jobs. Many detailed studies have found that tobacco growing is much less profitable and sustainable than the tobacco industry claims. Tobacco farmers throughout the world have successfully transitioned to other crops, although the transition often requires temporary or additional support from the government or other stakeholders. The extent of such support is moderated by the reality that transition from tobacco to other crops is a long-term consideration. CHAP T ER 4. PO LI T I C AL ECO N OMY 233 4.6 EARMARKING TOBACCO TAX REVENUES TO FUND HEALTH 4.6.1 INTRODUCTION Earmarking tax revenues involves the separation of all or a portion of revenue from a tax or group of taxes to be put aside for a specific purpose (137). Globally, more than 80 countries earmark for health (138), and 37 earmark tobacco tax revenues for health (27). There are two main types of earmarks: hard – also called substantive – and soft, or symbolic (139). Hard earmarks link the expenditure with a revenue source in legislation. This can limit funding if the earmarked revenues are the main source of funding, or it can cause surpluses to accrue wastefully when more revenues are raised than may be expended for the earmarked purpose. Soft earmarks include dedicated funds or commitments to use funds for a particular purpose. They are not necessarily legally binding. For example, in France, the ma- jority of tobacco tax revenue is used to fund social security (which includes health insurance and health care), but there is no hard, formal earmark (140). Earmarks can also be some combination of hard and soft. In the Philippines, tobacco tax earmarks are legally binding, but earmarked revenues go to the general fund, and the Department of Health must submit an annual budget for covered programs as part of its budget request. Earmarking is a broad and contentious topic that goes beyond the specifics of tobacco tax earmarking. Discussions on the topic fall within the ambit of public financial management, and earmarking generally is not encouraged. From a tobacco control perspective, however, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. It is a tool to improve the political economy of tobacco taxation; it is a secondary issue only, after the primary goal of reducing demand for tobacco. One way to use earmarking to improve the political economy of tobacco taxation is to link the payment of tax by tobacco users to benefits they will receive through the funding of complementary tobacco control programmes, such as cessation support, or through increased funding for health programmes on which they will rely disproportionately. This is known as the benefit principle. Earmarking for tobacco control makes sense, as its financial cost is relatively small and tobacco tax reduces demand more effectively when implemented within a package of complementary tobacco control measures. Another way earmarking improves the political economy of tobacco taxation is by safeguarding against any perceived or potential negative ramifications of the tax itself. This is important for neutralizing erroneous but often convincing tobacco industry arguments against effective tobacco tax policies. For example, the Philippines earmarks the bulk of the additional revenues from sin taxes for the health insurance premiums of the poor. In addition, a portion 234 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the country’s tobacco tax revenues is earmarked to provide for the economic well-being of tobacco growers and tobacco growing regions, with the general aim of promoting economically viable alternatives to tobacco farming and manufacturing as a safeguard against the potential for reduced domestic tobacco demand (141). Tobacco tax earmarks are complex, however, and care is required when determin- ing whether a particular earmark is needed on the grounds of political economy and justified on the grounds of equity and economic efficiency. In assessing tobacco tax earmarks, many of the same criteria that have been used to assess the appropriateness of generic earmarks also apply. In the rest of this section, these criteria are set out and matched with reasons for the ability of well-designed tobacco tax earmarks to fulfil them. The types and structures of tobacco tax earmarking are explored alongside descriptions of country experiences to provide guidance on how tobacco tax earmarks are used, when they are justified and the best ways to design them. 4.6.2 CIRCUMSTANCES IN WHICH EARMARKS MAY BE SUITABLE Scepticism about earmarking is both long-standing and justified, but much of the debate concerns earmarking generally and is not specifically concerned with the merits of tobacco tax earmarking (138). The main concerns raised about earmark- ing are listed in Table 4.8, accompanied by suggestions for how earmarks may be structured to address these concerns. Table 4.8 Concerns about earmarking and suggested safeguards to avoid the concerns MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Democratic accountability and oversight: earmarks undermine democratic processes by impeding legislative and executive oversight over expenditure. Establishing proper oversight and accountability procedures is important to ensure funds are not mismanaged (138). Additionally, if a soft earmark structure, which transfers revenue to the general fund from which it is then allocated, is adopted, this will not be a concern. Budget rigidity: earmarking may create budget rigidity that can lead to inefficient allocation of resources (138). An earmark’s particular design determines how much rigidity is introduced (138). Flexible soft earmarks are less prone to introducing rigidity than hard earmarks. Concerns about rigidity can be reduced by the inclusion of a sunset clause that ensures that the earmark is automatically discontinued or reviewed after a set period of time has elapsed (138). A further safeguard is to establish the earmark as a waterfall account, with any excess revenue over a set amount being allocated to the general fund. CHAP T ER 4. PO LI T I C AL ECO N OMY 235 MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Fragmentation: earmarking can result in fragmented and uncoordinated expenditures. This means policies complementary to the earmarked purpose but outside of its purview may be unfunded (138, 142). This is a legitimate concern. The negatives of fragmentation cannot be entirely eliminated, but they may be outweighed by the other merits of tobacco tax earmarking. That said, proposals for tobacco tax earmarks should be scrutinized to ensure that the funded purpose is at least cost-effective. Decreased equity: equity will decrease if individual access to benefits is narrowly defined according to payments made. This issue is not likely to arise with tobacco tax earmarks but is conceivable and something that should be guarded against in an earmark’s design. Capture by special interests: because earmarks are often the result of political expediency, an earmarked purpose may be determined by powerful special interests promoting a tax’s passage rather than careful prioritization of resources (138). Well-designed earmarks will guarantee funding for underresourced programmes and high- priority programmes. While the above concerns may be valid and design does matter, tobacco tax and other health-promoting taxes are not subject to the same concerns when it comes to the justifiability of earmarking their revenue (138, 143). Some of the factors that distinguish tobacco tax earmarks from more general critiques of earmarking are listed in Table 4.9 (138). Table 4.9 Concerns about earmarking and distinguishing factor for tobacco tax earmarks GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Procyclicality: earmarked revenues are often procyclical and susceptible to booms and busts (138–139, 142). Tobacco tax revenues are generally not cyclical (they are recession-proof ), and revenue is predictable relative to most other indirect and direct taxes (103). Budget rigidity Tobacco tax earmarks necessarily involve only a relatively small proportion of the budget; therefore, the effect of any rigidity will be relatively insignificant. Partly because of the relatively small amounts involved, there is only limited real-world evidence of tobacco tax earmarks having introduced harmful rigidity (143).24 24 See also the subsection on the amount of money associated with tobacco tax earmarks in section 4.6.3. 236 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Capture by special interests All earmarks should be scrutinized to ensure that their funded purpose is cost-effective. In the case of tobacco tax earmarks, however, political economy considerations may mean that it is sufficient for a low-priority purpose to be funded if the funding will unlock the political will needed for effective tobacco tax increases. In these cases, special interests are being purposefully catered to in order to ensure that tobacco tax increases occur. Of course, arguments against tobacco taxes and tobacco tax earmarking are led by special interests such as the tobacco industry (144–147). Insufficient revenue: the earmarked revenue source may become insufficient for funding its purpose (139, 142). Even though revenue may decrease in the long term when more tobacco users quit, such decrease is expected to be gradual. See Section 4.4 for details on how, with effective design, revenue will generally increase even with declining consumption. In addition to these reasons why general concerns about earmarking do not fully apply to well-designed tobacco tax earmarks, there are a number of compelling reasons for tobacco tax earmarking to finance tobacco control or public health that argue in favour of its implementation: • Significant increases in excise taxes are the most effective, as well as the most cost-effective mechanism for reducing consumption, but they are best implemented as a part of a package of complementary tobacco control measures, such as the WHO MPOWER package. Earmarking tobacco tax revenue for interventions that may not be funded otherwise can strengthen overall tobacco demand reduction (148). • The political economy of tobacco tax increases also makes earmarks attractive: – People have been shown to be more supportive of tobacco tax increases when they know the revenues will be used for targeted social programmes (143, 149–150). Earmarking tax revenue for health or tobacco control frames tobacco tax as a public health intervention in the minds of the public, which may otherwise view it as merely a revenue source (138). Research has shown that using earmarking to link a tobacco tax to health can also help raise awareness about the dangers of tobacco use (143). – When tobacco tax revenue is earmarked for programmes that benefit vulnerable groups, the tax becomes more equity-enhancing. Although lower socioeconomic groups and young adults receive disproportionate health and economic benefits from tobacco tax increases over the medium term, these groups will expend a greater share of their income in the CHAP T ER 4. PO LI T I C AL ECO N OMY 237 short term because of tobacco taxes. Earmarking tobacco tax revenue for programmes such as UHC or cessation services that provide immediate benefits to these groups neutralizes some critiques of tobacco taxation (e.g. the 2009 United States federal excise tax increase and the 2012 Philippines Sin Tax Reform illustrate how equity-enhancing earmarking facilitated passage of substantial tax rises) (93, 138, 143, 150). 4.6.3 EARMARKING PRACTICES AND COUNTRY EXAMPLES Earmarking tax revenues for health is a common practice in 80 countries. In 2018, 37 countries from all regions of the world earmarked tobacco tax revenues for health purposes.25 Case studies in the political economy of tobacco tax earmarking In 2012, the Philippines comprehensively reformed tobacco and alcohol excise taxes. Tobacco taxes were increased significantly, and numerous tax tiers were reduced to only one tier by 2017. Although increasing revenue was a foremost motive for some officials, the reform was explicitly framed around boosting UHC funding and advancing public health by reducing alcohol and tobacco consumption. Earmark- ing of tax revenue for UHC was essential to the political compromise that made this trailblazing tax increase a reality. It ensured that the increase, which may have otherwise been perceived as regressive, was framed as a progressive public health measure in the public imagination, while also appeasing tobacco growers and their political representatives. Earmarking was also important because the earmark en- sured high-level support for the tax by achieving a key political priority (151). Its soft-earmark structure meant it was not a blank cheque to the Ministry of Health, and this addressed concerns within the Ministry of Finance. Similarly, in Australia, earmarking of revenue helped overcome community objections to tobacco taxes and tobacco control more generally that resulted from the tobacco industry’s sponsorship of sports and the arts in the 1980s. Attempts to completely ban tobacco advertising and sponsorship had been unsuccessful due to strong pressure from sports, arts and racing lobbies that claimed that a ban would harm these activities. States, starting with Victoria, responded by earmarking funding for Health Promotion Foundations that took over the tobacco industry’s sponsorship activities and also paid for antismoking campaigns. In 1997, these earmarks ended after a High Court ruling that the Constitution did not allow states to collect excise taxes. However, in recognition of the successful work of the Foundations, the federal 25 Details about earmarked taxes by country are provided at https://www.who.int/tobacco/global_report/ Table-9-4-Use-of-earmarked-tobacco-taxes.xls?ua=1. 238 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government began funding them directly from the federal budget (152). Although this example may be difficult to replicate precisely, it shows how earmarks with built-in sunset clauses for piloting cost-effective interventions can potentially graduate to funding from the general budget once they have proven their effectiveness. It also more generally shows how earmarks can disarm community objections, reframe tobacco tax increases and unlock the political will needed to advance effective tobacco control measures. Structures for managing earmarked tobacco taxes A 2016 review of nine countries’ tobacco tax revenue earmarking experiences identified three arrangements for governance and allocating revenue (151). Table 4.10 presents some examples of these allocation arrangements. In some countries, earmarked tobacco tax revenues are combined with alcohol tax revenues. Table 4.10 Illustrative arrangements for allocating earmarked tobacco tax revenues POSSIBILITIES FOR ALLOCATING TOBACCO TAX EARMARK REVENUE Forms of budget allocation Revenue goes to the general fund and is later assigned to the official actor(s) specified in the earmark. Revenues do not go through the general budget but are instead paid into a separate account belonging to the official actor(s) specified in the earmark. Earmarked tax revenue is paid directly to the account of the entity managing an autonomous or semi- autonomous fund. Examples In the Philippines, revenue goes to the general fund before being allocated to the Ministry of Health following submission of a budget for its use (140). In Romania, revenue goes directly into a Ministry of Health account that is distinct from the general fund. In Panama, revenue is paid into subaccounts of the three recipient agencies (the Ministry of Health, the National Cancer Institute and the Customs Authority). In Thailand, ThaiHealth directly receives the earmarked revenues in its own account. In Viet Nam, the Viet Nam Tobacco Control Fund receives the revenues directly into a subaccount it manages but that belongs to the Ministry of Health. Source: (151). Where is the money being spent? Earmarked tobacco tax revenues are used for a variety of health purposes, including tobacco control, health promotion and UHC. A wide variety of other programmes have also been funded with earmarks from tobacco taxes, including disaster relief (e.g. hospital medical supplies and equipment to treat COVID-19 in India), youth pro- grams, sports and craft jobs in Yemen, social cohesion in Morocco, health and social programs in areas dependent on tobacco growing in Argentina, health promotion CHAP T ER 4. PO LI T I C AL ECO N OMY 239 and tobacco control in Thailand and alternative livelihood programs for tobacco farmers as well as economic projects in tobacco-growing provinces in the Philippines. Table 4.11 shows the three main categories of health programmes to which ear- marked tobacco tax revenue is allocated, as well as a fourth miscellaneous category, with country-specific examples for each.26 Table 4.11 Programmes to which earmarked tobacco tax revenue is allocated TOBACCO CONTROL NCD PREVENTION AND CONTROL PROGRAMMES (otherwise indicated between brackets) HEALTH COVERAGE EXPANSION (e.g. through health insurance coverage) OTHER, MORE GENERAL OR UNSPECIFIED HEALTH PROGRAMMES Costa Rica, Côte d’Ivoire, the Islamic Republic of Iran, Madagascar, Panama (tobacco cessation and fighting illicit trade), Switzerland, Viet Nam Cook Islands, Costa Rica, Mauritania (anti-cancer research), Palau (NCD prevention only), Panama (National Institute of Oncology), Paraguay Colombia, Congo, Egypt, Palau, Philippines Algeria, Argentina, Bangladesh, Botswana, Cabo Verde, Chad (programmes delivering antiretroviral drugs), Colombia (sports), Comoros (sports, hospital emergencies), Congo (sports), Côte d’Ivoire (AIDS programme), El Salvador, Estonia (sports), Guatemala, Indonesia, the Islamic Republic of Iran (sports), Ireland, Jamaica, Lithuania (sports), Madagascar (sports), Morocco, Nepal, Paraguay (sports), Republic of Korea (health promotion), Romania, Thailand (health promotion), United States, Yemen (sports) Note: Countries appear in more than one column when their earmarked tax revenues are used in more than one specific health programme. Source: (27). The amount of money associated with tobacco tax earmarks Case studies of the experiences of nine countries in tobacco tax earmarking show that earmarked funds are relatively small in comparison with government spend- ing on health (see Table 4.12) and, consequently, even smaller in terms of GDP. Therefore, the argument that tobacco tax earmarks would introduce rigidity into public financial management may not apply. 26 Details about how the tobacco tax revenues are earmarked are given in Annex 4.2. 240 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.12 Proportion of earmarked tobacco tax funds in government expenditures COUNTRY ESTIMATED ANNUAL TOTAL FUNDS FROM EARMARKED TAX ANNUAL FUNDS FROM TOBACCO TAX EARMARKS as a % of general government expenditure on health in 2013 Botswana 2014–2015: 4 million pula (US$ 0.48 million) NA Egypt 2013–2014: 392 million Egyptian pounds (US$ 52.06 million); earmarked taxes only 1.8% of total taxes on cigarettes 1.086% Iceland 2014: 108.3 million kronor (US$ 0.89 million) 0.083% Panama 2014: US$ 27.8 million 1.322% Philippines 2014: 50.18 billion Philippine pesos (US$ 1.18 billion) NA Poland 2013: 1 million złoty (US$ 0.316 million) from general budget  0.001% Romania 2014: 1.1 million lei (US$ 0.33 million); 14.4% of total health budget 0.004% Thailand 2014: 4064.74 million baht (US$ 125.15 million); 1.78% of Ministry of Health budget and 1.84% of National Health Security Fund 0.932% Viet Nam 2014: 299.171 billion dong (US$ 13.91 million); 0.5% of national health budget 0.335% Source: (151). 4.6.4 CONCLUSIONS Despite the initial principled resistance to earmarking by some ministries of finance, experience has shown that the use of revenue from tobacco taxes and other taxes on the consumption of products that have negative externalities can ensure political as well as public support. Successful earmarking needs a well-developed structure for the use of funds for health purposes. Even intergovernmental organizations that are opposed to earmarking (e.g. the IMF) have acknowledged the justifiability of well-designed tobacco tax earmarks when revenue is directed to specific cost- effective programmes (153–154). The amounts of tobacco tax revenue effectively earmarked for health have been relatively small and could hardly introduce the feared rigidity in government budgets. Moreover, in some countries, those funds have helped to implement much-needed health programmes (e.g. Australia, the Philippines, Thailand). More governments are considering this option as a stable medium-term source of secure funding for programmes such as tobacco control. The payoffs will be seen in the future as fewer people fall ill and less medical care for tobacco-related illnesses is needed. In Australia, CHAP T ER 4. PO LI T I C AL ECO N OMY 241 an earmarked tax was used to fund a needed and underresourced programme that proved to be successful, effective and impactful; the programme is now sustainably funded, embedded in the federal budget. Earmarking is desirable in a particular political economy when it enables the implementation of effective tobacco taxation that will increase price and reduce consumption. It will, however, also be rational as a matter of public financial man- agement, economic efficiency and democratic governance when concerns such as the following are considered. Although not every question needs an affirmative answer, policy-makers who can answer yes to many of the following questions will likely be considering an effective and rational tobacco tax earmark:27 • Does the tobacco tax earmark’s purpose rationally connect with the recipient programme’s purpose? Earmarks that fund tobacco control or other health programmes are more economically rational under the benefit principle than those that fund unrelated programmes such as childhood education, even when the popularity of the unrelated programmes may make a tax increase politically palatable. • Does the tobacco tax earmark’s amount rationally connect with the needs of the recipient programme? Earmarked funds that cannot be absorbed by the recipient programme are, in effect, money taken away from other needs. • When a tobacco tax earmark funds health programmes, is this clearly com- municated to the public to ensure that the framing of the tobacco tax increase as a health measure reinforces the demand-reduction effect? • Is the programme being funded by the tobacco tax earmark a politically neglected but highly cost-effective or crucially needed programme that, once established as a proof of concept, has a chance of being funded out of the general budget? • Does the tobacco tax earmark’s purpose rationally connect with the effects of the tax itself? Earmarks that fund programmes that disproportionately benefit lower socioeconomic groups or that fund alternative livelihoods for former tobacco workers and farmers will have equity-enhancing effects that will reinforce the already progressive nature of tobacco taxes. • Is the scope of the earmark’s purpose narrow enough that it can be funded mostly from the tobacco tax earmark, to ensure that the revenue is additive and does not merely substitute for spending that would otherwise come from the general fund? • Does the design of the tobacco tax earmark provide for flexibilities that ensure that windfall revenue collection is not squandered on a purpose already saturated with overfunding? 27 Adapted from and informed by References 138–139, 143, 150. 242 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Does the design of the tobacco tax earmark include a sunset clause that triggers its automatic end or review? The presence of such a clause will guard against inefficiency in allocation arising and being left uncorrected due to political inertia. CHAP T ER 4. PO LI T I C AL ECO N OMY 243 REFERENCES 1. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tob Control. 2013; 22(2): 144–153 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 31 January 2021). 2. LaFaive MD, Drenkard S, Nesbit T. Strong link between cigarette tax and illegal smuggling rates [website]. Midland: Mackinak Center for Public Policy; 2018 (https://www.mackinac.org/strong- link-between-cigarette-tax-and-illegal-smuggling-rates#:~:text=An%20increase%20of%2011%20 percentage,packs%20to%2030.6%20million%20packs.&text=Tax%20evasion%20and%20avoidance%20 is,high%20excise%20taxes%20on%20cigarettes., accessed 31 January 2021). 3. Ross H, Blecher E. Illicit trade in tobacco products need not hinder tobacco tax policy reforms and increases. Chicago: University of Illinois at Chicago; 2019 (Tobacconomics White Paper; https:// tobacconomics.org/research/illicit-trade-in-tobacco-products-need-not-hinder-tobacco-tax- policy-reforms-and-increases-white-paper/#:~:text=additional%20related%20research-,Illicit%20 Trade%20in%20Tobacco%20Products%20Need%20Not%20Hinder%20Tobacco,Reforms%20 and%20Increases%20( White%20Paper)&text=The%20paper%20cites%20research%20 that,consumption%2C%20and%20improves%20overall%20health, accessed 31 January 2021). 4. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 12 October 2020). 5. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 6. Joossens L, Chaloupka FJ., Merriman D, Yürekli A. Issues in smuggling of tobacco products. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac00Box 370064B00PUBLIC0.pdf, accessed 18 February 2021). 7. Merriman D, Yürekli A., Chaloupka, FJ. How big is the worldwide cigarette smuggling problem. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac 00Box370064B00PUBLIC0.pdf, accessed 18 February 2021). 8. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 9. Merriman D. Understand, measure, and combat tobacco smuggling. Washington (DC): World Bank Group; 2003 (Economics of tobacco toolkit 7; https://openknowledge.worldbank.org/bitstream/ handle/10986/16267/805800NWP070Sm0Box0379807B00PUBLIC0.pdf?sequence=1&isAllowed=y, accessed 31 January 2021). 10. Progress report on the implementation of the Commission Communication. Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products – a comprehensive EU strategy. Brussels: European Commission; 2017 (COM(2017) 235 final; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:52017DC0235&from=en, accessed 31 January 2021). 11. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 12. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Reuter P, Majmundar M. (editors). Washington (DC): National Academies Press; 2015. 13. von Lampe K. The illegal cigarette trade. In: Natarajan M, editor. International crime and justice. New York: Cambridge University Press. 2010:148–54. 14. Joossens L, Raw M, From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–34 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 31 January 2021). 15. Gomis B, Lee K, Carrillo Botero N, Shepherd P, Iglesias RM. “We think globally”: the rise of Paraguay’s 24 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tabacalera del Este as a threat to global tobacco control. Global Health. 2018;14(1):110 (https:// globalizationandhealth.biomedcentral.com/articles/10.1186/s12992-018-0412-3, accessed 31 January 2021). 16. Communication from the commission to the Council and the European Parliament: Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products - A comprehensive EU Strategy. Brussels: European Commission; 2013 (https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52013DC0324&from=EN, accessed 31 January 2021). 17. Ross H, Vellios N, Smith KC, Ferguson J, Cohen JE. A closer look at ‘cheap white’ cigarettes. Tob Control. 2016;25(5):527–31 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/527.full.pdf, accessed 31 January 2021). 18. Nagelhout GE, van den Putte B, Allwright S, Mons U, McNeill A, Guignard R, et al. Socioeconomic and country variations in cross-border cigarette purchasing as tobacco tax avoidance strategy. Findings from the ITC Europe Surveys. Tob Control. 2014;23:i30–i38 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/23/suppl_1/i30.full.pdf, accessed 1 February 2021). 19. Gilmore AB, Rowell A, Gallus S, Lugo A, Joossens L, Sims M. Towards a greater understanding of the illicit tobacco trade in Europe: a review of the PMI funded ‘Project Star’ report. Tob Control. 2014;23:e51–e61 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/e1/e51.full.pdf, accessed 1 February 2021). 20. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Revista Panam Salud Publica. 2016;40(4):243–9 (https://www.scielosp.org/pdf/ rpsp/2016.v40n4/243-249, accessed 1 February 2021). 21. Operação Huno combate mercado clandestino de cigarros em sete Estados [Operação Huno fights clandestine cigar market in seven states]. Brasilia: Federal Police of Brazil; 2015 (in Portuguese) (http:// www.pf.gov.br/agencia/noticias/2015/11/operacao-huno-combate-mercado-clandestino-de-cigarros- em-seis-estados, accessed 12 October 2020). 22. Gilmore AB, Gallagher AWA, Rowell A. Tobacco industry’s elaborate attempts to control a global track and trace system and fundamentally undermine the Illicit Trade Protocol. Tob Control. 2019;28:127–40 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/28/2/127.full.pdf, accessed 1 February 2021). 23. Technical assessment of the experience made with the Anti-Contraband and Anti-Counterfeit Agreement and general release of 9 July 2014 among Philip Morris International and affiliates, the Union and its Member States. Brussels: European Commission; 2016 (SWD (2016) 44 final; https://ec.europa.eu/anti- fraud/sites/antifraud/files/technical_assessment_pmi_24022016_en.pdf, accessed 1 February 2021). 24. Gallagher AWA, Evans-Reeves KA, Hatchard JL, Gilmore AB. Tobacco industry data on illicit tobacco trade: a systematic review of existing assessments. Tob Control. 2019;28(3):334–45 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/28/3/334.full.pdf, accessed 1 February 2021). 25. Joossens L, Merriman D, Ross H, Raw M. How eliminating the global illicit cigarette trade would increase tax revenue and save lives. Paris: International Union Against Tuberculosis and Lung Disease; 2009 (https://theunion.org/sites/default/files/2020-08/How%20Eliminating%20the%20Global%20Illicit%20 Cigarette%20Trade%20Would%20Increase%20Tax%20Revenue%20and%20Save%20Lives%20-%20 EN.pdf, accessed 1 February 2021). 26. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 27. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 28. Prieger JE, Kulic J. Cigarette taxes and illicit trade in Europe. Econ Inquiry. 2018;56(3):1706–23 (https:// papers.ssrn.com/sol3/papers.cfm?abstract_id=2718519, accessed 18 February 2021). 29. Stoklosa M. Prices and cross-border cigarette purchases in the EU: evidence from demand modelling. Tob Control. 2020;29:55–60. 30. Passport: illicit trade in tobacco products. In: Euromonitor International [website]. London: Euromonitor International; 2018 (restricted access). 31. The Global Illicit Trade Environment Index. London: The Economist Intelligence Unit; 2018 (http:// illicittradeindex.eiu.com/documents/EIU%20Global%20Illicit%20Trade%20Environment%20Index%20 CHAP T ER 4. PO LI T I C AL ECO N OMY 245 2018%20-%20Overall%20Results%20White%20Paper%20June%206%20FINAL.pdf, accessed 1 February 2021). 32. Chaloupka FJ, Yürekli A, Fong GT. Tobacco taxes as a tobacco control strategy. Tob Control. 2012;21:172– 80 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/172.full.pdf, accessed 1 February 2021). 33. Chernick H, Merriman D. Using littered pack data to estimate cigarette tax avoidance in NYC. Natl Tax J. 2013;66(2):635–68 (https://www.ntanet.org/NTJ/66/3/ntj-v66n03p635-668-littered-pack-data- cigarette-tax.pdf, accessed 1 February 2021). 34. Liber AC, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia Survey. Tob Control. 2015;24:iii83–iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed 1 February 2021). 35. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 36. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 37. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 38. Paraje G, Araya D, Drope J. Illicit cigarette trade in metropolitan Santiago de Chile. Tob Control 2020;29:68–73. 39. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 40. Saenz de Miera Juarez BS, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020(0):1-7 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 1 February 2021). 41. Lee K, Collin J. ‘Key to the future’: British American Tobacco and cigarette smuggling in China. PLoS Medicine. 2006;3(7):e228 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1502159/pdf/pmed.0030228. pdf, accessed 2 February 2021). 42. Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38– i43 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i38.full.pdf, accessed 2 February 2021). 43. Aziani A, Kulick J, Norman N, Prieger JE. Empty discarded pack data and the prevalence of illicit trade in cigarettes; 2017. (https://ssrn.com/abstract=2906015, accessed 12 October 2020). 44. Philip Morris International. The Perspective of PM International on Smoking and Health Issues. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1985. Bates no. 2023268328-2023268349 (https://www.industrydocuments.ucsf.edu/tobacco/docs/#id=pxpb0040, accessed 5 February 2021). 45. Mandal S, Gilmore A, Collin J, Weishaar H, Smith K, McKee M. Block, amend, delay: tobacco industry efforts to influence the European Union’s Tobacco Products Directive (2001/37/EC). Smokefree Partnership/Cancer Research UK; 2009 (http://www.smokefreepartnership.eu/images/Reports-and- Position-Papers/EU_TI_TPD_report_May_2012.pdf, accessed 15 October 2020). 46. Bishop DJ. Key area paper: indirect taxation – British American Tobacco. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1994. Bates no. 503911771-503911780 (http:// bat.library.ucsf.edu/data/h/f/k/hfk11a99/hfk11a99.pdf, accessed 15 October 2020). 47. Zhou S, Liberman JD, Ricafort E. The impact of the WHO Framework Convention on Tobacco Control in defending legal challenges to tobacco control measures. Tob Control. 2009:28:s113-s118 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/28/Suppl_2/s113.full.pdf, accessed 2 February 2021). 48. Gilmore AB, Fooks G, Drope J, Bialous SA, Jackson RR. Exposing and addressing tobacco industry 246 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N conduct in low-income and middle-income countries. Lancet. 2015;385:1029–43. 49. Ulucanlar S, Fooks G, Gilmore A. The policy dystopia model: an interpretive analysis of tobacco industry political activity. PLoS Med. 2016:13(9):e1002125 (https://journals.plos.org/plosmedicine/ article?id=10.1371/journal.pmed.1002125, accessed 2 February 2021). 50. Apollonio D, Glantz S. Tobacco manufacturer lobbying to undercut minimum price laws: an analysis of internal industry documents. Tob Control. 2020;29:e10-e17 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC7374022/pdf/nihms-1590806.pdf, accessed 2 February 2021). 51. Crosbie E, Glantz S. Tobacco industry argues domestic trademark laws and international treaties preclude cigarette health warning labels, despite consistent legal advice that the argument is invalid. Tob Control. 2012;23:e7 (https://www.researchgate.net/publication/233769722_Tobacco_industry_ argues_domestic_trademark_laws_and_international_treaties_preclude_cigarette_health_warning_ labels_despite_consistent_legal_advice_that_the_argument_is_invalid, accessed 2 February 2021). 52. Puig S. Tobacco litigation in international courts. Harv Int Law J. 2016;57(2):1–68. 53. Crooked nine: nine ways the tobacco industry undermines health policy. New York: Vital Strategies; 2019 (https://exposetobacco.org/wp-content/uploads/2019/09/Crooked-9-STOP.pdf, accessed 2 February 2021). 54. Balwicki L, Stokłosa M, Balwicka-Szczyrba M, Tomczak W. Tobacco industry interference with tobacco control policies in Poland: legal aspects and industry practices. Tob Control. 2016;25(5):521–26 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/521.full.pdf, accessed 2 February 2021). 55. Sud SR, Brenner JE, Shaffer ER. Trading away health: the influence of trade policy on youth tobacco control. J Pediatrics. 2015;166(5):1303–07 (https://www.jpeds.com/action/showPdf?pii =S0022-3476%2815%2900017-7, accessed 2 February 2021). 56. Lee S, Ling PM, Glantz SA. The vector of the tobacco epidemic: tobacco industry practices in low and middle-income countries. Cancer Cause Control. 2012;23:117–29 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC3332051/pdf/nihms-367996.pdf, accessed 2 February 2021). 57. Zhou S, Liberman J. The global tobacco epidemic and the WHO Framework Convention on Tobacco Control—the contributions of the WHO’s first convention to global health law and governance. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018 (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3339718, accessed 18 February 2021). 58. Legislative Consultation with Constitutional Division of the Supreme Court, No. 2012-003918. Constitutional Division of the Supreme Court of Costa Rica; 20 March 2012 (https://www. tobaccocontrollaws.org/files/live/litigation/830/CR_Legislative%20Consultation%20with%20.pdf, accessed 2 February 2021). 59. British American Tobacco Kenya Ltd v Ministry of Health, Petition No. 143 of 2015. High Court of Kenya; 2016 (https://www.tobaccocontrollaws.org/files/live/litigation/2643/KE_British%20American%20 Tobacco%20Kenya.pdf, accessed 2 February 2021). 60. British American Tobacco v Ministry of Health, et al, Civil Appeal No. 112 of 2016. Kenyan Court of Appeal; 2017 (https://www.tobaccocontrollaws.org/files/live/litigation/2521/KE_British%20 American%20Tobacco%20Ltd%20v.pdf, accessed 2 February 2021). 61. Thailand – customs and fiscal measures on cigarettes from the Philippines: Report of the Panel. Geneva: World Trade Organization; 2010 (WT/DS371/R; https://docs.wto.org/dol2fe/Pages/SS/directdoc. aspx?filename=Q:/WT/DS/371R.pdf&Open=True, accessed 2 February 2021). 62. Tribunal Constitucional de Chile [Constitutional Court of Chile] (1995 Rol No. 219-95; https://www. tribunalconstitucional.cl/descargar_sentencia2.php?id=3074, accessed 4 April 2020.) 63. McGrady B. International investment law. In: Voon T, Mitchell AD, Liberman J, editors. Regulating tobacco, alcohol and unhealthy foods: the legal issues. New York: Routledge; 2014. 64. Marvin Roy Feldman Karpa v United Mexican States, Case No. ARB(AF)/99/1 International Centre for Settlement of Investment Disputes; 2002 (https://www.italaw.com/sites/default/files/case-documents/ ita0319.pdf, accessed 2 February 2021). 65. Feldman Karpa v Mexico (2002 ICSID Arbitral Tribunal) (https://www.italaw.com/sites/default/files/ case-documents/ita0319.pdf, accessed 18 February 2021). 66. Undang-Undang Nomor 11 Tahun 1995 tentangCukai (Lembaran Negara Republik Indonesia Tahun1995 CHAP T ER 4. PO LI T I C AL ECO N OMY 247 Nomor 76) (Excise Law,in Indonesian). Article 5. 67. Excise Regulation (2011) Reference No. PMK Number 167/PMK.011/2011 (in Indonesian) (https:// peraturan.bpk.go.id/Home/Details/145897/pmk-no-167pmk0112011, accessed 3 February 2021). 68. Mahkamah Agung. Indonesia Supreme Court, Case No. 73/PDT/2018/PT SBY (https://putusan3. mahkamahagung.go.id/direktori/putusan/74efff9cfe18d60990fcde19583ecd5f.html, accessed 6 February 2021). 69. Maya D. Analisis kebijakan kenaikam tarif cukai dan penyesuaian batasan harga jual eceran hasil tembakau [Policy analysis on excise tax increases and price limit adjustments retail tobacco products]; 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-09/S45937-Deacy%20Maya accessed 3 March 2020). 70. Indriawati EN, Susanti L, Qurnia MA. Kebijakan tarif cukai hasil tembakau terhadap perekonomian [Tobacco excise tariff policy on the economy]; 2014 (in Indonesian). (https://www.academia. edu/6611977/MAKALAH_CHT_kel_6, accessed 15 October 2020). 71. Arroyani DW. Evaluasi Kebijakan Penetapan Tarif Cukai Hasil Tembakau Berdasarkan PMK [Evaluation of Tobacco Excise Tariff Determination Policy Based on PMK].167/PMK.011/2011. 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-08/S44664-Dwira%20Wanti%20Arroyan, accessed 15 October 2020). 72. Ngo Ngo Ha and Another v. State of New South Wales and Others (Matter No S 45 of 1996); Walter Hammond & Associates Pty Ltd v. State of New South Wales and Others (Matter No S 165 of 1996). Constitutional law; (1997) 146 ALR 355: 5 August 1997 (https://staging.hcourt.gov.au/assets/ publications/judgments/1997/029--NGO_NGO_HA_AND_ANOR_v_STATE_OF_NEW_SOUTH_WALES_&_ ORS_(Matter_No_S_45_of_1996);_WALTER_HAMMOND--(1997)_146_ALR_355.html, accessed 2 February 2021). 73. McGrady B. Trade and public health: the WTO, tobacco, alcohol, and diet. New York: Cambridge University Press; 2011. 74. Giorgetti C. Health and international investment law. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018. 75. British American Tobacco v Attorney General and Center for Human Rights and Development. Constitutional Court of Uganda Petition No.46 of 2016; 2017 (https://media.ulii.org/files/judgments/ ugcc/2019/4/2019-ugcc-4_0.pdf, accessed 18 February 2021). 76. Lim CL, Ho J, Paparinskis M. International investment law and arbitration: commentary, awards and other materials. Cambridge: Cambridge University Press; 2018. 77. Dominican Republic – measures affecting the importation and internal sale of cigarettes : Report of the Panel. Geneva: World Trade Organization; 2004 (WT/DS302/R; https://docs.wto.org/dol2fe/ Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302R.pdf&Open=True, accessed 21 February 2021) 78. Dominican Republic — measures affecting the importation and internal sale of cigarettes: Report of the Appellate Body. Geneva: World Trade Organization; 2005 (WT/DS302/AB/R; https://docs.wto. org/dol2fe/Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302ABR.pdf&Open=True, accessed 21 February 2021). 79. Gilmore A. Tobacco and transition: understanding the impact of transition on tobacco use and control in the former Soviet Union [Thesis]. London: London School of Hygiene and Tropical Medicine, June 2005 (https://escholarship.org/uc/item/3rw2c04c, accessed 2 February 2021). 80. MacKenzie R, Collin J, Sopharo C, Sopheap Y. ‘Almost a role model of what we would like to do everywhere’: British American Tobacco in Cambodia. Tob Control. 2004;13(suppl II): ii112–ii117 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/13/suppl_2/ii112.full.pdf, accessed 2 February 2021). 81. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tobacco Control 2013;22(2):144-53 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 21 February 2021). 82. Price and tax. In: TobaccoTactics [website]. Bath: University of Bath; 2020 (https://tobaccotactics.org/ topics/price-and-tax/, accessed 15 December 2020). 83. Vulovic V, Chaloupka FJ. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation—commentary. Tob Control. 24 June 2020. 248 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 84. Verguet S, Kearns PKA, Rees VW. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation. Tob Control. 23 June 2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/08/11/tobaccocontrol-2019-055315. full.pdf, accessed 2 February 2021).. 85. Palipudi KM, Gupta PC, Sinha DN, Andes LJ, Asma S, McAfee T. Social determinants of health and tobacco use in thirteen low and middle income countries: evidence from Global Adult Tobacco Survey. PLoS ONE. 2012;7(3):e33466 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3306395/pdf/ pone.0033466.pdf, accessed 2 February 2021). 86. Systematic review of the link between tobacco and poverty. Geneva: World Health Organization; 2014 (http://www.who.int/tobacco/publications/economics/9789241507820/en, accessed 21 October 2020). 87. Goodchild M, Sinha P, Gill Munish V, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/bitstream/handle/10665/331836/whoseajph2020v9i1p73. pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 88. Faruque GM, Ahmed M, Huq I, Parven R, Wadood SN, Chowdhury SR, et al. The economic cost of tobacco use in Bangladesh: a health cost approach. Dhaka: Bangladesh Cancer Society; 2020 (https://www. researchgate.net/publication/342445467_The_Economic_Cost_of_Tobacco_Use_in_Bangladesh_A_ Health_Cost_Approach, accessed 2 February 2021). 89. John RM, Sung H, Max WB, Ross H. Counting 15 million more poor in India, thanks to tobacco. Tob Control. 2011;20:349–52 (https://www.researchgate.net/publication/49809290_Counting_15_million_ more_poor_in_India_thanks_to_tobacco, accessed 2 February 2021). 90. Liu Y, Rao K, Hu TW, Sun Q, Mao Z. Cigarette smoking and poverty in China. Soc Sci Med. 2006;Dec;63(11):2784–90. 91. Fuchs A, González Icaza F, Paz D. Distributional effects of tobacco taxation – a comparative analysis. Washington (DC): World Bank Group; 2019 (Policy research working paper 8805; https://openknowledge. worldbank.org/bitstream/handle/10986/31534/WPS8805.pdf?sequence=4&isAllowed=y, accessed 2 February 2021). 92. Smith CE, Hill SE, Amos A. Impact of population tobacco control interventions on socioeconomic inequalities in smoking: a systematic review and appraisal of future research directions. Tob Control. 29 September 2020 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/09/28/ tobaccocontrol-2020-055874.full.pdf, accessed 2 February 2021). 93. Furman J. Six lessons from the U.S. experience with tobacco taxes – World Bank conference “Winning the tax wars: global solutions for developing countries”. Washington (DC): Obama White House Archives; 24 May 2016 (https://obamawhitehouse.archives.gov/sites/default/files/page/files/20160524_cea_ tobacco_tax_speech.pdf, accessed 21 October 2020). 94. Önder Z, Yürekli A. Who pays the most cigarette tax in Turkey. Tob Control. 2016;25(1):39–45. 95. Verguet S, Gauvreau CL, Mishra S, MacLennan M, Murphy SM, Brouwer ED, et al. The consequences of tobacco tax on household health and finances in rich and poor smokers in China: an extended cost-effectiveness analysis. Lancet Glob Health. 2015;3:e206–16 (https://www.thelancet.com/action/ showPdf?pii=S2214-109X%2815%2970095-1, accessed 2 February 2021). 96. Fuchs Tarlovsky A, González Icaza F. The welfare and distributional effects of increasing taxes on tobacco in Vietnam. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/ curated/en/758391561477648302/pdf/The-Welfare-and-Distributional-Effects-of-Increasing-Taxes- on-Tobacco-in-Vietnam.pdf, accessed 2 February 2021). 97. Expanding the global tax base: taxing to promote public goods: tobacco taxes. Washington (DC): World Bank Group; 2016 (http://pubdocs.worldbank.org/en/799981480947591462/Taxing-to-Promote- Public-Goods-tobacco-taxation-chapter-conference-25-nov-2016-final-version-pmarquez-002.pdf, accessed 2 February 2021). 98. Blecher E. Increasing tobacco taxes significantly will increase revenues. Chicago: University of Illinois at Chicago; 2018. (Tobacconomics policy brief; https://tobacconomics.org/files/research/470/Laffer- Curve-Policy-Brief_v2.2.pdf, accessed 2 February 2021). 99. Savedoff W, Lopert R. Sugar, rum, and tobacco: domestic resource mobilization for low-income countries through excise taxes. Washington (DC): Centre for Global Development; 2020 (CGD Policy Paper CHAP T ER 4. PO LI T I C AL ECO N OMY 249 164, https://www.cgdev.org/sites/default/files/PP164-Savedoff-Health-Taxes-DRM.pdf, accessed 2 February 2021). 100. Lavado R, Postolovska I, Sotomayor RE. Beyond the perceived regressivity of higher tobacco taxes, turning short-term losers into long-term winners? In: Tobacco reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 101. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 102. Tonga NCD related taxation policy assessment preliminary findings. Presentation made at the innovations and actions against NCDs conference ADB headquarters, Manila, 17 July 2018. 103. Moreno-Dodson B. Raising tobacco taxes: why, what, and how: where public health and economics converge. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e- library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 104. Fuchs A, Del Carmen G, Mukong AK. Long-run impacts of increasing tobacco taxes: evidence from South Africa. Policy Research Working Paper 8369. Washington [DC]: World Bank; 2018 (https://openknowledge. worldbank.org/bitstream/handle/10986/29497/WPS8369.pdf?sequence=4&isAllowed=y, accessed 18 February 2021) . 105. van Walbeek C. Case studies in illicit tobacco trade: South Africa. Tobacconomics fact sheet. Chicago: University of Chicago, Tobacconomics, Health Policy Center; 2020 (https://tobacconomics.org/files/ research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 106. van Loggerenberg, J. Tobacco Wars. Cape Town: NB Publishers; 2019. 107. Marquez P. Global tobacco control: a development priority. In: Tobacco tax reform at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads. pdf, accessed 2 February 2021). 108. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington [DC]: World Bank; 2017 (https:// tobacconomics.org/files/research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 109. Scollo M, Bayly M. Revenue from tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia. org.au/chapter-13-taxation/13-6-revenue-from-tobacco-taxes-in-australia, accessed 21 October 2020). 110. Scollo M, Bayly M. Tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia.org.au/ chapter-13-taxation/13-2-tobacco-taxes-in-australia, accessed 21 October 2020). 111. Breton E, Richard L, Gagnon F, Jacques M, Bergeron P. Fighting a tobacco-tax rollback: a political analysis of the 1994 cigarette contraband crisis in Canada. Journal of Pub Health Policy. 2006;27:77–99. 112. Fanta E, Garcia M. Preventing and controlling illicit tobacco flows. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB- Report-CrossRoads.pdf, accessed 2 February 2021). 113. Goodchild M, Perucic AM, Zheng R, Blecher E, Paul J. The health impact of raising tobacco taxes in developing countries. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/ taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 114. Ross H. Tracking and tracing tobacco products in Kenya. Prev Med. 2017;105:S15–S18 (https://reader. elsevier.com/reader/sd/pii/S0091743517301457?token=424D9EC1D42A9C69BDB55B90090C3F7853AA 20A4970D0E379C91EE64C73A273A0FA949557F4F120A0B636701581EC055, accessed 2 February 2021). 115. Mamudu H, Hammond R, Glantz S. Tobacco industry attempts to counter the world bank report curbing the epidemic and obstruct the WHO framework convention on tobacco control. Soc Sci & 250 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Med. 2008;67(11):1690–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2662513/pdf/nihms81164. pdf, accessed 2 February 2021). 116. Gilmore A. Understanding the vector in order to plan effective tobacco control policies: an analysis of contemporary tobacco industry materials. Tob Control. 2012;21:119–26 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/21/2/119.full.pdf, accessed 2 February 2021). 117. The economics of tobacco taxation and employment in Indonesia: policy implications technical brief. Washington (DC): World Bank Group; 2018 (https://openknowledge.worldbank.org/bitstream/ handle/10986/29814/126158-WP-P154568-PUBLIC-add-series-WBGPNIndoEmploymentUPDATEFin alWeb.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 118. Sahadewo GA, Drope J, Li Q, Nargis N, Witoelar F. Tobacco or not tobacco: predicting farming households’ income in Indonesia. Tob Control. 2020. 119. The economics of tobacco farming in Indonesia: health, population, and nutrition global practice. Washington (DC): World Bank Group, 2017 (http://documents1.worldbank.org/curated/ en/161981507529328872/pdf/120307-REVISED-PUBLIC-WBGIndoEconomicsTobaccoFarming.pdf, accessed 2 February 2021). 120. The revenue and employment outcome of biri taxation in Bangladesh. Dhaka: National Board of Revenue, Government of the People’s Republic of Bangladesh; 2019 (https://www.who.int/docs/ default-source/searo/bangladesh/pdf-reports/cat-2/biri-study-report-03-12-2019.pdf?sfvrsn=b8fea69_2, accessed 2 February 2021). 121. Magati P, Lencucha R, Li Q, Drope J, Labonte R, Appau A, et al. Costs, contracts and the narrative of prosperity: an economic analysis of smallholder tobacco farming livelihoods in Kenya. Tob Control. 2019;28(3):268–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6512316/pdf/nihms-1012064. pdf, accessed 2 February 2021). 122. Ghaus K, Iqbal MA, Sabir M, Saleem W, Ali A, Aamir N. Macroeconomic impacts of tobacco use in Pakistan. Karachi: The Social Policy and Development Centre; 2018 (https://tobacconomics.org/files/ research/483/Macroeconomic-impact-of-Tobacco-in-Pakistan.pdf, accessed 2 February 2021). 123. Ross H. The effects of tobacco taxes on employment and labor productivity. In: Tobacco tax reform: at the crossroads of health and development. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021) 124. Bigawanto M. Tobacco industry interference undermined tobacco tax policy in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance (SEATCA); 2019 (https://seatca.org/dmdocuments/ Indonesia%20TII%20in%20Tax.pdf, accessed 2 February 2021). 125. Ahsan A, Wyono N. Analysis of the impact of higher cigarettes prices on employment in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance; 2007 (https://seatca.org/dmdocuments/An%20 Analysis%20of%20the%20Impact%20of%20Higher%20Cigarette%20Prices%20on%20Employment%20 in%20Indonesia.pdf, accessed 2 February 2021). 126. Higashi H, Khuong TA, Ngo AD, Hill PS. Evidence and decision making: tobacco control policy and legislation in Vietnam. Int J of Health Plann. 2013;2(1):e72–e94. 127. Buck D, Raw M, Godfrey CA, Sutton M. Tobacco and jobs: the impact of reducing consumption on employment in the UK. York: University of York; 1995 (https://www.researchgate.net/ publication/4814523_Tobacco_and_jobs_the_impact_of_reducing_consumption_on_employment_ in_the_UK, accessed 2 February 2021). 128. Petkova R, Ileva R and Rusenov A. Tobacco control policies and tobacco industry employment in Bulgaria. In: Economic, social and health issues in tobacco control. Report of a WHO international meeting, Kobe, Japan, 3-4 December 2001. Geneva: World Health Organization; 2003. 129. Jha V, Narayanan BG, Wadhwa D, Tesche J. Economic and environmental effects of reduction in smoking prevalence in Tanzania. Tob Control; 2020;29:24–8 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/29/1/24.full.pdf, accessed 2 February 2021). 130. Lencucha R, Makoka D, Drope J, Appau A, Labonté R, Li Q, et al. Costs, revenues and profits: an economic analysis of smallholder tobacco farmer livelihoods in Malawi. Tob Control. 2017;26(6)634–40 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC5957074/pdf/nihms965842.pdf, accessed 2 February 2021). 131. Goma FM, Labonté R, Drope J, Li Q, Zulu R, Kangwa E. The economics of tobacco farming in Zambia: CHAP T ER 4. PO LI T I C AL ECO N OMY 251 tobacco farmers survey report. Lusaka: University of Zambia and American Cancer Society; 2019 (https://www.cancer.org/content/dam/cancer-org/research/economic-and-healthy-policy/zambia- report-7-16-19.pdf, accessed 28 October 2020). 132. Chavez JJ, Drope J, Li Q, Aloria MJ. The economics of tobacco farming in the Philippines. Quezon City: Action for Economic Reforms and Atlanta: American Cancer Society, 2016 (https://www.cancer. org/content/dam/cancer-org/research/economic-and-healthy-policy/the-economics-of-tobacco- farming-in-the-philippines.pdf, accessed 22 October 2020). 133. Augstin CP, Cardenas RP, Cortez JB, Gimongala GC, Ramos AR. The effects of the sin tax reform law of 2012 to tobacco farmers of Amulung, Cagayan. IJARMSS. 2017;6(12):57–75. 134. Mercado A. Tobacco tax: good for the tobacco farmers, too. Business World; 2019 [website] (https:// www.bworldonline.com/tobacco-tax-good-for-the-tobacco-farmers-too/, accessed 25 October 2020). 135. Yürekli A, Önder Z, Elibol M, Erk N, Cabuk A, Fisunoglu M, et al. The economics of tobacco and tobacco taxation in Turkey. Paris: International Union Against Tuberculosis and Lung Disease; 2010 (https:// www.who.int/tobacco/en_tfi_turkey_report_feb2011.pdf, accessed 2 February 2021). 136. Latif E, Warner V. Legislation and smoking prevention. In: Loddenkemper R, Kreuter M, editors. The Tobacco Epidemic: progress in Respiratory Research. 2015;42:136–148. 137. Tandon A, Cashin C. Assessing public expenditure on health from a fiscal space perspective. Washington (DC): World Bank Group; 2010 (http://documents1.worldbank.org/curated/en/333671468330890417/ pdf/560530WP0Box341penditureFiscalSpace.pdf, accessed 2 February 2021). 138. Cashin C, Sparkes S, Bloom D. Earmarking for health: from theory to practice. Geneva: World Health Organization; 2018 (https://apps.who.int/iris/bitstream/handle/10665/255004/9789241512206-eng. pdf?sequence=1, accessed 2 February 2021). 139. Wilkinson M. Paying for public spending: is there a role for earmarked taxes? Fiscal Studies. 1994;15(4):119–35 (https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1475-5890.1994.tb00213.x, accessed 2 February 2021). 140. ITAF: impôts et taxes affectées à la protection sociale. Les impôts et taxes affectés (ITAF) [taxes and duties allocated to social protection]. Vie publique: au coeur du débat public [website]. Paris: Directorate of Legal and Administrative Information; 2018 (in French) (https://www.vie-publique.fr/fiches/22010- impots-et-taxes-affectees-itaf-protection-sociale, accessed 31 October 2020). 141. Seventeenth congress of the Philippines, Metro Manila, RA 11346, 23 July 2018 (https://www. officialgazette.gov.ph/downloads/2019/07jul/20190725-RA-11346-RRD.pdf, accessed 31 October 2020). 142. McCleary W. The earmarking of government revenue: a review of some World Bank experience. World Bank Res Obs. 1991;6(1):81–104 (http://documents1.worldbank.org/curated/en/109871468764672767/ pdf/multi-page.pdf, accessed 2 February 2021). 143. Strategic investment of tobacco tax revenue. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/global/pdfs/en/strategic_investment_tobacco_tax_revenue. pdf, accessed 31 October 2020). 144. Bialous SA, Glantz SA. Arizona’s tobacco control initiative illustrates the need for continuing oversight by tobacco control advocates. Tob Control. 1999;8:141–51 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/8/2/141.full.pdf, accessed 2 February 2021). 145. Goldman LK, Glantz SA. The passage and initial implementation of Oregon’s measure 44. Tob Control. 1999;8:311–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1763939/pdf/v008p00311.pdf, accessed 2 February 2021). 146. Heiser PF, Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health. 1997;87(6):968–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380932/pdf/amjph00505-0082. pdf, accessed 2 February 2021). 147. Landman A, Bialick P. Tobacco industry involvement in Colorado. Chicago: American Lung Cancer Association; 2004 (https://escholarship.org/uc/item/9d880448#main, accessed 18 February 2021). 148. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010. (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 149. Vardavas CI, Filippidis FT, Agaku IT, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the 252 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N importance of earmarking the revenue to health care and tobacco control. Tob Induc Dis. 2012;10:21 (http://www.tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the- revenue-to-health-care-and-tobacco,66110,0,2.html, accessed 31 October 2020). 150. Bird RM. Tobacco and alcohol excise taxes for improving public health and revenue outcomes: marrying sin and virtue?. Washington (DC): World Bank Group; 2015 (Policy Research Working Paper 7500; https://openknowledge.worldbank.org/bitstream/handle/10986/23464/ Tobacco0and0al0ing0sin0and0virtue00.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 151. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 152. The establishment and use of dedicated taxes for health. Manila: World Health Organization Western Pacific Regional Office; 2004 (https://www.who.int/health_financing/documents/dedicated_taxes. pdf?ua=1, accessed 2 February 2021). 153. Teja RS. The case for earmarked taxes: theory and example. Washington (DC): International Monetary Fund; 1988 (IMF Working Paper; https://www.elibrary.imf.org/doc/IMF001/14514- 9781451922677/14514-9781451922677/Other_formats/Source_PDF/14514-9781455212002.pdf, accessed 2 February 2021). 154. Keen M. Taxation and development – again. Washington (DC): International Monetary Fund; 2012(IMF Working Paper WP/12/220O; (https://www.imf.org/external/pubs/ft/wp/2012/wp12220.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 253 ANNEX 4.1 METHODS TO ASSESS THE NATURE AND SIZE OF THE ILLICIT TOBACCO TRADE A4.1 DIRECT MEASUREMENT A4.1.1 SMOKER INTERCEPT AND PACK OBSERVATION SURVEYS Illicit trade can be measured directly by examining the cigarette packs of smokers. The smokers themselves can provide information on purchasing patterns, brand preferences and prices paid. Researchers can select individuals or retailers to survey based on a convenience sample (i.e. a sample that may not be representative) or a probability-based sample (i.e. a sample selected to be statistically representative of an underlying population). Data collected from a pack could reveal whether the pack is compliant or non- compliant with the local tax laws. Information can be obtained from objective markings such as brands, public health warning labels, tax stamps, foreign language labels or duty-free labels. During these stops, researchers can record demographic information (e.g. age and gender of the smoker), smoking-related history (e.g. number of cigarettes smoked per day) and price information. This is helpful in understand- ing the profile of smokers who are able and willing to avoid cigarette taxes. Pack observations can be used in conjunction with population-based household surveys to obtain population-based estimates of the illicit tobacco trade (1,2). For example, as part of a regular national health survey, Kaplan et al. conducted a cross-sectional study of smokers in Turkey, using a face-to-face interviewer-administered survey and pack observation (3). They were able to collect sociodemographic, lifestyle and medical details along with pack observations as part of the study protocol. Advantages and disadvantages of smoker intercept and pack observation surveys A primary advantage of conducting pack observation is that it is direct and objective, and smokers are not subject to any value judgements (2). Paired with survey data, pack observation can appropriately account for respondents who are not residents of the area in which they are surveyed (4). Disadvantages include the difficulty of identifying areas that are representative of the tobacco use population and the difficulty of sampling important subpopulations such as elderly and immobile smok- ers. Also, surveys conducted in the daytime may discount the number of youthful smokers who are in school. Another disadvantage is that a sizeable number of smokers may refuse to show their last-purchased pack (2). Kaplan et al. found that 24% of smokers sampled in Turkey did not show their cigarette pack to the study interviewer (3). This issue may be mitigated by asking users to provide information on the brand purchased, whether any public health warnings were posted and the 254 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N price paid (5). Although the responses are based on recall, they may still yield useful information. For example, Joossens et al. allowed smokers who did not show their packs to provide self-reported information and found no statistical differences in illicit packs between these respondents and those who did show their packs (2). Another obstacle to accurate measurement based on smoker intercepts is the inability to determine the tax payment of smokers who purchase single cigarettes, since these smokers are generally not given packs. However, information can still be captured in self-report surveys by asking smokers to report the brand purchased and price paid. Key study for readers to refer to for additional guidance: Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–447. A4.1.2 PACK RETURN AND SWAP SURVEYS Pack return and pack swap surveys fall within the broader category of pack ob- servation studies that use survey sampling techniques to examine smokers’ pack characteristics and to determine whether they are tax compliant. For these surveys, the unit of analysis is the individual. The main differences between pack swap and pack return surveys is that swap surveys offer the smoker a replacement pack, whereas pack returns are built into mail surveys and allow respondents to mail in their unopened packs. Pack swap and pack return surveys use probability and nonprobability sampling procedures. Probability sampling allows researchers to generalize to the broader population. Governments can use this method to rapidly assess the availability of illicit products in a given geographic area or to measure the share of the illicit market- place. Rapid assessment may be performed in instances where there is an emerging tobacco product (e.g. a new cheap white brand) or suspected counterfeiting of tax stamp features. Rapid assessment using a convenience sampling strategy could place researchers near busy intersections where they could ask smokers for permission to look at their cigarette packs or to take photographs that could be analysed later. A population-based study requires a sample that closely mirrors the tobacco use population. Advantages and disadvantages of pack return and pack swap surveys Pack swap and pack return surveys may help to overcome the stigma associated with traditional smoking surveys. For example, when researchers ask smokers to see their cigarette packs (or when they take photographs), no value judgements are made. These surveys are good rapid-assessment tools that can be used to examine CHAP T ER 4. PO LI T I C AL ECO N OMY 255 the effectiveness of physical features of a pack designed to deter illicit trade (e.g. packs that have tracking and tracing technology or high-tech stamps). In addi- tion, they can be supplemented with population-based tobacco use surveys. When coupled with such survey data, these methods allow researchers to obtain relevant information about the context of illicit purchases, including, for example, sources (e.g. street, peer networks, retail stores) and prices. Mail-in surveys are filled out in the comfort of the respondent’s home without the presence of family members or passers-by, which may assure them that responses will be kept confidential. A potential disadvantage is that smokers who purchase both illicit cigarettes and tax-paid cigarettes may disproportionately mail back compliant packs. In addition, in LMICs, this mode of survey distribution may be unreliable because of issues associated with mail delivery systems. Key study for readers to refer to for additional guidance: Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61-66. A4.1.3 LITTERED-PACK SURVEYS Littered-pack surveys, also known as empty discarded pack surveys, are used pre- dominantly in high- and middle-income countries (e.g. the United States, France, Canada, New Zealand, Mexico and Poland). This unobtrusive method relies on the premise that smokers publicly discard packs (e.g. on streets, sidewalks and in public trash cans). The packs bear characteristics that illustrate whether they are tax compliant (e.g. tax stamps, health warnings). For example, an Albanian health warning label on a cigarette pack discarded in Greece provides evidence that the pack was destined for the Albanian market. The pack may have been smuggled into Greece by criminal entrepreneurs or it may have been brought by a visitor. Collecting discarded packs from a representative geographic sample and examining these characteristics can provide estimates of tax compliance. Operationally, this data collection method uses an ecological approach whereby geographies are the units of analysis. Geographical units are meant to represent the smokers in the city/ country and can be administratively defined (e.g. by the country’s census bureau or transit zones) or may reflect researcher-defined neighbourhoods (e.g. half-mile buffer zones near bus stops or activity spaces). Researchers in Canada have expanded the littered-pack method to include col- lection and analysis of cigarette butts on 25 postsecondary campuses. The cigarette butts provide information on the brands sold (or lack thereof) and allow researchers to distinguish between legal and illegal products (6). A recent innovative expansion 256 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the littered-pack methodology is the collection of packs from cigarette retailers. John and Ross collected empty packs of tobacco products from a sample of registered and unregistered retailers in India (7). Collecting packs from retailers was relevant given that single cigarettes dominate the illicit market in India. Smokers who buy single cigarettes would be unable to provide a pack in a pack swap or street intercept survey, so collecting littered packs from the ground would undercount sales of single cigarettes. The feasibility of this method is dependent on the relationship between researchers and retailers (enhanced trust) and the efforts taken to ensure confidentiality. In some countries, retailer compliance with this research method might be strained because of concerns regarding confidentiality, since retailers may face criminal and civil penalties, depending on the research findings. Advantages and disadvantages of littered-pack surveys Littered-pack surveys are generally advantageous for governments because they facilitate comparison with industry estimates. This is one of the most-preferred methods because it yields estimates that are less likely to be biased due to issues of social desirability, recall error and confidentiality that plague survey research, and they are much less expensive than face-to-face interviews used in smoker intercept or household surveys. However, there are some issues regarding these surveys, espe- cially in high-income countries, including the inability to differentiate between tax avoidance and tax evasion (8). For example, a pack in Berlin that bears a Vietnamese tax stamp may have been smuggled in mass quantity or brought in by a temporary visitor. Researchers have circumvented this issue and broadened the umbrella to measure cigarette tax noncompliance considering the potential biases introduced by tourism. Another disadvantage of littered-pack surveys is that larger budgets are needed to employ field researchers to collect, code and analyse the data. Not all countries employ tax stamps on their cigarette packaging, which may make it difficult to measure tax compliance. These surveys also can underestimate the markets in low-income countries such as India, where the main item of illegal trade is single cigarettes (7). The surveys may also overestimate illicit trade if littering behaviour is correlated with willingness to engage in illicit trade. Finally, littered-pack surveys and butt collections provide information on the proportion of butts and packs that are illegal, not the proportion of smokers that purchase illegal cigarettes (6). Key studies for readers to refer to for additional guidance: Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38–i43. Merriman D. The micro-geography of tax avoidance: evidence from littered cigarette CHAP T ER 4. PO LI T I C AL ECO N OMY 257 packs in Chicago. Am Econ J Econ Policy. 2010;2(2):61–84. Stoklosa M., Paraje G., Blecher E., A Toolkit on Measuring Illicit Trade in Tobacco Products. A Tobacconomics and American Cancer Society Toolkit. Chicago, IL:Tobacconomics, Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/621/uic-illicit-trade-tool-kit-eng-v2.0-2. pdf, accessed 18 February 2021). A4.1.4 SELF-REPORT POPULATION SURVEYS Self-report surveys, when distributed to a representative sample of the population, can provide meaningful data on the prevalence of tax noncompliance. The surveys can be distributed to individuals or households in various ways, including face-to-face, telephone, mail and internet. Questions that specifically address illicit purchases can be added as supplementary questions to existing health or tobacco surveys. Some countries include such questions in their adult and youth tobacco surveys to estimate tax evasion/avoidance. For example, Canada’s annual Youth Smoking Survey asks smokers about the frequency of their purchases of First Nations/Native brand cigarettes (9-10). Davis et al. used data from the New York Adult Tobacco Survey to measure the source of purchase of the last cigarette pack purchased (i.e. Native American Reservations, lower-tax neighbouring states or countries, toll-free telephone numbers, the internet, duty-free shops) and the price paid (11). Twenty- eight nations currently use surveys to measure tax noncompliance as part of the ITC Project (12). Similar analyses can be conducted using questions from the Global Adult Tobacco Use Surveys. For example, Iglesias et al. used the Brazil Global Adult Tobacco Use Surveys to compare self-reported prices with a defined threshold retail price to estimate the proportion of illicit cigarette use among smokers in Brazil (13). Countries are encouraged to use existing global health surveys or to incorporate similar types of questions pertaining to illicit trade in their annual health surveys. Asking respondents about price paid per pack (including taxes), brand name and location where cigarettes were purchased (e.g. duty free shop, unlicensed vendor, internet) can contribute to a better understanding of the illicit tobacco trade. Advantages and disadvantages of self-report surveys Self-report surveys can be repeated over time to measure purchasing trends and progress associated with increases in cigarette taxation. Well-designed surveys can also provide generalizable estimates at the national level. Depending on the size of the sample, a self-report survey can provide comparable data across geographies that can help governments target resources. For example, findings that illicit cigarettes are more common in urban areas could lead to additional education campaigns and targeted enforcement. 258 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Limitations of self-report surveys include the possibility of bias due to the social stigma associated with participating in the illicit trade, which could lead survey respondents to underreport participation. Additionally, surveys may be unable to gauge whether individuals are associated with tax avoidance versus tax evasion. Finally, there is evidence that self-report household surveys may underrepresent smokers. Key studies for readers to refer to for additional guidance: Callaghan RC, Veldhuizen S, Ip D. Contraband cigarette consumption among adolescent daily smokers in Ontario, Canada. Tob Control. 2011;20(2):173–174. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany (NY): New York State Department of Health, Tobacco Control Program; 2006. A4.1.5 COVERT-PURCHASES SURVEYS A number of studies in high-, middle- and low-income countries use covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes in public and semi-private spaces (14-17). This method is also used by the tobacco industry in the United States to identify retailers who sell counterfeit cigarettes (18). Covert-purchases surveys do not provide estimates of the size of the illicit trade (i.e. market volume). Instead, they serve as a surveillance tool to identify where illicit cigarettes are sold and the extent to which they have infiltrated legal businesses. For example, a covert-purchases survey can examine whether illicit cigarettes are sold through legal retailers. It can also be used to measure compliance with emerging tobacco control policies that focus on, for example, product standardization or new regulations on flavours (e.g. plain packaging or bans on flavoured tobacco products). Covert-purchases surveys use trained researchers to visit a selected sample of retailers and directly purchase or inquire about the availability of illicit tobacco products. Retailers are not informed about the goals of the studies. Methods for determining the availability of illicit product vary. For example, in some studies, covert buyers do not directly inquire about illicit products. Instead, they purchase packs of tobacco products, paying full price, to determine whether retailers are selling illicit products under the guise that they are licit (14). The research team then examines the packs to determine whether they are legal. In the United States, researchers have observed that some consumers are paying full price for illicit untaxed packs smuggled from lower-tax states (14). Other research protocols directly ask retailers for illicit products (14,16-17). In Guatemala, Arevalo et al. specifically asked retailers for “imported cigarettes” (17). The ways covert buyers ask for illicit products may also vary geographically. For example, in some countries covert buyers may ask for “cheaper” packs or for illicit whites such as Jin Ling. CHAP T ER 4. PO LI T I C AL ECO N OMY 259 Advantages and disadvantages of covert-purchases surveys Covert purchasing allows researchers to directly identify sources of illicit cigarettes. It also allows them to measure and test the dynamics between buyer and seller. For example, researchers can experiment to see if repeated attempts to purchase products increase the likelihood of purchase (known as the familiarity protocol) (16). One methodological challenge associated with covert purchases is that it is difficult to create a sampling frame for illicit sources because some may be unknown (e.g. pubs or homes). The traditional approach is to make purchases in legal outlets, which may bias estimates. Another issue with this method is that it is difficult for buyers (also called raters) to purchase products if they are unfamiliar with the seller or do not fit the typical demographics of purchasers. Therefore, researchers using covert- purchases surveys must have detailed knowledge of the marketplace, including the ways individuals specifically ask for illicit tobacco products, and they must know whether they mirror the demographics of the neighbourhood. For example, in a study of South Bronx smokers, von Lampe et al. found that smokers looked for certain clues to assess whether they were being sold illicit cigarettes (19). Overall, this method can be quite costly because it requires training researchers, travelling to retailers and purchasing product. Covert-purchases surveys do not enable researchers to estimate the level of illicit trade, but they can provide information on availability of supply. Key studies for readers to refer to for additional guidance: Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Global Health. 2016;12(1):78. A4.1.6 SEIZURES OF GOODS Seizures are the result of enforcement activity carried out by local, national and international organizations that confiscate tobacco products that are illegally manu- factured, transported and sold. Seizures are meant to reduce the profits associated with illicit trade by confiscating proceeds (e.g. cash, cars or houses) and the tools of the trade (e.g. print and tobacco machinery). Seizures can occur at various points in the supply chain. Seizure data are often tallied by the responsible agencies and used to measure program effectiveness or as justification for requesting additional resources (e.g. personnel). Some of the data may be supplied to international customs organiza- tions, including the WCO (20). The quality of recordkeeping varies. For example, some agencies may maintain criminal files in databases that detail dates of seizure, brand names and laboratory testing. 260 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Seizures provide preliminary data on the scope of criminal activity and can help identify key trends to guide law enforcement agencies’ efforts. For example, seizures can identify trends on the modus operandi of smugglers, including transporta- tion methods (e.g. sea cargo versus trucks), point of entry and brand preference. Seizure statistics can also be used as a preliminary test to measure the efficacy of interventions. For example, Stoklosa and Ross used seizure data from the Canadian province of Nova Scotia to test the impact of a 2015 menthol ban. He found no statistically significant change in the number of menthol cigarettes seized before and after the ban (1). Advantages and disadvantages of seizures of goods Generally, seizure statistics can be readily obtained from law enforcement agencies through formal requests to agency gatekeepers (e.g. public information officers). Seizure data, however, generally do not provide a representative picture of illicit activity. For example, certain geographies may yield higher seizures because that is where the bulk of operations are being conducted. Police agencies may focus on certain geographies (e.g. locations near borders) rather than randomly inspecting, and their findings may be limited to those specific regions. Seizure data may also be skewed by the type of investigation procedures utilized. Large seizures may be the result of long-term investigations (i.e. wiretaps or culling confidential infor- mants), while smaller seizures may come from anti-smuggling cases that involve cross-border purchases of low quantities of cigarettes (less than 1 000) (2). Seizures can also be skewed by industry cooperation with law enforcement agencies. For example, the tobacco industry may be more likely to support law enforcement on counterfeit seizures rather than smuggling cases because counterfeiting impacts their brand integrity. A4.2 RESIDUAL METHODS Because the illicit tobacco trade is often decentralized, it can be difficult to observe directly. However, researchers are sometimes able to make inferences about its size without direct observation by comparing observed tobacco tax revenues with the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. The difference between observed and hypothetical revenues is called the residual and can be used as an indicator of the magnitude of illicit trade. Even when the residual is only an approximate measure, changes in its size may be a reliable indicator of changes in the size of the trade. When actual tobacco tax revenues are reliably observed, the main challenge for residual methods is that of producing accurate estimates of the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. CHAP T ER 4. PO LI T I C AL ECO N OMY 261 A4.2.1 GAP ANALYSIS Gap analysis is the preferred residual methodology because it is intuitive, straight- forward and relatively easy to explain to policy-makers and the general public, and it has been widely employed in government studies (4). Researchers using gap analysis compare survey-based self-reported consumption data with observed (usually administrative) data on tax-paid sales. The basic premise is that if both self-reports and observed data are accurate, any difference between reported consumption and tax-paid sales can be explained by legal imports of non-taxed cigarettes (such as duty-free sales), exports of taxed cigarettes, tax evasion or tax avoidance. The greatest research challenge in implementing gap analysis – as with most residual methods – is obtaining reliable and accurate estimates of tobacco consump- tion. In its simplest implementation, gap analysis calculates the residual as the difference (which should be a minimum of zero) between the amount of tobacco consumption reported in surveys and tax-paid sales, which are generally available from administrative sources, minus exports. This simple calculation, however, is generally flawed, since surveys of reported tobacco consumption underestimate true consumption. Underreporting of tobacco consumption may result from survey respondents’ reticence about disclosing behaviour that is viewed as unhealthy and potentially socially undesirable. Depending on the legal and cultural context, cer- tain groups (e.g. women or youth) may be more likely than others to underreport consumption. Other groups (e.g. rebellious young men and women) may accurately report or even overestimate consumption. Reuter and Majmundar measured actual consumption by total national taxed sales in the United States, where both legal untaxed imports and exports of taxed tobacco are widely believed to be very small, and found that the ratio of self-reported consumption to actual consumption was only 65% (4). After incorporating this survey underreporting into their analyses and considering the evidence from their gap analysis and the literature, Reuter and Majmundar found that the illicit market in the United States, which largely consists of avoidance or evasion of subnational state taxes, is between 8.5% and 21% of consumption (4). The higher range of the estimate is consistent with prior estimates using population-based pack observation studies (21). Researchers using gap analysis for countries or regions where legal imports of untaxed tobacco (such as duty-free products) or (legal or illegal) exports of taxed cigarettes are more significant should attempt to incorporate data about, or estimates of, these factors into their calculations. Legal untaxed imports of tobacco should be subtracted from reported consumption (after adjustment for underreporting), and exports of taxed tobacco should be subtracted from taxed sales. Obtaining data about legal untaxed imports and exports of taxed tobacco may be challenging, because these imports and exports may be the result of decentralized decisions of individual 262 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N travellers as they cross tax borders. Data on these activities will not necessarily be collected through normal administrative activities. Despite these challenges, gap analyses may still prove useful. For example, if researchers have reason to believe that misreporting of tobacco consumption and the amounts of legal untaxed imports and taxed exports are relatively stable or follow known trends (e.g. are falling) over time, gap analyses can be used to provide estimates or lower (or upper) bounds on illicit trade when several years of data on taxed sales and reported consumption are available. Data sources may be country tax administrators who have access to sales data and health departments that have access to population-level studies of reported tobacco con- sumption. In this context, multiple years of data on reported consumption and tax-paid sales can allow researchers to estimate changes in the size of the illicit trade even when it is difficult to measure the absolute level. Paraje used the 2008 Global Adult Tobacco Survey and the 2013 National Health Survey to measure reported tobacco consumption in Brazil (22-23). Advantages and disadvantages of gap analysis A major advantage of gap analysis is that when quality data are available, it is simple, easily reproduced and explainable to policy-makers and the general public. How- ever, high-quality data on reported consumption may not be available, especially in low-income countries. In many cases, gap analysis does not provide reliable information on the size of the illicit market but only on changes in the size over time (22). Additionally, some low-income countries may not have reliable estimates of tax-paid cigarette sales, and secondary data repositories of cigarette sales may not be transparent about their methodology (24). Another disadvantage of gap analysis is that it generally cannot be used to obtain separate estimates of tax avoidance and tax evasion. Biased estimates may also result if surveys of tobacco consumption are not representative of the population (25). Moreover, it is generally not possible to quantify the precision of the estimates or uncertainty associated with the estimates, because of both statistical uncertainty resulting from the use of samples to imperfectly represent populations (e.g. the share of the population that smokes) and uncertainty about key facts such as the degree to which survey respondents understate their tobacco consumption. Key studies for readers to refer to for additional guidance: Szklo A, Iglesias RM, Carvalho de Souza M, Szklo M, Maria de Almeida L. Trends in illicit cigarette use in Brazil estimated from legal sales, 2012–2016. Am J Public Health. 2018;108(2):265–269. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine and Tob Res. 2019;21(8):1079–86 CHAP T ER 4. PO LI T I C AL ECO N OMY 263 A4.2.2 ECONOMETRIC MODELLING There is a long tradition of using data to estimate parameters of demand functions that relate the quantity of goods consumed to the prices faced by consumers, their incomes and other variables. Because of the addictive nature of tobacco – and because of important public health and public policy concerns relating to tobacco use – economists have paid particular attention to the estimation of cigarette demand functions (26). As the literature on this topic developed, it became apparent that taxed tobacco sales would be a biased indicator of tobacco consumption if some consumers obtained their tobacco in illicit markets. Similarly, the price of cigarettes in the legal market might overestimate the price paid by consumers if some sales were not tax-paid. While economists generally cannot observe sales in the illicit market, they have been able to develop models that predict conditions under which consumers avoid tobacco taxes. They reason that the relative size of illicit tobacco markets depends primarily on two variables: the relative price of taxed and untaxed consumption and the ease of obtaining lower-cost (untaxed) tobacco. Other variables, includ- ing the social stigma from evading tax laws and the perceived relative quality of illicit tobacco, could also influence the demand for it. While illicit trade cannot be directly observed, it can be estimated from the difference between tax-paid sales and predicted consumption. Tax-paid sales can be less than predicted consump- tion when retailers or consumers evade taxes. They can be greater if some tax-paid cigarettes are bought within the jurisdiction and then consumed in areas where after-tax prices are higher. Econometric modelling estimates of illicit trade must be tailored to the situation in the country that is being studied, and therefore the data requirements may differ substantially from case to case. Researchers using this method should be familiar with the literature and should also understand the conditions in the areas they are researching. They must always include some measure of tobacco consumption or sales and some measure of the price of tobacco in the home country, as well as other variables (e.g. income) that are known to affect the demand for tobacco. It is also generally necessary to include variables that measure the availability and relative price of illicit tobacco, which can often be measured by comparing tobacco taxes in the home country with those in areas that are the source of illicit tobacco. Advantages and disadvantages of econometric modelling The major advantage of econometric modelling is that it is consistent with a long tradition of economic theory and practice, and the quality of the modelling techniques and empirical estimates can therefore be evaluated against widely accepted criteria. Empirical analyses provide estimates of price elasticities, income elasticities and price elasticities of tax avoidance. A substantial literature base makes it possible to 264 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N rigorously quantify uncertainty about the estimates and to test their robustness to various assumptions made in the modelling process. Estimates can be compared to other estimates available in the literature, and the results of these analyses can be used to simulate the impacts of policy changes (including tax and/or enforcement increases) on both consumption and tax avoidance. Because this methodology provides direct estimates of the uncertainty of the results, researchers can specify their level of confidence in the findings. A shortcoming of econometric modelling is that it requires high-quality data on a variety of important variables over a period of time, as well as advanced econo- metric modelling expertise. Also, because results from the econometric models are based on statistical inference and economic theory rather than direct observation (e.g. the proportion of packs without tax stamps), it can be difficult to explain to policy-makers and the general public. Key studies for readers to refer to for additional guidance: Becker GS, Grossman M, Murphy KM, (1994). An empirical analysis of cigarette addiction. Amer Econ Review. 1994;84(3):396–418. Schafferer C, Yeh CY, Chen SH, Lee JM, Hsieh CJ. A simulation impact evaluation of a cigarette excise tax increase on licit and illicit cigarette consumption and tax revenue in 36 European countries. Public Health. 2018;162:48–57. A4.2.3 EXPERT OPINION (KEY-INFORMANT SURVEYS AND INTERVIEWS) Insight on illicit trade dynamics can come from experts in the field, including researchers (e.g. in economics, criminal justice and public health), taxation depart- ments, enforcement agencies, product manufacturers, wholesalers and retailers. Other key informants include journalists and academics who have secured confi- dential informants. Experts can provide novel information about emerging trends (e.g. new smuggling routes). In some cases, researchers can obtain interviews with incarcerated or active offenders (27-29). For example, researchers studying cigarette smuggling in eastern Africa conducted interviews with more than 150 Ugandan tobacco smugglers (29). Experts can be queried through surveys or semi-structured interviews. When sampling frames are available (e.g. directories of tax department employees), surveys are more expedient than interviews. However, when experts are hard to find, non- random sampling strategies coupled with interviews are recommended. Identifying experts may require recruiting a gatekeeper who is tasked with helping researchers find additional experts; or purposive sampling, where individuals are identified based on set criteria (e.g. they are taxation experts employed by local governments) (30). CHAP T ER 4. PO LI T I C AL ECO N OMY 265 Advantages and disadvantages of key-informant surveys and interviews Informant interviews can be a useful starting point for identifying trends in the marketplace (e.g. venues where illicit cigarettes are sold or modes of entry). One of the disadvantages of relying on informants is that the information solicited from them may not be generalizable. Expert knowledge may be outdated or limited by the informants’ experience. Furthermore, the opinions of experts are subjective and may be biased by the experts’ employment status and the sampling methods used. For example, persons working in law enforcement may overestimate the extent of bootlegging in order to secure additional funding for future operations. Similarly, manufacturers looking to defeat taxes may overestimate the illicit trade to illustrate the links between taxation and illicit behaviour. Alternatively, tobacco control advocates may underestimate illegal market measures in order to support the argument that taxes do not increase illicit trade. Key studies for readers to refer to for additional guidance: Joossens L, Raw M. Cigarette smuggling in Europe: who really benefits? Tob Control. 1998;7:66–71. doi:10.1136/tc.7.1.66 PMID: 9706757. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20(3):226–232. A4.3 MIXED AND MULTIMETHOD STUDIES Given the shortcomings of the aforementioned methods for assessing the nature and size of the illicit tobacco trade, governments may want to validate their findings by using mixed or multiple methodologies. Mixed methods use two methodological paradigms, qualitative and quantitative, as tools for exploration and explanation. For example, mixed method studies can use littered-pack surveys to measure the size of the market along with self-report surveys of smokers to understand patterns of purchasing, including sources, frequency and social norms. For example, Stoklosa and Ross estimated the share of the illicit market in Poland using a population- based self-report survey and a littered-pack survey (1). Using both types of survey simultaneously enables governments to assess their validity in estimating the size of the illicit market. Alternatively, governments can employ multimethod research, i.e. the use of multiple methods that are similar in tradition (e.g. focus groups and semi-structured interviews) (31). Saenz de Miera et al. used face-to-face interviews (households), litter collection and observation of single-stick sellers, which enabled them not only to cross-validate the two major methodologies, but also to see if the brand of the single stick was a good measure of licit versus illicit trade (33). 266 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Advantages and disadvantages of mixed and multimethod studies Mixed and multimethods studies enable researchers to check the validity of their findings. Multiple methods are preferred in contexts where illicit trade estimates are politicized. For example, low estimates may be challenged by the tobacco indus- try, while high estimates may be challenged by tobacco control researchers and/or proponents. Mixed and multiple methods (e.g. littered-pack surveys and informant interviews) can enable governments to understand the situational context in which the illicit trade operates, including the actors involved and venues of sale. One disadvantages of using mixed and multiple methods is cost. Governments that are constrained by tight budgets may choose to use a single method that provides the most accurate information. However, given the issues faced by each method, this may not be feasible – each method has limitations. Instead, governments can pair a high-cost method with a lower-cost method (e.g. pairing interviews with empty pack surveys, or law enforcement seizure data with face-to-face consumer surveys). Key study for readers to refer to for additional guidance: Zaloshnja E, Ross H, Levy DT. The impact of tobacco control policies in Albania. Tob Control. 2010;19:463–468. CHAP T ER 4. PO LI T I C AL ECO N OMY 267 REFERENCES 1. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 2. Joossens L, Lugo A, La Vecchia C, Gilmore AB, Clancy L, Gallus S. Illicit cigarettes and hand-rolled tobacco in 18 European countries: a cross-sectional survey. Tob Control. 2014;23:e17–e23 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC3812425/pdf/nihms491463.pdf, accessed 2 February 2021). 3. Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–7. 4. Reuter P, Majmundar M. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Washington (DC): National Academies Press; 2015. 5. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 6. Barkans M, Lawrance KA. Contraband tobacco on post-secondary campuses in Ontario, Canada: analysis of discarded cigarette butts. BMC Pub Health. 2013;13(1):335 (https://bmcpublichealth. biomedcentral.com/articles/10.1186/1471-2458-13-335, accessed 2 February 2021). 7. John RM, Ross H. Illicit cigarette sales in Indian cities: findings from a retail survey. Tob Control. 2018;27(6), 684–688. 8. Davis KC, Grimshaw V, Merriman D, Farrelly MC, Chernick H, Coady MH, et al. Cigarette trafficking in five northeastern US cities. Tob Control. 2014;23(e1):e62–e68. 9. Callaghan RC, Veldhuizen S, Leatherdale S, Murnaghan D, Manske S. Use of contraband cigarettes among adolescent daily smokers in Canada. CMAJ. 2009;181(6-7):384–6 (https://www.cmaj.ca/content/ cmaj/181/6-7/384.full.pdf, accessed 2 February 2021). 10. Guindon GE, Burkhalter R, Brown KS. Levels and trends in cigarette contraband in Canada. Tob Control. 2017;26(5):518–25 (https://www.researchgate.net/publication/307890937_Levels_and_trends_in_ cigarette_contraband_in_Canada, accessed 2 February 2021). 11. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany: New York State Department of Health; 2006 (https://www. health.ny.gov/prevention/tobacco_control/docs/cigarette_purchasing_patterns.pdf, accessed 2 February 2021). 12. International Tobacco Control Evaluation Project (2018). Surveys. Waterloo: University of Waterloo; 2018 (http://www.itcproject.org/surveys, accessed 12 October 2020). 13. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 14. Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. 15. Brown J, Welding K, Cohen JE, Cherukupalli R, Washington C, Ferguson J, et al. An analysis of purchase price of legal and illicit cigarettes in urban retail environments in 14 low-and middle-income countries. Addiction. 2017;112:1854–60 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5600117/pdf/ADD- 112-1854.pdf, accessed 2 February 2021). 16. Scollo M, Bayly M, Wakefield M. Availability of illicit tobacco in small retail outlets before and after the implementation of Australian plain packaging legislation. Tob Control. 2015;24(e1):e45–e51. 17. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Globalization and Health. 2016;12(1):78 (https://globalizationandhealth. biomedcentral.com/articles/10.1186/s12992-016-0219-z, accessed 2 February 2021). 18. Phillip Morris USA Inc. v. Shalabi. United States, District Court, C.D. California; 2004 (https://www. casemine.com/judgement/us/5914b6cbadd7b0493477b3da, accessed 2 February 2021). 19. von Lampe K, Kurti M, Johnson J, Rengifo AF. ‘I wouldn’t take my chances on the street’ navigating illegal cigarette purchases in the South Bronx. J Res Crime Delinq. 2016;53(5):654–80 (https://www. 268 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N researchgate.net/publication/298515449_I_Wouldn’t_Take_My_Chances_on_the_Street_Navigating_ Illegal_Cigarette_Purchases_in_the_South_Bronx, accessed 2 February 2021). 20. Customs and Tobacco Report 2009. Brussels: World Customs Organization; 2009. (http://www.wcoomd. org/en/media/newsroom/2010/june/~/media/83967DFEB9F74D388924A4C61F279DC4.ashx, accessed 12 October 2020). 21. Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ, et al. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61–i66 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3984758/pdf/nihms567943.pdf, accessed 2 February 2021). 22. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine Tob Res. 2019;21(8):1079–86. 23. National Survey of Health. Instituto Brasileiro de Geografia e Estatística; 2013 (in Portuguese) (https:// www.ibge.gov.br/en/statistics/social/health/16840-national-survey-of-health.html?=&t=downloads, accessed 27 November 2020). 24. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 25. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 26. Chaloupka FJ, Warner KE. The economics of smoking. In: Arrow KJ and Intriligator MD, editors. Handbook of Health Economics. Amsterdam: Elsiver; 2000. pp1539–1627. 27. Antonopoulos GA. Cigarette smugglers: a note on four ‘unusual suspects’. Glob. Crime. 2007;8(4):393–8 (https://www.researchgate.net/publication/248955367_Cigarette_Smugglers_A_Note_on_ Four_’Unusual_Suspects’, accessed 2 February 2021). 28. Antonopoulos GA. The Greek connection(s): the social organization of the cigarette-smuggling business in Greece. Eur J Criminol. 2008;5(3):263–88 (https://www.researchgate.net/publication/249752218_ The_Greek_ConnectionsThe_Social_Organization_of_the_Cigarette-Smuggling_Business_in_Greece, accessed 2 February 2021). 29. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20:226–32 (https://www.researchgate.net/publication/49809289_ Blood_cigarettes_Cigarette_smuggling_and_war_economies_in_central_and_eastern_Africa, accessed 2 February 2021). 30. Babbie, ER. The basics of social research. Boston: Cengage Learning; 2013. 31. Tashakkori A, Teddlie C., editors. Sage handbook of mixed methods in social & behavioral research. New York: Sage; 2010. 32. Saenz de Miera Juarez B, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 269 ANNEX 4.2 HOW ARE THE TOBACCO TAX REVENUES EARMARKED? The introduction of earmarking of tobacco tax revenue is almost always combined with an increase in excise taxes (or a new surcharge) rather than reallocation of existing revenues (1). Table A4.1 provides examples of the different approaches used by several countries to earmark tobacco tax revenues. Table A4.1 Approaches used to earmark tobacco tax revenues FUNDING SOURCES/ TYPE OF TAX EXAMPLES OF TAX BASE AND RATES As part of the excise system (tobacco, alcohol) Specific Republic of Korea: 841 won (US$ 0.75) per pack or 29% of the specific excise rate Costa Rica: 467.8 cólones (US$ 0.83) per pack or 100% of the specific excise rate Congo: 20 CFA francs (US$ 0.036) per pack or 50% of the specific excise rate Ad valorem Colombia: 10% of retail price (equivalent to 100% of the ad valorem rate) New levy (surcharge on the existing excise or completely new levy) Specific Egypt: additional 0.75 Egyptian pounds (US$ 0.042) per pack Ad valorem Thailand: surcharge of 2% over the excise tax base Botswana: new tobacco levy of 30% of the cost of production or CIF Percentage of excise revenue Cook Islands: 50% of revenues from the excise tax on tobacco Guatemala: 100% of revenues from the excise tax on tobacco Note: Conversions of amounts from the local currency were made using the official exchange rates from the IMF as of 31 July 2018 (date of the data collection). Source: (2). REFERENCES 1. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 2. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 270 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 271 CHAPTER 5. Best practices in tobacco tax policy and administration TAX POLICY Use excise tax increases to achieve the public health goal of reducing the death and diseases caused by tobacco use Extensive research has clearly demonstrated the effectiveness of higher tobacco product taxes and prices in reducing tobacco use and its harmful consequences, particularly among the poor and the young. In fact, tobacco excise tax increases are the single most effective and cost-effective policy for reducing tobacco use. Excise taxes are the most significant taxes applied on tobacco products because of their ability to raise both absolute and relative prices. Tobacco excise tax increases also generate sizeable new revenues that will be sustained in the short to medium term. In the long term, continued increases in tobacco taxes – coupled with implementa- tion of other evidence-based tobacco control policies and programmes – will lead to even larger reductions in tobacco use and its consequences. Include significant tobacco excise tax increases as part of a comprehensive strategy to reduce tobacco use Governments should adopt a comprehensive tobacco control strategy that includes objectives for reducing adult tobacco use and preventing youth tobacco use. Rais- ing excise taxes significantly is the most effective, as well as the most cost-effective, measure for reducing consumption. When combined with other demand reduction interventions, the impact of tax increases on tobacco use is even stronger. Such interventions include comprehensive smoke-free policies in all public spaces, total bans on tobacco advertising, promotion and sponsorship by tobacco companies, large graphic health warnings about the consequences of tobacco use, plain packaging, broad efforts to help current users quit and mass media public education campaigns. Implementation of a comprehensive strategy to reduce tobacco use leads to greater reductions in the harmful consequences of tobacco use, builds public and political support for higher taxes and maximizes the effectiveness of tax increases in achieving public health objectives. 272 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Involve the competent authority from the start when considering the revision of a tax policy Competent authorities such as tax administrations and customs authorities are key partners in the effective implementation of a tax policy. Policy-makers need to ensure that those competent authorities are consulted and involved in the tax policy revision process so that their concerns about the impacts of policy change on enforcement can be taken into account from the beginning. This can also help identify and address possible loopholes early on in the enforcement process. Coordination among relevant bodies, including close cooperation and sharing of information, will optimize enforcement of tax policy and tax collection. To streamline the process of cooperation and exchanges of information, a basis in law needs to be established. Additionally, the involvement of tax administration authorities in the entirety of the tax revision policy process is important to ensure effective implementation of the policy. Promote greater policy coherence across sectors such as agriculture, industry, trade, finance and labour Greater multisectoral integration and policy coherence is needed at the country level to achieve effective health improvements. In particular, it is important to ensure that public policies and interventions in non-health sectors (e.g. agriculture, industry, trade, finance and labour) do not act against the intended public health impact of tobacco control and taxation (such interventions include providing subsidies to tobacco growing or manufacturing). TAX DESIGN Tax structure matters and simpler is better Complex tax structures are difficult to administer, create opportunities for tax avoid- ance and evasion and are less effective than simpler structures in achieving public health and revenue goals. Simplifying the structure of tobacco excise taxes will facilitate tax administration, reduce tax avoidance and evasion, enhance revenues and have a greater impact on tobacco use by reducing incentives to substitute among tobacco products or brands in response to tax increases. Countries with multiple tiers of tobacco tax rates based on product characteristics (e.g. price level, length, weight, type of tobacco) should reduce and eventually eliminate these differential tax rates. An appropriate transition strategy is to reduce the variations in tax rates over time with the aim of implementing a uniform tax (i.e. a single rate applies whether excise is ad valorem or specific) on a given tobacco product. Applying a uniform tax to all brands of a given tobacco product also sends a clear message that they are equally harmful. CHAP T ER 5. B E S T PR AC T I CE S 273 Rely more on specific tobacco excises to drive price increases Greater reliance on specific excise taxes maximizes the impact of tobacco taxes on public health by reducing the gap in prices between premium and low-priced alternatives and limiting opportunities for users to switch down in response to tax increases. For countries that currently rely on an ad valorem tax, an appropriate first step would be to shift to a mixed system by adding a sizeable specific component or introducing a high minimum specific excise tax (an excise tax floor). For countries that rely on a mix of ad valorem and specific taxes, the specific tax component should be increased regularly so that it accounts for a greater share of the total excise tax. Increase tobacco taxes significantly to reduce the affordability of tobacco products To maximize the public health impact of higher tobacco taxes while at the same time generating higher revenues, governments should significantly raise taxes to increase prices and reduce the affordability of tobacco products. In many LMICs, tobacco use increases with incomes, and since incomes rise faster than tobacco product prices, these products are becoming more affordable. To reduce afford- ability, tax increases need to result in real price increases that are higher than the increases in real incomes. Where revenue increases are a goal, rely on regular excise tax increases If governments want to increase tobacco revenues, they must increase excise taxes regularly. From the tax revenue perspective, the important determinant is the tax base elasticity, which has three key components: the price elasticity of demand of tobacco, the share of the tax in the retail price and the degree of pass-through of the excise tax rate increase on to retail price. Tax increases will increase revenues at least in the short to medium term, because demand is price inelastic, tax levels are generally low as a proportion of retail prices and the pass-through of tax increases on to retail prices is unlikely to be higher than the tax increase itself (i.e. there is no overshifting). In addition, increasing tax rates is the only policy measure that can reverse reduced revenues in a declining market that has strong tobacco control policies. Automatically adjust specific tobacco taxes for inflation and income growth Unless specific tobacco taxes are regularly adjusted, their real value will fall over time as general price levels increase. When this happens, their effectiveness in reducing tobacco use will be diminished. Governments should establish a mechanism for automatically adjusting specific taxes to keep pace with inflation. Recently, some governments have begun to extend this indexation to include income growth as well, further ensuring that tobacco does not become more affordable over time. 274 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Pricing regulations cannot be considered an alternative to excise tax. However, in some specific contexts, pricing regulations could be used in conjunction with excise taxes to help ensure the effective implementation of tax increases In certain contexts where increasing taxes is challenging or the tax structure is weak, non-tax policies such as pricing regulation (specifically, minimum mark-ups and price floors/minimum prices) may be seen as a second-best alternative to ensure a high price level and dissuade consumption of tobacco products. These policies, however, do not necessarily lead to the desired price level, nor do they protect consumers and government from industry manipulation. However, in the context of powerful multinationals that sell brands across all market segments and could easily undershift a tax increase to cheaper brands – or where price promotions cannot be banned – minimum price policies may help increase the effectiveness of tax increases, especially if the minimum prices are increased regularly. Implement nontax policies affecting price levels, such as banning promotional discounts for tobacco products and the sale of single sticks of cigarettes The banning of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Do not allow concerns about the inflationary impact of higher tobacco taxes to deter tax increases Given that wages or some government spending may be tied to a price index, govern- ments can reduce concerns about the inflationary impact of a tobacco tax increase by using a price index that excludes tobacco products. TAX PARITY Tax all tobacco products in a comparable way Increasing excise taxes on some tobacco products but not on others results in changes in the relative prices of these other products. This induces substitution towards relatively less-expensive products – for example, from expensive manufactured cigarettes to other, cheaper tobacco products such as RYO tobacco, bidis, cheroots or chewing tobacco. As a result, the overall reduction in tobacco use is smaller than it would have been had all taxes increased by comparable amounts. Comparable increases in the taxes on all tobacco products maximize the public health impact of tobacco tax increases by minimizing opportunities for substitution. Moreover, increases in taxes on all tobacco products will generate larger increases in revenues. CHAP T ER 5. B E S T PR AC T I CE S 275 Strictly regulate new and emerging tobacco and nicotine products where they are not banned and impose an excise tax In recent years, the world has been experiencing the rise of new and emerging tobacco and nicotine products including ENDS, ENNDS and HTPs. The tobacco industry claims these new products are safer than traditional tobacco products, but the evidence so far suggests that they could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of newer products – as well as initiation, smoking cessation and switching behaviour among different socioeconomic groups – are not yet clear. Best practices for taxing new and emerging tobacco and nicotine products, based on current knowledge, are that: 1. HTPs should be taxed at the same level as cigarettes and, in terms of structure, through a specific excise per unit regardless of tobacco content. HTPs contain tobacco and should be treated as a tobacco product. 2. ENDS/ENNDS products should be taxed in a manner that discourages up- take by youth and non-users. Nicotine- and non-nicotine-delivery systems containing e-liquids should be taxed equally. 3. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While these newer products create additional challenges for tobacco control, it is important to remember that cigarettes remain by far the predominant tobacco product and that raising taxes and prices on cigarettes – and thereby reducing their use – should remain the top priority. MONITORING AND EVALUATION Know your market Know your market well. The type of tax structure you choose and the impacts it will have on consumption and tax revenue are shaped by the particular dynamics of your market. Understanding the nature and degree of competition in your market is vital to selecting the appropriate type of tax structure and policies to achieve your public health and revenue objectives. This knowledge will also facilitate more accurate estimates of the impacts of a tax increase, as well as better anticipation of industry responses. Assess the impact of your policies to design and implement the most effective tobacco excise tax policies Monitoring and evaluation are essential for effective tobacco taxation, and they 276 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N should be built into the initial design – or redesign – of tobacco tax policies. A number of tools exist to help policy-makers pre-emptively assess the effects of a proposed tobacco control policy on consumption, smoking prevalence and lives saved. The WHO TaXSiM uses target simulations to assist governments in predict- ing how specific tax changes will impact consumer prices, consumption and tax revenue in their market. Adopt indicators that help you measure improvements in tax policy and its impact Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the improvement of their policies and determine if those policies have an impact on tobacco use over time. The tax share of the retail price for a particular tobacco product is a key indicator that should be used in conjunction with an af- fordability indicator. A recommended target for countries to aspire to is to have an excise tax that represents at least 70% of the retail price of tobacco products. Another useful indicator to assess the performance of the tax policy overall is the use of a tax scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). TAX ADMINISTRATION Implement best practice approaches in general tax administration to make tobacco tax administration more effective and efficient Best practice approaches include (1) defining clearly the roles and responsibilities of competent authorities, (2) ensuring effective coordination among relevant bodies at the national and international levels and (3) undertaking evaluation of performance and accountability against pre-agreed indicators to identify points for improvement. Ensure compliance and accuracy of information on the tax compliance cycle To achieve this, implement the following actions: • Require licences for manufacturing, importing, exporting, retailing, growing, transporting, wholesaling, brokering, warehousing and distributing tobacco products. This will help secure the supply chain while obtaining valuable infor- mation, e.g. through access to companies’ accounting and inventory systems. • Make sure all persons and entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment keep complete and accu- rate records of all relevant transactions and details of materials used in the production of tobacco products. CHAP T ER 5. B E S T PR AC T I CE S 277 • Ensure that tax declarations collect as much information as possible on the taxpayer. • Collect taxes close to the point of production and import to limit the number of taxpayers a competent authority needs to manage. • Maintain a system of authorization for warehousing to carry out controls in production and storage facilities to ensure that taxes are paid. • Use electronic methods, through the best available IT, for declarations and collection of taxes. This allows for cross-check of information provided in dec- larations with information from other government agencies and third parties. Ensure control and enforcement on the supply chain To achieve this, implement the following actions: • Include control and enforcement as a fundamental pillar in the strategic plan of the tax administration overall. • Use a risk-based approach by choosing defined targets for enforcement and control, such as those who have a higher probability of noncompliance. • In the licensing process, ensure that purchases from unlicensed suppliers or sales to unlicensed purchasers are not allowed. Ensure also that the validity of licences is limited in time and require renewals or reapplication to maintain a high level of control. • Use tax stamps with strong security features to reduce the risk of stamp counterfeiting. These markings facilitate the collection of excise taxes, audits and enforcement actions. • Implement a tracking and tracing system for tobacco products. A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, and monitoring and control- ling the movement of tobacco products and their legal status. • Implement anti-forestalling measures so that forestalling does not delay a tax increase and its intended effect on revenues and consumer behaviour. • Control import and export of tobacco products and manufacturing equipment by allowing only duly licensed natural persons or legal entities to conduct such activities. • Strengthen border control, e.g. by utilizing non-invasive tools such as X-ray scanners and dogs to detect tobacco products. • Limit or tightly control and, ideally, ban activities related to production and trade of tobacco products in tax-free zones to avoid opportunities for tax evasion. • Prohibit intermingling of tobacco products with non-tobacco products in a single container or any other similar transportation unit when removed from tax-free zones. 278 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Prohibit the sale to or import by international travellers of tax-free or duty- free tobacco products. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and adversely affect government revenues by creating a loophole in the tax structure. Clearly define procedures to follow after detecting illicit trade of tobacco • Take immediate action to seize and destroy smuggled and/or illicit tobacco and collect due taxes. • Ensure certain, swift and severe sanctions for those caught engaging in illicit trade in tobacco products, such as penalties, fines and withdrawal of licences. It can also be effective to consider illicit trade in tobacco products by law as a source of money-laundering. Become a Party to and/or implement the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products The WHO FCTC Protocol provides a blueprint of best practices and policies for dealing with illicit trade and should be part of any strategy to fight it. Implement, to the extent possible, the same rules and regulations for tax administration and enforcement for all tobacco products, as well as new and emerging nicotine and tobacco products Implement broad policies for ensuring a good tax system that will trickle down to good tax administration of tobacco products by: • ensuring proper resourcing of competent authorities; • having strict rules and regulations to detect corruption and to punish both personnel and taxpayers who are engaged in corrupt practices; and • ensuring a strong judicial system that is independent in fact and in perception, where disputes are solved quickly. The appeal process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially for illicit trade. POLITICAL ECONOMY Beyond the technical soundness of best practices in tax policy and administration, a critical factor in advancing tobacco taxes is the ability to get the political buy-in of the highest instances in the government. One key strategy is to address concerns around the political economy of tobacco taxation, which are often exploited by the tobacco industry to block major reforms. CHAP T ER 5. B E S T PR AC T I CE S 279 SCARE tactics The tobacco industry uses SCARE tactics to dissuade governments from implement- ing tobacco tax increases. These include smuggling and illicit trade (S), court and legal challenges (C), anti-poor rhetoric (A), revenue reduction (R) and employment impact (E). Best practices for countering these tactics are described below. S: Smuggling and illicit trade Do not allow concerns over the impact of increasing excise taxes on illicit trade in tobacco affect your decision to increase them. Rely on your own estimates of the level and nature of illicit trade and not on the industry’s estimates. Illicit trade in tobacco products continues to be a major concern for tax administrators because of the difficulties associated with accurate and independent measurement of it, as well as with its elimination. Industry figures provide a distorted understanding of the extent of the problem, along with a monocausal explanation of the link between illicit trade and tobacco taxation. It is therefore recommended that governments (1) assess independently and with the best statistical practices the size of the illicit trade to assess the scope of the problem; (2) address directly the country-specific institutional and/or governance challenges, including multilateral coordination, and improve tax and customs administrations practices; and (3) implement best practices to fight illicit trade, contained in the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products. Ideally, accede to the Protocol if not yet a Party. C: Court and legal challenges Do not let tobacco industry threats of court and legal challenges to tax increases or reforms prevent you from improving your tax policy. Closely follow legal requirements for design, procedure and consultation to strengthen your legal position and minimize the possibility that any challenge will be raised. Health-protective and non-discriminatory tobacco excise taxes are legally defensible, and industry threats will usually be baseless. Your legal position can be strengthened, however, by exercising care with a tax measure’s procedure, design and consultation: (1) determine the standard of consultation required under domestic law and any applicable international obligations; (2) distance the tobacco industry from the policy-making process to the extent that this is permissible; (3) avoid unnecessary and unjustified discrimination towards foreign tobacco products or investors in the design, implementation or enforcement of a tax measure; and (4) do not offer investment incentives in the form of inducements or contractual undertakings, as these may be binding in and of themselves or grounds for a challenge under an international investment agreement. 280 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A: Anti-poor rhetoric Do not allow concerns about the regressivity of higher tobacco taxes prevent tobacco tax increases. In recent years, there has been an overwhelming increase in the evidence concern- ing the positive distributional impact of tobacco taxes and tax increases. Indeed, tobacco taxation and tax increases are actually a progressive or pro-poor policy once these wider considerations are properly accounted for. In its effort to lobby against tax increases, the tobacco industry often claims that tobacco taxation will hurt the poor. This argument is based on the concept of regressivity in relation to taxation. Conceptually, a tax is regressive if it means lower-income people must pay a relatively greater proportion of their household income to meet the tax liability than wealthy people. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use that are disproportionately experienced by lower socioeconomic groups. Second, higher tobacco taxes and prices can induce behavioural change in the population, as reflected in the price elasticity of demand, which means that lower-income smokers will curtail their smoking the most and thus will benefit disproportionately in terms of health gains from reduced tobacco consumption and use. In fact, these broader considerations make tobacco taxation a progressive, rather than regressive, public health intervention. R: Revenue reduction Do not let fears of potential revenue reductions prevent you from increasing excise taxes on tobacco products. Tax increases, even in countries with already high taxes, bring in additional revenue. Arguments by tobacco control opponents that tax increases will not result in increases in revenue are unfounded. The relatively price inelastic nature of cigarette demand, combined with the low tax share and no overshifting of the tax, means that for most, if not all, countries, increases in revenues will accompany increases in taxes. If tax increases are carefully designed and tax administration is functional, it is extremely unlikely that tax increases will lead to revenue decreases. E: Employment impact Do not allow concerns about employment impact to prevent tobacco tax increases. The tobacco industry often seeks to frame tobacco taxes as an economic issue rather than a public health issue. Particular emphasis is placed on the alleged threat tax increases pose to employment in tobacco farming and manufacturing, as well as related industries. This so-called choice between health and jobs, however, is largely based on exaggeration. The tobacco industry exaggerates the importance of tobacco CHAP T ER 5. B E S T PR AC T I CE S 281 employment relative to total national employment and overstates the impact that domestic demand reduction from local taxes will have on tobacco farmers serving a global market. The argument used by the industry also ignores the fact that expendi- tures on tobacco do not disappear but rather are redistributed to other consumption that can produce a similar or higher number of jobs. Case studies demonstrate the possibility and methods for governments to support farmers in transitioning to other crops that provide similar and often better returns with greater sustainability. Earmarking Consider earmarking tobacco tax revenues for health-focused programmes, especially if it helps advance tobacco control efforts and, more specifically, efforts to implement large tobacco tax increases and tax reforms. This could have the additional benefit of funding health programmes where they are poorly funded or not prioritized. From a tobacco control perspective, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. Earmarking is a tool to improve the political economy of tobacco taxation; it is only a secondary issue, after the primary goal of reducing demand for tobacco. Evidence shows that public support for higher tobacco taxes is greater when at least some of the increased revenues are explicitly used to support health-focused programmes. Current evidence shows that the amounts effectively earmarked for health have been relatively small and therefore unlikely to introduce rigidity in government budgets. At the same time, in some countries, those funds have helped to imple- ment much needed underresourced health programmes. The payoffs will be seen in the future as fewer people fall ill and need less medical care for tobacco-related illnesses. Earmarking tobacco tax revenues to fund high-burden/low-priority health programmes could pave the way for raising awareness about the importance of such programmes and their effectiveness, thereby convincing governments to redefine their priorities and commit to including the programmes in their regular budget. 282 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) Focus on tobacco taxation’s purposes Tobacco tax policy should aim not only to increase revenues but also to decrease consumption and improve health. To both raise revenue and reduce consumption, you need to (1) simplify tobacco tax structures, (2) significantly increase rates to impact price levels, and (3) regularly adjust rates to at least account for inflation and income growth. Analyse your tax structure and identify its weaknesses You need to analyse and identify the problems of your current tax structure to know which steps to consider next. Which tax structure do you have: specific, ad valorem, mixed, or no excise? Identify the improvements to be made to the existing tax policy/structure Your present tax structure and tax situation will provide you with the steps you would ideally take next to achieve the aims in Step 1. Specific: 1. Ensure that the tax automatically adjusts for inflation and income growth effects. 2. Ensure that all price promotions are banned. Ad Valorem: 1. Ensure that the tax base of the ad valorem is retail price. 2. Introduce a high specific excise component (and a minimum specific excise). 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. Mixed: 1. Ensure that the tax base of the ad valorem component is retail price. 2. Ensure that you are using a high specific excise component and a minimum specific excise. 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. No Excise: 1. Introduce a high specific excise. 2. Ensure that the rate automatically adjusts for inflation and income growth effects. 3. Ensure that all price promotions are banned. Assess tobacco taxation’s political economy Reform must begin with an assessment of tobacco taxation’s political economy: (1) learn from past successes and failures – what went wrong, what went right, what you can do differently this time; (2) assess the reform’s strengths and weaknesses, likely opportunities and risks; (3) determine who the main supporters and opponents of reform inside and outside of government have been and may be, based on past reforms and current situation; and (4) anticipate argu- ments that will be used against the reform (refer to SCARE tactics). STEP 1 STEP 2 STEP 3 STEP 4 Prepare a plan for realizing the reform Focusing on the overall aims identified in Step 1, the steps for achiev- ing them identified in Steps 2 and 3 and the political economy around this reform as identified in Step 4, prepare your plan: 1. Be clear on the non-negotiable objectives for the reform and the trade-offs you are prepared to make to realize them. 2. Develop a plan to approach potential allies and win them over to the reform efforts. 3. Develop the counterarguments that will be needed in response to the SCARE arguments identified earlier. 4. Prepare the evidence you will need ahead of time. To do this, get support from academics and relevant intergovernmental agencies. Mobilize a coalition for reform 1. Formulate a strategic communications plan: aim for political support both at the highest levels and among the public (framing tobacco taxation as a health issue has helped win political support in many countries). 2. Identify champions in government: ensure that finance and health officials are on the same page; involve implementing departments, such as enforcement agencies, from the start. 3. Mobilize allies from academia, civil society and the private sector to counter the anticipated pushback from the tobacco industry, its proxies and its allies. Monitor and evaluate To make the most well-informed policy decisions, a reform effort should be monitored to assess its overall impact and its effect on key indicators; this will help identify issues to be fixed while also creating a strong evidence base for further reform efforts. Get and analyse the relevant data to better understand the market situation and its dynamics: 1. Monitor the market and its evolution (e.g. retail prices, duty-paid sales, market shares). 2. Get regular estimates of price elasticity (including cross-price elasticity), income elasticity and tax base elasticity to evaluate any changes in tobacco demand. Use relevant tools to assess the impact of the tax policy on consumption and revenue: 1. Use specific tools on the impact of excise on price, consumption and revenue (e.g. the WHO TaXSiM). 2. Use global tools to assess the tax increase’s impact on prevalence (e.g. the WHO ISPT). Monitor key indicators closely to assess improvements over time: 1. Tax as a percentage of retail price. 2. Change in affordability of tobacco products over time. 3. Change in the tobacco tax scorecard, which combines a mix of best practices in tax policy. 4. Change in sales, prevalence and illicit trade in tobacco products. 5. Improvements in MPOWER package achievement. STEP 5 STEP 6 STEP 7

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION WHO technical manual on tobacco tax policy and administration ISBN 978-92-4-001918-8 (electronic version) ISBN 978-92-4-001919-5 (print version) © World Health Organization 2021 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/ rules/). Suggested citation. WHO technical manual on tobacco tax policy and administration. Geneva: World Health Organization; 2021. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Document design by Ana Sabino. iii Contents Foreword v Acknowledgements vii Acronyms viii Executive summary xi CHAPTER 1. Why this manual? 1 CHAPTER 2. Tobacco excise tax policy 11 2.1 Global overview of tobacco tax practices 11 2.2 Designing excise tax policy 18 2.3 Domestic and regional policy integration 55 2.4 New and emerging nicotine and tobacco products 59 2.5 Conclusions 75 ANNEX 2.1 Countries that apply different types of excise tax structures 89 ANNEX 2.2 Analytics of the tax base elasticity 90 ANNEX 2.3 Elements of the devices that make up ENDS/ENNDS products 91 CHAPTER 3. Tobacco tax administration 93 3.1 Introduction 93 3.2 Institutional arrangements 94 3.3 The tax compliance cycle 102 3.4 Control and enforcement 115 3.5 Tax administration of other tobacco products 152 3.6 The broader elements of a good tax system 155 3.7 Conclusions 156 ANNEX 3.1 Composition of tobacco products 166 ANNEX 3.2 Example of forestalling and countermeasures 172 CHAPTER 4. Political economy 175 4.1 SCARE tactic S: Smuggling and illicit trade 176 4.2 SCARE tactic C: Court and legal challenges 198 4.3 SCARE tactic A: Anti-poor rhetoric or regressivity 212 4.4 SCARE tactic R: Revenue reduction 217 4.5 SCARE tactic E: Employment impact 228 4.6 Earmarking tobacco tax revenues to fund health 233 ANNEX 4.1 Methods to assess the nature and size of the illicit tobacco trade 253 ANNEX 4.2 How are the tobacco tax revenues earmarked? 269 CHAPTER 5. Best practices 271 TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) 283 iv v Foreword In 1999, the World Bank’s Curbing the Epidemic was the first report by an interna- tional organization to recognize that increasing tobacco excise taxes was the most effective and cost-effective measure to reduce tobacco use and save lives. Over the two decades since, the evidence base supporting this claim, especially in low- and middle-income countries, has been steadily growing. Meanwhile, the credibility of the tobacco industry’s arguments against tobacco taxation has been slowly waning. In short, health-promoting tobacco taxation has come of age, and the evidence has consistently shown that it is a win for public health, a win for revenue and a win for the economy overall. But we must be cautioned against complacency. Although the evidence on tobacco taxation is irrefutable and there are now signs that the tide is turning on the global tobacco epidemic, tobacco taxation was, in 2018, the WHO MPOWER1 measure that was least implemented at the highest level of achievement. Even more concerningly, cigarettes have become more, rather than less, affordable in many low- and middle-income countries over the past decade. Many countries set rates at insufficient levels and increase them too infrequently, while others still use complex and inefficient taxation structures. This failure to advance tobacco taxation able to effect significant price increases constitutes a loss for governments in revenues, a loss for public health and a win for the tobacco industry. To overcome this inertia, this manual charts the way forward for policy-makers, finance officials and others involved in tobacco tax policy development. It equips them with the information and evidence needed for the realization of their coun- tries’ tobacco tax policy objectives. It also analyses the tobacco industry’s tactics for influencing the political economy of tobacco taxation and shows the limitations and exaggerations of the arguments used against tax increases. The manual serves as an update of the 2010 WHO technical manual on tobacco tax administration by adding new evidence on the successes of tobacco taxation in all parts of the world and broadening its scope to capture more material relevant to developing and implementing more effective tobacco tax policy. 1 The WHO MPOWER package of technical measures and resources that comprises (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco. vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Its contributions are particularly timely in a COVID-19-stricken world. As the pandemic has been worsened by the global burden of noncommunicable disease, and revenue is now desperately needed, taxing tobacco should be more palatable than ever. This manual shows policy-makers how to seize this unique opportunity to use tobacco taxation to build back better, save lives and strengthen health systems while increasing revenue. Dr Naoko Yamamoto Assistant Director-General UHC/Healthier Populations Division vii Acknowledgements This manual was developed under the direction of Jeremias Paul Jr and Anne-Marie Perucic. The following contributed to the content of the manual (in alphabetical order): • WHO: Evan Blecher, Annerie Bouw, Mark Goodchild, Roberto Iglesias, Juliette McHardy, Jeremias Paul Jr, Anne-Marie Perucic and Robert Totanes. • External authors: Mauricio Cardenas (Columbia University), Sophia Delipalla (University of Macedonia), Luk Joossens (tobacco control expert, Belgium), Marin Kurti (Eastern Connecticut State University), Enrique Fanta (former World Bank senior specialist), David Merriman (University of Illinois at Chicago) and Jean Tesche (University of Cape Town). WHO would like to thank the following reviewers for their invaluable comments (in alphabetical order): • External: Jo Birckmayer (Bloomberg Philanthropies), Adriana Blanco Mar- quizo (WHO Framework Convention on Tobacco Control), Frank Chaloupka (University of Illinois at Chicago), Yoni Dekker (WHO Framework Conven- tion on Tobacco Control), Jeffrey Drope (University of Illinois at Chicago), Ceren Ozer (World Bank), Corne van Walbeek (University of Cape Town), Chonlathan Visaruthvong (Ministry of Finance, Thailand) and Rodrigo Santos Feijo (WHO Framework Convention on Tobacco Control). • WHO: – Headquarters: Douglas Bettcher, Itziar Belausteguigoitia, Ranti Fayokun, Joseph Kutzin, Benn McGrady, Vinayak Prasad and Susan Sparkes. – Regional offices: Nina Dela Cruz (WPRO), Fatimah El-Awa (EMRO), Charles Frasier (EMRO), Lee Lily Joung-Eun (WPRO), Jagdish Kaur (SEARO), Elizaveta Lebedeva (EURO), Maxime Roche (AMRO) and Rosa Sandoval (AMRO). WHO would also like to thank Amal Amoune-Naal for the administrative support, Alison Goldstein for the technical editing and Janet DeLand for the copy-editing. Production of this document has been supported by a grant from Bloomberg Phi- lanthropies. The contents of this document are the sole responsibility of WHO and should not be regarded as reflecting the position of Bloomberg Philanthropies. viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ACRONYMS AEO authorized economic operator AFRO WHO Regional Office for Africa AMRO WHO Regional Office for the Americas ATO Australian Taxation Office BAT British American Tobacco CCTV closed-circuit television CIF cost, insurance and freight COP Conference of the Parties CPI consumer price index CVA Customs Valuation Agreement (Thailand) DGCE Directorate General of Customs and Excise (Indonesia) DIY do-it-yourself ECBA World Bank Extended Cost-Benefit Analysis EIU Economist Intelligence Unit EMRO WHO Regional Office for the Eastern Mediterranean ENDS electronic nicotine delivery systems ENNDS electronic non-nicotine delivery systems EU European Union EURO WHO Regional Office for Europe FDA Food and Drug Administration (United States) FET fair and equitable treatment GCC Cooperation Council for the Arab States of the Gulf GDP gross domestic product HTP heated tobacco product HMRC Her Majesty’s Revenue and Customs (United Kingdom) IARC International Agency for Research on Cancer IIA international investment agreement IMF International Monetary Fund IRS Internal Revenue Service (United States) ISO International Organization for Standardization IT information technology ITC International Tobacco Control JTI Japan Tobacco International KRA Kenya Revenue Authority LMICs low- and middle-income countries ACRONYMS ix MFN most favoured nation MOP Meeting of the Parties (to the Protocol) MPOWER (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco NCDs noncommunicable diseases NCI National Cancer Institute NT national treatment OECD Organisation for Economic Co-operation and Development OST other smoking tobacco PMI Philip Morris International PPP purchasing power parity QR quick response RGTE WHO Report on the global tobacco epidemic RYO roll-your-own SACU Southern African Customs Union SCARE (S) smuggling and illicit trade; (C) court and legal challenges; (A) anti-poor rhetoric; (R) revenue reduction; and (E) employment impact SDGs Sustainable Development Goals SEARO WHO South-East Asia Regional Office SII Internal Revenue Service of Chile TADAT Tax Administration Diagnostic Assessment Tool TTC transnational tobacco company UAE United Arab Emirates UHC universal health coverage VAT value added tax WAEMU West African Economic and Monetary Union WCO World Customs Organization WHO World Health Organization WHO FCTC WHO Framework Convention on Tobacco Control WHO ISPT WHO interactive smoking projection and target-setting tool WHO TaXSiM WHO tobacco tax simulation model WPRO WHO Regional Office for the Western Pacific WTO World Trade Organization x W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N xi Executive summary This WHO technical manual on tobacco tax policy and administration builds upon the 2010 WHO technical manual on tobacco tax administration by further detailing the strategies for effective tobacco tax policy development, design, implementation and administration. This 2021 edition also serves as an update to the 2010 manual, incorporating the latest developments in science, technology and policy, as well as providing illustrative recent examples from a variety of countries. The best practices laid out in this manual are designed to inform governments on the development of their tobacco taxation policy, facilitating the achievement of their health and revenue objectives while also supporting their overall development strategy. Tobacco taxes have long been seen as a source of revenue for governments, but as evidence of the harms caused by tobacco has accumulated over the years, public perception has evolved. Increasingly, governments, as well as the general public, are recognizing that taxation of tobacco is not only a revenue source but also an effective public health intervention to reduce tobacco consumption and its associated harms. The profile of tobacco taxation as a health policy tool has increased greatly since the publication of the 2010 WHO technical manual on tobacco tax administration. Multiple global commitments have been adopted over the past decade to address tobacco use specifically – as well as noncommunicable diseases (NCDs) and the Sustainable Development Goals (SDGs) more broadly – through tax and price measures to reduce demand for tobacco products, save lives and fund develop- ment. Global development institutions, including the World Bank, the International Monetary Fund (IMF) and major philanthropic foundations, also agree with WHO on the importance of emphasizing and strengthening tobacco taxation as a key health policy tool. The COVID-19 pandemic has further fuelled this shift in the narrative on tobacco taxation by revealing how the global economy is inextricably linked with population and planetary health. Investing in health is fundamental to any economic recovery, and fiscal policy will be a key driver in addressing the socioeconomic consequences of COVID-19. Interventions such as tobacco taxation – which leads to reduced tobacco consumption, improved population health and increased revenues for governments – should be part of a comprehensive strategy for a build back better recovery. The evidence is clear: significant increases in excise taxes that lead to price increases have consistently proven to be the most effective, as well as the most xii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cost-effective, mechanism for reducing tobacco consumption. This manual will guide readers through the necessary steps to create and implement the strongest tobacco taxation policies for their specific countries. There are many factors to consider when developing tobacco taxation policy. Understanding the market is a fundamental step before deciding which form of taxation to use. Specifically, the choice between ad valorem and specific taxation is influenced by the market structure in a given country. At the same time, tax structure also shapes the market structure. Over the past decade, there has been a trend towards countries adopting specific excise taxes or mixed excise systems that rely more heavily on the specific component, which the latest global data associate with the highest average prices. Tobacco tax structures can be simple, with one flat rate across the board, or com- plex, with multiple tiers for products with different characteristics. In 2018, 31 coun- tries used complex, multitiered structures. But evidence demonstrates that simpler excise tax structures – utilized in all high-income countries – leave the least room for industry manipulation or tax avoidance and brand/product switching by consumers. Not only is it important to set taxes at a high level to discourage consumption, specific excise tax policies must include regular adjustments to increase the tax rate so that it keeps up with inflation and income growth in a country over time. Excise tax increases should aim to reduce the affordability of tobacco products. The base on which the tax is applied is also important. For specific taxation, the tax base should be the quantity in clearly defined units. For ad valorem (or mixed) taxation, the best practice is to use the retail price as the tax base and introduce a minimum excise tax. With regard to non-tax regulations that affect the price of tobacco products, pric- ing regulation may be considered to prevent the tobacco industry from exercising differential tax shifting, which it uses to ensure that large price gaps exist between premium and cheap cigarettes. However, pricing policies cannot be used alone. If con- sidered, they should be used only as complements to significant excise tax increases. Other non-tax regulations include banning promotional discounts for tobacco products and banning the sale of single cigarettes. To assuage concerns that tax increases will increase inflation – as well as to reflect the declining trend in consump- tion of tobacco products – it is good practice to exclude tobacco products from the basket of items that are used to develop consumer price indexes. Finally, in order to make excise tax on tobacco products more effective in reducing overall tobacco use, all tobacco products must be taxed in a comparable way. Regular assessment, evaluation and monitoring of the impact of tobacco tax policies over time are essential components of effective tax policy development and analysis. Governments need to have accurate estimates of price, income and tax base elasticities in order to anticipate the impact of a tax increase on consumption and EXECUTIVE SUMMARY xiii tax revenue. Ideally, other factors such as non-price policies should also be taken into account when estimating price and income elasticities for a specific country. A variety of tools and indicators exist to measure impact and monitor progress, and these are described in Chapter 2 of this manual. When developing tobacco tax policy, it is also important to take the broader policy context into consideration at both the domestic and the regional level. Domestically, cooperation is needed across sectors to ensure that policies and interventions in the areas of agriculture, trade, finance and labour do not work against the public health objectives of tobacco control and taxation. For countries that are part of a regional bloc, regional harmonization of tobacco taxation is a useful tool to prevent tax revenue erosion, tax avoidance and tax evasion, as well as to protect population health. Tax harmonization must be designed carefully, however, to be effective. The experience of the European Union (EU) demonstrates that both a declining consumption trend and stable revenues can be achieved with harmonized minimum excise tax rates. Discussions of policy development and implementation for new and emerging nicotine and tobacco products such as heated tobacco products (HTPs) and electronic nicotine and non-nicotine delivery systems (ENDS/ENNDS), are complicated, by their constantly changing technology and market dynamics. Policies and regulations need to be developed carefully and adjusted accordingly. Where HTPs are not banned, the current recommendation is to tax them at the same level as cigarettes on a per-unit basis, regardless of tobacco content. Early evidence from the United States shows that demand for e-cigarettes, a subcategory of ENDS/ENNDS products, is possibly even more price-responsive than the demand for conventional cigarettes, meaning that taxes can be used as an effective deterrent to ENDS/ENNDS products use. While there is preliminary evidence of substitutability between conventional cigarette use and e-cigarette use, further research is needed to understand substitutability effects among users of both conventional cigarettes and ENDS/ENNDS products. It is essential to imple- ment regulation of ENDS/ENNDS products along with any tax policy to safeguard public health. In countries where they are not banned, ENDS/ENNDS products must be regu- lated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS taxation. Nicotine-containing and non-nicotine-containing e-liquids should be taxed equally. Ultimately, while the policy implications of these newer products require careful consideration, the fact remains that conventional tobacco products constitute the overwhelming share of consumption (more than 97% in 2018). Tobacco tax administration must be both efficient and effective to ensure that health objectives are met and the desired level of tax revenue is raised. Since the xiv W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation of tobacco taxation often involves numerous agencies within a country, clearly defined roles and responsibilities are essential to maximize efficiency. Coordination among the different agencies involved, as well as with neighbouring countries, is required for tobacco tax administration to be effective. Performance evaluation and accountability for competent authorities is also necessary, and many tools and indicators exist to facilitate these processes (described in Chapter 3). There are a number of steps authorities should take to ensure efficiency and effectiveness at each stage throughout the tax compliance cycle (see Chapter 3, section 3.3). Control and enforcement are the main functions of tax administration, and these can best be achieved through the use of a strategic plan and a risk-based approach. Controls can be exercised through licensing and due diligence, fiscal markings (e.g. tax stamps), tracking and tracing, implementation of anti-forestalling measures, national audits and specific controls for imports and exports, as well as for free zones and transhipment points. Once smuggling or illicit trade is detected, actions such as seizing and destroying smuggled and/or illicit tobacco and col- lecting due taxes must be taken immediately. To deter further illegal activities, a comprehensive audit must also be carried out, including all those involved in the illicit acts. Penalties and sanctions must be sufficient to deter illegal activities. The Protocol to Eliminate Illicit Trade in Tobacco Products provides invaluable guidance for tobacco tax administration, control and enforcement that is applicable even for countries that are not Parties to it. The broader elements of a good tax system include proper resourcing of competent authorities, strict rules and regulations to detect and punish corruption and a strong judiciary system capable of resolving disputes as soon as possible. In its efforts to oppose tobacco tax increases, the tobacco industry utilizes many SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric, (R) revenue reduction and (E) employment impact – to influence the political economy of tobacco. Chapter 4 provides detailed analyses of these issues with supporting evidence that belies the SCARE tactics, as well as guidance for tax and other relevant authorities on how to anticipate and respond to industry arguments. This manual also provides tools and methodologies to help tax authorities define and evaluate the problem of illicit trade of tobacco products in their countries, inde- pendent of the tobacco industry’s generally inflated estimates. Price (and tax) levels are not a key determinant of illicit trade; rather, the problem is exacerbated by the lack of governance and tax administration capacity. Refraining from increasing taxes is not the solution. Countries should instead respond with a comprehensive strategy to fight illicit trade, including undertaking independent estimates of illicit trade levels and implementing good tax administration practices such as those discussed in Chapter 3 and contained in the WHO Framework Convention on Tobacco Con- trol (WHO FCTC) Protocol to Eliminate the Illicit Trade in Tobacco Products. When it comes to court challenges, the tobacco industry is less likely to challenge excise taxes than other tobacco control measures, because taxation is a comparatively well-established regulatory measure. The industry will, however, exploit the slightest vulnerability in the design, adoption or implementation of tax measures. For this reason, measures to strengthen regulators’ legal position are described that will enable authorities to protect themselves from potential legal challenges. The industry argument of regressivity, or the notion that tobacco tax increases hurt the poor because they have to pay a larger share of their income in taxes than the rich, has two fundamental limitations. First, the notion of regressivity does not take into consideration the broader health and economic harms caused by tobacco use that exacerbate the impoverishment of lower-income smokers. These harms are actually reduced when tobacco consumption decreases following a tax increase. Second, the tobacco industry argument ignores the fact that higher tobacco taxes and prices can induce behaviour change as is reflected in the price elasticity of demand. Evidence consistently shows that lower-income smokers are more sensitive to price and therefore more likely to reduce smoking in response to a tax and price increase. Including these factors shows tobacco taxation to be, in fact, a progressive public health intervention that disproportionately benefits the poor. While essentially admitting that a tobacco tax increase may have the desired effect of reducing consumption, the industry also tries to argue that a tax increase will also reduce revenues. In fact, the price inelastic demand for tobacco makes tobacco tax increases a win-win for both public health and finance. This manual presents several country examples that demonstrate how well-designed and well-implemented tobacco tax increases lead to increases rather than decreases in revenue in the short to medium term. In addition, the reduced consumption resulting from a tax increase results in reductions of other tobacco-related government expenditures as well. The final tactic used by the tobacco industry to challenge proposed tax increases is to frame tobacco taxes as an economic rather than a public health issue. This false choice between health and jobs is based on faulty assumptions that 1) tobacco is a significant source of domestic employment; 2) job creation relies on tobacco consumption and 3) tobacco-related livelihoods are prosperous, sustainable and irreplaceable. Earmarking can be a useful tool for improving the political economy of tobacco tax increases. While the primary goal of tobacco tax increases is to reduce demand for tobacco, setting aside portions of tax revenue to fund other tobacco control efforts or relevant health programmes can help convince the public, politicians and officials of the value of significant tobacco tax increases. Earmarking can also be EXECUTIVE SUMMARY xv x vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N used to counter tobacco industry arguments about potential negative effects of tax increases – for example, by supporting tobacco farmers in transitioning to other crops. Tobacco taxes work. This is why the industry invests so much money and effort in blocking large tax increases and other effective tax policy reforms. Policy-makers must not be swayed by industry pressure but need only to follow the facts. This manual provides all the information policy-makers need to make the right deci- sions at each step of the process – from designing, evaluating, implementing and administering tax policy to refuting specious industry attacks and communicating the value of tobacco taxation to legislators and the broader population. An effectively designed and efficiently administered tobacco tax policy will not only produce the direct results of reducing tobacco consumption among smokers and raising revenue for governments, its effects will be felt much more broadly. Indeed, raising tobacco taxes is a SMART policy: it Saves lives; Mobilizes resources; Addresses health inequities; Reduces burdens on health systems; and Targets tobacco use, a major risk factor for NCDs. • Saves lives: Tobacco use is the leading cause of preventable deaths globally – it claims 8 million lives each year. Tobacco taxation is the most effective mechanism for reducing tobacco consumption and its associated health burden worldwide. • Mobilizes resources: Despite being the single most effective tobacco control measure, tobacco taxation is largely underutilized as a policy mechanism. Based on available data on the price and taxation of cigarettes, it is estimated that excise taxes on cigarettes generated a worldwide total of US$ 361 billion in revenues in 2018, including US$ 162 billion in revenues for low- and middle- income countries (LMICs). If all countries were to raise cigarette excise tax rates by the equivalent of US$ 1 per pack, the amount of excise revenue from cigarettes would increase by US$ 178–219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from such tax increases, with excise revenues increasing by 82–103%, providing governments in these countries with an extra US$ 133–167 billion. This shows the substantial revenue potential of tobacco taxes. • Addresses health inequities: Tobacco taxation and tax increases are effectively progressive or pro-poor policies because of their positive distributional impact. Lower-income smokers benefit disproportionately from reduced tobacco consumption and use in terms of health gains and income retention. • Reduces burdens on health systems: The worldwide economic cost of tobacco use was US$ 1.4 trillion in 2012. Tobacco taxes reduce tobacco-related burdens on governments and health systems through population-based preventive measures. • Targets tobacco use: Tobacco taxation directly targets and reduces tobacco use, which is a major risk factor for several deadly NCDs. In summary, significant tobacco tax increases, designed and implemented according to the latest guidance and best practices presented in this technical manual – and as a strong component of a comprehensive tobacco control strategy – will bring about substantial reductions in tobacco use and the health and economic harms it causes. EXECUTIVE SUMMARY xvii x viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 1 CHAPTER 1. Why this manual? BACKGROUND Tobacco taxes are not new. Governments around the world have been applying taxes on tobacco and tobacco products practically since the idea of excise was conceived. And rightly so: tobacco is not a necessity, it is easy to tax and the demand for it is relatively inelastic. These characteristics, along with the substantial revenues tobacco taxes generate, have made tobacco a highly appropriate object of taxation. As evidence of the harms of tobacco has accumulated over the years, the public perception of tobacco taxes has evolved. Now tobacco taxes are not only seen as a revenue source, but, more importantly, they are recognized as an effective public health intervention to reduce tobacco consumption. This trend reflects the reasons excise taxes exist in the first place – to discourage harmful behaviour and to mitigate the associated negative externalities (1–2). Many governments view tobacco taxes as a significant and stable source of rev- enue, which may explain why there is often a degree of hesitation whenever tobacco tax reform is proposed. Historically, many governments have relied on revenues from tobacco taxes and have even adjusted the level of taxation according to their revenue needs (3). However, some countries are beginning to recognize the value of applying high tobacco taxes primarily as a public health tool, viewing revenues as a secondary consideration (4). Arguments against tobacco tax hikes or improvements to the tax structure are often economic in nature: such tax changes will allegedly decrease revenues, wipe out jobs, increase illicit trade and harm local industries, among other claims. But the evidence has consistently shown that such claims are simply not true in an overwhelming majority of situations. The tobacco industry, in particular, frequently portrays this conflict as a false dichotomy between public health and the economy – as if prioritizing health comes at the expense of the economy. In fact, studies and real-world experiences have shown that increasing tobacco taxes not only improves public health but also has a net positive impact on the economy and development of a country – a true win-win scenario (5–6). As an update to the first WHO technical manual on tobacco tax administration published in 2010, this manual aims to help readers better navigate the various 2 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N issues surrounding tobacco taxes and their implementation. The primary intended audience includes policy-makers, finance officials, tax authorities, customs officials and other relevant persons/bodies involved in the formulation and implementation of tobacco tax policy. The manual provides a detailed guide to the design of tobacco tax policy and describes how to effectively administer these taxes to maximize impact. Detailed discussions of the political economy considerations and the hurdles that need to be overcome before and during implementation are included as well. The overarching goal is to equip those working in the tax policy and implementation spheres with sufficient information to help realize the health and revenue objectives of a government’s tobacco tax policy in line with its overall development strategy. THE CONSEQUENCES OF TOBACCO USE AND THE NEED FOR INTERVENTION Most people are aware that smoking and tobacco use are harmful to health, but few truly comprehend the scale of this harm. The tobacco epidemic claimed more than 100 million lives in the last century (7), with updated estimates now reaching 8 million deaths annually from tobacco use and exposure to second-hand smoke (8). As much as 80% of these deaths occur in low- and middle-income countries (LMICs) (6), revealing how the developing world carries much of the global burden. Tobacco use is a major risk factor for many chronic conditions, including heart disease, cancer, diabetes and chronic lung disease – collectively known as noncom- municable diseases (NCDs). NCDs account for about 15 million premature deaths (between ages 30 and 69) worldwide, killing people in their most productive years. As the leading cause of preventable deaths, tobacco use remains one of the foremost public health challenges of our time. The consequences of tobacco use also present enormous economic, development and social costs that wreak havoc on families, communities and societies. The annual economic cost of smoking was estimated at US$ 1.4 trillion in 2012, equivalent to 1.8% of the global gross domestic product (GDP) (9). With these figures likely to have increased since then, the massive health and economic burdens of tobacco use provide justification for governments to intervene and strictly regulate the market for tobacco products. The purview of tobacco control extends beyond the strong imperative to protect people’s health and well-being; it should also strive to contain the market failures and negative externalities of tobacco use, particularly since these effects can significantly impact a country’s development trajectory. The mounting evidence of the enduring destruction caused by tobacco in the 20th century provided compelling reasons for a strong global response, which led countries to negotiate the World Health Organization Framework Convention on Tobacco Control (WHO FCTC). The WHO FCTC came into force in 2005 as the CHAP T ER 1. WHY T HIS M ANUAL? 3 first public health treaty under the auspices of WHO. To facilitate its implementation at the country level, WHO packaged a set of demand-reduction measures directly taken from the treaty (7). These interventions, collectively known as MPOWER, are as follows: (M) monitoring tobacco use and prevention policies; (P) protecting people from tobacco smoke (smoke-free laws); (O) offering help to quit tobacco use (cessation services); (W) warning about the dangers of tobacco (including graphic pack warnings and plain packaging); (E) enforcing bans on tobacco advertising, promotion and sponsorship; and (R) raising taxes on tobacco products. Specifically, under Article 6 of the WHO FCTC, Parties recognized that price and tax measures are an effective and important means of reducing tobacco consumption for various segments of the population – in particular, among young persons (10). The severity of the tobacco epidemic and its ongoing damage to health and economies are clear justifications for governments to actively intervene and correct market failures. The scale of the burden and the rate at which lives are being destroyed necessitates urgent and aggressive action on tobacco control, using measures that most countries have committed to implementing and that are proven to be effective in reducing tobacco use. WHY TOBACCO (EXCISE) TAXES ARE CRUCIAL Among the different tobacco control interventions, raising excise taxes has been identified as the most effective as well as the most cost-effective measure to reduce consumption (6). While other interventions are certainly important components of a comprehensive tobacco control strategy, the direct impact of significant tax increases on consumption is by far the strongest. On average, a tax increase that causes prices to go up by 10% reduces consumption by 4% in high-income countries and 5% in LMICs (6). When implemented at scale, this demonstrates the enormous power of tobacco taxation and its potential to save lives. Tobacco taxes differ from other interventions in that their impact can increase and build over time – even if taxes are already relatively high, their rates need to be continuously increased to retain and amplify their effectiveness. However, this should not be taken as a suggestion that governments considering tobacco control interventions should focus solely on taxes. Taxes are even more effective when implemented as part of a comprehensive package of measures such as MPOWER, which covers distinct but complementary intervention points. Among the different taxes applied on tobacco products, excise taxes are the most significant because they raise both absolute and relative prices (6). This is important when considering health objectives, since it is the magnitude of the price increase of tobacco products that determines the reduction in consumption. An excise tax is typically applied on a limited set of products, designed to discourage their use by 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N raising the price significantly over that of other products available in the market. This is in contrast to value added taxes (VAT) or sales taxes, which apply to most goods and services. Seeking to raise the prices of tobacco products through VAT or sales taxes would fail to increase relative prices, making this an ineffective and inefficient method. Customs or import duties on tobacco products are also utilized, but their impact is waning with the global trend towards bilateral and regional agreements aimed at trade facilitation. The application of these duties varies across countries, but overall, they are not applicable to locally produced tobacco products. As a tool to increase prices, import duties cannot substitute for excise taxes, since they are not specifically designed to reduce consumption. When viewed as a public health policy tool, tobacco taxation is highly cost-effective, since it delivers significant impact yet is relatively inexpensive to implement (11). The costs of implementing tobacco taxation are much lower than those of clinical NCD interventions such as cancer treatments or maintenance medications, since the commodity and human capital requirements are less substantial (12). Moreover, increasing tobacco taxes actually generates additional revenue for a government. Tobacco taxes are also very effective in pre-empting or reducing consumption among groups of people who are especially price-sensitive – youth in particular, who are prevented from initiating a lifelong addiction if taxes and prices are sufficiently high (13). This is also true for the poor, who are more prone to catastrophic health expenditures than the wealthy are. Preventing initiation or encouraging cessation by imposing high taxes provides an escape route from the vicious cycle of tobacco use and poverty (6). A DECADE OF PROGRESS AND COMMITMENT TO ACTION Since 2010, when the first WHO technical manual on tobacco tax administration was published, numerous developments have raised the profile of tobacco taxes as an essential public health intervention. The Conference of the Parties (COP) to the WHO FCTC adopted guidelines for implementation of Article 6 of the treaty, which focuses on price and tax measures to reduce the demand for tobacco. Also within this period, three high-level meetings on the prevention and control of NCDs by the United Nations General Assembly, as well as the endorsement of the Global NCD Action Plan in 2013 by the World Health Assembly, have resulted in strong global commitments to implement measures, such as increased tobacco taxes to protect people’s health. The 2030 Agenda for Sustainable Development, which contains 17 goals known as the Sustainable Development Goals (SDGs), describes the global development strategy for the next decade. Within the SDGs, two specified targets are highly relevant for tobacco control: strengthening the implementation of the WHO FCTC (target 3.a) CHAP T ER 1. WHY T HIS M ANUAL? 5 and reducing premature mortality from NCDs by 30% (target 3.4). Furthermore, the Addis Ababa Action Agenda1, which aims to provide a global framework for financing the SDGs, also highlights tax and price measures on tobacco as key mechanisms to reduce demand and save lives while increasing domestic resources for develop- ment. Another important milestone was the 2018 entry into force of the Protocol to Eliminate Illicit Trade in Tobacco Products. These key events, along with several outcome documents and policy declarations in the area of tobacco control and the wider development sphere, have introduced tobacco taxation into the consciousness of a much larger share of policy-makers. As detailed in subsequent chapters, numerous countries have imposed sufficiently high tobacco tax rates while applying best practices in tax policy design and imple- mentation over the past decade (8, 10). For example, sustained and substantial tax increases have reduced tobacco use in LMICs such as Brazil (14), Turkey (15) and the Philippines (16). High-income countries also continued their leadership in this area, comprising 23 of the 38 countries judged to have sufficiently high tobacco taxes in 2018 (8). However, much remains to be done. The 2019 WHO report on the global tobacco epidemic (RGTE) shows that tobacco taxes are still the most underutilized tobacco control policy among the MPOWER measures (8), with only 14% of the world’s population being covered by sufficiently high tobacco taxes. Substantial progress has also been made in building the tools and evidence base for tobacco taxation. Volume 14 of the International Agency for Research on Cancer (IARC) handbooks of cancer prevention, Effectiveness of tax and price policies for tobacco control, published in 2011, is a key review of the literature published as of May 2010 on the effectiveness of tax and price policies in reducing tobacco use. The National Cancer Institute (NCI)-WHO Monograph on the economics of tobacco and tobacco control, published in 2016, details the evidence accumulated over the years from various countries, focusing not only on tax and price policies, but on all aspects of the economics of tobacco and tobacco control. In addition, numerous published studies from LMICs provide a comprehensive picture of the impact of tobacco taxation in different contexts. The updated Appendix 3 of the Global NCD Action Plan explains the cost-effectiveness of tobacco taxation (11), while the Global NCD Business Plan, Saving lives, spending less, built on this work by estimating a dollar figure for the return on investment expected from implementing the best-buy interventions for tobacco control, including taxation (12). The past decade has seen major steps forward for tobacco taxation in terms of global commitments, the number of countries implementing best practices and the 1 The Addis Ababa Action Agenda of the Third Conference on Financing for Development. Third Inter- national Conference, 13-16 July 2015, Addis Ababa, Ethiopia (https://sustainabledevelopment.un.org/ content/documents/2051AAAA_Outcome.pdf, accessed 17 February 2021). 6 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N expansion of the evidence base on tobacco taxes, particularly in LMICs. Within the development sphere, institutions such as the World Bank, the International Monetary Fund (IMF) and many other multilateral agencies are aligned with WHO on the importance of tobacco taxation and the need to improve its implementation (17–18). Considerable challenges remain; although it appears that the world is headed in the right direction, progress needs to accelerate at a much quicker pace in order to achieve the SDG targets by 2030. SHAPING A “NEW NORMAL” FOR TOBACCO TAXATION The global upheaval caused by the COVID-19 pandemic has cast an unprecedented spotlight on how well governments around the world prepared for and responded to the crisis. It has exposed glaring health systems vulnerabilities and highlighted the struggles of many countries to control the spread of the virus. But perhaps more than anything, the pandemic has demonstrated how the economy, trade, science, politics and many other aspects of our societies are very much interdependent and interconnected with the health of the population. It is clear that an individual’s state of health can significantly determine their susceptibility to disease and their ability to overcome it. People with NCDs are more vulnerable to becoming severely ill with a number of conditions, which also appears to be the case with COVID-19 (19). Tobacco use is a major risk factor for NCDs, and available research suggests that smokers are at higher risk of developing severe illness and dying from COVID-19 (20). Just as the different aspects of society are interconnected, so too are people’s health, the existence of health-promoting environments and the government policies and agencies that shape these environ- ments. This critical moment presents a unique opportunity and renewed motivation to discourage the use of harmful products such as tobacco and to further improve tobacco control measures, especially tobacco tax policy. Moving forward, a business-as-usual approach to tobacco taxation will not be sufficient. Responding to this new reality and preparing for the next pandemic entails implementing measures that promote healthier populations. Like the COVID-19 pandemic, any future pandemic will likely exacerbate health inequities, bring about more economic uncertainty and put pressure on governments’ fiscal capacities. Interventions such as higher tobacco taxes, which protect people’s health while generating more revenues and economic benefits, become even more important in such crisis situations. Given this context and the stakes involved, ministries of finance and tax authorities are in a unique and powerful position – one of saving not only livelihoods, but also lives. The importance of increasing tobacco taxes – one of the most effective public health tools available – cannot be overstated. The traditional approach of treating CHAP T ER 1. WHY T HIS M ANUAL? 7 tobacco tax exclusively as a revenue source has no place in the new normal. One cannot deny the scale of the tobacco epidemic, the necessity to correct market failures and the overwhelming evidence of tobacco taxation’s benefits to health and to the economy. The positive trend in the changing narrative around tobacco taxation needs to continue. Tobacco taxation should not be viewed in isolation from the rest of government policies, but rather as an important part of the whole, an essential piece in working towards our common goal of better health for all. OVERVIEW OF SUBSTANTIVE CHAPTERS This manual is primarily designed for policy-makers, finance officials, tax authorities and customs officials. It may also be useful for officials within health ministries or other government agencies, as well as nongovernmental organizations working in this area, including tobacco control advocates. Significant effort is made to present real-world examples and recent experiences from a wide range of countries to demonstrate success stories and lessons learned in raising tobacco taxes. A sub- stantial amount of evidence has been generated in LMICs over the past few years that supports and augments the existing evidence base, providing a much broader body of knowledge than was available when the first WHO technical manual on tobacco tax administration was released. Chapter 2 delves into the theory, practice and empirical evidence on tobacco excise tax policy, including current global trends. The chapter offers a detailed analysis of the various elements that constitute tax structure, aiming to provide policy-makers with a comprehensive understanding of the factors affecting prices, consumption and the market. It describes the key components to keep in mind when designing tobacco tax policy to maximize the impact of tax increases and improve the tax structure. The chapter also includes updated global price and tax data, specific examples from various countries and a discussion of tax base elasticity, automatic excise tax adjustments and pricing regulations, as well as descriptions of new and emerging nicotine and tobacco products, including electronic nicotine- and non- nicotine delivery systems (ENDS/ENNDS) and heated tobacco products (HTPs). Chapter 3 focuses on tobacco tax administration. It provides an in-depth discus- sion of the fundamental components that make tobacco tax collection effective and efficient, ensuring achievement of the health and revenue objectives of tax policy. It highlights the importance of cooperation among the various agencies involved in the implementation of tobacco taxes within countries and across borders. Building on country and regional experiences from previous decades, the chapter outlines specific measures and recommendations to maintain oversight of the whole tax compliance cycle. Also included are actions to facilitate control and enforcement, such as licensing, fiscal markings (e.g. tax stamps), tracking and tracing systems 8 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N and import and export controls. Finally, the discussion pivots towards the broader elements of a good tax system, such as proper resourcing of competent authorities, a strong judiciary and strict rules regarding corruption. Chapter 4 deals with the important challenges in the area of political economy when countries attempt to increase tobacco taxes or simplify the tax structure. The tobacco industry often relies on identified patterns of argumentation and tactics to obstruct such reforms, i.e. SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric (regressivity), (R) revenue reduc- tion and (E) employment impact – each of which is discussed thoroughly in this chapter. Also included is a detailed discussion of the measurement of illicit trade and a discussion of earmarking tobacco tax revenue for health purposes. Finally, Chapter 5 presents a comprehensive list of the best practices in tobacco tax policy and administration discussed throughout this manual. The list is intended to serve as a practical guide and quick reference to the salient points presented. CHAP T ER 1. WHY T HIS M ANUAL? 9 REFERENCES 1. Ranson K, Jha P, Chaloupka FJ, Nguyen SN. The effectiveness and cost-effectiveness of price and other tobacco control policies. In: Jha P, Chaloupka FJ, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000:427–447 (https://www.paho.org/hq/dmdocuments/2010/ Cost-effectiveness%20of%20price%20increases.pdf, accessed 4 February 2021). 2. Effectiveness of tax and price policies for tobacco control. Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications.iarc.fr/Book-And-Report-Series/Iarc- Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And-Price-Policies-For-Tobacco-Control-2011, accessed 4 February 2021. 3. Tobacco taxation in the United States. In: Lynch BS, Bonnie RJ, editors. Institute of Medicine (US) Committee on Preventing Nicotine Addiction in Children and Youths. Growing up tobacco free: preventing nicotine addiction in children and youths. Washington (DC): National Academies Press (US); 1994 (https://www.ncbi.nlm.nih.gov/books/NBK236771/, accessed 10 November 2020). 4. Sin tax reform. Manila: Department of Finance (Philippines); 2012 (https://www.dof.gov.ph/advocacies/ sin-tax-reform/, accessed 10 November 2020). 5. Goodchild M, Perucic AM, Nargis N. Modelling the impact of raising tobacco taxes on public health and finance. Bull World Health Organ. 2016; 94:250–7 (https://www.who.int/bulletin/ volumes/94/4/15-164707.pdf, accessed 4 February 2021). 6. The economics of tobacco and tobacco control. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute, NIH Publication No. 16-CA-8029A; 2016 (National Cancer Institute tobacco control monograph 21; https://cancercontrol.cancer.gov/ brp/tcrb/monographs/monograph-21, accessed 17 December 2020). 7. WHO report on the global tobacco epidemic, 2008: the MPOWER package. Geneva: World Health Organization; 2008 (https://www.who.int/tobacco/mpower/mpower_report_full_2008.pdf, accessed 10 November 2020). 8. WHO report on the global tobacco epidemic, 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/tobacco/global_report/en/, accessed 10 November 2020). 9. Goodchild M, Nargis N, Tursan d’Espaignet E. Global economic cost of smoking-attributable diseases. Tob Control. 2018;27:58–64 (https://tobaccocontrol.bmj.com/content/27/1/58, accessed 4 February 2021). 10. Guidelines for implementation of Article 6 of the WHO FCTC. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 4 February 2021). 11. Tackling NCDs: ‘best buys’ and other recommended interventions for the prevention and control of noncommunicable diseases. Geneva: World Health Organization; 2017 (https://apps.who.int/iris/ handle/10665/259232, accessed 10 November 2020). 12. Saving lives, spending less: a strategic response to noncommunicable diseases. Geneva: World Health Organization; 2018 (https://www.who.int/publications/i/item/WHO-NMH-NVI-18.8, accessed 4 February 2021). 13. Chaloupka FJ, Warner KE. The economics of smoking. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics. Elsevier; 2000;1(1):1539–1627. 14. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Rev Panam Salud Publica. 2016;40(4):243–9 (https://iris.paho.org/bitstream/ handle/10665.2/31306/v40n4a09_243-9.pdf?sequence=1&isAllowed=y, accessed 17 February 2021). 15. Cetinkaya V, Marquez PV. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 16. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington (DC): World Bank Group; 2016 (http://documents.worldbank.org/curated/en/638391468480878595/pdf/106777-PUB-PUBLIC- PUBDATE-7-26-2016.pdf, accessed 10 November 2020). 17. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 10 November 2020). 10 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 18. Irwin A, Marquez P, Jha P, Peto R, Moreno-Dodson B, Goodchild M, et al. Tobacco tax reform: at the crossroads of health and development – a multisectoral perspective. Washington (DC): World Bank Group; (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report- CrossRoads.pdf, accessed 2 February 2021). 19. Information note: COVID-19 and NCDs. Geneva: World Health Organization; 23 March 2020 (https:// www.who.int/publications/m/item/covid-19-and-ncds, accessed 10 November 2020). 20. WHO statement: tobacco use and COVID-19. Geneva: World Health Organization; 11 May 2020 (https:// www.who.int/news-room/detail/11-05-2020-who-statement-tobacco-use-and-covid-19, accessed 10 November 2020). 11 CHAPTER 2. Tobacco excise tax policy 2.1. GLOBAL OVERVIEW OF TOBACCO TAX PRACTICES A well-designed tax policy is key to having an effective tax policy. Any government that is planning to reform its tax policy must first understand the fundamental components of a good tax policy, as well as consider the strengths and weaknesses of different approaches to taxation, how they impact price and the requirements for tax administration. Understanding how the tobacco market operates in a country is equally important for policy-makers because of the inevitable interaction between the market and tax structures. Beyond the political considerations that strongly influence tobacco tax policy development, this chapter focuses on the technical aspects of tobacco taxation – excise tax in particular. Section 2.1 provides an overview of tobacco tax practices at the global level, focusing on the different ways countries structure excise tax. Section 2.2 emphasizes the importance of carefully designing excise tax policy, highlighting not only the significance of tax increases but also excise tax structure and its impact on prices, taking into account how market structure influences trends. This section also discusses the importance of measuring impact as another aspect of tax policy development, and it presents the crucial elements for performing measurement, as well as the relevant indicators available to monitor progress. Section 2.3 describes external policy considerations in the design phase to ensure that the goals of tobacco control and taxation are achieved. Intersectoral policy integration and coherence at the domestic level is discussed as a strategy to ensure that policies of other sectors do not inhibit or obstruct public health policy objectives. This section also reviews the current state of regional tax harmonization based on the experience of existing regional blocs and draws conclusions on the best policy approaches to preserve the public health interests of individual countries. Section 2.4 discusses new and emerging nicotine and tobacco products, in par- ticular HTPs and ENDS/ENNDS. It reviews the latest evidence on the health impacts of these products and current approaches to regulation. Key policy considerations are identified, and recommendations are provided for adopting an appropriate excise tax policy for these products. Section 2.5 summarizes the issues covered in the chapter and the key takeaways. 12 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.1.1 TAXES APPLIED ON TOBACCO PRODUCTS Taxes are classified as either direct or indirect. Direct taxes are imposed on the profit, income, property or wealth of persons or companies, whereas indirect taxes are imposed on the price of goods and services. Indirect taxes are most relevant to tobacco products taxation, because they directly influence price. A variety of types of indirect taxes can be applied to tobacco products. These include: • excise taxes – taxes that apply to a few selected commodities (they can also be applied to alcohol, fuel, sugar-sweetened beverages, etc.). • VAT or sales taxes – VAT is a multistage tax on all consumer goods and services that is applied proportionally to the price the consumer pays for a product. It is a tax on the amount by which the value of an article has been increased at each stage of its production or distribution. Some countries impose sales taxes instead of VAT. Unlike VAT, which is collected at every stage of the supply chain, sales taxes are generally levied at the point of retail on the total value of goods and services purchased. Ultimately, the consumer ends up paying the tax, whether it is a VAT or a sales tax. • import duties – taxes on selected goods imported into a country to be consumed in that country (i.e. goods that are not in transit to another country). In general, import duties are collected from the importer at the point of entry into the country. • other taxes – other indirect taxes, such as environmental taxes, that do not fall into any of the categories listed above. One of the most well-established and widely understood points in tax policy is that tobacco products should be subject to excise taxation. The focus of this chapter – and of this manual overall – is on excise taxes. They are the most important type of indirect taxes for tobacco control because they are applied directly to tobacco products and contribute the most to increasing the price of tobacco products relative to other goods and, subsequently, to reducing consumption. There are two basic types of excise taxes: • specific – levied as a monetary value per quantity of the product being taxed (e.g. 1 000 cigarettes, pack of 20 sticks, kilogram of tobacco); and • ad valorem – levied as a percentage of the value (e.g. retail price, or the producer/ex-factory price or the cost, insurance and freight [CIF] value1) of the product being taxed. These types of excise tax can be applied at a uniform or a differential (tiered) rate and on their own or in combination (i.e. a mixed system). 1 CIF is the value of an imported product as declared to customs upon entry into a territory. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 13 Given the widespread use of cigarettes – almost the only tobacco product used in some parts of the world – and the scarce availability of data for other tobacco products, this chapter focuses mainly on cigarettes. But there are a few examples and recommendations for other tobacco products, including those that are more prevalent in specific parts of the world (e.g. bidis or smokeless tobacco in South-East Asia and waterpipe tobacco in the Eastern Mediterranean region). 2.1.2 CIGARETTE TAXES AND RECENT TRENDS WORLDWIDE Tax and national income levels: the higher the income level, the higher the taxes and prices At the global level, cigarette price and tax levels correlate positively with a country’s income level: prices and taxes are higher in higher-income countries and lower as income level decreases. This trend has not changed over the years since 2008. Figure 2.1 presents the levels of price and tax by income groups for 2018, using the World Bank classification of income groups. Fig. 2.1 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by income group, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in purchasing power parity (PPP) adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Excise tax per pack Other taxes Retail price minus taxes Pr ic es a nd ta xa tio n pe r p ac k (P PP $ ) High-income Low-income 4.25 0.68 Middle-income 2.06 Total taxes: 5.30 (67.9% of pack price) Total taxes: 2.91 (58.3% of pack price) Total taxes: 1.18 (38.1% of pack price) PPP $ 7.80 PPP $ 4.99 PPP $ 3.09 Global 2.48 PPP $ 5.53 Total taxes: 3.36 (60.8% of pack price) 0 1 2 3 4 5 6 7 8 14 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tax and price levels: the higher the tax share, the higher the price Globally, cigarette prices correlate positively with tax percentage levels: as the total tax share (of which excise represents the largest part) as a percentage of retail price increases, the price of cigarettes generally also increases (see Fig. 2.2 below).2 This indicates that taxes do influence prices. Fig. 2.2 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by total tax levels, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Tax level and WHO regional classification: taxes and prices are highest in the European region, followed by South-East Asia, the Americas and the Western Pacific, with the lowest levels in the Eastern Mediterranean and African regions At the regional level (WHO regional classification), average levels of prices and taxes vary greatly. The highest level can be seen in the European region, which includes the European Union (EU) countries. The EU’s unified tax structure includes high levels of minimum taxes – which lead to high prices – and encourage member 2 This is a general trend and does not apply for every country; there are countries that have a large tax share but low prices for cigarettes. Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax ≥ 75% Total tax ≤ 25% 0 7 4 1 2 3 6 5 50% ≤ Total tax < 75% 25% ≤ Total tax < 50% Excise tax Other taxes Retail price minus taxes 4.44 2.24 1.57 PPP $ 5.07 PPP $ 5.33 PPP $ 7.07 PPP $ 2.60 0.17 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 15 countries to regularly increase their taxes to meet their obligations. However, with the current minimum level now being reached by all EU member countries, the motivation to increase excise taxes may wane. Indeed, the minimum cigarette excise amount in the EU has not been adjusted since it went into effect on 1 January 2014, and it is suffering from inflation erosion. Member States of the EU acknowledged this in June 2020 by stating that action at the EU level is required to ensure that minimum excise duty rates regain traction to effectively reduce the consumption of tobacco products and that the minimum rates of excise duties on a number of tobacco products would be increased (2). Excise taxes are lowest in the African and Eastern Mediterranean regions. And China – reported separately due to its size – has lower tax rates than the Western Pacific region (see Fig. 2.3). Fig. 2.3 Weighted average retail prices and taxation (excise and total) of most-sold brand of cigarettes, by region, 2018 Notes: China is represented separately from the Western Pacific Regional Office (WPRO) average because of its exceptionally large number of smokers compared with the number in other countries in the region. AFRO is the African Region, AMRO is the Region of the Americas, EMRO is the Eastern Mediterranean Region, EURO is the European Region, SEARO is the South-East Asia Region. Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Retail price Excise tax Total tax share % AFRO AMRO EMRO EURO SEARO WPRO w/o China China All 3. 80 4. 02 4. 24 4. 89 4. 02 1. 45 2. 89 2. 65 2. 48 7. 53 7. 27 5. 61 2. 51 1. 95 1. 05 42.2% 55.8% 61.2% 72.9% 63.2% 55.7% 60.8% 56.9% Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax share % 5. 53 0 1 2 3 4 5 6 7 8 9 10 40 30 50 60 70 80 20 16 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Global tax structures trend: more countries are adopting specific excise taxes or mixed excise systems that rely more on the specific component Since 2008, the number of countries that rely solely on ad valorem taxes or apply no excise taxes at all has decreased as more countries have adopted specific or mixed systems. More of the countries that have implemented a mixed system have increased the specific component of the tax structure relative to the ad valorem component (see Figs. 2.4 and 2.5).3 Fig. 2.4 Changes in excise tax structure, 2008–2018 Fig. 2.5 Changes in reliance on specific versus ad valorem component in mixed systems, 2008–2018 Source: (1). 3 For information about countries that applied each type of excise tax structure in 2018, see Annex 2.1. N um be r o f c ou nt ri es Specic excise Ad valorem Mixed excise No excise 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 54 56 57 63 50 45 24 23 24 21 19 15 49 47 44 42 41 55 57 59 56 60 63 62 Mixed excise Relying more on specic Relying more on ad valorem N um be r o f c ou nt ri es 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 45 22 23 23 24 22 26 27 27 27 32 35 37 50 54 56 57 63 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 17 The imposition of a minimum specific excise tax: half of the countries that have a mixed or ad valorem structure impose an excise tax floor A minimum specific excise tax ensures that at least a certain minimum amount of tax is paid, irrespective of price level. Almost half of the 101 countries that impose either ad valorem or a mixed excise for which data on minimum excise are available (47 countries) set a minimum specific excise tax.4 Nearly two thirds of those that set a minimum specific excise tax (29 countries) are high-income countries; most of them are in the EU, which requires its members to impose a minimum specific excise tax. The choice of tax base worldwide: almost half of the countries that apply a mixed or ad valorem excise system use retail price as the base Setting the base for applying a specific excise is relatively easy: most countries use a defined quantity of sticks for cigarettes, the weight in kilograms for tobacco and the weight in grams for other tobacco products (1). Different bases for ad valorem excises are applied in different countries. Nearly half of the 105 countries that implement either ad valorem or a mixed excise for which data are available (47 countries) use the retail price5 as the tax base for the ad valorem part, and most of those (28 countries) are high-income countries. Using the retail price as the base for the excise ad valorem tax is more effective than using the producer price or the CIF value. Unlike retail prices, which are easy for tax administrators to ascertain by monitoring the market, the producer price or CIF value is prone to undervaluation by producers or importers, who may pass on their margins to related parties further down the supply chain and successfully reduce their tax burden. This tactic is also known as transfer pricing. Additionally, global-level data show that the excise ad valorem on the retail price seems to lead to higher retail prices on average compared with an excise ad valorem applied on other bases, such as the producer price or CIF value (see Fig. 2.8 below). On complex tiered structures: 31 countries still apply complex, multitiered excise taxes on tobacco products As of 2018, 31 countries imposed excise taxes that varied according to defined char- acteristics of cigarettes, including price level, type of production, type of package and length of cigarette (Table 2.1). Some countries use more than one criterion to differen- tiate the tax rates. Indonesia, for example, imposes differential rates based on volume 4 This means that countries with a mixed system impose an overall minimum specific excise tax (where the yield of the specific plus the ad valorem excise cannot be below the set minimum specific excise tax), in addition to the excise on a specific component. 5 Countries that impose ad valorem on retail price exclusive of VAT are also included, since retail prices are easy to determine and VAT rates are known variables. 18 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N produced, type of cigarette and price level. In other countries, such as Member States of the EU, differential rates for cigarettes are prohibited by law, and the rate of the ad va- lorem tax and the amount of specific excise duty must be the same for all cigarettes (3). Table 2.1 Criteria used by countries for tiered excise taxes, 2018 BASE OF TIERS COUNTRY Retail price Bangladesh, Belarus, Indonesia, Jordan, Mozambique, Myanmar, Pakistan, Thailand Cigarette grade (e.g. premium, mid-grade, economy) Egypt, Japana, Mali Producer price China, Lao People’s Democratic Republic Production volume Indonesia Type filter/non-filter Belarus, Georgiaa, India, Kenya, Republic of Moldova, Nepal, Papua New Guinea hand/machine made India, Indonesia kretek/white cigarette Indonesia tobacco content (dark/ blonde or dark/light) Algeria, Bolivia (Plurinational State of ) Packaging soft/hard Mozambique, Uganda Cigarette length India, Nepal, Sri Lanka Trade (domestic/imported) Iran (Islamic Republic of ), Lebanon, Myanmar, Solomon Islands, Tonga, Uzbekistan Leaf content (domestic/imported) Fiji, United Republic of Tanzania a Japan and Georgia were using a tiered excise tax structure when these data were collected in 2018, but as of 2020, that is no longer the case. Source: (1). 2.2 DESIGNING EXCISE TAX POLICY Significantly increasing the taxes on and prices of tobacco products is the most effective and most cost-effective policy to control tobacco use (4). Increased taxes that are passed on to tobacco users as higher prices reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face challenges ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues such as addressing the SCARE6 tactics of the tobacco industry. This section provides guidance for policy-makers regarding the best tax structure to use from a health perspective, taking into consideration all the appropriate tax designs. It also proposes recommended indicators to consider when formulating policy change. 6 SCARE tactics are the tactics most commonly used by the tobacco industry when countries plan to increase tobacco taxes. They are described, and refuted, in detail in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 19 2.2.1 THE IMPORTANCE OF THE TYPE OF EXCISE TAX STRUCTURE The existing theoretical and empirical evidence on approaches to the choice of (uniform) specific and ad valorem excises is reviewed below, along with their effects on price, consumption, perceived quality and variety of tobacco products, govern- ment revenue and tax administration. The use of the word “quality” in this chapter does not refer in any way to the health impact of a tobacco product. It refers rather to the consumers’ perceptions of quality and their decision to buy a product, which they may evaluate based on the packaging, the blend used for the cigarette or anything that makes the product more appealing to them. Just to be clear, from a public health perspective, all cigarettes are equally harmful even if perceived by consumers as having higher or lower quality. The choice between ad valorem and specific taxation is influenced by the market structure, i.e. the nature and degree of competition in the market for goods and services. Although each country has its own specific characteristics, the tobacco market structure is typically a monopoly or an oligopoly where firms have the power to control prices – and hence exploit the tax structure – to their benefit. For example, China, the largest producer and consumer of tobacco products in the world (5), has a state monopoly. In Viet Nam, foreign brands are produced under licence by the state monopoly. In Thailand and Egypt, despite the presence of foreign companies, the market is dominated by the state-owned company. In Uruguay, the oligopoly is led by a domestically owned company. In Bangladesh, the oligopoly consists of domestically owned companies competing with foreign companies (6). In most of Africa, the market consists of transnational tobacco companies (7). The impact of tax structure on final price: uniform specific versus uniform ad valorem The choice between specific and ad valorem taxes is a long-standing issue in tax policy, as the level and structure of excises have different implications for the interests and goals of various groups. Given the market structure of the tobacco industry – typically a monopoly or oligopoly for most products in most countries – different excises may have different effects on government revenue, manufacturer profit, consumer price, perceived product quality and variety and tax administration (8–16). Consequently, the two types of excise taxes – specific and ad valorem – may have different implications for public health to the extent that they affect individual consumption via their impact on perceived product quality, variety and prices. Moreover, governments have the potential to influence tobacco excises to manage demand, raise revenue and promote public health. 20 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax on prices and in turn on consumption is also influenced by price and income elasticities, as well as consumer perceptions of quality (“perceived quality”) and the variety of available products, which, in turn, are closely linked to the type of tax structure adopted. Tax structure is affected by both the price elasticity of demand and the price elasticity of supply. The price elasticity of demand measures the responsiveness of consumer demand to changes in prices. The price elasticity of supply measures how sensitive producers are to changes in prices. Tobacco tax structure is also influenced by market structure. In a monopoly, the profit-maximizing firm sets the price, considering the price elasticity of demand: the lower the price elasticity (in absolute value) – i.e. the less sensitive the consumer is to price changes – the higher the price the monopolist can set. Profits are typically abnormal in a monopolist market structure, meaning total sales revenue is higher than total cost (where total cost includes a normal profit). A monopolist producer therefore receives more than the minimum reward required to invest its (physical and human) capital and undertake business risks. Economic theory predicts that in a private monopoly, prices are higher than in an oligopolistic market. This is not, however, necessarily true when the monopoly is owned by the state and the government’s objective is not straightforward profit maximization: the government might have other considerations, such as preserving jobs (e.g. in China) or keeping prices low for low-income consumers (e.g. in Egypt). Under a monopoly, an ad valorem taxation structure enables the monopolist producer to set prices lower than would be possible under a specific tax structure. This is feasible because under ad valorem taxation, when supply increases and price falls, the price reduction is not fully borne by the producer. Rather, the price reduc- tion is partly shared by the government since, as supply increases, the tax per unit of product sold falls. In other words, ad valorem taxation leads to lower prices and higher consumption relative to revenue-equivalent specific taxation. Technically, this means that the supply function is less elastic under ad valorem taxation. In contrast, under a specific taxation structure, any increase in the monopolist producer price will go to the producers as revenue, which incentivizes them to increase prices. The same logic also applies to an oligopolistic market structure, where profits again are, in general, abnormal. KEY TAKEAWAY 1 In a monopoly or an oligopoly, specific taxation incentivizes industry to set prices higher than it would with ad valorem taxation. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 21 Understanding oligopolies, however, is more complicated, since they are characterized by strategic interdependence among a few firms. This strategic interdependence extends to the relationship between industry and regulators. Industry anticipates the government’s regulatory policy – whether through tax or other interventions – and acts accordingly. For example, competitors may coordinate and lobby against a certain tax structure reform or tax rate increase. Under an oligopoly market structure, ad valorem taxation is a relatively more efficient tool for transferring part of the profits to the government as tax revenue, since it acts like both an excise and a profit tax. In contrast, a specific tax has a smaller (negative) effect on profits. This explains why we observe multinationals that are leaders in high-priced brands (e.g. Philip Morris International [PMI]) lobbying in favour of specific taxation (17). As an example, in the countries of the Cooperation Council for the Arab States of the Gulf (GCC), the tobacco industry has been trying for a long time to lobby governments to introduce a specific excise (18–21). After years of consideration and discussions on the possible introduction of excise taxes, the GCC adopted the Common Excise Tax Agreement of the States of the Gulf Cooperation Council in November 2016 (21), which introduced an ad valorem excise on tobacco products. Tobacco companies’ support for excise tax structures ultimately depends on the market segments they control in a particular country. A company selling mainly premium brands will favour specific excises, whereas a company that sells mid-priced or economy brands would favour ad valorem excise (17). When oligopolistic firms produce identical products, a specific tax has a stronger positive effect on price and is more likely to be overshifted to consumer prices than an ad valorem tax (13). Overshifting means that the price increases by more than the tax increase itself. Empirical evidence supports this (22–26). KEY TAKEAWAY 2 In an oligopoly, prices are likely to increase by more than the amount of the specific tax increase when demand is relatively inelastic. In general, demand for a product depends not only on prices but also on consumer perceptions of quality and preferences for variety. For example, the most popular brand in GCC countries is Marlboro, a premium brand (1). Consumers differ in their willingness to pay, depending on their respective perceptions of quality, which influence whether they ultimately purchase high- or low-priced brands. A tax-induced price increase can cause the following plausible responses from consumers or users of tobacco products: (1) a group of consum- ers will quit; (2) a group of consumers will reduce their overall consumption; 22 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N (3) another group, most likely high-income users, will switch to an upgraded version of the tobacco product if its relative price (compared to the cheaper brand) has been reduced, which is the case under a specific tax increase; there might also be a group of consumers in the lower income range who switch to lower-priced variants of the tobacco product if the price gap increases, as is the case under an ad valorem tax increase; and (4) another group might switch to the illegal market or buy products in a neighbouring country with a lower tax where possible. Consumers’ decisions to purchase are also affected by their preferences for variety – meaning preferences among products that consumers perceive as equal in quality but are given different characteristics by the producers to account for consumer taste preferences. Thus, it is possible that a tax increase that leads to an increase in average prices will lead to an increase in the total quantity demanded in the market because of an increase in the variety of product choices available to consumers. Variety enables new consumers to be captured, especially in an environment lacking certain regulations (e.g. without plain packaging and flavour bans). The tobacco industry was able to capture a new group of consumers when it introduced menthol cigarettes into the tobacco market. There is more than sufficient evidence that menthol cigarettes increased youth smoking initiation, increased nicotine dependence and reduced adult smoking cessation (27). To prevent this from happening in their countries, Member States of the EU have prohibited characterizing flavours other than tobacco in tobacco products (28). It is therefore important to consider the broader effects that the structure and level of an excise tax can have on average price, perceived quality and the variety of cigarette brands and other emerging substitutes. When consumers make choices based on dimensions other than quantity, the two types of tax structures are not equivalent, even in a perfectly competitive market where firms have no market power (29–30). To illustrate this point, consider a US$ 1 cost to improve consumer perceptions of quality for a tobacco product. This will lead to an equivalent price increase under specific taxation but not under ad valorem taxation. At an ad valorem rate of 20%, the price must increase by more than US$ 1, or by 1/(1 – 0.2) to cover the US$ 1 cost of improvement, due to the multiplier effect. A specific tax induces consumers to reduce the quantity demanded, but they might still choose to pay a higher price in exchange for a product that they perceive to be of better quality. An ad valorem tax, on the other hand, leads to a reduction in both quantity and perceived quality, not a substitution between them. An ad valorem tax has only an income effect and – unlike specific taxation – does not lead to substitution between perceived quality and quantity. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 23 KEY TAKEAWAY 3 Under specific taxation, the industry has incentives to create upgraded variants of tobacco products that attract new consumers and encourage consumption. When firms produce differentiated products, as the tobacco industry does, economic theory provides ambiguous results regarding specific versus ad valorem taxation. The relative effects of the two types of tax are not as straightforward as in the case of oligopolistic firms producing a homogeneous product. With differentiated products, the relative effects of the tax types depend on various assumptions: whether or not firms face symmetric costs, whether the number of firms is fixed or new firms can enter the market and the level of the tax revenue requirement. When firms face different costs, ad valorem taxes exacerbate the absolute differences in marginal costs between them. The high-cost brand is not considered a perfect substitute for the low-cost brand. A sufficiently high ad valorem tax rate may lead to a relative underproduction of the high-cost products (31). The effect of specific and ad va- lorem taxes on consumer perceptions of quality depends on market structure and the price and income elasticities of demand across various qualities. The relative price of the cheapest product does not necessarily remain unchanged or increase; it might fall (32). Empirical evidence showing that increases in the specific tax lead to a lower market share for the cheaper generic brands and an upward shift to premium brands (33–34) usually considers gradual tax increases and ignores income effects. Chaloupka et al. (35) found that in 21 EU countries that impose a mixed tax system, the price gap between premium and low-priced brands – while not reflecting the full distribution of cigarette prices – is smaller when the specific component of the mixed structure dominates.7 Although the price gaps are narrower under specific taxation, there is evidence that firms sometimes respond by introducing new, very cheap (subvalue) brands, or they exercise differential tax shifting. This practice has been evident in India for quite some time, with the Indian Tobacco Company launching a number of cheaper variants of its flagship cigarette brand, Gold Flake, to take advantage of a lower excise tax rate in the so-called microcigarette (< 60 mm length) market (36). Consequently, the cheapest end of India’s cigarette market has expanded significantly in recent years due in part to the marketing of new brand variants like Gold Flake Century. 7 The EU countries impose a mixed tax structure with a minimum tax floor. Some countries rely on the specific component more than others, but they remain within a given range (the specific component must be between 5% and 76.5% of total tax share of the weighted average price). 24 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As another example, transnational tobacco companies, which have sold ultra- low-priced brands in the United Kingdom since 2006, have managed to double their market share in a few years: their real price did not increase, since they absorbed part of the tax increases (37). The share of ultra-low-priced brands increased between 2001 and 2009 from 5% to 10%, while the market share of economy brands increased from 40% to 50% and the market share of premium brands and mid-priced brands decreased during the same period (from 35% to less than 25% for the former and from 15% to 5% for the latter). In order to keep the price of discount brands low and certain consumers in the market, firms may overshift the tax for premium products and undershift it for the lower-priced products (37–40). KEY TAKEAWAY 4 Evidence suggests that the price gap between brands is narrower under a specific tax structure. As the tobacco industry simultaneously consolidates producers and widens its portfolio of products, evidence is emerging that it is introducing cheaper brands while increasing the price of its expensive brands, therefore paradoxically widening the price gap within its products. The extent of the impact is still unclear, however, and this evidence does not negate the overall conclusion that a specific tax structure reduces price gaps. The impact of tax structure on final price: uniform specific, ad valorem and mixed systems Evidence from the 2019 RGTE (1) data suggests that the average price of the most- sold brand of cigarettes – weighted by the number of smokers – is the highest in countries implementing a mixed system that relies more on specific excise, followed by countries applying specific excise taxes only, followed by countries applying a mixed system that relies more on ad valorem and then by countries that apply ad valorem excise only (Fig. 2.6). The price is lowest in countries that have no excise at all. In past WHO reports on the global tobacco epidemic, countries that applied specific excise only had the highest price, on average. The trend may have changed partly because more countries are adopting mixed excise systems. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 25 Fig. 2.6 Weighted average price and excise for a pack of the most-sold brand of cigarettes, by excise tax structure, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Shang et al. (41) describe and compare price distributions, using data from 16 countries of the International Tobacco Control (ITC) Project that impose different cigarette tax structures. Specific uniform taxation tends to result in less variability in prices than all other structures (ad valorem tax, mixed tax, tiered tax). In general, structures other than uniform specific tax give rise to more opportunities for brand switching and tax avoidance. Reliance on complicated systems is likely to be as- sociated with wider price distribution, leading to greater tax avoidance, as there are more opportunities for substitution with cheaper brands when taxes rise. KEY TAKEAWAY 5 Evidence suggests that the tax structures most likely to lead to higher prices are uniform specific excise tax structures or mixed systems that rely more on specific excises. Mixed system relying more on specic excise Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Mixed system relying more on ad valorem excise Ad valorem excise No excise 2.21 Specic excise 2.62 3.26 0 7 6 5 4 3 2 1 1.66 Excise tax Other taxes Retail price minus taxes PPP $ 7.56 PPP $ 5.66 PPP $ 4.84 PPP $ 3.73 PPP $ 2.52 26 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax structure on final price: uniform excise versus tiered tax systems Another aspect of tax structure that impacts final price is the use of tiered taxation, i.e. tax rates that vary according to product characteristics. The characteristics can vary, from price level to the type of tobacco leaf contained in the cigarette, the size of production volume, the packaging, etc. Table 2.1 (earlier) lists the criteria used by 31 countries as the basis for different tax rates. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax (see Fig. 2.7). Fig. 2.7 Weighted average price and excise for a pack of the most-sold brand of cigarettes for countries with and without tiered taxation, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Uniform specific tax structures are likely to lead to relatively higher prices with less variability in price distribution. Compared with tiered tax structures that have differential rates based on brand characteristics, uniform taxation may reduce consumers’ incentive to switch to cheaper brands (leading to higher quit rates and lower prevalence), as well as decreasing manufacturers’ incentive to reduce their tax liabilities by changing their pricing strategies, production process or size (42–45). In Indonesia, for example, where small producers were taxed more fa- vourably, manufacturers had an incentive to reduce their scale of production but increase the number of affiliated small companies. The issue was resolved when tax authorities considered the aggregate production of all affiliated companies in the application of differential tax rates. By 2017, there were 786 active factories, while Excise uniform Price and taxation per pack of 20 sticks (PPP $) Excise tiers 3.28 1.94 Excise tax Other taxes Retail price minus taxes PPP $ 4.99 PPP $ 6.30 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 27 there had been 4 198 factories in 2006.8 Abolishing the differential tax rates would have been even more effective in removing the incentive for tax avoidance, as well as helping improve both public health and government finances. KEY TAKEAWAY 6 Evidence suggests that applying a uniform excise tax on cigarettes is not only easier to administer than tiered systems but also more likely to lead to higher cigarette prices. The impact of tax structure on final price: the significance of the choice of the tax base It is important for the excise tax to be applied to the base that leads to the greatest possible effect on price and revenue. For specific taxation, the tax base is the quantity of tobacco products. The quantity of cigarettes, cigars and bidis is measured in number of sticks; for other tobacco products, such as smokeless tobacco or roll-your-own (RYO), it is measured in the weight of the tobacco. When the tax is ad valorem, the choice of the tax base is important not only for health considerations – due to its effect on consumption – but also for tax revenue generation and industry profits. An ad valorem tax that is based on the ex-factory price (or CIF value) provides tobacco manufacturers with opportunities to reduce their tax liability, especially when they control the distribution system. Tobacco producers may sell cigarettes to distributors who are related parties at a reduced price, which then serves as the basis for calculating their ad valorem tax liability. Distributors, however, can then set high prices and share the extra profit with the producers (46). Because of the potential for such trade mispricing, the best practice is to use the retail price as the tax base and introduce a minimum excise tax per pack. Data in the 2019 WHO RGTE (1) show that, on average, the price level of a pack of cigarettes and the excise level are both much higher in countries that use retail price as the base for their ad valorem excise (Fig. 2.8). The maximum retail sales price, which includes all taxes, is used as the ad valorem tax base in the EU. That price also forms the tax base for ad valorem taxes in a growing number of LMICs, including Brazil, Egypt, Thailand, Turkey and Rwanda. 8 Indonesian Ministry of Finance, personal communication, 2017. 28 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.8 Weighted average price of the most-sold brand of cigarettes in countries that use retail price as the base for their ad valorem excise, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 7 The base on which the excise is applied is important. For specific excise, the base needs to be clearly defined (for cigarettes, cigars and bidis, it is the number of sticks; for other tobacco products, such as smokeless tobacco or RYO, it is the weight of tobacco). For ad valorem excise – where the base is typically either retail price, CIF value or producer price – evidence suggests that countries that apply the excise tax on the retail price of cigarettes tend to have higher prices than those that apply the tax on other bases. CIF and producer prices are difficult for government authorities to ascertain and are prone to undervaluation. The tax impact on final price: the significance of the minimum excise tax The use of a minimum excise tax in countries with ad valorem or mixed systems is another important factor in determining final price. On average, the price of a pack of cigarettes – as well as the excise level – is much higher in countries that impose a minimum specific excise than in those that do not (see Fig. 2.9). While more than half of the 47 countries that apply a minimum excise are members of the EU, removing EU countries from the average calculations produces the same conclusions. Ad valorem/mixed with retail price as base 3.72 PPP $ 5.01 PPP $ 6.41 Ad valorem/mixed with other base 1.79 Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 29 Fig. 2.9 Weighted average price of the most-sold brand of cigarettes in countries with and without a minimum specific excise tax, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 8 Among countries that apply an ad valorem or mixed excise tax on cigarettes, evidence suggests that those that impose a minimum specific excise tax tend to have higher prices than those that do not. The minimum excise tax also helps guarantee minimum excise revenues. Summarizing the advantages, disadvantages and impacts of the choice of excise tax structure for tobacco products Table 2.2 summarizes the characteristics of different types of tobacco excise taxes and the advantages and disadvantages of each type in relation to its impact on quantity demanded, perceived quality of brands offered, price, certainty and stability of revenue, administration and enforcement and opportunities for tax avoidance and tax evasion as they are predicted by the economic theory of imperfect competition and observed in real life. Ad valorem/mixed with minimum speci c 1.80 4.23 Ad valorem/mixed without minimum speci c Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes PPP $ 4.82 PPP $ 7.41 30 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 2.2 Characteristics of different types of tobacco excise taxes Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) TA X B A SE The unit of product (e.g. 1 000 cigarettes) The value of the product (e.g. retail, wholesale or manufacturer price) Excise is calculated on an ad valorem basis; however, if the calculated tax falls below a specified minimum amount, a specific tax rate applies Unit and value of product Both unit and value, unless the calculated tax falls below a specified minimum, in which case the tax base is the unit A D M IN IS TR AT IV E R EQ U IR EM EN TS The tax should be collected at the point of manufacturing or at the time of importation Low, as only the volume of the products needs to be ascertained Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as with a pure ad valorem regime Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises, as well as minimum specific excise tax compliance U N D ER VA LU AT IO N Not an issue Susceptible to undervaluation Provides an easy tool to prevent undervaluation of low-priced brands subject to the minimum specific excise The ad valorem part of the excise collection may be susceptible to undervalua- tion, depending on the choice of tax base The minimum specific excise prevents possible ad valorem tax base undervaluation of low-priced brands CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 31 Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IM PA C T O N P ER C EI V ED PR O D U C T Q U A LI TY Upgrading effect tends to reduce the relative tax on higher-priced brands Multiplier effect provides a disincentive to costly so-called quality improvement No incentive to upgrade higher- priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands IM PA C T O N P R IC E Tends to lead to relatively higher prices, particularly for low-priced cigarettes Tends to lead to relatively lower prices; price reductions will be subsidized if the multiplier effect is strong Tends to lead to relatively higher price increases for low-priced cigarettes An increase in the specific tax will to lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem payment if the base of the ad valorem includes excise An increase in the specific tax will lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem tax amount if the base of the ad valorem includes excise. Increases in the ad valorem and /or specific tax will raise the minimum tax paid if the minimum is a percentage of the total tax on, for example, weighted average price; they will reduce price gaps, given impact on perceived quality 32 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IN FL AT IO N The real value of the excise tax will be eroded unless the tax is adjusted in line with inflation The real value of the excise tax will be preserved as prices increase, at least to the extent that tobacco product prices follow inflation The real value of the minimum specific excise will be eroded over time unless the excise is adjusted in line with inflation The real value of the specific excise will be eroded unless the excise is adjusted in line with inflation The real value of the specific excise and the minimum specific excise will be eroded unless the excises are adjusted in line with inflation H EA LT H B EN EF IT S Will discourage consumption of tobacco products irrespective of the price band May encourage more trading down in favour of cheaper cigarettes, reducing the health benefit The minimum specific excise reduces incentives for trading down May reduce trading down Reduces trading down Source: (47). 2.2.2 OTHER TAX DESIGN CONSIDERATIONS The significance of automatic adjustments and indexation of specific tax to inflation Specific taxation does not depend on price and therefore, unlike the ad valorem tax, is not automatically adjusted for inflation. The real value of a specific tax is eroded over time as the price of the taxed product increases. Therefore, especially in countries with rapid growth in inflation, the nominal value of the specific tax must be increased regularly in order for the tax to maintain its real value. This is of great importance for both public health and public revenues, especially in countries where manufacturers do not increase prices regularly and/or low-priced tobacco products are the dominant products in the market. Table 2.3 lists countries that include automatic adjustments to their excise in order to avoid the erosion of the specific excise over time, using different units of adjustment and based on different frequencies. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 33 Table 2.3 Countries that include automatic adjustments to the specific excise COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Argentina Inflation (consumer price index [ CPI]), on a quarterly basis Armenia Minimum specific excise set to increase in the Tax Code by 15% on average between 2019 and 2021 Australia Wages – excise rates on tobacco and tobacco products increase in March and September each year, based on average weekly ordinary time earnings Bosnia and Herzegovina Specific excise rate is increased annually by at least 7.50 convertible marks per 1 000 cigarettes; minimum excise tax is increased annually to be at least 60% of the weighted average price Canada Inflation – federal tobacco tax rates are to be increased every five years, indexed to Canada’s CPI starting in 2019 Chile Inflation Colombia Specific tax set to 1 400 pesos, increased to 2100 pesos in 2018; starting in 2019, it will increase yearly by the CPI plus 4 points Costa Rica Inflation Dominican Republic Inflation, on a quarterly basis France Increase from 2017 to reach an average price for cigarettes of €10 per pack by 2020 Honduras Inflation, annually to December of the previous year Italy Minimum tax burden calculated every year in March on the basis of the weighted average price of cigarettes sold in the previous year New Zealand Inflation annually plus 10% annually from 2017 to 2020 Nicaragua Updated annually as of 1 January 2017, taking the highest among the annual devaluation of the official exchange rate of Cordoba with respect to the US dollar, published by the Central Bank of Nicaragua, and the annual inflation rate of the CPI published by the National Development Information Institute, observed in the last 12 months available North Macedonia Specific and minimum specific rate increase by 0.2 denars per cigarette on 1 July each year until 2023 Philippines Agreed tax increases and rates for specific excise tax between 2020 and 2023, with a 5% indexation thereafter Romania Inflation, annual (1 January) adjustment of the total excise according to inflation calculated on 1 October of the previous year Serbia Inflation, every six months Southern African Customs Union (SACU) – Botswana, Eswatini, Lesotho, Namibia and South Africa Inflation, on an annual basis9 Sweden Inflation 9 While the adjustment is not strictly automatic in the SACU, it is greatly informed by the inflation rate. The Treasury has some discretion. In recent years, the increases have typically been slightly above inflation. 34 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Turkey Producer Price Index Ukraine Agreement to increase the specific component 20% annually between 2019 and 2025 United Kingdom Increase by 2% above the retail price index (measure of inflation) for the 2015–2020 Parliament Source: (1). KEY TAKEAWAY 9 To avoid erosion of their specific excise tax, countries need to regularly – and, ideally, automatically – adjust the excise to inflation. The significance of automatic adjustments and indexation of specific tax to income growth In addition to the risk of erosion due to inflation, the effect of a (specific) tax can be significantly reduced if the tax is not adjusted for increases in consumer income. Income growth makes products more affordable – thereby encouraging consump- tion – especially in countries with rapid income growth. Australia is one of the rare countries that explicitly adjusts its specific excise rates according to wage growth (see Table 2.3). However, a number of countries have adopted automatic adjustments that are higher than inflation and sometimes largely cover income growth as well (see also Table 2.3). Adjusting tax for income growth contributes to increases in prices that make tobacco products less affordable (see section 2.2.3). KEY TAKEAWAY 10 The specific excise tax needs to be adjusted to reflect income growth so that tobacco products do not become more affordable over time. Measures for specific contexts: the role of pricing and other non-tax regulation Emerging evidence indicates that the tobacco industry finds ways to mitigate the impact of higher taxes on prices. For example, despite the heavy reliance on specific taxation in the United Kingdom, a price differential between premium and cheap cigarettes still exists. There is evidence that the tobacco industry does not always pass tax increases on to cheaper products (37, 48). Differential shifting among price categories is also observed in the EU (49–50), New Zealand (38) and the United States (51). Therefore, the public health community has suggested that pricing regulation could be considered as a method of eliminating inexpensive tobacco products that are often used by the young and the poor (52). Three types of pricing regulation are described below: minimum mark-up, price floor and price ceiling. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 35 Pricing regulation • Minimum mark-up – It has been argued that a minimum mark-up of wholesale tobacco prices could be a better strategy to raise prices of tobacco products than excise tax increases. Minimum mark-up laws aim to discourage the sale of products below an assessed cost by imposing a mark-up to the cost declared at different levels of the supply chain. Some studies in the United States have shown that minimum mark-ups do not increase average cigarette prices (53–54). However, a recent study of the impact of minimum mark-up/ price laws has shown that these laws are linked with higher prices, especially for the cheapest brands, and could be used as an effective tool to mitigate the impact of the industry’s price-reducing promotions (55). Another concern related to minimum mark-ups is that they can be manipulated by manufac- turers and are likely to lead to higher profits for the industry, as well as extra administrative costs for the government (56). • Price floor – A few studies suggest that setting a price floor, or a minimum price, is an alternative strategy for increasing tobacco taxes, particularly with respect to reducing health inequities (57–60). A price floor, imposed by the government or as a vertical restraint imposed by the supplier upon retailers, is a price that firms cannot legally undercut. Governments impose price floors to restrain unfair competition or, in the case of services, to increase quality. It is difficult, however, to find the right floor or to anticipate unintended conse- quences or an industry’s adjustments. A study in Malaysia, where a minimum price for cigarettes was imposed in 2010, found that the policy did not seem to have a meaningful impact on prices: licit brand prices remained well above the minimum price, while illicit brands remained well below it. This outcome may be a result of the floor being set too low or the proportion of illicit trade being high, either of which would reduce the effectiveness of the policy (52). In the EU, imposing minimum retail sale prices for cigarettes could be a breach of harmonized legislation concerning the internal market, as minimum prices would distort competition. Therefore, increasing minimum excise duties is recommended instead, to discourage consumption (61). Increasing the minimum excise duties would also result in the additional revenue going to the governments instead of contributing to industry profits. A price floor would probably lead to increased industry profits – giving the industry greater funds for its marketing strategies – and lower tax revenue for governments, reducing their ability to cover costs associated with tobacco use. By reducing price competition, the price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifica- tions). Competition among firms may prevent them from raising their prices, 36 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N but a government that imposes a price floor does this for them. Minimum pricing is likely to create windfall profits for manufacturers and retailers. It can even help manufacturers sustain a cartel. If the industry uses the money to increase promotions, advertising or grant proposals for related research, this could undo some of the potential benefits of the policy.10 Some recent evidence shows that, at least in the case of the United Kingdom, increased concentration of power among a handful of multinational corpora- tions is enabling them to undermine tax increases through increased price segmentation, and that requiring minimum prices might be a good way to address the problem. A longitudinal analysis of price data from the United Kingdom (48) has shown that despite regular excise tax increases over time, average real prices for cheaper segments of the tobacco market (in this case, cigarettes and RYO) did not increase – indicating an undershifting of the tax increases in those segments that resulted in increased sales volume. At the same time, average prices for more-expensive market segments increased, indicating overshifting of the tax increases that resulted in decreased sales volume. This industry strategy ensures that the most price-sensitive consumers remain addicted, while encouraging initiation and discouraging cessation. Furthermore, segmenting the market further by overshifting the tax increase on premium brands while undershifting it for cheaper brands mitigates the impact of declining consumption resulting from higher taxes while increasing overall industry margins and profitability. Another situation where setting minimum prices can be a useful policy is specific to the United States. Banning marketing and promotions11 is not possible under the freedom of expression protections of the Constitution of the United States (Amendment I), and it was estimated in 2008 that more than 82% of all advertising and promotional spending by the tobacco industry was focused on reducing the price of their products at the point of sale (62). This limitation on how government can set policy has paved the way for the implementation of minimum price policies in many states and cities to counter the detrimental impact of price promotions on consumption and on the tax policy itself. Huang et al. (55) found that the presence of minimum price laws was associated with higher cigarette prices. They also noted that cigarette prices were even higher than prices resulting from minimum price laws in states that also prohibit industry from engaging in other price-reduction strategies, 10 See, for example, the PMI strategy of setting up the Foundation for a Smoke-Free World and grant proposals for related research. 11 See section below on banning promotional discounts for tobacco products for further discussion about marketing and promotions. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 37 such as below-cost combination sales, using trade discounts to reduce the base cost of cigarettes and distributing below-cost coupons to consumers. In the contexts of both price segmentation and price promotions, the imposi- tion of minimum prices as a complementary policy to excise tax increases – not as a policy alternative – may help guarantee that taxes do indeed lead to the intended reduction in consumption. Nonetheless, more evidence is needed to support the effectiveness of this policy. • Price ceiling – Concerns about differential tax shifting have led to suggestions that a price cap may benefit public health by limiting the tobacco industry’s ability to reduce average prices by differentially shifting tax increases among various price segments (63–64). Because tobacco manufacturers operate across international markets, however, they could maintain low prices in one coun- try but maintain overall profitability by selling more premium products in another country. Additionally, limiting price increases does not fit the public health purpose of reducing consumption. It is worth noting that maximum retail prices are sometimes used as a base for calculating the ad valorem tax payments in countries with ad valorem or mixed tax systems. In a systematic review of the literature on non-tax policy approaches to raising prices, Golden et al. (63) hypothesized how such policies would influence price dispersion and average prices. Their study found that minimum price policies combined with promotion bans have the potential to increase average prices. This is, of course, relevant in a context where price promotions are present. From either a theoretical or a practical standpoint, however, it is clear that price policies cannot be used alone and should always be considered as complements to excise tax increases. Significantly increasing taxes is the most effective way to dissuade consumption, correcting whatever bias may exist. Significant tax increases also provide the added benefit of raising money for the government rather than profits for the tobacco industry. Nonetheless, a minimum price might help narrow the gap between cheap and pre- mium cigarettes when applied to all tobacco products to avoid product substitution. Other non-tax regulation • Banning promotional discounts for tobacco products – The sale of tobacco products at a discount rate – such as through reduced-price coupons or buy-one-get-one-free offers – encourages consumption and undermines tax increases. Such practices should be completely banned. They often exist outside the realm of the finance sector because they are considered a type of marketing – promotional discounts are usually addressed in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. 38 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N According to the 2019 RGTE (1), 118 countries out of 195 with all levels of income had such a provision implemented as of 31 December 2018. • Banning the sale of single sticks of cigarettes – Article 16 of the WHO FCTC, “Sales to and by minors,” paragraph 3, requires Parties to “prohibit the sale of cigarettes individually or in small packets which increase the affordability of such products to minors”. Some smokers opt for buying single sticks partly because of the lower im- mediate costs of buying cigarettes individually (65). Internal (unpublished) analysis of single-stick prices collected by WHO for the 2012, 2014 and 2016 editions of the WHO RGTE shows that, in fact, the aggregate price of 20 single sticks of cigarettes sold separately is generally higher than the price of a 20-cigarette pack sold in the market of a specific country. Despite this fact, single-stick sales – and sales of small-sized packs – make cigarettes accessible to consumers with limited disposable income. De Ojeda (66) found in a study conducted in Guatemala that single-cigarette sales are associated with increased cigarette accessibility for less-educated, lower-income populations and minors. Single-stick sales are also a feature of many markets in South-East Asia, including most notably Bangladesh and India, but also in other parts of the world, e.g. South Africa. Single-stick sales also reduce the impact of a tax increase, since the in- crease per stick is much smaller than the increase per pack (67). In a study investigating how smokers in New York City responded to a tax increase of US$ 1.25 per pack in 2008, Coady et al. (68) found that 15% of smokers bought more single cigarettes than they had previously.12 By allowing single stick sales, governments risk losing part of the ad valorem taxes if the tax base is the retail selling price; the retail price of single sticks is much more difficult to monitor than the retail price of packs of cigarettes, on which, for example, tax stamps with prices can be applied. An internal WHO analysis of the most recent tobacco control laws in 2018 in 195 countries found that 86 countries impose by law a ban on the sale of single sticks of cigarettes (36% of the countries are high-income, and 64% are LMICs). In addition to banning the sale of single sticks of cigarettes, 67 of the 86 countries specify a minimum size for packs of cigarettes. Most (52 countries) use the 20 cigarettes per pack standard, but minimum sizes 12 Before 2018, the use of single sticks was possible, but it has since been banned. See New York City Administrative Code. chapter 7: regulation of tobacco products, subchapter 1: Tobacco Product Regula- tion Act, §17-704.a-1. New York: New York Legal Publishing Corporation; 2020 (http://library.amlegal. com/nxt/gateway.dll/New%20York/admin/title17health/chapter7regulationoftobaccoproducts?f=tem plates$fn=default.htm$3.0$vid=amlegal:newyork_ny$anc=JD_T17C007, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 39 range from five sticks in Ghana up to 25 sticks in Papua New Guinea. Another 15 countries do not specifically ban the sale of single sticks but do specify the minimum size for packs of cigarettes. According to the WHO FCTC, in order to reduce affordability, single stick sales should be banned and a minimum number of cigarettes contained per pack should also be defined. KEY TAKEAWAY 11 A number of non-tax measures are closely connected to tax policies, including price regulations, bans on promotional discounts for tobacco products and bans on the sale of single sticks of cigarettes. The price policies discussed are (1) minimum mark-up, (2) price floors and (3) price ceilings. Current evidence does not yet demonstrate that minimum mark-ups and price floors lead to increases in average price. Nonetheless, they may be relevant in some specific contexts as complementary policies to excise tax increases. Price ceilings limit price increases, which can mitigate their impact on consumption. Price marketing strategies such as promotional discounts and the sale of single sticks undermine the effect of tax policies and should be banned. A minimum pack size should also be required by regulators. Tax increases and their possible impact on inflation At times, the inflationary impact of tax increases on cigarettes and other tobacco products is raised as an argument for not increasing these taxes. This may be a concern in countries where wages and/or a significant share of government spend- ing is indexed to inflation (e.g. for public pension payments) or where government policy is to keep inflation low. The extent to which tobacco product tax increases lead to increases in inflation depends on several factors, most notably the share of these taxes in prices and the weight tobacco prices are given in computing a price index. For example, if taxes account for 25% of tobacco product prices, a doubling of the tax (100% increase) will increase prices by 25%. If the weight given to tobacco products in the price index is 3%, the index will rise by 0.75% in response to the tax increase. As tobacco taxes account for a larger share of tobacco product prices, the inflationary impact of a tax increase will be greater. Similarly, as tobacco products are given more weight in computing a price index, a given tax increase will have a greater inflationary effect. In general, for most countries, the inflationary impact of tobacco product tax increases would be relatively small (47). 40 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Consumer price indexes have multiple purposes. They are an important economic indicator for most countries and are often a key determinant of monetary policy. Inflation rates have a direct impact on interest rates and exchange rates. In many countries, changes in wages, social security benefits and other payments are tied to inflation as measured by a price index. Price indexes are used to provide more accurate comparisons of changes in expenditures, incomes and prices for specific goods over time, as well as to allow comparisons across countries. Given the many uses of consumer price indexes and the potential inflationary impact of tobacco tax increases, some governments have developed alternatives that exclude tobacco (and sometimes other goods) for some uses. For example, since 1992, France has excluded tobacco products from the price index used for adjust- ing minimum wages (47). However, many countries continue to include tobacco product prices in their consumer price indexes. Excluding tobacco products from the basket of goods used in developing key price indexes would greatly reduce concerns about their impact on inflation. In addition, with declining consumption of tobacco products, the inclusion of their prices in key price indexes results in a distorted measure of price for many consumers. KEY TAKEAWAY 12 If governments are concerned about the potential inflationary impact of a tobacco tax increase because wages or some government spending may be tied to a price index, they can use a price index that excludes tobacco products. The importance of taxing cigarettes and other tobacco products in a comparable way While cigarettes are the most commonly used tobacco product globally, other tobacco products are as prevalent and sometimes more prevalent than cigarettes in some parts of the world. Bidis and smokeless tobacco are the main products consumed in some countries in South-East Asia – Bangladesh and India in particular – and waterpipes are widely used for smoking tobacco in the Eastern Mediterranean region (4). These products, as well as RYO, have historically been taxed much less than cigarettes (see, for example, Fig. 2.10 for Bangladesh and India, where the excise tax and prices of bidis and smokeless tobacco are much lower than those for cigarettes). This differential taxation undermines the health impact of excise taxes on tobacco products because (1) it encourages users to switch from cigarettes to the lower-taxed product (see the case of Thailand below); (2) it is not effective in reducing tobacco use in general, especially if the most widely used product in the country is not cigarettes; (3) it can encourage tax avoidance by companies that may CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 41 redefine products that are similar to cigarettes so that they fall within the lower- taxed product category (see the EU example below) and (4) it reduces the size of government revenues since those products could have been taxed at higher levels. Fig. 2.10 Price and tax of the most-sold brand of cigarettes, bidis and smokeless tobacco in Bangladesh and India, 2018 Source: (1). In Thailand, for example, the price of cigarettes has been raised quite successfully through taxation over a number of years, while taxes and prices of loose or RYO tobacco have until very recently remained unchanged. Indigenous tobacco used for RYO cigarettes has historically been exempt from excise, while foreign tobacco was taxed at a low level relative to that of manufactured cigarettes. Consequently, Thailand experienced growth in the RYO market even though cigarette consumption had been falling.13 The Thai government eventually took strong action to address this issue. First, the exemption for indigenous tobacco was removed in 2018. The Cabinet then approved an increase in the excise rate on small producers (of indigenous tobacco) from 0.005 baht per gram to 0.025 baht per gram in 2020, with another increase to 0.1 baht per gram scheduled for 2021 (69). In the EU, the minimum excise duty levels for cigars and cigarillos is significantly lower than that for cigarettes. The Member States of the EU are required to levy an 13 WHO Country Office for Thailand, personal communication, 2019. 12.8 0 190 54.9 64 24 Price Excise tax amount 80 12.8 3.8 44 Lo ca l C ur re nc y 0 50 100 150 200 Cigarettes (20 sticks) Bangladesh, taka India, rupees Bidis (20 sticks) Bidis (20 sticks) Cigarettes (20 sticks) Smokeless (20 g) 42 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N excise duty of at least €90 per 1 000 cigarettes, which should be 60% of the weighted average retail selling price of cigarettes released for consumption. For cigars and cigarillos, only €12 per 1 000 items, or an excise duty of 5% of the retail selling price, is required. As a result, the excise tax share on cigarettes is much higher in many EU countries than the share for cigars and cigarillos. In response, some companies started to market so-called borderline cigarillos. These products have characteristics similar to cigarettes but can be sold at a lower price because for excise purposes, they are considered as cigarillos. Although this issue seems to be largely solved by amendments to the definitions of these products at the EU level and a change in tax structures in some countries, it is important to be aware of the unintended incentives that can be created by large gaps in excise tax levels between product categories (70–71). For more details on industry tactics to undermine tax increases, see Box 2.1. KEY TAKEAWAY 13 To make excise tax on tobacco products more effective in reducing overall tobacco use and to avoid substitution between products, all tobacco products need to be taxed in a comparable way. The Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce the demand for tobacco) (73) recommend that all tobacco products should be taxed in a comparable way. Box 2.1 Industry tactics used to undermine tax increases Tax increases reduce the demand for tobacco products and present a threat to the tobacco industry’s high profits. The industry responds by using various strategies (17, 46, 48), including the following (46): Stockpiling (forestalling/front-loading) – Before the implementation of an an- nounced tax increase, manufacturers overproduce tobacco products, paying the pre-tax-increase rate. As a consequence, sales and tax revenue decline immediately but temporarily after the tax increase (while sales and revenues had increased sub- stantially just before the tax increase) and the industry attributes this drop in revenue to the emergence or increase of illicit trade. This practice results in tax avoidance if there is no law prohibiting it (see also the discussion on anti-forestalling in Chapter 3). Changing certain product characteristics (for example, weight or length) and/ or adjusting the production process – When tobacco products are taxed at different rates or are subject to different tax increases, the industry can, for example, re-label one type of tobacco product as another product that has a lower tax burden (as in the example of the EU above). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 43 Choosing the time of a price increase announcement strategically – The industry may raise prices in anticipation of a tax rate increase, generating extra profits in the period until the tax is actually implemented. When the tax increase is implemented, consumption and tax revenue will fall, but prices will not change, so the industry can claim that the tax policy was ineffective in reducing demand. Adopting price-discriminating strategies or price-related promotions – The industry may offer discounts, retailer rebates or added value (gifts) to tobacco purchases to minimize the loss of price-sensitive consumers. This, however, is not possible in countries where strict bans on tobacco advertising, promotions and sponsorship are implemented. Using brand proliferation (for example, launching a low-priced brand) and price segmentation – Manufacturers can choose to reduce prices of certain brands or introduce new, even cheaper ones to keep price-sensitive consumers in the market. There is evidence that firms introduce new cheaper products and use price-marking – printing the price directly on packs of tobacco products – to lock in their price (48). Such practices compromise both public health and revenue objectives. Differential shifting of tax increases across different price segments, depending on the market circumstances – The industry may increase the price of a product by more than the amount of the tax increase (tax overshifting) and blame the govern- ment for the total increase. Tax overshifting is profitable when demand is inelastic, that is, when the price increase more than offsets the reduction in sales. The industry may overshift the tax increase for higher-priced brands, which are expected to be more price inelastic than lower-priced brands. Additionally, to keep price-sensitive consumers in the market, the industry may temporarily absorb part (or all) of the tax increase on lower-priced brands. The differential tax shifting will lead to different responses in the demand for the different brands (37, 48). Lobbying government to distort interventions – Government policy might be influenced by tobacco industry lobbying, directly or indirectly. Policy-makers are not simply welfare or revenue maximizers; they also value political support. Industry lobbying might lead to adopting a favourable type of taxation, postponing tobacco tax increases or distorting the tax rate downwards (17). Article 5.3 of the WHO FCTC, “On the protection of public health policies with respect to tobacco control from commercial and other vested interests of the tobacco industry”, and its guidelines provide useful guidance on how to address tobacco industry interference. In fact, all 181 countries that are Parties to the WHO FCTC have a legal obligation to implement the requirements of Article 5.3. Having correct expectations about industry responses is important for estimating the impact of a tax increase on consumption and tax revenue. 4 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.2.3 MEASURING IMPACT AND RECOMMENDED INDICATORS Governments need to consider a number of indicators when formulating policy changes. Inappropriate assumptions about consumer behaviour, market structure and industry behaviour can lead to faulty policy analysis. Measuring impact on price and demand Consumption habits, local traditions and industry characteristics – such as the number of different brands offered, the possibilities of cross-border shopping and the presence and level of illicit trade – all affect the shape of the demand and supply of tobacco products, thereby determining the value of the price elasticities. Price elasticity, together with the industry’s pricing strategies – for example, the degree of tax shifting – and the tax share in the retail price, determine the elasticity of the tax base, regardless of whether the base is determined by quantity (for specific taxation) or transaction value (for ad valorem taxation). The importance of elasticity estimates Different types of elasticity should be considered: • price elasticity – own-price elasticity – measures the response of consumers’ demand for a product following a change in the price of the product. – cross-price elasticity – measures the response of consumers’ demand for a product when the price of another product changes. Cross-price elasticity can also occur between different brands or price segments for the same product. • income elasticity – the response of consumers’ demand for a product when their income level changes. Correct estimates of price and income elasticities are important for policy-makers who need to anticipate the impact of a tax increase on consumption and tax rev- enue. Estimates will vary depending on a number of factors, including whether responses are considered in the short run versus the long run, the functional form of the demand function used, whether factors such as addiction or tax evasion are accounted for and the way data are constructed. For example, details such as the degree of aggregation of data, whether gender- or age-specific data are used, the time span covered and which estimation procedures are used (e.g. ordinary least squares, two-stage least squares or generalized method of moments) will all affect the results of the estimate (72). Price elasticities may change over time, as well because of changes in any of the other factors affecting demand, such as income or tobacco control measures, and also because of changes in estimation techniques and the types or sources of data used. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 45 Moreover, what is of most interest is the price elasticity of total demand. A tax increase may reduce tax-paid retail sales but not necessarily total consumption. For example, smuggling can significantly bias price elasticities when the elasticities are estimated using legal sales data; not accounting for possible illicit trade might lead to overestimation. Similarly, when cross-border shopping is included, the price elasticity of demand is lower (in absolute value) (74). Estimating the total price elasticity of demand for legal and illegal consumption can be done by using cross- sectional data from nationally representative household surveys. However, this approach also has its weaknesses. For example, respondents tend to underreport their consumption of tobacco, which leads to bias in the size of demand. Price endogeneity14 is another technical problem that can be challenging to address. To comprehensively estimate the total effect of a tax increase on demand for all tobacco products as well as on tax revenue, the degree of substitutability between them needs to be estimated (55). Cross-price elasticity measures how the quantity demanded of a particular tobacco product changes when the price of another tobacco product increases. When this elasticity is positive, the products are substitutes; the higher the value of the elasticity, the closer substitutes the products are to one another. For example, positive cross-price elasticity between RYO and manufactured cigarettes implies that the demand for RYO increases as the price for cigarettes increases. Substitutability may also arise between different cigarette brands – when the relative price of economy brands increases, demand for premium brands may increase. This effect can be exacerbated when differential (tiered) taxation is ap- plied on different types of cigarettes, further widening the gap in prices between brands and segments and encouraging substitution. The substitutability between traditional and new and emerging tobacco and nicotine products is currently of great interest (see section 2.4 below). In some countries, different tobacco products can also be complementary rather than substitute goods. This means that when the price of a tobacco product increases, the demand for its complement drops because users are unlikely to use the complementary tobacco product alone. For example, a number of studies have found manufactured and indigenous bidi cigarettes to be complementary goods in India (75–76). The sign and magnitude of income elasticity vary across time, countries and demographic groups. For example, in the United States, a high-income country, income elasticity over time has changed from positive to negative, and cigarettes have switched from being a normal good to an inferior good (77–78). On the other hand, among LMICs, where prevalence of smoking tends to be relatively higher, 14 Price is endogenous because it is not an independent variable: it is estimated by dividing expenditure on tobacco by consumption of tobacco, with consumption being a dependent variable in the estimation of price elasticity. 46 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarettes might still be a normal good, with consumption increasing as income levels increase (positive income elasticity) (79–80). There are unobservable characteristics that differentiate higher-income smok- ers from lower-income smokers, such as differences in time and risk preferences, differences in associating a social stigma with smoking and differences in taste for smoking as a pleasurable activity. When these characteristics are ignored, estimates of the correlation between income and smoking-related outcomes are biased. Kenkel et al. (81), using techniques that estimate the causal effect of income on smoking among low-income adults, found that tobacco is a normal (even a luxury) good: higher income is associated with a higher probability of smoking participation and a lower probability of smoking cessation. These results are consistent with those regarding the impact of the business cycle – periods of expansion or recession in economic activity – on health be- haviour and outcomes. Ruhm (82–83), for example, found that smoking declines during temporary economic downturns and increases during economic expansions. Tarantilis et al. (84) found that estimates of income elasticities of demand in Greece were higher after the economic crisis of 2010 than before it. The financial crisis and the austerity measures shifted the demand for cigarettes downwards and turned cigarettes into a more income-elastic good. Interestingly, evidence from Germany suggests that the propensity to become a smoker significantly increases during an economic downturn. However, among those who are already smokers, cigarette consumption actually decreases (85). Ideally, when estimating price and income elasticities, the effect of non-price policies should also be accounted for. A recent study from South Africa shows that failing to take non-price policies into account will overstate the price effect (86). The NCI/WHO Monograph (4) suggests that price elasticity of demand for tobacco is on average -0.4 in high-income countries (ranging from -0.2 to -0.6). Estimates for LMICs are more variable, clustering around -0.5 (ranging from -0.2 to -0.8). A price elasticity of -0.5 means that a 10% increase in price would lead to a 5% reduction in consumption. KEY TAKEAWAY 14 Policy-makers need to know the elasticity of demand – including price elasticity (own-price and cross-price) and income elasticity – for tobacco products in their country in order to correctly assess the impacts of potential policy changes on consumption and subsequent revenues. These estimates need to be made on a regular basis to capture changes in demand over time. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 47 The importance of the tax base elasticity The tax base elasticity measures the sensitivity to a change in the tax rate of the base on which the tax is imposed – the base being tobacco consumption in the case of specific taxation and tobacco expenditure in the case of ad valorem taxation. The magnitude of the elasticity of the tax base depends on price elasticity of demand, the tax structure, the level of the tax rate and its share in price, along with the industry response through its decision to absorb, pass through or overshift the tax on to the retail price. Consumers’ preferences and income, the availability of substitutes and other non-price tobacco control measures also influence the tax base elasticity, essentially through the price elasticity of demand. The magnitude of the elasticity of the tax base also depends on social motivations, including price and tax expectations, which are ultimately impacted by successful tobacco control measures that affect consumers’ willingness to pay taxes or prices. In addition, the tax-base elasticity depends on smokers’ perceptions of the prob- ability of detection and tax enforcement when using illegal products, as well as the availability and accessibility of opportunities for tax evasion and avoidance. Finally, consumers’ willingness to pay taxes depends on their perceptions regarding the use of the tax revenue (87). Therefore, the tax base elasticity is largely influenced by government policy choices. Increasing the tax share in prices is recommended by WHO as a tool to achieve the public health objective of reduced tobacco use: a higher tax share in prices increases the tax base elasticity, all else remaining constant, and therefore increases the reduction in the tax base through the resulting reduction in smoking. However, manufacturers can be expected to attempt to manipulate the tax base elasticity through their pricing policies, such as tax shifting. As discussed earlier in this chapter, industry behaviour is itself affected by government tax policy and regulations. A number of factors need to be taken into account when considering tax pass- through. As discussed earlier, tax is more likely to be overshifted within a specific tax structure than within an ad valorem structure. There is also evidence of industry overshifting the tax for premium or expensive cigarette brands while undershifting the tax for cheaper brands. This indicates that within a given market, the industry’s decision on the extent of tax pass-through will vary based not only on the tax structure but also on the structure of the market. It will also vary by brand. But this does not give an indication about the impact of the tax increase on the average price of a tobacco product. In the context of the tax base elasticity and the impact of tax increases on revenues, it is important to assess how tax increases affect average prices. The example of South Africa is very useful here. Over the past two decades, South Africa has been consistently increasing its specific excise tax on cigarettes, which has led to large price increases. An analysis of the effect of excise tax increases on 48 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarette prices between 2001 and 2015 (26) shows that while there is evidence of tax overshifting, at least within a specified period of time, overall, the tax has been undershifted in real terms. This undershifting is due partly to increased competitive- ness in the market and partly to the introduction of low-priced brands. Of course, this encouraged some consumers to downshift their consumption to cheaper products, but it also pushed manufacturers of more expensive brands to absorb part of the tax increase to reduce the impact on price. A change in the level of the tax rate – with all other factors that influence con- sumption held constant – will result in a change in the tax revenue.15 Estimates of tax base elasticity help governments predict changes in tobacco tax revenues following a tax increase (see details in Annex 2.2). Under specific taxation: • tobacco consumption – the tax base – is expected to be price inelastic (17, 47); • prices increase by less than the tax increase, on average (there is no tax overshifting overall); and • consumption – the tax base – is also expected to be tax inelastic: the quantity of consumption falls less than proportionately to the tax increase, and the tax revenue increases. Under ad valorem taxation: • the tax base is the total consumer expenditure (or, equivalently, the industry sales revenue) on (legal) tobacco consumption – that is, the tax base under ad valorem taxation is determined by both price and quantity, which is itself a function of price; • the sign of the tax base elasticity – which can be either negative or positive – depends on the magnitude of the price elasticity of demand; • since evidence suggests that tobacco demand is price inelastic, the tax base elasticity is positive; • when the ad valorem tax rate increases, both price and quantity adjust, but quantity falls less than proportionately to the price increase, and tax revenue increases; and • a tax rate increase leads to both a higher level of revenue and a lower level of consumption; the value of the elasticity – and hence the tax revenue – increases with the degree of tax shifting. 15 This concept has been used by Laffer to argue that tax increases that are too high will reduce excise tax revenues (the so-called Laffer curve). For a detailed discussion on the Laffer curve, see section 4.4 in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 49 Taxation serves as an instrument for achieving both fiscal and public health objectives. If, after successful tobacco control interventions, prices reach levels where demand becomes elastic, the tax base is still most likely to be inelastic due to tax undershift- ing, since overshifting is not a good pricing policy when demand is elastic. In other words, a tax rate increase – in combination with non-price tobacco control measures that make consumers more sensitive to price (tax) increases – leads to decelerating but still positive marginal revenues. For an example of a tobacco taxation success story, see Box 2.2. For further details of countries’ experiences with tax increases and their impact on revenues, see section 4.4 in Chapter 4. KEY TAKEAWAY 15 Policy-makers’ key policy tool to control demand is tax. Therefore, it is essential they assess not only the impact of price on demand but, more appropriately, the impact of tax on demand: this is the tax base elasticity. The tax base elasticity is essentially determined by (1) the price elasticity of demand, (2) the degree to which the industry will pass the tax on to the retail price and (3) the tax as a share of the retail price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. Box 2.2 A tobacco taxation success story: Turkey Turkey is an example of a country that has been increasing taxes regularly and sig- nificantly over a relatively short period of time and has reaped the benefits of this policy. As shown in Fig. 2.11, the excise tax per pack of cigarettes more than doubled in real terms over 10 years, with the real price almost doubling as well. In parallel, tobacco excise revenues increased by 67% and cigarette sales decreased by 20%. Since the beginning of the country’s Health Transformation Program in 2003, Turkey has successfully increased public health spending and collected more tobacco tax revenue. According to the latest available figures, in 2015, tobacco tax revenue was equivalent to 42% of the country’s public health expenditure and 1.5% of GDP (88). 50 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.11 Tobacco excise revenue and consumption in Turkey (base year 2008), 2008–2018 Sources: Reference 1 for the price of the most-sold brand, Ministry of Finance for the sales and revenue data and IMF world economic outlook, April 2020. See https://www.imf.org/en/Publications/WEO/weo- database/2020/April for the adjustment for inflation. Impact on affordability While price increases clearly have an impact on consumption, when the effects of increasing per capita income of a population are not considered, the price impact may not be as strong as expected. Increases in a population’s income also increase its purchasing power. And, as indicated earlier, tobacco products generally behave like a normal good. Consequently, as income increases, it is expected that tobacco consumption will increase as well. To mitigate this effect, price increases (following tax increases) need to be greater than increases in income. This is where the concept of affordability comes in. Affordability examines the effects of both increasing prices and increasing incomes on consumer behaviour. A common and easy way to calculate affordability, made popular by Blecher and van Walbeek (89), is to use the percentage of GDP per capita required to buy 2000 cigarettes (or 100 packs of 20 cigarettes) in a given year. An increase in this proportion over time will indicate that cigarettes are becoming less affordable and should lead to reductions in consumption. Changes in trends in affordability of cigarettes over time help policy-makers understand how prices are evolving 0 1 2 3 4 5 6 Tu rk is h lir as p er p ac k N um ber of sticks, 10 000 000 Turkish liras 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real excise tax amount Real price Real tobacco excise revenues Cigarettes, per capita sales 1 100 500 1 700 2 300 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 51 relative to a population’s ability to purchase cigarettes and enable them to revise their policies accordingly. Recent studies in India, for example, highlight the wide price differential between manufactured cigarettes and indigenous tobacco products such as bidis and chewing tobacco, as well as the propensity for these indigenous products to become more affordable over time due to favourable or more lenient tax policies towards them (90–91). Figure 2.12 shows the change in affordability of a pack of the most-sold brand of cigarettes by country income group between 2008 and 2018. During this time period, affordability declined in almost 70% of high-income countries, while it declined in slightly more than 35% of middle-income countries and only 26% of low-income countries. Fig. 2.12 Number of countries that have experienced a change in affordability of cigarettes between 2008–2018, by income level Note: Change in affordability was computed as the least squares rate of change in the per capita GDP required to purchase 2000 cigarettes of the most-sold brand in local currency in a given year. The trend rate of growth was computed for countries with four or more years of data, including 2018. Affordability was assessed as not having changed if the least squares trend in the per capita GDP required to purchase 2000 cigarettes over the period 2008–2018 was not statistically significant at the 5% level. Source: (1). KEY TAKEAWAY 16 From a health perspective, in addition to examining the impact of a tax increase on the levels of price, demand and revenues, policy-makers should consider a tax hike that will lead to prices rising more than increases in their population’s income; a tax increase should make tobacco products less affordable to consumers so that demand will be effectively reduced. Cigarettes became less aordable Cigarettes became more aordable Aordability did not change Could not be assessed due to insucient data High income Middle income Low income 39 36 9 13 5 7 37 23 6 13 2 5 52 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Projecting impact on consumption, smoking prevalence and lives saved The WHO interactive smoking projection and target-setting tool (WHO ISPT) The WHO ISPT enables national policy-makers and tobacco control experts to explore the potential impact of proposed tobacco control policies. It uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. The WHO ISPT provides projections of (1) tobacco smoking rates and (2) tobacco-smoking-related deaths in a country under different policy settings and for different time periods. It was designed to promote multisec- toral collaboration within countries by enabling experts from various ministries (for example, health, education, finance, national statistics), civil society, academia and media to explore options for medium- and long-term tobacco control planning together with WHO experts. Use of the WHO ISPT enables strong partnerships for policy change advocacy, program development and evaluation.16 In particular, it can help policy-makers in the Ministry of Finance assess the specific contribution of tax policies – within overall tobacco control policies – towards achieving specific targets in tobacco prevalence reduction. Projecting impact on excise revenue The WHO tobacco tax simulation model (WHO TaXSiM) The WHO TaXSiM is a simple but data-intensive Excel-based tool that helps policy- makers analyse their tobacco tax policy and assess the impact of any excise tax increase or change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Using detailed data about the market – including the majority of brands found in the market, their market share and price levels and the applicable tax – and assumptions about price elasticity of demand, the WHO TaXSiM predicts the impact of tax changes on consumer prices, consumption volume and tax revenues generated by each brand and market segment for the following year. The exercise can be done for multiple years.17 By exploring market data in detail, in addition to assessing the potential rev- enue impact of changes in excise tax, the WHO TaXSiM is a useful instrument for highlighting weaknesses and opportunities in an existing tax system and market. It can also encourage policy-makers to create administrative databases that can be periodically updated to monitor the dynamics of the cigarette market. 16 The WHO ISPT is not available publicly, but WHO will work directly with interested countries upon request to use it to produce data-to-action-type plans. 17 For more information about the methodology, see https://www.who.int/tobacco/economics/tax- sim_background.pdf, accessed 29 September 2020. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 53 KEY TAKEAWAY 17 Policy-makers can use available tools to simulate the impact of tax increases on prices, consumption and revenues, as well as smoking prevalence and lives saved. Recommended indicators to monitor tobacco taxation progress MPOWER WHO publishes a biannual RGTE, which monitors global progress in tobacco con- trol. In particular, the report focuses on the implementation of the policy package MPOWER, a set of proven demand-reduction measures in line with the key provi- sions of the WHO FCTC (1). While raising taxes on tobacco (component R) is proven to be the most effective and cost-effective policy to reduce tobacco use (4), implementing the entire MPOWER package at the best practice level will reinforce the impact of R. For example, as mentioned earlier in this chapter, banning promotional discounts as part of the E measure (enforce bans on tobacco advertising, promotion and sponsorship) will favour price increases following a tax increase. If all the MPOWER tobacco control measures except R were implemented at the best practice level, all else remaining constant, revenues would be expected to decline. Thus, in order to maintain revenue levels, it is important to raise excise taxes on tobacco products regularly to compen- sate for the decline in tobacco use from the other four tobacco control measures. Tax share The main indicator in the R policy in the RGTE (1) is the total share of indirect taxes in the retail price of the most-sold brand of cigarettes.18 Countries whose most-sold brand of cigarettes has a total tax that is equal to or greater than 75% of the retail price are considered to be at the highest level of achievement. While total taxes include excise taxes, VAT (or sales taxes), import duties (when applicable) and other indirect taxes (where applicable), it is preferable to focus on excise taxes, since they are the component that most influences the relative price of tobacco. The share of excise tax in the retail price can be extracted from the RGTE database.19 The 2010 WHO technical manual on tobacco tax administration recommended making excise taxes account for at least a 70% share of excise taxes in the retail price of tobacco products (47). 18 For more details about how this indicator was compiled, see Technical Note III of the RGTE 2019 (https://www.who.int/tobacco/global_report/Technical-Note-III.pdf?ua=1). 19 See taxes and retail price for a pack of 20 cigarettes, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.1-Taxes-and-retail-price-for-a-pack- of-20-cigarette-most-sold-brand.xls?ua=1, accessed 29 September 2020). 54 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Affordability As discussed previously, the share of tax in the retail price is not enough to ensure that a policy will be successful in reducing demand. Any tax increase should lead to an increase in price that will effectively discourage consumption. While global trends indicate that a high tax share is positively correlated with a high price level (see Fig. 2.2 in section 2.1.2), this may not necessarily apply to a particular country; a tax share can be high, while at the same time tobacco products remain afford- able. For this reason, it is important to monitor not only tax increases but also whether those increases led to a price increase that is greater than income increases. As described in section 2.2.3 of this chapter, a common indicator is the percentage of GDP per capita required to buy 100 packs of 20 cigarettes in a given year.20 Other indicators As discussed in detail in sections 2.2.1 and 2.2.2, a good tax structure can make a tax policy more effective in increasing prices and decreasing affordability of tobacco products. Indicators can include whether a uniform excise is applied, whether it is a specific excise and whether it is adjusted regularly for inflation. A number of such indicators are also monitored through the RGTE and can be downloaded online.21 A tobacco tax indicator compiled in 2020 combines the various elements that form a good tobacco tax policy. The Tobacconomics Cigarette Tax Scorecard (92) rates a country’s tobacco tax policy performance based on best practices. The four components that determine the level of performance are (1) cigarette price (in PPP), (2) changes in the affordability of cigarettes over time, (3) the share of taxes (total and excise) in retail cigarette prices and (4) the structure of cigarette taxes (i.e. whether excise is applied; whether it is uniform or tiered; whether excise is specific, ad valorem or mixed; and, for the ad valorem component, if the tax is applied on the retail price and if there is a minimum specific excise and, for the specific component, if tax is automatically adjusted upwards). Each of the four components is given a score, using a five-point index, with the total score reflecting an average of the four component scores. The closer the total score is to 5, the bet- ter the tobacco tax policy performance is in a given country. While this published scorecard is currently applied only on cigarettes, it can be easily applied on other tobacco products, provided the needed data are available. 20 This indicator has also been compiled in the RGTE; see (https://www.who.int/tobacco/global_report/ Table-9.6-Affordability.xls?ua=1, accessed 29 September 2020). 21 See supplementary information on taxation, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.5-Supplementary-information-on-taxation. xls?ua=1, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 55 KEY TAKEAWAY 18 Tobacco taxation works best if implemented as part of a comprehensive MPOWER package. MPOWER is an overall indicator that incorporates all the key demand-side tobacco control measures. KEY TAKEAWAY 19 In addition to assessing the potential impact of a tax increase, policy-makers need to monitor progress over time. The share of the tax in the retail price is an indicator of progress. However, it is important to remember that an effective tax increase must translate into higher prices in order to make tobacco products less affordable. Combining all the components of a good tax policy into one scorecard can also be useful for assessing tobacco tax policy as a whole. 2.3 DOMESTIC AND REGIONAL POLICY INTEGRATION While it is essential to design tobacco tax policies with the utmost consideration of all the aforementioned factors, it is also important to consider how external factors can impact or even impede public health policy objectives. As Chapter 3 explains, cooperation among the various agencies that are directly involved in tax administration, collection and enforcement is important for effective and efficient tax policy implementation. But in the design phase, it is also essential to engage with agencies and other policy-makers that are not directly involved with taxation. Domestically, coordination is required to ensure that policies in non-health sectors do not negatively impact or even counteract tobacco control initiatives. For countries that are part of a regional bloc, harmonization of tobacco taxation is essential to protect the single market – as well as the health of the population – and to prevent tax revenue erosion, tax avoidance and tax evasion. 2.3.1 INTERSECTORAL COOPERATION ON DOMESTIC POLICY Domestic policies in agriculture, industry, trade, finance and labour have the po- tential to create or support incentives at different stages of tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. For example, subsidies provided to farmers or manufacturers involved in growing or processing tobacco can reduce prices and incentivize continued participation or even increase development in these areas, which is counterproductive to the goals of making tobacco products less affordable and reducing tobacco consumption. Multisectoral integration and policy coherence are needed at the country level to ensure that public policies and interventions in non-health sectors do not act against the intended public health impact of tobacco control and taxation. 56 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 20 Greater policy coherence in agriculture, industry, trade, finance and labour should be promoted to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. 2.3.2 REGIONAL TOBACCO TAX HARMONIZATION Policy integration is driven by the recognition that cooperation on domestic policies can substantially increase the gains from forming a regional bloc. Harmonization is desirable and may be necessary in certain areas with spillover effects, such as tax policy, the possibility of a so-called “race to the bottom” or threats to public health. Harmonization could be as simple as setting minimum standards and requirements based on global norms and best practices. Harmonization of tobacco taxation is required to ensure the establishment and proper functioning of a single market, prevent tax revenue erosion, prevent tax avoidance and tax evasion and protect people. When barriers to trade between countries are removed, harmonized tax rates support the single market because they improve the ability of consumers, producers and investors to make decisions that are not distorted by taxation but reflect real opportunity costs. Tax competition – where countries simply undercut each other’s tax rate – could prevent governments from raising sufficient funds to pursue social policy. To avoid such a race to the bottom, countries can establish minimum tax rates within the customs union (93). Even if tax competition is not present, when substantial tax differences exist in neighbouring countries, there is a clear incentive to trade across borders in order to reduce tax payments legally or illegally. The experiences of established regional economic communities offer important policy lessons, not only in terms of the general integration process but also for the process and extent of tax policy coordination. The EU implemented a successful regional tax harmonization scheme. Over the years, the focus in harmonization of tobacco taxes has broadened from the elimination of tax obstacles to the fight against harmful tax competition, tax avoid- ance and tax evasion and, more recently, to public health protection. Naturally, addressing these issues requires increasing convergence in fiscal policy and tax administration. Although price differentials still exist, setting a minimum on the share of taxes in the final price of tobacco products as well as a minimum excise tax has helped countries reach some level of harmonization. The EU experience confirms that both a declining tobacco consumption trend and stable revenues can be achieved with harmonized minimum excise rates (94). Moreover, the harmonization process has CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 57 offered opportunities for the enforcement authorities (tax or customs) to obtain information that can be used in the fight against fraud and tax evasion. On the other hand, the experience of harmonization efforts in the West African Economic and Monetary Union (WAEMU) shows how the absence of a supranational body (like the EU) or a hegemonic member state (see the SACU example below) can slow down policy integration that would benefit all member countries (95). The eight countries of the WAEMU are bound by a Tax Directive22 that requires them to impose an ad valorem excise on the CIF value or producer price of tobacco products, which is subject to under-declaration and is difficult to ascertain. Additionally, a maximum excise rate is imposed, and some members apply additional taxes to deal with this constraint. The Directive was revised in 2017 (96), but unfortunately the tax structure remains the same, and the maximum rate was not removed but rather has been increased. The Southern African Customs Union (SACU), which has five member countries, is the oldest existing customs union, established in 1910. Thanks to the hegemonic lead of South Africa, a country with a sophisticated administration system and an aggressive tobacco tax policy, SACU adopted a well-integrated tax policy that has benefited all its members (95). The GCC, established in 1981, is a regional intergovernmental political and economic union consisting of six states of the Persian Gulf. Home to one fifth of the global oil supply (97), the GCC has never relied on taxation as a source of revenue; no direct or indirect taxes were applicable in the region. Although there was no excise on tobacco products, as a customs union, the GCC countries have a common external tariff. This common tariff includes harmonized rates but also a harmonized structure. The import duty is 100% of the CIF value of tobacco products imported in the region, with a minimum tax per quantity imported. However, in recent years, to reduce their dependence on income from oil, GCC countries have considered diversifying their sources of income, including by de- veloping reliance on indirect taxes such as excise and VAT. In 2015, a decision was adopted at the 36th GCC summit meeting to implement selective taxes on all imported tobacco products and cultivated raw tobacco grown domestically (GCC Decision number 963/1). A follow-up decision in December 2016 formally agreed to the introduction of an excise tax on tobacco and other products such as sugary and energy drinks, as well as special goods (alcohol and pork meat), in all GCC countries. The decisions at the national level to implement this subregional decision came into force gradually in all GCC countries, starting with Saudi Arabia, which began 22 Directive No. 03/98/CM/WAEMU on the harmonization of Member States’ legislation of excise duties was adopted 22 December 1998. It was amended by Directive No. 03/2009/CM/WAEMU of 27 March 2009 with the objective of harmonizing excise duties within WAEMU. 58 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation in June 2017, followed by Bahrain and the United Arab Emirates (UAE) later that year, Qatar in 2018 and Oman in 2019 (1). Only Kuwait has yet to adapt its national laws accordingly. The excise introduced by the GCC countries has a structure somewhat similar to the import duty on tobacco products: the rate is 100%, but the base was changed from the CIF value to the retail price excluding taxes. The introduction of the excise led to large increases in the price of the most- sold brand of cigarettes in member countries between 2016 and 2018 – by 33% in Bahrain, more than 80% in the UAE and more than 100% in Saudi Arabia (1). In federations such as Canada and the United States – where the central govern- ment has real taxing power and some financial and regulatory control over the states or provinces – tobacco taxes are not harmonized (98–99). Even though there are significant interjurisdictional differences in taxes and prices, and tax harmonization holds great potential to reduce the scope of illicit transactions in the tobacco market, there is little evidence that Canadian provinces or individual states in the United States are interested in tobacco tax harmonization. Tax harmonization is most relevant in the context of further economic integration within a group of countries that are already part of a customs union, but it needs to be planned well to be effective. Discrepancies in law interpretation and a lack of standardization of tobacco product definitions and tax base lead to suboptimal situations. Tax rate alignment, or setting minimum rates, should come after tax structure alignment. It is important that governments support the move towards harmonization and are committed to dedicating enough financial resources and skilled personnel to oversee the entire process. KEY TAKEAWAY 21 In the context of regional economic integration and ongoing discussions regarding the possibility of harmonizing tobacco excise taxation among member countries, the experiences of existing groups can be instructive. So far, only the EU, SACU, WAEMU and, more recently, the GCC have effectively implemented a harmonized approach to excise taxation of tobacco products. Lessons learned indicate that harmonization should be planned well and should not come at the expense of tobacco control. Setting a common minimum specific excise tax, adjusted over time, is the best approach. This ensures that taxes and prices are above a minimum level, encouraging equalization of price levels and at the same time reducing affordability across countries. On the other hand, agreeing on maximum tax rates is a bad policy. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 59 Countries that wish to raise their taxes further for revenue purposes, health concerns or both should be given the space to do so. Concerns about illicit trade provoked by higher tax rates are best dealt with by strong cooperation in administration and enforcement, information sharing and adoption of new technology with common or interoperable information systems. 2.4 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In recent years, awareness of tobacco risks and harms, implementation of tobacco control provisions – especially under the WHO FCTC – and tightening of regulations have resulted in declining sales of cigarettes, primarily in high-income economies. This has changed the dynamics of the tobacco market. In response to these effective tobacco control measures, the tobacco industry has diversified its business by promot- ing a new portfolio of products, which they claim to be technological innovations that supposedly reduce the harms and risks associated with conventional tobacco products, particularly cigarettes. So-called novel tobacco products have been promoted by the tobacco industry as “cleaner alternatives,” “safer alternatives” and “reduced harm/risk products” with no smoke and no ash. On the basis of these claims, they negotiate for less-restrictive regulatory environments within countries. Some of the new products are also mar- keted or promoted for smoking cessation, despite the evidence of this outcome being inconclusive. Where these products are not banned, one of the debates in the global health community concerns the issue of their regulation and taxation. 2.4.1 HEATED TOBACCO PRODUCTS (HTPs) HTPs are tobacco products that produce aerosols containing nicotine and toxic chemicals upon heating of the tobacco or activation of a device containing the tobacco. These aerosols are inhaled by users sucking on or smoking the device. They contain the highly addictive substance nicotine (found in tobacco) as well as non-tobacco additives and are often flavoured. The tobacco in HTPs may be in the form of specially designed cigarettes (e.g. so-called heat sticks or Neo sticks) or pods or plugs. These products include IQOS from PMI, Ploom TECH from Japan Tobacco International (JTI), glo from British American Tobacco (BAT) and PAX from PAX Labs. HTPs differ not only from con- ventional cigarettes but also from ENDS – some of which are called e-cigarettes – as ENDS do not contain tobacco but rather a nicotine solution (see next subsection). However, the boundaries between the different products are becoming increasingly 60 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N difficult to define, given the emergence of so-called hybrid tobacco products that contain both nicotine solution and tobacco. HTPs are currently available in more than 40 countries and are banned in fewer than 10 countries. Even in countries where they are regulated, there is significant variation in the approaches taken to regulation. A variety of factors affect a country’s ability to control and regulate the use of HTPs, including national regulatory pow- ers, enforcement capacity regulatory frameworks, country capacity and tobacco industry interference (1). Most countries tax HTPs at a lower rate than cigarettes and on the kilogram of tobacco as a base when applying a specific or mixed excise (see Table 2.4). The use of such a base may be quite challenging for tax collection, especially because of the difficulty of checking the tobacco content in each stick. In the past, some countries taxed cigarettes per kilogram of tobacco, but today it is common practice to tax them per stick regardless of tobacco content. Table 2.4 Excise taxation of HTPs, first collected for July 2018 – updated for July 2020 OVERALL COMPARISON WITH CIGARETTES Type of excise Base unit is kg, overall rate lower than cigarettes Base unit is sticks, rate is the same as cigarettes Base unit is sticks, rate is lower than cigarettes Other Specific excise Albania, Austria, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Greece, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Montenegro, Netherlands, New Zealand, North Macedonia, Romania, Russian Federation, Slovakia, Slovenia, Sweden, United Kingdom Azerbaijan a, Japan, Ukraine b Armenia, Hungary, Jordan, Italy c, Philippines, Republic of Korea d Montenegro e, Republic of Moldova f , Serbia g Ad valorem excise (base is retail price unless specified other- wise between brackets) Spain, Switzerland Saudi Arabia and United Arab Emirates (base is retail price exclusive of excise and VAT) Indonesia h CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 61 Mixed system (ad valorem compo- nent based on retail price unless specified other- wise between brackets) France, Germany, Poland, Portugal Colombia, Georgia, Israel and West Bank and Gaza Strip (ad valorem excise base is wholesale price) a The specific excise rate applied is the same as for imported cigarettes, higher than the rate applied to domestically produced cigarettes. b The rate is the same as the minimum excise on cigarettes per 1 000 pieces. Rate and structure were effective as of 1 January 2021. c The specific excise rate is defined as 25% of the excise tax on cigarettes based on an equivalency used between cigarettes and HTPs. There are planned increases of this proportion to 40% by 2023. d In 2020 the specific excise rate was only 11% lower than cigarettes. e The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. f Specific excise rate is higher than for cigarettes but, unlike HTPs, cigarettes also face an ad valorem excise. Overall effect of excise is a slightly lower for HTPs. g The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. There is a planned phased increase of this proportion aiming equalization with cigarettes by 2025. h While cigarettes face a specific excise tax rate, HTPs face an ad valorem rate, the highest rate as defined by law, on the basis of a pre-defined minimum price. Sources: (1, 100, 101, WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021 and the Campaign for Tobacco-Free Kids website on Taxation and Price for Heated Tobacco Products https://www.tobaccofreekids.org/what-we-do/global/taxation-price/staging-tax-gap). A study by Liber (102) compared prices of HTPs and cigarettes in 34 countries and showed that while taxes have been systematically lower for HTPs than for cigarettes, prices were higher in half of the countries surveyed. KEY TAKEAWAY 22 HTPs, when taxed, are usually taxed lower than cigarettes, although they generally seem to be priced higher than cigarettes. It is important to remember that HTPs are tobacco products, and the same provisions that apply to tobacco products should apply to them as well. This is articulated in WHO’s information sheet on HTPs (103), which provides guidance on how these products should be regulated, as well as Decision FCTC/COP8(22) for novel and emerging tobacco products. Moreover, MPOWER measures, which help WHO Member States to implement the demand-reduction articles of the WHO FCTC, are applicable to HTPs, in particular, Article 6 for taxation. Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. Furthermore, HTPs contain chemicals not found in cigarette smoke, the health effects of which are not yet known. Independent assessment of industry data demonstrates that more than 20 harmful and potentially harmful chemicals 62 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N are significantly higher in HTP emissions than in cigarette smoke (104). Therefore, there is a need to learn more about these products and the health impacts of their emissions, as well as the impacts of exposure to these emissions. KEY TAKEAWAY 23 Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. From both public health and tax administration perspectives, HTPs should be taxed at the same level and in the same way as tobacco cigarettes. Some countries have already adopted this approach and are taxing HTPs at the same rate per stick as cigarettes (Azerbaijan, Colombia, Georgia, Israel, Japan, Ukraine and West Bank and Gaza Strip). Saudi Arabia and the UAE, which have recently introduced an excise tax on tobacco products as part of the GCC, are now applying the same import duty rate and excise tax structure for cigarettes and HTPs. Continuing developments in technology and changes in products have led to a recommendation to tax HTPs per unit. The definition of unit may vary by product within the HTP category. For example, one unit of IQOS is one heat stick, for Glo it is one Neo Stick and for Ploom TECH it is one tobacco pod. Governments will need to determine the exact definition of a unit for each product allowed on the market. The potential complexity of the market strongly supports limiting the types of HTPs allowed in a country and setting strict regulations to standardize the products as much as possible. Countries can also consider taxing the devices used to consume HTPs, i.e. the holder and the charger (see product description in Annex 3.1). KEY TAKEAWAY 24 HTPs are tobacco products, and they need to be treated as such. Where they are not banned, HTPs need to be strictly regulated and taxed. The recommendation is to tax them at the same level as cigarettes on a per-unit basis. Countries can also consider taxing the devices used for HTP consumption. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 63 2.4.2 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS)23 Products like ENDS and ENNDS have evolved rapidly over the past decade. ENDS heat a solution (e-liquid) containing nicotine, but not tobacco, and other chemicals that may be toxic to people’s health to create an aerosol, which is inhaled by the user. Examples of ENDS include Juul from Juul Labs, Vype from BAT and blu from Imperial Brands (1). Electronic non-nicotine delivery systems (ENNDS) are essentially the same as ENDS, but the e-liquid used generally does not contain nicotine. Upon testing, however, many so-called zero-nicotine solutions are found to contain nicotine (105–107). While generally considered as a single product class, ENDS products constitute a diverse group with potentially significant differences in the production of toxicants and delivery of nicotine. There are several coexisting types of devices for ENDS/ ENNDS on the market, including first-generation or so-called cigalikes, second- generation tank systems and even-larger third-generation or personal vaporizers. Collectively, they are also often referred to as e-cigarettes, vapes or vape pens. Other categories of ENDS include e-hookahs, e-pipes and e-cigars – hence, ENDS is an all-encompassing term for multiple product categories. Some of the products resemble their conventional tobacco counterparts – cigarettes, cigars, cigarillos, pipes or hookahs – while others are shaped more generically like pens, USB memory sticks or basic cylinders. Different forms of nicotine are also used in these ENDS, the most recent one being nicotine salts, which deliver high levels of nicotine (1). There are two types of ENDS/ENNDs products: open systems and closed systems. Open systems are devices that allow the user to buy e-liquids and fill their device with the mixtures they want (with no nicotine, different nicotine concentrations and/or flavours). Closed systems are products that come with a prefilled container (called a cartridge, pod or tank). For the past decade, divisive debates have been waged over the effectiveness of ENDS as smoking cessation aids – especially for tobacco users who are unable to give up the habit – as well as the possibility of ENDS playing a role in public health. However, the evidence remains inconclusive. Despite the tobacco industry and other related industries promoting these products as tools for quitting smoking, current evidence does not support their use as part of a population-based cessation strategy (108). Accordingly, the United States Surgeon General, in January 2020, concluded that 23 It is worth noting that ENDS are not tobacco products and not exactly new products – the technology has been around since the late 1980s (e.g. Premier, Eclipse and Accord). However, the recent generation of these products is new and has more or less piggybacked on the success of e-cigarettes. 64 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N E-cigarettes, a continually changing and heterogeneous group of products, are used in a variety of ways. Consequently, it is difficult to make generalizations about efficacy for cessation based on clinical trials involving a particular e- cigarette, and there is presently inadequate evidence to conclude that e-cigarettes, in general, increase smoking cessation (109). The evidence on the adverse health effects associated with use of ENDS is mounting, and when ENDS are used in combination with smoking – which is the practice of the majority of ENDS users (110) – the adverse health effects of two or more products are combined. However, there are insufficient data to understand the full breadth of these effects, as ENDS have not been on the market long enough for their long-term effects to be established. Nevertheless, the evidence is clear that the aerosols of the majority of ENDS and ENNDS, some of which are cancer causing chemicals. ENDS also contain nicotine, which is highly addictive. In addition, ENDS are associated with increased risk of cardiovascular diseases and lung disorders, as well as adverse effects on the developing fetus during pregnancy (108, 110). For adolescents, the use of nicotine can lead to dependence and may harm brain development. Use of ENDS could also lead to a new generation of nicotine and tobacco users, as seen in some countries, especially since these products are designed to appeal to young people. Although the specific level of risk associated with ENDS has not yet been determined conclusively, these products are undoubtedly harmful. Therefore they should be strictly regulated if allowed to be sold in domestic markets, and must be kept away from children. Taxation will be a key component of regulation, since it is an effective tool for influencing consumer behaviour. Some countries have taken the bold decision to completely ban these products. Approaches that have been taken range from partial to comprehensive bans, and ENDS/ENNDS products were banned in more than 30 countries in 2018.24 In other countries, they are regulated as, for example, consumer products, pharmaceutical products or tobacco products, or they are completely unregulated. WHO recom- mends that where ENDS/ENNDS are not banned, they should be regulated to achieve the following objectives: 1. prevent the initiation of ENDS/ENNDS by non-smokers and youth, with special attention to vulnerable groups; 2. minimize as much as possible potential health risks for ENDS/ENNDS users and protect non-users from exposure to their emissions; 3. prevent unproven health claims being made about ENDS/ENNDS; and 24 Data collected for the WHO RGTE 2019. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 65 4. protect tobacco control activities from all commercial and other vested interests related to ENDS/ENNDS, including the interests of the tobacco industry. ENDS/ENNDS present a risk to youth, who have taken up their use in high numbers in some countries, including Canada and the United States (1, 111). The Juul brand, for example, has quickly gained a significant e-cigarette market share in the United States (112–113). Its marketing and popularity have led the United States Food and Drug Administration (FDA) to raise serious concerns and to seek solutions to effectively prevent youth from taking up the use of ENDS/ENNDS (114). The city of San Francisco banned the sale of e-cigarettes in June 2019 (115). In addition to posing a risk for initiation by youth, ENDS can attract non- tobacco users or prevent current smokers from quitting. Taxation could play a role in preventing the uptake of these products, specifically among non-smokers, vulnerable groups, children and adolescents. KEY TAKEAWAY 25 The long-term health effects of ENDS/ENNDS products are still unknown, but they are clearly harmful to health. Furthermore, evidence on the effectiveness of ENDS products as a smoking- cessation aid remains inconclusive. Taxing these products could play a role in preventing their uptake, specifically among non-smokers, vulnerable groups, children and adolescents. Price elasticity of demand for ENDS products In the context of taxation, it is important to ask whether demand for ENDS is price-responsive. Preliminary evidence, although almost exclusively focused on e-cigarette data from the United States, indicates that this is the case: demand for e-cigarettes may be even more price-responsive than the demand for conventional cigarettes, so taxes can be used to deter initiation by never-users (116–123). Most of the studies of price elasticity of demand for ENDS products also demonstrate that e-cigarettes and conventional cigarettes are partial substitutes – that is, they show positive cross-price elasticity. The magnitude of the elasticity indicates the degree of substitutability between products: the higher its value, the closer the products are to being substitutes, with higher cigarette prices being associated with increased e-cigarette sales. Some of the studies also show a substitutability effect in the other direction, with increased prices for e-cigarettes leading to an increase in conven- tional cigarette use (117, 120). All of the studies show evidence of substitutability except for one (124), which differentiates between exclusive and dual users and shows no evidence of substitution between e-cigarettes and conventional cigarettes. 66 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The presence of concurrent (or dual) use – consumers using both conventional cigarettes and ENDS products – complicates results and highlights the need for more research in this area to better disentangle the different effects. Liber et al. (125) analysed sales prices in a sample of 45 countries and concluded that comparable units of conventional cigarettes cost less than disposable e-cigarettes. The units considered for pricing e-cigarettes included both the e-liquid and the rechargeable device. Taken alone, the price of e-liquids is on average much lower than that of cigarettes in high-income countries and the same in LMICs. The time needed to buy back a rechargeable device is estimated to be less than two weeks in most countries. One can argue that increasing price differentials by further increasing taxes on regular cigarettes could be effective in driving current smokers of regular cigarettes to e-cigarettes (126) as a potentially lower-risk alternative (127). However, the ef- fectiveness of ENDS as smoking cessation devices is still being debated; a study by Sweet et al. (128) shows that dual use of e-cigarettes as a potential tool for cessation was effective only in the short term. Moreover, significantly more smokers said they would quit if cigarette prices doubled and e-cigarettes were not available (122) or that they would never have become addicted to nicotine if e-cigarettes had not been so readily available (129). Once an e-cigarette user is addicted to nicotine, there is a risk of initiating traditional tobacco products use (130). In general, cessation can be better facilitated by governments via stronger implementation of the other tobacco control policies that have been proven effective at reducing use. KEY TAKEAWAY 26 Few studies are available on the price elasticity of ENDS products, and the available data come almost exclusively from the United States. These early studies indicate that demand for e-cigarettes will go down as the price of e-cigarettes increases. Generally, the results also show that cigarettes and e-cigarettes are partial substitutes, where an increase in cigarette price would increase the demand for e-cigarettes while reducing demand for cigarettes. But these results do not differentiate between people who are exclusive cigarettes or e-cigarette users and those who are users of both products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 67 Tax structure Different countries impose different tax structures on ENDS/ENNDS products25 (see details in Table 2.5). The Republic of Korea, for example, imposes a specific tax per millilitre of ENDS/ENNDS e-liquid (131), while Indonesia imposes an ad valorem tax on the retail price of the e-liquid; the maximum rate allowed by law for tobacco products (132). In the United States, there is no common way to tax e-cigarettes among the states that do tax them (133–134). The situation is similar in the EU, where new and emerging nicotine and tobacco products are not currently covered by the tobacco tax directive, and Member States may apply a national tax as they see fit under their own rules. All the EU countries that tax ENDS products apply a specific excise per millilitre of e-liquid. These different tax treatments have the potential to distort the functioning of the internal market. In February 2020, the European Commission concluded that the current provisions of the harmonized directive are no longer relevant for the taxation of ENDS and HTPs, and this is a source of concern from the internal market perspective (135). In June 2020, the Member States of the EU reiterated that it is urgent and necessary to upgrade the EU regulatory framework by harmonizing defini- tions and the tax treatment of novel products such as ENDS/ENNDS and HTPs (2). Table 2.5 Types of excises applied on ENDS/ENNDS products e-liquids globally and in individual states in the United States, as of July 2019 (updated as of July 2020 for all countries except the United States) TYPE OF EXCISE COUNTRIES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Albania, Kazakhstan, Kyrgyzstan, Pakistan, Portugal, Republic of Korea, Romania, Russian Federation, Slovenia, Sweden Azerbaijan, Cyprus, Estonia, Finland, Georgia, Greece, Hungary, Italyb, Latvia, Lithuania, Montenegro, Morocco, North Macedonia, Philippines, Serbia Ad valorem (% of retail price or import value) Bahraina Indonesia, Jordan, United Arab Emirates, Yemen TYPE OF EXCISE INDIVIDUAL STATES IN THE UNITED STATES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Delaware, Illinois (Chicago), Cook County, Louisiana, Ohio, Puerto Ricoc, Connecticutc Kansas, North Carolina, Washington, West Virginia, Wisconsin 25 The focus is on the e-liquid used for ENDS/ENNDS products. 68 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ad valorem (% of wholesale/ distributor price) Alaska (Juneau, Matanuska- Susitna Borough), California, Illinois, Maine, Maryland (Montgomery County), Minnesota, Nevada, Pennsylvania, Vermont, Washington DC, Virgin Islandsc New York Mixed New Jersey, New Mexico a Tax applied to e-shisha (or e-hookah) because e-cigarettes are banned in Bahrain. b Italy imposes differential rates for nicotine and non-nicotine containing liquids. c States in which it is unclear if only ENDS or both ENDS and ENNDS products are taxed with an excise. Sources: (1, 135, complementary data from Frank Chaloupka and WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021). Table 2.6 provides reference material on the pros and cons of different considerations for determining the tax structure and base of ENDS/ENNDS products e-liquids. Table 2.6 Excise tax options for ENDS/ENNDS products e-liquids TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Specific Volume of nicotine- containing e-liquid (regardless of concentration) 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only the volume needs to be determined 1. Difficult to compare if tax equivalencya with cigarettes is sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarettes and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed 5. May encourage more do-it- yourself (DIY) products where e-liquids are mixed by the users themselves, which increases the risk of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 69 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of e-liquid regardless of nicotine presence 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only volume needs to be determined 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Does not require laboratory capacity to detect the presence of nicotine in liquids 1. Difficult to compare if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed Volume of all e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL are affected the most; also reduces the price gap between different products 2. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 3. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 4. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 5. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to user 1. Difficult to compare with cigarettes if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed 70 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of nicotine- containing e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL affected the most; also reduces the price gap between different products 2. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as the nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 3. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 4. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to the user 1. Difficult to compare if tax equivalency with cigarettes is being sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids as there is no simple way currently available to determine whether the e-liquid contains nicotine; self-declarations by industry are not sufficient 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed Ad valorem Producer price/ CIF value of nicotine- containing e-liquid (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 71 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Producer price/ CIF value of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Retail price of nicotine- containing e-liquids (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 2. Requires capacity to monitor the market to assess market prices 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled in an e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risk of accidents, illness and death 72 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Retail price of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires capacity to monitor the market to assess retail prices 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Ad valorem with minimum specific or mixed Ad valorem + min specific 1. Volume of nicotine- containing e-liquids will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of nicotine- containing e-liquids will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up. b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products 1. Requires capacity to monitor the market to assess retail prices 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotin; (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Difficult to set a minimum specific excise amount/specific excise amount, especially if tax equivalency with cigarettes is sought 5. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 73 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Ad valorem + min specific 1. Volume of all e-liquids regardless of nicotine presence will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of all e-liquids regardless of nicotine presence will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 1. Requires capacity to monitor the market to assess retail prices 2. Difficult to set a minimum specific excise amount/specific excise amount, especially if equivalency with cigarettes is sought 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels a Tax equivalency can be measured in different ways: (1) in terms of tax burden (as % of the retail price) or (2) as the exact amount of tax for equivalent quantities (assuming an equivalency between a certain volume of e-liquid and a pack of cigarettes). b There is also the option to use the producer price/CIF value as a base for the ad valorem component, but it is a weaker option because the base is difficult to ascertain and therefore prone to undervaluation. Note: Table compiled following a WHO Expert Meeting on Taxation of Electronic Nicotine and Non-Nicotine Delivery Systems (ENDS/ENNDS), Geneva, Switzerland, 2–4 September 2019. There is currently a lack of evidence on the practical challenges being faced by countries favouring one approach over the other. Furthermore, such data are difficult to obtain because the nature of the market is constantly changing. However, a clear recommendation can be made with regard to which e-liquids to tax. As indicated in Table 2.5, some countries tax all e-liquids – whether or not they contain nicotine (ENDS and ENNDS products) – while some tax only nicotine- containing e-liquids (ENDS products). As shown in Table 2.6, there is evidence that in a number of instances, ENNDS products do contain some nicotine. Additionally, ENNDS products are not harmless (136–137). It is therefore recommended that all e-liquids be taxed for both ENDS and ENNDs products. The question of whether to employ differential taxation based on nicotine content seems reasonable from a health perspective, since nicotine is a toxic substance. However, this would likely create an additional burden for tax administrators as 74 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N they would need to determine the nicotine concentration of e-liquids on the market. Additionally, this may no longer be relevant, as advancements in technology indicate that other features of the product can influence nicotine delivery beyond the actual concentration of the e-liquid. It is now possible to increase nicotine delivery at low nicotine concentrations by increasing battery power (by reducing resistance or increasing voltage) (138). In terms of implementation, while most countries seem to have adopted a specific excise tax on ENDS/ENNDS e-liquids per millilitre, one benefit of implementing ad valorem taxation is that it seems relatively easier to regulate in the context of a large heterogeneity of products. However, it is essential that the tax be applied on the retail price value of the products, as this base is easier to ascertain than any other value that could be declared by the manufacturer. It is also important to add that regulation of the characteristics of ENDS/ENNDS products is essential, and it should be implemented along with any tax policy adopted. Regulations should include: 1. setting a maximum nicotine concentration per millilitre to safeguard public health, including reducing the risk of dependence, especially among youth; 2. setting a maximum volume for cartridges to reduce toxicants exposure and possibly limit use; 3. setting a maximum capacity for refill containers to reduce toxicants exposure and possibly limit use; 4. setting a maximum battery power to reduce the possibility of influencing nicotine and toxicant delivery; and 5. taxing nicotine regardless of its source (e.g. tobacco, eggplant, synthetic). Countries may choose to impose an excise tax on ENDS and ENNDS devices26 as well. The easiest type of tax would be an ad valorem tax based on the declared retail price. If countries choose not to impose an excise tax on these products, they should at least impose the regular VAT or sales tax rate. Imposing an excise tax on devices can be challenging from an administrative perspective, as all components need to be clearly defined and classified as devices for ENDS/ENNDS consumption. For example, if the device is assembled after importation and some parts may be used for other purposes than ENDS/ENNDS consumption, authorities may face a challenge in detecting and differentiating which component parts would be subject to excise tax and which would not. 26 See Annex 2.3 for an overview of elements of devices used in ENDS/ENNDS products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 75 KEY TAKEAWAY 27 There is currently not enough evidence to recommend one tax structure over another for ENDS/ENNDS products. However, it is clear that taxing the e-liquids used for consumption is key. The excise tax should be applied on all e-liquids, whether or not they contain nicotine. If the preferred type of excise tax is ad valorem, it should be applied to the retail price. Countries can consider taxing devices as well, but they need to adequately assess their administrative capacity to do so. Policy-makers need to be mindful of the diversity and rapid evolution of ENDS/ ENNDS products and adjust accordingly. Regulation must reflect this reality so that loopholes will not be exploited by the industry. For example, ENDS/ENNDs products include not only e-cigarettes, vapes and vape pens but also other categories such as e-hookahs, e-pipes and e-cigars. Lawmakers need to be clear about how ENDS/ENNDS products are defined so that subcategories do not fall under the radar when regulation comes into effect. Definitions will also be relevant when it comes to taxation. An unclear definition can lead to a seemingly contradictory situation, such as in Bahrain, where e-cigarettes are banned but e-hookahs are not.27 Finally, while policy-makers need to be mindful of the emergence of new products and must take appropriate actions to protect the health of their citizens, it is important to remember that the overwhelming share of nicotine consumption remains that of tobacco products, especially cigarettes. The total market value of ENDS/ENNDS and HTPs sales in 2018 was less than 2.2% of the total market value, while cigarette sales alone accounted for 91% of the same total market value (139–140). 2.5 CONCLUSIONS An overview of excise tax application globally reveals a broad variety of price and tax levels, as well as structures used for taxing tobacco products, in particular, cigarettes. Some trends, however, indicate that tax and price levels are higher among higher- income countries. The rate of taxes also matters: higher tax rates are correlated with higher prices, and higher prices change behaviour, which leads to a reduction in consumption. More countries are moving away from ad valorem taxes and towards either mixed or specific excise systems, and there are few countries that do not impose any excise tax on cigarettes. 27 In Bahrain, the Ministry of Production and Trade Decision 38 of 2013 banned e-cigarettes, while the official list of excisable products from the Ministry of Finance includes e-shishas (or e-hookahs), making them apt to be taxed and therefore considered legal. 76 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Significantly increasing the taxes and prices of tobacco products is the most effective and cost-effective policy to control tobacco use. Increased taxes – which are passed on to smokers as higher prices – reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face several challenges, ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues, as well as the perceived contribution of the tobacco sector to economic development. In designing tax policy, the tax structure adopted not only affects consumption overall, it also shapes the market structure. Ad valorem taxation incentivizes industry to set prices lower than specific taxation does. Evidence suggests that under a specific tax, the price gap between premium and lower-priced products is narrower, therefore reducing incentives for substitution to lower-priced products following a tax increase. However, as industry consolidates producers and widens its portfolio of products, new evidence indicates that the industry is introducing cheaper brands while increasing the price of its expensive brands, therefore, paradoxically, widening the price gap between its products. Evidence also suggests that prices are higher under a specific excise tax structure. Additionally, from a tax administration perspective, a specific tax is easier to imple- ment, since only the quantity produced needs to be ascertained rather than the value of the product. Another aspect of tax structure is the use of tiered taxation – that is, tax rates that vary on the basis of different product characteristics. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax. Tiered taxation encourages substitution from premium to cheaper brands, maintaining smoking prevalence and reducing the health im- pact of tax rate increases. In addition to leading to lower prices, tiered taxation is difficult to administer and creates opportunities for the tobacco industry to avoid and evade taxes. The design of a tax structure must also consider the base on which tax is applied. The choice of base should lead to the highest possible effect on price and revenue. For specific taxation, the tax base is the quantity. When the tax is ad valorem, the CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 77 choice of the tax base is important not only for health considerations, through its effect on consumption, but also for tax revenue generation, as well as industry profits. An ad valorem tax based on the producer price, or CIF value, gives tobacco manufacturers opportunities to reduce their tax liability, especially when they control the distribution system through related parties. The best practice in an ad valorem (or mixed) excise structure is to use the retail price as the tax base and introduce a minimum excise tax per pack. Other tax design considerations include the importance of using automatic adjust- ments and indexation to inflation and income growth for the specific excise tax in order to avoid erosion of the tax over time. Emerging evidence indicates that tobacco taxation does not always achieve the intended results, because the tobacco industry finds ways to circumvent it. Non- tax policies such as pricing regulation (in particular, minimum mark-ups or price floors/minimum prices) may be seen as a complementary approach to ensuring a high price level and discouraging consumption of tobacco products. So far, these policies have not proven to increase average prices. A price floor is likely to lead to increased industry profits, giving the industry greater funds for its marketing strategies (such as the introduction of new products), and lower tax revenues for governments. By reducing price competition, a price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifications). However, where powerful multinationals are operating in certain markets with presence in all market segments and with the capability to overshift a tax on some brands while undershifting the tax on others, or where price promotions cannot be banned, minimum price policies may help increase the effectiveness of tax increases. Other non-tax policies affecting price levels are those relating to promotional dis- counts for tobacco products and the sale of single sticks of cigarettes. Both should be completely banned. The ban of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Higher taxes are the most effective way to dissuade consumption, with the added benefit of raising money for the government – money that can be earmarked for health and education programs, rather than going as profits to the tobacco industry. Additionally, in order to make excise tax on tobacco products more effective in reduc- ing overall tobacco use and in line with the recommendation of the Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce 78 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N the demand for tobacco), all tobacco products need to be taxed in a comparable way; the focus should not be on cigarettes only. Tax choices and reforms have various and sometimes conflicting consequences for the market. For example, there might be a trade-off between quantity and variety or perceived quality implications. It is important for the government to recognize that firms respond strategically to changes in tax policy. Close monitoring of the market is necessary to form correct expectations about industry responses and enable estimates of the impact of a tax increase on consumption and tax revenue. To estimate the total effect of a tax increase on demand for tobacco products and tax revenue, it is important to use correct estimates of the own-price elasticity of demand, the cross-price elasticity and the income elasticity of demand. It is also important to use updated estimates of demand elasticities, as the environment within which consumers make decisions continues to change. For example, financial crises or successful tobacco control interventions can be expected to shift demand and change elasticity. Another key measure of the impact of tax policy is the tax base elasticity. Policy- makers need to be mindful of the three key components of tax base elasticity: (1) the price elasticity of demand of tobacco, (2) the share of the tax in the consumer price and (3) the degree of pass-through of the excise tax rate increase to consumer price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. It is important to acknowledge that if tax increases lead to increases in prices be- low concurrent increases in income levels, they will not be effective in reducing consumption, as tobacco remains a normal good in most countries. Policy-makers need to account for affordability when considering tax increases. They should ensure that tax increases are high enough to increase prices above income growth so that consumption goes down effectively. When designing tax policy and deciding on the right level to impose, policy-makers need to assess and project the impact of their policy decisions. Monitoring and evaluation are important. Tools for measuring impact can be very helpful, and several such tools exist. The WHO ISPT, for example, looks not only at the impact of tax policy but also at a set of tobacco control policies, and this enables national policy-makers and other tobacco control experts to explore the potential impact of CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 79 future tobacco control policies. The tool uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. More specific to tobacco tax policy, the WHO TaXSiM assesses the impact of any excise tax increase and change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the effectiveness of their policies and whether those policies have an impact on tobacco use over time. The implementation of the MPOWER package is one useful indicator for assessing tobacco control overall. Tobacco taxation works best if it is implemented as part of a comprehensive MPOWER package. The tax share in the retail price of a selected tobacco product is one indicator of the effectiveness of tax policy, but a more important one is affordability, that is, whether tax increases do lead to price increases that are above income and general price increases. A useful indicator to assess the performance of the tax policy overall is the Tobacconomics Cigarette Tax Scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). Domestic policies in agriculture, industry, trade, finance and labour all have the potential to create or support incentives at different stages in tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. Greater domestic policy coherence should be pro- moted across different sectors of the government to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. Differential tax structures and rates have the potential to distort the functioning of the internal market. Harmonization of tobacco taxation ensures the establishment and proper functioning of a single market; prevents tax revenue erosion, tax avoidance and tax evasion; and protects people’s health. In this context, tax competition, where countries simply undercut each other’s tax rate, might prevent governments from achieving their tobacco control objectives and raising sufficient funds to pursue public health policies. To avoid such a race to the bottom, countries can establish minimum tax rates on all tobacco products. A common high minimum specific excise tax is the best approach to ensure that taxes and prices are above a minimal level. 80 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N In recent years, the world has experienced the rise of new and emerging tobacco and nicotine products, including ENDS/ENNDS and HTPs, which the industry claims are safer than traditional tobacco products. The evidence so far suggests that these products could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of these products as well as initiation, cessation and switching of tobacco use behaviours among different socioeconomic groups, are not yet clear. Until more evidence for the claimed benefits of these tobacco products is available, caution should be taken in developing tax policy. Therefore, the current recommendation is for HTPs to be taxed at the same level as cigarettes on a per-unit basis regardless of tobacco content. In countries where they are not banned, ENDS/ENNDS products must be regulated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS products taxation. Nicotine- and non-nicotine-containing e-liquids should be taxed equally. If ad valorem excise is chosen as the structure, the base should be applied on the retail price. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While the evolution of new and emerging tobacco and nicotine delivery systems merits the attention of administrators and regulators, it is worth remembering that the bulk of the world’s tobacco consumption is still in conventional cigarettes. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 81 REFERENCES 1. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 2. Outcome of proceedings: Council conclusions concerning the structure and rates of excise duty applied to manufactured tobacco. Brussels: Council of the European Union; 2 June 2020 (8483/20; https://www.consilium.europa.eu/media/44235/st08483-en20.pdf, accessed 28 September 2020). 3. Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco: Article 7, paragraph 2. Brussels: OJEU. 2011;176:24–36 (https://eur-lex.europa. eu/LexUriServ/LexUriServ.do?uri=OJ:L:2011:176:0024:0036:EN:PDF, accessed 28 September 2020). 4. U.S. National Cancer Institute and World Health Organization.The economics of tobacco and tobacco control. NIH Publication No. 16-CA-8029A. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute; 2016 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/brp/tcrb/monographs/monograph-21, accessed 28 September 2020). 5. Drope J, Schluger N, Cahn Z, Drope J, Hamill S, Islami F, et al. The tobacco atlas. Atlanta: American Cancer Society and Vital Strategies; 2018 (https://tobaccoatlas.org/wp-content/uploads/2018/03/ TobaccoAtlas_6thEdition_LoRes_Rev0318.pdf, accessed 24 January 2021). 6. Sector reports, cigarettes. In: GlobalData [website]. London: John Carpenter House; 2019 (restricted access). 7. Vellios N, Ross H, Perucic AM. 2018 trends in cigarette demand and supply in Africa. PLOS ONE. 2018;13(8):e0202467 (https://doi.org/10.1371/journal.pone.0202467, accessed 28 September 2020). 8. Keen MJ. The balance between specific and ad valorem taxation. Fiscal Studies. 1998;19(1):1–37 (https://www.ifs.org.uk/fs/articles/fsmick1.pdf, accessed 24 January 2021). 9. Kay JA, Keen MJ. The structure of tobacco taxes in the European community. IFS report series. Institute for Fiscal Studies; 1982. 10. Kay JA, Keen MJ. How should commodities be taxed?: Market structure, product heterogeneity and the optimal structure of commodity taxes. Euro Econ Rev. 1983;23(3):339–58. 11. Kay JA, Keen MJ. Commodity taxation for maximum revenue. Public Finan Q. 1987;15(4):371–85. 12. Kay JA, Keen MJ. Product quality under specific and ad valorem taxation. Public Finan Q. 1991;19(2):238–47. 13. Delipalla S, Keen MJ. The comparison between ad valorem and specific taxation under imperfect competition. J Public Econ. 1992;49(3):351–67. 14. Suits DB, Musgrave RA. Ad valorem and unit taxes compared. Q J Econ. 1953;67(4):598–604. 15. Skeath SE, Trandel GA. A pareto comparison of ad valorem and unit taxes in noncompetitive environments. J Public Econ. 1994;53(1):53–71. 16. Myles G. Imperfect competition and the optimal combination of ad valorem and specific taxation. Int Tax Public Finance. 1994;3:29–44. 17. Effectiveness of tax and price policies for tobacco control. Geneva: WHO Press, World Health Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications. iarc.fr/Book-And-Report-Series/Iarc-Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And- Price-Policies-For-Tobacco-Control-2011, accessed 28 January 2021) 18. Bevan J. GCC taxation. Aylesbury (UK): Rothmans Exports Limited; 1992 (https://www.industrydocuments. ucsf.edu/docs/yxpy0198, accessed 29 September 2020). 19. Morton IT. Note from IT Morton to I Birks regarding agreement on specific taxation; 8 February 1992 (https://www.industrydocuments.ucsf.edu/docs/#id=nxpy0198, accessed 29 September 2020). 20. Galgut R. Oman. BAT (UK and Export) Limited; 16 April 1992 (https://www.industrydocuments.ucsf. edu/docs/xxpy0198, accessed 29 September 2020). 21. Common excise tax agreement of the states of the Gulf Cooperation Council [website]. Abu Dhabi: UAE Ministry of Finance; 2016 (https://www.mof.gov.ae/En/lawsAndPolitics/govLaws/Documents/ Common%20Excise%20Tax%20Agreement.pdf, accessed 29 September 2020). 82 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 22. Brock B, Choi K, Boyle RG, Moilanen M, Schillo B. Tobacco product prices before and after a statewide tobacco tax increase. Tob Control. 2016;25(2):166–73. 23. Hanson A, Sullivan R. The incidence of tobacco taxation: evidence from geographic micro-level data. Natl Tax J. 2009;62(4):677–98. 24. Delipalla S, O’Donnell O. Estimating tax incidence, market power and market conduct: the European cigarette industry. Int J Ind Organ. 2001;19(6):885–908. 25. Keeler TE, Hu TW, Barnett PG, Manning WG, Hai-Yen S. Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxation. J Health Econ. 1996;15:499–512. 26. Linegar DJ, van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2017;0:1–7 (https://tobaccocontrol.bmj.com/content/27/1/65, accessed17 February 2021). 27. Villanti AC, Collins LK, Niaura RS, Gagosian SY, Abrams DB. Menthol cigarettes and the public health standard: a systematic review. Public Health. 2017;17:983 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5747135/, accessed 24 January 2021). 28. Directive 2014/40/EU of the European Parliament and of the Council of 3 April 2014 on the approximation of the laws, regulations and administrative provisions of the member states concerning the manufacture, presentation and sale of tobacco and related products and repealing Directive 2001/37/EC: Article 7. Brussels: OJEU. 2014;57:1–38 (https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=OJ:L:2014:127:FULL&from=EN, accessed 24 January 2021). 29. Delipalla S, Keen MJ. Product quality and the optimal structure of commodity taxes. J Public Econ Theory. 2006;8:547–54. 30. Obara T, Tsugawa S. The welfare comparison of ad-valorem tax and specific tax with both quality and quantity choice of a consumer. Theor Econ Lett. 2017;7:1797–1813 (https://www.scirp.org/journal/ paperinformation.aspx?paperid=79862, accessed 24 January 2021). 31. Anderson S, de Palma A, Kreider B. Tax incidence in differentiated product oligopoly. Journal of Public Econ. 2001;81(2):173–92. 32. Razzolini L, Shughart WF, Tollison RD. On the third law of demand. Econ Inq. 2003;41:2. 33. Sobel RS, Garrett TA. Taxation and product quality: new evidence from generic cigarettes. J Polit Econ. 1997;105:880–7 (http://faculty.citadel.edu/sobel/All%20Pubs%20PDF/Generic%20Cigarettes. pdf, accessed 24 January 2021). 34. Nargis N, Fong GT, Chaloupka FJ, Li Q. The choice of discount brand cigarettes: a comparative analysis of International Tobacco Control (ITC) surveys in Canada and the United States (2002-2005). Tob Control. 2014;23:i86–i96 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/suppl_1/i86. full.pdf, accessed 24 January 2021). 35. Chaloupka FJ, Peck R, Tauras JA, Xu X, Yurekli A. Cigarette excise taxation: the impact of tax structure on prices, revenues and cigarette smoking. Cambridge: National Bureau of Economic Research; 2010 (NBER Working Paper 16287; https://www.nber.org/system/files/working_papers/w16287/w16287. pdf, accessed 24 January 2021). 36. Cigarettes in India, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 37. Gilmore AB, Tavakoly B, Taylor G, Reed H. Understanding tobacco industry pricing strategy and whether it undermines tobacco tax policy: the example of the UK cigarette market. Addiction. 2013;108(7):1317–26 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3746125/, accessed 24 January 2021). 38. Marsh L, Cameron C, Quigg R, Hoek J, Doscher C, McGree R, et al. The impact of an increase in excise tax on the retail price of tobacco in New Zealand. Tob Control. 2016;25(4):458–634. 39. Brock B, Schillo BA, Moilanen M. Tobacco industry marketing: an analysis of direct mail coupons and giveaways. Tob Control. 2015;24:505–8. 40. Chaloupka FJ, Warner KE. The economics of smoking. Handbook Health Econ. 2000;1:1539–1627. 41. Shang C, Chaloupka FJ, Fong G, Thompson M, O’Connor RJ. The association between tax structure and cigarette price variability: findings from the ITC Project. Tob Control. 2015;24:iii88–iii93 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/24/Suppl_3/iii88.full.pdf, accessed 24 January 2021) 42. Li Q, Hyland A, Fong GT, Jiang Y, Elton-Marshall T. Use of less expensive cigarettes in six cities in China: findings from the International Tobacco Control (ITC) China Survey. Tob Control. 2010;19(suppl CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 83 2):i63–i68 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/19/Suppl_2/i63.full.pdf, accessed 24 January 2021). 43. Hyland A, Laux FL, Higbee C, Hastings G, Ross H, Chaloupka FJ, et al. Cigarette purchase patterns in four countries and the relationship with cessation: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15(suppl III):iii59–iii64 (https://tobaccocontrol.bmj.com/ content/tobaccocontrol/15/suppl_3/iii59.full.pdf, accessed 24 January 2021). 44. Tsai YW, Yang CL, Chen CS, Liu, TC, Chen PF. The effect of Taiwan’s tax induced increases in cigarette prices and brand switching and the consumption of cigarettes. Health Econ. 2005;14:627–41. 45. Cummings KM, Hyland A, Lewit E, Shopland D. Use of discount cigarettes by smokers in 20 communities in the United States, 1988-1993. Tob Control. 1997;6(suppl 2):S25–S30 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/6/suppl_2/S25.full.pdf, accessed 24 January 2021). 46. Ross H, Tesche J. Undermining government tax policies: common strategies employed by the tobacco industry in response to increases in tobacco taxes. Economics of Tobacco Control Project, School of Economics, University of Cape Town and Tobacconomics, Health Policy Center, Institute for Health Research and Policy. Chicago: University of Illinois at Chicago; 2016 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC5747349/, accessed 26 January 2021). 47. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 48. Hiscock R, Branston JR, McNeill A, Hitchman SC, Partos TR, Gilmore AB. Tobacco industry strategies undermine government tax policy: evidence from commercial data. Tob Control. 2017;0:1–10 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 49. Filippidis F, Laverty A, Hone T, Been JV, Millett C. Association of cigarette price differentials with infant mortality in 23 European Union countries. JAMA Pediatr. 2017;171(11):1100–1106 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5710372/#:~:text=Cigarette%20price%20increases%20across%20 23,priced%20cigarettes%20during%20this%20period, accessed 26 January 2021). 50. López-Nicolás A, Belén Cobacho M, Fernández E. The Spanish tobacco tax loopholes and their consequences. Tob Control. 2013;22(E1):e21. 51. Ballester L, Auchincloss A, Robinson L, Mayne S. Exploring impacts of taxes and hospitality bans on cigarette prices and smoking prevalence using a large dataset of cigarette prices at stores 2001–2011, USA. Int J Environ Res Public Health. 2017;14(3):e318 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5369154/, accessed 24 January 2021). 52. Liber A, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia survey. Tob Control. 2015;24:iii83-iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed on 26 January 2021). 53. Feighery E, Ribisl K, Schleicher N, Zellers L, Wellington N. How do minimum cigarette price laws affect cigarette prices at the retail level? Tob Control. 2005;14:80–85 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/14/2/80.full.pdf, accessed 26 January 2021). 54. Tynan MA, Ribisi KM, Loomis BR. Impact of cigarette minimum price laws on the retail price of cigarettes in the USA. Tob Control. 2012;22:e1. 55. Huang J, Chriqui JF, DeLong H, Mirza M, Diaz MC, Chaloupka FJ. Do state minimum markup/price laws work? Evidence from retail scanner data and TUS-CPS. Tob Control. 2016;25:i52–i59 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i52.full.pdf, accessed 26 Jan 2021). 56. Whitehead R, Brown L, Riches E, Rennick L, Armour G, McAteer J, et al. Strengths and limitations of tobacco taxation and pricing strategies. Edinburgh: NHS Health Scotland; 2018 (http://www. healthscotland.scot/media/1829/rapid-evidence-review-strengths-and-limitations-of-tobacco- taxation-and-pricing-strategies.pdf, accessed 27 January 2021). 57. Golden S, Farrelly M, Luke D, Ribisl K. Comparing projected impacts of cigarette floor price and excise tax policies on socioeconomic disparities in smoking. Tob Control. 2016;25:i60-i66 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i60.full.pdf, accessed 27 January 2021). 58. Doogan NJ, Wewers ME, Berman M. The impact of a federal cigarette minimum pack price policy on cigarette use in the USA. Tob Control. 2018;27(2):203–08. 84 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 59. Marynak K, Xu X, Wang X, Holmes CB, Tynan MA, Pechacek T. Estimating the impact of raising prices and eliminating discounts on cigarette smoking prevalence in the United States. Public Health Rep. 2016;131:536–43 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4937114/#:~:text=We%20found%20 that%20each%20price,smokers%20aged%20%E2%89%A526%20years., accessed 27 January 2021). 60. McLaughlin I, Pearson A, Laird-Metke E, Ribisl K. Reducing tobacco use and access through strengthened minimum price laws. Am J Public Health. 2014;104(10):1844–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4937114/#:~:text=We%20found%20that%20each%20price,smokers%20aged%20 %E2%89%A526%20years., accessed 27 January 2021). 61. European Commission v. French Republic, C-197/08, European Court of Justice 4 March 2019 (https:// eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A62008CC0197, accessed on 27 January 2021). 62. Tobacco price promotion: policy responses to industry price manipulation. Boston: Public Health and Tobacco Policy Center; October 2011 (http://tobaccopolicycenter.org/documents/Price%20 Promotion%20Overview%20FINAL.pdf, accessed 28 September 2020). 63. Golden SD, Holt Smith M, Feighery EC, Roeseler A, Rogers T, Ribisl KM. Beyond excise taxes: a systematic review of literature on non-tax policy approaches to raising tobacco product prices. Tob Control. 2016;25:377–85. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/4/377.full.pdf, accessed 27 January 2021). 64. Branston J, Gilmore S. The case for Ofsmoke: the potential for price cap regulation of tobacco to raise ₤500 million per year in the UK. Tob Control. 2014;23(1):45–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3812939/, accesed 27 January 2021). 65. von Lampe K, Kurti M, Johnson J. “I’m gonna get me a loosie” understanding single cigarette purchases by adult smokers in a disadvantaged section of New York City. Prev Med Rep. 2018;12:182–5. (https:// www.sciencedirect.com/science/article/pii/S2211335518302195, accessed 27 January 2021). 66. de Ojeda A, Barnoya J, Thrasher JF. Availability and costs of single cigarettes in Guatemala. Nicotine Tob Res. 2013;15(1):83–7 . 67. Lal P, Kumar R, Ray S, Sharma N, Bhattarcharya B, Mishra D, et al. The single cigarette economy in India - a back of the envelope survey to estimate its magnitude. Asian Pac J Cancer Prev. 2015;16(13):5579-82 (http://journal.waocp.org/article_31295_9ebd3981652934bbf5d78ec066b6a39d.pdf, accessed 27 January 2021). 68. Coady MH, Chan CA, Sacks R, Mbamalu IG, Kansagra SM. The impact of tax on purchasing behaviors among New York City smokers. Am J Public Health. 2013;103(6):e54-e60 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC3698745/, accessed 27 January 2021). 69. Cigarettes in Thailand, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 70. Cigars in Germany, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 71. Report from the Commission to the Council on Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2018 (COM(2018) 17 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/report_excise_duty_manufactured_ tobacco_12012018_en.pdf, accessed 28 September 2020). 72. Gallet CA, List JA. Cigarette demand: a meta-analysis of elasticities. Health Econ. 2003;12(10):821–35. 73. Guidelines for implementation of article 6 of the WHO FCTC. Geneva: World Health Organization; 2016 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 29 September 2020). 74. DeCicca P, Kenkel D, Liu F. Excise tax avoidance: the case of state cigarette taxes. J Health Econ. 2013;32(6):1130–41 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3855906/, accessed 27 January 2021). 75. Jawad M, Lee JT, Glantz S, Millett C. Price elasticity of demand of non-cigarette tobacco products: a systematic review and meta-analysis. Tob Control; 2018;27(6):689–95 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6056338/, accessed 27 January 2021). 76. John RM. Price elasticity estimates for tobacco products in India. Health Policy Plan. 2008;23(3):200–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2800993/, accessed on 27 January 2021). 77. Wasserman J, Manning WG, Newhouse JP, Winkler JD. The effects of excise taxes and regulations on CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 85 cigarette smoking. Journal of Health Econ, Elsevier. 1991;10(1):43–64. 78. Cheng K, Kenkel D. U.S. cigarette demand: 1944-2004. B.E. Journal of Econ Analysis & Policy. 2010;10(1):10.2202/1935-1682.2438 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3820165/, accessed 28 January 2021). 79. Bobak M, Jha P, Nguyen S, Jarvis M. Poverty and smoking. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000. 80. Peck, RM. Equity issues, tobacco, and the poor. Yurekli A, de Beyer J, editors. Washington (DC): World Bank Group; 2013. (Economics of Tobacco Toolkit, Tool 6; https://openknowledge.worldbank.org/ handle/10986/16265, accessed 28 January 2021). 81. Kenkel DS, Schmeiser MD, Urban C. Is smoking inferior?: evidence from variation in the earned income tax credit. J Hum Res. 2014;49:1094–1120. 82. Ruhm, CJ. Healthy living in hard times. Journal of Health Econ. 2005;24(2):341–63. 83. Ruhm, CJ. Are recessions good for health? The Quarterly Journal of Econ. 2000;115(2):617-50. 84. Tarantilis F, Athanasakis K, Zavras D, Vozikis A, Kyriopoulos I. Estimates of price and income elasticity in Greece. Greek debt crisis transforming cigarettes into a luxury good: an econometric approach. BMJ Open. 2015;5(1):e004748. (https://bmjopen.bmj.com/content/bmjopen/5/1/e004748.full.pdf, accessed 28 January 2021). 85. Kaiser M, Reutter M, Sousa-Poza A, Strohmayer K. Smoking and the business cycle: evidence from Germany. Discussion Paper Series. Bonn, Germany: IZA Institute of Labour Economics; 2017 (http:// ftp.iza.org/dp10953.pdf, accessed 28 September 2020). 86. Tingum EN, Mukong AK, Mdege N. The effects of price and non-price policies on cigarette consumption in South Africa. Tob Induc Dis. 2020;18(July):62 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC7398599/, accessed 28 January 2021). 87. Vardavas CI, Filippidis F, Agak, I, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the importance of earmarking revenue to health care and tobacco control. Tob Induc Dis. 2012;10(1):21 (http://www. tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the-revenue-to- health-care-and-tobacco,66110,0,2.html, accessed 28 January 2021). 88. Marquez P, Catinkaya V. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 89. Blecher EH, van Walbeek CP. An international analysis of cigarette affordability. Tob Control. 2004;13:339– 46 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/13/4/339.full.pdf, accessed 28 January 2021). 90. Goodchild M, Sinha P, Munish VG, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/handle/10665/331836, accessed 28 January 2021). 91. John RM, Dauchy E. Trends in affordability of tobacco products before and after the transition to GST in India. Tob Control Online First. 20 March 2020. 92. Chaloupka FJ, Drope J, Siu E, Vulovic V, Stoklosa M, Mirza M, et al. Tobacconomics cigarette tax scorecard. Chicago: Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/636/uic-tobacco-scorecard-report-final.pdf, accessed 28 January 2021). 93. Kanbur R, Keen MJ. Jeux sans frontieres: tax competition and tax coordination when countries differ in size. Am Econ Rev. 1993;83(4):877–92 (https://www.researchgate.net/publication/4901220_Jeux_ Sans_Frontieres_Tax_Competition_and_Tax_Coordination_When_Countries_Differ_in_Size, accessed 28 January 2021). 94. Bouw A. Tobacco taxation in the European Union: an overview (English). Washington (DC): World Bank Group; 2017 (http://documents.worldbank.org/curated/en/493581492415549898/Tobacco- taxation-in-the-European-Union-an-overview, accessed on 28 September 2020). 95. Blecher E, Drope J. The rewards, risks and challenges of regional tobacco tax harmonization. Tob Control. 2014;0:1–5. 96. Directive No.n° 01/2017/CM/UEMOA of 22 December 2017. Paris: Droit-Afrique; 2017 (https://www. 86 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N droit-afrique.com/uploads/UEMOA-Directive-2017-01-droits-accise-tabacs.pdf, accessed 29 September 2020). 97. Mirzoev T, Zhu L, YangY, Pescatori A, Matsumoto A, Callen T. The future of oil and fiscal sustainability in the GCC region. Middle East and Central Asia Research Department. Washington (DC): International Monetary Fund; 2020 (https://www.imf.org/-/media/Files/Publications/DP/2020/English/FOFSGCCEA. ashx, accessed 28 January 2021). 98. State cigarette excise tax rates & rankings. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/factsheets/0097.pdf, accessed 28 September 2020). 99. A map and table comparing cigarette prices in Canada (March 2018). Toronto: Non-Smokers’ Rights Association/Smoking and Health Action Foundation; 2018 (https://nsra-adnf.ca/wp-content/ uploads/2018/03/180302_map_and_table.pdf, accessed 4 February 2021). 100. European Commission excise duty rates on manufactured tobacco. Brussels: European Commission; 31 December 2017 (https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco- energy/excise-duties-tobacco_en, accessed 1 July 2018). 101. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2019 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study-on-the-tobacco-taxation-directive-2019_en.pdf, accessed 28 September 2020). 102. Liber AC. Heated tobacco products and combusted cigarettes: comparing global prices and taxes. Tob Control; 2018;28(6). 103. Heated tobacco products: information sheet - 2nd edition. Geneva: World Health Organization; 2020 (https://www.who.int/publications/i/item/WHO-HEP-HPR-2020.2, accessed 28 January 2021). 104. Kass Lempert L, Glantz S. Analysis of FDA’s IQOS marketing authorisation and its policy impacts. Tob Control 2020;0:1–9. 105. Chivers E, Janka M, Franklin P, Mullins M, Larcombe A. Nicotine and other potentially harmful compounds in “nicotine-free” e-cigarette liquids in Australia. Med J Aust. 2019;210(3):118–9 (https://onlinelibrary. wiley.com/doi/epdf/10.5694/mja2.12059, accessed 28 September 2020). 106. Wenzl T, Zelinkova Z, Regueiro J, Giri A. Scientific-technical support activities to DG TAXUD-C-2 on the option to include e-cigarettes within the scope of excisable goods for the Impact Assessment on a possible revision of Directive 2011/64/EU. Administrative Arrangement TAXUD/2015/DE/329. Luxembourg: Publications Office of the European Union; 2017 (https://publications.jrc.ec.europa.eu/ repository/bitstream/JRC105828/kjna29062enn.pdf, accessed 28 September 2020). 107. Lisko JG, Tran H, Stanfill SB, Blount BC, Watson CH. Chemical composition and evaluation of nicotine, tobacco alkaloids, pH, and selected flavors in e-cigarette cartridges and refill solutions. Nicotine Tob Res. 2015;17(10):1270–78. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573955/, accessed 28 January 2021). 108. Ghebreyesus TA. Progress in beating the tobacco epidemic. Lancet. 17 August 2019;394:548–49 (https:// www.thelancet.com/pb-assets/Lancet/pdfs/S0140673619317301.pdf?hss_channel=tw-27013292, accessed 29 September 2020). 109. Smoking cessation, a report of the Surgeon General. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2020. (https://www.hhs.gov/sites/default/ files/2020-cessation-sgr-executive-summary.pdf, accessed 8 October 2020). 110. E-cigarettes are harmful to health. Geneva: World Health Organization. 5 February 2020 (https://www. who.int/news-room/detail/05-02-2020-e-cigarettes-are-harmful-to-health, accessed 29 September 2020). 111. Shan L, Manzione LC, Azagba S. Psychological well-being and dual-use of cigarettes and e-cigarettes among high school students in Canada. J Affect Disord. 2020;265:357–63 112. King BA, Gammon DG, Marynak KL, Rogers T. Electronic cigarettes sales in the United States, 2013- 2017. JAMA. 2018;320(13):1379-80 (https://jamanetwork.com/journals/jama/fullarticle/2705175, accessed 28 January 2021). 113. LaVito A. Popular e-cigarette Juul’s sales have surged almost 800 percent over the past year [website]. CNBC. 2 July 2018 (https://www.cnbc.com/2018/07/02/juul-e-cigarette-sales-have-surged-over-the- past-year.html, accessed 28 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 87 114. Statement from FDA Commissioner Scott Gottlieb, M.D., on new enforcement actions and a youth tobacco prevention plan to stop youth use of, and access to, JUUL and other e-cigarettes. Washington (DC): US Food and Drug Administration; 23 April 2018 (www.fda.gov/NewsEvents/Newsroom/ PressAnnouncements/ucm605432.htm, accessed 29 September 2020). 115. Paul K. San Francisco becomes first US city to ban sale of e-cigarettes. The Guardian. 25 June 2019 (https://www.theguardian.com/us-news/2019/jun/25/san-francisco-e-cigarettes-vaping-juul-ban, accessed 29 September 2020). 116. Pesko MF, Courtemanche CJ, Maclean JC. The effects of traditional cigarette and e-cigarette tax rates on adult tobacco product use. J Risk Uncertain. 2020;60:229–258. 117. Cotti CD, Courtemanche CJ, Maclean JC, Nesson ET, Pesko MJ, Tefft N. The effects of e-cigarettes taxes on e-cigarettes prices and tobacco product sales: evidence from retail panel data. Cambridge: National Bureau of Economic Research; 2020 (NBER Working Paper 26724; https://www.nber.org/ system/files/working_papers/w26724/w26724.pdf, accessed 28 January 2021). 118. Saffer H, Dench DL, Grossman M, Dave DM. E-cigarettes and adult smoking: evidence from Minnesota. Cambridge: National Bureau of Economic Research; 2019 (NBER Working Paper 26589; https://www. nber.org/system/files/working_papers/w26589/w26589.pdf, accessed 28 January 2021). 119. Huang J, Gwarnicki C, Xu X, Caraballo RS, Wada R, Chaloupka FJ. A comprehensive examination of own- and cross-price elasticities of tobacco and nicotine replacement products in the U.S. Prev Med. 2018;117:107–14 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6195827/, accessed 28 September 2020). 120. Zheng Y, Zhen C, Dench D, Nonnemaker JM. U.S. demand for tobacco products in a system framework. Health Econ. 2017;26:1067–86. 121. Stoklosa M, Drope JM, Chaloupka FJ. Prices and e-cigarette demand: evidence from the European Union. Nicotine Tob Res. 2016;18(10):1973–80. 122. Grace RC, Kivell BM, Laugesen M. Estimating cross-price elasticity of e-cigarettes using a simulated demand procedure. Nicotine Tob Res. 2015;17(5):592–8. (https://www.researchgate.net/ publication/270291421_Estimating_Cross-Price_Elasticity_of_E-Cigarettes_Using_a_Simulated_ Demand_Procedure, accessed 28 January 2021). 123. Huang J, Tauras J, Chaloupka FJ. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob Control. 2014;23:iii41–iii47 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/23/suppl_3/iii41.full.pdf, accessed 17 February 2021). 124. Cheng KW, Chaloupka FJ, Shang C, Ngo A, Fong GF, Borland R, et al. Prices, use restrictions and electronic cigarette use—evidence from wave 1 (2016) US data of the ITC Four Country Smoking and Vaping Survey. Addiction. 2019;114(suppl 1):115–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC6661221/, accessed 28 January 2021). 125. Liber AC, Drope JM, Stoklosa M. Combustible cigarettes cost less to use than e-cigarettes: global evidence and tax policy implications. Tob Control. 2016;0:1–6 (https://www.researchgate.net/ publication/299473181_Combustible_cigarettes_cost_less_to_use_than_e-cigarettes_Global_ evidence_and_tax_policy_implications, accessed 28 January 2021). 126. Chaloupka FJ, Sweanor D, Warner KE. Differential taxes for differential risks — toward reduced harm from nicotine yielding products. N Engl Journal Med. 2016;373(7):594-7. 127. Goniewicz ML, Knysak J, Gawron M, Kosmider L, Sobczak A, Kurek J, et al. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob Control. 2014;23:133–9. 128. Sweet L, Brasky TM, Cooper S, Doogan N, Hinton A, Klein EG, et al. Quitting behaviors among dual cigarette and e-cigarette users and cigarette smokers enrolled in the tobacco user adult cohort. Nicotine Tob Res. 2019;21(3):278–84 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6379027/, accessed 28 September 2020). 129. Wills TA, Knight R, Williams RJ, Pagano I, Sargent JD. Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics. 2014;135(1):e43–e51 (https://www. ncbi.nlm.nih.gov/pmc/articles/PMC4279062/, accessed 28 January 2021). 130. Electronic Nicotine Delivery Systems: report by WHO. Geneva: World Health Organization; 2014. (FCTC/ COP/6/10 Rev.1; https://apps.who.int/gb/fctc/PDF/cop6/FCTC_COP6_10Rev1-en.pdf, accessed 28 January 2021) 88 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 131. WHO report on the global tobacco epidemic 2017: monitoring tobacco use and prevention policies. Geneva: World Health Organization; 2017 (https://www.who.int/tobacco/global_report/2017/en/, accessed 28 January 2021). 132. Regulation of the Minister of Finance of the Republic of Indonesia, number 146, PMK.010/2017 on excise tax rates of tobacco products. Jakarta: Ministry of Finance Indonesia; 2017. 133. Gourdet CK, Chriqui JF, Chaloupka FJ. A baseline understanding of state laws governing e-cigarettes. Tob Control. 2014;23:iii37-iii40 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4078672/, accessed 28 January 2021). 134. An act to ensure that nicotine products are taxed equally. 127th Maine Legislature: first regular session-2015. H.P. 670, Legislative Document No. 973. 2015 (https://www.mainelegislature.org/legis/ bills/getPDF.asp?paper=HP0670&item=1&snum=127&PID=, accessed 28 January 2021). 135. Evaluation of the Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco. Commission Staff Working Document. Brussels: European Commission; 2020 (SWD(2020) 32 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/10- 02-2020-tobacco-taxation-report.pdf, accessed 28 September 2020). 136. Electronic nicotine delivery systems and electronic non-nicotine delivery systems (ENDS/ENNDS): report by WHO. Geneva: World Health Organization; 2016. (FCTC/COP/7/11; https://www.who.int/ fctc/cop/cop7/FCTC_COP_7_11_EN.pdf, accessed 28 September 2020). 137. E-cigarette use among youth and young adults: a report of the Surgeon General. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2016 (https://e-cigarettes. surgeongeneral.gov/documents/2016_SGR_Full_Report_non-508.pdf, accessed 28 September 2020). 138. Wagener TL, Floyd EL, Stepanov I, Driskill LM, Frank SG, Meier E, et al. Have combustible cigarettes met their match? The nicotine delivery profiles and harmful constituent exposures of second-generation and third-generation electronic cigarette user. Tob Control 2017;26:e23–e28 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/26/e1/e23.full.pdf, accessed 28 September 2020). 139. Vapour products data. In: Euromonitor International [website]. London: Euromonitor International; 2019 (restricted access). 140. Global market data: segment insights. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 89 ANNEX 2.1 Table A2.1 Countries that apply different types of cigarette excise tax structures, 2018 SPECIFIC EXCISE (65 COUNTRIES) AD VALOREM EXCISE (42 COUNTRIES) MIXED EXCISE (63 COUNTRIES) NO EXCISE (15 COUNTRIES) Albania, Andorra, Australia, Azerbaijan, Barbados, Belarus, Belize, Bolivia (Plurinational State of ), Burundi, Canada, Cook Islands, Dominica, Ecuador, Eswatini, Fiji, Gambia, Honduras, Iceland, India, Indonesia, Jamaica, Japan, Jordan, Kazakhstan, Kenya, Kiribati, Kyrgyzstan, Lesotho, Malaysia, Mauritius, Mongolia, Mozambique, Myanmar, Namibia, Nepal, New Zealand, Nicaragua, Norway, Pakistan, Palau, Papua New Guinea, Peru, Philippines, Republic of Korea, Saint Lucia, Saint Vincent and the Grenadines, Samoa, Seychelles, Singapore, Solomon Islands, South Africa, Sri Lanka, Suriname, Tajikistan, Timor-Leste, Tonga, Trinidad and Tobago, Uganda, United Republic of Tanzania, USA, Uruguay, Uzbekistan, Vanuatu, Yemen, Zimbabwe Argentina, Armenia, Bahrain, Bangladesh, Benin, Burkina Faso, Cabo Verde, Cambodia, Cameroon, Chad, Comoros, Côte d’Ivoire, Cuba, Democratic Republic of the Congo, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Ghana, Grenada, Guatemala, Guinea-Bissau, Liberia, Madagascar, Mali, Mauritania, Niger, Panama, Paraguay, Saint Kitts and Nevis, Saudi Arabia, Senegal, Sierra Leone, Sudan, Syrian Arab Republic, Togo, Turkmenistan, Tuvalu, United Arab Emirates, Venezuela (Bolivarian Republic of ), Viet Nam, Zambia Algeria, Austria, Belgium, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Central African Republic, Chile, China, Colombia, Congo, Costa Rica, Croatia, Cyprus, Czechia, Denmark, Dominican Republic, Egypt, El Salvador, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iran (Islamic Republic of ), Ireland, Israel, Italy, Lao People’s Democratic Republic, Latvia, Lebanon, Lithuania, Luxembourg, Malta, Mexico, Montenegro, Morocco, Netherlands, Nigeria, North Macedonia, Poland, Portugal, Republic of Moldova, Romania, Russian Federation, Rwanda, Sao Tome and Principe, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Thailand, Tunisia, Turkey, Ukraine, United Kingdom, West Bank and Gaza Strip Afghanistan, Angola, Antigua and Barbuda, Democratic People’s Republic of Korea, Iraq, Kuwait, Libya, Maldives, Marshall Islands, Micronesia (Federated States of ), Nauru, Niue, Oman,a Qatar,a Somalia a This table shows the status of cigarette excise tax structures as of July 2018 and does not account for changes occurring after that date, in particular for the cases of Qatar and Oman, which introduced excise on tobacco in January 2019 and June 2019, respectively. Source: WHO RGTE. 90 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 2.2 THE ANALYTICS OF THE TAX BASE ELASTICITY Assume tax revenue R = tsQ or R = tvPQ , where Q is the quantity consumed, ts is the specific tax, tv is the ad valorem tax and P is the consumer price. The following equations can help to illustrate the different components of the tax base elasticity. Under a specific excise regime, change in revenue depends essentially on the change in consumption: where R is the tobacco tax revenue, is the specific excise tax and is the tobacco tax base elasticity. The tax base elasticity is made of: where ε, the price elasticity = , is the degree of pass-through of the specific excise tax rate increase on consumer price and is the tax-price ratio. Under an ad valorem excise regime, change in revenue depends essentially on the change in tobacco expenditure: where R is the tobacco tax revenue, tav is ad valorem excise tax and ηav is the tobacco tax base elasticity. The tax base elasticity here is made of: where is the degree of pass-through of the ad valorem excise tax rate increase on consumer price, is the tax-price ratio and ε the price elasticity = . ∂R = Q (1+ηs)∂ ts ηs = ε ts P ∂P ∂ts tav P ηav = (1 + ε) tav P ∂P ∂tav∂P ∂tav ∂P ∂ts ΔQ ΔP P Q ΔQ ΔP P Q ts P ∂R = P Q (1+ηav)∂tav CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 91 ANNEX 2.3 ELEMENTS OF THE DEVICES THAT MAKE UP ENDS/ENNDS PRODUCTS The main components of any ENDS/ENNDS kit include essentially: • USB charger (not a car charger) • Inbuilt battery Additionally, For open systems • Tanks (refillable containers) with removable atomizer (often sold bundled with atomizers) • Clearomizers/refillable pods (no removable atomizer) • E-liquid For closed systems • Disposable e-cigarettes: not rechargeable, thrown away after e-liquid is finished • Nondisposable e-cigarettes: – Pre-filled cartomizers (cartridges designed to go with the cigalike kit) – Pre-filled tank refills/pods (pods or cartridges designed to go with the prefilled tank/pod kits) Some definitions: • Atomizer: uses a heating element to vaporize the e-liquid • Cartomizer: combines the cartridge/tank and the atomizer • Clearomizer: same as cartomizer, uses different technology • Cartridge/tank/pod: container that includes the e-liquid In summary, ENDS/ENNDS product devices include the following: • USB charger (not a car charger) • Inbuilt battery • Disposable e-cigarettes • Atomizer • Cartomizer/clearomizer • Cartridge/tank/pod with or without atomizer • Pre-filled cartridge/tank/pod (for closed systems, includes e-liquid) • E-liquid (added in the cartridge/tank/pod in open systems) Source: ECigIntelligence, 2020. Information also obtained from vaping websites, including https://www. misthub.com/blogs/vape-tutorials/76788357-tutorial-atomizer-vs-cartomizer-vs-clearomizer, http:// www.bestclearomizer.com/clearomizer-vs-cartomizer-vs-atomizer/, https://wayofleaf.com/accessories/ vapes/atomizer-vs-clearomizer-vs-cartomizer, https://wayofleaf.com/accessories/vapes/atomizer-vs- clearomizer-vs-cartomizer, accessed 15 July 2020. 92 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 93 CHAPTER 3. Tobacco tax administration 3.1 INTRODUCTION Imposing excise taxes on tobacco products usually serves more than one purpose. Governments often find themselves balancing interests between financial and public health objectives. Both objectives can best be achieved by an efficient and effective competent authority with strong technical capacity to enforce and collect taxes. A competent authority is the agency, organization or department that is legally as- signed to complete a particular activity; in the case of administering tobacco taxes, the competent authority is often a tax administration, revenue authority, customs department or ministry of finance. Article 6 of the WHO FCTC (1), along with its guidelines (2), provides a solid foundation for sound tax administration. As stated under section 1.5 in the guidelines, tobacco tax systems should be efficient and effective. They should be structured to minimize the costs of compliance and administration, while ensuring that the desired level of tax revenue is raised and health objec- tives are achieved. Efficient and effective administration of tobacco tax systems enhances tax compliance and collection of tax revenue while reducing tax evasion and the risk of illicit trade. Efficiency in tax administration refers to minimizing the costs per unit of tax revenue collected. It is measured by comparing the resources used with the revenues gener- ated. Effectiveness in tax administration refers to a high level of compliance – also described as taxpayers meeting their obligations. Thus, an efficient and effective competent authority collects the tax at a minimum cost while ensuring conformity to the rules. Tobacco taxation is the single most effective tobacco control measure for re- ducing tobacco use and is best implemented as part of a comprehensive tobacco control plan (3). Illicit trade – including smuggling and illicit manufacturing – and tax avoidance undermine the effectiveness of tax policies and their objectives (4). The impact on illicit trade is often cited by opponents of tax increases, who argue that increasing taxes increases illicit trade. They contend that illicit trade can lead 94 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N to lower revenues for governments and lower prices. The challenge faced by the competent authorities is to ensure that due taxes are declared and collected on all tobacco products that are manufactured in and/or imported into its jurisdiction, while at the same time detecting tobacco products that are illegally manufactured in and/or imported into its jurisdiction, stopping such activity and prosecuting the responsible parties. This chapter describes the shared characteristics of good tax administrations, including best practices based on country experiences. It regularly refers to the WHO FCTC, and – given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade – draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (5). Any practice or conduct prohibited by law and related to production, shipment, receipt, possession, distribution, sale or purchase of tobacco products – including any practice or conduct intended to facilitate such activity – is considered as illicit trade (Article 1). The objective of the Protocol is to eliminate and prevent all forms of illicit trade in tobacco products. At the same time, the Protocol includes measures for tobacco tax administration based on international best practices, which makes it relevant for all countries, even those that are not Parties to it. The Protocol was adopted at the fifth session of the COP to the WHO FCTC in 2012 and entered into force on 25 September 2018. As indicated in the Preamble, it was developed in response to the increasing international illicit trade in tobacco products (5). The Protocol covers three main areas: (1) measures to control the supply chain (Part III); (2) measures dealing with offences, including sanctions (Part IV) and (3) international cooperation (Part V). Different provisions of the Protocol are discussed in detail throughout this chapter, and section 3.4 is devoted specifically to control and enforcement. 3.2 INSTITUTIONAL ARRANGEMENTS Competent authorities that collect taxes effectively in an efficient way share a number of attributes. The organizational structures of these authorities contain clearly defined roles, responsibilities and rules for coordination among relevant bodies. Moreover, competent authorities collect data regularly and manage information needed for assessing risks. The key to successful risk management is to share this information among relevant authorities both within a country and between countries. Effective and efficient competent authorities also regularly evaluate their performance and accountability according to key performance indicators to identify areas for improve- ment. These characteristics are discussed in greater detail in the following sections. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 95 3.2.1. CLEARLY DEFINED ROLES AND RESPONSIBILITIES OF COMPETENT AUTHORITIES The designation of competent authorities for the implementation and enforcement of tax laws – including clear definitions of the boundaries of authority among numerous agencies within a country – is essential for efficient collection of taxes. Areas where different agencies need to cooperate and share data must also be defined. Overlap of activities by different authorities leads to inefficient use of resources, whereas gaps create opportunities for fraud, leading to ineffective tax laws. The importance of clearly defined roles and responsibilities applies not only to tax authorities and customs but also to law enforcement agencies, including police and border control forces. The implementation and enforcement of taxation is organized differently in various countries. The most common structure separates customs and tax administration. The trend since the 1990s, however, has been to combine these functions into one agency, such as Her Majesty’s Revenue and Customs (HMRC) in the United King- dom, SUNAT in Peru1 and AFIP in Argentina.2 Several countries have increased coordination between tax and customs by creating a revenue secretariat and also implementing systems to share tax records as a single taxpayer account. Coordina- tion between tax policy and tax administration authorities has also increased. One can think of combining both into one department within the ministry of finance or ensure that tax administration authorities are consulted during the tax policy process. Some tasks, such as licensing, may be handled by other ministries such as the ministries of health, agriculture or trade. For example, the Ministry of Health of Brunei and the Health Science Authority of Singapore are responsible for the licensing of importers of tobacco products (6). In some federal countries, including Colombia and the United States, excise taxes – including tobacco taxes – are collected and enforced by local or state tax administrations. Other countries have organized the administration of national taxes by establishing a single unified revenue body. Particularly in larger economies, that body is often responsible for both direct and indirect taxes, including excise taxes, and reports to the ministry of finance. All the functions needed for effective and efficient tax administration are established within these bodies (7). No matter what the institutional arrangements may be, it is vital that the agencies cooperate and exchange information and that their competencies find their basis in law. More information on this topic is provided in section 3.2.2. 1 Law Decreto Supremo 061-2002-PCM - Disponen fusión por absorción de la Superintendencia Na- cional de Administración Tributaria – SUNAT con la Superintendencia Nacional de Aduanas - Aduanas [Supreme decree year 2002 about the merger between Tax and Customs Administration]. Lima: El Peruano, 12, July 2002 (in Spanish) (http://www.sunat.gob.pe/legislacion/sunat/ds061-2002-PCM.pdf, accessed 13 November 2020). 2 Administracion Federal de Ingresos Publicos, Decreto 618/1997 [Federal Administration of Public Revenue, Decree 618] (in Spanish) (http://servicios.infoleg.gob.ar/infolegInternet/an- exos/40000-44999/44432/norma.htm, accessed 13 November 2020). 96 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Many countries, however, have separate bodies for the collection of taxes and customs duties. A 2015 survey of 135 tax administrations worldwide found that only 36% of them were responsible for both tax administration and customs ad- ministration (8). In most countries, customs authorities are more likely to collect excise duties on imports, and in many countries, VAT or sales tax is collected jointly with tobacco tax, particularly for imported products. This simplifies controls and creates synergy by unifying common processes and procedures, resulting in cost savings for tax administrations and taxpayers. The involvement of multiple bodies in tax collection requires especially good collaboration and information-sharing to ensure efficient and effective collection of taxes and duties. KEY TAKEAWAY 1 Institutional arrangements with clearly defined roles and responsibilities – designed to prevent overlaps and voids – contribute to effective and efficient tax administration. 3.2.2. EFFECTIVE COORDINATION AMONG RELEVANT BODIES Coordination at the national level Coordination among relevant bodies is key to effective tobacco tax administration. This means not only clearly defined roles and responsibilities, as described in the previous section, but also coordination among the competent authority, customs and those responsible for formulating, analysing and implementing tax policy. Regardless of the institutional arrangements – whether the responsible parties are all within the ministry of finance or in separate government agencies – all parties need to cooperate and exchange information to optimize tax collection and enforcement of tax policy. In practice, this means that information should be shared among, for example, customs, local government units that issue licences and health authori- ties – particularly those that regulate the sale of tobacco products. For tax authorities, the most relevant information concerning excise taxes in- cludes the identity of taxpayers and those involved in the trade of tobacco (import and export data, licences, criminal records, tax returns, bank statements, etc.); the category, quantity, value and location of manufactured goods; and the movement of those goods until all taxes are paid. Legal impediments to obtaining this informa- tion – such as bank secrecy or privacy regulations – should be kept in mind, and where needed, exceptions for fiscal procedures should be incorporated into law. Seizure data are also a valuable source of information; more details on this are provided in section 3.4. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 97 Tax authorities should regularly coordinate with law enforcement agencies – such as the police and border control forces, depending on a country’s laws – to properly monitor tobacco-related activities and enforce the tax laws. Often, the competent tax authority and customs authorities work in close cooperation with anti-fraud teams.3 Coordination and sharing of information can be required in legislation or regulations to ensure a streamlined process and avoid confusion. This can be done on an ad hoc basis as needed or with formal planned exchanges of information and regular meetings. It is recommended that at least a legal basis for exchange or access to information among government bodies be established to prevent claims during legal procedures that evidence was obtained unlawfully. Some countries go beyond exchanging information and cooperation. In the Neth- erlands, for example, customs authorities not only carry out work for the Ministry of Finance, they also carry out non-fiscal tasks for seven other departments, including the Ministry of Agriculture, Nature and Food Quality; the Ministry of Justice and Security; and the Ministry of Foreign Affairs (9). These activities are often based on bilateral agreements between the Ministry of Finance and the other departments. In other countries, such as the United States and Canada, Customs and Border Protection are not part of the Ministry of Finance; they are part of the Department of Homeland Security in the United States and the Ministry of Public Safety and Emer- gency Preparedness in Canada. These agencies also carry out many non-fiscal tasks. Along with the implementation of new tobacco control and tax laws, several countries have also created high-level committees to ensure good coordination and implementation of the laws. Led by health and finance ministries, committees ensure coordination and fine-tuning to achieve desired results. Botswana, Chile, Colombia, Indonesia and Senegal, among other countries, have successfully started with coordina- tion, planning and monitoring of tobacco laws’ implementation through periodic com- mittee meetings. The committees usually include representatives from the ministries of health, finance, tax and customs, police, transport and, in some cases, education. Coordination across borders Effective approaches to control smuggling in tobacco products require interventions at the borders of jurisdictions and therefore must involve the border agencies. However, with the globalization of trade, there is a need for close coordination not only between tax and border control authorities but also between different jurisdictions. Recent cases have demonstrated that an absence of formal cooperation frameworks may expose a market to financial crime, including money-laundering and financing of terrorism (10). 3 See, for example, Focus on tax fraud. Customs administration of the Netherlands, tax and customs administration. 2017;2 (https://customsnl-insight.nl/article/309563676, accessed 3 October 2020). 98 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Accession to international cooperation agreements such as the Protocol, the Organisation for Economic Co-operation and Development (OECD) Multilateral Convention on Mutual Administrative Assistance in Tax Matters and other regional arrangements will contribute greatly to the effective exchange of information and cooperation among enforcement agencies. An effective exchange of market data and information from participating jurisdictions can prevent potential cross-border crimes and loss of domestic revenue. International cooperation reinforces domestic measures to stop illicit trade and raise much-needed revenues. Parties to the Protocol have a commitment to cooperate with one another and to share information to meet their obligations under the Protocol (Article 20). The Protocol itself is the legal instrument that allows Parties to cooperate and share information across borders. Authorities of governments that are not Parties to the Protocol or another coopera- tion agreement that represents a legal instrument to exchange information could conclude a mutual assistance agreement or exchange of information agreement to guide the procedures under which information exchange can take place effectively. The Revised Kyoto Convention of 2010 promulgated by the World Customs Organization (WCO) recommends that jurisdictions that enter into bilateral agree- ments require the other jurisdiction to provide pre-arrival information on goods bound for their customs territory. A survey of 87 WCO members in 2013 found that the vast majority of customs administrations had the legal authority to share information related to the supply chain of tobacco products with other administrations (11). Some economic blocs have also established harmonized legislation applying to all of their Member States to provide administrative cooperation to efficiently cooperate on tax matters (12). Coordination can include the establishment of a special agency to ensure the safety and proper functioning of external borders, such as the European Border and Coast Guard Agency, also known as Frontex (from the French frontières extérieures, “external borders”). In some of the Frontex-led operations, EU and non-EU countries cooperate together with international organizations to target cross-border crime, including the smuggling of cigarettes and raw tobacco (13). Criminals who engage in illicit trade of tobacco products are usually also en- gaged in related criminal activities such as bribery, money laundering, corruption, obstruction of justice and even financing of terrorist organizations (14). A number of international treaties provide the legal framework for addressing such conduct through mechanisms that tackle illicit trade from a criminal justice perspective, such as the United Nations Convention against Transnational Organized Crime, the United Nations Convention against Corruption and the International Convention for the Suppression of the Financing of Terrorism. Table 3.1 summarizes the types of structures available for such coordination. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 99 Table 3.1 Structures for coordinating mechanisms TYPE OF COORDINATION BASIS INVOLVED ACTORS National coordination Agreements with a basis in law between national agencies Customs authorities, ministries of finance and those responsible for formulating, analysing and implementing tax policy; law enforcement agencies, such as police and border control forces; and anti-fraud teams Agreements between ministries or a basis in law or regulation on the establishment of high- level committees Ministry of health, finance, revenue, justice, transport and sometimes education and enforcement entities such as customs and police Bilateral coordination Bilateral cooperation agreements National governments Regional coordination Regional arrangements such as • Harmonized legislation applying to all Member States of an economic bloc to provide administrative cooperation in taxation to efficiently cooperate on tax matters • Regulation to jointly establish a special agency to ensure the safety and functioning of external borders EU Member States, the European Border and Coast Guard Agency (Frontex), customs, law and border enforcement agencies International coordination International treaties or conventions such as • The Protocol • OECD multilateral Convention on Mutual Administrative Assistance in Tax Matters • United Nations Convention against Transnational Organized Crime • United Nations Convention against Corruption • International Convention for the Suppression of the Financing of Terrorism Parties to international treaties and conventions, law and border enforcement agencies KEY TAKEAWAY 2 Regardless of differing institutional arrangements, coordination and cooperation within a country and across jurisdictions are essential to optimize tax collection and enforcement of tax policy. 100 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 3.2.3. EVALUATION OF PERFORMANCE AND ACCOUNTABILITY Key strategic indicators are useful for assessing the performance of a competent authority. Performance indicators can include measures such as net revenue col- lected, total expenditures compared with budgeted amounts, the ratio of costs to collection, measures of filing and payment compliance and taxpayer satisfaction (15). Several international organizations, including the IMF, the World Bank, the Inter-American Development Bank and OECD have developed tools to evaluate tax and customs with key performance indicators. This section provides information on some of the indicators that are particularly useful for measuring performance related to tobacco taxes, including the cost of collection ratio, tax gap analysis and tax revenue targets. Cost of collection ratio Collection costs vary among countries. The cost of collection ratio is the total ex- penditure as a percentage of the total net taxes collected. This ratio is often used as a measure of efficiency and effectiveness of competent authorities. In Table 3.2, the cost of collection ratio is calculated for country groups by income level, based on an annual IMF survey. The numbers in the table give an indication of resources used and revenues collected for taxes in general. The same definition of cost of collection was used for all countries. The tax revenue excludes VAT and excise taxes on imported products, so it reflects internal taxes only: personal and corporate income taxes, VAT and excise on domestic production. Customs duties are also not included. The results show the differences among countries at various income levels. Other contributing factors include differences between tax systems, economic situations and compliance levels. Table 3.2 Cost of collection ratio in 2015 per 100 units (ratio of average recurrent budget to revenue collecteda) GROUP (SAMPLE SIZE) 2015 Low-income countries (6) 1.3 Lower-middle-income countries (15) 1.6 Upper-middle income countries (18) 0.9 High-income countries (36) 0.9 All (76) 1.1 a Does not include VAT or excise on imports Source: (Reference 8, Appendix Table 12). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 101 As one would expect, given lower levels of automation and resources, the ratio is higher for low- and lower-middle-income countries, greater than 1.0 (more than 1.0 currency unit needed to collect 100 currency units). The ratios for upper-middle income and high-income countries are below 1.0, indicating more efficient and/or effective collection systems. The cost of collection might be less relevant for taxes that are introduced with other than solely financial objectives, such as influencing a change in behaviour. In particular, in the case of excise taxes applied on tobacco products, the cost of tax collection does not reveal the full picture. If excise tax rates are increased substan- tially – or at least increased above inflation and income growth – consumption will be reduced. As a result, health care costs will be reduced due to reduced tobacco- related mortality and morbidity and increased productivity. These savings are not factored into the ratio of cost of collection to revenue, but governments do benefit from these lower expenses overall. Nevertheless, the cost of collection can be used as an indicator of the efficiency of a competent authority. Tax gap analysis Tax gap analysis is another method of determining how effectively taxes on tobacco products are collected. The tax gap is the difference between the tax due and the tax that is collected. For example, the theoretical tax due under an ad valorem tax on the retail price of cigarettes would be the average price of a pack of cigarettes multiplied by the number of packs sold (estimated from household expenditure surveys, for example) multiplied by the tax rate. This outcome can then be compared to the actual revenues collected (16). The effectiveness of tax collection can also be determined by using the macro- economic input-output matrix, measuring the added value of the economic sector – tobacco in this case – and the theoretical VAT due and then comparing the result with the real VAT collection. This methodology is valid for measuring domestic tax evasion (more information on the use of this method to estimate illicit trade is provided in Chapter 4, section 4.1). Tax revenue target The performance of a competent authority can also be evaluated by determining whether the tax revenue target has been met, if mandated, for a given tax period. Although revenue forecasts are often used as targets, caution is advised. Forecast revenues could include assumptions such as economic growth, inflation and amount collected. Forecasting is a good practice, however, and competent authorities should provide input to the government for the forecasting of revenues to improve the quality of the estimates. Competent authorities should monitor the actual collections 102 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N in comparison with the forecasted revenues, but the theoretical base may not be attainable for a variety of reasons. In addition, a revenue target could provide an incentive for some customs and competent authorities to simply aim to reach the target amount, rather than making efforts to collect the maximum amount possible with the available resources. 3.3 THE TAX COMPLIANCE CYCLE For any tax, there are associated compliance, control and enforcement processes. The compliance cycle usually includes registration and licensing, tax declarations, recordkeeping, storage in warehouses, duty suspension, collection of tax and tax refunds. Figure 3.1 illustrates the typical stages of the tax compliance cycle. Fig. 3.1 Tax compliance cycle 3.3.1 REGISTRATION AND LICENSING Along with regulating and ensuring the integrity of those who deal with controlled substances or goods, the main objective of licensing is to regulate the supply chain. Licensing is a powerful tool for obtaining more information and securing the supply chain of tobacco products. Parties to the Protocol are committed to licensing the manufacturing, import and export of tobacco products and manufacturing equipment (Article 6). In addition, Parties are committed to endeavouring to license – as considered appropriate – the persons involved in the growing of tobacco and the retailing, transporting, wholesal- ing, brokering, warehousing and distribution of tobacco products or manufacturing Registration and licensing (renewal) Tax declaration Authorities: audit and control Taxpayers: recordkeeping Audit and control of information provided in tax declaration Payment and collection of tax Refund Licensed activities, for example: import, production, transport, storage, export, etc. 1 2 3 456 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 103 equipment (Article 6). To ensure an effective licensing system, Parties shall monitor and collect, where applicable, any licence fees that may be levied and consider using them in effective administration and enforcement of the licensing system, for public health or for any other related activity in accordance with national law. If feasible, each Party shall require that retailers and tobacco growers – except for traditional growers working on a noncommercial basis – maintain complete and accurate records of all relevant transactions in which they engage, in accordance with its national law (Article 9.4). Article 6.3(b) of the Protocol provides a list of information to be requested from the applicant of the licence, including: • relevant identity information on the applicant • business location of the manufacturing unit or warehouse and production capacity • detailed list of tobacco products and equipment used • description of where the manufacturing equipment will be installed and used • documentation or declaration of any criminal records • information on bank accounts to be used for transactions and payments • description of intended use and intended market of sale of the tobacco products. To make it easier for authorities to collect all the information they need, rules of confidentiality could be exempted in the licensing process. Licences can be general – covering all activities requiring a licence – or issued for each activity separately, such as different licences for manufacturing, importing and retail. A general licence is less burdensome for the licensing authority, whereas licences for each type of activity offer greater control but at the cost of more adminis- tration (17). The cost of implementing the licensing system should be proportionate to the potential impact of the system. Not only should the type of licences be taken into consideration, the process and information needed to obtain a licence should be carefully considered to ensure proportionality. The more stringent the process is – in terms of the information required and the obligations the system imposes on licensees – the more burdensome the regime will be on both businesses and the authorities who must administer and enforce it. The more information is collected, the higher the compliance and administrative burden will be. It is recommended that the added value of the information be balanced with the additional compliance, administrative and/or enforcement burden. The level of stringency should be decided with consideration of factors such as the level of risk of the activity and the availability of enforcement capacity. A more stringent regime might be justifiable for activities that pose a higher risk for the 104 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government in terms of potential loss of tax revenues – such as the import, production and handling of excisable products on which the excise taxes have not yet been paid. Authorities could consider setting licence fees at a high enough level to cover the costs of administering and enforcing the system. For an example of a system that relies mainly on licensing and permissions, see the case study of Australia in Box 3.1. Wholesalers, distributors and retailers of tobacco products could also be required to obtain a licence before they can engage in the trade of those products. This would enable the competent authority to require reports on, for example, transactions relating to the purchase and sale of tobacco products. Moreover, it would allow the authorities to complete the audit trail of the entire supply chain and to obtain data that will help tax and health policy-makers properly and effectively monitor tobacco products. Governments could also require a licence for entities dealing with raw materials or growing tobacco, including farmers. If licensing of tobacco farmers is deemed appropriate and subsequently required in a country, the farmers have to identify and register their farm areas and location to obtain a licence. The benefit of requiring licences for farmers is that the control of the legitimate supply chain is extended to the identification of the source of the raw material for tobacco products. It also makes it more difficult to divert raw tobacco from the licit to the illicit supply chain. Licences are issued by different agencies across the world. In Brazil, for example, the Health Surveillance Agency is responsible for providing licences. Operators need to obtain approval of the layout of manufacturing and warehousing facilities before they can operate. In addition, they must demonstrate how they will comply with other laws and regulations – for example, by showing the design of product packaging, including the pack, carton and master case. The factory location must be identified before manufacturers can obtain a licence. Finally, a licence is required for the importation of machinery to produce tobacco products (18). Licences can be a source of useful information if authorities establish the informa- tion that applicants must supply in order to obtain the licence. Such information could include the quantity, price and how the tobacco harvests are disposed, as well as the identity of the buyers. It is recommended that an effective licensing regime collect information to establish both the identity and characteristics of applicants by requiring criminal records on relevant offences, such as previous noncompliance with tobacco licences or fraud. To avoid loopholes for monitoring raw tobacco, importers of tobacco leaf could also be licensed or at least required to register and report information on quanti- ties, sources and sales. In some countries, this information is already collected by a government agency other than tax authorities, for example, by the ministry of agriculture. Duplication of requirements and reporting should be avoided through CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 105 legislation and coordination among agencies. If licensing of (small-scale) farmers is difficult to implement, subsequent purchasers (first processors) in the supply chain could be licensed and regulated instead. Licensing first processors is often less burdensome to enforce for competent authorities because, in general, there are far fewer first processors than there are growers. For example, in the EU, between 50 and 100 first processors have been identified, compared with 55 000 farmers (19). Countries could also consider requiring registration of persons or entities engaged in the manufacture and import or sale of materials used for the manufacturing of tobacco products, such as cigarette papers, tobacco leaves, additives, adhesives, acetate or any other type of filters used for cigarettes, tipping paper and cellophane or plastic wraps, as well as materials for packing the cigarettes into packs, reams and master cases. In addition, tobacco manufacturers could be required to obtain a licence before they can purchase these materials. The Parties to the Protocol should decide on appropriate measures, depending on research as to whether key inputs that are essential for manufacturing of tobacco products exist and can be identified and subject to effective controls. KEY TAKEAWAY 3 The objective of licensing is to regulate and secure the supply chain. It is a powerful tool for obtaining information for verification, further investigation and audits. Ideally, all persons involved in the growing of tobacco and retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. Licensing requisites Based on case studies and best practices – including experiences from managing bonded warehouses where the value of merchandise or suspended duties or taxes is high – the following kinds of information could be required to obtain a licence, in particular, for producers, warehouses and distributors of tobacco products: • certification of safety of installations, perimeter security for production and storage (may include CCTV [closed-circuit television]4 access for tax administration) • certification of financial solvency • detailed online, real-time inventory of tobacco products and main raw materi- als, accessible by tax administration • electronic accounting systems 4 The term “closed-circuit television” is used generically to describe surveillance camera systems. 106 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • detailed lists of owners and managers • banking and other financial records • periodic electronic reports of transactions for tobacco products • anytime tax administration right of entry for inventories • mandatory electronic tax returns and payments • mandatory prior-to-arrival customs declarations for tobacco products • declarations of compliance with the tax stamp system (if applicable) • for those involved in import or export, authorized economic operator (AEO) certification • proof of compliance with the bond or guarantee regime • agreement to finance reasonable cost of inspections and tracking and tracing. Box 3.1 Case study of licensing in Australia Australia has taken an approach to controlling tobacco taxes that differs from that in many other countries. It has not used fiscal marks or tracking and tracing.5 Instead, it administers tobacco taxes through licensing and permission-based systems aimed at facilitating operations by lower-risk entities while preventing or tightly controlling commerce involving higher-risk entities. The domestic tax agency, the Australian Taxation Office (ATO), is responsible for most of the controls. These controls cover tobacco that is grown or manufactured and imported as finished goods or as leaf for manufacturing in Australia. In fact, the legal tobacco market in Australia consists only of imported finished tobacco products. In 2006, all tobacco-growing licences were cancelled by the ATO because manufacturers switched to cheaper leaf from external suppliers. The last domestic cigarette manufacturers closed in 2015 and 2016, and there has been no legal domestic tobacco growing or manufacture since then. The ATO administers all other functions relating to the import of tobacco and tobacco products, including licensing of bonded warehouses used to store imported products and issuance of permissions to undertake movement of bonded tobacco products between licensed bonded warehouses or to places of export. Importers must apply for a licence for a bonded warehouse to store imported tobac- co. The applicant must meet general criteria such as fitness, recordkeeping and security.6 These criteria are designed to ensure that only low-risk entities are able to enter the ex- cise tax system. Risk levels are also kept at an acceptable level through provisions allow- ing the suspension or cancellation of licences, subject to appeal. Licences are valid for a 5 See sections 3.4.4 and 3.4.5 for detailed discussions on fiscal marks and tracking and tracing. 6 The entity must not have been charged with an offence under the Excise Act or any Commonwealth, State or Territory Act that carries a penalty in excess of US$ 105 000 in the previous 12 months (or convicted in the previous 10 years), has shown a history of compliance with tax law in the previous four years, has had no previous cancellation of a licence, has adequate financial resources and is not in receivership. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 107 three-year period, with automatic renewal for licensees with demonstrated compliance. The permission system relies on post-transaction audits of commercial records. Criteria used to assess risk include the size of the duty liability, the compliance record of both parties and the possibility of diversion into the market. When there is a perceived risk of revenue loss, the application can be denied or a financial security deposit can be required. Exports of tobacco products are also subject to an export declaration process with the Australian Border Force. An approved export declaration is required for the products to be able to leave the country. Following recommendations from a government task force in 2017, the status of tax-suspended, bonded tobacco was eliminated as of 1 July 2019. In addition, an import licensing regime was introduced, and commercial tobacco imports without a licence are banned. Importers are required to identify their duty liabilities at import and make immediate payment; there are no credit terms available. Full payment of duties and taxes to the Australian Border Force are required prior to a release of tobacco products into the country. Sources: (20–21). As mentioned above, certification as an AEO could be requested as part of the licensing process. Most customs authorities are familiar with the concept of AEOs. Created by the WCO, AEO principles were initially focused on security concerns (22). Having a special licensing regime for operators of the tobacco supply chain is recom- mended due to the special nature of the product. For countries that have no system in place, AEO certification could be a starting point for setting up such a regime. An AEO is defined by the WCO SAFE Framework of Standards (22) as a party involved in the international movement of goods – in whatever function – that has been approved by, or on behalf of, a national customs administration as complying with WCO or equivalent supply chain security standards. AEOs include, inter alia, manufacturers, importers, exporters, brokers, carriers, consolidators, intermediaries, ports, airports, terminal operators, integrated operators, warehousers and distributors. For many years – in some cases, even since the 1970s – customs administrations have been increasingly involved in the security of the international trade supply chain. More recently, customs administrations have developed security programmes in a global context. The AEO is part of these programmes, and in 2005, the WCO adopted the SAFE Framework of Standards. Since then, a number of traders have been required to make substantial investments in order to obtain AEO status and must continue to invest to maintain that status. The AEO program is also recognized by the Trade Facilitation Agreement, a multilateral agreement signed by 174 countries (23). 108 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Some regional blocs have further specified the standards for AEOs and provide clear and well-structured information on their websites to guide and encourage operators to apply for AEO status. A good example of this practice is the website of the Revenue Commissioners of the Republic of Ireland, which contains the in- formation shown in Box 3.2. Box 3.2 AEO: Republic of Ireland Tax and Customs What are AEOs? AEO status is a certified standard authorization issued by customs administrations in the European Union (EU). It certifies that an economic operator has met certain standards in relation to: • safety and security • systems to manage commercial records • compliance with customs rules • financial solvency • practical standards of competence or professional qualifications. This is primarily a trade facilitation measure that recognizes reliable operators and encourages best practices in the international supply chain. As an AEO, an operator could benefit from: • recognition worldwide as a safe, secure and compliant business partner in international trade; • lower risk scores in risk analysis systems when profiling; • priority treatment if physical controls are conducted; • mutual recognition of AEO programmes under Joint Customs Cooperation Agreements, which could result in faster movement of goods through third- country borders; • reduced data sets for entry and exit summary declarations (this applies only to AEO safety and security); • easier access to simplified procedures; • reduction or waiver of comprehensive guarantees. The conditions for AEO status apply to all businesses regardless of size. Manufacturers, exporters, freight forwarders, warehouse keepers, clearance agents, carriers and importers may all apply for AEO status. Source: (24). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 109 3.3.2 DATA COLLECTION, DECLARATIONS AND ACCOUNTING The effectiveness of risk analysis depends on the quality and reliability of the available data. This is also the case for risk analysis in relation to tobacco taxes. Obtaining reliable data can be a challenge in many countries, but the use of electronic sys- tems to collect and manage data is increasing in most competent authorities. The introduction of VAT in many countries around the world has greatly improved the availability of data that can be used for tobacco tax analysis, since reporting is done along the supply chain on, for example, the value, quantity of goods and transaction date. Most countries applying excise duties also have a VAT system in place. In addition, more countries are becoming Parties to the Protocol. With the implementation of the Protocol, more data will become available because countries will be obliged to implement, among other measures, licensing systems with report- ing requirements and tracking and tracing systems. More information on tracking and tracing systems is provided in section 3.4. The obligations of the Protocol will also assist in monitoring the stock of tobacco products. Ideally, all entities involved in the tobacco product supply and distribution chains should be licensed and required to record every transaction that occurs. As this might be burdensome for both tax authorities and taxpayers, the use of automated and electronic systems is recommended in order to decrease the costs of compliance. An accurate inventory system for all raw materials, machinery, goods in process and finished products can be required. It is even more important to have good recordkeeping of the required data. As the volume of reported data increases, a good information technology (IT) system will be needed. The use of IT for periodic tax declarations, accounting, inventory and financial data is critical for obtaining accurate information and decreasing costs for the entire reporting system. Most countries now have some level of automation that can facilitate data analysis. An emerging trend is the use electronic invoices, issued by traders, as part of online real-time information for tax administration. Countries generally start by using electronic invoices at public utility companies and then later expand the use to large companies. Electronic invoices minimize the use of paper, contribute to automated recordkeeping and give accurate and timely information about transactions for tax administration. Several countries began using electronic invoices for companies on a voluntary basis and later made their use mandatory, especially for large companies with a high number of transactions – including the tobacco industry. Electronic invoices have been implemented successfully in EU countries and almost all Latin American countries, as well as several Asian countries.7 7 Electronic Invoicing in Latin America: English Summary of the Spanish Document; Inter-American Development Bank, Inter-American Center of Tax Administrations, 2018 (https://publications.iadb.org/ publications/english/document/Electronic-Invoicing-in-Latin-America.pdf ). 110 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N To verify that information is accurate, competent authorities could systematically cross-check declared information against third-party information (e.g. from banks, financial institutions, employers) or match the data with the information in registers of other government agencies. Processes of cross-checking and data matching could also be automated to minimize the administrative burden (25). KEY TAKEAWAY 4 Reliable data are essential for effective risk analysis. While obtaining these data can be challenging, electronic systems can help reduce the burden by automating procedures of data collection and cross-checking of information with different sources. 3.3.3 RECORDKEEPING Parties to the Protocol are committed to requiring, as appropriate, that all persons or entities engaged in the supply chain of tobacco, tobacco products and manu- facturing equipment keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of tobacco products (Article 9). Relevant information includes market volumes, trends, forecasts of tobacco products and quantities of tobacco products and manufacturing equipment kept in stock in tax and customs warehouses in transit, transhipment and under duty suspension. This information should be required from the persons and entities engaged in the supply chain and submitted to the competent authority on a regular basis, as provided for in the law. The competent authority can use the submitted information to monitor compliance with tobacco regulations and payment of taxes. A registry with this level of detail can realistically be kept only in electronic form. Records must provide full accountability for materials used in the production of tobacco products. The intention is that tax authorities and manufacturers should be able to reconcile the production quantities with the inputs used in production – thereby providing confidence that no unrecorded or illicit production has occurred. Obligations should also be imposed on suppliers of key inputs to show that supply is commensurate with demand (17). KEY TAKEAWAY 5 To monitor compliance and payment of taxes, all persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions, as well as details of materials used in the production of tobacco products. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 111 3.3.4 WAREHOUSING, STORAGE AND DISTRIBUTION According to Article 6.2 of the Protocol, all Parties shall endeavour to license persons involved in any wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Maintaining a system of authorization allows the authorities to carry out controls in production and storage facilities to ensure that taxes are paid (2). The approval process to obtain an authorization could include an evaluation of the layout of the plant or warehouse, the machinery that will be used and the flow of production, warehousing and shipping, including the points of entry and exit of raw materials and finished products. The basic method of monitoring production and ensuring that only tax-paid products are released to the market from the premises is to identify the production facilities and to control the entry and exit points. From time to time, the competent authority should conduct a physical inventory of the goods contained therein to check whether all documentation was duly prepared and approved and to determine the accuracy and completeness of the records kept. If the jurisdiction requires tax stamps to be placed on the tobacco products, only products with the proper stamps affixed can be withdrawn. Generally, tobacco products for which the required taxes have not been paid and, if required, fiscal marks have not been affixed should not be allowed into warehouses. For practical reasons, many countries allow suspension of excise duties, meaning that prior authorized persons can produce, send, receive and store tobacco products on which the excise duty has not yet been paid. The relevant authorities could also require that products on which the taxes have been paid should not be stored in the same areas as the products under duty suspension. Obviously, products under suspension of payment of excise duties are at high risk, which could justify stricter requirements for production, trade, storage and handling. Australia, which has a strict system of licensing and requirements for permission to move tobacco products, has migrated to a new system that eliminates bonded warehouses from the supply chain as of 1 July 2019. Importers are required to have an import licence and to pay excise taxes on cigarettes immediately upon import (see Box 3.1 in section 3.3.1). 3.3.5 DUTY SUSPENSION Many countries require authorization of natural or legal persons (as authorized warehouse keepers) to produce, process, hold, receive and dispatch products sub- ject to excise duty during their business. Producing, processing, holding, receiving and dispatching excise goods often take place under suspension of the excise duty. Guarantees can be requested from authorized persons to secure the payment of taxes. Features of such a system may include strict criteria for granting authorization, 112 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N warehouse pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. The use of a computerized system for monitoring movements of excise goods under suspension of excise duty can be a control as well. Different licences for products under duty suspension could also be considered. This would make enforcement easier and less burdensome for both authorities and operators. In general, it is recommended to allow the handling of excise goods under suspension of duties only if strict criteria are met. Such criteria could include pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. In principle, the movements of tobacco products should also be covered by the tracking and tracing system. Considering the high risk related to these products, additional monitoring could be considered appropriate, such as a computerized system monitoring the movements of excise goods under suspension of excise duty. In the design of such a system, it is recommended that close attention be paid to customs procedures for import and export to ensure alignment and avoid a vacuum in monitoring. An example of a computerized system is the EU’s Excise Movement and Control System, which follows the movement of all excise products – including manufactured tobacco products – for which excise taxes have not been paid. The system records the movement in real time and is thereby an important tool for combatting fraud. In addition, this system is indispensable for the exchange of information and co- operation between the relevant authorities of Member States of the EU (26). Finally, authorization is required before tobacco products can be produced, imported or stored under suspension of excise duties (27). KEY TAKEAWAY 6 Products under duty suspension of excise taxes are at a higher risk of tax evasion, which can justify stringent measures such as requesting guarantees to ensure the payment of taxes, additional licensing requirements, compliance with computerized systems to monitor the movement of excise goods under suspension and on-site authorization and audits. 3.3.6 COLLECTION OF TAXES To reduce the complexity of tax collection systems, it is recommended that excise taxes be imposed at the point of manufacture, import or release from storage or production warehouses for consumption. This is common practice in the majority of countries that impose excise taxes. Collecting taxes at this level of the supply chain CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 113 greatly limits the number of taxpayers and thus the resources needed to control them. Encouraging taxpayers to use electronic payment methods can also increase the chances of collecting all taxes. The same applies to requiring guarantees for certain high-risk activities, such as the handling of goods under duty suspension. Many countries decide on a case- by-case basis the level of the guarantee, depending on the situation of the requestor and the level of risk (quantity or value and potentially due excise taxes) that the regular business activities represent in a given time frame. Some countries allow a reduction of guarantees for operators with a track record of good compliance. It should be noted that a guarantee is not a limitation of the liability; taxpayers can still be requested and liable to pay an amount far above the level of the guarantee. Tax payments should be required by law to be remitted at fixed intervals after sales or on a fixed date each month (2). Many countries have a specialized collection enforcement unit that works full-time on the collection of taxes. It is important to have a stop-filer or payment control that can act immediately when noncompliance occurs, by sending a message and phone call of late declaration or late payment to the taxpayer. This increases the likelihood of keeping taxpayers compliant. If nondeclaration or nonpayment persists, the bond or guarantee could be executed. Another reason for collecting excise taxes around the time of production or import is that quantities can be monitored more effectively at these points. There are different options for monitoring the supply chain of tobacco products. The decision about what kind of monitoring system to use depends on the country’s financial, technical and human resources. The weakest form of monitoring is in- dustry self-declaration. Activities to verify compliance and ensure the collection of the full amount of taxes due can include, for example, physical checks, audits, cross-checking of declared information with third-party data and inspection of administration and recordkeeping. In general, in countries with poor administration systems, enforced compliance is carried out by imposing physical control over the production or manufacturing process. The cost of physical control increases when there is a potential for fraud by excise officers. However, fraud can be diminished significantly when excise officers are rotated frequently among different locations and supervisors make surprise visits. Historically, some countries (e.g. India) have posted tax administration staff at production facilities to monitor production and removals. In India, a staff member of the competent authority is placed in cigarette and large bidi manufacturing facili- ties around the clock. Each officer records the daily production and the quantity of cigarettes/bidis that leaves the factory and reports to the next officer.8 8 Ministry of Finance India, personal communication, 2009. 114 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A better option is to monitor production remotely. The competent authority can require the installation of CCTV cameras in strategic places throughout the manufacturing and warehousing facilities. With these, the authority can establish a central command post from which the facilities and activities can be continu- ously monitored and documented. In addition, the competent authority can carry out physical inventory controls from time to time and – if electronic invoices are implemented – cross-checking between invoices and declared inventory. This is also an effective way to prevent collusion between staff of a competent authority and manufacturers or importers. For example, in 2015, the Bureau of Internal Revenue of the Philippines required all tobacco companies to install CCTV cameras in their production lines and warehouses. This decision was taken in response to large seizures of untaxed cigarettes, with the objective of monitoring production to ensure the payment of all taxes. The collection process must also be supported by IT systems. These systems must provide for transparency and accuracy to ensure a safe process for the flow of payments from taxpayers to the tax treasury. Most countries have implemented automated electronic systems for tax payments linked to each declaration, for both domestic and import. It is key for tax administrations to have a comprehensive agreement with the banking system in order to obtain lower transaction costs, if applicable. Some countries have implemented a state payment web portal that allows citizens to pay their taxes and other fees such as county fees, fees for car permits and licences and agricultural, health and environmental fees online. KEY TAKEAWAY 7 Excise taxes should be imposed at the point of manufacture, import or release from storage or production warehouses for consumption, to ensure that quantities can be monitored effectively. This also reduces the complexity of tax collection systems by limiting the number of taxpayers and thus the resources needed to control them. 3.3.7 TAX REFUNDS Refunds for VAT, excise taxes and customs duties are a common process in most countries, under the principle that consumption taxes are not exported. Frequency and methods of refund vary by country. It is common to have monthly refunds (if there are exports during the period), and the reimbursements may be sent directly to the exporter or reserved as a credit to pay other taxes. An alternative used by some countries that have a high volume of exports is a so-called zero rate, or suspension, meaning that indirect taxes (VAT, excise taxes and customs duties) are suspended for the whole chain – from import of raw materials to production and packing until CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 115 export. This regime requires a special licensing process. Since the tobacco industry has an export component, the refund process for this sector requires special atten- tion for tax administration. 3.4 CONTROL AND ENFORCEMENT Control and enforcement are the main functions of tax administration. In fact, most tax laws include the objective “to control and enforce tax compliance” and, for customs, “to control and enforce tax and duty payments at the border” or similar phrases. The Protocol provides guidance for control and enforcement of tobacco taxes. Efficient and effective competent authorities often have a strategic plan to ensure compliance, a risk-based approach to identify the problematic points in the chain and the ability to direct resources accordingly to high-risk or high-value areas. Tasks that can play a role in control and enforcement include controlling the registration and licensing process, due diligence, verifying declarations and collec- tion of taxes. Production and distribution controls including tracking and tracing, fiscal markings, audits and import and export controls all play a role in control and enforcement. This section describes the main activities for improving control and enforcement, focusing on the tobacco supply chain. The procedures and penalties that can be enacted once illicit trade in tobacco has been detected are also discussed. 3.4.1 CONTROL AND ENFORCEMENT PLANNING Strategic plan In modern tax administrations, it is common to have a strategic plan, with control and enforcement as pillars. Appropriate control of the compliance cycle is key to keeping taxpayers in compliance and preventing illicit trade and tax avoidance. For this reason, most tax administrations focus a majority of their resources on preventive policy. Some examples of this can be found in the strategic plans of the United Kingdom and the United States’ Internal Revenue Service (IRS). In the United Kingdom, HMRC has had a well-developed strategic plan for years. A key pillar of the plan focuses on keeping taxpayers compliant. This is the concept of prevention: controlling initial minor noncompliant behaviour for the majority of taxpayers, while using strong enforcement for the minority on the noncompliant side (28). The strategic plan of the IRS has a similar approach, with a focus on control. If noncompliance is detected, data analysis and behavioural insights are used to identify the best way to address noncompliance. Early intervention or self-correction are examples of ways to address detected noncompliance. The IRS also highlights the importance of resolving noncompliance to ensure taxpayer confidence in the tax system and protecting the integrity of the system (29). 116 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Risk-based approach Following the establishment of a strategic plan, an enforcement and control plan must be drafted. This plan should include definitions of the activities that will be enforced, the taxpayers upon whom they will be enforced and the circumstances under which they will be enforced, as well as allocating resources for staffing, audit- ing, infrastructure and IT. Targets must also be defined, including the number of interventions and the amount of additional collected revenue or reduction of tax evasion. Several tax administrations elaborate annual plans with periodic perfor- mance reviews aimed at improving results, and they correct allocations and targets as needed. Clear targeting of interventions is needed for better results, more efficient use of resources, lower costs for taxpayers and more effective collection. In other words, the point is to focus interventions on those who have a higher probability of noncompliance. Using a risk-based approach can be particularly beneficial. Tax risk management is a key element of control strategy in modern tax admin- istration. A risk is a possible threat to reaching objectives such as collecting taxes in an effective and efficient way for competent authorities. Risk assessment is the process of analysing risks and deciding on the best way to manage an identified risk. The responses can vary from acceptance to mitigation to avoidance. Proper risk assessment allows competent authorities to use their available resources most efficiently and to become more effective in dealing with risks. It can be used to improve compliance by identifying taxpayers or types of activities with a high risk of noncom- pliance. Groups of taxpayers with the same characteristics often have similar risks. Groups with a high risk of noncompliance could then be subject to greater review. Areas of potentially greater risk of noncompliance in the tobacco supply chain include import, export and transfers to and from warehouses, particularly when they take place under duty suspension. Gathering risk-related information from internal and external sources is a best practice in compliance risk management. Such sources could include third-party information (e.g. from banks, credit card companies, transport companies), studies on taxpayer behaviour and research on compliance issues, tax gap analysis, tax audits and declarations (30). Risk management uses these different sources of data along with algorithms to find patterns of high noncompliance. Risk analysis can indicate reduced risk as well. Lower-risk areas are likely to need less governance to ensure compliance, which allows for resources to be directed elsewhere. Risk assessment can therefore help with strategic allocation of limited resources to the areas of greatest risk while at the same time reducing the burden on lower-risk taxpayers. Risk management has always been done by competent authorities, but data availability and statistical methods to identify patterns have changed the way risks CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 117 can be assessed. Although many risk assessment systems are still done manually or include manual elements, the use of intensive data techniques allows systematic, deeper and more targeted analysis (31). Modern risk assessment makes use of electronic data on taxpayers, tax payments, declarations from other taxes, such as VAT, and third-party information. With these data, tax authorities can identify indicators that suggest where further activities might be required to ensure compliance. For example, VAT invoices can be used to match reported purchases of inputs of tobacco leaf to sales invoices of tobacco leaf wholesalers. For taxpayers (i.e. those who are licensed and provide required reports), competent authorities can create a business analytics program to determine whether the data reported are consistent on each side of the transactions. Moreover, in countries with a VAT system, competent authorities can compare the data reported by taxpayers under the VAT system with data reported under the tobacco excise tax system to detect any inconsistency. VAT invoices can also be used to verify inputs and sales data. If VAT is collected at all levels of the supply chain, it is easier for govern- ments to monitor the supply chain for the enforcement of excise duty obligations. Regular surveys on tobacco consumption that use the same methodology can also provide indications about the level of compliance with excise tax policy. A sudden drop in revenue that is not reflected in consumption data could be an indication of illicit manufacturing, illegal imports, cross-border shopping or forestalling. In addition, seizure data can provide valuable information on areas and activities at high risk of noncompliance. The structure of tobacco tax policies should also be taken into account when con- ducting risk analyses. If excise tax rates are increased, there might be a greater risk of forestalling or front-loading (see the discussion on anti-forestalling later in this section). Differentiated excise tax rates based on product or packaging characteristics – such as distinctions between soft and hard packs or filter and nonfilter cigarettes – are also prone to manipulation by operators, which could affect tax revenues. One of the options to mitigate these risks is to amend the excise tax policy and apply a uniform tax rate. For customs transactions, the use of risk management is a key element in target- ing merchandise and support declarations to be inspected. Before the 1990s, most customs agencies used random criteria for selecting targets for inspection. Since that time, many countries have implemented risk-based approaches for selecting inspections. Historical data on importers and trade communities, complemented by artificial intelligence technology, show that risk management tools dramatically increased the effectiveness of physical inspections. Most modern customs agencies have implemented such techniques, allowing for more effective control processes while facilitating smoother processes for those transactions that are in compliance. Box 3.3 details some of the recent changes in risk management processes. 118 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.3 Changes in risk management The OECD developed Fig. 3.2 to show the framework and key steps for understanding compliance risks in 2004 (32). The same approach is still used to identify, assess and prioritize risk. However, many competent authorities now use new technologies and advanced data analytics, along with more information sources, including external data from banks, employers and sales invoices for VAT, for example. Fig. 3.2 Compliance risk management process Source: (32). The methods of identifying risks and the analysis of compliance behaviour have also changed. Traditionally, competent authorities used audits to identify high-risk cases. With more diverse and better data, competent authorities can now use more evidence-based approaches to examine risk patterns. Success of compliance activi- ties is now more often measured in terms of their impact on the overall compliance environment, rather than only on increased revenues.9 9 For more information on effective risk management with several indicators and a checklist of questions, see the Tax Administration Diagnostic Assessment Tool (68). OPERATING CONTEXT Assess and prioritize risks Evaluate compliance outcomes: - Registration - Filing - Reporting - Payment Monitor performance against plan Analyse compliance behaviour (causes, options for treatment) Determine treatment strategies Identify risks Plan and implement strategies CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 119 In Indonesia, use of the compliance risk management process reduced the share of illicit trade in total consumption of cigarettes from 12% to 3%. More information on this can be found in the case study of Indonesia in Box 3.13 later in this chapter. Understanding the products – as well as the supply and distribution chains – al- lows competent authorities to identify which areas along a chain pose the greatest risk and therefore require more resources. Detailed information on the composition of selected tobacco products is given in Annex 3.1. KEY TAKEAWAY 8 Risk analysis helps identify the points of intervention that have higher probabilities of noncompliance. A risk-based approach with targeted interventions allows for better results and more efficient use of resources to ensure effectiveness of tax collection. 3.4.2 CONTROLS OVER THE TOBACCO SUPPLY CHAIN As defined in Article 1 of the Protocol, the supply chain covers the manufacture of tobacco products and manufacturing equipment – as well as their import or export – and may be extended, where relevant, to one or more of the following activities when so decided by a Party: 1. retailing of tobacco products 2. growing of tobacco, with the exception of traditional small-scale growers, farmers and producers 3. transporting of commercial quantities of tobacco products or manufacturing equipment 4. wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Article 4.1 of the Protocol requires parties to “adopt and implement effective mea- sures to control or regulate the supply chain of tobacco products to prevent, deter, detect, investigate and prosecute illicit trade in such goods and to cooperate with one another to this end”. Concrete measures to regulate the supply chain, as well as best practices in this regard, are discussed further below. Figure 3.3 shows the main places for reporting and monitoring along the supply chain: import, ex-factory and removals from warehouses. Manufacturers could be required to report imported inputs at the border, as importers of finished products do. If components are subject to licensing, information can be required as part of the licensing process. The arrows in Fig. 3.3 represent transporting, which is also part of the supply chain. 120 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 3.3 Cigarette supply chain from manufacture or import to retail sale 3.4.3 LICENSING10 AND DUE DILIGENCE A licensing system is effective only if it is properly controlled. Most tax administra- tions have experience with licensing processes for excise taxes on products such as alcoholic beverages and energy products. It is strongly recommended that lessons learned with the licensing process of such products be applied when implementing and enforcing tobacco-related licensing. Licensing provides timely and accurate data that can serve as the basis for audits because it identifies and controls legitimate operators. For new operators, the process to obtain a licence could include visits and verification of production factories, storage facilities and distribution premises. Countries that have no licensing system in place and would like to start applying licences could allow a transitional period for existing operators to comply with the new licensing requirements. The process of licensing control must be carried out and updated periodically, in particular by controlling the validity of bonds or guarantees, the proper functioning of the required systems (CCTV, for example) and recordkeeping. 10 Licensing is discussed here in the context of due diligence and enforcement. Details about how licensing can be set up and what information could be requested are presented in section 3.3.1. Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blend d. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: wher he tobac o will be roll d into st cks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 121 Where licences are required, the law should include a provision specifying that purchases from unlicensed suppliers – or sales to unlicensed purchasers – are not allowed. This means that both suppliers and purchasers would need to verify those with whom they are doing business. This requirement provides enforcement au- thorities with an entry point to enforce the licensing system at both ends. Also, a licensing requirement for manufacturing equipment assists authorities in identifying and prosecuting illegal manufacturing of tobacco products, reducing the burden of proof substantially. In many countries, the presence of manufacturing equipment is not sufficient proof that illegal manufacturing is taking place; the machinery has to be in operation and producing illegal tobacco products when authorities inspect the location. With a licensing requirement, however, the presence of machinery without a licence is sufficient for authorities to act. The validity of licences should be time-limited, requiring renewals or reapplica- tion, to maintain a high level of control. Adherence to the conditions required for a licence should be controlled by the authorities, and penalties for noncompliance – for example, suspension or withdrawal of a licence – should be severe enough to act as a deterrent (33). Regulations for licensing should provide for inspection of the licensee’s products and premises, with penalties for noncompliance, which could include criminal and civil prosecution for serious or repeated offences. As stated in Article 6.3(a) of the Protocol, Parties need to establish or designate a single authority or multiple authorities to issue, renew, suspend, revoke and/or cancel licences. In accordance with Article 7 of the Protocol, persons engaged in the supply chain are required by law to conduct due diligence before and during business relationships. They also must report to the competent authorities any evidence that a customer is engaged in activities in contravention of its obligations arising from the Protocol. This requirement includes customer identification, monitoring of sales to ensure that the quantities are commensurate with demand for such products within the intended market and taking measures to ensure compliance. Knowledge of the demand of a market is indispensable for determining if there is a case of oversupplying. If the supply of tobacco products to a lower-taxing foreign market exceeds the demand, it creates a higher risk that these products will be smuggled back into a higher-taxing country, undermining the objectives and effectiveness of the higher-taxing jurisdiction. In the past, some governments decided to impose a fine on tobacco companies if the quantities supplied were significantly higher than the demand and the risk of being smuggled back into their jurisdiction was judged to be high (34,35).11 11 Excise duty rates applied in all the EU countries can be found on the European Commission’s webpage: https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco-energy/ excise-duties-tobacco_en. 122 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 9 Licensing helps to identify and control legitimate operators. The data obtained from licensing can serve as a basis for audits. Licences should be controlled on a regular basis and updated periodically to ensure their validity. 3.4.4 FISCAL MARKINGS (E.G. TAX STAMPS) Fiscal markings are another important tool for controlling and monitoring pro- duction and import of tobacco products. Their use is generally considered to be appropriate for increasing compliance with tax laws. Fiscal markings can also be of help in distinguishing between genuine and illicit tobacco products. Tax stamps or other fiscal markings affixed to packs of cigarettes or tobacco products facilitate the collection of excise taxes, as well as audits and enforcement actions. The pres- ence of fiscal markings enables both the competent authority and the public to monitor whether the taxes on tobacco products were properly paid. It thus assists the competent authority in investigating illicit trade and prosecuting violations. Fiscal markings include tax stamps, enhanced tax stamps (banderols) and digital tax stamps. Examples of fiscal marks are tobacco stamps, tax stamps, excise stamps, tax stickers and banderols. Box 3.4 presents details on the different types and features of tax stamps. Tobacco products for export are often required to be marked that they are for export. Box 3.5 provides useful information regarding the International Organization for Standardization (ISO) standard for excise tax stamps. The terminology “fiscal mark” holds no indication of the characteristics of the mark. A fiscal marking is affixed to each pack of tobacco product. Requiring a standard package size can facilitate the application of the markings (2). Fraudsters can be deterred from attempting to re-use fiscal markings (in particular stamps) by having the marking affixed to each pack of cigarettes (or other tobacco product) before the pack is wrapped with cellophane (36). In most cases, tax stamps are purchased by the producer or importer and applied to each product sold as proof of excise tax payment (33). Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Box 3.4 Types and features of tax stamps Over time, tax stamps and markings have become more sophisticated. In the past, tax stamps were often paper-based and easy to counterfeit. New tax stamps use additional security features to make them more difficult to counterfeit. Authentica- tion solutions against counterfeiting can utilize various security features, including: • overt features – features that can be verified by the naked eye; CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 123 • covert features – features that can be authenticated only by using dedicated and specialized electronic readers; • semi-covert features – features requiring a simple tool that does not involve extensive training; and • forensic features – features that can be identified through laboratory analysis. Tax markings can be either physical or digital: • physical markings – the information is contained in the document or device attached to the package. • digital markings – information is obtained through a link with a database and by decrypting with the tools and keys used for creation of the data. The term “digital tax stamp” sometimes leads to confusion, as some paper-based stamps with digital components are also described as digital tax stamps. Tax markings that are fully digital do not contain information in the document or device attached to the package. It is probably too simplistic to say that digital tax stamps are more secure than paper- based tax stamps. Both types have advantages and weaknesses. For instance, both physical and digital tax stamps can be weak or strong on security features. Neverthe- less, the management, production, sales, transport and monitoring of physical tax stamps require increased attention. Tax stamps have the same value as banknotes and are a possible attraction for theft, loss and fraud. In Belgium, for example, the Court of Audit severely criticized the lack of control of the production and stock management of tax stamps in 2015. The Court of Au- dit concluded that tax stamps issuance should operate under recognized security practices and procedures relative to the security risk associated with the various production, distribution and issuance processes. Moreover, it was noted that new printing technology of digital tax stamps on packs may facilitate stock management and lead to less fraud. In March 2016, Belgium changed its stamps. The printing became an in-house process by the financial federal government department. The new stamps are still printed with a watermark, but they also have a digital component. The change resulted in a cost reduction by standardizing the sizes and optimizing the production process. Some of the more advanced fiscal marking technologies include embedded threads and watermarks; special inks and coatings, such as so-called invisible inks, holograms and foils; and calculated or changeable content. Because of their enhanced security features, these stamps can be more expensive than traditional stamps. In the state of California in the United States, the traditional stamps cost US$ 0.42 per 1 000 stamps. The cost of the first generation of high-tech stamps was 10 times higher, at US$ 4.77 per 1 000 stamps. This price nearly doubled for the second-generation encrypted stamp, to US$ 8.20 per 1 000 stamps. Nevertheless, California collected 124 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N about US$ 450 million of additional tax revenue in the first decade following the implementation of encrypted tax stamps. This additional revenue was far greater than the costs of implementation and enforcement. Other jurisdictions have also revised their tax stamps to incorporate new technolo- gies. The state of Michigan, for instance, replaced heat-applied cigarette tax stamps with digital pressure-applied stamps in 2015. Michigan deployed a tax stamp with several overt and covert security features and a unique quick response (QR) code and serial number. QR codes (machine-readable codes consisting of an array of black and white squares, typically used for storing URLs or other information) can have purposes beyond tracking and tracing. The QR code can be read by consumers with a smartphone or tablet application to access information on smoking-cessation programs, report violations of the state’s youth access policies, connect to a tip line to report noncompliant packs and learn about the harms from illicit tobacco sales and purchases. Enforcement authorities can validate stamps using the smartphone- based eTRACS (Electronic Tax Reporting and Audit Compliance System). As part of the system’s implementation, the Michigan State Police department created teams of enforcement officers in each of the state’s seven districts and the state Department of Treasury created its own enforcement team. Sources: (37–42). Box 3.5 The ISO standard on excise tax stamps The ISO published its excise tax stamp standard (ISO/TC 292/SC) in October 2018. The purpose of the ISO standard is to assist tax and finance authorities in enhancing compliance with excise tax regulations. A tax stamp is defined as a visible tax stamp, label or mark placed on certain types of consumer goods to show that the applicable excise tax has been paid. The ISO standard applies to tax stamps that are physical in nature – not to digital markings, which are directly printed on to packs without a physical component. “Authentication” in this standard refers to the authentication of the tax stamp, not the product on which the tax stamp is affixed. In other words, authentication of a tax stamp on a cigarette pack means that the tax stamp is authentic but does not guarantee that the pack is authentic. In addition, control measures are needed at the time of the application of the stamp to verify the conformity of the tax stamp with the corresponding product. The standard provides guidance on the content, security, issuance and examination of physical tax stamps used to indicate that the required taxes have been paid and that the tax stamp is authentic. The use of stamps to facilitate tracking and tracing within the supply chain is not described. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 125 Specifically, the ISO standard deals with the following issues: • defining the functions of a tax stamp • identifying and consulting with stakeholders • planning the procurement process and selection of suppliers • the design and construction of tax stamps • the overt and covert security features that provide protection of the tax stamp • the finishing and application processes for the tax stamp • security of the tax stamp supply chain • serialization and unique identifier codes for tax stamps • examination of tax stamps • monitoring and assessing tax stamp performance. A stamp may fulfil many functions, but the core business of tax stamps is to ensure and facilitate the collection of revenue. The tax stamp must use a combination of security features. The tax authority should ensure that the tax stamp can be authen- ticated and that counterfeit, altered, tampered or otherwise fraudulent tax stamps can be detected. The standard provides detailed information on the different components of the tax stamp such as the substrate, inks, adhesives, laminate, authentication or security features and the unique identifier that should enable checks on the payments of the required tax. The process of procurement is discussed in detail in the standard. The tax author- ity should ensure that the procurement process is open, transparent and meets the sustainability objectives. The tax authority should set out the goals and requirements to give tendering organizations more leeway in proposing optimum solutions that might be different from those the authority would specify. The standard is not prescriptive; rather, it provides a catalogue of options. It does not, for instance, recommend specific security features, but it does describe the different types of features that are necessary for a tax stamp to be secure. Tax officials still need to make decisions and choose the option that suits them best, but the standard remains recommended reading for those who would like to introduce tax stamp programmes in their jurisdiction. ISO standards are not freely available but can be purchased at the ISO Store (www.ISO.org) or from an ISO national member body. Source: (43). According to Article 8 of the Protocol, each Party shall require that unique, secure and nonremovable identification markings – such as codes or stamps – are affixed to or form part of all unit packets, packages and any outside packaging of cigarettes for the implementation of the tracking and tracing system within a period of five 126 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N years, and of other tobacco products within a period of 10 years, of entry into force of the Protocol for that Party. The Protocol specifies that at least the following information shall form part of the unique marking: • date and location of manufacture • manufacturing facility • product description • where available, the intended market of retail sale. In several countries, QR codes are used as fiscal markings for tobacco and alcohol tax control. Each stamp has a unique identifier code and a QR code. The data stored in the QR code provide the following product information: • manufacturer • production location • stamp order date • tax status and class • brand • intended market • unique identifier (serial number). KEY TAKEAWAY 10 The use of fiscal markings is generally considered to be an appropriate tool for increasing compliance with tax laws. Fiscal markings can also be helpful for distinguishing between genuine and illicit tobacco products. 3.4.5 TRACKING AND TRACING A tracking and tracing system assists authorities in determining the origin of tobacco products – and the point of diversion, if applicable – as well as monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to enable authorities to have information on all transactions through the entire tobacco product supply chain until duties are paid or other obligations are discharged. Traceability is not used only for tobacco products. It is also used to improve the supply chain function, as in the case of parcel services, as well as for product safety reasons, to manage potential product recalls and for regulatory reasons. Tracking is the process that monitors where a product is at all times while also creating a time and location record for all movements. Tracing is the ability to identify the past locations of a product, so that the product’s route can be followed CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 127 back to its origin (44). In other words, traceability is “the ability to trace the history, application or location of an object” (45). A tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it is possible to unambiguously register that product’s movements. Other necessary characteristics include the ability to share the registered move- ment information and to authenticate products. This enables a product’s status to be captured through the supply chain and its history to be identified and verified retrospectively. According to Article 8.4.1 of the Protocol, Parties should require the following information to be available: • the date and location of manufacture • the manufacturing facility • the machine used • the production shift or time of manufacture • the name, invoice, order number and payment records of the first customer not affiliated with the manufacturer • the product description and intended market of retail sale • any warehousing and shipping • the identity of any known subsequent purchaser • the intended shipment route, date, destination, point of departure and consignee. A good tracking and tracing system enables the government to properly monitor the supply chain, improves its ability to ensure collection of the proper duties and taxes, provides it with the ability to authenticate whether the identification marking is genuine and matches the product and improves its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. The following elements are required for an effective tracking and tracing system (46): • A serialized unique identification marking for each package of product. These identifiers are a distinctive combination of numbers, letters or both. They cannot be predictable or used more than once. The representation of the identifier on the package can be human-readable (letters or numbers) or machine-readable (barcodes). Generation of codes and encryption that are part of a tobacco industry patent should be excluded. • A data carrier with the serialized unique identifier and other information such as date and location of manufacture, manufacturing facility, product description and, where available, the intended retail market. This informa- tion should be readable by authorized agencies of any Party to the Protocol. The data carrier should comply with quality standards and be suitable for 128 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N high-speed production lines. Two-dimensional barcodes, for example, meet these standards and are readable with inexpensive equipment. • A link and parent-child relationship (called aggregation) between different packaging units that offers the option to trace a pallet without the need to scan all the packs and master cases of that pallet. • Recordkeeping of all shipping and receiving events along the supply chain. This includes, for example, the departure location and the arrival location, as well as the involved operators. International standards from the ISO are recommended for the capture and exchange of data and events. • The use of international standards for key information that is encoded in the data carrier (5). An example of a unique and internationally recognized identifier for products is a Global Trade Item Number. The following details on information storage and sharing are drawn from various sections of the Protocol. Data and events along the supply chain must be stored in an independent database that is controlled by competent government authorities. At the global level, national and/or regional databases can be interconnected to facilitate international inquiries by competent authorities. Parties to the Protocol agree to establish a global information-sharing focal point located at the Conven- tion Secretariat of the WHO FCTC, accessible to all Parties, enabling them to make enquiries and receive relevant information. Each Party shall ensure that the information recorded under paragraph 5 of Article 8 of the Protocol is accessible to the global information-sharing focal point on request, subject to paragraph 9, through a standard electronic secure interface with its national and/or regional central point. The global information-sharing focal point shall compile a list of the competent authorities of Parties and make the list available to all Parties. The cost of tracking and tracing systems is a concern for many countries, but as indicated in paragraph 14 of Article 8 of the Protocol, jurisdictions may require the tobacco industry to bear any costs associated with putting in place the tracking and tracing system in a country (46). In Brazil, the cost for cigarette manufacturers was US$ 0.0185 per pack (42). In Kenya, the cost for manufacturers was US$ 0.024 per pack (42). Along with considering the characteristics of a tracking and tracing system in selecting a particular one, it is important to avoid conflicts of interest, ensure fair and transparent dealing with suppliers, implement a zero-tolerance policy for corruption or anti-competitive behaviour and ensure compliance. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, which deals with industry interference, and Article 8 of the Protocol. Article 8.13, which states that “each Party shall ensure that its competent authorities, in participating CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 129 in the tracking and tracing regime, interact with the tobacco industry and those representing the interests of the tobacco industry only to the extent strictly necessary in the implementation of this Article”. Box 3.8 provides a cautionary example of a tracking and tracing system that is not compliant with Article 5.3 of the WHO FCTC. While the Protocol contains a great deal of information on the requirements that a tracking and tracing system should meet, questions come up in relation to the implementation of such systems. To achieve the objectives of the Protocol, the Meeting of the Parties (MOP) to the Protocol, as the governing body of the treaty, has the prerogative to establish subsidiary bodies, such as expert groups and working groups. In decision FCTC/MOP1(6), the MOP established a working group for the development and implementation of tracking and tracing systems in accordance with Article 8 of the Protocol, including the global information-sharing focal point (Article 8.1) and unique identification markings for cigarette packets and pack- ages (Article 8.3), to further elaborate on the next steps. The working group will produce a comprehensive report compiling good practices and experiences on the implementation of tracking and tracing systems, as well as unique identification markings for cigarette packets and packages at national or regional levels. The working group was also given a mandate to prepare a conceptual analysis of how a global information-sharing focal point could be set up. Implementing a complete tracking and tracing system with fiscal markings takes time. In most of the countries that have already implemented tracking and tracing, it took several years from starting with the legal framework to final implementation. Several hurdles need to be overcome: • Legal framework approval is usually delayed by the tobacco industry. • Knowledge of tracking and tracing and associated technologies is scarce at tax administrations. • Tender and bidding processes are complex. • Coordination between domestic tax authorities and customs is weak. Even though the process might be lengthy, the investment in a tracking and tracing system will be repaid with the amount of tobacco taxes that are not lost due to evasion. When implementing a new tracking and tracing system, tax administration should ask for collaboration and technical assistance from intergovernmental organizations and countries that have successfully implemented such systems, in order to speed up and ensure success of the process. Examples of tracking and tracing systems implementation in Chile, Kenya and the EU are detailed in Boxes 3.8, 3.9 and 3.10. For countries that already have fiscal markings in place, the potential interaction between the markings and the implementation of a tracking and tracing system should be taken into account. Further information on this interaction is presented in Box 3.6. 130 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.6 Tracking and tracing and fiscal markings It is becoming increasingly common for stamps to contain some tracking and trac- ing features, such as unique identification markings and basic information on the product that describes the company, tax status or the product itself. The intention is to mark each pack with a unique identification marking so it can be monitored from the point of production to the retailer, including each step in between, thereby creating a complete time and location history. Although a tax stamp could meet the requirements of Article 8 of the Protocol and have tracking and tracing features, in general, the focus of tax stamp systems differs from that of tracking and tracing systems. Tracking and tracing is more than the unique, secure and nonremovable identification markings on the packages of tobacco products. It implies reading or scanning the codes; linking the codes between packs, cartons, master cases and pallets; uploading the information to a database; recording of any shipping and receiving events along the supply chain; and interconnecting the different databases. While new tax stamp programs contain tracking and tracing features, they are primarily intended to facilitate tax collection on the domestic market and not to track duty-suspended cross-border trade or the export of products. The focus of tax stamp systems is on authenticity and the proof that taxes are paid. The focus of tracking and tracing systems is on unique identification and on control of the movements in the supply chain by monitoring and investigating the past and future location of products. Tax stamp programs focus on stock management, verification (that the stamps correspond to the product) and authentication (that the stamps are genuine), while the focus of tracking and tracing systems is on the origin, intended route, first customer and final destination. The focus of tax stamps is primarily on individual packs intended for the duty-paid domestic market, while the focus of tracking and tracing systems is on all packaging (packs, cartons, master cases, pallets) and certainly – but not exclusively – for the duty-suspended export market. Nevertheless, sometimes there can be synergies. For example, the EU countries that require a tax stamp or national identification mark for fiscal purposes have the option to use it as the security feature for tracking and tracing purposes, provided that the requirements are met. In summary, tax stamps can be converted to or be part of a tracking and tracing system when the converted system provides aggregation between packs, cartons and master cases and records all movement along the sup- ply chain. For the export market, a unique identification marking should be added. Sources: (39–42). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 131 Box 3.7 What not to do: use the industry solution to tracking and tracing Codentify is a serialization system used to produce cigarette pack markers for the purpose of verifying whether cigarette packs are legal. It was patented by PMI but subsequently licensed at no cost to other major cigarette manufacturers. In 2016, Codentify was transferred to Inexto, which is an affiliate of the French group Impala. What is the problem with Codentify/Inexto? The main issue is that Codentify/Inexto’s links to the tobacco industry make it incom- patible with the Protocol, which came into force in September 2018. The Protocol specifies that obligations assigned to a Party shall not be performed by or delegated to the tobacco industry. Additionally, many elements indicate it is an ineffective means of authentica- tion. For example, the 12-character digital codes generated by Codentify can be easily duplicated or cloned and used as originals on either a counterfeit or genuine pack, which can then pass the system’s basic verification test. The codes are also produced by relatively unsecured, commercially available equipment and do not include high-security features capable of protecting the authenticity of identifier numbers. Systems that use multilayered, advanced security solutions that enable distributors, retailers, customers and authorities to identify noncompliant products are more secure. Another problem is that Codentify/Inexto cannot track products as efficiently as other available systems. It requires a much larger enforcement capacity to achieve the same detection rates as other systems that are not linked to the tobacco indus- try. Authorities would have to inspect significantly more packs marked under the Codentify system than is necessary under some other systems to achieve the same certainty of not missing a fraudulent pack. In addition, not all stakeholders will be able to verify that a pack marked under the Codentify system is genuine, while other available systems do offer this possibility. Source: (47). 132 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.8 A successful tobacco traceability system: SITRAF, Chile The Servicio de Impuestos Internos (SII, Internal Revenue Service of Chile) has success- fully coordinated and implemented a tracking and tracing system for tobacco products. Application of a compliance management model In the framework of a compliance management model, a traceability system for tobacco products utilizes a structural measure to reduce tax evasion. It is estimated that evasion of taxes (VAT and excise taxes) in the cigarette market in Chile amounts to 16.6% of the country’s total market annually – approximately US$ 300 million. The traceability system implemented in Chile (SITRAF (TAB2)) allows authorities to know in a certain and timely manner the quantities of cigarettes produced or imported into the country. Moreover, it helps authorities to distinguish between counterfeit products and original products that did not comply with payment of the tax. In 2018, the implementation of the traceability system was awarded to a company through public bidding. The company is in charge of the implementation and operation of the system for five years, according to a contract signed with the SII, and must maintain a team of 20 people available for the project. Direct markings are applied to items produced in Chile for national consump- tion, and stamps are used for imported products. For both types of product, the marking is based on a data matrix code, which is printed using security ink that is distinguishable from any other type of ink with specific devices that are provided by the awarded company. Although products for export are not subject to marking, they are controlled and accounted for by the traceability system. For national production, devices are installed on each production line that rec- ognize the type of pack being produced, print a unique code on each pack and then read it (activation) to save all the information on servers located in the production plant. This information is transmitted to the central servers of the system and then to the SII. It is also available for on-site inspection. In the case of imported products, the stamps must be acquired in Chile by each importer and then sent to its producer abroad, which is responsible for adhering them to each pack of cigarettes prior to wrapping the packs with cellophane, us- ing applicators on the production lines. Once the cigarettes enter Chile, the tax determination process has been completed in the service (Provisional Free Transit Guide) and the corresponding taxes have been paid to customs (Import Declara- tion), the importer must enter the data on the stamps used by the importer on the platform of the traceability system. After validation, the stamps can be activated in the system – that is, they are recognized as valid for commercialization. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 133 In addition to the devices provided by the company, SII has developed a smart- phone application for verification by citizens. Although the application cannot verify the authenticity of the ink used, it is able to verify whether a code is correctly gener- ated and display the information contained in the traceability system for the code (brand, variety, quantity of cigarettes, products/importer) so that the taxpayer can verify its consistency. Progress of the compliance management model Implementation of the traceability system has required a coordinated effort both within the service – for the generation of instructions, procedures and computer developments – and with other institutions, such as the National Customs Service and the Ministry of Health. It is an unprecedented project in terms of coordinating the implementation of the system in the production lines of the different tobacco companies in the country and the provider company. Some of the main milestones of the project are: September 2014: Law 20,780 on Tax Reform establishes an obligation to implement the system within a term of six months, after the publication of the resolution determining the obligated tax payers. May 2015: Resolution No. 47 determines obligated taxpayers. June 2015: Circular No. 47 describes obligation to incorporate stamps or distinctive marks as a traceability mechanism. February 2016: Law 20,899 on Tax Reform simplifies the definition of the system, allowing the system to be outsourced or provided by the SII, in addition to making the type of traceability more flexible. August 2016: Traceability system regulation D.S. 1,027 is issued (published on 28 December 2016). March 2017: Exempt Resolution No. 49 of the Ministry of Finance authorizes the SII to outsource all or part of the traceability system. June 2017: Bidding bases in public market are published. February 2018: Tender is awarded to selected company. June 2018: Decision of contract is made by General Comptroller of the Republic. August 2018: Resolution No. 61 determines taxpayers obliged to apply the trace- ability system. August 2018: Holding of first workshops for detailed definitions of the project, with the participation of Customs, Ministry of Health, provider and SII. September – October 2018: Visits of plants to coordinate with producers and define adaptations to production lines for system implementation. January 2019: Resolutions No. 6, 7 and 8 passed, with response to a request for an 134 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N extension of producers; start of system implementation in all production lines in the country. February 2019: Resolution No. 16 establishes a term to commercialize the remaining stock without marking. 2019: Resolution No. 24, with request for extension to importer, includes training of customs staff. March 2019: The traceability system is started up. June 2019: Stamp process begins for all imported cigarettes. Currently, the system is installed and operating in all production lines in the coun- try, placing traceability markings on virtually 100% of the cigarettes produced and imported. During 2019, the traceability system enabled controlling approximately 1 175 million packs: 744 million produced for national consumption, 409 million produced for export and 22 million imported yearly. Source: (48). Box 3.9 Case study of Kenya’s implementation of a tracking and tracing system Kenya’s current tracking and tracing system was preceded by a series of reforms in both tax structure and administration of excise taxes. The reforms included electronic cargo monitoring of exports, which allowed for automatic monitoring and reporting. The system appears to be highly effective because it requires less capacity and is less prone to manipulation than earlier systems. The experience of Kenya shows that a lower-middle-income country can successfully implement a sophisticated system capable of decreasing illicit trade. It also shows the importance of other measures such as strengthening enforcement, increasing cooperation and communication among different agencies and increasing penalties for noncompliance. Illicit trade, as measured by the Kenya Revenue Authority (KRA), was estimated to be around 15% of total consumption in the market during the initial reform period. After the introduction of the new system in 2015, it dropped to 5%. Timeline of the major reforms: 2003: In this period, the paper tax stamps used had a unique identifier and were colour-coded to indicate the type of product. Regular compliance checks were in- troduced. In 2007, the cost of a stamp was 2.124 Kenyan shillings or US$ 0.023 per pack. However, the stamps were found to be easily counterfeited and could not be linked to specific brands. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 135 2008: The KRA proposed a tracking and tracing system and increasing tax rates. The new system was introduced gradually. 2010: Enhanced security features, including ultraviolet markings, were added to the paper stamps. The stamps were to be clearly visible when packs were displayed for sale and placed so that opening a pack would destroy the stamp. The stamps were verified at four different points along the supply chain. The costs were just slightly higher than those of the previous stamps at US$ 0.024 per pack. Licensing was introduced for domestic manufacturers, subject to annual renewal. Importers were required to register with the KRA. Licences required submission of details on the company directors, inventories and equipment, accounting systems, input-to-production ratios and brands produced. Penalties for noncompliance were increased and included up to three years in prison. An electronic cargo tracking system was launched. Electronic seals were affixed on containers or trucks, and GPS technology was used for tracking. A bond was payable on exports to cover excise and VAT taxes. The bond was released only when the goods reached the final destination and taxes were paid. Verification involving both countries of the business deal takes place at the bor- ders. The electronic system provides information about the departure and arrival of the goods and the disarming of the seals. Authorities in the importing country are notified before the shipment leaves the domestic production facility. The system reduces the number of checkpoints and staff needed and generates arrival reports that can be verified with VAT refund requests. As a result of these changes, three factories and seven of the 10 importers were shut down due to noncompliance. Exports to Côte d’Ivoire, Eritrea, Mali and Sudan stopped because companies could not provide evidence that the goods reached the final destination and taxes were paid. More than US$ 11 million in excise tax losses was recovered in 2011. The KRA estimated that illicit trade dropped to 8%. 2013: A contract was signed to introduce a tracking and tracing system for tobacco and alcohol, the Excisable Goods Management System, in April. The system added production counting, tracking and tracing, stock control, processing and other data collection to the existing system. Infrastructure requirements included high-speed broadband internet at production facilities, warehouses, the KRA and ports, along with reliable power or backup generators at those points. Implementation was planned in three stages: • Stage 1 – A new electronic digital stamp with a unique identifier was introduced. It included a data matrix code plus overt markings (holograms, fluorescent fibres, a security link for KRA authentication and visible two-dimensional codes for verification and activation), semi-covert markings (UV features, fluorescent prints detectable by specialized devices, mini text printing for retailers and 136 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N distributors) and forensic taggants for use in prosecutions. The stamp also included human-readable codes for verification by short message service using the KRA web portal. • Stage 2 – Control and monitoring systems were automated in February 2014. Manufacturers had to install photosensitive readers on production lines, with data automatically sent to the KRA in real time. Each stamp was activated and associated with a brand and package size on the line. The KRA database is automatically updated every 15 minutes. • Stage 3 – Market surveillance began, with 83 officers given powers to seize illicit cigarettes and make arrests. The officers were equipped with hand-held devices that transmitted data to the KRA for authentication. Distributors and retailers became liable for selling products without an excise stamp and were subject to fines plus prison sentences of up to three years for noncompliance. In 2016, a smartphone application became available with which the public could authenticate cigarette packs. Importers must now buy digital stamps and send them to export facilities in other countries to be affixed. Tax liability is due at removal from a factory or at import. The electronic cargo monitoring system is still in effect. 2016: The Excise Duty and Tax Procedures Acts clarified new obligations and penalties. 2017: A new integrated customs management system was launched. The KRA estimates that illicit trade levels are now around 5%. The current, more comprehensive digital system is cheaper than the previous paper tax stamp system. Manufacturers pay for the production monitoring system, but it counts as a busi- ness expense on corporate tax returns. In 2018, two manufacturers and 10 licensed importers were operating in Kenya. In 2018, aggregation between the markings of packs, cartons and master cases had not yet been implemented but was expected to be forthcoming. Source: (36) Box 3.10 Case study of the new EU tracking and tracing system, May 2019 Cigarette smuggling and other forms of illicit trade in the EU is estimated to cause a loss of €10 billion in revenue annually. In 2018, 4.2 million packs (20 sticks per pack) of illegal cigarettes were seized by customs in the EU. Illicit tobacco production was also increasing: an illegal factory in Ireland, dismantled in 2018, was capable of producing 250 000 cigarettes per hour. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 137 The EU tobacco control policy is described in the Tobacco Products Directive and is influenced by the Protocol. Article 15 of the Directive calls for the traceability of cigarettes and RYO tobacco products by May 2019 and of other tobacco products by May 2024. The EU tracking and tracing system is sufficiently flexible to be imple- mented at both the regional and the single-country level. Countries can choose among providers as long as the basic requirements are met. The policy provides a high level of protection against any attempts at manipulating the data. The report- ing obligations cover all the economic operators involved in the manufacture and distribution of tobacco products. The EU system requires all unit packets of tobacco products to be marked with a unique identifier Information on the movements of those products is to be stored by third-party data storage providers. The suppliers of the unique identifiers and data stor- age are to be financially and legally independent from the tobacco industry. The data are to be fully accessible to authorities of EU Member States for enforcement purposes. The generation of unique identifiers, as well as all other codes required for pre- registration of economic operators, facilities and machines, will be done at the Member State level by designated identifier issuers. Manufacturers and importers are required to supply information relating to the product and production lines when requesting unique identifiers from the issuers. The issuers will then generate and deliver batches of unique identifiers. On the production line, manufacturers of tobacco products will complete each unique identifier with a marking indicating the date and time. The unique identifier will be a machine-readable, optical, one- or two-dimensional barcode. An anti-tampering device, capable of creating an unalterable independent record of the verification process, must have been installed previously. This additional record will be accessible to public authorities for potential investigation and inspection. Unit packets, as well as aggregated packages such as cartons, master cases or pallets, can be tracked and traced throughout the supply chain. Tracking is also allowed at an aggregated packaging level as long as unit packets remain traceable. During transport, each dispatch and arrival up to the final dispatch to the first retail outlet must be recorded and reported. All recorded information must be submitted to the independent third-party data storage facility, generally within three hours, and 24 hours before dispatch and transloading. Costs, including operational costs, are shifted to the tobacco industry, in line with Article 8 of the Protocol. The EU system of tobacco traceability and security features became operational on 20 May 2019. Sources: (36, 49–50). 138 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 11 A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as monitoring and controlling the movement of tobacco products and their legal status. 3.4.6 ANTI-FORESTALLING “Forestalling” is a term that describes increases in the production or stock of products in anticipation of a tax increase (2). Other terms referring to this practice include “stockpiling” and “front-loading”. Forestalling occurs when manufacturers or importers increase their tax-paid stock or oversupply the market by increasing production or imports in order to pay the previous lower rate. It reduces and delays the effective- ness of tax measures. The effective starting date of the new tax rate will be delayed, revenues will be lower and the possible effect on prices and thus consumer behaviour will also be postponed. To illustrate how anti-forestalling measures function, an example is presented in Annex 3.2. A legal basis must exist for anti-forestalling measures; otherwise, the govern- ment cannot prevent the industry from forestalling. Legal measures to deal with forestalling include (51–52): 1. Limiting the amount of tobacco products that can be released subject to the old tax rate and levying the new tax on the products exceeding that limit. 2. Levying the new tax rate on all goods that are still in stock and not yet sup- plied to the final consumer. 3. Limiting the number of tax stamps issued at the rate that was in effect before the increase or limiting the time that products with a tax stamp with the old rate can be sold. 4. Requiring producers and importers to buy new tax stamps annually or after a tax increase. Under the first three measures, the competent authority determines the limit for taxation at the previous (lower) rate. The quantity allowed may be based on the shelf life of tobacco products – around six months for cigarettes – or normal inventory levels, such as an average over the previous three years. The first measure, limiting the amount of tobacco products that can be released, requires resources from the competent authority for enforcement. Authorities may decide to post inspectors in each production facility, but even without posting inspectors, procedures are necessary for determining when the allowed quantity has been exceeded and what subsequent actions to take. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 139 The second measure, levying the new tax on goods in stock and not yet supplied to the final consumer, might be difficult to implement. The competent authority is required to monitor the manufacturing process and, at the very least, to conduct a stock-taking of all players in the supply chain – including the various manufacturers, importers, wholesalers and retailers – in a very short window of time. If monitoring covers only the stock of manufacturers and importers, this measure could easily be circumvented by ensuring the stock is sold to others in the supply chain or by setting up separate distribution companies to purchase the stock. Controlling stock at the retail level is burdensome and might not be administratively feasible given the large number of cigarette retailers. It becomes even more burdensome if there is no licensing requirement for retailers because they will first have to be identified. The third measure accomplishes the same thing as the first if a country uses tax stamps. The fourth also requires tax stamps and is simpler for the competent authority but somewhat more burdensome for the tobacco companies, since stamps must be purchased every year. Box 3.11 provides examples of anti-forestalling measures in EU countries. Box 3.11 Examples of anti-forestalling measures in EU countries Several EU Member States have taken measures to limit forestalling. A cautious ap- proach seems to be required for designing such measures to ensure that they comply with EU legislation and the general principles of EU law – in particular, the principle of proportionality. No disputes have occurred concerning the right of initiative of EU Member States to implement anti-forestalling measures. Nevertheless, several EU Member States had to defend their measures in front of the Court of Justice of the European Union, the institution that ensures all national legislation is in line with EU law and a consistent application of that law (53). The Court acknowledged that anti-forestalling measures are appropriate to combat tax evasion and tax avoidance. Moreover, the Court emphasized that fiscal legislation is an important and effective instrument for discouraging consumption of tobacco products and therefore for protecting public health (54–55). However, the measures taken should be proportion- ate to the objectives. The principle of proportionality means that only the action needed to achieve the objective should be taken, and it should not exceed what is necessary. This principle regulates the measures taken within the EU and is included in the Treaty on the European Union. The Court demanded that Portugal amend its legislation to ensure compliance with the principle of proportionality. Belgium, Estonia and Hungary were also urged to change their anti-forestalling measures to bring them in line with EU legislation (56–57). 140 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The majority of the EU Member States have anti-forestalling measures in place, but there is no harmonization of these measures. The following are some examples:12 • Portugal limits the quantity of cigarettes that can be released in the last four months of a year to the average of the previous 12 months plus 10%. In addition, manufacturers and importers must sell cigarette packages with a tax marking of the preceding calendar year within three months. For other tobacco products, longer limits apply. • Denmark limits the number of tax stamps issued before a tax increase at the old rate to 20% more than are usually purchased in the two months before the end of the year. • In Poland, tax stamps are valid only for the current calendar year, and cigarettes with the old stamp can be sold only through February of the following year. • In Romania, companies must apply for approval to release for consumption from the customs office. Source: (58). KEY TAKEAWAY 12 Forestalling reduces and delays the effectiveness of tax measures. Implementing anti-forestalling measures can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. 3.4.7 ADDITIONAL NATIONAL AUDITS AND CONTROLS In addition to the previously described measures, several periodic audits and controls could be implemented to increase compliance with tax laws. The most common audits and controls are the following: • Cost audit – The cost audit method provides expected VAT and tobacco tax collection by simulating the intermediate and final cost of cigarettes. It starts with the inventories of raw materials and estimates added values and final cost, then matches the results with real collection from the tobacco supply chain. Annex 3.1 provides more information about the components that make up some selected (tobacco) products. • Transfer pricing audit – To ensure companies pay their fair share of tax, prices of transactions between related companies should be assessed, and when prices are not in line with the market conditions, they should be corrected. Companies that operate at the international level (transnational companies), including many tobacco companies, can manipulate import or export prices 12 Considering the frequency of court cases concerning anti-forestalling incidents, these measures might have been replaced or amended. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 141 of merchandise or raw material to related companies or branches in other countries, with the objective of lowering profits in countries with higher tax rates, and can transfer those profits to countries with lower taxes. • Price and market monitoring – Retail price surveys can provide information about variance from the market price in certain locations, highlighting areas of potential tax avoidance or illicit trade. Physical control of such locations requires rapid response teams, as implemented in the Philippines. To monitor the tax compliance of its taxpayers, the competent authority in the Philippines needs to understand the tobacco market; it must have information on brands, market segments and prices of products. This information enables authori- ties to estimate the impact of tax and price changes on consumer behaviour and revenues. Market data can be analysed as part of risk management and anti-fraud analysis to determine whom to investigate for noncompliance and when to do so. Sales data can be triangulated to validate other data sources, such as household surveys on prevalence. Market data and trends are also useful indicators for determining whether there is a case of oversupplying. See also section 3.4.3. • Consumer control – Involving the public via awareness campaigns has also been shown to be effective. Consumers have the right to be assured that the products available in the market are authentic and come from legitimate sources. Thus, it is in the consumers’ interest to understand and be able to verify that they are buying genuine products. The features of the fiscal marks on tobacco products should help consumers distinguish between genuine and illicit products. Some countries – Kenya, for example (see the case study in Box 3.9) – use a smartphone application to allow anyone to check both covert and overt features and to report any cigarettes with incorrect markings. Other countries, such as the Netherlands, have developed a smartphone application that allows anyone to report a suspected case of excise tax fraud. • Cross-check controls – Competent authorities should consider using multiple sources to obtain market data and determine if these data are consistent with tax declarations. VAT declarations can be used to verify that suppliers and purchasers of raw materials and final products are reporting the same amounts. Bank information can be used to verify both sides of transactions along the supply chain. Any discrepancy can alert the competent authority to conduct further investigation for possible illicit trade of tobacco or tax evasion. 142 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 13 Several different types of periodic audits and controls that can be carried out to increase compliance, including cost audits, transfer pricing audits, price and market monitoring, consumer controls and cross-check controls. 3.4.8 IMPORT AND EXPORT CONTROLS Parties to the Protocol should allow import and export of tobacco products and manufacturing equipment only by duly licensed natural persons or legal entities (Article 6.1). A well-known strategy used by fraudsters is to declare products for export so that no duties are due to the country of export. These products are subse- quently transported through other countries, using the in-transit regime that allows temporary suspension of duties until the goods arrive at their final destination. Before arriving at their final destination – where the excise duties would be due – the goods disappear or are lost while being diverted to the illegal supply chain. The goods may never leave the country, or they may be smuggled back into the country from which they were exported without declaring or paying duties. This risk of loss of revenue can be mitigated by requiring a guarantee or bond, which will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless the required fiscal marking (such as a tax stamp or export label) is affixed on the pack, according to the national law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Illicit tobacco trade could be decreased significantly if the various competent authorities that have jurisdiction over manufacturers and exporters of tobacco prod- ucts and manufacturing equipment would provide the competent authority at the destination with prior information when a shipment has been authorized and is about to take place. The information could include the name of the consignee, a description of the item shipped and the quantity. Also, the competent authority at the destination should inform the competent authority having jurisdiction over the shipper that a shipment was received along with the pertinent information relating to the shipment. A good IT system is also required for import and export. Electronic processing of prior-to-arrival manifest and import declarations is recommended. Most countries have implemented an online customs system to process import and export declara- tions, including all required data such as country of origin or export, description of merchandise, value, weight, cargo insurance, carrier, importer or exporter and broker identification, detailed tax duties to pay and final destination. The World Trade Organization (WTO) Trade Facilitation Agreement provides sev- eral tools for better controls, including collaboration between customs administrations, CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 143 exchange of information, use of non-invasive devices and prior-to-arrival import declaration (23). Non-invasive detection equipment at customs posts is highly capable of detect- ing contraband merchandise. The most common tools are X-ray scanners that are used for small parcels, containers, trucks and trains. Most modern ports have also implemented the use of X-ray scanners, and such technology is improving the speed of controls to as little as two minutes per container. Although the cost of scan- ners is declining, it remains inaccessibly high for countries with limited resources. Fortunately, scanners are often available for lease, making them accessible for tax administrations in those countries. Less sophisticated and less costly detection equipment includes endoscopes, mirrors, night vision equipment, cameras and automatic licence plate readers (33). A still less expensive alternative is the use of dogs, which can be trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied in order to reduce contra- band. These include the separation of processing operations from the sealed storage of taxed and untaxed products. Physical and direct control by officials of the excise authority during a part or the whole of an operation can be applied (for example, physical escort of the transit consignment from border to border by individual trucks or in a convoy, or application of radio or satellite tracking systems such as GPS-enabled devices to goods, conveyances, vehicles or containers). Control at borders is essential and should include integrated technology and cooperation with agencies at the border station. Frontline officers should be sup- ported by appropriate intelligence, guidance and supervision from management, as well as technical aids to enforcement. Within a country, mobile excise control units are helpful for verifying excisable goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. These operations require close coordination between police, border guards and other public services. Exports also require special attention, in particular if VAT and tax refunds are granted to the export of tobacco products. Validating the real exit in the declared amount is essential to avoid illegal re-entry to a territory and the improper refund of taxes. For any tax refund, an audit including tax credit information must be carried out. The audit may include the invoices for the whole chain involved in the export, including tobacco farmers, first processors, manufacturers, wholesalers, storage and transport. 14 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 14 To ensure control of import and export, it is recommended that only duly licensed persons and entities be allowed to import and export tobacco products and manufacturing equipment. 3.4.9 FREE ZONES AND TRANSHIPMENT POINTS The term “free zone” is very broad and can refer to a number of different types of areas. The Financial Action Task Force listed the following types in its 2010 report (59): free trade, export processing, enterprise, free ports, foreign trade, special economic zones and bonded warehouses. A number of these areas can include tobacco manufacturing and trade. By definition, controls such as regulation and oversight within free zones are less strict than in other areas. This can make them appealing to persons involved in illegal cigarette manufacturing or trade (17). In fact, illicit activities related to free zones (not limited to tobacco) are regularly documented by organizations that recognize the linkage. These activities include money laundering, tax evasion and trade in counterfeit goods or other illicit goods (60). A report from the European Parliament (61) referring to free ports in particular, mentioned that the motivation for using them included a “high degree of secrecy and deferral of import duty and indirect taxes”. The report even proposed the “urgent phasing out of free ports”. In the European Parliament report, free ports are free zones that function as (semi-) permanent storage areas for high-value goods. The Protocol includes a time-bound provision of effective controls on all manu- facturing and transactions of tobacco products in free zones (Article 12). Free zones are defined as a part of the territory of a Party where goods are considered to be outside the customs territory for import duties and taxes (Article 1.5). This is the same definition used in the International Convention on the Simplification and Harmonization of Customs Procedures (Revised Kyoto Convention) (62). Parties to the Protocol must implement effective controls in free zones within three years of entry into force of the Protocol. For countries not yet Parties to the Protocol, stringent controls of manufacturing and transactions involving tobacco products in free zones are an important component of an effective and efficient tax administration. One of the measures for dealing with free zones within the Protocol includes implementing “effective controls on all manufacturing of, and transactions in, tobacco and tobacco products, in free zones, by use of all relevant measures as provided in this Protocol”. As indicated in an Interpol report (17), a significant vulnerability of free zones is the fact that different economic operations (e.g. manufacturing, assembly, re-packaging and warehousing) take place outside the control of authorities. It is therefore essential for customs administrations to exercise their authority in free zones CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 145 to effectively identify and fight illicit trade in tobacco products. Relevant measures listed in the Protocol should be applied. These include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a track- ing and tracing regime. Removing exemptions on excise taxes is an additional way to increase control and remove incentives for using free zones as a means for tax evasion. Indonesia, for example, imposes excise taxes on cigarette manufacturing in its free trade zone.13 Parties to the Protocol shall also prohibit the intermingling of tobacco products with non-tobacco products in a single container or any other such similar transporta- tion unit when removed from free zones. Finally, each Party is to “adopt and apply control and verification measures to the international transit or transhipment of tobacco products and manufacturing equipment in conformity with the provisions of the Protocol”. Article 13 of the Protocol, which covers all duty-free sales of tobacco products, requires Parties to the WHO FCTC to consider prohibiting or restricting the sale to or import by international travellers of tax-free or duty-free tobacco products, as mentioned in Article 6 of the WHO FCTC. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect government revenues by creating a loophole in the tax structure (2). KEY TAKEAWAY 15 Customs administrations should exercise their authority in free zones to prevent different economic operations from taking place outside the control of authorities. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a tracking and tracing regime. 3.4.10 PROCEDURES AFTER DETECTING ILLICIT TRADE OF TOBACCO The procedures described in previous sections are intended to increase compliance and to prevent illicit trade. When smuggling or illicit trade is detected – through, for example, audits, tracking and tracing systems, verification of declarations or border control – actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. To deter further illegal behaviour, a comprehensive audit of everyone and everything involved in the illicit acts must also be carried out. Assets and vessels involved in the illicit activity can be seized, and financial accounts can be frozen. Some countries, including the United Kingdom, Canada and Chile, have also adopted a strategy known as “follow the 13 Indonesian Ministry of Finance, personal communication, January 2020. 146 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N money” to obtain more information on those who finance illicit trade. This strategy is intended to have a further-reaching effect by targeting those who finance the transport, production and storage of illicit products. The United Kingdom, Canada and Chile all created special teams tasked with identifying and targeting the offenders. Furthermore, Article 18 of the Protocol provides for the confiscation and destruction of tobacco, tobacco products and manufacturing equipment. One of the difficulties faced by competent authorities in exercising the authority to seize and forfeit products and/or equipment used in the manufacture or distribution of tobacco products is the cost of keeping or storing the goods and/or machinery before destruction. Thus, the law should also provide for a mechanism and timetable for the disposal and/or destruction of seized and forfeited goods or machinery, while prescribing a mechanism by which these properties can still be presented as admissible evidence in a judicial proceeding. Boxes 3.12 and 3.13 provide examples of successful efforts to combat illicit trade in the United Kingdom and Indonesia. KEY TAKEAWAY 16 As soon as smuggling or illicit trade in tobacco products is detected, actions such as collecting taxes and seizing and destroying smuggled and/or illicit tobacco must be taken. Box 3.12 The United Kingdom’s experience in fighting illicit trade in tobacco products In 2000, illicit cigarettes accounted for 22% of the cigarette market in the Unit- ed Kingdom. To deal with the problem, Her Majesty’s Customs and Excise14 implemented a major anti-smuggling effort. The strategy was refreshed with ad- ditional resources and measures in 2011 and reviewed in 2015. The result was a steady decline in the illicit cigarette market to 10% by 2013/2014. The measures taken were comprehensive and included hiring 1 000 new customs officers and investigators. In addition, tobacco supply chain legislation was introduced, aimed at discouraging tobacco manufacturers from facilitating smuggling. Tougher sanctions included increased fines of up to £5 million levied on a manu- facturer, criminal prosecution with sentences up to seven years, confiscation of assets as part of the proceeds of the crime, payment of duty on the confiscated goods plus penalties up to 100% of the duty, prohibition of the sale of tobacco products for 14 By the time of the renewed strategy, the respective bodies were the HMRC and the United Kingdom Border Agency. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 147 up to six months, unlimited fines for selling tobacco without the United Kingdom duty-paid fiscal mark after 13 March 2015, referral for withdrawal of the transporter’s licence, use of immigration sanctions to refuse entry to the United Kingdom for tobacco smugglers and civil action, including bankruptcy. The cost of these measures was £209 million over the first three years of the program and around £100 million annually by 2008/2009. This figure covers only HMRC and excludes any costing of the United Kingdom Border Agency. In 2013/2014, tobacco tax revenues were £9.5 billion. After a review of the strategy in 2015, the controls on raw tobacco were strength- ened by the introduction of an approval system in 2017. Anyone who manufactures, purchases, acquires, owns or is in the possession of a tobacco products manufacturing machine must be licensed with customs as of 1 August 2018 (63). The United Kingdom ratified the Protocol on 27 June 2018. It was the 40th country to ratify, which was the trigger point for the Protocol to enter into force. Fig. 3.4 Estimate of the illicit cigarette market and United Kingdom tax-paid consumption Sources: (16, 64). Bi lli on c ig ar et te s UK tax paid consumption Illicit market 20 00 -01 20 01 -02 20 02 -03 20 03 -04 20 04 -05 20 05 -06 20 06 -07 20 07 -08 20 08 -09 20 09 -10 20 10 -11 20 11 -12 20 12 -13 20 13 -14 20 14 -15 20 15 -16 20 16 -17 20 17 -18 0 10 20 30 40 50 60 148 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.13 A success story: Indonesia reduced illegal cigarettes from 12% to 3% of the market In some countries, most of the illegal cigarettes are imported, but in Indonesia most of them are produced within the country by unregistered manufacturers that are usually home-based and relatively small. In several territories, specifically on the Island of Java, illegal cigarettes have been produced for generations. This practice is supported by the availability of raw tobacco materials and cloves, as well as cheap labour costs, especially for female workers. Indonesia also produces cigarette products that are not available in most other countries. These products – known as handmade clove cigarettes (sigaret kretek tangan) – contain cloves, and the production process covers blending, rolling and packing. Parts of the process are done by hand, and 99% of the labourers are women. The strategy for combatting the illegal cigarette trade in Indonesia is divided into two main parts: preventive actions and responsive actions. Indonesia’s success in effectively tackling illicit trade is attributed to the following key factors. Monitoring and surveillance Preventive actions consist of administrative measures – such as issuance of permits and the excise stamp purchasing mechanism – that use risk management by optimizing the Excise Service Information System (ExSis). With this IT system, the Directorate General of Customs and Excise (DGCE) can oversee both daily transactions and daily production from factories. When information of suspicious activities is obtained, DGCE can suspend the purchase of excise stamps. The efforts to fight the illegal cigarette trade also invite stakeholders to be involved by supplying information regarding high-risk areas and regional governments. Strategic communications and community involvement DGCE continuously disseminates information and conducts public education to fight illegal cigarettes. These efforts are conducted every year, using a special campaign slogan. In 2019, the slogan was “Gempur Rokok Ilegal” (“Fight Illegal Cigarettes”). Key performance indicators for DGCE units and offices Parallel with the above-mentioned preventive actions, DGCE also continuously conducts responsive actions: enforcement, investigation and audit activities in cigarette factories. Enforcement activities are planned and measured by consider- ing the limited human resources and the large scale of the monitored territories. To demonstrate the effectiveness of administrative and enforcement measures in curbing the trade of illegal cigarettes, both types of activities are translated into key CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 149 performance indicators for all DGCE working units and offices, including regional offices and personnel. Use of technology and intelligence Coordination between different DGCE offices responsible for monitoring the produc- tion and marketing of cigarettes is maintained by sophisticated IT applications that enable efficient distribution of information and investigation activities. The applica- tions are the Customs Intelligence and Tactical Centre for data and analysis and the Centre for Command and Control (Pusat Komando dan Pengendalian/Puskodal) for ensuring that sea patrols work effectively and efficiently. Independent evaluation To evaluate the efforts and activities to reduce the circulation of illegal cigarettes in Indonesia (e.g. cigarettes without stamps, with fake stamps or with used stamps), in 2016, the government commissioned the University of Gadjah Mada in Yogyakarta to conduct a survey using a stratified random sampling method. To maintain objectiv- ity and independence, an independent body from this well-known university was appointed to conduct the survey. The survey results showed that the level of illegal cigarette circulation in Indonesia was 12.1% of total consumption. In 2018, the DGCE commissioned the University of Gadjah Mada to conduct another survey. Results showed that circulation of illegal cigarettes had been reduced to 7.0%. In 2019, using the same method the university used, DGCE conducted a survey that showed a reduction to 3.0%. Fig. 3.5 shows the results of the surveys. Fig. 3.5 Share of illicit trade in total cigarette consumption in Indonesia, 2016–2019 Source: Customs and excise department, Ministry of Finance, Indonesia, personal communication, 2020. 2016 2017 2018 2019 12.1% 10. 9% 7% 3% 150 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The results of the actions taken can be used as feedback for DGCE in developing excise policies for both service and enforcement, including policies for excise tariffs. DGCE will provide recommendations for an optimum policy format for controlling consumption, maintaining labour protections, optimizing revenue and, most impor- tantly, constantly reducing the consumption of illegal cigarettes on a national level. 3.4.11 PENALTIES Penalties and sanctions must be sufficient to deter illegal activities. Otherwise, finan- cial penalties may simply be paid as a cost of doing business while the illegal activity continues. The Protocol specifies commitments for Parties and provides information on best practices for non-Parties. Article 14.1 in Part IV of the Protocol requires each Party to establish unlawful activities, including manufacturing, wholesaling, broker- ing, selling, transporting, distributing, storing, shipping and importing or exporting tobacco products or manufacturing equipment without the payment of applicable duties or taxes or without using fiscal stamps or other required markings or labels. Articles 14.2 and 15 of the Protocol mandate Parties to determine which of the types of unlawful conduct set out in Article 14.1 shall be criminal offences. Parties must adopt legislative and other measures to give effect to such determinations, as well as to define whether the liability for committing illicit trade in tobacco is a criminal, civil or administrative offence. Article 17 further provides that the Parties shall consider adopting measures as needed to authorize competent authorities to levy penalties in an amount proportionate to lost taxes and duties resulting from the commission of illicit trade. Box 3.14 provides a case study of how Colombia used penalties to fight illicit trade. Box 3.14 The use of penalties to combat illicit trade in Colombia In 2017, the specific tax on cigarettes in Colombia was doubled, increasing from COL$ 700 per pack in 2016 to COL$ 1 400 in 2017. The tax rate was tripled from 2016 levels in 2018, reaching COL$ 2 100 per pack. A provision was added to increase taxes annually after 2018 at the rate of inflation plus 4%. In 2015, before the tax increase, Law 1762 introduced a number of measures to fight illicit trade more effectively. The length of imprisonment for dealing in contraband cigarettes was increased from 3–5 years to 4–12 years. Moreover, government officials who facilitate illicit trade – or anyone involved in transporting or retail sales of illicit tobacco – face similar prison terms. The law allows vehicles used for smuggling to be confiscated, and penalties were increased for illicit trade that is conducted through areas such as special economic zones. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 151 Under the law, illicit trade is considered to be a source of money laundering, which means that the Financial Intelligence Unit can use the same methods it uses to investigate other illegal financial activities. This practice is not common. The law specifically created new sanctions related to alcohol and tobacco excise tax evasion, including the seizure of goods, fines, closure of retail outlets and the suspension or cancellation of licences, authorizations or registries. Arrests and seizures have increased under the new law. Indeed, since its enactment, law authorities reported that between 2016 and 2018, five criminal organizations were dismantled, 53 individuals were apprehended and 72 assets were confiscated. In addition, 2 236 individuals were apprehended and 503 vehicles transporting smuggled goods were confiscated, as transport of such goods is now also considered a crime under the law. More importantly, thanks to the large tax increases, consumption decreased while revenues increased substantially in 2017 and 2018 (see Fig. 3.6). It is estimated that illicit trade in cigarettes in five Colombian cities in 2016 constituted 3.5% of total consumption, a much lower estimate than the industry data suggest. In 2017, after nine months of the tax increase implementation, a similar study found that illicit cigarettes remained low, at 6.4% of total consumption. Fig. 3.6 Packs sold and tobacco tax revenue before and after the tax increase in Colombia, 2016–2018 Sources: (65–67 and Ministry of Finance, Colombia (Direccion de Apoyo Fiscal), personal communication, 2020). 2016 2017 2018 Cigarette sales, million packs Cigarette excise revenue, million US$ current 0 100 200 300 400 500 600 700 800 673.7 521 446.4 195.1 301.2 386.2 33.7% reduction 97.9% increase 152 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N For consumers in possession of illicit tobacco, the minimum penalty should be confiscation and destruction of the illicit tobacco products found in their possession, with payment required for the unpaid tax and duties on those products. In the state of California in the United States, it is illegal to possess a tobacco product on which taxes are due and not yet paid. The burden of proving that taxes have been paid is on those who have the products in their possession. The provi- sion is enforced by the California Department of Tax and Fee Administration and local law enforcement agencies. A violation is a misdemeanour, with a maximum fine of US$ 5 000 and/or up to one year in prison. Illegal packages are subject to seizure and forfeiture. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are using this type of legislation to address illicit trade of tobacco. Asset confiscation and increased penalties for involvement in illicit trade are becoming more common as well. Withholding or even confiscation of trucks involved in smuggling is also common in several countries. KEY TAKEAWAY 17 Penalties and sanctions imposed should be sufficient to deter illegal tobacco trade activities. Penalties should be levied in amounts proportionate to lost taxes and duties resulting from illicit trade. 3.5 TAX ADMINISTRATION OF OTHER TOBACCO PRODUCTS In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. However, there is a lack of standard- ization of other products and sometimes large informal markets. For example, it is estimated that two thirds of waterpipe tobacco in the EU is non-duty-paid (19). Other tobacco products – such as bidis in South-East Asia, waterpipe tobacco in the Eastern Mediterranean region and snus in Sweden – are considered part of a country’s traditions. This sometimes leads to situations where governments are hesitant to strongly regulate and tax these products. Some products, such as kreteks (clove cigarettes) in Indonesia and bidis in India and Bangladesh, are mainly sold in one market. Other tobacco products are more likely to be produced by hand on a small scale, making it difficult to detect and collect taxes on them. The same applies to RYO tobacco, which can be produced on a small scale by hand or with the use of small machinery. The trade in raw tobacco and small-scale home production of tobacco often take place outside of monitoring and control systems (19). As mentioned in section 3.3.1, countries have found various solutions to address this problem, including prior approval for purchase or sale of raw materials and reg- istering, authorizing or licensing of all operators and growers that handle raw tobacco. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 153 3.5.1 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In principle, adding a product to an existing tax framework is not likely to impose significant costs. It is reasonable to expect that challenges similar to those faced in dealing with conventional tobacco products will be faced in the collection of taxes on new products, as market players will attempt to use loopholes in tax regulation to avoid or evade taxes whenever possible. However, new challenges are expected to arise when those new products involve rapidly changing technology and where their market dynamics are widely unknown. Furthermore, taxation of new tobacco products may require additional capacity, as a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. 3.5.2 HEATED TOBACCO PRODUCTS (HTPs) Many countries apply a specific excise on tobacco products according to tobacco weight (see Table 2.4 in Chapter 2). With HTPs, this is likely to impose a challenge, since assessing the content of tobacco in a heated tobacco stick will be an additional burden. From a tax administration perspective, it will be easier for authorities to apply taxes per stick or per unit, as is done for cigarettes. 3.5.3 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS) PRODUCTS Some countries tax only nicotine-containing e-liquids while others tax both nicotine- and non-nicotine-containing e-liquids. Taxing only nicotine-containing e-liquids re- quires laboratory capacity to detect the presence of nicotine (see Table 2.5 of Chapter 2). Self-declarations by industry are not sufficient, since some e-liquids labelled as nicotine-free have been found to contain nicotine (see section 2.4.2). Therefore, it is simpler to tax both nicotine- and non-nicotine-containing e-liquids. One likely challenge of taxing all e-liquids will be the capacity to detect and differentiate whether the e-liquids used in ENDS/ENNDS are falsely declared as being for other purposes at the import and manufacturing levels. More information on advantages and disadvantages of different excise tax policies is given in Table 2.6 of Chapter 2. The challenge in taxing the other components of ENDS/ENNDS products is their diversity (see section 2.4.2 of Chapter 2) and the possibility that some parts may be used for other purposes (e.g. in batteries). As indicated earlier, rapidly changing technology and the lack of control and knowledge of the market make taxation of ENDS/ENNDS devices challenging. It may be for this reason that the majority of countries that tax those products address only the e-liquids. When applying a tax on these newer products, countries should be aware that many customers buy their products online. It is therefore recommended that countries 154 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N deciding to tax these products draw up a proper implementation plan, including how taxes will be collected on imported products and online sales. Online cross- border sales of tobacco products are not permitted in all countries. Several EU countries have banned such sales, which include online sales from retail outlets to consumers in another country. Of course, such bans makes sense only if there is also capacity to enforce them. As in the case of taxing tobacco products, the following actions will be important to more effectively impose taxes on these products: 1. implementing strong enforcement mechanisms such as licensing, recordkeep- ing and control of the supply chain, which can include but is not limited to: a) imposing strict licensing of retailers, importers and manufacturers; ideally, licensing of all those involved in the supply chain and developing a tracking and tracing regime for ENDS/ENNDS products and HTPs (to share cost, this can be done in tandem with the system developed for cigarettes); b) exercising the right to set the frequency and type of audits or controls; c) exercising the right to confiscate goods; and d) imposing sanctions such as penalties, fines and/or withdrawal of licences (if applicable) if legislation is not respected. Specific to ENDS/ENNDS products: 2. implementing highly consequential sanctions for producers who declare nicotine-containing e-liquids as “non-nicotine-containing”; and 3. requiring a fee (contribution to the costs) for laboratory tests when a new product is brought on the market or when there are significant modifications to an existing one. More information on the policy options to apply excise taxes on ENDS/ENNDS is given in Chapter 2, section 2.4.2, Table 2.6. KEY TAKEAWAY 18 In principle, the administration of taxes on new and emerging nicotine products and tobacco products should be similar to that for cigarettes. Due to the lack of standardization of these products, however, a rapid and constantly evolving understanding of them and their supply chain will be required to achieve effective and efficient administration of taxes. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 155 3.6 THE BROADER ELEMENTS OF A GOOD TAX SYSTEM 3.6.1 PROPER RESOURCING OF COMPETENT AUTHORITIES In addition to legal tools and a legal basis on which to act and enforce, the authori- ties in charge of implementing excise tax laws should be provided with sufficient resources to hire the necessary staff to properly implement and enforce them. The necessary staffing could encompass multiple agencies and will often require coop- eration between agencies, since some aspects – such as regulation, licensing and border control – may be performed by agencies other than the competent authority. Staff of competent authorities need the necessary tools, equipment, training and supplies to carry out their functions. This requirement includes the means to build or purchase and maintain a software system that will allow taxpayers to submit required information electronically. Electronic filing has benefits for both taxpayers and authorities. It minimizes the compliance cost for taxpayers and can therefore support voluntary compliance (68). To identify risks of noncompliance, the software system should offer competent authorities the ability to analyse the data submitted by taxpayers and cross-check it with data from other taxes – such as VAT – and third-party sources, such as banks and household surveys. Another option for authorities is to make more efficient use of existing resources. For example, an authority could optimize the risk management system by switching to a risk-based approach: resources could be saved by auditing taxpayers who are more likely to be noncompliant based on risk analysis rather than auditing all of them. Other problems that challenge the effective functioning of a competent authority are lack of a coherent strategy and problems with professionalism related to lack of training or corruption (69). Having a strategy avoids directing resources towards less- important areas. The strategy should always be aligned with the objectives, so that com- petent authorities can identify which steps they should take and in which order they should take them to reach these objectives. A strategy is indispensable to prioritizing and organizing resources so that identified issues or risks can be addressed efficiently. 3.6.2 CORRUPTION Competent authorities should implement tax laws with integrity and have strict rules and regulations for detecting corruption. Strict rules and regulations should also be in place for the punishment of both agency personnel and taxpayers who engage in corrupt practices. Corruption within a competent authority results in the improper monitoring of tax compliance and is one of the causes of the proliferation of illicit trade in tobacco products. It also erodes confidence in competent authorities and ultimately in governments overall. In addition to effective laws and regulation, strong internal audits covering prevention, investigation and sanctions should be 156 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implemented. To improve prevention, a risk map should be created that highlights areas of misconduct and possible leakages. An action plan to update controls should be established to improve areas of weakness detected in procedures and systems. The audits should also be scheduled regularly. Prepared internal auditors with pow- ers to conduct investigations are necessary. Sanctions for corruption, including administrative sanctions and criminal prosecution, must be strong. 3.6.3 A STRONG JUDICIARY The judicial system should be honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. The appeals process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially in the context of illicit trade. KEY TAKEAWAY 19 Broader elements of a good tax system include (1) proper resourcing of competent authorities to hire staff and obtain necessary equipment and systems, (2) strict rules and regulations to detect and punish corruption among both agency personnel and taxpayers and (3) ensuring that the judicial system is honest and independent, with disputes being solved as quickly as possible. 3.7 CONCLUSIONS Policies are more effective if they are properly implemented and enforced. Com- petent authorities have a key role in the achievement of financial and public health objectives of excise taxes. Given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade, this chapter draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (i.e. the Protocol). The Protocol provides a blueprint of measures to address the problem of illicit trade and can be used as a model even by countries that are not Parties to it. Qualities of an effective and efficient tax administration include institutional ar- rangements where roles and responsibilities of competent authorities are clearly defined to avoid overlap and voids. Additionally, effective collaboration among relevant bodies must be facilitated. At the national level, within any organizational arrangement, it is vital that agencies cooperate and exchange information and that their competences find their basis in law. A legal basis for exchange or access to information between government bodies should be ensured. At the international level, especially for border control, the role of customs is key, and access to international CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 157 cooperation agreements such as the Protocol is very useful. An organizational tax administration structure must include a system of performance evaluation and accountability through pre-defined key indicators. To ensure compliance, the accuracy of information for the tax compliance cycle is key, including clear and straightforward taxpayer registration and licensing, declara- tion, recordkeeping, warehousing, distribution, collection and tax refund processes. • Licensing is a powerful tool for obtaining information and securing the supply chain of tobacco products. Ideally, all persons involved in the growing of tobacco and the retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. • Collecting as much information as possible on the business of tobacco and recording all transactions are key to reducing tax evasion, but this may be burdensome for authorities. The use of IT for periodic tax declarations, ac- counting, inventory and financial information is critical for obtaining accurate information and can help decrease the cost of the whole reporting system. • Recordkeeping should be ensured. All persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of those products. • Maintaining a system of authorization for warehousing allows the authori- ties to carry out controls in production and storage facilities to ensure that taxes are paid. Ideally, bonded warehouses should be eliminated from the supply chain. • Duty suspension – which is often applied during the producing, processing, holding, receiving and dispatching of excise goods – should be granted only if strict criteria are met (e.g. for granting authorization, warehouse pre-authoriza- tion visits, adequate stock control measures, checking the origin of excise prod- ucts and the entire production process and coding and marking of products). • To limit the number of taxpayers a competent authority has to manage, tax collection should take place close to the point of production and import. • Refunds for VAT, excise taxes and customs duties are common in most coun- tries, under the principle that taxes are not exported. The refund process must be closely monitored to avoid opportunities for tax evasion. Control and enforcement – key components of tax administration – include a number of measures to secure the supply chain: licensing and due diligence, fiscal markings, 158 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N track and trace, anti-forestalling measures, audits and controls, import and export control and attention to free zones and transhipment points. Control and enforce- ment need to be included as pillars in the strategic plan of the tax administration. Enforcement and control plans must be designed to define the activities and taxpayers that are subject to enforcement and to allocate staffing, auditing, infrastructure and IT resources. Targets must be defined, including the number of interventions and any additional collection or reduction of tax evasion. This includes choosing interventions for those who have a higher probability of noncompliance (the risk- based approach). In the tobacco supply chain, import, export and transfers to and from warehouses may be areas at greater risk of noncompliance. • Licensing provides timely and accurate data that can serve as the basis for audits, since it identifies and controls legitimate operators. The process of licensing control must be carried out and updated periodically – in particular, by controlling the validity of bonds or guarantees and the proper functioning of the required systems and recordkeeping. Where licences are required, the law should include a provision that disallows purchases from unlicensed suppliers or sales to unlicensed purchasers. This means that both suppliers and purchasers will need to verify those with whom they are doing business. This will substantially help to reduce the burden of proof for authorities. In addition, to maintain a high level of control, the validity of licences should be limited in time, making renewals or reapplication required. • Another important measure for controlling and monitoring production and import of tobacco products is the use of fiscal markings (e.g. tax stamps). In addition to increasing compliance with tax laws, fiscal markings can help distinguish between genuine and illicit tobacco products. The use of fis- cal marks enables both the competent authority and the public to monitor whether the taxes on tobacco products have been properly paid. In addition to locally produced and imported products, tobacco products for export should also be required to be marked, but with an indication that they are for export. Requiring a standard package size can facilitate the application of fiscal markings. To lower the chance that fraudsters attempt to re-use fis- cal markings (in particular, stamps) the marking should be applied to each pack of cigarettes (and other tobacco products) before the pack is wrapped with cellophane. Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Fiscal markings should include several security features to make them more difficult to counterfeit. These can include overt, covert, semi-covert and/or forensic features. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 159 • Tracking and tracing systems assist authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as in monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to provide authorities with information on all transactions throughout the entire tobacco product supply chain until duties are paid or other obligations are discharged. Any tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it becomes possible to unambiguously register that product’s movements. A good tracking and tracing system enables the government to properly monitor the supply chain, improve its ability to ensure collection of the proper duties and taxes, authenticate whether the identification marking is genuine and matches the product, improve its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. To reduce the financial burden of implementing such a system, jurisdictions could require the tobacco industry to bear the cost. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, and governments should ensure that the system is independent from the tobacco industry. While the objectives of fiscal markings and tracking and tracing systems are different, stamps increasingly contain tracking and tracing features. • Implementing legal measures to prevent forestalling can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. Forestalling, stockpiling or front-loading occur when manufacturers or im- porters increase their tax-paid stock or oversupply the market by increasing production or imports before a tax increase in order to pay the previous lower rate. • Periodic audits and controls can be implemented to increase compliance. These include cost audits, transfer price audits, price and market monitoring, consumer controls and cross-check controls. • Import and export of tobacco products and manufacturing equipment should be allowed only for duly licensed natural persons or legal entities. The risk of loss of revenue can be mitigated by requiring a guarantee or bond that will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless required fiscal markings (such as tax stamps or export labels) are affixed on the pack, according to the law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Exchange of information between jurisdictions on the movement of goods can also reduce the risk of evasion. 160 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Non-invasive detection equipment (such as X-ray scanners) can be used at customs posts to detect contraband merchandise. A cheaper alternative is the use of dogs that are trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied to reduce contraband. Such measures include the separation of processing operations from the sealed storage of taxed and untaxed products. Within a country, mobile excise control units are helpful in verifying excis- able goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. Physical control operations require close coordination between police, border guards and other public services. • Controls such as regulation and oversight are usually less strict in free zones and transhipment points. This can make free zones appealing to persons involved in illegal cigarette manufacturing or trade. Customs administrations should exercise their authority in free zones to effectively identify and fight illicit trade in tobacco products. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing tracking and tracing regimes and removing exemptions from excise taxes. Other actions include the prohibition of intermingling of tobacco products with non-tobacco products in a single container or other similar transportation unit when the products are removed from free zones. Sale of tax-free or duty-free tobacco products to international travellers should be prohibited, as these sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect govern- ment revenues by creating a loophole in the tax structure. Procedures after detection of illicit trade of tobacco products should be clearly defined. If smuggling or illicit trade is detected through audits, tracking and tracing systems, verification of declarations or border control, actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. It is also important that penalties and sanctions be sufficient to deter illegal activities. Low financial penalties may simply be paid as a cost of doing business while the illegal activity continues. The minimum penalty for consumers in possession of illicit tobacco products should be confiscation and destruction of the products found in their possession and required payment for the unpaid tax and duties on those products. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are taking advantage of this type of legislation and using it to address illicit trade of tobacco as well. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 161 In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. The challenge for taxation of other products includes the lack of standardization of those products and sometimes large informal markets. Knowledge of the product and the supply chain greatly help to facilitate effective tax administration. The trade in raw tobacco and small-scale home production of RYO and other products such as bidis often takes place outside of monitoring and control systems. The best way to address this challenge is to enforce prior approval for purchase or sale of raw materials and a requirement to register, obtain an authorization or license all operators and growers handling raw tobacco. In principle, adding new and emerging nicotine and tobacco products to an exist- ing tax framework is not expected to impose significant costs. It is reasonable to expect that similar challenges will be faced in the collection of taxes on these newer products, as market players will attempt to use the current loopholes in tax regulation to avoid or evade taxes on these products whenever possible. However, challenges are expected to arise, as newer products involve rapidly changing technology, and their market dynamics are widely unknown. Furthermore, additional capacity may be required, since a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. Because the newer nicotine and tobacco products are widely purchased online, countries deciding to tax these products should draw up a proper implementation plan that includes rules on how taxes will be collected on imported products and online sales. Online cross-border sales are not permitted in some countries. The elements of a good tax system include (1) proper resourcing of competent authorities sufficient for hiring the necessary staff to properly implement and en- force excise tax laws; (2) implementation of tax laws with integrity and with strict rules and regulations to detect corruption and for the punishment of both agency personnel and taxpayers who are engaged in corrupt practices; and (3) ensuring that the judicial system is honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. 162 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N REFERENCES 1. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 2. WHO FCTC guidelines for implementation of Article 6 Price and tax measures to reduce the demand for tobacco. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/treaty_instruments/ Guidelines_Article_6_English.pdf?ua=1, accessed 5 October 2020). 3. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 4. Joossens L, Raw M. From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–234 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 5 October 2020). 5. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 6. Reyes IP. Measures to control the tobacco supply chain in the ASEAN: a review of excise tax stamps, track and trace mechanisms, and licensing systems in the ASEAN region. Bangkok: Southeast Asia Tobacco Control Alliance; 2014 (https://seatca.org/dmdocuments/SEATCA%20MEASURE%20BOOK%20. pdf, accessed 5 October 2020). 7. Italy’s tax administration: a review of institutional and governance aspects. Paris: Organisation for Economic Co-operation and Development; 2016 (https://www.oecd.org/tax/administration/italy- tax-administration-a-review-of-institutional-and-governance-aspects.pdf, accessed 5 October 2020). 8. Crandall W, Gavin E, Masters A. ISORA 2016: understanding revenue administration. Washington (DC): International Monetary Fund; 2019 (Departmental Paper Series No. 19/05; (https://www.imf.org/en/ Publications/Departmental-Papers-Policy-Papers/Issues/2019/03/07/ISORA-2016-Understanding- Revenue-Administration-46337, accessed 5 October 2020). 9. Dutch customs in 2018. Heerlen: Customs Administration of the Netherlands; 2019 (https://over-ons. belastingdienst.nl/wp-content/uploads/2019/04/Dutch-Customs-in-2018.pdf, accessed 8 October 2020). 10. ESAs transform the way competent authorities cooperate with each other on AML/CFT matters [press statement]. European Banking Authority; 16 December 2019 (https://eba.europa.eu/esas-transform- way-competent-authorities-cooperate-each-other-amlcft-matters, accessed 5 October 2020). 11. Illicit trade report 2013. Brussels: World Customs Organization; 2014 (http://www.wcoomd.org/-/ media/wco/public/global/pdf/topics/enforcement-and-compliance/activities-and-programmes/ illicit-trade-report/itr_2013_en.pdf?la=en, accessed 5 October 2020). 12. Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation and repealing Directive 77/799/EEC. Brussels: OJEU; 2011. (L 64/1; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:32011L0016&from=en, accessed 29 January 2021). 13. 439 stolen cars seized in 12 days as part of Frontex-led operation. Frontex News Release; 14 October 2019 (https://frontex.europa.eu/media-centre/news-release/439-stolen-cars-seized-in-12-days-as- part-of-frontex-led-operation-vvvirI, accessed 5 October 2020). 14. Focus on: the illicit trafficking of counterfeit goods and transnational organized crime. Vienna: United Nations Office on Drugs and Crime; 2014 (https://www.unodc.org/documents/counterfeit/FocusSheet/ Counterfeit_focussheet_EN_HIRES.pdf, accessed 5 October 2020). 15. Crandall W. Revenue administration: performance measurement in tax administration. Washington (DC): International Monetary Fund; 2010 (Technical Notes and Manuals 10/11; https://www.imf.org/ external/pubs/ft/tnm/2010/tnm1011.pdf, accessed 7 October 2020). 16. Measuring tax gaps 2019 edition – tax gap estimates for 2017-18. London: HM Revenue and Customs; 2019 (https://webarchive.nationalarchives.gov.uk/20200701215139/https://www.gov.uk/government/ statistics/measuring-tax-gaps, accessed 29 January 2021). 17. Countering illicit trade in tobacco products: a guide for policy-makers. Lyon: INTERPOL; 2014 (Legal handbook series; https://www.aph.gov.au/DocumentStore.ashx?id=704d9168-2aa7-4df4-bb81- CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 163 0fe1f357dfa6&subId=408832, accessed 29 January 2021). 18. Resolução- RDC No. 226 de 30 de Abril de 2018 [Resolution of April 30, 2018]. Ministério da Saúde/ Agência Nacional de Vigilância Sanitária/Diretoria Colegiada; 2018 (in Portuguese) (https://www.in.gov. br/materia/-/asset_publisher/Kujrw0TZC2Mb/content/id/12527411/do1-2018-05-02-resolucao-rdc- n-226-de-30-de-abril-de-2018-12527407, accessed 17 February 2021). 19. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 20. Preece R. Addressing the illicit flow of tobacco products in Australia. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 21. Excise on tobacco [website]. Canberra: Australian taxation office; 2019 (https://www.ato.gov.au/ business/excise-on-tobacco/, accessed 5 February 2021). 22. Safe Framework of Standards. Brussels: World Customs Organization; 2018 (http://www.wcoomd. org/-/media/wco/public/global/pdf/topics/facilitation/instruments-and-tools/tools/safe-package/ safe-framework-of-standards.pdf?la=en, accessed 29 January 2021). 23. The Trade Facilitation Agreement. Geneva: World Trade Organization; 2017 (https://www.wto.org/ english/docs_e/legal_e/tfa-nov14_e.htm, accessed 29 January 2021). 24. Authorised economic operator (AEO) [website]. Kildare: Irish Tax and Customs; 2020 (https://www. revenue.ie/en/customs-traders-and-agents/authorised-economic-operators/index.aspx, accessed 7 October 2020). 25. Performance outcome area 1: integrity of the registered taxpayer base. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www. tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 26. Commission implementing regulation (EU) 2016/323 of 24 February 2016 laying down detailed rules on cooperation and exchange of information between member states regarding goods under excise duty suspension pursuant to Council Regulation (EU) No 389/2012. Brussels: OJEU; 2018 (http://data. europa.eu/eli/reg_impl/2016/323/2018-02-15, accessed 7 October 2020). 27. Council Directive 2008/118/EC concerning the general arrangement for excise duty repealing Directive 92/12/EEC. Brussels: OJEU; 2009 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:3 2008L0118&from=en, accessed 29 January 2021). 28. Our strategy. London: HM Revenue & Customs, 2017 (https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/629941/HMRC-Strategy.pdf, accessed 31 January 2021). 29. Strategic plan FY2018-2019. Washington (DC): Internal Revenue Service, 2017 (https://www.irs.gov/ pub/irs-pdf/p3744.pdf, accessed 31 January 2021). 30. Performance outcome area 2: effective risk management. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/ assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 31. Tax administration 2017: comparative information on OECD and other advanced and emerging economies. Paris: Organisation for Economic Co-Operation and Development; 2017. 32. Compliance risk management: managing and improving tax compliance. Paris: Organisation for Economic Co-Operation and Development; 2017 (https://www.oecd.org/tax/administration/33818656. pdf, accessed 7 October 2020). 33. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 34. Evans P. U.K. Fines British American Tobacco for Oversupplying Cigarettes in Belgium. The Wall Street Journal. 13 November 2014 (https://www.wsj.com/articles/u-k-fines-british-american-tobacco-for- oversupplying-cigarettes-in-belgium-1415873470, accessed 31 January 2021). 35. Doward J. BAT fined for oversupplying tobacco in low-tax European jurisdictions. The Guardian. 164 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 16 November 2014 (https://www.theguardian.com/business/2014/nov/16/bat-fined-for-oversupplying- tobacco-in-low-tax-european-jurisdictions, accessed 31 January 2021). 36. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021) 37. Cour des comptes. Accises sur les produits du tabac, Bruxelles [Court of Auditors, Excise duties on tobacco products, Brussels]. Bruxelles: Cour des comptes; 2015 (in French) (https://www.ccrek.be/ docs/2015_27_AccisesProduitsDuTabac.pdf, accessed 7 October 2020). 38. Chaloupka FJ, Edwards SM, Ross H, Diaz M, Kurti M, Xu X, et al. Preventing and reducing illicit tobacco trade in the United States. Atlanta: Centers for Disease Control and Prevention; 2015 (https://www. cdc.gov/tobacco/stateandcommunity/pdfs/illicit-trade-report-121815-508tagged.pdf, accessed 29 January 2021). 39. The Tax Stamp Forum 2015. Hexham: Reconnaissance International (https://10times.com/tax-stamp- forum-miami, accessed 10 September 2020). 40. Tax stamps: a technical study and market report. London: Reconnaissance international; 2012. 41. ЕАНС/2013/Health/ll concerning the provision of an analysis and feasibility assessment regarding EU systems for tracking and tracing of tobacco products and for security features. Brussels: European Commission, 2015 (https://ec.europa.eu/health/sites/health/files/tobacco/docs/2015_tpd_tracking_ tracing_frep_en.pdf, accessed 29 January 2021). 42. Ross H. Controlling illicit tobacco trade: international experience. Economic research informing tobacco control policy. Cape Town: University of Cape Town; 2015 (https://tobacconomics.org/ uploads/misc/2015/05/Ross_International_experience_05.28.15.pdf, accessed 18 December 2020). 43. Security and resilience — authenticity, integrity and trust for products and documents — guidelines for the content, security, issuance and examination of excise tax stamps. Geneva: International Organization for Standardization; 2018. 44. Borkowski F, Twomey C. European Union: an update on EU policies. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 45. Quality management systems — Fundamentals and vocabulary. ISO 9001:2015(en). Geneva: International Organization for Standardization; 2015 (https://www.iso.org/obp/ui/#iso:std:iso:9001:ed- 5:v1:en, accessed 16 February 2021). 46. Protocol to Eliminate Illicit Trade in Tobacco Products: questions & answers. Geneva: World Health Organization; 2019 (https://www.who.int/fctc/protocol/faq/en/, accessed 7 October 2020). 47. Ross H, Eads M, Yates M. Why governments cannot afford Codentify to support their tracking and tracing solutions. Tob Control. 2018;27(6):706–08. 48. Ser vicio de Impuestos Internos, Chi le, Noticias, 28 May 2019. (http://www.si i .c l/ noticias/2019/280519noti01er.htm, accessed 20 October 2020). 49. Directive 2014/40/EU of the European Parliament and of the Council. OJEU. 2014;127:1–38 (https:// ec.europa.eu/health/sites/health/files/tobacco/docs/dir_201440_en.pdf, accessed 7 October 2020). 50. Report on the EU customs enforcement of intellectual property rights: results at the EU border. Luxembourg: Publications Office of the European Union, 2019 (https://ec.europa.eu/taxation_customs/ sites/taxation/files/2019-ipr-report.pdf, accessed 7 October 2020). 51. Matthews P. Forestalling ahead of property tax changes. London: Office for Budget Responsibility; 2016 (OBR Working Paper No. 10; https://obr.uk/docs/dlm_uploads/Working-paper-No.10-1.pdf, accessed 29 January 2021). 52. Ross H, Tesche J, Vellios N. Undermining government tax policies: common strategies employed by the tobacco industry in response to tobacco tax increases. Prev Med. 2017;105S:S19–S22 (https://www. sciencedirect.com/science/article/pii/S0091743517302165?via%3Dihub, accessed 17 February 2021). 53. European Commission v Portugal supported by Belgium, Estonia and Poland, Case C-126/15. Judgement of the Court (Fifth Chamber); 29 June 2017 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?u ri=CELEX:62015CJ0126&from=EN, accessed 31 January 2021). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 165 54. Amalia Valesko v Zollamt Klagenfurt, Case C-140/05. Judgement of the Court (Second Chamber); 5 October 2006 (http://curia.europa.eu/juris/showPdf.jsf;jsessionid=02DEE17CC11845BAE0F42FC7 CFD24C85?text=&docid=65599&pageIndex=0&doclang=en&mode=lst&dir=&occ=first&part=1&c id=2591459, accessed 29 January 2021). 55. European Commission v French Republic, Case C-197/08. Judgement of the Court (Third Chamber); 4 March 2010 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:62008CJ0197&qid =1612223322830&from=EN, accessed 31 January 2021). 56. Commission requests that Belgium amend its legislation on tobacco: letter of formal notice. Brussels: European Commission; 8 November 2018. (Memo 18/6247; https://ec.europa.eu/commission/ presscorner/detail/EN/MEMO_18_6247, accessed 29 January 2021). 57. Commission requests that Belgium amend its legislation on tobacco, MEMO/18/6247 of 8 November 2018. 58. Pederson H, Floristean A, Iseppi L, Dawkins R, Smith C, Mørup C, et al. Study on the measuring and reducing of administrative cost for economic operators and tax authorities and obtaining in parallel a higher level of compliance and security in imposing excise duties on tobacco products. Brussels: European Commission; 2014 (https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/ ramboll-tobacco-study.pdf, accessed 3 October 2020). 59. Money laundering vulnerabilities of free trade zones. Paris: Financial Action Task Force; 2010. https:// www.fatf-gafi.org/media/fatf/documents/reports/ML%20vulnerabilities%20of%20Free%20Trade%20 Zones.pdf, accessed 7 October 2020). 60. Omi K. ‘Extraterritoriality’ of free zones: the necessity for enhanced customs involvement. World Customs Organization; 2019 (WCO research paper no. 47; http://www.wcoomd.org/-/media/wco/public/global/ pdf/topics/facilitation/ressources/permanent-technical-committee/225-226/itemixb_fzs_wco_e. pdf?la=en, accessed 29 January 2021). 61. European Parliament resolution of 26 March 2019 on financial crimes, tax evasion and tax avoidance. European Parliament; 2019 (2018/2121(INI)); http://www.europarl.europa.eu/doceo/document/TA- 8-2019-0240_EN.html, accessed 7 October 2020). 62. International Convention on the Simplification and Harmonization of Customs Procedures. Brussels: World Customs Organization; 2008 (http://www.wcoomd.org/Topics/Facilitation/Instrument%20 and%20Tools/Conventions/pf_revised_kyoto_conv/Kyoto_New, accessed 7 October 2020). 63. Evidence Notice 2004: tobacco duty – tobacco products manufacturing machine licensing scheme. London: HM Revenue & Customs, 2018 (https://www.gov.uk/government/publications/excise-notice- 2004-tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme/excise-notice-2004- tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme, accessed 31 January 2021). 64. Tackling illicit tobacco: from leaf to light: the HM Revenue & Customs and Border Force strategy to tackle tobacco smuggling. London: HM Revenue and Customs; 2015 (https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/418732/Tackling_illicit_ tobacco_-_From_leaf_to_light__2015_.pdf, accessed 7 October 2020). 65. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 66. Cardenas M. Case studies in illicit tobacco trade: Colombia. Chicago: University of Illinois at Chicago; 2020 (Tobacconomics Fact Sheet; https://tobacconomics.org/files/research/606/UIC_Colombia-Illicit- Trade-Fact-Sheet_v1.4.pdf, accessed 29 January 2021). 67. Maldonado N, Llorente BA, Escobar D, Iglesias RM. Smoke signals: monitoring illicit cigarettes and smoking behaviour in Colombia to support tobacco taxes. Tob Control. 2019;29:s243-s248 ( https:// tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/s243.full.pdf, accessed 29 January 2021). 68. Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 69. Pellechio A, Tanzi V. The reform of tax administration. Washington (DC): International Monetary Fund; 1995 (Working Paper no. 95/22; https://www.elibrary.imf.org/view/IMF001/07242- 9781451843941/07242-9781451843941/07242-9781451843941_A001.xml?language=en&redirect=true, accessed 29 January 2021). 166 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.1 COMPOSITION OF TOBACCO PRODUCTS To implement and enforce tobacco taxes in the most efficient way, competent authori- ties should be familiar with all of the components of tobacco products, including each of the raw materials used in their manufacture, production inputs and tobacco manufacturing machinery. Knowledge of the components of excisable products and machinery provides valuable information to identify activities at high risk for non- compliance, implement measures to ensure all taxes are paid and prevent illicit trade. Nicotine, non-nicotine and tobacco products and their component parts In most countries, the ministry of finance determines tax policy, including which tobacco products are taxed, while the ministry of health is responsible for product and use regulation. This could lead to different definitions of the same product, depending on which ministry is responsible for a given law or regulation. Wherever possible, a clear and common definition should be developed to simplify procedures and avoid confusion. Tobacco products take various forms, and not all may be regulated or subject to excise tax in a specific jurisdiction. In addition to cigarettes, other traditional tobacco products include smokeless tobacco – such as chewing tobacco, snuff and snus – as well as bidis and kreteks (clove cigarettes), which can be hand-rolled or manufactured, pipes, hookah or waterpipe and cigars. Cigarettes It is important to understand the materials and component parts of the tobacco products most commonly used in a particular country. Cigarettes are the most common and significant tobacco products in terms of volume and tax revenues in most jurisdictions. A cigarette stick is composed of: • the tobacco blend of various types of tobacco plant (leaves and stem and other plant parts) and additives (including flavours); • the cigarette paper used to wrap the tobacco blend to make up the tobacco rod; • the acetate filter that forms the white portion at the tip of a filtered cigarette, which is in direct contact with the smoker’s mouth; • the tipping paper or wraps around the filter; and • the adhesive that secures the cigarette paper around the tobacco blend and the tipping (1). Each manufacturer follows a specific process to produce cigarettes. Aside from the tobacco blend, manufacturers also vary the size of cigarette paper and tipping paper and the length of acetate filter used per stick (1). In some countries, these elements are standardized. In an ideal regulatory framework, a manufacturer would CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 167 be required to submit information on the specific process for each brand and variant of the tobacco product that it manufactures to the competent authorities as part of the licensing requirement (see section 3.4.3). Authorities could, for example, require manufacturers to submit this information in order to obtain a licence. The minimum requirements of the administration and the information that should be included could be laid down in law or lower regulation to ensure that authorities have the information in their possession for all licensed manufacturers. This information contributes to verifying whether a company is reporting the actual quantity of cigarettes manufactured for sale and sold cigarettes by comparing the amount of materials used for production and the quantities used per cigarette with the total number of manufactured cigarettes. Eventually, this information also contributes to validating whether the taxes are properly paid. Figure A3.1 shows the component parts of a typical machine-made traditional cigarette. Fig. A3.1 Component parts of a machine-made cigarette Source: Author’s compilation. Photo by Walter Klerx. Not all parts of tobacco products are subject to the same level of control. According to Article 6.5 of the Protocol, five years following the entry into force of this Protocol, the MOP shall ensure at its next session that evidence-based research is conducted to ascertain whether any key inputs exist that are essential to the manufacture of tobacco products, are identifiable and can be subject to an effective control mechanism. On the basis of such research, the MOP shall consider appropriate action. In addition to the component parts of tobacco products, materials needed for packaging a specific number of sticks into a pack of cigarettes, usually 20 per pack, can be monitored. These materials include the foil paper, the package paper (which could bear the brand name, design and health warnings), the fiscal marking (if required) and the plastic or cellophane wrap. A fixed number of packs of cigarettes, Filter Tipping paper Cigarette paper Tobacco Tobacco rodFiltration zone 168 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N normally 10 packs, are packed into cartons, also called reams. These cartons are usu- ally made of soft paperboard or cardboard, possibly with branding, and are wrapped in plastic wrap or cellophane. Fifty cartons are packed in master cases, which are made of sturdier and thicker paperboard and stacked on pallets (usually 50 master cases to a pallet). An effective regulatory framework would require manufacturers and importers to provide information to the competent authorities on packaging and design, as well as the number of sticks per pack, carton and master case. Prior approval for purchase or sale of materials used in the cigarette production process can also be required. In the Philippines, suppliers of such raw materials, including those providing tobacco papers and filter components, are required to have a licence (2). In some of the Member States of the EU, raw tobacco is also subject to fiscal and legal requirements. For example, in Slovakia and Poland, raw tobacco can be handled only by authorized operators. While authorized operators do not have to pay excise duties on raw tobacco, if raw tobacco is detected by an unauthorized operator, excise duties will be due. Hungary, Italy and the United Kingdom require registration or authorization for all operators and growers handling raw tobacco (3). In addition to knowing the quantities of inputs required to produce a specific amount of a regulated product (e.g. cigarettes), the competent authority also needs to understand the supply, manufacturing and distribution chains to be able to properly monitor, regulate and determine whether taxes have been paid (see also Fig. 3.3). Novel and emerging nicotine, non-nicotine and tobacco products In recent times, new products have been introduced to several markets, namely, ENDS, ENNDS and HTPs. ENDS usually comprise a nicotine-containing e-liquid but do not contain to- bacco. ENNDS are essentially the same but do not (ostensibly) contain nicotine. The WHO COP requested the Convention Secretariat to invite Parties to monitor and report on scientific, regulatory and market developments such as initiation, cessation, advertising and promotion of ENDS and ENNDS. Furthermore, the COP requested WHO to report on the development of methods by regional and international standards-development organizations for the testing and measuring of contents and emissions of these products (4). There are different types of e-cigarettes – the most common type of ENDS and ENNDS – and currently there are four generations of products. However, they can be divided into two broad categories: open systems and closed systems. Both types of e-cigarette use a wick and a heat source to generate an aerosol. The wick is saturated with e-liquid, and a microprocessor is used to control operations (not all include this). Some e-cigarettes also have an LED light to imitate the burning end of a conventional cigarette (5). Fig. A3.2 presents examples of open and closed systems. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 169 E-liquid pod cover Rechargeable battery Heating element (heats solution, aerosolizing nicotine) Mouthpiece E-liquid pod Fig. A3.2 Examples of open and closed systems of ENDS/ENNDs products Open ENDS/ENNDS system (e-cigarette) Closed ENDS system (e-cigarette) Source: (6). Unlike ENDS/ENNDS, HTPs do contain tobacco. HTPs produce aerosols containing nicotine and toxic chemicals when tobacco is heated or when a device containing tobacco is activated (7). HTPs are composed of two elements: the sticks or pods that contain the tobacco and the device used to heat the tobacco. Both are necessary for the product to be used. Fig. A3.3 shows an example of a heated tobacco product. HTPs are tobacco products and are therefore subject to the regulatory measures contained in the WHO FCTC. Rechargeable battery Power button (to start vaping) Mouthpiece Atomizer / Heating element (heats solution, aerolizing nicotine) E-liquid tank (refillable e-liquid nicotine tank) 170 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. A3.3 Components of an HTP Holder: Tobacco stick: Note: PLA: polyactic acid, MPF: mouthpiece filter. Sources: (8-9). More information on tax administration of other tobacco products is presented in section 3.5. Tobacco stick Holder Charger Casing Control Electronics Battery Heating Blade Heatstick total lenght: 45mm MPF (7 mm) PLA (18 mm) Tobacco plug (12 mm) Hollow acetate tube (8 mm) Outer paper Diameter max 7.42 mm Tipping paper CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 171 REFERENCES 1. Memorandum by Gallaher Group Plc: the tobacco industry and the health risks of smoking. Select Committee on Health Minutes of Evidence. London: UK Parliament, 2000 (https://publications. parliament.uk/pa/cm199900/cmselect/cmhealth/27/0011323.htm, accessed 7 October 2020). 2. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 3. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco.. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 4. Electronic nicotine delivery systems and electronic non-nicotine delivery systems. Geneva: World Health Organization; 2016 (Decision FCTC/COP7(9) of the Conference of the Parties to the WHO Framework Convention on Tobacco Control; https://www.who.int/fctc/cop/cop7/FCTC_COP7_9_EN.pdf?ua=1, accessed 7 October 2020). 5. Brown CJ, Cheng JM. Electronic cigarettes: product characterization and design considerations. Tob Control. 2014;23:ii4-ii10 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3995271/pdf/ tobaccocontrol-2013-051476.pdf, accessed 31 January 2021). 6. E-cigarettes and vapor products [webpage]. King County, Washington (USA); 2019 (https://www. kingcounty.gov/depts/health/tobacco/data/e-cigarettes.aspx, accessed 9 October 2020). 7. Heated tobacco products. Geneva: World Health Organization; 2020 (WHO/HEP/HPR/2020.2 Information sheet; https://apps.who.int/iris/bitstream/handle/10665/331297/WHO-HEP-HPR-2020.2- eng.pdf?sequence=1&isAllowed=y, accessed 7 October 2020). 8. Tobacco heating system (IQOS) briefing document. Silver Spring: US Federal Drug Administration; 2018 (https://www.fda.gov/media/110377/download, accessed 31 January 2021). 9. Premarket tobacco product application: technical project lead review. Silver Spring: US Federal Drug Administration; 2017 (https://www.fda.gov/media/124247/download, accessed 7 October 2020). 172 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.2 EXAMPLE OF FORESTALLING AND COUNTERMEASURES This example is hypothetical but inspired from the situation in the Philippines. The amounts and prices have been altered, however, and it is assumed that the normal inventory kept by a tobacco manufacturer is two months.  The excise tax imposed on a pack of cigarettes in the current year is US$ 3.00; it will be increased to US$ 3.30 at the beginning of the new fiscal year (January in this example). The monthly production of Brand Y cigarettes of company X, which it declares for tax purposes, is as follows: MONTH OF THE CURRENT YEAR PACKS OF CIGARETTES January 10 000 000 February 10 500 000 March 9 900 000 April 11 000 000 May 10 200 000 June 10 600 000 July 9 700 000 August 10 100 000 September 9 900 000 October 10 100 000 November 20 000 000 December 25 000 000 Since the normal inventory is two months, the quantity in the two months prior to the implementation of the new excise tax rate is disregarded. The shelf life of tobacco products is approximately six months. The average of the six months prior to November is computed to obtain the quantity presumed to be produced or imported if there was no tax increase. The quantity from May to October (inclusive) divided by 6 is 10 100 000 packs. Thus, any quantity produced beyond 10 100 000 packs for the months of November and December (the months prior to the implementation of the new tax rate) is assessed using the new tax rate. In this example, 10 100 000 of the packs produced in November will be taxed at the old rate of $3.00, and 9 900 000 packs will be taxed at the new rate of $3.30. For December, 10 100 000 packs will be taxed at $3.00, while 14 900 000 packs will be taxed at $3.30. Without imposing these measures, the government would have been deprived of the excise tax increase on 24 800 000 packs. In addition, the effect of the increase on prices and consumers would have been delayed by approximately two months. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 173 In countries using tax stamps, the withholding of the issuance of stamps is a well-known approach to counter forestalling. Another practical solution is to allow the competent authorities to request advances from the industry to cover revenue shortfalls, provided there is a legal basis for such requests. 174 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 175 CHAPTER 4. Political economy As with any proposed government action, policy-makers need to navigate the political environment of tobacco taxation at every stage of policy development, implementa- tion and administration. While every country’s distinct history, culture, systems and structural forces shape its unique political landscape, there are some universal themes when it comes to tobacco control and particularly tobacco taxation. These themes boil down to the distribution of money, power and resources. The tobacco industry, as both a political and economic player, understands these themes well. The industry has been effective in using principles of the political economy of tobacco taxation in its efforts to block important advancements in tobacco control. Nevertheless, the savvy policy-maker can see through the industry arguments by considering who benefits from industry-favoured policy measures and interven- tions. The industry’s challenges to tobacco tax policies can be organized into the five categories of SCARE tactics. This chapter provides a road map to help policy-makers navigate the political economy of tobacco taxation through each of these themes. The first five sections dissect the tobacco industry framing of each issue, pinpoint- ing the flaws in each argument, identifying the extent to which each concern has merit and suggesting how a responsible government can address each one. These discussions are supported by unbiased evidence from independent, peer-reviewed research, as well as specific examples from country experiences. Sections 4.1 through 4.5 on SCARE tactics will equip policy-makers with the tools they need to proceed with confidence that their tobacco tax policy – developed and implemented fol- lowing the guidelines spelled out in this technical manual – will bring about the greatest health and economic benefits for their constituents, regardless of industry attempts to thwart them. Section 4.6 further buoys policy-makers’ efforts to ensure the beneficial impacts of their policies, as it describes how earmarking can improve the political economy of tobacco taxation by funding programmes and initiatives that promote and support the health and well-being of the population. 176 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.1 SCARE TACTIC S: SMUGGLING AND ILLICIT TRADE 4.1.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases will inevitably result in an increase in the illicit trade in tobacco products (1–2). They claim that higher tax rates and higher prices strengthen the financial incentives for criminal enterprises to supply cigarettes from lower-tax jurisdictions, boost domestic tax evasion and encourage smokers to seek cheaper illegal cigarettes. The industry also challenges the argument that tobacco tax hikes increase government revenue by claiming that the presence of an illicit tobacco market will actually reduce revenue collection following a tax increase. More recent versions of this argument – adapted to address public health concerns about tobacco use – claim that illicit market growth also offsets reductions in smoking prevalence that would otherwise be brought about by tobacco tax increases. In summary, the tobacco industry and its allies claim that raising tobacco taxes are ineffective – and even counterproductive – because they are circumvented by illicit markets, which prevents the government from achieving its public health objective and reduces rather than increases tax revenues. When a country considers a proposal to increase tobacco excises, the tobacco industry and its allies frequently make exaggerated claims about the size and scope of illicit tobacco trade in that country. Opponents of tobacco tax increases argue that price differentials are the exclusive – or at least the dominant – cause of illicit trade. Influenced by this fear-inducing faulty diagnosis, tax authorities frequently find it difficult to make decisions about tobacco taxes. However, the industry diagnosis always contains the same erroneous elements. First, the illicit trade in a country is frequently less than the industry portrays it to be, and the country’s tax enforcement policy towards tobacco products is rarely unique or in any way different from the norm in the country (3). Second, the scale of illicit trade in tobacco is not exclusively or even primarily determined by tax or price differentials. Typically, it results from a set of governance problems characterized by government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, lack of dissuasive sanctions and/or weak norms regarding participation in illegal and informal markets (4–7). This section provides guidance for tax and other relevant authorities on how to respond to the tobacco industry SCARE tactic that increasing tobacco taxes will lead to smuggling and illicit trade in their countries. Tax authorities need to know the nature, causes and extent of illicit trade so that they can define the problem properly and formulate an appropriate response. This chapter addresses the available tools for better defining and understanding specific illicit trade problems, particularly tools that facilitate independent assessment of the magnitude of that trade. Improvements CHAP T ER 4. PO LI T I C AL ECO N OMY 177 to governance within the realm of tax authorities – such as best practices in tobacco tax administration and policies to improve the effectiveness of fiscal regulations or norms regarding participation in informal and illegal markets – are discussed in Chapter 3. This section first describes the nature of the illicit tobacco trade to highlight some of its complexities and identify complementary policies for tackling the problem. Next, evidence that calls into question the link between illicit trade and high prices or tax rate changes is discussed. Finally, to help tax authorities assess their own situation, several different methodologies are presented to estimate the scope of the illicit tobacco trade and to evaluate estimates of that trade for a particular country or tax jurisdiction. 4.1.2 THE NATURE AND EXTENT OF THE ILLICIT TOBACCO TRADE The WHO FCTC defines illicit trade as: any practice or conduct prohibited by law and which relates to production, ship- ment, receipt, possession, distribution, sale or purchase, including any practice or conduct intended to facilitate such activity (8). Non-duty-paid tobacco products found in a jurisdiction (i.e. through littered-pack surveys) could be the result of either of two related but distinct activities: tax evasion and tax avoidance. Tax evasion is a set of unlawful actions seeking the non-payment of tobacco taxes and duties, whereas tax avoidance comprises legal actions with the purpose of avoiding payment of some or all taxes, such as bringing an amount of cigarettes up to the legal allowance from a lower- into a higher-tax jurisdiction. Tax avoidance is not illegal and is therefore not considered part of illicit trade in tobacco products.1 The focus of this section is on tax evasion activities, which can occur in the movement across borders or in domestic production and distribution. When tax evasion happens across borders, it is known as smuggling (9) and can be done on a large scale or a small scale. Tax evasion in the domestic market can be partial, when licensed and authorized producers or distributors comply with only part of their tax obligations, or total, when the whole production and distribution system is illegal and out of sight of tax administrators (5, 10–14).2 Large-scale tax evasion schemes can be run by different types of producers and their associated distributors, such as the transnational tobacco companies (TTCs) and their national subsidiaries, 1 Tax avoidance practices –common among states in the United States and countries in the EU – are not analysed in this section. 2 Tax evasion is normally considered as illicit manufacturing in the literature (14). 178 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N other local tobacco companies producing their own brands and illegal factories that normally counterfeit other brands or produce illicit brands. Large-scale smuggling involves, for example, taking advantage of tax-free zones and mislabelling shipping products prior to or during transit (11) or using so- phisticated clandestine networks. This form of tax evasion is systemic and can be carried out by TTCs (12), by local companies producing in countries with low tax enforcement, e.g. Paraguay (15) that feed neighbouring countries and regional illicit hubs through a network of clandestine distributors (15) or by companies located in tax-free zones like Jebel Ali and Dubai in the UAE (16) or such zones in Russia or Cyprus (17). The origins, routes and quantities of large-scale smuggling frequently change as affected countries or markets react by strengthening enforcement and seeking bilateral cooperation with the jurisdictions of origin. For example, the 2013 comprehensive strategy of the EU (10) enhanced bilateral cooperation with major source and transit countries of illicit cigarettes coming to Europe, including Russia, Belarus and Ukraine. These agreements improved day-to-day cross-border coopera- tion, reduced illegal flows and introduced gradual tobacco excise rate adjustments in those three countries to bring them to European levels. Small-scale smuggling (also known as ant smuggling or bootlegging) is the cross-border trafficking of cigarettes in quantities that are larger than the allowable limits (e.g. two cartons) but smaller than large shipments (e.g. truckloads, cargo containers), normally for the purpose of selling at a profit (11). This type of illicit trade may exist in places where there are opportunities within neighbouring tax jurisdictions. For example, small-scale smuggling is commonly done by individuals living in French and German provinces near lower-taxed countries (e.g. Belgium, Luxembourg, Switzerland, Spain, Poland and Czechia) (18). Counterfeiting is a form of illicit manufacturing that involves the production of tobacco products (including packaging and tobacco filler) without the approval of the trademark holder (13). Another product of illicit manufacturing is so-called cheap or illicit whites. Cheap whites are branded (e.g. Jin Ling) or unbranded cigarettes that are legally or illegally produced3 and knowingly sold in the illicit market (17). Cheap whites are not usually produced by TTCs (17, 19).4 They are produced by small tobacco produc- tion companies in one country and often sold in illegal markets of neighbouring 3 Ross et al. (17) analysed this issue and found that the sale to the first purchaser is usually legal. Their analysis covers the production in free zones (i.e. in the UAE, Russia and Cyprus) and production exported from Viet Nam, Indonesia and China. In those cases, there is no need to make the first sale illegally. However, cheap white production in Paraguay is sold to domestic distributors, and most of those sales are completely illegal. 4 Ross et al. (17) and Gilmore et al. (19) identify some cheap white brands sold by TTCs, such as President (PMI), produced in Ukraine, and Esse (Korea Tobacco & Ginseng Company, KT&G), produced in Indonesia. CHAP T ER 4. PO LI T I C AL ECO N OMY 179 countries. For example, in Paraguay, cheap whites are produced on a large scale by a few companies under the guise that they are marketed domestically, but a large share is smuggled into Uruguay and Brazil (4). Iglesias et al. (20) showed how TTCs’ cheap brands were illicitly shipped through Paraguay to be sold in the Brazilian and Argentine markets in the 1990s. This contributed to increased production of cheap whites in Paraguayan firms, which continued the illicit business even after Brazilian legislation obstructed the illegal activity of the TTCs. Domestic tax evasion is a pervasive phenomenon, particularly in LMICs. Partial tax evasion in tobacco products can be found at any level of tax rates or prices and is generally the result of defective legislation or weak tax enforcement.5 Complete or total tax evasion occurs when producers and distributors are clandestine or when there are serious institutional challenges to tax enforcement between two tax jurisdictions, such as between the United States and Native American Reservations. Evidence of illicit manufacturing has increased in recent years in several places in the world, including the EU (10) and Brazil (21). TTCs were predominant in illicit trade activity until the end of the 20th century, and even with the entrance of new actors into the illicit business, TTCs have not entirely exited. Gilmore et al. (22) analysed industry-funded data and seizure data and concluded that TTCs are still involved in illicit trade in Europe, despite the Anti-Contraband and Anti-Counterfeit Agreements (the “Agreements”) signed between the four TTCs and the EU (23).6 Using industry-funded data, Gilmore et al. show that 58% of illicit EU cigarettes can be attributed to the four main TTCs. When seizure data are used, 69% to 73% of illicit EU cigarettes can be attributed to these firms (22). It is always difficult to assess the extent of the global illicit tobacco trade because of its illegality, its global and changing nature and problems with data collection (24). Before the 21st century, when TTCs were almost unique actors in the large-scale smuggling of well-known cigarette brands, the difference between global exports and imports of cigarettes could provide a good approximation of the size of this problem globally (7). However, with the growth of illicit manufacturing in general, the manufacturing of cheap whites and the illegal movements of those products 5 This occurs when licensed and authorized producers underreport actual quantities and sell the non- duty-paid produced quantities through illegal channels. It can also include instances when producers do not report quantities at all, as in many ad valorem systems of LMICs. 6 From 2000 on, the European Commission and 10 Member States launched court cases regarding smuggling and money laundering against several TTCs. To end the court cases, the Anti-Contraband and Anti-Counterfeit Agreements were signed, which required the TTCs to exercise stringent control over their supply chain (through tracking and tracing, due diligence and anti-money-laundering and reporting obligations), share operational intelligence with Member States and the EU and pay penalties for seizures, as well as annual payments over a period of 12 years. The agreement with PMI has ended, the one with JTI will end in 2022 and the others with Imperial Tobacco and BAT will run until 2030 (23). 180 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N over the past two decades, trade statistics are no longer as useful as they were in the past. Joossens et al. (25) tried to estimate the size of the global illicit cigarette market by adding different types of national estimations prepared around 2007. They found that the estimated size was 657 billion cigarettes per year, or 11.6% of the global cigarette market. According to Joossens et al., illicit trade costs govern- ments US$ 40.5 billion in tax revenue worldwide, and eliminating illicit tobacco trade would recover US$ 13 billion in immediate revenue in high-income countries and US$ 18.3 billion in LMICs.7 Descriptions of the types of illicit trade are useful for developing the first com- ponent of a strategy to fight it: assess the nature and size of the problem. Table 4.1 presents all the main components of a strategy to fight illicit trade. To make progress in this first component – knowing the problem – authorities could use and adapt existing instruments of health surveillance or seek partnerships with academia and independent specialists to investigate the issues involved, using different methodolo- gies (see subsection 4.1.4 and Annex 4.1 on methodologies to assess the nature and size of the problem). Knowing the nature of the problem requires the cooperation of different government actors – for example, to investigate both the financial and criminal operations of organized crime behind the illicit trade. The gathering of qualitative information on the nature of the illicit trade should start simultaneously with the statistical work of measuring the magnitude of the problem. Table 4.1 Components of a strategy to fight the illicit tobacco trade 1. Assess the nature and size of the problem Use and adapt existing health surveillance and other existing national surveys to assess the problem Seek partnerships with academia and independent specialists to find ways to rigorously study illicit trade Use financial and police investigations to identify and fight organized crime operating in illicit trade 2. Start identifying and implementing appropriate country-specific policies and strategies to address illicit trade Improve tax and customs administration to close the legal and administrative loopholes facilitating illicit trade Implement other appropriate policies to deal with country-specific problems 3. Become a Party and/or implement the Protocol to Eliminate Illicit Trade in Tobacco Products Adapt the Protocol supply-chain control obligations Adjust national penalties for illicit trade offences Seek and build international cooperation 7 The WCO publishes an Illicit Trade Report annually, with the main characteristics and trends of illicit flows in key products, including tobacco, using data based on customs seizures. CHAP T ER 4. PO LI T I C AL ECO N OMY 181 Methodologies available to estimate the nature and size of illicit trade are discussed in subsection 4.1.4. This is the first step for dealing with SCARE tactic S. Chapter 3 discusses at length the relevant tax administration measures and best practices to minimize opportunities for illicit trade in tobacco products. Table 4.2 presents examples of appropriate policies and strategies targeted to address specific types of illicit trade in addition to the best practices described in Chapter 3. After completing the first step of this strategy, tax, health and justice authorities should discuss how to face country-specific problems, considering not only tax and customs administration measures but also social, law enforcement and international cooperation policies and strategies. Table 4.2 Suggested policies and strategies to address country-specific illicit trade problems MAIN TYPE OF ILLICIT TRADE IN THE JURISDICTION PROBLEMS POLICIES/STRATEGIES TO USE Bootlegging Neighbouring low-tax jurisdiction Bilateral negotiations to harmonize tobacco tax systems Difficulty of controlling people’s movements in countries with extensive land borders Identify and establish suitable social protection or employment policies for targeted populations in border regions Extensive land border with multiple accesses Bilateral cooperation with law enforcement and border control forces, monitoring of access routes to main consumption markets Large-scale smuggling from neighbouring jurisdiction Neighbouring low-tax jurisdiction and difficulties in controlling borders Bilateral negotiations to harmonize tobacco tax systems and bilateral law enforcement cooperation Producers and distributors in the lower-tax jurisdiction aiming to supply the high-tax jurisdiction Bilateral cooperation to harmonize tax systems and control producers and distributors in the origin country, create conditions for legal exports and taxed imports Large-scale smuggling from a third country or tax-free zones Producers and distributors aiming to supply non-duty-paid tobacco products wherever possible Customs and other forms of international cooperation to control and monitor exports from identified areas Domestic tax evasion Existence of many small informal or semi-formal producers Encourage business concentration through producer associations and cooperatives, create incentives for formalization and establish licensing rules and basic electronic information systems for raw material and production 182 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Underreporting from formal producers Improve tax administration with policies such as basic electronic information systems for inputs and production, establish neutral procedures to verify production, improve audit systems, increase third-party information on inputs and production of tobacco products Clandestine factories Law enforcement investigation of commercial associations with raw- material and machine producers and distributors Governments should try to identify the incentives and governance problems that encourage and allow illicit trade movement inside their country. As seen in Table 4.2, the design and implementation of policies to deal with those problems do not depend exclusively on tax and customs authorities; they also depend on the efforts of the police and law enforcement, the Justice Department and the judicial apparatus. In other words, a great deal of coordination and consultation among different types of government bodies and expertise is needed to produce an adequate response. It is also clear from Table 4.2 that domestic tax evasion by formal producers can be tackled by tax authorities and is mainly related to the supply-chain-control provisions of the Protocol. The Protocol (26) builds upon and complements Article 15 of the WHO FCTC, which addresses means of countering illicit trade in tobacco products as a key aspect of a comprehensive tobacco control strategy. The Protocol is a blueprint of measures to deal with this problem, and its provisions should be part of any strategy for fighting the illicit market. It is a legally binding treaty in its own right that entered into force on 25 September 2018. As described in detail in Chapter 3, the Protocol has three main lines of action: supply-chain controls, recommendations on how to treat unlawful conduct related to the illicit tobacco trade and suggested mechanisms to seek and build international cooperation to fight that trade. Countries can start implementing Article 15 of the WHO FCTC and the appropriate polices or strate- gies recommended by the Protocol even before acceding to it, selecting those most suitable to the nature and extent of their particular problem. Such transitional work will facilitate the eventual implementation of the Protocol, because any plan to correct loopholes in tax and customs practices will bring government authorities closer to the best practices recommended in the Protocol. 4.1.3 DETERMINANTS OF TAX EVASION: THE ROLE OF PRICE LEVELS The argument that price and tax rates are the main determinants of the illicit tobacco trade has persuaded some governments (e.g. Uruguay and Georgia in the past) to avoid policies that may lead to cigarette price increases (e.g. excise tax rate increases) (4). CHAP T ER 4. PO LI T I C AL ECO N OMY 183 Some governments (e.g. Canada in 1994, Brazil in 1999 and Pakistan in 2017) (20, 27) have even reduced tax rates in attempt to reduce the illicit trade. The wider scholarly literature demonstrates that illicit trade is not a monocausal phenomenon (7) but is the result of many factors, most of them related to gover- nance issues. Government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, weak norms regarding participation in illegal and informal markets and conflicts between neighbouring countries (5) all contribute to the existence and growth of the illicit tobacco trade. It is difficult to isolate the role of price from each of the other factors because (1) obtaining prices and quantity measures of illicit trade is inherently challenging; (2) in most countries, there are many cigarette brands, and prices vary between and even among brands; and (3) there is a lack of good measures to deal with nonprice factors affecting illicit trade, such as government corruption and ineffective criminal justice. These constraints make it challenging to develop rigorous empirical evidence about how price and other factors affect illicit trade. Despite these fundamental challenges, the economic literature has produced credible evidence that price is always only one factor – and often not the most important factor – determining the extent of illicit trade. Many econometric studies about the influence of price and other factors have focused on cross-border shopping (or small-scale bootlegging from low- to high-tax jurisdictions), given the availability in the United States and Europe8 of sales data for low- and high-tax jurisdictions, classified in a convenient way by geographical zones – i.e. close to or far from the borders. Those studies attempted to explain the illicit trade flows or the relatively higher sales in low-tax jurisdictions as a function of price and tax differentials between the lower-tax and surrounding higher-tax jurisdictions, after controlling for other important factors affecting cross-border sales such as proximity to borders and levels of corruption (6, 7, 11, 28).9 The main conclusion of the studies is that illicit trade flows are not linked solely to price (29). Some show a significant effect of price differentials together with other factors, but others do not find significant price differential effects. The important policy implication of these analyses is that decreasing tobacco tax rates and real prices in higher-tax jurisdictions could have minimal or no effect on illicit market shares.10 8 This was a traditional strand of the literature in the United States on trade among states, and to a lesser extent in European countries, most of which used conventional but inaccurate illicit trade measurements. 9 Recently, PMI-Altria financed some studies of factors affecting cross-border sales. One of those studies, Prieger and Kulic (28), criticized Merriman et al. (2000) (9) and arrived at the conclusion that in cross-border shopping, price differentials are important for determining the magnitude of illicit trade. 10 Brazil decreased tax rates and real prices at the beginning of the 21st century to fight illicit trade coming from lower-tax jurisdictions. After this action, however, the government lost revenues, and the size and scope of illicit trade remained unaltered, according to industry sources (20). 184 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Observational and case studies provide information that may improve public policy even when they are unable to produce compelling evidence of causal relation- ships. Some observational studies have correlated price levels with illicit market shares, using large samples of countries. Joossens et al. (25) found that countries with high taxes and prices normally have lower shares of illicit trade than countries with lower tax shares and prices. In their sample, high-income countries generally have relatively high cigarette prices and tax shares, but their favourable results (i.e. lower levels of illicit trade) are related to effective tax administration and lower corruption levels. In contrast, LMICs generally have lower prices and tax shares, along with significant illegal market shares. Joossens et al. attribute difficulties in fighting illicit trade to weak tax and customs administrations and, in most cases, institutional and legal challenges (25). Figure 4.1.1 illustrates the relationship between price and illicit trade, using the price (in US$) per pack of the most-sold brand of cigarettes and the estimated level of illicit trade for 94 countries in 2018.11 There is no apparent unique association between the two variables. Running a linear regression with retail price as the explana- tory variable and share of illicit trade as the dependent variable shows an inverse, but not statistically significant, relationship between price and illicit market share.12 Figure 4.1.1 illustrates some particular cases: • Many countries with low prices (i.e. lower than US$ 2 per pack) have the highest levels of illicit trade in the sample, e.g. Brazil (BRA) ($1.33 and 46.3% illicit share), Pakistan (PAK) ($0.39 and 40%), Ethiopia (ETH) ($0.55 and 32.9%), Ghana (GHA) ($1.06 and 29%) and Cameroon (CMR) ($0.89 and 25%). • In contrast, many of the countries with prices between US$ 4 and $8 – which could be considered high enough for financial incentives to operate – have illicit trade shares of less than 10% of total consumption. These countries include the Republic of Korea (KOR) ($4.02 and 0.8%), Czechia (CZE) ($4.31 and 2.9%) and Sri Lanka (LKA) ($6.89 and 1.6%). • All countries that have very high prices – higher than US$ 8 – except for Ireland, register illicit trade shares below 20%. These countries include France (FRA) ($9.39 and 17.8%), Switzerland (CHE) ($8.71 and 5.5%), Singapore (SGP) ($10.35 and 3.7%) and Norway (NOR) ($14.51 and 9.6%). 11 National estimates of the magnitude of illicit trade are controversial. The tobacco industry’s numbers overestimate the problem and are based on questionable methodologies. Estimates with a rigorous and transparent methodology are not available for a large sample of countries for the same year. In order to compare price levels with illicit market shares, Euromonitor’s estimations of illicit market share were selected, for two reasons: they are comparable estimates for a large sample of countries in a given year, and no one could argue that they are biased towards tobacco control’s points of view. The use of Euromonitor data does not imply that WHO fully agrees with all the details and methodologies used to obtain them. 12 Other factors must be taken into account to transform this observational analysis into a rigorous analysis of cause and effect. CHAP T ER 4. PO LI T I C AL ECO N OMY 185 Fig. 4.1.1 Share of illicit trade versus retail price of the most-sold brand of cigarettes in US$, by country, 2018 Note: The extent of illicit trade in cigarettes is measured by Euromonitor as the estimated quantity of illegal cigarettes consumed in a country divided by the estimated total consumption of cigarettes in that country. Sources: (27, 30). 51 2 3 4 6 7 8 9 10 11 12 13 14 1615 0 10 20 25 15 5 30 35 40 45 50 MYS BRA ECU ECUPAK ETH PAN CRI URY GRCIND GTM SLV ARE LVAHND NGA ZAF CAN FRA LBNAGO IRL DOM GHA CMR Retail price, USD per pack Ili ci t t ra de % IRQ VNM COL MMR AUT SWE EST PER ESP ROM POL CHL ISR GBR KEN LAO KHM EGY MKD TZA THA TUN BGD UZB UKR GEO KAZ CHN KWT HRVJOR CZE SVK KOR JPN BLR SRB TUR OMN FIN NLD CHE SGP SAU USA LKA DNK DEU ITA HUN SVN PRT BOL DZAIDN CIV BEL MAR ARG AUS NOR NZL LTU BIH MEX BGRPHLAZE RUS 186 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As indicated in numerous studies and analyses of illicit trade in tobacco products (4–5), the most effective way to tackle the problem is not to forgo tax increases but rather to strengthen the capacity to fight the trade. Therefore, it is important to consider the relationship between good governance and illicit trade. The more capacity a country has to counter illicit trade in general, the lower the level of that trade will be. An index compiled by the Economist Intelligence Unit (EIU), the Global Illicit Trade Environment Index, measures countries’ structural capacity to fight illicit trade overall. The EIU indicator is a combination of four indicators or categories designed to assess countries’ performance in those areas; the closer the overall indicator is to 100, the better the country’s capacity to fight illicit trade. The four categories are:13 1. government policy, which measures the government’s commitment to pro- actively monitoring and preventing illicit trade; 2. supply and demand, which measures the extent to which the domestic en- vironment discourages or encourages supply and demand for illicit goods; 3. transparency and trade, which measures transparency and the degree of governance applicable to free-trade zones and transhipments; and 4. customs environment, which measures how effectively customs services facilitate legitimate trade while at the same time preventing illicit trade. Figure 4.1.2 illustrates the relationship between the EIU indicator and the estimated level of illicit trade in cigarettes in a set of countries. There is an inverse and statisti- cally significant relationship between the indicator and the estimated level of illicit trade in cigarettes. This suggests that as the capacity to fight illicit trade in general increases, the illicit trade in cigarettes falls.14 13 For more details about this indicator, visit http://illicittradeindex.eiu.com/. 14 The association was significantly different from zero at a 90% confidence level, using a linear regression between the two variables. CHAP T ER 4. PO LI T I C AL ECO N OMY 187 Fig. 4.1.2 Share of illicit trade versus the EIU indicator in 70 countries, by country, 2018 Sources: (30–31). % Il ic it tr ad e ci ga re tt es MYS 0 10 15 5 20 25 30 35 40 45 50 55 60 EIU GIT indicator 20 3010 40 50 60 70 80 90 JPN KOR HRV SAU UKR BLR KHM LAO MMR IRQ GTM DOM VNM MAR SRB PER BGR MEX ROM COL ZAF BIH PHL IDN ITA PRT HUN BEL DEU ESP ARG AUS SWECHL TW POL ISR AUTLTU FRA CAN IRL LVA ARE GRC IND URY CRI PAN PAK ECU BRA EST GBR FIN SVN NLD SVK TUN RUS DZA TUR CHNKAZ THA CZE USA SGP NZL DNK 188 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Rigorous independent research has established that despite the challenges of illicit trade, taxation of tobacco products is an effective public health intervention that substantially reduces tobacco use and generates government revenue (5). Further, when cigarette taxes increase, governments generate higher revenue and consumption is reduced (32–33). However, ineffective tax administration can allow illicit trade to grow and can undermine some of the benefits of tobacco taxation by making cheaper cigarettes available. For example, the average street price of smuggled cigarettes in Malaysia is 55% lower than its legal tax-paid equivalent (34). Illicit tobacco trade also reduces government tax revenue and may increase health costs associated with smoking and costs associated with policing. 4.1.4 MEASURING ILLICIT TRADE IN TOBACCO PRODUCTS The magnitude of illicit trade is a powerful argument in tax policy discussions, and for this reason the tobacco industry funds estimation of illicit trade in countries or regions of particular interest to itself (i.e. Project Sun and Project Star in the EU and Oxford Economics in East Asia). However, a recent systematic review of industry data on illicit trade finds substantial methodological weaknesses in industry-commissioned reports (24). Furthermore, Blecher et al. (35) argue that industry-funded studies tend to systematically overestimate the size of illicit trade to persuade authorities to abandon tobacco tax reforms. Independent researchers have also uncovered inconsistencies in tobacco-industry-funded estimates (36). Some examples of inflated industry-linked illicit trade estimates are given in Table 4.3, which compares peer-reviewed and independent studies with estimates funded by the tobacco industry. Because some countries have several industry estimates from different sources or years, Table 4.3 presents the estimate included in the article that published the independent study, because it was considered as representative and adequate to illustrate the overestimation. In all cases, the industry estimates exceed those of the independent studies. Measuring the scale of illicit trade can be a daunting task for governments because different methods are employed by independent researchers, governments and the tobacco industry. Nonetheless, it is worth investing in these studies because they drive policy discussions and can be used to evaluate the impact of policies (e.g. tax increases, plain packaging and health warnings). CHAP T ER 4. PO LI T I C AL ECO N OMY 189 Table 4.3 Illicit market share estimated in independent studies compared with estimates in tobacco-industry-funded studies COUNTRY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDEPENDENT STUDIES SOURCE OF THE INDEPENDENT STUDY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDUSTRY- FUNDED STUDIES INSTITUTION RESPONSIBLE FOR THE INDUSTRY- FUNDED STUDIES Colombia 3.5% of the total market in five cities, 2016 Maldonado et al., 2018 (37) 13% of the total market, 2014 FND and INVAMER, 2015 Chile 16.3% of the total market in the Metropolitan Region of Santiago, 2017 Paraje et al., 2020 (38) 24.3% of the total market, 2017 Observatorio del Comercio Ilícito BATC, 2017 Brazil 28.8% of the total market, 2014 Iglesias et al, 2017 (39) 34%, of the total market, 2014 BAT public statement, 2015 Mexico 8.8% of the total market in eight major cities, 2017 Saenz de Miera Juarez et al., 2020 (40) 16.6% of the total market, 2012 Confederación de Cámaras Industriales, 2012 As shown in Table 4.4, methodologies to measure illicit trade can be grouped into three types: (1) direct measurement; (2) residual methods and (3) expert opinion (12). Direct measurements rely on evidence directly linked to actual illicit behaviour and pack observation; residual methods infer evasion based on theory and evidence about consumption and legal sales; and expert opinion distills information garnered from talking to individuals with the most direct knowledge of the tobacco market. Each method has advantages and disadvantages. No single method is unambigu- ously superior to others, but direct measurement and residual methods are more conducive to determining the size of the illicit market, whereas expert opinion could provide insight into the details of the market’s operations.15 Table 4.4 presents the relative amount of resources and the degree of expertise required to implement each main measurement method, as well as the primary purpose, data collection characteristics, sampling features and unit of analysis. A brief description of each of the methods is presented in Annex 4.1. Merriman (11) and Ross (9) provide more expansive details. There is no simple selection rule for deciding what measurement method to use. The major factors to consider when selecting a method or methods include (1) the nature and characteristics of the illicit trade problem (i.e. where and how the 15 In interviews with experts from the tobacco industry, provisions of Article 5.3 of the FCTC and its Guidelines need to be followed. 190 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N problem manifests and whether domestic tax evasion or illegal inflows of foreign brands or a combination of both predominates), (2) previously collected data, (3) available budget and (4) expertise of available analysts. Available budget and staff skills are often the main restrictions that governments face. Therefore, Table 4.4 orders the measurement methods according to resources needed and available expertise. For example, residual methods and expert opinion can provide crude but useful estimates at low cost and require the lowest levels of technical sophistication. Another low-cost option for countries that employ popula- tion health surveillance surveys is to add questions to measure illicit trade, such as brand name, value and quantities of the last purchase. In contrast, the direct measurement approach often requires sophisticated research designs and expensive (and time-consuming) field research. Table 4.4 Overview of resources and expertise needed and main purpose of measurement methods METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Seizures (D) $ Low Identify trends in types of products, transporta- tion methods, points of entry and brand names Secondary data use Non- probability Shipments Use of existing health sur- veillance surveys – self- reported consump- tion (D) $ Low, only additional questions Size of illicit trade, adding or improving questions on brands, value and quantities of the last purchase Additional primary data collection Probability Individuals Gap analysis (R) $ Medium Provides a measure of changes in illicit trade Secondary data use Universe Nations Econo- metric modelling (R) $ High Estimation of price elasticity of substitution from tax- paid to illicit products Secondary data use Universe Geography Expert interviews (E) $ Low Characteristics of the illicit trade Primary data collection Non- probability Individuals Smoker intercepts and pack observa- tion surveys (D) $$$ Medium Size and characteristics of illicit trade, probability-based sample to be representative of population Primary data collection Probability Individuals CHAP T ER 4. PO LI T I C AL ECO N OMY 191 METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Pack return and swap surveys (D) $$$ Medium Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Littered- pack surveys (D) $$$ Medium Size and characteristics of illicit trade, compa- rability with industry estimation using empty-pack surveys Primary data collection Probability Individuals Covert purchases (D) $$$ Medium Type of products and trade channels of illicit trade Primary data collection Probability Geography Self-report consumer surveys (D) $$$ High Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Notes: Universe includes total population; D = direct measurement, R = residual method, E = expert opinion. Scale for resource costs assuming a moderately sized study (e.g. a representative study of a region of several million): $ (cheapest) – weeks of skilled labour hours; $$ (moderately expensive) – 1 to 2 months of skilled labour hours; and $$$ (most expensive) – 6 to 12 months of skilled and unskilled labour hours. A more detailed description of the different measurement methods is given in Annex 4.1. To further assist responsible authorities in deciding which method to select, Table 4.5 presents the key characteristics of each of the measurement methods, along with the main advantages and disadvantages of each. Countries may begin with methods that require fewer resources and less skills to obtain an overview of the problem. Seizures – which are a by-product of law enforcement efforts – pro- vide a first step, and countries can analyse the information obtained (origin of the products, brands, location, etc.) and report the results to increase public awareness of the problem.16 Alternatively, countries can add questions related to illicit trade to existing and funded health surveillance surveys conducted regularly by health surveillance authorities and statistical authorities. In that way, cooperation in using existing measurement methods between health authorities – the tobacco control office and health surveillance unit – tax and customs authorities and the national 16 Seizures are useful for obtaining qualitative information about the illegal activity, but they have to be treated very cautiously in projecting the size of the problem. Countries may think they have a very large problem because they have competent authorities doing an extraordinary job at finding illicit goods. On the other hand, countries can have less-efficient authorities making few seizures, and in these environments, seizures tell nothing about the size and nature of the problem. 192 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N statistical office could be a starting point for identifying the nature and size of the illicit trade problem in the country. Direct observation of packs has been increasingly implemented in many LMICs, through different types of surveys such as intercepts of smokers or retailers, pack return, littered-pack inspections and covert purchases of cigarettes. These activities have expanded the skills of independent researchers and academia and increased knowledge of these methods. Also, increasingly cheaper digital technologies allow interviewers to take pictures and record pack characteristics in direct observation surveys or in larger national self-report consumer surveys. Table 4.5 Key characteristics, advantages and disadvantages of illicit trade measurement methods METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Seizures (D) Statistics of tobacco products confiscated by local and national authorities Readily available from law enforcement agencies May not provide a representative picture of the size and/or nature of illicit trade Using existing health surveillance surveys to obtain self-reported consumption (D) Adding or improving questions about brand names, quantities, prices, locale of purchase and other factors Produces good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Gap analysis (R) Compare self-reported consumption data with observed (usually administrative) data about tax-paid sales When quality data are available, is simple and easily reproduced (providing for measurements over time) and explainable Data on tax-paid sales and/or consumption are frequently inaccurate and in many cases do not provide information on the size of the illicit market, but only on changes over time Econometric modelling (R) Estimated according to the difference between tax-paid sales and predicted consumption given by the model Because it is consistent with a long tradition of economic theory, empirical estimates can be evaluated Requires high-quality data on a variety of important variables over a period of time and advanced econometric modelling expertise CHAP T ER 4. PO LI T I C AL ECO N OMY 193 METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Expert interviews (E) Experts include researchers (e.g. in economics, criminal justice and public health), journalists, tax and enforcement specialists, product manufacturers and wholesalers Useful for identifying the nature of and trends in the marketplace (e.g. venues where illicit cigarettes are sold, modes of entry), and the interviews can be useful for defining the method to assess the size of the illicit trade Information obtained may not be generalizable, and expert knowledge may be outdated or limited by the experts’ experience; also, experts often have strong biases Smoker/retailer intercepts and pack observation surveys (D) Examining the packs of smokers and cigarette retailers, convenience or probability-based sample Is direct and objective, and smokers do not suffer from any value judgements The difficulty of identifying areas representative of the tobacco use population and sampling important subpopulations such as elderly and immobile smokers, but household surveys could overcome sampling issues Pack return and swap surveys (D) Also a pack observation study using survey sampling techniques to examine smokers’ pack characteristics May decrease the stigma associated with traditional smoking surveys In LMICs, survey distribution may be unreliable because of the mail delivery system Littered pack surveys (D) Also known as empty-discarded-pack surveys; publicly discarded packs bear characteristics (e.g. tax stamps, public health warnings) that indicate whether they are tax compliant Yields estimates that are less likely to be biased from issues of social desirability, recall error and confidentiality Significant budgets could be needed to employ field researchers to collect, code and analyse the data; surveys do not provide information about the smoker and the price paid Covert purchases (D) Uses covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes Directly identifies sources of illicit cigarettes It is difficult to create a sampling frame of retailers for illicit sources or to know what smokers are actually buying and how much Self-report consumer surveys (D) Surveys can be distributed to individuals or households, using various modes of distribution Good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Notes: D = direct measurement, R = residual method, E = expert opinion. 194 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ultimately, when capacity allows, more solid estimates will need to be made using more than one methodology at a given point in time. Ideally, estimates will be made on a regular basis in order to assess the evolution of illicit trade over time and its possible connection to policy changes. 4.1.5 CRITIQUING STUDIES THAT MEASURE THE SIZE OF THE ILLICIT MARKET Measuring illicit trade is a challenge for researchers, industry and governments, because the trade is, by definition, hidden from plain sight. Buyers and dealers sometimes go to great lengths to ensure that their participation in illicit activity is concealed. Over the years, researchers and government agencies have been increas- ingly interested in estimating the size of illicit markets and identifying effective interventions. As consumers of research, governments should critically examine available studies and evaluate them on their scientific rigor and methodological transparency. Measurement issues are particularly acute with respect to the illicit tobacco trade because it is a politicized topic. High estimates may raise questions about the tobacco industry’s ability to control the supply chain, its involvement in illicit diversion, the impact of taxation policies and the effectiveness of enforcement strategies. While the industry has portrayed itself as taking an active stance in measuring and fighting illicit trade (e.g. Project Star, conducted by KPMG LLC but paid for by PMI, later followed by Project Sun), in the past it has used smuggling as a strategy to enter closed markets – for example, in China and Russia (19, 41). Govern- ments should carefully scrutinize evidence about the illicit tobacco trade produced by industry or quasi-industry sources and are advised to seek alternative evidence. Quasi-industry reports are studies commissioned by the industry but published by private research companies (e.g. Ernst and Young, Oxford Economics) (11). Characteristics of good analyses One of the main characteristics of a good analysis is scientific rigor, which involves the use of relevant theoretical frameworks, sound statistical methods and examination of the robustness of findings (e.g. sensitivity analyses). High-quality research reports provide transparent explanations about their methodology and statistical analysis steps undertaken, as well as supplementary analyses that established the robustness of the findings. For example, Joossens et al. (25) clearly describe the data sources used (limitations and advantages and where they can be found) and calculations performed on the number of lives that would be saved if the global market share of illicit cigarettes was eliminated. Explanations should be detailed enough to allow future researchers to scrutinize the analysis and replicate the findings. Replicability is another hallmark of good science. For example, littered-pack studies should detail where and when data collection took place, how many packs CHAP T ER 4. PO LI T I C AL ECO N OMY 195 were collected per geographical unit, the protocol of identifying the illicit packs (e.g. characteristics of the warning labels, brands, tax stamps, etc.) and details of statistical analyses. There should also be explanations of the representativeness of the selected geographical areas. Failure to provide this depth of information may call into question the generalizability of a study and whether there are faults with the chosen method. In the context of policy decisions regarding illicit trade, the most useful data provide information about a representative sample of individuals and geographies. Studies that are limited to, for example, one group of individuals based on specific characteristics or a given geography may yield biased information. Research reports also should be clear about the study’s limitations. For example, studies that measure illicit trade often do not measure product counterfeiting and do not include non- cigarette tobacco products in their estimates (42). When statistical estimates are included, they should provide confidence intervals as well as point estimates to account for uncertainty resulting from simple random chance (11). Characteristics of flawed analyses Flawed analyses can convolute and distort scientific knowledge about illicit trade. Flaws usually manifest in the data, methodology, statistical analysis and/or interpreta- tion of the results (11). Studies may be purposefully designed with methodological flaws to yield high or low estimates of the trade. For example, research showing that illicit trade constitutes a large share of the total market may be used to support arguments that taxes cause sharp increases in illicit trade, whereas lower estimates may be used to support arguments that certain governmental interventions (e.g. increased retail inspections) are effective. Pressures to skew data may also be tied to funding. For example, high estimates can sway governments to provide more resources for law enforcement activities. Analyses can be purposefully skewed by using data sources or data collection methods that will provide biased estimates. Flawed studies sometimes provide incomplete or inaccurate descriptions of their methodology. They may lack detail regarding the quality of the data used or information about how the data were collected and analysed. For example, a common weakness in industry-funded research on discarded packs is that the methods of collection and forensic analysis are not reported, ostensibly because they are “proprietary” information (43). However, these methodological details are key to assessing whether a study’s findings are biased by sampling error, model misspecification, measurement error, non-response or other flaws. It may be impossible to assess measurement error if researchers fail to disclose questions included in a survey instrument. Survey items used to measure the illicit tobacco trade may be imprecise. For example, asking respondents the frequency with 196 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N which they purchase “cheap” cigarettes may yield biased estimates, particularly if consumers can purchase cigarettes at discounted prices by using coupons. To more ac- curately measure tax evasion, surveys must include questions about the location of last purchase, purchase price, presence of public health warning labels and brand names. There are other ways that flawed studies can inadvertently or purposefully distort estimates of illicit trade. For example, data collectors can intentionally oversample areas known to be hot spots of illicit sales or sites that residents from lower-tax jurisdictions visit. Researchers can collect discarded cigarette packs close to the borders of countries with lower taxes to (inadvertently or purposefully) demonstrate the undesired side effects of tax policies. Studies published in non-peer-reviewed or lightly peer-reviewed outlets such as edited book volumes or policy briefs should be viewed with more scepticism than those published in highly regarded peer- reviewed outlets. 4.1.6 CONCLUSIONS Globally, the illicit tobacco trade continues to be a major concern for tax admin- istrators because of the challenges it generates to collecting higher revenues as well as the challenges to accurate and independent measurement. Industry figures provide distorted conclusions regarding the extent of the problem – frequently with a monocausal explanation of the link between illicit trade and tobacco taxation. Illicit trade comprises multisystemic issues and requires multiple strategies. Worldwide, countries at different levels of economic development have implemented a variety of effective measures to combat the illicit trade in tobacco products. The Philippines and the United Kingdom, for example, have addressed illicit trade as part of their overall tobacco tax reform (4). Price (and tax) levels are not a key determinant of illicit trade, the presence of which is exacerbated by the lack of tax administration capacity. Refraining from increasing taxes is not the solution; countries should instead respond with a com- prehensive strategy that includes at least these three main components: 1. It should identify – independently from the industry – the nature and dimen- sions of the problem. It is necessary to assess scientifically and with the best statistical practices the size of the illicit trade to understand the characteristics and scope of the problem. 2. It should identify and implement appropriate policies and strategies targeted at addressing the specific type of illicit trade the country is experiencing. It should address directly the country-specific institutional and/or governance challenges – as well as the lack of multilateral coordination that can exacer- bate illicit trade – and improve tax and customs administration practices as described in Chapter 3. CHAP T ER 4. PO LI T I C AL ECO N OMY 197 3. It should implement best practices contained in the WHO FCTC Protocol to Eliminate the Illicit Trade in Tobacco Products and accede to the Protocol if the country is not yet a Party. There are proper methods and policies with which to address the illicit tobacco trade. If countries start implementing the appropriate policies, they can raise tobacco taxes and reap health and revenue benefits even in the presence of illicit trade. 198 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.2 SCARE TACTIC C: COURT AND LEGAL CHALLENGES 4.2.1 INTRODUCTION The tobacco industry views well-designed and significant tax increases as a threat to the profit, growth and long-term sustainability of its business. As noted by PMI in 1985: Of all the concerns there is one – taxation – that alarms us the most. While [other restrictions] … do depress volume, in our experience taxation depresses it much more severely (44). The industry is, however, less likely to launch direct legal challenges to excise taxes than to other tobacco control measures (see Box 4.2.1 for details), because taxation – and excise tax in particular – is a comparatively well-established regulatory measure; in many jurisdictions, taxes have been levied on tobacco products for more than a century. There is also less unanimity in opposition to taxes among tobacco industry actors, because differences in the market position of different tobacco companies affect their interests in tax policy. This, in turn, decreases the likelihood that they will act collectively on the issue (45). BAT’s stated strategy in the early 1990s was to influence governments with regard to the level and structure of tobacco taxation in order to promote market growth and to secure competitive advantage (46). Nevertheless, tobacco industry actors will still legally challenge, or at least legally threaten, significant tax measures when vulnerabilities in their design, adoption or implementation are apparent. Box 4.2.1 Court and legal challenges to tobacco tax measures Evidence suggests that the tobacco industry and its allies instigate fewer legal actions against tax measures than against other tobacco control measures: 1. The Campaign for Tobacco-Free Kids’ tobacco control laws database contains only a handful of cases concerning tobacco tax measures, but hundreds on other tobacco control topics. This pattern can also be seen in a 2018 review of tobacco control legal challenges that examined this and two other databases to select 96 cases relevant to the question of the WHO FCTC’s usefulness in litigation (47). Only 6 of these 96 cases were challenges related to tax measures. CHAP T ER 4. PO LI T I C AL ECO N OMY 199 2. A 2013 systematic review of empirical studies on tobacco industry interference with tobacco tax policy found that only 9 of 36 relevant articles reported the specific use of litigation as a tobacco industry tactic (1). All 9 concerned constitutional challenges to earmarking provisions for tobacco tax initiatives in the United States (1). 3. A 2015 study on industry interference in LMICs cited legal challenges to tobacco control measures in 15 countries as examples of industry interference, but none of the challenges concerned a tobacco tax measure (48). 4. A 2016 analysis of papers published in systematic reviews of industry inter- ference with tax and marketing measures found that only 5 of 65 papers concerning tobacco tax related to the use of litigation or threats of litigation to interfere with tobacco tax measures (49). The tobacco industry makes extensive use of legal experts (1, 50–52) who study all relevant laws and regulations closely to determine their likely and arguable boundar- ies for the purpose of manipulating regulations and regulators (1, 50–52). Based on this expert advice, tobacco companies know when regulations remain within the bounds of both international and domestic obligations but can still argue that legally permissible tobacco control measures would be defeated in litigation if passed (48, 51–52). As the threat of a legal challenge alone can be used to the industry’s advantage, recourse to litigation is seldom needed or desirable (1, 45, 48, 51, 53–56). Even when litigation is launched, the objective may be to delay or weaken a measure rather than to win on the merits of the case (1, 45, 48, 53). To counter actual and threatened legal challenges, policy-makers need to be aware of relevant legal obliga- tions when preparing and implementing tobacco control measures. Fortunately, the tobacco industry playbook is relatively predictable. Tax and other tobacco control measures can thus be designed to strengthen the regulators’ legal position against genuine threats and enable them to dismiss baseless industry threats. 4.2.2 COUNTRY EXPERIENCES WITH LEGAL CHALLENGES TO TOBACCO TAXATION Legal obligations that are relevant to tobacco taxation include those under do- mestic law and international instruments such as international trade agreements and international investment agreements (IIAs).17 Some of the legal issues that a tax measure may encounter are outlined in Table 4.6. Case studies from various countries illustrate how these legal issues have and have not been avoided in the 17 Relevant international trade agreements include the WTO Agreement and custom unions such as the EU, the East African Customs Union and Mercosur. Relevant IIAs include bilateral investment treaties and the investment chapters in free trade agreements and within custom unions. 200 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N passage, design and implementation of tobacco taxes. These issues are not the norm, however, and should not give rise to undue apprehension. The case studies are rated as positive, mixed or negative based on the extent to which the legal decision upheld the taxation measure in question. Table 4.6 Potential legal issues for tobacco tax measures VULNERABILITIES LEGAL OBLIGATIONS CASE STUDIES Inadequate consultation and other procedural vulnerabilities Domestic procedural law 1, 2 Due process protections for investors under IIAs None Procedural requirements under WTO Agreements and Custom Unions 3 Discrimination against imports or investors Nondiscrimination obligations under WTO Agreements and Customs Unions 8, 9, 10 Nondiscrimination obligations under IIAs 11 Investment incentives or inducements Arbitration mechanisms under investor-state contracts 12 Fair and equitable treatment clauses of IIAs None Other substantive breaches Constitutional rights and restrictions on taxation 4 Statutory restrictions on the imposition of taxation 6 Expropriation clauses of IIAs 5 Ultra vires (the scope of legal authority) 7 Avoiding procedural vulnerabilities in tax laws Procedural defects can be avoided by taking great care in progressing and imple- menting regulatory or legislative provisions. Procedural concerns pose a dilemma for tobacco control regulators. Article 5.3 of the WHO FCTC and the COP guide- lines for its implementation state that policy-makers and regulators should interact with the tobacco industry only when and to the extent strictly necessary (57). For taxation measures, interaction might be necessary because consultative and de- liberative processes could be prescribed under domestic constitutional provisions and procedures for good governance, due process requirements of IIAs and some international trade agreements. The tobacco industry may use these requirements as leverage to delay, distort or hijack the rule-making process in contravention of Article 5.3. Accordingly, interactions with the tobacco industry should be limited to strictly necessary consultation conducted in a transparent or public manner but with care that this does not come at the expense of a measure’s defensibility. The proper balance will depend on the jurisdiction in question, since constitutional, statutory and applicable international legal obligations vary. CHAP T ER 4. PO LI T I C AL ECO N OMY 201 CASE STUDY 1 (MIXED): Industry manipulation of legislative procedures In 2012, a bill stipulating, among other things, the creation of a new specific excise tax on cigarettes passed its final reading in Costa Rica’s Legislative Assembly. Passage of the bill had, however, proceeded under “urgency” and notwithstanding a pending constitutional enquiry (a constitutional query is meant to prevent passage of a bill).18 ISSUE MAJORITY DECISION MINORITY DECISION LESSON Whether the court could consider the enquiry despite passage of the bill and the effect the bill’s passage could have despite the enquiry. The enquiry was taken up by the Supreme Court’s Constitutional Division’s majority (58). The signing and publication of the bill by the executive was suspended by the Constitutional Division pending their decision on the merits of the case – which, in the end, found any question of the bill’s constitutionality baseless (58). The enquiry was inadmissible by reason of having been filed too late and notice of its filing having not been received by the legislature prior to the reading of the bill (58). In disagreement with the majority, the minority held that the court could not consider the enquiry or suspend the bill’s signing by the executive – the final step in becoming law. This challenge demonstrates how the tobacco industry’s defenders may attempt to frustrate and impede a tax measure’s passage. In this case, the challenge seemed to have been a delaying tactic, as it was posted on the same day as the final reading of the bill. Its authors may have either wanted its pending nature to cause the legislature to delay or, as occurred, to create conditions for a procedural and constitutional challenge in the absence of delay. All the grounds of the challenge itself were found to be without merit. Although such frivolous challenges cannot be prevented, they can and should be anticipated to ensure that they do not lead to a tax measure’s defeat. 18 “Urgency” is a procedure under which a bill is progressed through a legislature in an expedited fashion. 202 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N CASE STUDY 2 (POSITIVE): Adhering to domestic procedural requirements Kenya’s tobacco control regulations required the tobacco industry to pay a levy to compensate the state for health care and other negative externalities of smoking. In a 2016 challenge brought against these regulations, the plaintiff, BAT, was unsuccessful on every count (59–60). Even though the levy was not considered a tax measure by the court, the case study is instructive on how regulators may safeguard tax measures against procedural challenges. ISSUES LAWS AND ARGUMENTS DECISIONS LESSONS Whether the government’s consultations on the measure were adequate. Asserting that the Constitution and the Statutory Instruments Act together meant that “appropriate consultations with persons who are likely to be affected” were required because of the measure’s likely substantial effect on business. BAT claimed that this standard was not met. Kenya’s government claimed that it was under no obligation to undertake special or extensive consultation with the tobacco industry. The judge found in favour of Kenya’s government, noting that (1) the requirement to consult does not imply that any particular view needs to prevail; (2) dissatisfaction with the level of consultation is not decisive; (3) on the facts, industry was allowed, and often invited, to send representatives to all relevant public consultative meetings and parliamentary committee hearings; and (4) consultation on the regulations was adequate (59). The tobacco industry carefully scrutinizes legislative and regulatory processes for defects. In this case, Kenyan government officials appropriately distanced themselves from the tobacco industry by not permitting its representatives special consideration but did permit their attendance at public meetings and the ability to submit their views under usual procedures. In this way, both the principles behind WHO FCTC Article 5.3 and the requirement for consultation under Kenyan law were observed. CHAP T ER 4. PO LI T I C AL ECO N OMY 203 Avoiding procedural issues in tax administration CASE STUDY 3 (NEGATIVE): Contravening procedural requirements in international obligations In 2010, a WTO panel held that Thailand violated the Customs Valuation Agreement (CVA) by the process it used to value cigarettes that Phillip Morris (PM) Thailand imported into the country from a related party, PM Philippines. Customs values are important as they are the tax base for tariffs and can feed into the base for other taxes levied against the value of the good, such as ad valorem excise taxes and VAT. Transaction values declared by PM Thailand were rejected by Thai tax authorities as influenced by the relationship between the parties and a customs value determined by deduction was substituted (61). ISSUE LEGAL OBLIGATION DECISION LESSONS Whether Thailand adequately consulted with PM Philippines before rejecting its declared transaction value (61). The CVA requires good faith exchange of reasons and information, with opportunities for response (61). Thailand had failed to properly explain its reasons for rejecting the transaction value, as well as its belief that price was influenced by the relationship between the two parties (61). This was a violation of the CVA. Thailand did not appeal these findings. Thailand’s authorities needed to take greater care in their dealings with the tobacco industry to ensure they met the pertinent procedural obligations. In this instance, a specific and high standard of consultation – the provision of detailed reasons and an opportunity for response – was prescribed by the CVA and Thailand failed to meet it. 204 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ensuring compliance with substantive requirements Rules found in domestic and international law also establish substantive obliga- tions. This subsection focuses on substantive obligations found in domestic law. International obligations concerning discrimination and investment incentives are considered in the next subsections. CASE STUDY 4 (POSITIVE): Tax measure found to be consistent with the Constitution The Chilean government introduced a substantial increase in tobacco and fuel excise, and in 1995, a coalition of taxpayers brought a Constitutional challenge to the measure (62). ISSUE DECISION LESSON Whether the tax was, per article 19 of the Chilean Constitution, “obviously disproportionate or unjust” (62). The excise tax increase did not violate the Constitution, as it was neither confiscatory nor manifestly irrational. Generally applicable excise taxes are not vulnerable to challenges for being excessive, unfair or disproportionate. CASE STUDY 5 (POSITIVE): Failure to grant tax rebates not an expropriation under an IIA This case study is an example of a claim for breach of an expropriation clause in an investment treaty. Such clauses protect foreign investors against measures that can be construed as directly or indirectly seizing an investment or depriving it of its value (63). In the case, an investor was, for more than a decade, denied tax rebates by the Mexican government. This affected the profitability of the business of purchasing and reselling Mexican cigarettes abroad, and the investor brought the claim to an investment agreement arbitral tribunal in 2002 (64). ISSUE DECISION LESSONS Whether Mexico’s failure to grant rebates to the investor exceeded the bounds of valid regulation to constitute indirect expropriation of the investor’s investment (65). There was no expropriation. The arbitral tribunal noted that not all business problems are violations: the investor had no right to participate in the “grey market” export of cigarettes and there were sound reasons to restrict that market (65). Further, the investor was able to participate in other business ventures and actually continued to have business success (65). Claims of indirect expropriation made under IIAs are unlikely to be successful, as generally applicable tax measures are a legitimate form of regulation. A mere loss of profit will not suffice. Claims of expropriation will not succeed unless a substantial or significant deprivation of the investment results. CHAP T ER 4. PO LI T I C AL ECO N OMY 205 CASE STUDY 6 (NEGATIVE): A regulation contrary to superior domestic legislation In 2011, an Indonesian tobacco industry association group, FORMASI, challenged a new excise regulation. Since 2009, the government had been implementing a tiered specific excise tax system based on a set of characteristics (size of production, type of cigarettes and price levels). In 2011, excise rates were increased in nearly all of the 19 tiers, but the reference prices were not accordingly adjusted. This gave rise to a legal issue. ISSUE DECISION LESSONS Whether new excise regulations breached a 57% ceiling for the rate of excise on the retail sale price of tobacco products under the superior Excise Law (66–70). The challenge specified that excise exceeded this ceiling for hand-rolled domestic clove cigarettes (kreteks) (68, 71). The Court found in favour of the tobacco industry association, and the government was required to immediately revoke the 2011 regulation. (69–70). It is advisable to stay within the rules and be aware of legal hierarchies – including superior domestic legislation. The tobacco industry scrutinizes all increases in tobacco taxes. In this case, a breach of a legislative requirement for a single category of tobacco product resulted in Indonesia suffering lost revenue and a setback in its efforts to reduce tobacco consumption. Ensuring a tax measure is within an authority’s legal power A tax measure is ultra vires when it goes beyond the legal power of the enacting body. As with case study 6, this is a legal issue that involves legal hierarchies. In ultra vires cases, however, instead of centring on conflict between inferior and superior law, the issue is whether an authority that enacts a tax measure is authorized to do so. This issue may arise when a tax measure is enacted by a subnational jurisdiction or by an executive acting under a statutory delegation. CASE STUDY 7 (NEGATIVE): Tobacco taxation contrary to the Australian Constitution ISSUE DECISION LESSONS Whether New South Wales’ licensing and penalty fees regime constituted an excise tax by other means contrary to the Australian Constitution’s exclusive grant of that power to the federal government (72). The court found that state licensing fees were excise taxes and that this was contrary to the Australian Constitution (72). Authorities enacting tobacco tax measures must act within the scope of their legal power. 206 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Avoiding discrimination against imports and foreign investments Although inherently discriminatory, customs duties may be used subject to the agreed upper limits in a country’s trade agreements. Excise and other taxes designed with the aim of raising tobacco prices to reduce demand and advance human health should be origin-neutral: they should not seek to tax local products less than foreign products or aim to treat foreign products differently from one another. Tobacco tax measures are pursuing objectives other than health when they aim to raise the price of imports more than that of local products or seek to burden favoured market participants less than others. Solely health-protective tobacco taxes will not ordinarily violate Articles III:2 and I:1 of the GATT (the WTO’s General Agreement on Tariffs and Trade), which prohibit discriminatory taxation (in light of general exceptions). Nor will solely health-protective tobacco taxes directly violate anti-discrimination protections for investors found in the national-treatment (NT), most-favoured-nation (MFN), expropriation and fair-and-equitable-treatment (FET) clauses of IIAs (63, 73–74). It is possible to make claims for breach of international obligations on grounds other than discrimination, but such claims are generally highly unlikely to succeed. CASE STUDY 8 (NEGATIVE): BAT v Uganda (2017 East African Court of Justice) DISCRIMINATION LESSON Uganda established a higher level of excise taxes on imported cigarettes – including those from Partner states of the East African Customs Union (75) – than on local cigarettes. Its implementation was discrimination contrary to Article 15 of the Customs Union Protocol (75). Differential taxation explicitly based on origin can be construed as protectionist discrimination in violation of international obligations. The tobacco industry can also turn to international trade agreements outside of the WTO – in particular, customs union mechanisms. Difficulties arise when ostensibly origin-neutral and health-protective tobacco taxes result in dissimilar taxation of tobacco products (73). Discrimination does not exist simply because there is dissimilar taxation – the taxation must adversely impact imported goods more than local products, the imports of one nation more than another or a particular investor’s products more than comparable products. Where dissimilar taxation between product categories results in discrimination, the tax will ordinarily still be lawful if the dissimilar taxation is based solely on a legitimate regulatory distinction between the product categories in question.19 19 The precise applicable rules vary depending on the nature of the legal obligations in question. Under the GATT, dissimilar taxation of like or directly competitive products can be justified based on scientifically grounded distinctions between products under Article III:2 and, in the alternative, discrimination that is necessary under the explicit carve-out for health-protective measures, Article XX(b) (73). For the MFN and NT clauses of IIAs, differential taxation can be argued as nondiscriminatory on the basis that difference in harm means the products are not “alike” or, in the alternative, discrimination is justified based on scientific evidence of differences in harm and rational reasons for the health-protective role of differential taxation (63). CHAP T ER 4. PO LI T I C AL ECO N OMY 207 Where discrimination is inadvertent, lack of an intention to discriminate is not sufficient as a defence for breach of obligations under IIAs or the GATT (73, 76). Policy-makers should carefully scrutinize measures to determine: 1. whether an aspect of a tax measure’s design or implementation may be more to the detriment of imports or foreign investors than of local products or domestic investors; 2. whether the potentially discriminatory aspect of the tax measure serves any useful purpose in supporting the tax measure (i.e. it is needed to achieve the health goal); 3. whether there is any reasonable alternative that could achieve the same effect without the potential for discrimination; (i.e. it is indispensable) and 4. when it is needed and indispensable there is a good chance that it will be defensible. The case studies below provide examples of discrimination arising in connection with a tobacco tax measure. CASE STUDY 9 (NEGATIVE): Thailand – Customs and Fiscal Measures on Cigarettes from the Philippines (2010 WTO panel) The facts of this case are presented in case study 3. This case study examines claims of discrimination rather than the procedural issues. DISCRIMINATION EXPLANATION LESSONS Thailand implemented its policy for determining the tax base for VAT on cigarettes inconsistently (61). Thailand applied a methodology in fixing the tax base, in particular a marketing cost component, of imported cigarettes that differed from that for local products (61). This resulted in the marketing cost component for the imported cigarettes being higher than it would have been under the general methodology. This difference in treatment was insufficiently justified and therefore considered discriminatory. As there is potential for inadvertent discrimination when the base for an ad valorem tax is fixed, tax base determinations must be consistent and well- reasoned (61). This case study demonstrates how policy- makers need to take care in designing and implementing ad valorem taxes to ensure they are nondiscriminatory and legally defensible. 208 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Thailand’s VAT rebate policy imposed a potentially higher tax burden and also created more burdensome administrative requirements for imported cigarettes (61). Resellers of cigarettes produced by a government entity were granted an exemption from VAT (61). Although resellers of imported cigarettes would be eligible for a tax credit on their VAT, this was not an automatic process (61). The distinct treatment of resellers of imported cigarettes and those of local cigarettes resulted in the risk that there would be a higher VAT burden for the former (61). The distinct treatment also imposed an additional administrative burden on resellers of imported cigarettes and altered conditions of competition (61). Rules for the collection and enforcement of tax obligations should be the same, or as similar as practicable, in both form and effect for domestic and imported tobacco products. CASE STUDY 10 (NEGATIVE): Dominican Republic – Measures Affecting the Importation and Internal Sale of Cigarettes (2004 WTO panel; 2005 WTO Appellate Body) Under article XX(d) of the GATT, discrimination that is necessary to secure com- pliance with a legitimate tax measure will be justified provided there is no less- discriminatory alternative. In this case, this justification was used unsuccessfully. DISCRIMINATION EXPLANATION LESSONS The Dominican Republic’s tax stamp regulations were discriminatory towards imported goods (77). Under the regulations, all cigarette packs had to be affixed with tax stamps, but imported cigarettes were to be affixed with tax stamps under the supervision of local tax authorities following importation, while locally manufactured cigarettes could be affixed with a tax stamp in the course of production. This de facto distinction between local and imported products modified the conditions of competition to the detriment of imported cigarettes by (1) increasing costs for importers and (2) impairing the aesthetics of imported products (77). The panel did not consider this discrimination justified: it was not necessary for the enforcement of tax measures, because less restrictive alternatives were available such as permitting importers to affix tax stamps during the course of production (Dominican Republic – measures affecting) (77). The panel’s findings were upheld on appeal (78). Policies crafted to ensure compliance with tax measures need to also be nondiscriminatory. Discrimination claims can arise when compliance costs are higher for imports than for local products and this de facto distinction is avoidable. It is important to consider whether less burdensome alternatives may achieve the same objective. CHAP T ER 4. PO LI T I C AL ECO N OMY 209 CASE STUDY 11 (NEGATIVE): Feldman Karpa v Mexico (2002 ICSID [International Centre for Settlement of Investment Disputes] Arbitral Tribunal) Arbitral tribunals have accepted differences in treatment accorded to investors protected by IIAs when there is a legitimate connection between the distinctions drawn and public welfare objectives (76). The facts of this case are presented in case study 5. This case study examines aspects of the case involving the investor’s claim of discrimination, rather than the substantive issue of expropriation. Claims of discrimination are made on different grounds than claims for expropriation, which is why the case was decided differently on this claim. DISCRIMINATION LESSONS Denial of foreign investors’ claims for tax rebates. Tax rebate claims were granted to similar local investors (65), which was a violation of an IIA’s national treatment clause (65). Foreign and local investors must be treated similarly, and consistent and well-documented policies must be used to guide administrative decisions. The denial of the rebates may have been justified, but the government was unable to establish this due to a lack of documentation. Avoiding the investment incentives trap Investor-state contracts between the tobacco industry and governments should be avoided. They are not merely “contractual” in the domestic law sense, as even in the absence of an applicable IIA, they can be internationalized to provide inves- tors the right to (1) remove dispute settlement from the state’s court in favour of independent arbitration and (2) remove the dispute from the state’s legal framework in favour of general principles of law (63, 76). Commitments under these clauses cannot, therefore, be legislatively moderated or extinguished, nor can liability be limited within domestic courts that may be more likely to favour the state’s right to regulate in favour of public health (76). Investor-state contracts and other noncontractual inducements can be further internationalized by umbrella clauses within IIAs. Such clauses make reneging on undertakings assumed towards investors a breach of the IIA (76). Moreover, even in the absence of an umbrella clause, contracts and inducement can underpin a claim for legitimate expectation and breach of fair and equitable treatment and can also strengthen an investor’s claim for indirect expropriation (63). Arbitral awards make clear that although taxes can be expected to vary and tobacco will be regulated, investors can have the legitimate expectation that states will abide by formal inducements and written contractual undertakings. A common clause within investor-state contracts, the stabilization clause, is ruinous to evidence-based tobacco control’s most effective measure: excise tax in- creases. Stabilization clauses purport to freeze specific domestic law from the time 210 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of investment (63). Seemingly less onerous, economic equilibrium clauses require contracting states to compensate for regulatory changes that negatively affect an investment’s value (63). There is little difference in effect between these two types of clauses: liability for the cost of breaching an equilibrium economic clause can be onerous enough to make it fiscally challenging and politically unpalatable. CASE STUDY 12: An investor-state contract A state entered into an investment agreement with a TTC in 2001 on the privatiza- tion of its state-owned tobacco enterprise and creation of a joint venture. This investment was to provide economic benefits under the agreement: the joint venture would increase exports and profit using the TTC’s cash and expertise while also ensuring prioritization of local employment, manufacturing and resources. The final investor-state contract included a form of economic equilibrium clause under which any increase in the excise tax rates applied to the company’s tobacco products before a set date would be compensable. While the agreement was not removed from the state’s law, it provided for independent arbitration in case of a dispute over its compensation. In addition, there is a bilateral investment treaty between the host state and another state in which the TTC’s subsidiary has residence that includes a FET clause – this could buttress, if needed, the protection provided by the stand-alone arrangements of the investor-state contract. There were similar less-formal inducements offered to a separate TTC. The extent to which incentives have been granted to the tobacco industry is unknown, but contracts and inducements are likely to be offered in the context of the privatization of state-owned tobacco interests and in dealings between investors and state-owned tobacco enterprises (63). Although countries have been entrapped by their incentives to industry, the investor-state contract provides the clearest example of how undertakings and inducements with the tobacco industry under- mine tobacco control (56, 79–80). States should avoid offering industry incentives and, in particular, entering into contractual undertakings with the industry. More systematically, government should consider avoiding IIAs that elevate incentives and inducements above sensible and reasonable regulation. 4.2.3 CONCLUSIONS Health-protective and origin-neutral tobacco excise taxes are legally defensible, and industry threats are usually baseless. There are, however, certain rules governing procedure, design and consultation that governments may need to consider: 1. Governments should be aware of the standard of consultation required under do- mestic law and any applicable international obligations (case studies 1, 2 and 3). CHAP T ER 4. PO LI T I C AL ECO N OMY 211 It is important to distance the tobacco industry from the policy-making process to the extent that this is permissible. Do not grant the industry special consideration, but do ensure that it is consulted with as required – for example, by providing public meetings, timely information and the ability to submit industry views – while being aware of potential procedural manipulation (case studies 1, 2 and 3). 2. Excise tax is generally safe from challenges that claim it is confiscation or expropriation under domestic or international law (case studies 4 and 5). But express limits on taxation can be found in other laws or a country’s constitution or in the limits of the power to tax granted to an authority (case studies 6 and 7). 3. Explicit and de facto discrimination against foreign tobacco products or investors must be avoided in the design, implementation or enforcement of tax measures (case studies 8, 9, 10 and 11). Legal issues may arise not from the tax measure itself, but rather from ancillary measures that support its implementation (case studies 9 and 10). 4. Explicit differentiation between products based on their effect on health may be challenged as discrimination if it falls heaviest on imported products and has to be justified on the basis of evidence of impact on health and a lack of alternatives. 5. Investment incentives in the form of inducements or contractual undertak- ings should not be offered, as these may be binding (case study 12) or may ground a challenge under an IIA; they are also contrary to the WHO FCTC Article 5.3 Guidelines. 212 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.3 SCARE TACTIC A: ANTI-POOR RHETORIC (REGRESSIVITY) 4.3.1 INTRODUCTION In their efforts to lobby against tax increases, the tobacco industry and its affiliates often claim that increases in tobacco taxation will hurt the poor (81–82). This argu- ment is based on the concept of regressivity in relation to taxation. Conceptually, a tax can be regressive if it means that lower-income people pay a greater proportion of their household income to meet the tax burden than do wealthy people. In other words, the tax burden tends to be relatively higher for lower-income households than for middle- and high-income households. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use. Second, higher tobacco taxes and prices can induce behaviour change among the population, as reflected in the price elasticity of demand (83–84). In combination, these broader considerations effectively make tobacco taxation a progressive – rather than regressive – public health intervention. 4.3.2 REGRESSIVITY AND THE BROADER PERSPECTIVE In a narrow sense, tobacco taxation can be seen as regressive because lower-income people must allocate a relatively greater proportion of their household income than wealthy people to pay for tobacco products when those products become more expensive following a tax increase. In many countries, people from lower-income groups use tobacco more than other people (85). A systematic literature review by WHO found a robust association between lower income and a higher prevalence of current smoking among adults, both men and women (86). This finding was consistent across three decades of studies, across most geographic regions and across countries of different income classifications. For example, in India, high rates of tobacco use – i.e. use by more than 30% of the adult population – are found only in lower-income states such as Assam and Odisha, where net state domestic product is still below 100 000 rupees per capita (see Fig. 4.3.1) (87). CHAP T ER 4. PO LI T I C AL ECO N OMY 213 Fig. 4.3.1 Relationship between adult tobacco use and net state domestic product per capita in states and union territories of India, 2016–2017 Source: (87). However, this finding does not account for broader health and economic factors that determine the full impact on households. Tobacco taxation can in fact be viewed as a progressive – or pro-poor – policy when these wider considerations are properly ac- counted for and explained. In terms of health concerns, the relatively high use of tobac- co among low-income populations translates into a much greater burden of tobacco- attributable diseases for these populations, including higher morbidity and mortality. Low-income groups are also less able to afford medical care to treat tobacco- attributable diseases, and large out-of-pocket medical expenditures can further impoverish many families. Consequently, many poor individuals do not get or even seek the medical care they need. One study found that in Bangladesh, 55% of patients diagnosed with a tobacco-attributable illness did not seek further medical care. This lack of health care utilization was attributed in part to prohibitively high out-of-pocket treatment costs (88). The combination of high rates of tobacco use and lack of access to affordable medical care means that tobacco use measurably contributes to the poverty rate in a number of high-tobacco-burden countries, including China and India (89–90). A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India 214 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The poor are also known to be more price-sensitive than the wealthy; lower-income smokers exhibit higher price elasticities than their higher-income counterparts. This is demonstrated in recent studies by the World Bank, findings of which are shown in Fig. 4.3.2 (91). The poor respond more strongly to higher tobacco taxes and prices by reducing their use of tobacco products more than others, and thus they benefit disproportionately in terms of avoiding tobacco-related deaths, diseases and associated medical costs. A similar conclusion was drawn in a systematic review of the population impact of tobacco control policies on socioeconomic inequities in high-income countries at the late stage of the tobacco epidemic (92). The review found 16 relevant studies relating to taxation, only one of which found a regressive association between tax and the social economic gradient (seven found a progressive impact, while the others produced mixed results). Fig. 4.3.2 Price elasticity of tobacco consumption, medium estimate, by decile Source: (91). This wider economic perspective is explained in the World Bank’s Extended Cost- Benefit Analysis (ECBA) framework, which assesses the distributional impact of tobacco tax increases on health, among other factors (82, 83). That is, the ECBA framework looks beyond the simple or partial definition of regressivity (i.e. impact El as tic it y Deciles -1.2 1 2 3 4 5 6 7 8 9 10 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina CHAP T ER 4. PO LI T I C AL ECO N OMY 215 on household expenditure by income levels) to capture the full distribution of benefits, including improved health and income. The ECBA framework has been applied in studies of various countries, including Bangladesh, Bosnia and Herzegovina, Chile, Indonesia, Republic of Moldova, South Africa, the Russian Federation, Ukraine and Viet Nam. The evidence from these studies supports the view that effective tobacco tax policies can generate pro-poor and welfare-improving outcomes. When reductions in medical expenditures and additional years of working life that result from lower smoking-related mortality are taken into account, the overall policy of tobacco tax increases becomes progressive rather than regressive (see Fig. 4.3.3) (84). A similar conclusion has been reached in studies of high-income countries, such as the United States, where a tobacco tax increase was enacted in 2009 (93). Fig. 4.3.3 Impact of a 100% price increase, with medium elasticities, by deciles Source: (91). Tobacco tax increases will also often lead wealthier smokers to contribute relatively more than poorer smokers to the overall amount of tax revenue collected. This is because poorer smokers reduce their consumption the most, since they are more price- sensitive and wealthier smokers also tend to purchase premium (higher-priced and In co m e ga in s (% ) Deciles -1 2 3 54 6 7 8 9 10 0 1 2 3 4 5 1 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina 216 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N taxed) tobacco products (94). Hence, higher tobacco taxes can be seen as progressive in terms of additional revenue collection and health gains resulting from cessation, as well as from preventing the young from taking up smoking in the first place. One study from China suggests that a 50% tax increase would raise US$ 703 billion over 50 years, with just 14% of this increase being borne by smokers in the lowest income quintile (95). In addition, the tax increase would yield a savings of US$ 24 billion in expenditures on tobacco-related diseases, with about 28% of these savings being enjoyed by smokers in the lowest income quintile. The ECBA framework actually presents a rather conservative profile of the net benefits of raising tobacco taxes, since it does not include other sources of gain, such as reduced harm from exposure to second-hand smoke, increased productivity and the potential for poor households to benefit from social programmes funded through increased tax revenues (96). Assessments of the distributive impact of the 2009 tobacco tax increase in the United States found that the overall progressivity of the increase was enhanced by the tax rev- enue being used to expand health insurance coverage for children of low- and middle- income families (97). Accounting for this expanded coverage added to the progressiv- ity of the overall legislative package, the bottom line being that the impacts are positive for lower-income quintiles and greatest, on average, for low-income households (93). Similarly, a large proportion of the tobacco tax revenues from the Philippines’ so-called Sin Tax Reform was used to subsidize universal health coverage (UHC) for poor and near-poor families. Globally, 37 countries are known to earmark some tobacco tax revenues for health programs, with many of these programs indirectly benefiting the poor and less-advantaged disproportionately more than other groups (27) (for details on earmarking, see section 4.6). 4.3.3 CONCLUSIONS Contrary to the perception of tobacco taxation being regressive, it is a strong pro- poor policy when the broader economic impacts are taken into consideration. The tax burden is not a complete indicator of regressivity, since it does not include the negative health and economic impacts of tobacco-attributable diseases or the positive impacts of behaviour change in response to tax and price increases. The health and economic burdens of tobacco-attributable diseases fall dispropor- tionately on the poor, who tend to have higher tobacco use and are also the least able to afford the necessary medical care. Because the poor tend to be more price-sensitive, they curtail their use and consumption more significantly than wealthier smokers in response to tax increases, which in turn reduces their downstream health and economic costs. Tobacco taxation can be made even more progressive by earmarking or allocating tobacco tax revenues for social goods and services that benefit the poor (see section 4.6). CHAP T ER 4. PO LI T I C AL ECO N OMY 217 4.4 SCARE TACTIC R: REVENUE REDUCTION 4.4.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases result in reduced tax revenues for the government. According to them, the reduction in revenues is caused either by substitution to cheaper, lower-taxed or smuggled tobacco products or by reductions in consumption overall (98–99). The tobacco industry often refers to the Laffer curve to make this argument. According to this curve, revenues increase along with tax rates up to a certain point, after which further increasing tax rates leads to declining revenues. When considering tobacco taxes, the tobacco industry assumes that countries are already approaching or are even beyond the critical tax rate level (98). However, the argument rests on a narrow theoretical and empirically unsubstanti- ated foundation (98–100). The price inelastic demand for tobacco and the relatively low tax share in prices in many countries explain the win-win for public health and finance, i.e. that declines in consumption and increases in revenues can occur simultaneously (98, 101). Furthermore, many country examples (see case studies below) demonstrate that well-designed and well-implemented tobacco tax increases lead to increases in revenue, at least in the short to medium term (98, 100). Although consumption will diminish with a tobacco tax increase, the percentage increase in excise tax per unit is greater than the percentage decrease in tobacco consumption, cancelling out at least some of the effect of reduced consumption on revenue (98–99). A change in the tax rate, with all other factors influencing consumption kept constant, corresponds to a change in the tax revenue and is represented by a move- ment along the Laffer curve. As the tax rate changes, so does the elasticity of the tax base; each point on the Laffer curve corresponds to a different tax base elasticity. When one or more of the other factors changes, this affects the position of the curve, and the tax base elasticity changes at a given tax rate. For example, a successful smoke-free policy or advertising ban that reduces the demand for tobacco shifts the curve down, reducing the tax revenue potential for each tax rate. To demonstrate that few, if any, countries are beyond the revenue-maximizing point on the Laffer curve, Table 4.7 shows the revenue impact of increasing excise taxes under different scenarios, using different price elasticities of demand, different levels of tax increases and different starting tax shares, depending on country income levels. This is the tax base elasticity approach from which the Laffer curve is derived (for more details, see section 2.2.3 and Annex 2.2). The total and excise tax shares shown are weighted averages for each country income group, calculated from the RGTE dataset. The revenue gains were simulated using progressive levels of excise tax increases (25%, 50%, 75% and 100%) and varying price elasticities of demand (-0.4 to -1.2). 218 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.7 Percent increase in excise revenues under different scenarios of tax levels, tax increases and price elasticities20 Income group Total tax as % of retail price Excise tax as % of retail price Increase in excise tax Increase in excise revenue when price elasticity of demand is: -0.4 -0.6 -0.8 -1.0 -1.2 LOW INCOME 38% 22% 25% 22% 20% 19% 17% 16% 50% 43% 39% 36% 33% 29% 75% 63% 57% 52% 46% 41% 100% 82% 74% 66% 59% 51% MIDDLE INCOME 58% 41% 25% 19% 17% 14% 11% 9% 50% 37% 31% 26% 20% 15% 75% 54% 45% 36% 27% 19% 100% 71% 57% 45% 34% 23% HIGH INCOME 68% 55% 25% 18% 15% 11% 8% 5% 50% 35% 27% 21% 14% 8% 75% 50% 39% 29% 19% 10% 100% 65% 50% 36% 23% 11% Source: Authors’ calculations using data from the RGTE (27).21 Substantial revenue increases occurred in all the scenarios that were considered in the simulation. These results reaffirm much of what is already known, i.e. that higher tax increases generate higher revenue gains, and that these gains increase with the increasing inelasticity of demand. Even when demand is relatively price elastic (-1.2), the simulation predicts a gain in revenue. The tax share in price also affects revenue potential. The lower the tax share in price, the larger the revenue potential. This suggests that revenue reductions as a result of an excise tax increase will occur only if the scenario is extreme (i.e. a very elastic demand coupled with a very high current tax share). It is important to note that the vast empirical literature 20 These projections use 2018 data from 185 countries. The countries were classified according to World Bank income group, with the average total tax share, excise tax share and VAT/sales tax share for each country weighted according to the number of current adult cigarette smokers. To calculate the projected revenue for each stated elasticity, it was assumed that there would be full pass-through of the excise tax increase, along with constant percentages of non-excise taxes (VAT/sales tax) as a share of the retail price. The consequent changes in price were multiplied against the respective elasticities to derive the expected change in consumption. The projected revenues could be easily computed by multiplying the new consumption figures against the increased excise tax rates. 21 These calculations do not take into account brand substitution (cross-price elasticities), income ef- fects or illicit trade. The excise tax was assumed to be a specific tax, while the non-excise taxes (VAT and others) were bundled and treated as an ad valorem tax with retail price as the tax base. The difference between retail price minus all taxes was also assumed to be constant, with full pass-through of the tax increase to consumers. CHAP T ER 4. PO LI T I C AL ECO N OMY 219 shows tobacco to be universally inelastic; thus the extreme scenario should not be given credence by policy-makers. Furthermore, as demonstrated by the data in Table 4.7, tax shares in most countries are relatively low and reinforce the revenue potential of tobacco tax increases. The revenue potential of tobacco taxes is indeed quite significant. It is estimated that in 2018, excise taxes on cigarettes generated a total of US$ 361 billion in revenues worldwide, including US$ 162 billion in LMICs. If all countries were to raise excise rates by the equivalent of US$ 1 per pack of cigarettes, the amount of excise revenue would increase by between US$ 178 billion and US$ 219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from these tax increases, with excise revenues in these countries increasing by US$ 133 billion to US$ 167 billion, or by 82–103%.22 Revenue reduction in the countries examined was due to other causes, not the tax increase per se. For example, Tonga significantly increased its excise tax on cigarettes in 2016 and saw a very sharp decrease in its consumption (40% decrease), followed by a revenue decrease. This occurred because 20% of smokers switched to an untaxed, cheap local loose tobacco product called Tapaka Tonga (102). The lesson learned was that Tonga needed to tax all its tobacco products at the same level to avoid substitution to lower-price/untaxed tobacco products. Another example of revenue decrease that was not related to tax increases but rather to tax administration mismanagement is the case of South Africa (see explanation in the case study later in this section). Finally, declines in revenue due to long-term declining trends in tobacco use should not be confused with being beyond the revenue-maximizing point of the Laffer curve. For example, in the United Kingdom, where long-term declines in tobacco use are being experienced, a nominal decline in revenues occurred between 2017 and 2018 even though excise taxes remained unchanged. Conversely, even countries with very high tobacco excise rates experience increases in revenues as a result of tobacco tax increases (see the case study of Australia below) (98, 100). This suggests that few countries, if any, are beyond the revenue-maximizing point on the Laffer curve. Tobacco consumption is expected to be tax inelastic, even if demand becomes effectively price elastic as a result of successful tobacco control interventions. Taxation serves as an instrument for both fiscal and public health objectives. If after successful tobacco control interventions, prices reach levels where demand is elastic, the tax base is still most likely to be inelastic due to tax undershifting, since overshifting is not a good pricing policy when demand is elastic (for a more detailed discussion on the shifting of tax, see section 2.2.2). In other words, a tax rate increase in combination 22 Goodchild M, Perucic AM, Paul J. Tobacco taxation as a strategy to achieve global targets for smoking prevalence. Unpublished manuscript. October 2020. 220 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N with non-price tobacco control measures, which make consumers more sensitive to price (tax) increases, leads to declining but still positive marginal revenues. In the long run, tobacco control policies, including price and tax measures, may be so successful in reducing consumption that revenues will plateau or fall. This is ultimately the long-term policy goal. Reducing the impact of the policies and ending the global tobacco epidemic is the aim of tobacco control and not something to be avoided. However, ending the global tobacco epidemic is unfortunately not foreseen in the short to medium term; therefore, governments can currently rely upon tobacco taxes as a reliable source of revenue (103). 4.4.2 THE REVENUE IMPACT OF EXCISE TAX INCREASES: CASE STUDIES The following case studies illustrate four key points: (1) large and regular tax increases result in large and consistent revenue increases; (2) countries with high taxes and falling prevalence of tobacco use can still increase revenue with tax increases; (3) countries that reduce taxes experience revenue declines; and (4) countries that increase taxes in the face of illicit trade still increase revenue. Large and regular tax increases usually mean large and consistent revenue increases South Africa’s experience shows how successive tax increases, well above inflation and year after year, generate additional revenues even after taxes have been increased substantially. After two decades of declining real revenue in the 1970s and 1980s as real excise per pack declined, South Africa implemented successive excise tax increases from 1994 until 2011 (Fig. 4.4.1) (98, 104). After adjusting for inflation, this resulted in a real excise tax revenue increase of 245% (98). Revenues began to plateau from 2012 as tax increases stalled. They began to decline after 2015 – not due to tax increases, however, but due to a dramatic decline in administrative capacity and enforcement measures exacerbated by large-scale corruption in the government, including the tax administration authority (105). The rapid and catastrophic decline in tax administration and enforcement has been the subject of much attention (106). CHAP T ER 4. PO LI T I C AL ECO N OMY 221 Fig. 4.4.1 Real excise tax per pack of cigarettes and real excise tax revenue in South Africa, 1961–2020 Source: Data shared by University of Cape Town, 2020. Similarly, the Philippines provides a compelling example of how large and regular tax increases alongside reforms to tax structure can lead to large and consistent revenue increases – in this case, also through an accompanying reform to the tax structure (Fig. 4.4.2) (98). The 2012 Sin Tax Law consolidated the country’s four tax tiers into two by 2013 and established a uniform structure by 2017. The same law provided for large, progressive increases across the board, but in particular for the lowest tax categories (98). Not only were the revenue gains substantial, they exceeded all the projections for 2013–2017 made prior to the law’s passage (98). Excise tax per pack Excise tax revenue Ra nd s pe r p ac k (c on st an t 2 02 0 ra nd s) Excise revenue (constant 2020 rands) 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11 20 13 20 15 20 17 20 19 0 0 5 6 10 10 15 14 20 18 222 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.2 Real excise tax per pack of cigarettes (multitier, two-tier, unitary) and total tobacco real excise tax revenue in the Philippines, 2012–2018 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2012 as the base year. Sources: (107, 108 and data shared by the Philippines Department of Finance, September 2020). Ukraine is another example of a country that has regularly increased taxes over the past 10 years and has experienced increased revenues along with decreases in consumption and the number of smokers. Figure 4.4.3 the shows the trends in excise tax, revenues, cigarette sales and number of smokers in 2008–2017. Increases in excise rates were consistently accompanied by increases in revenues. In 2014–2015, excise tax was not increased above inflation (and inflation, especially in 2015, was very high, at 48.7%), so real values of excise and revenues went down. But it is evident from the data that revenues closely follow the path of excise levels even when sales go down. 12 Re al e xc is e ta x pe r p ac k, P hi lip pi ne p es os (2 01 2 ba se ) Billions (Philippine pesos) in tobacco real excise tax revenue (2012 base) 00 20 60 100 140 2012 HIGH PREMIUM SIN TAX LAW RA 10354 TRAIN LAW RA 10963 UNITARY RATE 2013 2014 2015 2016 2017 2018 5 10 15 20 25 30 32 .1 9 68 .6 6 71 .0 9 93 .5 2 87 .2 0 96 .0 1 11 4. 39 24.4 25.4 26.2 26.8 26.9 27.5 11.7 16 23.1 19.6 LOW Real tobacco excise revenues Dierent levels of real tobacco excise tax per pack MEDIUM 7.6 2.7 28.3 CHAP T ER 4. PO LI T I C AL ECO N OMY 223 Fig. 4.4.3 Average real cigarette excise tax rates, real cigarettes excise tax revenues (base year 2008) and cigarette sales and number of cigarette smokers in Ukraine, 2008–2017 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2008 as the base year. Source: Data provided by Konstantin Krasovsky, July 2020. Countries with high tax and falling prevalence of tobacco use can still increase revenue with tax increases Countries with already high tobacco taxes and rapidly diminishing tobacco use can still increase revenue by increasing taxes (98, 109). Australia has implemented comprehensive tobacco control policies and enacted consistent tobacco tax increases on top of what were already some of the highest tax rates in the world (see Fig. 4.4.4). Between 2001 and 2010, revenue increased with increasing tax rates, but in real terms (inflation-adjusted) it remained static (109). Then, in 2010, a 25% excise tax increase was introduced, with large annual increases scheduled from 2013 onward (98, 109). The result of this tax policy has been consistent and large increases in revenue year after year for nearly a decade, even when the increases were being made on already high tax rates.23 23 The apparent reduction in revenues in 2012 and 2013 was due to a change in the source of the data for 2001–2011 and 2012–2016. Data for 2012 and 2013 do not include customs duty, while all other years do. 73 95 125 112 88 82 75 76 67 11.8 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Number of daily smokers, in millions Real tobacco excise revenue, in billions of Ukrainian hryvnia Real average excise per cigarette packs of 20, Ukrainian hryvnia Number of taxed cigarettes (sales), in billions of sticks 0.6 3.6 10.1 9.2 8.7 8.6 8.4 8.1 7.3 6.2 6.5 6.3 7.8 10.3 11.2 12.1 13 1.4 2.2 2.5 2.9 3.5 3.2 2.7 3.3 4 12.7 13.4 74 9.7 224 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.4 Real excise tax and customs duty per stick and real total revenue (all tobacco products) in Australia, 2001–2019 Notes: Rates published by Australian Taxation Office and Australia Department of Immigration and Border Protection, adjusted using Australian Bureau of Statistics Consumer Price Index rates. The 2011–2012 and 2012–2013 figures do not include customs duty, which explains the apparent decline in revenue. Using official disclosures, Scollo and Bayly estimate that duties in these years were $7397.2 and $7687.2 respectively (110). Sources: (109, 110). Countries that reduced taxes and saw revenues decline Prior to 1982, Canada lowered taxes on cigarettes and experienced declining revenues as well as increased smoking – particularly among youth. Subsequent fivefold in- creases in cigarette taxes between 1982 and 1992 resulted in more revenue, increases in retail price and substantial reductions in consumption, with teenage smoking declining by nearly two thirds (5). In the early 1990s, a growing illicit trade in ciga- rettes emerged in which Canadian cigarettes exported to the United States were then smuggled back into Canada (5). The tobacco industry – which was later found to be complicit in and profiting from this illicit trade – sought to frame Canada’s high tax rates as the cause of smuggling (111–112) and succeeded in convincing the federal government, as well as six provincial governments, to make massive reductions in the tobacco tax (111–112). As a result, federal tax revenues fell significantly – more than twice as much as the government had predicted – and smoking rates among both adults and youth began to increase (5, 112). The Canadian government later changed its strategy, and the federal excise tax was restored, resulting in increased Specic excise/ duty per stick (cigarettes/cigars less than 0.8 g) Total revenue (all tobacco products) To ta l c us to m s/ du ty re ve nu e (in a tio n, a dj us te d, 20 19 A us tr al ia n do lla rs , m ill io ns ) Total excise/duty per stick (ination, adjusted, 2019 A ustralian dollars) 0 3 000 6 000 9 000 12 000 $0.8 $0.6 $0.4 $0.2 $0 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 CHAP T ER 4. PO LI T I C AL ECO N OMY 225 revenues and decreased smoking (5). Canada’s focus then shifted to using customs enforcement, rather than tax rates, as the best means of countering illicit trade (5). Countries that increased taxes in the face of illicit trade and still increased revenue As discussed in section 4.1, the tobacco industry exploits illicit trade as a strategy to undermine tobacco tax policy, with the goal of deterring governments from increasing tobacco taxes. The narrative that has been created is that higher tobacco tax rates result in increased illicit trade and undermine the policy goals by resulting in lower (or no) declines in tobacco use or lower (or no) increases or even decreases in revenue. However, as shown in section 4.1, the empirical evidence does not sup- port the industry arguments. Furthermore, the evidence shows that the industry and its allies have consistently overstated and exaggerated the scale and extent of illicit trade (see section 4.1). As was the case in Canada, Brazil’s tobacco tax policy suffered from a fear that the illicit market would expand unless it was undercut by price competition in the legal market, which it was thought could be best encouraged through tax cuts (20). Real excise tax rates declined from 1999 until the mid 2000s, as nominal increases were below the rate of inflation. This resulted in declines in real tax revenues (20). In these years, the tobacco industry used the tax cuts to increase profit margins rather than decrease prices and outcompete the illicit market, while also exaggerating the size and scope of the illicit trade problem (20). This caused the industry’s argument on illicit trade and revenue to lose credibility and resulted in increases in tax rates from 2007 onwards, with a major reform passed in 2011 (20). Tobacco excise rates and minimum prices were scheduled by the law to increase at levels above expected inflation from 2011 until 2015 (20). This resulted in substantial increases in the tobacco excise per pack, as well as overall revenue, which by 2015 had more than doubled from its low point in 2013 – equating to more than 50% in real terms (see Fig. 4.4.5). The success of this reform shows that revenues can be increased by higher rates despite the presence of a sizeable illicit market (113). More recent data show that revenues in Brazil declined in 2015 and 2016, coinciding with an increase in illicit trade, but also with an exceptionally bad economic recession that saw GDP decline by more than 3% in those years. 226 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.5 Average real excise tax per pack, real excise tax revenues and domestic cigarette sales in Brazil, 1999–2014 Notes: Data based on domestic sales and tobacco excise revenues, Federal Revenue Secretariat, indexed to 2013 Brazilian reals, using Consumer Price Index. Revenue collection indexed to 2013 reals, using Consumer Price Index. Source: (20). Improvements in tax administration and enforcement can also generate increases in revenues. In Kenya, several measures, including fiscal markings and, later, an advanced tracking and tracing system, improved collection, resulting in increases in both legal sales and tax revenues and a reduction in illicit sales (114). Moreover, these examples of poor governance indicate that attention should be focused on countries where a significant loss in administrative and enforcement capacity un- dermined revenue collection. 4.4.3 CONCLUSIONS The tobacco industry uses revenue concerns as a SCARE tactic to avoid, dilute and/or delay tobacco tax increases. The argument that higher taxes will decrease revenue is theoretically plausible, but real-world examples have demonstrated that this has not occurred. Furthermore, simulations show that even large tax increases in current average tax shares yield substantial revenue gains. The use of the Laffer curve by the tobacco industry should be challenged and refuted. The relatively price inelastic nature of cigarette demand combined with the Real excise tax amount per pack Domestic sales Real excise tax revenue D om es tic s al es (b ill io n pa ck s) Re al e xc is e ta x re ve nu es (b ill io n 20 13 re ai s) Real excise tax am ount per pack (2013 reais) 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 0 1 2 3 4 5 6 0 1 2 3 CHAP T ER 4. PO LI T I C AL ECO N OMY 227 low tax share and no overshifting of the tax means that most – if not all – countries are still far from the revenue-maximizing point, indicating that increases in taxes will lead to increases in revenues. The case studies in this section refute each of the tobacco industry’s arguments regarding alleged potential revenue loss due to tax increases. The experiences of South Africa, the Philippines and Ukraine demonstrate that large and regular tax increases result in large and consistent revenue increases. Well-designed tax structures have also proven to play an important role in generating revenues. The experience of Australia shows that even countries with already high tax rates and declining prevalence of tobacco use can increase revenues with regular, large tax increases. The experience of Canada warns against following the advice of the tobacco industry to decrease taxes as a way to fight illicit trade. It demonstrates clearly that decreasing tobacco taxes will decrease revenue and encourage consumption, rather than counteract illicit trade. The experience of Brazil shows that countries with substantial illicit trade issues can still increase revenue by increasing taxes. Finally, in the few cases where revenue decreases were seen, the reasons for the decreases were not strictly linked to tax increases. This was the case in Tonga, where the increase in tax was applied only to cigarettes and not to their close substitute, loose tobacco – leading smokers to switch products. In South Africa, a decrease in revenue was the result of the weakening of government institutions. And in Ukraine, real revenues decreased only during the two years when taxes were not increased above inflation. 228 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.5 SCARE TACTIC E: EMPLOYMENT 4.5.1 INTRODUCTION In opposing tax increases, the tobacco industry often seeks to frame tobacco taxes as an economic rather than a public health issue (5, 48, 53, 115). Particular emphasis is placed on the alleged threat that tax increases pose to employment in tobacco farm- ing and manufacturing, as well as other related industries (5). This so-called choice between health and jobs is, however, based on several false premises, including (5): 1. tobacco is a significant source of jobs within the context of broader labour markets, and domestic tobacco tax increases will have a drastic effect on domestic employment (48, 53, 116–117); 2. tobacco consumption is an indispensable engine for job creation (5, 48, 54); and 3. tobacco provides highly prosperous, sustainable and irreplaceable livelihoods (5, 53, 118). In reality, the relationship between tobacco taxation and employment is consider- ably more complex than the industry makes it out to be. In fact, there is ample evidence to show that tobacco taxes are a win-win for public health and the fiscal space, without measurable risks to employment. 4.5.2 THE LINK BETWEEN TOBACCO EMPLOYMENT AND TOBACCO TAX RATES Tobacco farming, production and manufacturing (including hand-rolling in some countries, most of them in South-East Asia) constitute a small proportion of the labour force, even in countries where the industry is most heavily concentrated (5, 48, 103, 116). Employment in tobacco farming and manufacturing has been declining globally due to advances in technology, trade liberalization, market consolidation and the privatization of formerly state-owned tobacco companies (5, 103, 119). These same trends have led to the heavy concentration of tobacco growing and manufacturing in only a handful of countries – and within these countries, often in only a small number of regions (103, 119–121). Even in those countries that lead in tobacco growing and manufacturing, tobacco’s overall share of total agricultural and manufacturing employment is relatively small and is often decreasing as efficiencies in production reduce labour intensity (5, 117–118, 122). Similarly, the industry’s claim that tobacco taxes reduce employment is exag- gerated and typically overlooks wider trends driving tobacco industry employment. Indeed, tobacco industry developments and innovations have played a greater role in the reduction of employment in the tobacco industry than have tobacco control policies (103). Despite industry claims that tobacco taxes can affect employment, CHAP T ER 4. PO LI T I C AL ECO N OMY 229 characteristics of the location of production – such as market size, labour costs, growing conditions and leaf preferences – have much more to do with tobacco industry interests than with the tobacco tax rate (103, 123). Moreover, jobs in countries that produce tobacco primarily for export are not greatly affected by reductions in local consumption resulting from tax increases (5, 103, 116, 123). Finally, it has been demonstrated that tobacco tax increases do not have a significant effect on employment in the retail sector, as most retail businesses sell other goods (103). Estimates of the gross employment impact of tobacco tax hikes demonstrate that job losses that do occur can be more than compensated for by increases in revenue. A 2018 World Bank study estimated that in Indonesia, for example, an ambitious tax reform that would simplify tiers and increase prices by close to 50% would reduce gross employment in the tobacco manufacturing sector by less than 0.5% (a loss of 2 914 jobs). The government could provide income support to the displaced workers (for example, through training, temporary transport/mobility or income support) with less than 2% of the revenue gained from the tax increase (117). Similarly, a 2019 study by Bangladesh’s National Board of Revenue estimated that a substantial increase in tobacco taxation would cause 7 012 lost jobs, but that the total income associated with these job losses in the bidi industry would amount to only 3.5% of the revenue gained (120). Accordingly, increased revenue can more than compensate for the expenditure of supporting those who lose jobs and need to acquire new skills before transitioning to new employment (120). Box 4.5.1 Employment fears deployed to frustrate tobacco tax reform in Indonesia In 2017, the Indonesian Ministry of Finance decided to implement tobacco tax in- creases by 2019 and tier simplification by 2021 (124). This resolution was, however, abandoned within a year, after a concerted campaign by tobacco industry actors and their allies to reframe the increase as an economic issue with a focus on, among other things, the effect the tobacco tax increase would have on employment (124). This defeat for the tobacco tax initiative came despite estimations of how the loss of income associated with lost jobs would be dwarfed by the additional revenue gained by the tax (117). Earlier analysis had estimated an overall large net positive impact on employment from tobacco tax increases (125), which illustrates how evidence that challenges assumptions around the negative socioeconomic impacts of tobacco control tends to be discounted (126). In this case, tobacco industry arguments seem to have resonated strongly with politicians from the electoral districts of West Java, East Java, Central Java and West Nusa Tenggara, where employment in tobacco farming and manufacturing is concentrated (124). Although tobacco manufacturing 230 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N represented only 5.13% of total manufacturing employment, the concentration of the job and economic activity meant that arguments regarding employment were particularly salient (117,119). Accordingly, tobacco industry arguments that tobacco taxation would negatively impact employment and farmer livelihoods prevailed, despite strong opposing evidence (119). Concentration of tobacco industry activity within countries poses particular obstacles to overcoming industry arguments on employment and needs to be given careful attention. 4.5.3 THE EVIDENCE ON THE NET EFFECT OF TOBACCO TAX INCREASES ON EMPLOYMENT A proper analysis of the effect of tobacco tax increases on employment must examine their impact on net or economywide employment. Decreases in expenditures on tobacco associated with tobacco control do not mean that expenditures simply disappear; rather, they are redistributed towards consumption of other goods and services, thereby generating employment elsewhere in the economy (5, 103, 123). Similarly, though the effect of higher tobacco taxes on net consumption is arguably more ambiguous, revenues from this intervention do generate spending, invest- ment and employment in public services such as health and education (5). Tobacco control polices usually have a marginal neutral or positive effect on net employment, particularly in countries that are net importers of raw or manufactured tobacco products, as expenditures on these imported items tend to flow out of the country (5). Export-oriented tobacco producers are less sensitive to local demand and are not significantly affected by domestic tobacco tax measures, which likely have a near-neutral net impact (5). In some cases, the net employment impact is a very small negative number, typically less than 1% (127–128). A recent study estimated that in the United Republic of Tanzania – a large tobacco-producing and exporting country – a 30% reduction in smoking prevalence would result in a net employment decline of just 0.5% across the economy as a whole (129). A similar study of Pakistan found that, with some variance depending on where spending was redistributed from tobacco consumption, the overall net effect on employment from a significant reduction in expenditure on cigarette employment – 1 billion rupees – would be a gain of between 6 651 and 5 803 jobs (122). This increase would occur because expenditure on cigarettes produces much less employment in the broader economy than expenditure on food and education (122). In the United Republic of Tanzania, as elsewhere, increased revenue could be used to assist those who lose employment with transitioning to new livelihoods. CHAP T ER 4. PO LI T I C AL ECO N OMY 231 4.5.4 THE VIABILITY OF BETTER LIVELIHOODS In arguing against tobacco tax increases, the tobacco industry advances the myth that people employed in tobacco production – particularly tobacco farming, but also manufacturing – lack any other prospect for a comparably attractive livelihood. However, studies based on extensive survey data in Indonesia, Kenya, Malawi, the Philippines and Zambia have shown that despite needing to commit significant amounts of labour to their crop, tobacco farmers often suffer losses rather than gain profits (119, 121, 123, 130–132). Furthermore, the Indonesian studies demonstrate that tobacco farming has a negative impact on household income and opportunity compared with the experience of other farming households that have given it up (119). Declines in consumption as a result of tobacco tax increases are gradual and susceptible to the same progressive adaptation that has occurred for decades (5, 103). While there will be a need in some countries for the government to help farmers transition to other crops or industries in the longer term, this process will not be a major short-term shock to employment or the wider economy (123). Because tobacco growing and manufacturing can be concentrated in just a few locations within a country, job losses within the tobacco industry might have a disproportionate effect in one location, while employment gains from reduced consumption may be spread across the whole country (120, 123). A study of the employment effects of tobacco tax increases in Bangladesh estimated that up to 60% of all job losses would occur in only two districts – among the poorest in the country – due to the high level of industry concentration (120). Studying the need for support, as well as the means of delivery and funding of support, is particularly necessary in these circumstances. Beyond the need to ensure equity and support employment, a failure to provide for targeted relief can exacerbate fear of job losses and may prove fatal to a tobacco tax proposal (120). Box 4.5.2: Supporting alternative livelihoods in the Philippines The Philippines earmarked 15% of the revenue from a 2012 increase in tobacco taxes to supporting economically viable alternative livelihoods for tobacco farmers and workers (5). Tobacco farming in the Philippines is regionally concentrated, and the tobacco industry had previously been successful in deploying concern for smallholder tobacco farmers to undermine tobacco control measures (136). The provision of economic support was a politically effective countermeasure to tobacco industry SCARE tactics and eased the tax increase’s passage. Given the Philippines’ integration with global tobacco markets and demand, tobacco farmers have not been seriously affected by the tobacco tax increase and 232 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N reduced domestic demand (108, 132). Nevertheless, transfers to tobacco-growing regions from the earmarked tax have been substantial (108). These funds are support- ing gradual transitions to alternative livelihoods, with farmers being encouraged to take up alternative crops, as well as establishing infrastructure, such as market-to-farm roads, that will make these alternative crops more economically viable (108, 133–134). Supporting alternative livelihoods for farmers and other tobacco workers is important because it can offset the political effect of industry arguments, even though domestic tax increases usually have only a modest and gradual effect on employment. There are various models for supporting alternative livelihoods when employment in the tobacco sector gradually diminishes due to decreases in either global or national demand. The Philippines is exemplary, but many other countries have either implemented or experimented with supporting crop transitions. Turkey’s alternative crop pro- gramme, implemented in anticipation of the privatization of the country’s cigarette monopoly, has proven effective in supporting many tobacco farmers’ move to other crops (135). Smaller-scale crop substitution projects in Kenya and Yunnan Province in China have shown how financial, regulatory and infrastructure support from government can contribute to crop transitions (5, 53). Argentina, Bangladesh, Mexico and the state of Maryland in the United States provide additional case studies of how governments can support these transitions (5, 136). 4.5.5 CONCLUSIONS The tobacco industry exaggerates the importance of tobacco employment and over- states the impact that domestic demand reduction due to local taxes will have on tobacco farmers serving a global market. The industry also simplifies employment’s relationship with taxation by focusing only on gross employment in tobacco, which ignores the reality that expenditures on tobacco do not disappear but rather are redistributed for other consumption that can produce a similar or higher number of jobs. Many detailed studies have found that tobacco growing is much less profitable and sustainable than the tobacco industry claims. Tobacco farmers throughout the world have successfully transitioned to other crops, although the transition often requires temporary or additional support from the government or other stakeholders. The extent of such support is moderated by the reality that transition from tobacco to other crops is a long-term consideration. CHAP T ER 4. PO LI T I C AL ECO N OMY 233 4.6 EARMARKING TOBACCO TAX REVENUES TO FUND HEALTH 4.6.1 INTRODUCTION Earmarking tax revenues involves the separation of all or a portion of revenue from a tax or group of taxes to be put aside for a specific purpose (137). Globally, more than 80 countries earmark for health (138), and 37 earmark tobacco tax revenues for health (27). There are two main types of earmarks: hard – also called substantive – and soft, or symbolic (139). Hard earmarks link the expenditure with a revenue source in legislation. This can limit funding if the earmarked revenues are the main source of funding, or it can cause surpluses to accrue wastefully when more revenues are raised than may be expended for the earmarked purpose. Soft earmarks include dedicated funds or commitments to use funds for a particular purpose. They are not necessarily legally binding. For example, in France, the ma- jority of tobacco tax revenue is used to fund social security (which includes health insurance and health care), but there is no hard, formal earmark (140). Earmarks can also be some combination of hard and soft. In the Philippines, tobacco tax earmarks are legally binding, but earmarked revenues go to the general fund, and the Department of Health must submit an annual budget for covered programs as part of its budget request. Earmarking is a broad and contentious topic that goes beyond the specifics of tobacco tax earmarking. Discussions on the topic fall within the ambit of public financial management, and earmarking generally is not encouraged. From a tobacco control perspective, however, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. It is a tool to improve the political economy of tobacco taxation; it is a secondary issue only, after the primary goal of reducing demand for tobacco. One way to use earmarking to improve the political economy of tobacco taxation is to link the payment of tax by tobacco users to benefits they will receive through the funding of complementary tobacco control programmes, such as cessation support, or through increased funding for health programmes on which they will rely disproportionately. This is known as the benefit principle. Earmarking for tobacco control makes sense, as its financial cost is relatively small and tobacco tax reduces demand more effectively when implemented within a package of complementary tobacco control measures. Another way earmarking improves the political economy of tobacco taxation is by safeguarding against any perceived or potential negative ramifications of the tax itself. This is important for neutralizing erroneous but often convincing tobacco industry arguments against effective tobacco tax policies. For example, the Philippines earmarks the bulk of the additional revenues from sin taxes for the health insurance premiums of the poor. In addition, a portion 234 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the country’s tobacco tax revenues is earmarked to provide for the economic well-being of tobacco growers and tobacco growing regions, with the general aim of promoting economically viable alternatives to tobacco farming and manufacturing as a safeguard against the potential for reduced domestic tobacco demand (141). Tobacco tax earmarks are complex, however, and care is required when determin- ing whether a particular earmark is needed on the grounds of political economy and justified on the grounds of equity and economic efficiency. In assessing tobacco tax earmarks, many of the same criteria that have been used to assess the appropriateness of generic earmarks also apply. In the rest of this section, these criteria are set out and matched with reasons for the ability of well-designed tobacco tax earmarks to fulfil them. The types and structures of tobacco tax earmarking are explored alongside descriptions of country experiences to provide guidance on how tobacco tax earmarks are used, when they are justified and the best ways to design them. 4.6.2 CIRCUMSTANCES IN WHICH EARMARKS MAY BE SUITABLE Scepticism about earmarking is both long-standing and justified, but much of the debate concerns earmarking generally and is not specifically concerned with the merits of tobacco tax earmarking (138). The main concerns raised about earmark- ing are listed in Table 4.8, accompanied by suggestions for how earmarks may be structured to address these concerns. Table 4.8 Concerns about earmarking and suggested safeguards to avoid the concerns MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Democratic accountability and oversight: earmarks undermine democratic processes by impeding legislative and executive oversight over expenditure. Establishing proper oversight and accountability procedures is important to ensure funds are not mismanaged (138). Additionally, if a soft earmark structure, which transfers revenue to the general fund from which it is then allocated, is adopted, this will not be a concern. Budget rigidity: earmarking may create budget rigidity that can lead to inefficient allocation of resources (138). An earmark’s particular design determines how much rigidity is introduced (138). Flexible soft earmarks are less prone to introducing rigidity than hard earmarks. Concerns about rigidity can be reduced by the inclusion of a sunset clause that ensures that the earmark is automatically discontinued or reviewed after a set period of time has elapsed (138). A further safeguard is to establish the earmark as a waterfall account, with any excess revenue over a set amount being allocated to the general fund. CHAP T ER 4. PO LI T I C AL ECO N OMY 235 MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Fragmentation: earmarking can result in fragmented and uncoordinated expenditures. This means policies complementary to the earmarked purpose but outside of its purview may be unfunded (138, 142). This is a legitimate concern. The negatives of fragmentation cannot be entirely eliminated, but they may be outweighed by the other merits of tobacco tax earmarking. That said, proposals for tobacco tax earmarks should be scrutinized to ensure that the funded purpose is at least cost-effective. Decreased equity: equity will decrease if individual access to benefits is narrowly defined according to payments made. This issue is not likely to arise with tobacco tax earmarks but is conceivable and something that should be guarded against in an earmark’s design. Capture by special interests: because earmarks are often the result of political expediency, an earmarked purpose may be determined by powerful special interests promoting a tax’s passage rather than careful prioritization of resources (138). Well-designed earmarks will guarantee funding for underresourced programmes and high- priority programmes. While the above concerns may be valid and design does matter, tobacco tax and other health-promoting taxes are not subject to the same concerns when it comes to the justifiability of earmarking their revenue (138, 143). Some of the factors that distinguish tobacco tax earmarks from more general critiques of earmarking are listed in Table 4.9 (138). Table 4.9 Concerns about earmarking and distinguishing factor for tobacco tax earmarks GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Procyclicality: earmarked revenues are often procyclical and susceptible to booms and busts (138–139, 142). Tobacco tax revenues are generally not cyclical (they are recession-proof ), and revenue is predictable relative to most other indirect and direct taxes (103). Budget rigidity Tobacco tax earmarks necessarily involve only a relatively small proportion of the budget; therefore, the effect of any rigidity will be relatively insignificant. Partly because of the relatively small amounts involved, there is only limited real-world evidence of tobacco tax earmarks having introduced harmful rigidity (143).24 24 See also the subsection on the amount of money associated with tobacco tax earmarks in section 4.6.3. 236 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Capture by special interests All earmarks should be scrutinized to ensure that their funded purpose is cost-effective. In the case of tobacco tax earmarks, however, political economy considerations may mean that it is sufficient for a low-priority purpose to be funded if the funding will unlock the political will needed for effective tobacco tax increases. In these cases, special interests are being purposefully catered to in order to ensure that tobacco tax increases occur. Of course, arguments against tobacco taxes and tobacco tax earmarking are led by special interests such as the tobacco industry (144–147). Insufficient revenue: the earmarked revenue source may become insufficient for funding its purpose (139, 142). Even though revenue may decrease in the long term when more tobacco users quit, such decrease is expected to be gradual. See Section 4.4 for details on how, with effective design, revenue will generally increase even with declining consumption. In addition to these reasons why general concerns about earmarking do not fully apply to well-designed tobacco tax earmarks, there are a number of compelling reasons for tobacco tax earmarking to finance tobacco control or public health that argue in favour of its implementation: • Significant increases in excise taxes are the most effective, as well as the most cost-effective mechanism for reducing consumption, but they are best implemented as a part of a package of complementary tobacco control measures, such as the WHO MPOWER package. Earmarking tobacco tax revenue for interventions that may not be funded otherwise can strengthen overall tobacco demand reduction (148). • The political economy of tobacco tax increases also makes earmarks attractive: – People have been shown to be more supportive of tobacco tax increases when they know the revenues will be used for targeted social programmes (143, 149–150). Earmarking tax revenue for health or tobacco control frames tobacco tax as a public health intervention in the minds of the public, which may otherwise view it as merely a revenue source (138). Research has shown that using earmarking to link a tobacco tax to health can also help raise awareness about the dangers of tobacco use (143). – When tobacco tax revenue is earmarked for programmes that benefit vulnerable groups, the tax becomes more equity-enhancing. Although lower socioeconomic groups and young adults receive disproportionate health and economic benefits from tobacco tax increases over the medium term, these groups will expend a greater share of their income in the CHAP T ER 4. PO LI T I C AL ECO N OMY 237 short term because of tobacco taxes. Earmarking tobacco tax revenue for programmes such as UHC or cessation services that provide immediate benefits to these groups neutralizes some critiques of tobacco taxation (e.g. the 2009 United States federal excise tax increase and the 2012 Philippines Sin Tax Reform illustrate how equity-enhancing earmarking facilitated passage of substantial tax rises) (93, 138, 143, 150). 4.6.3 EARMARKING PRACTICES AND COUNTRY EXAMPLES Earmarking tax revenues for health is a common practice in 80 countries. In 2018, 37 countries from all regions of the world earmarked tobacco tax revenues for health purposes.25 Case studies in the political economy of tobacco tax earmarking In 2012, the Philippines comprehensively reformed tobacco and alcohol excise taxes. Tobacco taxes were increased significantly, and numerous tax tiers were reduced to only one tier by 2017. Although increasing revenue was a foremost motive for some officials, the reform was explicitly framed around boosting UHC funding and advancing public health by reducing alcohol and tobacco consumption. Earmark- ing of tax revenue for UHC was essential to the political compromise that made this trailblazing tax increase a reality. It ensured that the increase, which may have otherwise been perceived as regressive, was framed as a progressive public health measure in the public imagination, while also appeasing tobacco growers and their political representatives. Earmarking was also important because the earmark en- sured high-level support for the tax by achieving a key political priority (151). Its soft-earmark structure meant it was not a blank cheque to the Ministry of Health, and this addressed concerns within the Ministry of Finance. Similarly, in Australia, earmarking of revenue helped overcome community objections to tobacco taxes and tobacco control more generally that resulted from the tobacco industry’s sponsorship of sports and the arts in the 1980s. Attempts to completely ban tobacco advertising and sponsorship had been unsuccessful due to strong pressure from sports, arts and racing lobbies that claimed that a ban would harm these activities. States, starting with Victoria, responded by earmarking funding for Health Promotion Foundations that took over the tobacco industry’s sponsorship activities and also paid for antismoking campaigns. In 1997, these earmarks ended after a High Court ruling that the Constitution did not allow states to collect excise taxes. However, in recognition of the successful work of the Foundations, the federal 25 Details about earmarked taxes by country are provided at https://www.who.int/tobacco/global_report/ Table-9-4-Use-of-earmarked-tobacco-taxes.xls?ua=1. 238 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government began funding them directly from the federal budget (152). Although this example may be difficult to replicate precisely, it shows how earmarks with built-in sunset clauses for piloting cost-effective interventions can potentially graduate to funding from the general budget once they have proven their effectiveness. It also more generally shows how earmarks can disarm community objections, reframe tobacco tax increases and unlock the political will needed to advance effective tobacco control measures. Structures for managing earmarked tobacco taxes A 2016 review of nine countries’ tobacco tax revenue earmarking experiences identified three arrangements for governance and allocating revenue (151). Table 4.10 presents some examples of these allocation arrangements. In some countries, earmarked tobacco tax revenues are combined with alcohol tax revenues. Table 4.10 Illustrative arrangements for allocating earmarked tobacco tax revenues POSSIBILITIES FOR ALLOCATING TOBACCO TAX EARMARK REVENUE Forms of budget allocation Revenue goes to the general fund and is later assigned to the official actor(s) specified in the earmark. Revenues do not go through the general budget but are instead paid into a separate account belonging to the official actor(s) specified in the earmark. Earmarked tax revenue is paid directly to the account of the entity managing an autonomous or semi- autonomous fund. Examples In the Philippines, revenue goes to the general fund before being allocated to the Ministry of Health following submission of a budget for its use (140). In Romania, revenue goes directly into a Ministry of Health account that is distinct from the general fund. In Panama, revenue is paid into subaccounts of the three recipient agencies (the Ministry of Health, the National Cancer Institute and the Customs Authority). In Thailand, ThaiHealth directly receives the earmarked revenues in its own account. In Viet Nam, the Viet Nam Tobacco Control Fund receives the revenues directly into a subaccount it manages but that belongs to the Ministry of Health. Source: (151). Where is the money being spent? Earmarked tobacco tax revenues are used for a variety of health purposes, including tobacco control, health promotion and UHC. A wide variety of other programmes have also been funded with earmarks from tobacco taxes, including disaster relief (e.g. hospital medical supplies and equipment to treat COVID-19 in India), youth pro- grams, sports and craft jobs in Yemen, social cohesion in Morocco, health and social programs in areas dependent on tobacco growing in Argentina, health promotion CHAP T ER 4. PO LI T I C AL ECO N OMY 239 and tobacco control in Thailand and alternative livelihood programs for tobacco farmers as well as economic projects in tobacco-growing provinces in the Philippines. Table 4.11 shows the three main categories of health programmes to which ear- marked tobacco tax revenue is allocated, as well as a fourth miscellaneous category, with country-specific examples for each.26 Table 4.11 Programmes to which earmarked tobacco tax revenue is allocated TOBACCO CONTROL NCD PREVENTION AND CONTROL PROGRAMMES (otherwise indicated between brackets) HEALTH COVERAGE EXPANSION (e.g. through health insurance coverage) OTHER, MORE GENERAL OR UNSPECIFIED HEALTH PROGRAMMES Costa Rica, Côte d’Ivoire, the Islamic Republic of Iran, Madagascar, Panama (tobacco cessation and fighting illicit trade), Switzerland, Viet Nam Cook Islands, Costa Rica, Mauritania (anti-cancer research), Palau (NCD prevention only), Panama (National Institute of Oncology), Paraguay Colombia, Congo, Egypt, Palau, Philippines Algeria, Argentina, Bangladesh, Botswana, Cabo Verde, Chad (programmes delivering antiretroviral drugs), Colombia (sports), Comoros (sports, hospital emergencies), Congo (sports), Côte d’Ivoire (AIDS programme), El Salvador, Estonia (sports), Guatemala, Indonesia, the Islamic Republic of Iran (sports), Ireland, Jamaica, Lithuania (sports), Madagascar (sports), Morocco, Nepal, Paraguay (sports), Republic of Korea (health promotion), Romania, Thailand (health promotion), United States, Yemen (sports) Note: Countries appear in more than one column when their earmarked tax revenues are used in more than one specific health programme. Source: (27). The amount of money associated with tobacco tax earmarks Case studies of the experiences of nine countries in tobacco tax earmarking show that earmarked funds are relatively small in comparison with government spend- ing on health (see Table 4.12) and, consequently, even smaller in terms of GDP. Therefore, the argument that tobacco tax earmarks would introduce rigidity into public financial management may not apply. 26 Details about how the tobacco tax revenues are earmarked are given in Annex 4.2. 240 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.12 Proportion of earmarked tobacco tax funds in government expenditures COUNTRY ESTIMATED ANNUAL TOTAL FUNDS FROM EARMARKED TAX ANNUAL FUNDS FROM TOBACCO TAX EARMARKS as a % of general government expenditure on health in 2013 Botswana 2014–2015: 4 million pula (US$ 0.48 million) NA Egypt 2013–2014: 392 million Egyptian pounds (US$ 52.06 million); earmarked taxes only 1.8% of total taxes on cigarettes 1.086% Iceland 2014: 108.3 million kronor (US$ 0.89 million) 0.083% Panama 2014: US$ 27.8 million 1.322% Philippines 2014: 50.18 billion Philippine pesos (US$ 1.18 billion) NA Poland 2013: 1 million złoty (US$ 0.316 million) from general budget  0.001% Romania 2014: 1.1 million lei (US$ 0.33 million); 14.4% of total health budget 0.004% Thailand 2014: 4064.74 million baht (US$ 125.15 million); 1.78% of Ministry of Health budget and 1.84% of National Health Security Fund 0.932% Viet Nam 2014: 299.171 billion dong (US$ 13.91 million); 0.5% of national health budget 0.335% Source: (151). 4.6.4 CONCLUSIONS Despite the initial principled resistance to earmarking by some ministries of finance, experience has shown that the use of revenue from tobacco taxes and other taxes on the consumption of products that have negative externalities can ensure political as well as public support. Successful earmarking needs a well-developed structure for the use of funds for health purposes. Even intergovernmental organizations that are opposed to earmarking (e.g. the IMF) have acknowledged the justifiability of well-designed tobacco tax earmarks when revenue is directed to specific cost- effective programmes (153–154). The amounts of tobacco tax revenue effectively earmarked for health have been relatively small and could hardly introduce the feared rigidity in government budgets. Moreover, in some countries, those funds have helped to implement much-needed health programmes (e.g. Australia, the Philippines, Thailand). More governments are considering this option as a stable medium-term source of secure funding for programmes such as tobacco control. The payoffs will be seen in the future as fewer people fall ill and less medical care for tobacco-related illnesses is needed. In Australia, CHAP T ER 4. PO LI T I C AL ECO N OMY 241 an earmarked tax was used to fund a needed and underresourced programme that proved to be successful, effective and impactful; the programme is now sustainably funded, embedded in the federal budget. Earmarking is desirable in a particular political economy when it enables the implementation of effective tobacco taxation that will increase price and reduce consumption. It will, however, also be rational as a matter of public financial man- agement, economic efficiency and democratic governance when concerns such as the following are considered. Although not every question needs an affirmative answer, policy-makers who can answer yes to many of the following questions will likely be considering an effective and rational tobacco tax earmark:27 • Does the tobacco tax earmark’s purpose rationally connect with the recipient programme’s purpose? Earmarks that fund tobacco control or other health programmes are more economically rational under the benefit principle than those that fund unrelated programmes such as childhood education, even when the popularity of the unrelated programmes may make a tax increase politically palatable. • Does the tobacco tax earmark’s amount rationally connect with the needs of the recipient programme? Earmarked funds that cannot be absorbed by the recipient programme are, in effect, money taken away from other needs. • When a tobacco tax earmark funds health programmes, is this clearly com- municated to the public to ensure that the framing of the tobacco tax increase as a health measure reinforces the demand-reduction effect? • Is the programme being funded by the tobacco tax earmark a politically neglected but highly cost-effective or crucially needed programme that, once established as a proof of concept, has a chance of being funded out of the general budget? • Does the tobacco tax earmark’s purpose rationally connect with the effects of the tax itself? Earmarks that fund programmes that disproportionately benefit lower socioeconomic groups or that fund alternative livelihoods for former tobacco workers and farmers will have equity-enhancing effects that will reinforce the already progressive nature of tobacco taxes. • Is the scope of the earmark’s purpose narrow enough that it can be funded mostly from the tobacco tax earmark, to ensure that the revenue is additive and does not merely substitute for spending that would otherwise come from the general fund? • Does the design of the tobacco tax earmark provide for flexibilities that ensure that windfall revenue collection is not squandered on a purpose already saturated with overfunding? 27 Adapted from and informed by References 138–139, 143, 150. 242 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Does the design of the tobacco tax earmark include a sunset clause that triggers its automatic end or review? The presence of such a clause will guard against inefficiency in allocation arising and being left uncorrected due to political inertia. CHAP T ER 4. PO LI T I C AL ECO N OMY 243 REFERENCES 1. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tob Control. 2013; 22(2): 144–153 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 31 January 2021). 2. LaFaive MD, Drenkard S, Nesbit T. Strong link between cigarette tax and illegal smuggling rates [website]. Midland: Mackinak Center for Public Policy; 2018 (https://www.mackinac.org/strong- link-between-cigarette-tax-and-illegal-smuggling-rates#:~:text=An%20increase%20of%2011%20 percentage,packs%20to%2030.6%20million%20packs.&text=Tax%20evasion%20and%20avoidance%20 is,high%20excise%20taxes%20on%20cigarettes., accessed 31 January 2021). 3. Ross H, Blecher E. Illicit trade in tobacco products need not hinder tobacco tax policy reforms and increases. Chicago: University of Illinois at Chicago; 2019 (Tobacconomics White Paper; https:// tobacconomics.org/research/illicit-trade-in-tobacco-products-need-not-hinder-tobacco-tax- policy-reforms-and-increases-white-paper/#:~:text=additional%20related%20research-,Illicit%20 Trade%20in%20Tobacco%20Products%20Need%20Not%20Hinder%20Tobacco,Reforms%20 and%20Increases%20( White%20Paper)&text=The%20paper%20cites%20research%20 that,consumption%2C%20and%20improves%20overall%20health, accessed 31 January 2021). 4. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 12 October 2020). 5. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 6. Joossens L, Chaloupka FJ., Merriman D, Yürekli A. Issues in smuggling of tobacco products. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac00Box 370064B00PUBLIC0.pdf, accessed 18 February 2021). 7. Merriman D, Yürekli A., Chaloupka, FJ. How big is the worldwide cigarette smuggling problem. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac 00Box370064B00PUBLIC0.pdf, accessed 18 February 2021). 8. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 9. Merriman D. Understand, measure, and combat tobacco smuggling. Washington (DC): World Bank Group; 2003 (Economics of tobacco toolkit 7; https://openknowledge.worldbank.org/bitstream/ handle/10986/16267/805800NWP070Sm0Box0379807B00PUBLIC0.pdf?sequence=1&isAllowed=y, accessed 31 January 2021). 10. Progress report on the implementation of the Commission Communication. Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products – a comprehensive EU strategy. Brussels: European Commission; 2017 (COM(2017) 235 final; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:52017DC0235&from=en, accessed 31 January 2021). 11. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 12. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Reuter P, Majmundar M. (editors). Washington (DC): National Academies Press; 2015. 13. von Lampe K. The illegal cigarette trade. In: Natarajan M, editor. International crime and justice. New York: Cambridge University Press. 2010:148–54. 14. Joossens L, Raw M, From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–34 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 31 January 2021). 15. Gomis B, Lee K, Carrillo Botero N, Shepherd P, Iglesias RM. “We think globally”: the rise of Paraguay’s 24 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tabacalera del Este as a threat to global tobacco control. Global Health. 2018;14(1):110 (https:// globalizationandhealth.biomedcentral.com/articles/10.1186/s12992-018-0412-3, accessed 31 January 2021). 16. Communication from the commission to the Council and the European Parliament: Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products - A comprehensive EU Strategy. Brussels: European Commission; 2013 (https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52013DC0324&from=EN, accessed 31 January 2021). 17. Ross H, Vellios N, Smith KC, Ferguson J, Cohen JE. A closer look at ‘cheap white’ cigarettes. Tob Control. 2016;25(5):527–31 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/527.full.pdf, accessed 31 January 2021). 18. Nagelhout GE, van den Putte B, Allwright S, Mons U, McNeill A, Guignard R, et al. Socioeconomic and country variations in cross-border cigarette purchasing as tobacco tax avoidance strategy. Findings from the ITC Europe Surveys. Tob Control. 2014;23:i30–i38 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/23/suppl_1/i30.full.pdf, accessed 1 February 2021). 19. Gilmore AB, Rowell A, Gallus S, Lugo A, Joossens L, Sims M. Towards a greater understanding of the illicit tobacco trade in Europe: a review of the PMI funded ‘Project Star’ report. Tob Control. 2014;23:e51–e61 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/e1/e51.full.pdf, accessed 1 February 2021). 20. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Revista Panam Salud Publica. 2016;40(4):243–9 (https://www.scielosp.org/pdf/ rpsp/2016.v40n4/243-249, accessed 1 February 2021). 21. Operação Huno combate mercado clandestino de cigarros em sete Estados [Operação Huno fights clandestine cigar market in seven states]. Brasilia: Federal Police of Brazil; 2015 (in Portuguese) (http:// www.pf.gov.br/agencia/noticias/2015/11/operacao-huno-combate-mercado-clandestino-de-cigarros- em-seis-estados, accessed 12 October 2020). 22. Gilmore AB, Gallagher AWA, Rowell A. Tobacco industry’s elaborate attempts to control a global track and trace system and fundamentally undermine the Illicit Trade Protocol. Tob Control. 2019;28:127–40 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/28/2/127.full.pdf, accessed 1 February 2021). 23. Technical assessment of the experience made with the Anti-Contraband and Anti-Counterfeit Agreement and general release of 9 July 2014 among Philip Morris International and affiliates, the Union and its Member States. Brussels: European Commission; 2016 (SWD (2016) 44 final; https://ec.europa.eu/anti- fraud/sites/antifraud/files/technical_assessment_pmi_24022016_en.pdf, accessed 1 February 2021). 24. Gallagher AWA, Evans-Reeves KA, Hatchard JL, Gilmore AB. Tobacco industry data on illicit tobacco trade: a systematic review of existing assessments. Tob Control. 2019;28(3):334–45 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/28/3/334.full.pdf, accessed 1 February 2021). 25. Joossens L, Merriman D, Ross H, Raw M. How eliminating the global illicit cigarette trade would increase tax revenue and save lives. Paris: International Union Against Tuberculosis and Lung Disease; 2009 (https://theunion.org/sites/default/files/2020-08/How%20Eliminating%20the%20Global%20Illicit%20 Cigarette%20Trade%20Would%20Increase%20Tax%20Revenue%20and%20Save%20Lives%20-%20 EN.pdf, accessed 1 February 2021). 26. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 27. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 28. Prieger JE, Kulic J. Cigarette taxes and illicit trade in Europe. Econ Inquiry. 2018;56(3):1706–23 (https:// papers.ssrn.com/sol3/papers.cfm?abstract_id=2718519, accessed 18 February 2021). 29. Stoklosa M. Prices and cross-border cigarette purchases in the EU: evidence from demand modelling. Tob Control. 2020;29:55–60. 30. Passport: illicit trade in tobacco products. In: Euromonitor International [website]. London: Euromonitor International; 2018 (restricted access). 31. The Global Illicit Trade Environment Index. London: The Economist Intelligence Unit; 2018 (http:// illicittradeindex.eiu.com/documents/EIU%20Global%20Illicit%20Trade%20Environment%20Index%20 CHAP T ER 4. PO LI T I C AL ECO N OMY 245 2018%20-%20Overall%20Results%20White%20Paper%20June%206%20FINAL.pdf, accessed 1 February 2021). 32. Chaloupka FJ, Yürekli A, Fong GT. Tobacco taxes as a tobacco control strategy. Tob Control. 2012;21:172– 80 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/172.full.pdf, accessed 1 February 2021). 33. Chernick H, Merriman D. Using littered pack data to estimate cigarette tax avoidance in NYC. Natl Tax J. 2013;66(2):635–68 (https://www.ntanet.org/NTJ/66/3/ntj-v66n03p635-668-littered-pack-data- cigarette-tax.pdf, accessed 1 February 2021). 34. Liber AC, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia Survey. Tob Control. 2015;24:iii83–iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed 1 February 2021). 35. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 36. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 37. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 38. Paraje G, Araya D, Drope J. Illicit cigarette trade in metropolitan Santiago de Chile. Tob Control 2020;29:68–73. 39. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 40. Saenz de Miera Juarez BS, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020(0):1-7 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 1 February 2021). 41. Lee K, Collin J. ‘Key to the future’: British American Tobacco and cigarette smuggling in China. PLoS Medicine. 2006;3(7):e228 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1502159/pdf/pmed.0030228. pdf, accessed 2 February 2021). 42. Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38– i43 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i38.full.pdf, accessed 2 February 2021). 43. Aziani A, Kulick J, Norman N, Prieger JE. Empty discarded pack data and the prevalence of illicit trade in cigarettes; 2017. (https://ssrn.com/abstract=2906015, accessed 12 October 2020). 44. Philip Morris International. The Perspective of PM International on Smoking and Health Issues. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1985. Bates no. 2023268328-2023268349 (https://www.industrydocuments.ucsf.edu/tobacco/docs/#id=pxpb0040, accessed 5 February 2021). 45. Mandal S, Gilmore A, Collin J, Weishaar H, Smith K, McKee M. Block, amend, delay: tobacco industry efforts to influence the European Union’s Tobacco Products Directive (2001/37/EC). Smokefree Partnership/Cancer Research UK; 2009 (http://www.smokefreepartnership.eu/images/Reports-and- Position-Papers/EU_TI_TPD_report_May_2012.pdf, accessed 15 October 2020). 46. Bishop DJ. Key area paper: indirect taxation – British American Tobacco. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1994. Bates no. 503911771-503911780 (http:// bat.library.ucsf.edu/data/h/f/k/hfk11a99/hfk11a99.pdf, accessed 15 October 2020). 47. Zhou S, Liberman JD, Ricafort E. The impact of the WHO Framework Convention on Tobacco Control in defending legal challenges to tobacco control measures. Tob Control. 2009:28:s113-s118 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/28/Suppl_2/s113.full.pdf, accessed 2 February 2021). 48. Gilmore AB, Fooks G, Drope J, Bialous SA, Jackson RR. Exposing and addressing tobacco industry 246 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N conduct in low-income and middle-income countries. Lancet. 2015;385:1029–43. 49. Ulucanlar S, Fooks G, Gilmore A. The policy dystopia model: an interpretive analysis of tobacco industry political activity. PLoS Med. 2016:13(9):e1002125 (https://journals.plos.org/plosmedicine/ article?id=10.1371/journal.pmed.1002125, accessed 2 February 2021). 50. Apollonio D, Glantz S. Tobacco manufacturer lobbying to undercut minimum price laws: an analysis of internal industry documents. Tob Control. 2020;29:e10-e17 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC7374022/pdf/nihms-1590806.pdf, accessed 2 February 2021). 51. Crosbie E, Glantz S. Tobacco industry argues domestic trademark laws and international treaties preclude cigarette health warning labels, despite consistent legal advice that the argument is invalid. Tob Control. 2012;23:e7 (https://www.researchgate.net/publication/233769722_Tobacco_industry_ argues_domestic_trademark_laws_and_international_treaties_preclude_cigarette_health_warning_ labels_despite_consistent_legal_advice_that_the_argument_is_invalid, accessed 2 February 2021). 52. Puig S. Tobacco litigation in international courts. Harv Int Law J. 2016;57(2):1–68. 53. Crooked nine: nine ways the tobacco industry undermines health policy. New York: Vital Strategies; 2019 (https://exposetobacco.org/wp-content/uploads/2019/09/Crooked-9-STOP.pdf, accessed 2 February 2021). 54. Balwicki L, Stokłosa M, Balwicka-Szczyrba M, Tomczak W. Tobacco industry interference with tobacco control policies in Poland: legal aspects and industry practices. Tob Control. 2016;25(5):521–26 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/521.full.pdf, accessed 2 February 2021). 55. Sud SR, Brenner JE, Shaffer ER. Trading away health: the influence of trade policy on youth tobacco control. J Pediatrics. 2015;166(5):1303–07 (https://www.jpeds.com/action/showPdf?pii =S0022-3476%2815%2900017-7, accessed 2 February 2021). 56. Lee S, Ling PM, Glantz SA. The vector of the tobacco epidemic: tobacco industry practices in low and middle-income countries. Cancer Cause Control. 2012;23:117–29 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC3332051/pdf/nihms-367996.pdf, accessed 2 February 2021). 57. Zhou S, Liberman J. The global tobacco epidemic and the WHO Framework Convention on Tobacco Control—the contributions of the WHO’s first convention to global health law and governance. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018 (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3339718, accessed 18 February 2021). 58. Legislative Consultation with Constitutional Division of the Supreme Court, No. 2012-003918. Constitutional Division of the Supreme Court of Costa Rica; 20 March 2012 (https://www. tobaccocontrollaws.org/files/live/litigation/830/CR_Legislative%20Consultation%20with%20.pdf, accessed 2 February 2021). 59. British American Tobacco Kenya Ltd v Ministry of Health, Petition No. 143 of 2015. High Court of Kenya; 2016 (https://www.tobaccocontrollaws.org/files/live/litigation/2643/KE_British%20American%20 Tobacco%20Kenya.pdf, accessed 2 February 2021). 60. British American Tobacco v Ministry of Health, et al, Civil Appeal No. 112 of 2016. Kenyan Court of Appeal; 2017 (https://www.tobaccocontrollaws.org/files/live/litigation/2521/KE_British%20 American%20Tobacco%20Ltd%20v.pdf, accessed 2 February 2021). 61. Thailand – customs and fiscal measures on cigarettes from the Philippines: Report of the Panel. Geneva: World Trade Organization; 2010 (WT/DS371/R; https://docs.wto.org/dol2fe/Pages/SS/directdoc. aspx?filename=Q:/WT/DS/371R.pdf&Open=True, accessed 2 February 2021). 62. Tribunal Constitucional de Chile [Constitutional Court of Chile] (1995 Rol No. 219-95; https://www. tribunalconstitucional.cl/descargar_sentencia2.php?id=3074, accessed 4 April 2020.) 63. McGrady B. International investment law. In: Voon T, Mitchell AD, Liberman J, editors. Regulating tobacco, alcohol and unhealthy foods: the legal issues. New York: Routledge; 2014. 64. Marvin Roy Feldman Karpa v United Mexican States, Case No. ARB(AF)/99/1 International Centre for Settlement of Investment Disputes; 2002 (https://www.italaw.com/sites/default/files/case-documents/ ita0319.pdf, accessed 2 February 2021). 65. Feldman Karpa v Mexico (2002 ICSID Arbitral Tribunal) (https://www.italaw.com/sites/default/files/ case-documents/ita0319.pdf, accessed 18 February 2021). 66. Undang-Undang Nomor 11 Tahun 1995 tentangCukai (Lembaran Negara Republik Indonesia Tahun1995 CHAP T ER 4. PO LI T I C AL ECO N OMY 247 Nomor 76) (Excise Law,in Indonesian). Article 5. 67. Excise Regulation (2011) Reference No. PMK Number 167/PMK.011/2011 (in Indonesian) (https:// peraturan.bpk.go.id/Home/Details/145897/pmk-no-167pmk0112011, accessed 3 February 2021). 68. Mahkamah Agung. Indonesia Supreme Court, Case No. 73/PDT/2018/PT SBY (https://putusan3. mahkamahagung.go.id/direktori/putusan/74efff9cfe18d60990fcde19583ecd5f.html, accessed 6 February 2021). 69. Maya D. Analisis kebijakan kenaikam tarif cukai dan penyesuaian batasan harga jual eceran hasil tembakau [Policy analysis on excise tax increases and price limit adjustments retail tobacco products]; 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-09/S45937-Deacy%20Maya accessed 3 March 2020). 70. Indriawati EN, Susanti L, Qurnia MA. Kebijakan tarif cukai hasil tembakau terhadap perekonomian [Tobacco excise tariff policy on the economy]; 2014 (in Indonesian). (https://www.academia. edu/6611977/MAKALAH_CHT_kel_6, accessed 15 October 2020). 71. Arroyani DW. Evaluasi Kebijakan Penetapan Tarif Cukai Hasil Tembakau Berdasarkan PMK [Evaluation of Tobacco Excise Tariff Determination Policy Based on PMK].167/PMK.011/2011. 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-08/S44664-Dwira%20Wanti%20Arroyan, accessed 15 October 2020). 72. Ngo Ngo Ha and Another v. State of New South Wales and Others (Matter No S 45 of 1996); Walter Hammond & Associates Pty Ltd v. State of New South Wales and Others (Matter No S 165 of 1996). Constitutional law; (1997) 146 ALR 355: 5 August 1997 (https://staging.hcourt.gov.au/assets/ publications/judgments/1997/029--NGO_NGO_HA_AND_ANOR_v_STATE_OF_NEW_SOUTH_WALES_&_ ORS_(Matter_No_S_45_of_1996);_WALTER_HAMMOND--(1997)_146_ALR_355.html, accessed 2 February 2021). 73. McGrady B. Trade and public health: the WTO, tobacco, alcohol, and diet. New York: Cambridge University Press; 2011. 74. Giorgetti C. Health and international investment law. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018. 75. British American Tobacco v Attorney General and Center for Human Rights and Development. Constitutional Court of Uganda Petition No.46 of 2016; 2017 (https://media.ulii.org/files/judgments/ ugcc/2019/4/2019-ugcc-4_0.pdf, accessed 18 February 2021). 76. Lim CL, Ho J, Paparinskis M. International investment law and arbitration: commentary, awards and other materials. Cambridge: Cambridge University Press; 2018. 77. Dominican Republic – measures affecting the importation and internal sale of cigarettes : Report of the Panel. Geneva: World Trade Organization; 2004 (WT/DS302/R; https://docs.wto.org/dol2fe/ Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302R.pdf&Open=True, accessed 21 February 2021) 78. Dominican Republic — measures affecting the importation and internal sale of cigarettes: Report of the Appellate Body. Geneva: World Trade Organization; 2005 (WT/DS302/AB/R; https://docs.wto. org/dol2fe/Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302ABR.pdf&Open=True, accessed 21 February 2021). 79. Gilmore A. Tobacco and transition: understanding the impact of transition on tobacco use and control in the former Soviet Union [Thesis]. London: London School of Hygiene and Tropical Medicine, June 2005 (https://escholarship.org/uc/item/3rw2c04c, accessed 2 February 2021). 80. MacKenzie R, Collin J, Sopharo C, Sopheap Y. ‘Almost a role model of what we would like to do everywhere’: British American Tobacco in Cambodia. Tob Control. 2004;13(suppl II): ii112–ii117 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/13/suppl_2/ii112.full.pdf, accessed 2 February 2021). 81. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tobacco Control 2013;22(2):144-53 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 21 February 2021). 82. Price and tax. In: TobaccoTactics [website]. Bath: University of Bath; 2020 (https://tobaccotactics.org/ topics/price-and-tax/, accessed 15 December 2020). 83. Vulovic V, Chaloupka FJ. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation—commentary. Tob Control. 24 June 2020. 248 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 84. Verguet S, Kearns PKA, Rees VW. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation. Tob Control. 23 June 2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/08/11/tobaccocontrol-2019-055315. full.pdf, accessed 2 February 2021).. 85. Palipudi KM, Gupta PC, Sinha DN, Andes LJ, Asma S, McAfee T. Social determinants of health and tobacco use in thirteen low and middle income countries: evidence from Global Adult Tobacco Survey. PLoS ONE. 2012;7(3):e33466 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3306395/pdf/ pone.0033466.pdf, accessed 2 February 2021). 86. Systematic review of the link between tobacco and poverty. Geneva: World Health Organization; 2014 (http://www.who.int/tobacco/publications/economics/9789241507820/en, accessed 21 October 2020). 87. Goodchild M, Sinha P, Gill Munish V, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/bitstream/handle/10665/331836/whoseajph2020v9i1p73. pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 88. Faruque GM, Ahmed M, Huq I, Parven R, Wadood SN, Chowdhury SR, et al. The economic cost of tobacco use in Bangladesh: a health cost approach. Dhaka: Bangladesh Cancer Society; 2020 (https://www. researchgate.net/publication/342445467_The_Economic_Cost_of_Tobacco_Use_in_Bangladesh_A_ Health_Cost_Approach, accessed 2 February 2021). 89. John RM, Sung H, Max WB, Ross H. Counting 15 million more poor in India, thanks to tobacco. Tob Control. 2011;20:349–52 (https://www.researchgate.net/publication/49809290_Counting_15_million_ more_poor_in_India_thanks_to_tobacco, accessed 2 February 2021). 90. Liu Y, Rao K, Hu TW, Sun Q, Mao Z. Cigarette smoking and poverty in China. Soc Sci Med. 2006;Dec;63(11):2784–90. 91. Fuchs A, González Icaza F, Paz D. Distributional effects of tobacco taxation – a comparative analysis. Washington (DC): World Bank Group; 2019 (Policy research working paper 8805; https://openknowledge. worldbank.org/bitstream/handle/10986/31534/WPS8805.pdf?sequence=4&isAllowed=y, accessed 2 February 2021). 92. Smith CE, Hill SE, Amos A. Impact of population tobacco control interventions on socioeconomic inequalities in smoking: a systematic review and appraisal of future research directions. Tob Control. 29 September 2020 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/09/28/ tobaccocontrol-2020-055874.full.pdf, accessed 2 February 2021). 93. Furman J. Six lessons from the U.S. experience with tobacco taxes – World Bank conference “Winning the tax wars: global solutions for developing countries”. Washington (DC): Obama White House Archives; 24 May 2016 (https://obamawhitehouse.archives.gov/sites/default/files/page/files/20160524_cea_ tobacco_tax_speech.pdf, accessed 21 October 2020). 94. Önder Z, Yürekli A. Who pays the most cigarette tax in Turkey. Tob Control. 2016;25(1):39–45. 95. Verguet S, Gauvreau CL, Mishra S, MacLennan M, Murphy SM, Brouwer ED, et al. The consequences of tobacco tax on household health and finances in rich and poor smokers in China: an extended cost-effectiveness analysis. Lancet Glob Health. 2015;3:e206–16 (https://www.thelancet.com/action/ showPdf?pii=S2214-109X%2815%2970095-1, accessed 2 February 2021). 96. Fuchs Tarlovsky A, González Icaza F. The welfare and distributional effects of increasing taxes on tobacco in Vietnam. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/ curated/en/758391561477648302/pdf/The-Welfare-and-Distributional-Effects-of-Increasing-Taxes- on-Tobacco-in-Vietnam.pdf, accessed 2 February 2021). 97. Expanding the global tax base: taxing to promote public goods: tobacco taxes. Washington (DC): World Bank Group; 2016 (http://pubdocs.worldbank.org/en/799981480947591462/Taxing-to-Promote- Public-Goods-tobacco-taxation-chapter-conference-25-nov-2016-final-version-pmarquez-002.pdf, accessed 2 February 2021). 98. Blecher E. Increasing tobacco taxes significantly will increase revenues. Chicago: University of Illinois at Chicago; 2018. (Tobacconomics policy brief; https://tobacconomics.org/files/research/470/Laffer- Curve-Policy-Brief_v2.2.pdf, accessed 2 February 2021). 99. Savedoff W, Lopert R. Sugar, rum, and tobacco: domestic resource mobilization for low-income countries through excise taxes. Washington (DC): Centre for Global Development; 2020 (CGD Policy Paper CHAP T ER 4. PO LI T I C AL ECO N OMY 249 164, https://www.cgdev.org/sites/default/files/PP164-Savedoff-Health-Taxes-DRM.pdf, accessed 2 February 2021). 100. Lavado R, Postolovska I, Sotomayor RE. Beyond the perceived regressivity of higher tobacco taxes, turning short-term losers into long-term winners? In: Tobacco reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 101. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 102. Tonga NCD related taxation policy assessment preliminary findings. Presentation made at the innovations and actions against NCDs conference ADB headquarters, Manila, 17 July 2018. 103. Moreno-Dodson B. Raising tobacco taxes: why, what, and how: where public health and economics converge. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e- library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 104. Fuchs A, Del Carmen G, Mukong AK. Long-run impacts of increasing tobacco taxes: evidence from South Africa. Policy Research Working Paper 8369. Washington [DC]: World Bank; 2018 (https://openknowledge. worldbank.org/bitstream/handle/10986/29497/WPS8369.pdf?sequence=4&isAllowed=y, accessed 18 February 2021) . 105. van Walbeek C. Case studies in illicit tobacco trade: South Africa. Tobacconomics fact sheet. Chicago: University of Chicago, Tobacconomics, Health Policy Center; 2020 (https://tobacconomics.org/files/ research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 106. van Loggerenberg, J. Tobacco Wars. Cape Town: NB Publishers; 2019. 107. Marquez P. Global tobacco control: a development priority. In: Tobacco tax reform at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads. pdf, accessed 2 February 2021). 108. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington [DC]: World Bank; 2017 (https:// tobacconomics.org/files/research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 109. Scollo M, Bayly M. Revenue from tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia. org.au/chapter-13-taxation/13-6-revenue-from-tobacco-taxes-in-australia, accessed 21 October 2020). 110. Scollo M, Bayly M. Tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia.org.au/ chapter-13-taxation/13-2-tobacco-taxes-in-australia, accessed 21 October 2020). 111. Breton E, Richard L, Gagnon F, Jacques M, Bergeron P. Fighting a tobacco-tax rollback: a political analysis of the 1994 cigarette contraband crisis in Canada. Journal of Pub Health Policy. 2006;27:77–99. 112. Fanta E, Garcia M. Preventing and controlling illicit tobacco flows. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB- Report-CrossRoads.pdf, accessed 2 February 2021). 113. Goodchild M, Perucic AM, Zheng R, Blecher E, Paul J. The health impact of raising tobacco taxes in developing countries. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/ taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 114. Ross H. Tracking and tracing tobacco products in Kenya. Prev Med. 2017;105:S15–S18 (https://reader. elsevier.com/reader/sd/pii/S0091743517301457?token=424D9EC1D42A9C69BDB55B90090C3F7853AA 20A4970D0E379C91EE64C73A273A0FA949557F4F120A0B636701581EC055, accessed 2 February 2021). 115. Mamudu H, Hammond R, Glantz S. Tobacco industry attempts to counter the world bank report curbing the epidemic and obstruct the WHO framework convention on tobacco control. Soc Sci & 250 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Med. 2008;67(11):1690–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2662513/pdf/nihms81164. pdf, accessed 2 February 2021). 116. Gilmore A. Understanding the vector in order to plan effective tobacco control policies: an analysis of contemporary tobacco industry materials. Tob Control. 2012;21:119–26 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/21/2/119.full.pdf, accessed 2 February 2021). 117. The economics of tobacco taxation and employment in Indonesia: policy implications technical brief. Washington (DC): World Bank Group; 2018 (https://openknowledge.worldbank.org/bitstream/ handle/10986/29814/126158-WP-P154568-PUBLIC-add-series-WBGPNIndoEmploymentUPDATEFin alWeb.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 118. Sahadewo GA, Drope J, Li Q, Nargis N, Witoelar F. Tobacco or not tobacco: predicting farming households’ income in Indonesia. Tob Control. 2020. 119. The economics of tobacco farming in Indonesia: health, population, and nutrition global practice. Washington (DC): World Bank Group, 2017 (http://documents1.worldbank.org/curated/ en/161981507529328872/pdf/120307-REVISED-PUBLIC-WBGIndoEconomicsTobaccoFarming.pdf, accessed 2 February 2021). 120. The revenue and employment outcome of biri taxation in Bangladesh. Dhaka: National Board of Revenue, Government of the People’s Republic of Bangladesh; 2019 (https://www.who.int/docs/ default-source/searo/bangladesh/pdf-reports/cat-2/biri-study-report-03-12-2019.pdf?sfvrsn=b8fea69_2, accessed 2 February 2021). 121. Magati P, Lencucha R, Li Q, Drope J, Labonte R, Appau A, et al. Costs, contracts and the narrative of prosperity: an economic analysis of smallholder tobacco farming livelihoods in Kenya. Tob Control. 2019;28(3):268–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6512316/pdf/nihms-1012064. pdf, accessed 2 February 2021). 122. Ghaus K, Iqbal MA, Sabir M, Saleem W, Ali A, Aamir N. Macroeconomic impacts of tobacco use in Pakistan. Karachi: The Social Policy and Development Centre; 2018 (https://tobacconomics.org/files/ research/483/Macroeconomic-impact-of-Tobacco-in-Pakistan.pdf, accessed 2 February 2021). 123. Ross H. The effects of tobacco taxes on employment and labor productivity. In: Tobacco tax reform: at the crossroads of health and development. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021) 124. Bigawanto M. Tobacco industry interference undermined tobacco tax policy in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance (SEATCA); 2019 (https://seatca.org/dmdocuments/ Indonesia%20TII%20in%20Tax.pdf, accessed 2 February 2021). 125. Ahsan A, Wyono N. Analysis of the impact of higher cigarettes prices on employment in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance; 2007 (https://seatca.org/dmdocuments/An%20 Analysis%20of%20the%20Impact%20of%20Higher%20Cigarette%20Prices%20on%20Employment%20 in%20Indonesia.pdf, accessed 2 February 2021). 126. Higashi H, Khuong TA, Ngo AD, Hill PS. Evidence and decision making: tobacco control policy and legislation in Vietnam. Int J of Health Plann. 2013;2(1):e72–e94. 127. Buck D, Raw M, Godfrey CA, Sutton M. Tobacco and jobs: the impact of reducing consumption on employment in the UK. York: University of York; 1995 (https://www.researchgate.net/ publication/4814523_Tobacco_and_jobs_the_impact_of_reducing_consumption_on_employment_ in_the_UK, accessed 2 February 2021). 128. Petkova R, Ileva R and Rusenov A. Tobacco control policies and tobacco industry employment in Bulgaria. In: Economic, social and health issues in tobacco control. Report of a WHO international meeting, Kobe, Japan, 3-4 December 2001. Geneva: World Health Organization; 2003. 129. Jha V, Narayanan BG, Wadhwa D, Tesche J. Economic and environmental effects of reduction in smoking prevalence in Tanzania. Tob Control; 2020;29:24–8 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/29/1/24.full.pdf, accessed 2 February 2021). 130. Lencucha R, Makoka D, Drope J, Appau A, Labonté R, Li Q, et al. Costs, revenues and profits: an economic analysis of smallholder tobacco farmer livelihoods in Malawi. Tob Control. 2017;26(6)634–40 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC5957074/pdf/nihms965842.pdf, accessed 2 February 2021). 131. Goma FM, Labonté R, Drope J, Li Q, Zulu R, Kangwa E. The economics of tobacco farming in Zambia: CHAP T ER 4. PO LI T I C AL ECO N OMY 251 tobacco farmers survey report. Lusaka: University of Zambia and American Cancer Society; 2019 (https://www.cancer.org/content/dam/cancer-org/research/economic-and-healthy-policy/zambia- report-7-16-19.pdf, accessed 28 October 2020). 132. Chavez JJ, Drope J, Li Q, Aloria MJ. The economics of tobacco farming in the Philippines. Quezon City: Action for Economic Reforms and Atlanta: American Cancer Society, 2016 (https://www.cancer. org/content/dam/cancer-org/research/economic-and-healthy-policy/the-economics-of-tobacco- farming-in-the-philippines.pdf, accessed 22 October 2020). 133. Augstin CP, Cardenas RP, Cortez JB, Gimongala GC, Ramos AR. The effects of the sin tax reform law of 2012 to tobacco farmers of Amulung, Cagayan. IJARMSS. 2017;6(12):57–75. 134. Mercado A. Tobacco tax: good for the tobacco farmers, too. Business World; 2019 [website] (https:// www.bworldonline.com/tobacco-tax-good-for-the-tobacco-farmers-too/, accessed 25 October 2020). 135. Yürekli A, Önder Z, Elibol M, Erk N, Cabuk A, Fisunoglu M, et al. The economics of tobacco and tobacco taxation in Turkey. Paris: International Union Against Tuberculosis and Lung Disease; 2010 (https:// www.who.int/tobacco/en_tfi_turkey_report_feb2011.pdf, accessed 2 February 2021). 136. Latif E, Warner V. Legislation and smoking prevention. In: Loddenkemper R, Kreuter M, editors. The Tobacco Epidemic: progress in Respiratory Research. 2015;42:136–148. 137. Tandon A, Cashin C. Assessing public expenditure on health from a fiscal space perspective. Washington (DC): World Bank Group; 2010 (http://documents1.worldbank.org/curated/en/333671468330890417/ pdf/560530WP0Box341penditureFiscalSpace.pdf, accessed 2 February 2021). 138. Cashin C, Sparkes S, Bloom D. Earmarking for health: from theory to practice. Geneva: World Health Organization; 2018 (https://apps.who.int/iris/bitstream/handle/10665/255004/9789241512206-eng. pdf?sequence=1, accessed 2 February 2021). 139. Wilkinson M. Paying for public spending: is there a role for earmarked taxes? Fiscal Studies. 1994;15(4):119–35 (https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1475-5890.1994.tb00213.x, accessed 2 February 2021). 140. ITAF: impôts et taxes affectées à la protection sociale. Les impôts et taxes affectés (ITAF) [taxes and duties allocated to social protection]. Vie publique: au coeur du débat public [website]. Paris: Directorate of Legal and Administrative Information; 2018 (in French) (https://www.vie-publique.fr/fiches/22010- impots-et-taxes-affectees-itaf-protection-sociale, accessed 31 October 2020). 141. Seventeenth congress of the Philippines, Metro Manila, RA 11346, 23 July 2018 (https://www. officialgazette.gov.ph/downloads/2019/07jul/20190725-RA-11346-RRD.pdf, accessed 31 October 2020). 142. McCleary W. The earmarking of government revenue: a review of some World Bank experience. World Bank Res Obs. 1991;6(1):81–104 (http://documents1.worldbank.org/curated/en/109871468764672767/ pdf/multi-page.pdf, accessed 2 February 2021). 143. Strategic investment of tobacco tax revenue. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/global/pdfs/en/strategic_investment_tobacco_tax_revenue. pdf, accessed 31 October 2020). 144. Bialous SA, Glantz SA. Arizona’s tobacco control initiative illustrates the need for continuing oversight by tobacco control advocates. Tob Control. 1999;8:141–51 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/8/2/141.full.pdf, accessed 2 February 2021). 145. Goldman LK, Glantz SA. The passage and initial implementation of Oregon’s measure 44. Tob Control. 1999;8:311–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1763939/pdf/v008p00311.pdf, accessed 2 February 2021). 146. Heiser PF, Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health. 1997;87(6):968–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380932/pdf/amjph00505-0082. pdf, accessed 2 February 2021). 147. Landman A, Bialick P. Tobacco industry involvement in Colorado. Chicago: American Lung Cancer Association; 2004 (https://escholarship.org/uc/item/9d880448#main, accessed 18 February 2021). 148. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010. (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 149. Vardavas CI, Filippidis FT, Agaku IT, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the 252 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N importance of earmarking the revenue to health care and tobacco control. Tob Induc Dis. 2012;10:21 (http://www.tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the- revenue-to-health-care-and-tobacco,66110,0,2.html, accessed 31 October 2020). 150. Bird RM. Tobacco and alcohol excise taxes for improving public health and revenue outcomes: marrying sin and virtue?. Washington (DC): World Bank Group; 2015 (Policy Research Working Paper 7500; https://openknowledge.worldbank.org/bitstream/handle/10986/23464/ Tobacco0and0al0ing0sin0and0virtue00.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 151. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 152. The establishment and use of dedicated taxes for health. Manila: World Health Organization Western Pacific Regional Office; 2004 (https://www.who.int/health_financing/documents/dedicated_taxes. pdf?ua=1, accessed 2 February 2021). 153. Teja RS. The case for earmarked taxes: theory and example. Washington (DC): International Monetary Fund; 1988 (IMF Working Paper; https://www.elibrary.imf.org/doc/IMF001/14514- 9781451922677/14514-9781451922677/Other_formats/Source_PDF/14514-9781455212002.pdf, accessed 2 February 2021). 154. Keen M. Taxation and development – again. Washington (DC): International Monetary Fund; 2012(IMF Working Paper WP/12/220O; (https://www.imf.org/external/pubs/ft/wp/2012/wp12220.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 253 ANNEX 4.1 METHODS TO ASSESS THE NATURE AND SIZE OF THE ILLICIT TOBACCO TRADE A4.1 DIRECT MEASUREMENT A4.1.1 SMOKER INTERCEPT AND PACK OBSERVATION SURVEYS Illicit trade can be measured directly by examining the cigarette packs of smokers. The smokers themselves can provide information on purchasing patterns, brand preferences and prices paid. Researchers can select individuals or retailers to survey based on a convenience sample (i.e. a sample that may not be representative) or a probability-based sample (i.e. a sample selected to be statistically representative of an underlying population). Data collected from a pack could reveal whether the pack is compliant or non- compliant with the local tax laws. Information can be obtained from objective markings such as brands, public health warning labels, tax stamps, foreign language labels or duty-free labels. During these stops, researchers can record demographic information (e.g. age and gender of the smoker), smoking-related history (e.g. number of cigarettes smoked per day) and price information. This is helpful in understand- ing the profile of smokers who are able and willing to avoid cigarette taxes. Pack observations can be used in conjunction with population-based household surveys to obtain population-based estimates of the illicit tobacco trade (1,2). For example, as part of a regular national health survey, Kaplan et al. conducted a cross-sectional study of smokers in Turkey, using a face-to-face interviewer-administered survey and pack observation (3). They were able to collect sociodemographic, lifestyle and medical details along with pack observations as part of the study protocol. Advantages and disadvantages of smoker intercept and pack observation surveys A primary advantage of conducting pack observation is that it is direct and objective, and smokers are not subject to any value judgements (2). Paired with survey data, pack observation can appropriately account for respondents who are not residents of the area in which they are surveyed (4). Disadvantages include the difficulty of identifying areas that are representative of the tobacco use population and the difficulty of sampling important subpopulations such as elderly and immobile smok- ers. Also, surveys conducted in the daytime may discount the number of youthful smokers who are in school. Another disadvantage is that a sizeable number of smokers may refuse to show their last-purchased pack (2). Kaplan et al. found that 24% of smokers sampled in Turkey did not show their cigarette pack to the study interviewer (3). This issue may be mitigated by asking users to provide information on the brand purchased, whether any public health warnings were posted and the 254 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N price paid (5). Although the responses are based on recall, they may still yield useful information. For example, Joossens et al. allowed smokers who did not show their packs to provide self-reported information and found no statistical differences in illicit packs between these respondents and those who did show their packs (2). Another obstacle to accurate measurement based on smoker intercepts is the inability to determine the tax payment of smokers who purchase single cigarettes, since these smokers are generally not given packs. However, information can still be captured in self-report surveys by asking smokers to report the brand purchased and price paid. Key study for readers to refer to for additional guidance: Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–447. A4.1.2 PACK RETURN AND SWAP SURVEYS Pack return and pack swap surveys fall within the broader category of pack ob- servation studies that use survey sampling techniques to examine smokers’ pack characteristics and to determine whether they are tax compliant. For these surveys, the unit of analysis is the individual. The main differences between pack swap and pack return surveys is that swap surveys offer the smoker a replacement pack, whereas pack returns are built into mail surveys and allow respondents to mail in their unopened packs. Pack swap and pack return surveys use probability and nonprobability sampling procedures. Probability sampling allows researchers to generalize to the broader population. Governments can use this method to rapidly assess the availability of illicit products in a given geographic area or to measure the share of the illicit market- place. Rapid assessment may be performed in instances where there is an emerging tobacco product (e.g. a new cheap white brand) or suspected counterfeiting of tax stamp features. Rapid assessment using a convenience sampling strategy could place researchers near busy intersections where they could ask smokers for permission to look at their cigarette packs or to take photographs that could be analysed later. A population-based study requires a sample that closely mirrors the tobacco use population. Advantages and disadvantages of pack return and pack swap surveys Pack swap and pack return surveys may help to overcome the stigma associated with traditional smoking surveys. For example, when researchers ask smokers to see their cigarette packs (or when they take photographs), no value judgements are made. These surveys are good rapid-assessment tools that can be used to examine CHAP T ER 4. PO LI T I C AL ECO N OMY 255 the effectiveness of physical features of a pack designed to deter illicit trade (e.g. packs that have tracking and tracing technology or high-tech stamps). In addi- tion, they can be supplemented with population-based tobacco use surveys. When coupled with such survey data, these methods allow researchers to obtain relevant information about the context of illicit purchases, including, for example, sources (e.g. street, peer networks, retail stores) and prices. Mail-in surveys are filled out in the comfort of the respondent’s home without the presence of family members or passers-by, which may assure them that responses will be kept confidential. A potential disadvantage is that smokers who purchase both illicit cigarettes and tax-paid cigarettes may disproportionately mail back compliant packs. In addition, in LMICs, this mode of survey distribution may be unreliable because of issues associated with mail delivery systems. Key study for readers to refer to for additional guidance: Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61-66. A4.1.3 LITTERED-PACK SURVEYS Littered-pack surveys, also known as empty discarded pack surveys, are used pre- dominantly in high- and middle-income countries (e.g. the United States, France, Canada, New Zealand, Mexico and Poland). This unobtrusive method relies on the premise that smokers publicly discard packs (e.g. on streets, sidewalks and in public trash cans). The packs bear characteristics that illustrate whether they are tax compliant (e.g. tax stamps, health warnings). For example, an Albanian health warning label on a cigarette pack discarded in Greece provides evidence that the pack was destined for the Albanian market. The pack may have been smuggled into Greece by criminal entrepreneurs or it may have been brought by a visitor. Collecting discarded packs from a representative geographic sample and examining these characteristics can provide estimates of tax compliance. Operationally, this data collection method uses an ecological approach whereby geographies are the units of analysis. Geographical units are meant to represent the smokers in the city/ country and can be administratively defined (e.g. by the country’s census bureau or transit zones) or may reflect researcher-defined neighbourhoods (e.g. half-mile buffer zones near bus stops or activity spaces). Researchers in Canada have expanded the littered-pack method to include col- lection and analysis of cigarette butts on 25 postsecondary campuses. The cigarette butts provide information on the brands sold (or lack thereof) and allow researchers to distinguish between legal and illegal products (6). A recent innovative expansion 256 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the littered-pack methodology is the collection of packs from cigarette retailers. John and Ross collected empty packs of tobacco products from a sample of registered and unregistered retailers in India (7). Collecting packs from retailers was relevant given that single cigarettes dominate the illicit market in India. Smokers who buy single cigarettes would be unable to provide a pack in a pack swap or street intercept survey, so collecting littered packs from the ground would undercount sales of single cigarettes. The feasibility of this method is dependent on the relationship between researchers and retailers (enhanced trust) and the efforts taken to ensure confidentiality. In some countries, retailer compliance with this research method might be strained because of concerns regarding confidentiality, since retailers may face criminal and civil penalties, depending on the research findings. Advantages and disadvantages of littered-pack surveys Littered-pack surveys are generally advantageous for governments because they facilitate comparison with industry estimates. This is one of the most-preferred methods because it yields estimates that are less likely to be biased due to issues of social desirability, recall error and confidentiality that plague survey research, and they are much less expensive than face-to-face interviews used in smoker intercept or household surveys. However, there are some issues regarding these surveys, espe- cially in high-income countries, including the inability to differentiate between tax avoidance and tax evasion (8). For example, a pack in Berlin that bears a Vietnamese tax stamp may have been smuggled in mass quantity or brought in by a temporary visitor. Researchers have circumvented this issue and broadened the umbrella to measure cigarette tax noncompliance considering the potential biases introduced by tourism. Another disadvantage of littered-pack surveys is that larger budgets are needed to employ field researchers to collect, code and analyse the data. Not all countries employ tax stamps on their cigarette packaging, which may make it difficult to measure tax compliance. These surveys also can underestimate the markets in low-income countries such as India, where the main item of illegal trade is single cigarettes (7). The surveys may also overestimate illicit trade if littering behaviour is correlated with willingness to engage in illicit trade. Finally, littered-pack surveys and butt collections provide information on the proportion of butts and packs that are illegal, not the proportion of smokers that purchase illegal cigarettes (6). Key studies for readers to refer to for additional guidance: Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38–i43. Merriman D. The micro-geography of tax avoidance: evidence from littered cigarette CHAP T ER 4. PO LI T I C AL ECO N OMY 257 packs in Chicago. Am Econ J Econ Policy. 2010;2(2):61–84. Stoklosa M., Paraje G., Blecher E., A Toolkit on Measuring Illicit Trade in Tobacco Products. A Tobacconomics and American Cancer Society Toolkit. Chicago, IL:Tobacconomics, Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/621/uic-illicit-trade-tool-kit-eng-v2.0-2. pdf, accessed 18 February 2021). A4.1.4 SELF-REPORT POPULATION SURVEYS Self-report surveys, when distributed to a representative sample of the population, can provide meaningful data on the prevalence of tax noncompliance. The surveys can be distributed to individuals or households in various ways, including face-to-face, telephone, mail and internet. Questions that specifically address illicit purchases can be added as supplementary questions to existing health or tobacco surveys. Some countries include such questions in their adult and youth tobacco surveys to estimate tax evasion/avoidance. For example, Canada’s annual Youth Smoking Survey asks smokers about the frequency of their purchases of First Nations/Native brand cigarettes (9-10). Davis et al. used data from the New York Adult Tobacco Survey to measure the source of purchase of the last cigarette pack purchased (i.e. Native American Reservations, lower-tax neighbouring states or countries, toll-free telephone numbers, the internet, duty-free shops) and the price paid (11). Twenty- eight nations currently use surveys to measure tax noncompliance as part of the ITC Project (12). Similar analyses can be conducted using questions from the Global Adult Tobacco Use Surveys. For example, Iglesias et al. used the Brazil Global Adult Tobacco Use Surveys to compare self-reported prices with a defined threshold retail price to estimate the proportion of illicit cigarette use among smokers in Brazil (13). Countries are encouraged to use existing global health surveys or to incorporate similar types of questions pertaining to illicit trade in their annual health surveys. Asking respondents about price paid per pack (including taxes), brand name and location where cigarettes were purchased (e.g. duty free shop, unlicensed vendor, internet) can contribute to a better understanding of the illicit tobacco trade. Advantages and disadvantages of self-report surveys Self-report surveys can be repeated over time to measure purchasing trends and progress associated with increases in cigarette taxation. Well-designed surveys can also provide generalizable estimates at the national level. Depending on the size of the sample, a self-report survey can provide comparable data across geographies that can help governments target resources. For example, findings that illicit cigarettes are more common in urban areas could lead to additional education campaigns and targeted enforcement. 258 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Limitations of self-report surveys include the possibility of bias due to the social stigma associated with participating in the illicit trade, which could lead survey respondents to underreport participation. Additionally, surveys may be unable to gauge whether individuals are associated with tax avoidance versus tax evasion. Finally, there is evidence that self-report household surveys may underrepresent smokers. Key studies for readers to refer to for additional guidance: Callaghan RC, Veldhuizen S, Ip D. Contraband cigarette consumption among adolescent daily smokers in Ontario, Canada. Tob Control. 2011;20(2):173–174. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany (NY): New York State Department of Health, Tobacco Control Program; 2006. A4.1.5 COVERT-PURCHASES SURVEYS A number of studies in high-, middle- and low-income countries use covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes in public and semi-private spaces (14-17). This method is also used by the tobacco industry in the United States to identify retailers who sell counterfeit cigarettes (18). Covert-purchases surveys do not provide estimates of the size of the illicit trade (i.e. market volume). Instead, they serve as a surveillance tool to identify where illicit cigarettes are sold and the extent to which they have infiltrated legal businesses. For example, a covert-purchases survey can examine whether illicit cigarettes are sold through legal retailers. It can also be used to measure compliance with emerging tobacco control policies that focus on, for example, product standardization or new regulations on flavours (e.g. plain packaging or bans on flavoured tobacco products). Covert-purchases surveys use trained researchers to visit a selected sample of retailers and directly purchase or inquire about the availability of illicit tobacco products. Retailers are not informed about the goals of the studies. Methods for determining the availability of illicit product vary. For example, in some studies, covert buyers do not directly inquire about illicit products. Instead, they purchase packs of tobacco products, paying full price, to determine whether retailers are selling illicit products under the guise that they are licit (14). The research team then examines the packs to determine whether they are legal. In the United States, researchers have observed that some consumers are paying full price for illicit untaxed packs smuggled from lower-tax states (14). Other research protocols directly ask retailers for illicit products (14,16-17). In Guatemala, Arevalo et al. specifically asked retailers for “imported cigarettes” (17). The ways covert buyers ask for illicit products may also vary geographically. For example, in some countries covert buyers may ask for “cheaper” packs or for illicit whites such as Jin Ling. CHAP T ER 4. PO LI T I C AL ECO N OMY 259 Advantages and disadvantages of covert-purchases surveys Covert purchasing allows researchers to directly identify sources of illicit cigarettes. It also allows them to measure and test the dynamics between buyer and seller. For example, researchers can experiment to see if repeated attempts to purchase products increase the likelihood of purchase (known as the familiarity protocol) (16). One methodological challenge associated with covert purchases is that it is difficult to create a sampling frame for illicit sources because some may be unknown (e.g. pubs or homes). The traditional approach is to make purchases in legal outlets, which may bias estimates. Another issue with this method is that it is difficult for buyers (also called raters) to purchase products if they are unfamiliar with the seller or do not fit the typical demographics of purchasers. Therefore, researchers using covert- purchases surveys must have detailed knowledge of the marketplace, including the ways individuals specifically ask for illicit tobacco products, and they must know whether they mirror the demographics of the neighbourhood. For example, in a study of South Bronx smokers, von Lampe et al. found that smokers looked for certain clues to assess whether they were being sold illicit cigarettes (19). Overall, this method can be quite costly because it requires training researchers, travelling to retailers and purchasing product. Covert-purchases surveys do not enable researchers to estimate the level of illicit trade, but they can provide information on availability of supply. Key studies for readers to refer to for additional guidance: Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Global Health. 2016;12(1):78. A4.1.6 SEIZURES OF GOODS Seizures are the result of enforcement activity carried out by local, national and international organizations that confiscate tobacco products that are illegally manu- factured, transported and sold. Seizures are meant to reduce the profits associated with illicit trade by confiscating proceeds (e.g. cash, cars or houses) and the tools of the trade (e.g. print and tobacco machinery). Seizures can occur at various points in the supply chain. Seizure data are often tallied by the responsible agencies and used to measure program effectiveness or as justification for requesting additional resources (e.g. personnel). Some of the data may be supplied to international customs organiza- tions, including the WCO (20). The quality of recordkeeping varies. For example, some agencies may maintain criminal files in databases that detail dates of seizure, brand names and laboratory testing. 260 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Seizures provide preliminary data on the scope of criminal activity and can help identify key trends to guide law enforcement agencies’ efforts. For example, seizures can identify trends on the modus operandi of smugglers, including transporta- tion methods (e.g. sea cargo versus trucks), point of entry and brand preference. Seizure statistics can also be used as a preliminary test to measure the efficacy of interventions. For example, Stoklosa and Ross used seizure data from the Canadian province of Nova Scotia to test the impact of a 2015 menthol ban. He found no statistically significant change in the number of menthol cigarettes seized before and after the ban (1). Advantages and disadvantages of seizures of goods Generally, seizure statistics can be readily obtained from law enforcement agencies through formal requests to agency gatekeepers (e.g. public information officers). Seizure data, however, generally do not provide a representative picture of illicit activity. For example, certain geographies may yield higher seizures because that is where the bulk of operations are being conducted. Police agencies may focus on certain geographies (e.g. locations near borders) rather than randomly inspecting, and their findings may be limited to those specific regions. Seizure data may also be skewed by the type of investigation procedures utilized. Large seizures may be the result of long-term investigations (i.e. wiretaps or culling confidential infor- mants), while smaller seizures may come from anti-smuggling cases that involve cross-border purchases of low quantities of cigarettes (less than 1 000) (2). Seizures can also be skewed by industry cooperation with law enforcement agencies. For example, the tobacco industry may be more likely to support law enforcement on counterfeit seizures rather than smuggling cases because counterfeiting impacts their brand integrity. A4.2 RESIDUAL METHODS Because the illicit tobacco trade is often decentralized, it can be difficult to observe directly. However, researchers are sometimes able to make inferences about its size without direct observation by comparing observed tobacco tax revenues with the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. The difference between observed and hypothetical revenues is called the residual and can be used as an indicator of the magnitude of illicit trade. Even when the residual is only an approximate measure, changes in its size may be a reliable indicator of changes in the size of the trade. When actual tobacco tax revenues are reliably observed, the main challenge for residual methods is that of producing accurate estimates of the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. CHAP T ER 4. PO LI T I C AL ECO N OMY 261 A4.2.1 GAP ANALYSIS Gap analysis is the preferred residual methodology because it is intuitive, straight- forward and relatively easy to explain to policy-makers and the general public, and it has been widely employed in government studies (4). Researchers using gap analysis compare survey-based self-reported consumption data with observed (usually administrative) data on tax-paid sales. The basic premise is that if both self-reports and observed data are accurate, any difference between reported consumption and tax-paid sales can be explained by legal imports of non-taxed cigarettes (such as duty-free sales), exports of taxed cigarettes, tax evasion or tax avoidance. The greatest research challenge in implementing gap analysis – as with most residual methods – is obtaining reliable and accurate estimates of tobacco consump- tion. In its simplest implementation, gap analysis calculates the residual as the difference (which should be a minimum of zero) between the amount of tobacco consumption reported in surveys and tax-paid sales, which are generally available from administrative sources, minus exports. This simple calculation, however, is generally flawed, since surveys of reported tobacco consumption underestimate true consumption. Underreporting of tobacco consumption may result from survey respondents’ reticence about disclosing behaviour that is viewed as unhealthy and potentially socially undesirable. Depending on the legal and cultural context, cer- tain groups (e.g. women or youth) may be more likely than others to underreport consumption. Other groups (e.g. rebellious young men and women) may accurately report or even overestimate consumption. Reuter and Majmundar measured actual consumption by total national taxed sales in the United States, where both legal untaxed imports and exports of taxed tobacco are widely believed to be very small, and found that the ratio of self-reported consumption to actual consumption was only 65% (4). After incorporating this survey underreporting into their analyses and considering the evidence from their gap analysis and the literature, Reuter and Majmundar found that the illicit market in the United States, which largely consists of avoidance or evasion of subnational state taxes, is between 8.5% and 21% of consumption (4). The higher range of the estimate is consistent with prior estimates using population-based pack observation studies (21). Researchers using gap analysis for countries or regions where legal imports of untaxed tobacco (such as duty-free products) or (legal or illegal) exports of taxed cigarettes are more significant should attempt to incorporate data about, or estimates of, these factors into their calculations. Legal untaxed imports of tobacco should be subtracted from reported consumption (after adjustment for underreporting), and exports of taxed tobacco should be subtracted from taxed sales. Obtaining data about legal untaxed imports and exports of taxed tobacco may be challenging, because these imports and exports may be the result of decentralized decisions of individual 262 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N travellers as they cross tax borders. Data on these activities will not necessarily be collected through normal administrative activities. Despite these challenges, gap analyses may still prove useful. For example, if researchers have reason to believe that misreporting of tobacco consumption and the amounts of legal untaxed imports and taxed exports are relatively stable or follow known trends (e.g. are falling) over time, gap analyses can be used to provide estimates or lower (or upper) bounds on illicit trade when several years of data on taxed sales and reported consumption are available. Data sources may be country tax administrators who have access to sales data and health departments that have access to population-level studies of reported tobacco con- sumption. In this context, multiple years of data on reported consumption and tax-paid sales can allow researchers to estimate changes in the size of the illicit trade even when it is difficult to measure the absolute level. Paraje used the 2008 Global Adult Tobacco Survey and the 2013 National Health Survey to measure reported tobacco consumption in Brazil (22-23). Advantages and disadvantages of gap analysis A major advantage of gap analysis is that when quality data are available, it is simple, easily reproduced and explainable to policy-makers and the general public. How- ever, high-quality data on reported consumption may not be available, especially in low-income countries. In many cases, gap analysis does not provide reliable information on the size of the illicit market but only on changes in the size over time (22). Additionally, some low-income countries may not have reliable estimates of tax-paid cigarette sales, and secondary data repositories of cigarette sales may not be transparent about their methodology (24). Another disadvantage of gap analysis is that it generally cannot be used to obtain separate estimates of tax avoidance and tax evasion. Biased estimates may also result if surveys of tobacco consumption are not representative of the population (25). Moreover, it is generally not possible to quantify the precision of the estimates or uncertainty associated with the estimates, because of both statistical uncertainty resulting from the use of samples to imperfectly represent populations (e.g. the share of the population that smokes) and uncertainty about key facts such as the degree to which survey respondents understate their tobacco consumption. Key studies for readers to refer to for additional guidance: Szklo A, Iglesias RM, Carvalho de Souza M, Szklo M, Maria de Almeida L. Trends in illicit cigarette use in Brazil estimated from legal sales, 2012–2016. Am J Public Health. 2018;108(2):265–269. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine and Tob Res. 2019;21(8):1079–86 CHAP T ER 4. PO LI T I C AL ECO N OMY 263 A4.2.2 ECONOMETRIC MODELLING There is a long tradition of using data to estimate parameters of demand functions that relate the quantity of goods consumed to the prices faced by consumers, their incomes and other variables. Because of the addictive nature of tobacco – and because of important public health and public policy concerns relating to tobacco use – economists have paid particular attention to the estimation of cigarette demand functions (26). As the literature on this topic developed, it became apparent that taxed tobacco sales would be a biased indicator of tobacco consumption if some consumers obtained their tobacco in illicit markets. Similarly, the price of cigarettes in the legal market might overestimate the price paid by consumers if some sales were not tax-paid. While economists generally cannot observe sales in the illicit market, they have been able to develop models that predict conditions under which consumers avoid tobacco taxes. They reason that the relative size of illicit tobacco markets depends primarily on two variables: the relative price of taxed and untaxed consumption and the ease of obtaining lower-cost (untaxed) tobacco. Other variables, includ- ing the social stigma from evading tax laws and the perceived relative quality of illicit tobacco, could also influence the demand for it. While illicit trade cannot be directly observed, it can be estimated from the difference between tax-paid sales and predicted consumption. Tax-paid sales can be less than predicted consump- tion when retailers or consumers evade taxes. They can be greater if some tax-paid cigarettes are bought within the jurisdiction and then consumed in areas where after-tax prices are higher. Econometric modelling estimates of illicit trade must be tailored to the situation in the country that is being studied, and therefore the data requirements may differ substantially from case to case. Researchers using this method should be familiar with the literature and should also understand the conditions in the areas they are researching. They must always include some measure of tobacco consumption or sales and some measure of the price of tobacco in the home country, as well as other variables (e.g. income) that are known to affect the demand for tobacco. It is also generally necessary to include variables that measure the availability and relative price of illicit tobacco, which can often be measured by comparing tobacco taxes in the home country with those in areas that are the source of illicit tobacco. Advantages and disadvantages of econometric modelling The major advantage of econometric modelling is that it is consistent with a long tradition of economic theory and practice, and the quality of the modelling techniques and empirical estimates can therefore be evaluated against widely accepted criteria. Empirical analyses provide estimates of price elasticities, income elasticities and price elasticities of tax avoidance. A substantial literature base makes it possible to 264 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N rigorously quantify uncertainty about the estimates and to test their robustness to various assumptions made in the modelling process. Estimates can be compared to other estimates available in the literature, and the results of these analyses can be used to simulate the impacts of policy changes (including tax and/or enforcement increases) on both consumption and tax avoidance. Because this methodology provides direct estimates of the uncertainty of the results, researchers can specify their level of confidence in the findings. A shortcoming of econometric modelling is that it requires high-quality data on a variety of important variables over a period of time, as well as advanced econo- metric modelling expertise. Also, because results from the econometric models are based on statistical inference and economic theory rather than direct observation (e.g. the proportion of packs without tax stamps), it can be difficult to explain to policy-makers and the general public. Key studies for readers to refer to for additional guidance: Becker GS, Grossman M, Murphy KM, (1994). An empirical analysis of cigarette addiction. Amer Econ Review. 1994;84(3):396–418. Schafferer C, Yeh CY, Chen SH, Lee JM, Hsieh CJ. A simulation impact evaluation of a cigarette excise tax increase on licit and illicit cigarette consumption and tax revenue in 36 European countries. Public Health. 2018;162:48–57. A4.2.3 EXPERT OPINION (KEY-INFORMANT SURVEYS AND INTERVIEWS) Insight on illicit trade dynamics can come from experts in the field, including researchers (e.g. in economics, criminal justice and public health), taxation depart- ments, enforcement agencies, product manufacturers, wholesalers and retailers. Other key informants include journalists and academics who have secured confi- dential informants. Experts can provide novel information about emerging trends (e.g. new smuggling routes). In some cases, researchers can obtain interviews with incarcerated or active offenders (27-29). For example, researchers studying cigarette smuggling in eastern Africa conducted interviews with more than 150 Ugandan tobacco smugglers (29). Experts can be queried through surveys or semi-structured interviews. When sampling frames are available (e.g. directories of tax department employees), surveys are more expedient than interviews. However, when experts are hard to find, non- random sampling strategies coupled with interviews are recommended. Identifying experts may require recruiting a gatekeeper who is tasked with helping researchers find additional experts; or purposive sampling, where individuals are identified based on set criteria (e.g. they are taxation experts employed by local governments) (30). CHAP T ER 4. PO LI T I C AL ECO N OMY 265 Advantages and disadvantages of key-informant surveys and interviews Informant interviews can be a useful starting point for identifying trends in the marketplace (e.g. venues where illicit cigarettes are sold or modes of entry). One of the disadvantages of relying on informants is that the information solicited from them may not be generalizable. Expert knowledge may be outdated or limited by the informants’ experience. Furthermore, the opinions of experts are subjective and may be biased by the experts’ employment status and the sampling methods used. For example, persons working in law enforcement may overestimate the extent of bootlegging in order to secure additional funding for future operations. Similarly, manufacturers looking to defeat taxes may overestimate the illicit trade to illustrate the links between taxation and illicit behaviour. Alternatively, tobacco control advocates may underestimate illegal market measures in order to support the argument that taxes do not increase illicit trade. Key studies for readers to refer to for additional guidance: Joossens L, Raw M. Cigarette smuggling in Europe: who really benefits? Tob Control. 1998;7:66–71. doi:10.1136/tc.7.1.66 PMID: 9706757. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20(3):226–232. A4.3 MIXED AND MULTIMETHOD STUDIES Given the shortcomings of the aforementioned methods for assessing the nature and size of the illicit tobacco trade, governments may want to validate their findings by using mixed or multiple methodologies. Mixed methods use two methodological paradigms, qualitative and quantitative, as tools for exploration and explanation. For example, mixed method studies can use littered-pack surveys to measure the size of the market along with self-report surveys of smokers to understand patterns of purchasing, including sources, frequency and social norms. For example, Stoklosa and Ross estimated the share of the illicit market in Poland using a population- based self-report survey and a littered-pack survey (1). Using both types of survey simultaneously enables governments to assess their validity in estimating the size of the illicit market. Alternatively, governments can employ multimethod research, i.e. the use of multiple methods that are similar in tradition (e.g. focus groups and semi-structured interviews) (31). Saenz de Miera et al. used face-to-face interviews (households), litter collection and observation of single-stick sellers, which enabled them not only to cross-validate the two major methodologies, but also to see if the brand of the single stick was a good measure of licit versus illicit trade (33). 266 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Advantages and disadvantages of mixed and multimethod studies Mixed and multimethods studies enable researchers to check the validity of their findings. Multiple methods are preferred in contexts where illicit trade estimates are politicized. For example, low estimates may be challenged by the tobacco indus- try, while high estimates may be challenged by tobacco control researchers and/or proponents. Mixed and multiple methods (e.g. littered-pack surveys and informant interviews) can enable governments to understand the situational context in which the illicit trade operates, including the actors involved and venues of sale. One disadvantages of using mixed and multiple methods is cost. Governments that are constrained by tight budgets may choose to use a single method that provides the most accurate information. However, given the issues faced by each method, this may not be feasible – each method has limitations. Instead, governments can pair a high-cost method with a lower-cost method (e.g. pairing interviews with empty pack surveys, or law enforcement seizure data with face-to-face consumer surveys). Key study for readers to refer to for additional guidance: Zaloshnja E, Ross H, Levy DT. The impact of tobacco control policies in Albania. Tob Control. 2010;19:463–468. CHAP T ER 4. PO LI T I C AL ECO N OMY 267 REFERENCES 1. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 2. Joossens L, Lugo A, La Vecchia C, Gilmore AB, Clancy L, Gallus S. Illicit cigarettes and hand-rolled tobacco in 18 European countries: a cross-sectional survey. Tob Control. 2014;23:e17–e23 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC3812425/pdf/nihms491463.pdf, accessed 2 February 2021). 3. Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–7. 4. Reuter P, Majmundar M. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Washington (DC): National Academies Press; 2015. 5. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 6. Barkans M, Lawrance KA. Contraband tobacco on post-secondary campuses in Ontario, Canada: analysis of discarded cigarette butts. BMC Pub Health. 2013;13(1):335 (https://bmcpublichealth. biomedcentral.com/articles/10.1186/1471-2458-13-335, accessed 2 February 2021). 7. John RM, Ross H. Illicit cigarette sales in Indian cities: findings from a retail survey. Tob Control. 2018;27(6), 684–688. 8. Davis KC, Grimshaw V, Merriman D, Farrelly MC, Chernick H, Coady MH, et al. Cigarette trafficking in five northeastern US cities. Tob Control. 2014;23(e1):e62–e68. 9. Callaghan RC, Veldhuizen S, Leatherdale S, Murnaghan D, Manske S. Use of contraband cigarettes among adolescent daily smokers in Canada. CMAJ. 2009;181(6-7):384–6 (https://www.cmaj.ca/content/ cmaj/181/6-7/384.full.pdf, accessed 2 February 2021). 10. Guindon GE, Burkhalter R, Brown KS. Levels and trends in cigarette contraband in Canada. Tob Control. 2017;26(5):518–25 (https://www.researchgate.net/publication/307890937_Levels_and_trends_in_ cigarette_contraband_in_Canada, accessed 2 February 2021). 11. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany: New York State Department of Health; 2006 (https://www. health.ny.gov/prevention/tobacco_control/docs/cigarette_purchasing_patterns.pdf, accessed 2 February 2021). 12. International Tobacco Control Evaluation Project (2018). Surveys. Waterloo: University of Waterloo; 2018 (http://www.itcproject.org/surveys, accessed 12 October 2020). 13. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 14. Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. 15. Brown J, Welding K, Cohen JE, Cherukupalli R, Washington C, Ferguson J, et al. An analysis of purchase price of legal and illicit cigarettes in urban retail environments in 14 low-and middle-income countries. Addiction. 2017;112:1854–60 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5600117/pdf/ADD- 112-1854.pdf, accessed 2 February 2021). 16. Scollo M, Bayly M, Wakefield M. Availability of illicit tobacco in small retail outlets before and after the implementation of Australian plain packaging legislation. Tob Control. 2015;24(e1):e45–e51. 17. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Globalization and Health. 2016;12(1):78 (https://globalizationandhealth. biomedcentral.com/articles/10.1186/s12992-016-0219-z, accessed 2 February 2021). 18. Phillip Morris USA Inc. v. Shalabi. United States, District Court, C.D. California; 2004 (https://www. casemine.com/judgement/us/5914b6cbadd7b0493477b3da, accessed 2 February 2021). 19. von Lampe K, Kurti M, Johnson J, Rengifo AF. ‘I wouldn’t take my chances on the street’ navigating illegal cigarette purchases in the South Bronx. J Res Crime Delinq. 2016;53(5):654–80 (https://www. 268 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N researchgate.net/publication/298515449_I_Wouldn’t_Take_My_Chances_on_the_Street_Navigating_ Illegal_Cigarette_Purchases_in_the_South_Bronx, accessed 2 February 2021). 20. Customs and Tobacco Report 2009. Brussels: World Customs Organization; 2009. (http://www.wcoomd. org/en/media/newsroom/2010/june/~/media/83967DFEB9F74D388924A4C61F279DC4.ashx, accessed 12 October 2020). 21. Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ, et al. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61–i66 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3984758/pdf/nihms567943.pdf, accessed 2 February 2021). 22. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine Tob Res. 2019;21(8):1079–86. 23. National Survey of Health. Instituto Brasileiro de Geografia e Estatística; 2013 (in Portuguese) (https:// www.ibge.gov.br/en/statistics/social/health/16840-national-survey-of-health.html?=&t=downloads, accessed 27 November 2020). 24. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 25. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 26. Chaloupka FJ, Warner KE. The economics of smoking. In: Arrow KJ and Intriligator MD, editors. Handbook of Health Economics. Amsterdam: Elsiver; 2000. pp1539–1627. 27. Antonopoulos GA. Cigarette smugglers: a note on four ‘unusual suspects’. Glob. Crime. 2007;8(4):393–8 (https://www.researchgate.net/publication/248955367_Cigarette_Smugglers_A_Note_on_ Four_’Unusual_Suspects’, accessed 2 February 2021). 28. Antonopoulos GA. The Greek connection(s): the social organization of the cigarette-smuggling business in Greece. Eur J Criminol. 2008;5(3):263–88 (https://www.researchgate.net/publication/249752218_ The_Greek_ConnectionsThe_Social_Organization_of_the_Cigarette-Smuggling_Business_in_Greece, accessed 2 February 2021). 29. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20:226–32 (https://www.researchgate.net/publication/49809289_ Blood_cigarettes_Cigarette_smuggling_and_war_economies_in_central_and_eastern_Africa, accessed 2 February 2021). 30. Babbie, ER. The basics of social research. Boston: Cengage Learning; 2013. 31. Tashakkori A, Teddlie C., editors. Sage handbook of mixed methods in social & behavioral research. New York: Sage; 2010. 32. Saenz de Miera Juarez B, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 269 ANNEX 4.2 HOW ARE THE TOBACCO TAX REVENUES EARMARKED? The introduction of earmarking of tobacco tax revenue is almost always combined with an increase in excise taxes (or a new surcharge) rather than reallocation of existing revenues (1). Table A4.1 provides examples of the different approaches used by several countries to earmark tobacco tax revenues. Table A4.1 Approaches used to earmark tobacco tax revenues FUNDING SOURCES/ TYPE OF TAX EXAMPLES OF TAX BASE AND RATES As part of the excise system (tobacco, alcohol) Specific Republic of Korea: 841 won (US$ 0.75) per pack or 29% of the specific excise rate Costa Rica: 467.8 cólones (US$ 0.83) per pack or 100% of the specific excise rate Congo: 20 CFA francs (US$ 0.036) per pack or 50% of the specific excise rate Ad valorem Colombia: 10% of retail price (equivalent to 100% of the ad valorem rate) New levy (surcharge on the existing excise or completely new levy) Specific Egypt: additional 0.75 Egyptian pounds (US$ 0.042) per pack Ad valorem Thailand: surcharge of 2% over the excise tax base Botswana: new tobacco levy of 30% of the cost of production or CIF Percentage of excise revenue Cook Islands: 50% of revenues from the excise tax on tobacco Guatemala: 100% of revenues from the excise tax on tobacco Note: Conversions of amounts from the local currency were made using the official exchange rates from the IMF as of 31 July 2018 (date of the data collection). Source: (2). REFERENCES 1. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 2. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 270 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 271 CHAPTER 5. Best practices in tobacco tax policy and administration TAX POLICY Use excise tax increases to achieve the public health goal of reducing the death and diseases caused by tobacco use Extensive research has clearly demonstrated the effectiveness of higher tobacco product taxes and prices in reducing tobacco use and its harmful consequences, particularly among the poor and the young. In fact, tobacco excise tax increases are the single most effective and cost-effective policy for reducing tobacco use. Excise taxes are the most significant taxes applied on tobacco products because of their ability to raise both absolute and relative prices. Tobacco excise tax increases also generate sizeable new revenues that will be sustained in the short to medium term. In the long term, continued increases in tobacco taxes – coupled with implementa- tion of other evidence-based tobacco control policies and programmes – will lead to even larger reductions in tobacco use and its consequences. Include significant tobacco excise tax increases as part of a comprehensive strategy to reduce tobacco use Governments should adopt a comprehensive tobacco control strategy that includes objectives for reducing adult tobacco use and preventing youth tobacco use. Rais- ing excise taxes significantly is the most effective, as well as the most cost-effective, measure for reducing consumption. When combined with other demand reduction interventions, the impact of tax increases on tobacco use is even stronger. Such interventions include comprehensive smoke-free policies in all public spaces, total bans on tobacco advertising, promotion and sponsorship by tobacco companies, large graphic health warnings about the consequences of tobacco use, plain packaging, broad efforts to help current users quit and mass media public education campaigns. Implementation of a comprehensive strategy to reduce tobacco use leads to greater reductions in the harmful consequences of tobacco use, builds public and political support for higher taxes and maximizes the effectiveness of tax increases in achieving public health objectives. 272 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Involve the competent authority from the start when considering the revision of a tax policy Competent authorities such as tax administrations and customs authorities are key partners in the effective implementation of a tax policy. Policy-makers need to ensure that those competent authorities are consulted and involved in the tax policy revision process so that their concerns about the impacts of policy change on enforcement can be taken into account from the beginning. This can also help identify and address possible loopholes early on in the enforcement process. Coordination among relevant bodies, including close cooperation and sharing of information, will optimize enforcement of tax policy and tax collection. To streamline the process of cooperation and exchanges of information, a basis in law needs to be established. Additionally, the involvement of tax administration authorities in the entirety of the tax revision policy process is important to ensure effective implementation of the policy. Promote greater policy coherence across sectors such as agriculture, industry, trade, finance and labour Greater multisectoral integration and policy coherence is needed at the country level to achieve effective health improvements. In particular, it is important to ensure that public policies and interventions in non-health sectors (e.g. agriculture, industry, trade, finance and labour) do not act against the intended public health impact of tobacco control and taxation (such interventions include providing subsidies to tobacco growing or manufacturing). TAX DESIGN Tax structure matters and simpler is better Complex tax structures are difficult to administer, create opportunities for tax avoid- ance and evasion and are less effective than simpler structures in achieving public health and revenue goals. Simplifying the structure of tobacco excise taxes will facilitate tax administration, reduce tax avoidance and evasion, enhance revenues and have a greater impact on tobacco use by reducing incentives to substitute among tobacco products or brands in response to tax increases. Countries with multiple tiers of tobacco tax rates based on product characteristics (e.g. price level, length, weight, type of tobacco) should reduce and eventually eliminate these differential tax rates. An appropriate transition strategy is to reduce the variations in tax rates over time with the aim of implementing a uniform tax (i.e. a single rate applies whether excise is ad valorem or specific) on a given tobacco product. Applying a uniform tax to all brands of a given tobacco product also sends a clear message that they are equally harmful. CHAP T ER 5. B E S T PR AC T I CE S 273 Rely more on specific tobacco excises to drive price increases Greater reliance on specific excise taxes maximizes the impact of tobacco taxes on public health by reducing the gap in prices between premium and low-priced alternatives and limiting opportunities for users to switch down in response to tax increases. For countries that currently rely on an ad valorem tax, an appropriate first step would be to shift to a mixed system by adding a sizeable specific component or introducing a high minimum specific excise tax (an excise tax floor). For countries that rely on a mix of ad valorem and specific taxes, the specific tax component should be increased regularly so that it accounts for a greater share of the total excise tax. Increase tobacco taxes significantly to reduce the affordability of tobacco products To maximize the public health impact of higher tobacco taxes while at the same time generating higher revenues, governments should significantly raise taxes to increase prices and reduce the affordability of tobacco products. In many LMICs, tobacco use increases with incomes, and since incomes rise faster than tobacco product prices, these products are becoming more affordable. To reduce afford- ability, tax increases need to result in real price increases that are higher than the increases in real incomes. Where revenue increases are a goal, rely on regular excise tax increases If governments want to increase tobacco revenues, they must increase excise taxes regularly. From the tax revenue perspective, the important determinant is the tax base elasticity, which has three key components: the price elasticity of demand of tobacco, the share of the tax in the retail price and the degree of pass-through of the excise tax rate increase on to retail price. Tax increases will increase revenues at least in the short to medium term, because demand is price inelastic, tax levels are generally low as a proportion of retail prices and the pass-through of tax increases on to retail prices is unlikely to be higher than the tax increase itself (i.e. there is no overshifting). In addition, increasing tax rates is the only policy measure that can reverse reduced revenues in a declining market that has strong tobacco control policies. Automatically adjust specific tobacco taxes for inflation and income growth Unless specific tobacco taxes are regularly adjusted, their real value will fall over time as general price levels increase. When this happens, their effectiveness in reducing tobacco use will be diminished. Governments should establish a mechanism for automatically adjusting specific taxes to keep pace with inflation. Recently, some governments have begun to extend this indexation to include income growth as well, further ensuring that tobacco does not become more affordable over time. 274 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Pricing regulations cannot be considered an alternative to excise tax. However, in some specific contexts, pricing regulations could be used in conjunction with excise taxes to help ensure the effective implementation of tax increases In certain contexts where increasing taxes is challenging or the tax structure is weak, non-tax policies such as pricing regulation (specifically, minimum mark-ups and price floors/minimum prices) may be seen as a second-best alternative to ensure a high price level and dissuade consumption of tobacco products. These policies, however, do not necessarily lead to the desired price level, nor do they protect consumers and government from industry manipulation. However, in the context of powerful multinationals that sell brands across all market segments and could easily undershift a tax increase to cheaper brands – or where price promotions cannot be banned – minimum price policies may help increase the effectiveness of tax increases, especially if the minimum prices are increased regularly. Implement nontax policies affecting price levels, such as banning promotional discounts for tobacco products and the sale of single sticks of cigarettes The banning of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Do not allow concerns about the inflationary impact of higher tobacco taxes to deter tax increases Given that wages or some government spending may be tied to a price index, govern- ments can reduce concerns about the inflationary impact of a tobacco tax increase by using a price index that excludes tobacco products. TAX PARITY Tax all tobacco products in a comparable way Increasing excise taxes on some tobacco products but not on others results in changes in the relative prices of these other products. This induces substitution towards relatively less-expensive products – for example, from expensive manufactured cigarettes to other, cheaper tobacco products such as RYO tobacco, bidis, cheroots or chewing tobacco. As a result, the overall reduction in tobacco use is smaller than it would have been had all taxes increased by comparable amounts. Comparable increases in the taxes on all tobacco products maximize the public health impact of tobacco tax increases by minimizing opportunities for substitution. Moreover, increases in taxes on all tobacco products will generate larger increases in revenues. CHAP T ER 5. B E S T PR AC T I CE S 275 Strictly regulate new and emerging tobacco and nicotine products where they are not banned and impose an excise tax In recent years, the world has been experiencing the rise of new and emerging tobacco and nicotine products including ENDS, ENNDS and HTPs. The tobacco industry claims these new products are safer than traditional tobacco products, but the evidence so far suggests that they could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of newer products – as well as initiation, smoking cessation and switching behaviour among different socioeconomic groups – are not yet clear. Best practices for taxing new and emerging tobacco and nicotine products, based on current knowledge, are that: 1. HTPs should be taxed at the same level as cigarettes and, in terms of structure, through a specific excise per unit regardless of tobacco content. HTPs contain tobacco and should be treated as a tobacco product. 2. ENDS/ENNDS products should be taxed in a manner that discourages up- take by youth and non-users. Nicotine- and non-nicotine-delivery systems containing e-liquids should be taxed equally. 3. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While these newer products create additional challenges for tobacco control, it is important to remember that cigarettes remain by far the predominant tobacco product and that raising taxes and prices on cigarettes – and thereby reducing their use – should remain the top priority. MONITORING AND EVALUATION Know your market Know your market well. The type of tax structure you choose and the impacts it will have on consumption and tax revenue are shaped by the particular dynamics of your market. Understanding the nature and degree of competition in your market is vital to selecting the appropriate type of tax structure and policies to achieve your public health and revenue objectives. This knowledge will also facilitate more accurate estimates of the impacts of a tax increase, as well as better anticipation of industry responses. Assess the impact of your policies to design and implement the most effective tobacco excise tax policies Monitoring and evaluation are essential for effective tobacco taxation, and they 276 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N should be built into the initial design – or redesign – of tobacco tax policies. A number of tools exist to help policy-makers pre-emptively assess the effects of a proposed tobacco control policy on consumption, smoking prevalence and lives saved. The WHO TaXSiM uses target simulations to assist governments in predict- ing how specific tax changes will impact consumer prices, consumption and tax revenue in their market. Adopt indicators that help you measure improvements in tax policy and its impact Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the improvement of their policies and determine if those policies have an impact on tobacco use over time. The tax share of the retail price for a particular tobacco product is a key indicator that should be used in conjunction with an af- fordability indicator. A recommended target for countries to aspire to is to have an excise tax that represents at least 70% of the retail price of tobacco products. Another useful indicator to assess the performance of the tax policy overall is the use of a tax scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). TAX ADMINISTRATION Implement best practice approaches in general tax administration to make tobacco tax administration more effective and efficient Best practice approaches include (1) defining clearly the roles and responsibilities of competent authorities, (2) ensuring effective coordination among relevant bodies at the national and international levels and (3) undertaking evaluation of performance and accountability against pre-agreed indicators to identify points for improvement. Ensure compliance and accuracy of information on the tax compliance cycle To achieve this, implement the following actions: • Require licences for manufacturing, importing, exporting, retailing, growing, transporting, wholesaling, brokering, warehousing and distributing tobacco products. This will help secure the supply chain while obtaining valuable infor- mation, e.g. through access to companies’ accounting and inventory systems. • Make sure all persons and entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment keep complete and accu- rate records of all relevant transactions and details of materials used in the production of tobacco products. CHAP T ER 5. B E S T PR AC T I CE S 277 • Ensure that tax declarations collect as much information as possible on the taxpayer. • Collect taxes close to the point of production and import to limit the number of taxpayers a competent authority needs to manage. • Maintain a system of authorization for warehousing to carry out controls in production and storage facilities to ensure that taxes are paid. • Use electronic methods, through the best available IT, for declarations and collection of taxes. This allows for cross-check of information provided in dec- larations with information from other government agencies and third parties. Ensure control and enforcement on the supply chain To achieve this, implement the following actions: • Include control and enforcement as a fundamental pillar in the strategic plan of the tax administration overall. • Use a risk-based approach by choosing defined targets for enforcement and control, such as those who have a higher probability of noncompliance. • In the licensing process, ensure that purchases from unlicensed suppliers or sales to unlicensed purchasers are not allowed. Ensure also that the validity of licences is limited in time and require renewals or reapplication to maintain a high level of control. • Use tax stamps with strong security features to reduce the risk of stamp counterfeiting. These markings facilitate the collection of excise taxes, audits and enforcement actions. • Implement a tracking and tracing system for tobacco products. A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, and monitoring and control- ling the movement of tobacco products and their legal status. • Implement anti-forestalling measures so that forestalling does not delay a tax increase and its intended effect on revenues and consumer behaviour. • Control import and export of tobacco products and manufacturing equipment by allowing only duly licensed natural persons or legal entities to conduct such activities. • Strengthen border control, e.g. by utilizing non-invasive tools such as X-ray scanners and dogs to detect tobacco products. • Limit or tightly control and, ideally, ban activities related to production and trade of tobacco products in tax-free zones to avoid opportunities for tax evasion. • Prohibit intermingling of tobacco products with non-tobacco products in a single container or any other similar transportation unit when removed from tax-free zones. 278 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Prohibit the sale to or import by international travellers of tax-free or duty- free tobacco products. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and adversely affect government revenues by creating a loophole in the tax structure. Clearly define procedures to follow after detecting illicit trade of tobacco • Take immediate action to seize and destroy smuggled and/or illicit tobacco and collect due taxes. • Ensure certain, swift and severe sanctions for those caught engaging in illicit trade in tobacco products, such as penalties, fines and withdrawal of licences. It can also be effective to consider illicit trade in tobacco products by law as a source of money-laundering. Become a Party to and/or implement the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products The WHO FCTC Protocol provides a blueprint of best practices and policies for dealing with illicit trade and should be part of any strategy to fight it. Implement, to the extent possible, the same rules and regulations for tax administration and enforcement for all tobacco products, as well as new and emerging nicotine and tobacco products Implement broad policies for ensuring a good tax system that will trickle down to good tax administration of tobacco products by: • ensuring proper resourcing of competent authorities; • having strict rules and regulations to detect corruption and to punish both personnel and taxpayers who are engaged in corrupt practices; and • ensuring a strong judicial system that is independent in fact and in perception, where disputes are solved quickly. The appeal process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially for illicit trade. POLITICAL ECONOMY Beyond the technical soundness of best practices in tax policy and administration, a critical factor in advancing tobacco taxes is the ability to get the political buy-in of the highest instances in the government. One key strategy is to address concerns around the political economy of tobacco taxation, which are often exploited by the tobacco industry to block major reforms. CHAP T ER 5. B E S T PR AC T I CE S 279 SCARE tactics The tobacco industry uses SCARE tactics to dissuade governments from implement- ing tobacco tax increases. These include smuggling and illicit trade (S), court and legal challenges (C), anti-poor rhetoric (A), revenue reduction (R) and employment impact (E). Best practices for countering these tactics are described below. S: Smuggling and illicit trade Do not allow concerns over the impact of increasing excise taxes on illicit trade in tobacco affect your decision to increase them. Rely on your own estimates of the level and nature of illicit trade and not on the industry’s estimates. Illicit trade in tobacco products continues to be a major concern for tax administrators because of the difficulties associated with accurate and independent measurement of it, as well as with its elimination. Industry figures provide a distorted understanding of the extent of the problem, along with a monocausal explanation of the link between illicit trade and tobacco taxation. It is therefore recommended that governments (1) assess independently and with the best statistical practices the size of the illicit trade to assess the scope of the problem; (2) address directly the country-specific institutional and/or governance challenges, including multilateral coordination, and improve tax and customs administrations practices; and (3) implement best practices to fight illicit trade, contained in the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products. Ideally, accede to the Protocol if not yet a Party. C: Court and legal challenges Do not let tobacco industry threats of court and legal challenges to tax increases or reforms prevent you from improving your tax policy. Closely follow legal requirements for design, procedure and consultation to strengthen your legal position and minimize the possibility that any challenge will be raised. Health-protective and non-discriminatory tobacco excise taxes are legally defensible, and industry threats will usually be baseless. Your legal position can be strengthened, however, by exercising care with a tax measure’s procedure, design and consultation: (1) determine the standard of consultation required under domestic law and any applicable international obligations; (2) distance the tobacco industry from the policy-making process to the extent that this is permissible; (3) avoid unnecessary and unjustified discrimination towards foreign tobacco products or investors in the design, implementation or enforcement of a tax measure; and (4) do not offer investment incentives in the form of inducements or contractual undertakings, as these may be binding in and of themselves or grounds for a challenge under an international investment agreement. 280 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A: Anti-poor rhetoric Do not allow concerns about the regressivity of higher tobacco taxes prevent tobacco tax increases. In recent years, there has been an overwhelming increase in the evidence concern- ing the positive distributional impact of tobacco taxes and tax increases. Indeed, tobacco taxation and tax increases are actually a progressive or pro-poor policy once these wider considerations are properly accounted for. In its effort to lobby against tax increases, the tobacco industry often claims that tobacco taxation will hurt the poor. This argument is based on the concept of regressivity in relation to taxation. Conceptually, a tax is regressive if it means lower-income people must pay a relatively greater proportion of their household income to meet the tax liability than wealthy people. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use that are disproportionately experienced by lower socioeconomic groups. Second, higher tobacco taxes and prices can induce behavioural change in the population, as reflected in the price elasticity of demand, which means that lower-income smokers will curtail their smoking the most and thus will benefit disproportionately in terms of health gains from reduced tobacco consumption and use. In fact, these broader considerations make tobacco taxation a progressive, rather than regressive, public health intervention. R: Revenue reduction Do not let fears of potential revenue reductions prevent you from increasing excise taxes on tobacco products. Tax increases, even in countries with already high taxes, bring in additional revenue. Arguments by tobacco control opponents that tax increases will not result in increases in revenue are unfounded. The relatively price inelastic nature of cigarette demand, combined with the low tax share and no overshifting of the tax, means that for most, if not all, countries, increases in revenues will accompany increases in taxes. If tax increases are carefully designed and tax administration is functional, it is extremely unlikely that tax increases will lead to revenue decreases. E: Employment impact Do not allow concerns about employment impact to prevent tobacco tax increases. The tobacco industry often seeks to frame tobacco taxes as an economic issue rather than a public health issue. Particular emphasis is placed on the alleged threat tax increases pose to employment in tobacco farming and manufacturing, as well as related industries. This so-called choice between health and jobs, however, is largely based on exaggeration. The tobacco industry exaggerates the importance of tobacco CHAP T ER 5. B E S T PR AC T I CE S 281 employment relative to total national employment and overstates the impact that domestic demand reduction from local taxes will have on tobacco farmers serving a global market. The argument used by the industry also ignores the fact that expendi- tures on tobacco do not disappear but rather are redistributed to other consumption that can produce a similar or higher number of jobs. Case studies demonstrate the possibility and methods for governments to support farmers in transitioning to other crops that provide similar and often better returns with greater sustainability. Earmarking Consider earmarking tobacco tax revenues for health-focused programmes, especially if it helps advance tobacco control efforts and, more specifically, efforts to implement large tobacco tax increases and tax reforms. This could have the additional benefit of funding health programmes where they are poorly funded or not prioritized. From a tobacco control perspective, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. Earmarking is a tool to improve the political economy of tobacco taxation; it is only a secondary issue, after the primary goal of reducing demand for tobacco. Evidence shows that public support for higher tobacco taxes is greater when at least some of the increased revenues are explicitly used to support health-focused programmes. Current evidence shows that the amounts effectively earmarked for health have been relatively small and therefore unlikely to introduce rigidity in government budgets. At the same time, in some countries, those funds have helped to imple- ment much needed underresourced health programmes. The payoffs will be seen in the future as fewer people fall ill and need less medical care for tobacco-related illnesses. Earmarking tobacco tax revenues to fund high-burden/low-priority health programmes could pave the way for raising awareness about the importance of such programmes and their effectiveness, thereby convincing governments to redefine their priorities and commit to including the programmes in their regular budget. 282 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) Focus on tobacco taxation’s purposes Tobacco tax policy should aim not only to increase revenues but also to decrease consumption and improve health. To both raise revenue and reduce consumption, you need to (1) simplify tobacco tax structures, (2) significantly increase rates to impact price levels, and (3) regularly adjust rates to at least account for inflation and income growth. Analyse your tax structure and identify its weaknesses You need to analyse and identify the problems of your current tax structure to know which steps to consider next. Which tax structure do you have: specific, ad valorem, mixed, or no excise? Identify the improvements to be made to the existing tax policy/structure Your present tax structure and tax situation will provide you with the steps you would ideally take next to achieve the aims in Step 1. Specific: 1. Ensure that the tax automatically adjusts for inflation and income growth effects. 2. Ensure that all price promotions are banned. Ad Valorem: 1. Ensure that the tax base of the ad valorem is retail price. 2. Introduce a high specific excise component (and a minimum specific excise). 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. Mixed: 1. Ensure that the tax base of the ad valorem component is retail price. 2. Ensure that you are using a high specific excise component and a minimum specific excise. 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. No Excise: 1. Introduce a high specific excise. 2. Ensure that the rate automatically adjusts for inflation and income growth effects. 3. Ensure that all price promotions are banned. Assess tobacco taxation’s political economy Reform must begin with an assessment of tobacco taxation’s political economy: (1) learn from past successes and failures – what went wrong, what went right, what you can do differently this time; (2) assess the reform’s strengths and weaknesses, likely opportunities and risks; (3) determine who the main supporters and opponents of reform inside and outside of government have been and may be, based on past reforms and current situation; and (4) anticipate argu- ments that will be used against the reform (refer to SCARE tactics). STEP 1 STEP 2 STEP 3 STEP 4 Prepare a plan for realizing the reform Focusing on the overall aims identified in Step 1, the steps for achiev- ing them identified in Steps 2 and 3 and the political economy around this reform as identified in Step 4, prepare your plan: 1. Be clear on the non-negotiable objectives for the reform and the trade-offs you are prepared to make to realize them. 2. Develop a plan to approach potential allies and win them over to the reform efforts. 3. Develop the counterarguments that will be needed in response to the SCARE arguments identified earlier. 4. Prepare the evidence you will need ahead of time. To do this, get support from academics and relevant intergovernmental agencies. Mobilize a coalition for reform 1. Formulate a strategic communications plan: aim for political support both at the highest levels and among the public (framing tobacco taxation as a health issue has helped win political support in many countries). 2. Identify champions in government: ensure that finance and health officials are on the same page; involve implementing departments, such as enforcement agencies, from the start. 3. Mobilize allies from academia, civil society and the private sector to counter the anticipated pushback from the tobacco industry, its proxies and its allies. Monitor and evaluate To make the most well-informed policy decisions, a reform effort should be monitored to assess its overall impact and its effect on key indicators; this will help identify issues to be fixed while also creating a strong evidence base for further reform efforts. Get and analyse the relevant data to better understand the market situation and its dynamics: 1. Monitor the market and its evolution (e.g. retail prices, duty-paid sales, market shares). 2. Get regular estimates of price elasticity (including cross-price elasticity), income elasticity and tax base elasticity to evaluate any changes in tobacco demand. Use relevant tools to assess the impact of the tax policy on consumption and revenue: 1. Use specific tools on the impact of excise on price, consumption and revenue (e.g. the WHO TaXSiM). 2. Use global tools to assess the tax increase’s impact on prevalence (e.g. the WHO ISPT). Monitor key indicators closely to assess improvements over time: 1. Tax as a percentage of retail price. 2. Change in affordability of tobacco products over time. 3. Change in the tobacco tax scorecard, which combines a mix of best practices in tax policy. 4. Change in sales, prevalence and illicit trade in tobacco products. 5. Improvements in MPOWER package achievement. STEP 5 STEP 6 STEP 7

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION WHO technical manual on tobacco tax policy and administration ISBN 978-92-4-001918-8 (electronic version) ISBN 978-92-4-001919-5 (print version) © World Health Organization 2021 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/ rules/). Suggested citation. WHO technical manual on tobacco tax policy and administration. Geneva: World Health Organization; 2021. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Document design by Ana Sabino. iii Contents Foreword v Acknowledgements vii Acronyms viii Executive summary xi CHAPTER 1. Why this manual? 1 CHAPTER 2. Tobacco excise tax policy 11 2.1 Global overview of tobacco tax practices 11 2.2 Designing excise tax policy 18 2.3 Domestic and regional policy integration 55 2.4 New and emerging nicotine and tobacco products 59 2.5 Conclusions 75 ANNEX 2.1 Countries that apply different types of excise tax structures 89 ANNEX 2.2 Analytics of the tax base elasticity 90 ANNEX 2.3 Elements of the devices that make up ENDS/ENNDS products 91 CHAPTER 3. Tobacco tax administration 93 3.1 Introduction 93 3.2 Institutional arrangements 94 3.3 The tax compliance cycle 102 3.4 Control and enforcement 115 3.5 Tax administration of other tobacco products 152 3.6 The broader elements of a good tax system 155 3.7 Conclusions 156 ANNEX 3.1 Composition of tobacco products 166 ANNEX 3.2 Example of forestalling and countermeasures 172 CHAPTER 4. Political economy 175 4.1 SCARE tactic S: Smuggling and illicit trade 176 4.2 SCARE tactic C: Court and legal challenges 198 4.3 SCARE tactic A: Anti-poor rhetoric or regressivity 212 4.4 SCARE tactic R: Revenue reduction 217 4.5 SCARE tactic E: Employment impact 228 4.6 Earmarking tobacco tax revenues to fund health 233 ANNEX 4.1 Methods to assess the nature and size of the illicit tobacco trade 253 ANNEX 4.2 How are the tobacco tax revenues earmarked? 269 CHAPTER 5. Best practices 271 TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) 283 iv v Foreword In 1999, the World Bank’s Curbing the Epidemic was the first report by an interna- tional organization to recognize that increasing tobacco excise taxes was the most effective and cost-effective measure to reduce tobacco use and save lives. Over the two decades since, the evidence base supporting this claim, especially in low- and middle-income countries, has been steadily growing. Meanwhile, the credibility of the tobacco industry’s arguments against tobacco taxation has been slowly waning. In short, health-promoting tobacco taxation has come of age, and the evidence has consistently shown that it is a win for public health, a win for revenue and a win for the economy overall. But we must be cautioned against complacency. Although the evidence on tobacco taxation is irrefutable and there are now signs that the tide is turning on the global tobacco epidemic, tobacco taxation was, in 2018, the WHO MPOWER1 measure that was least implemented at the highest level of achievement. Even more concerningly, cigarettes have become more, rather than less, affordable in many low- and middle-income countries over the past decade. Many countries set rates at insufficient levels and increase them too infrequently, while others still use complex and inefficient taxation structures. This failure to advance tobacco taxation able to effect significant price increases constitutes a loss for governments in revenues, a loss for public health and a win for the tobacco industry. To overcome this inertia, this manual charts the way forward for policy-makers, finance officials and others involved in tobacco tax policy development. It equips them with the information and evidence needed for the realization of their coun- tries’ tobacco tax policy objectives. It also analyses the tobacco industry’s tactics for influencing the political economy of tobacco taxation and shows the limitations and exaggerations of the arguments used against tax increases. The manual serves as an update of the 2010 WHO technical manual on tobacco tax administration by adding new evidence on the successes of tobacco taxation in all parts of the world and broadening its scope to capture more material relevant to developing and implementing more effective tobacco tax policy. 1 The WHO MPOWER package of technical measures and resources that comprises (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco. vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Its contributions are particularly timely in a COVID-19-stricken world. As the pandemic has been worsened by the global burden of noncommunicable disease, and revenue is now desperately needed, taxing tobacco should be more palatable than ever. This manual shows policy-makers how to seize this unique opportunity to use tobacco taxation to build back better, save lives and strengthen health systems while increasing revenue. Dr Naoko Yamamoto Assistant Director-General UHC/Healthier Populations Division vii Acknowledgements This manual was developed under the direction of Jeremias Paul Jr and Anne-Marie Perucic. The following contributed to the content of the manual (in alphabetical order): • WHO: Evan Blecher, Annerie Bouw, Mark Goodchild, Roberto Iglesias, Juliette McHardy, Jeremias Paul Jr, Anne-Marie Perucic and Robert Totanes. • External authors: Mauricio Cardenas (Columbia University), Sophia Delipalla (University of Macedonia), Luk Joossens (tobacco control expert, Belgium), Marin Kurti (Eastern Connecticut State University), Enrique Fanta (former World Bank senior specialist), David Merriman (University of Illinois at Chicago) and Jean Tesche (University of Cape Town). WHO would like to thank the following reviewers for their invaluable comments (in alphabetical order): • External: Jo Birckmayer (Bloomberg Philanthropies), Adriana Blanco Mar- quizo (WHO Framework Convention on Tobacco Control), Frank Chaloupka (University of Illinois at Chicago), Yoni Dekker (WHO Framework Conven- tion on Tobacco Control), Jeffrey Drope (University of Illinois at Chicago), Ceren Ozer (World Bank), Corne van Walbeek (University of Cape Town), Chonlathan Visaruthvong (Ministry of Finance, Thailand) and Rodrigo Santos Feijo (WHO Framework Convention on Tobacco Control). • WHO: – Headquarters: Douglas Bettcher, Itziar Belausteguigoitia, Ranti Fayokun, Joseph Kutzin, Benn McGrady, Vinayak Prasad and Susan Sparkes. – Regional offices: Nina Dela Cruz (WPRO), Fatimah El-Awa (EMRO), Charles Frasier (EMRO), Lee Lily Joung-Eun (WPRO), Jagdish Kaur (SEARO), Elizaveta Lebedeva (EURO), Maxime Roche (AMRO) and Rosa Sandoval (AMRO). WHO would also like to thank Amal Amoune-Naal for the administrative support, Alison Goldstein for the technical editing and Janet DeLand for the copy-editing. Production of this document has been supported by a grant from Bloomberg Phi- lanthropies. The contents of this document are the sole responsibility of WHO and should not be regarded as reflecting the position of Bloomberg Philanthropies. viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ACRONYMS AEO authorized economic operator AFRO WHO Regional Office for Africa AMRO WHO Regional Office for the Americas ATO Australian Taxation Office BAT British American Tobacco CCTV closed-circuit television CIF cost, insurance and freight COP Conference of the Parties CPI consumer price index CVA Customs Valuation Agreement (Thailand) DGCE Directorate General of Customs and Excise (Indonesia) DIY do-it-yourself ECBA World Bank Extended Cost-Benefit Analysis EIU Economist Intelligence Unit EMRO WHO Regional Office for the Eastern Mediterranean ENDS electronic nicotine delivery systems ENNDS electronic non-nicotine delivery systems EU European Union EURO WHO Regional Office for Europe FDA Food and Drug Administration (United States) FET fair and equitable treatment GCC Cooperation Council for the Arab States of the Gulf GDP gross domestic product HTP heated tobacco product HMRC Her Majesty’s Revenue and Customs (United Kingdom) IARC International Agency for Research on Cancer IIA international investment agreement IMF International Monetary Fund IRS Internal Revenue Service (United States) ISO International Organization for Standardization IT information technology ITC International Tobacco Control JTI Japan Tobacco International KRA Kenya Revenue Authority LMICs low- and middle-income countries ACRONYMS ix MFN most favoured nation MOP Meeting of the Parties (to the Protocol) MPOWER (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco NCDs noncommunicable diseases NCI National Cancer Institute NT national treatment OECD Organisation for Economic Co-operation and Development OST other smoking tobacco PMI Philip Morris International PPP purchasing power parity QR quick response RGTE WHO Report on the global tobacco epidemic RYO roll-your-own SACU Southern African Customs Union SCARE (S) smuggling and illicit trade; (C) court and legal challenges; (A) anti-poor rhetoric; (R) revenue reduction; and (E) employment impact SDGs Sustainable Development Goals SEARO WHO South-East Asia Regional Office SII Internal Revenue Service of Chile TADAT Tax Administration Diagnostic Assessment Tool TTC transnational tobacco company UAE United Arab Emirates UHC universal health coverage VAT value added tax WAEMU West African Economic and Monetary Union WCO World Customs Organization WHO World Health Organization WHO FCTC WHO Framework Convention on Tobacco Control WHO ISPT WHO interactive smoking projection and target-setting tool WHO TaXSiM WHO tobacco tax simulation model WPRO WHO Regional Office for the Western Pacific WTO World Trade Organization x W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N xi Executive summary This WHO technical manual on tobacco tax policy and administration builds upon the 2010 WHO technical manual on tobacco tax administration by further detailing the strategies for effective tobacco tax policy development, design, implementation and administration. This 2021 edition also serves as an update to the 2010 manual, incorporating the latest developments in science, technology and policy, as well as providing illustrative recent examples from a variety of countries. The best practices laid out in this manual are designed to inform governments on the development of their tobacco taxation policy, facilitating the achievement of their health and revenue objectives while also supporting their overall development strategy. Tobacco taxes have long been seen as a source of revenue for governments, but as evidence of the harms caused by tobacco has accumulated over the years, public perception has evolved. Increasingly, governments, as well as the general public, are recognizing that taxation of tobacco is not only a revenue source but also an effective public health intervention to reduce tobacco consumption and its associated harms. The profile of tobacco taxation as a health policy tool has increased greatly since the publication of the 2010 WHO technical manual on tobacco tax administration. Multiple global commitments have been adopted over the past decade to address tobacco use specifically – as well as noncommunicable diseases (NCDs) and the Sustainable Development Goals (SDGs) more broadly – through tax and price measures to reduce demand for tobacco products, save lives and fund develop- ment. Global development institutions, including the World Bank, the International Monetary Fund (IMF) and major philanthropic foundations, also agree with WHO on the importance of emphasizing and strengthening tobacco taxation as a key health policy tool. The COVID-19 pandemic has further fuelled this shift in the narrative on tobacco taxation by revealing how the global economy is inextricably linked with population and planetary health. Investing in health is fundamental to any economic recovery, and fiscal policy will be a key driver in addressing the socioeconomic consequences of COVID-19. Interventions such as tobacco taxation – which leads to reduced tobacco consumption, improved population health and increased revenues for governments – should be part of a comprehensive strategy for a build back better recovery. The evidence is clear: significant increases in excise taxes that lead to price increases have consistently proven to be the most effective, as well as the most xii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cost-effective, mechanism for reducing tobacco consumption. This manual will guide readers through the necessary steps to create and implement the strongest tobacco taxation policies for their specific countries. There are many factors to consider when developing tobacco taxation policy. Understanding the market is a fundamental step before deciding which form of taxation to use. Specifically, the choice between ad valorem and specific taxation is influenced by the market structure in a given country. At the same time, tax structure also shapes the market structure. Over the past decade, there has been a trend towards countries adopting specific excise taxes or mixed excise systems that rely more heavily on the specific component, which the latest global data associate with the highest average prices. Tobacco tax structures can be simple, with one flat rate across the board, or com- plex, with multiple tiers for products with different characteristics. In 2018, 31 coun- tries used complex, multitiered structures. But evidence demonstrates that simpler excise tax structures – utilized in all high-income countries – leave the least room for industry manipulation or tax avoidance and brand/product switching by consumers. Not only is it important to set taxes at a high level to discourage consumption, specific excise tax policies must include regular adjustments to increase the tax rate so that it keeps up with inflation and income growth in a country over time. Excise tax increases should aim to reduce the affordability of tobacco products. The base on which the tax is applied is also important. For specific taxation, the tax base should be the quantity in clearly defined units. For ad valorem (or mixed) taxation, the best practice is to use the retail price as the tax base and introduce a minimum excise tax. With regard to non-tax regulations that affect the price of tobacco products, pric- ing regulation may be considered to prevent the tobacco industry from exercising differential tax shifting, which it uses to ensure that large price gaps exist between premium and cheap cigarettes. However, pricing policies cannot be used alone. If con- sidered, they should be used only as complements to significant excise tax increases. Other non-tax regulations include banning promotional discounts for tobacco products and banning the sale of single cigarettes. To assuage concerns that tax increases will increase inflation – as well as to reflect the declining trend in consump- tion of tobacco products – it is good practice to exclude tobacco products from the basket of items that are used to develop consumer price indexes. Finally, in order to make excise tax on tobacco products more effective in reducing overall tobacco use, all tobacco products must be taxed in a comparable way. Regular assessment, evaluation and monitoring of the impact of tobacco tax policies over time are essential components of effective tax policy development and analysis. Governments need to have accurate estimates of price, income and tax base elasticities in order to anticipate the impact of a tax increase on consumption and EXECUTIVE SUMMARY xiii tax revenue. Ideally, other factors such as non-price policies should also be taken into account when estimating price and income elasticities for a specific country. A variety of tools and indicators exist to measure impact and monitor progress, and these are described in Chapter 2 of this manual. When developing tobacco tax policy, it is also important to take the broader policy context into consideration at both the domestic and the regional level. Domestically, cooperation is needed across sectors to ensure that policies and interventions in the areas of agriculture, trade, finance and labour do not work against the public health objectives of tobacco control and taxation. For countries that are part of a regional bloc, regional harmonization of tobacco taxation is a useful tool to prevent tax revenue erosion, tax avoidance and tax evasion, as well as to protect population health. Tax harmonization must be designed carefully, however, to be effective. The experience of the European Union (EU) demonstrates that both a declining consumption trend and stable revenues can be achieved with harmonized minimum excise tax rates. Discussions of policy development and implementation for new and emerging nicotine and tobacco products such as heated tobacco products (HTPs) and electronic nicotine and non-nicotine delivery systems (ENDS/ENNDS), are complicated, by their constantly changing technology and market dynamics. Policies and regulations need to be developed carefully and adjusted accordingly. Where HTPs are not banned, the current recommendation is to tax them at the same level as cigarettes on a per-unit basis, regardless of tobacco content. Early evidence from the United States shows that demand for e-cigarettes, a subcategory of ENDS/ENNDS products, is possibly even more price-responsive than the demand for conventional cigarettes, meaning that taxes can be used as an effective deterrent to ENDS/ENNDS products use. While there is preliminary evidence of substitutability between conventional cigarette use and e-cigarette use, further research is needed to understand substitutability effects among users of both conventional cigarettes and ENDS/ENNDS products. It is essential to imple- ment regulation of ENDS/ENNDS products along with any tax policy to safeguard public health. In countries where they are not banned, ENDS/ENNDS products must be regu- lated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS taxation. Nicotine-containing and non-nicotine-containing e-liquids should be taxed equally. Ultimately, while the policy implications of these newer products require careful consideration, the fact remains that conventional tobacco products constitute the overwhelming share of consumption (more than 97% in 2018). Tobacco tax administration must be both efficient and effective to ensure that health objectives are met and the desired level of tax revenue is raised. Since the xiv W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation of tobacco taxation often involves numerous agencies within a country, clearly defined roles and responsibilities are essential to maximize efficiency. Coordination among the different agencies involved, as well as with neighbouring countries, is required for tobacco tax administration to be effective. Performance evaluation and accountability for competent authorities is also necessary, and many tools and indicators exist to facilitate these processes (described in Chapter 3). There are a number of steps authorities should take to ensure efficiency and effectiveness at each stage throughout the tax compliance cycle (see Chapter 3, section 3.3). Control and enforcement are the main functions of tax administration, and these can best be achieved through the use of a strategic plan and a risk-based approach. Controls can be exercised through licensing and due diligence, fiscal markings (e.g. tax stamps), tracking and tracing, implementation of anti-forestalling measures, national audits and specific controls for imports and exports, as well as for free zones and transhipment points. Once smuggling or illicit trade is detected, actions such as seizing and destroying smuggled and/or illicit tobacco and col- lecting due taxes must be taken immediately. To deter further illegal activities, a comprehensive audit must also be carried out, including all those involved in the illicit acts. Penalties and sanctions must be sufficient to deter illegal activities. The Protocol to Eliminate Illicit Trade in Tobacco Products provides invaluable guidance for tobacco tax administration, control and enforcement that is applicable even for countries that are not Parties to it. The broader elements of a good tax system include proper resourcing of competent authorities, strict rules and regulations to detect and punish corruption and a strong judiciary system capable of resolving disputes as soon as possible. In its efforts to oppose tobacco tax increases, the tobacco industry utilizes many SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric, (R) revenue reduction and (E) employment impact – to influence the political economy of tobacco. Chapter 4 provides detailed analyses of these issues with supporting evidence that belies the SCARE tactics, as well as guidance for tax and other relevant authorities on how to anticipate and respond to industry arguments. This manual also provides tools and methodologies to help tax authorities define and evaluate the problem of illicit trade of tobacco products in their countries, inde- pendent of the tobacco industry’s generally inflated estimates. Price (and tax) levels are not a key determinant of illicit trade; rather, the problem is exacerbated by the lack of governance and tax administration capacity. Refraining from increasing taxes is not the solution. Countries should instead respond with a comprehensive strategy to fight illicit trade, including undertaking independent estimates of illicit trade levels and implementing good tax administration practices such as those discussed in Chapter 3 and contained in the WHO Framework Convention on Tobacco Con- trol (WHO FCTC) Protocol to Eliminate the Illicit Trade in Tobacco Products. When it comes to court challenges, the tobacco industry is less likely to challenge excise taxes than other tobacco control measures, because taxation is a comparatively well-established regulatory measure. The industry will, however, exploit the slightest vulnerability in the design, adoption or implementation of tax measures. For this reason, measures to strengthen regulators’ legal position are described that will enable authorities to protect themselves from potential legal challenges. The industry argument of regressivity, or the notion that tobacco tax increases hurt the poor because they have to pay a larger share of their income in taxes than the rich, has two fundamental limitations. First, the notion of regressivity does not take into consideration the broader health and economic harms caused by tobacco use that exacerbate the impoverishment of lower-income smokers. These harms are actually reduced when tobacco consumption decreases following a tax increase. Second, the tobacco industry argument ignores the fact that higher tobacco taxes and prices can induce behaviour change as is reflected in the price elasticity of demand. Evidence consistently shows that lower-income smokers are more sensitive to price and therefore more likely to reduce smoking in response to a tax and price increase. Including these factors shows tobacco taxation to be, in fact, a progressive public health intervention that disproportionately benefits the poor. While essentially admitting that a tobacco tax increase may have the desired effect of reducing consumption, the industry also tries to argue that a tax increase will also reduce revenues. In fact, the price inelastic demand for tobacco makes tobacco tax increases a win-win for both public health and finance. This manual presents several country examples that demonstrate how well-designed and well-implemented tobacco tax increases lead to increases rather than decreases in revenue in the short to medium term. In addition, the reduced consumption resulting from a tax increase results in reductions of other tobacco-related government expenditures as well. The final tactic used by the tobacco industry to challenge proposed tax increases is to frame tobacco taxes as an economic rather than a public health issue. This false choice between health and jobs is based on faulty assumptions that 1) tobacco is a significant source of domestic employment; 2) job creation relies on tobacco consumption and 3) tobacco-related livelihoods are prosperous, sustainable and irreplaceable. Earmarking can be a useful tool for improving the political economy of tobacco tax increases. While the primary goal of tobacco tax increases is to reduce demand for tobacco, setting aside portions of tax revenue to fund other tobacco control efforts or relevant health programmes can help convince the public, politicians and officials of the value of significant tobacco tax increases. Earmarking can also be EXECUTIVE SUMMARY xv x vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N used to counter tobacco industry arguments about potential negative effects of tax increases – for example, by supporting tobacco farmers in transitioning to other crops. Tobacco taxes work. This is why the industry invests so much money and effort in blocking large tax increases and other effective tax policy reforms. Policy-makers must not be swayed by industry pressure but need only to follow the facts. This manual provides all the information policy-makers need to make the right deci- sions at each step of the process – from designing, evaluating, implementing and administering tax policy to refuting specious industry attacks and communicating the value of tobacco taxation to legislators and the broader population. An effectively designed and efficiently administered tobacco tax policy will not only produce the direct results of reducing tobacco consumption among smokers and raising revenue for governments, its effects will be felt much more broadly. Indeed, raising tobacco taxes is a SMART policy: it Saves lives; Mobilizes resources; Addresses health inequities; Reduces burdens on health systems; and Targets tobacco use, a major risk factor for NCDs. • Saves lives: Tobacco use is the leading cause of preventable deaths globally – it claims 8 million lives each year. Tobacco taxation is the most effective mechanism for reducing tobacco consumption and its associated health burden worldwide. • Mobilizes resources: Despite being the single most effective tobacco control measure, tobacco taxation is largely underutilized as a policy mechanism. Based on available data on the price and taxation of cigarettes, it is estimated that excise taxes on cigarettes generated a worldwide total of US$ 361 billion in revenues in 2018, including US$ 162 billion in revenues for low- and middle- income countries (LMICs). If all countries were to raise cigarette excise tax rates by the equivalent of US$ 1 per pack, the amount of excise revenue from cigarettes would increase by US$ 178–219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from such tax increases, with excise revenues increasing by 82–103%, providing governments in these countries with an extra US$ 133–167 billion. This shows the substantial revenue potential of tobacco taxes. • Addresses health inequities: Tobacco taxation and tax increases are effectively progressive or pro-poor policies because of their positive distributional impact. Lower-income smokers benefit disproportionately from reduced tobacco consumption and use in terms of health gains and income retention. • Reduces burdens on health systems: The worldwide economic cost of tobacco use was US$ 1.4 trillion in 2012. Tobacco taxes reduce tobacco-related burdens on governments and health systems through population-based preventive measures. • Targets tobacco use: Tobacco taxation directly targets and reduces tobacco use, which is a major risk factor for several deadly NCDs. In summary, significant tobacco tax increases, designed and implemented according to the latest guidance and best practices presented in this technical manual – and as a strong component of a comprehensive tobacco control strategy – will bring about substantial reductions in tobacco use and the health and economic harms it causes. EXECUTIVE SUMMARY xvii x viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 1 CHAPTER 1. Why this manual? BACKGROUND Tobacco taxes are not new. Governments around the world have been applying taxes on tobacco and tobacco products practically since the idea of excise was conceived. And rightly so: tobacco is not a necessity, it is easy to tax and the demand for it is relatively inelastic. These characteristics, along with the substantial revenues tobacco taxes generate, have made tobacco a highly appropriate object of taxation. As evidence of the harms of tobacco has accumulated over the years, the public perception of tobacco taxes has evolved. Now tobacco taxes are not only seen as a revenue source, but, more importantly, they are recognized as an effective public health intervention to reduce tobacco consumption. This trend reflects the reasons excise taxes exist in the first place – to discourage harmful behaviour and to mitigate the associated negative externalities (1–2). Many governments view tobacco taxes as a significant and stable source of rev- enue, which may explain why there is often a degree of hesitation whenever tobacco tax reform is proposed. Historically, many governments have relied on revenues from tobacco taxes and have even adjusted the level of taxation according to their revenue needs (3). However, some countries are beginning to recognize the value of applying high tobacco taxes primarily as a public health tool, viewing revenues as a secondary consideration (4). Arguments against tobacco tax hikes or improvements to the tax structure are often economic in nature: such tax changes will allegedly decrease revenues, wipe out jobs, increase illicit trade and harm local industries, among other claims. But the evidence has consistently shown that such claims are simply not true in an overwhelming majority of situations. The tobacco industry, in particular, frequently portrays this conflict as a false dichotomy between public health and the economy – as if prioritizing health comes at the expense of the economy. In fact, studies and real-world experiences have shown that increasing tobacco taxes not only improves public health but also has a net positive impact on the economy and development of a country – a true win-win scenario (5–6). As an update to the first WHO technical manual on tobacco tax administration published in 2010, this manual aims to help readers better navigate the various 2 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N issues surrounding tobacco taxes and their implementation. The primary intended audience includes policy-makers, finance officials, tax authorities, customs officials and other relevant persons/bodies involved in the formulation and implementation of tobacco tax policy. The manual provides a detailed guide to the design of tobacco tax policy and describes how to effectively administer these taxes to maximize impact. Detailed discussions of the political economy considerations and the hurdles that need to be overcome before and during implementation are included as well. The overarching goal is to equip those working in the tax policy and implementation spheres with sufficient information to help realize the health and revenue objectives of a government’s tobacco tax policy in line with its overall development strategy. THE CONSEQUENCES OF TOBACCO USE AND THE NEED FOR INTERVENTION Most people are aware that smoking and tobacco use are harmful to health, but few truly comprehend the scale of this harm. The tobacco epidemic claimed more than 100 million lives in the last century (7), with updated estimates now reaching 8 million deaths annually from tobacco use and exposure to second-hand smoke (8). As much as 80% of these deaths occur in low- and middle-income countries (LMICs) (6), revealing how the developing world carries much of the global burden. Tobacco use is a major risk factor for many chronic conditions, including heart disease, cancer, diabetes and chronic lung disease – collectively known as noncom- municable diseases (NCDs). NCDs account for about 15 million premature deaths (between ages 30 and 69) worldwide, killing people in their most productive years. As the leading cause of preventable deaths, tobacco use remains one of the foremost public health challenges of our time. The consequences of tobacco use also present enormous economic, development and social costs that wreak havoc on families, communities and societies. The annual economic cost of smoking was estimated at US$ 1.4 trillion in 2012, equivalent to 1.8% of the global gross domestic product (GDP) (9). With these figures likely to have increased since then, the massive health and economic burdens of tobacco use provide justification for governments to intervene and strictly regulate the market for tobacco products. The purview of tobacco control extends beyond the strong imperative to protect people’s health and well-being; it should also strive to contain the market failures and negative externalities of tobacco use, particularly since these effects can significantly impact a country’s development trajectory. The mounting evidence of the enduring destruction caused by tobacco in the 20th century provided compelling reasons for a strong global response, which led countries to negotiate the World Health Organization Framework Convention on Tobacco Control (WHO FCTC). The WHO FCTC came into force in 2005 as the CHAP T ER 1. WHY T HIS M ANUAL? 3 first public health treaty under the auspices of WHO. To facilitate its implementation at the country level, WHO packaged a set of demand-reduction measures directly taken from the treaty (7). These interventions, collectively known as MPOWER, are as follows: (M) monitoring tobacco use and prevention policies; (P) protecting people from tobacco smoke (smoke-free laws); (O) offering help to quit tobacco use (cessation services); (W) warning about the dangers of tobacco (including graphic pack warnings and plain packaging); (E) enforcing bans on tobacco advertising, promotion and sponsorship; and (R) raising taxes on tobacco products. Specifically, under Article 6 of the WHO FCTC, Parties recognized that price and tax measures are an effective and important means of reducing tobacco consumption for various segments of the population – in particular, among young persons (10). The severity of the tobacco epidemic and its ongoing damage to health and economies are clear justifications for governments to actively intervene and correct market failures. The scale of the burden and the rate at which lives are being destroyed necessitates urgent and aggressive action on tobacco control, using measures that most countries have committed to implementing and that are proven to be effective in reducing tobacco use. WHY TOBACCO (EXCISE) TAXES ARE CRUCIAL Among the different tobacco control interventions, raising excise taxes has been identified as the most effective as well as the most cost-effective measure to reduce consumption (6). While other interventions are certainly important components of a comprehensive tobacco control strategy, the direct impact of significant tax increases on consumption is by far the strongest. On average, a tax increase that causes prices to go up by 10% reduces consumption by 4% in high-income countries and 5% in LMICs (6). When implemented at scale, this demonstrates the enormous power of tobacco taxation and its potential to save lives. Tobacco taxes differ from other interventions in that their impact can increase and build over time – even if taxes are already relatively high, their rates need to be continuously increased to retain and amplify their effectiveness. However, this should not be taken as a suggestion that governments considering tobacco control interventions should focus solely on taxes. Taxes are even more effective when implemented as part of a comprehensive package of measures such as MPOWER, which covers distinct but complementary intervention points. Among the different taxes applied on tobacco products, excise taxes are the most significant because they raise both absolute and relative prices (6). This is important when considering health objectives, since it is the magnitude of the price increase of tobacco products that determines the reduction in consumption. An excise tax is typically applied on a limited set of products, designed to discourage their use by 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N raising the price significantly over that of other products available in the market. This is in contrast to value added taxes (VAT) or sales taxes, which apply to most goods and services. Seeking to raise the prices of tobacco products through VAT or sales taxes would fail to increase relative prices, making this an ineffective and inefficient method. Customs or import duties on tobacco products are also utilized, but their impact is waning with the global trend towards bilateral and regional agreements aimed at trade facilitation. The application of these duties varies across countries, but overall, they are not applicable to locally produced tobacco products. As a tool to increase prices, import duties cannot substitute for excise taxes, since they are not specifically designed to reduce consumption. When viewed as a public health policy tool, tobacco taxation is highly cost-effective, since it delivers significant impact yet is relatively inexpensive to implement (11). The costs of implementing tobacco taxation are much lower than those of clinical NCD interventions such as cancer treatments or maintenance medications, since the commodity and human capital requirements are less substantial (12). Moreover, increasing tobacco taxes actually generates additional revenue for a government. Tobacco taxes are also very effective in pre-empting or reducing consumption among groups of people who are especially price-sensitive – youth in particular, who are prevented from initiating a lifelong addiction if taxes and prices are sufficiently high (13). This is also true for the poor, who are more prone to catastrophic health expenditures than the wealthy are. Preventing initiation or encouraging cessation by imposing high taxes provides an escape route from the vicious cycle of tobacco use and poverty (6). A DECADE OF PROGRESS AND COMMITMENT TO ACTION Since 2010, when the first WHO technical manual on tobacco tax administration was published, numerous developments have raised the profile of tobacco taxes as an essential public health intervention. The Conference of the Parties (COP) to the WHO FCTC adopted guidelines for implementation of Article 6 of the treaty, which focuses on price and tax measures to reduce the demand for tobacco. Also within this period, three high-level meetings on the prevention and control of NCDs by the United Nations General Assembly, as well as the endorsement of the Global NCD Action Plan in 2013 by the World Health Assembly, have resulted in strong global commitments to implement measures, such as increased tobacco taxes to protect people’s health. The 2030 Agenda for Sustainable Development, which contains 17 goals known as the Sustainable Development Goals (SDGs), describes the global development strategy for the next decade. Within the SDGs, two specified targets are highly relevant for tobacco control: strengthening the implementation of the WHO FCTC (target 3.a) CHAP T ER 1. WHY T HIS M ANUAL? 5 and reducing premature mortality from NCDs by 30% (target 3.4). Furthermore, the Addis Ababa Action Agenda1, which aims to provide a global framework for financing the SDGs, also highlights tax and price measures on tobacco as key mechanisms to reduce demand and save lives while increasing domestic resources for develop- ment. Another important milestone was the 2018 entry into force of the Protocol to Eliminate Illicit Trade in Tobacco Products. These key events, along with several outcome documents and policy declarations in the area of tobacco control and the wider development sphere, have introduced tobacco taxation into the consciousness of a much larger share of policy-makers. As detailed in subsequent chapters, numerous countries have imposed sufficiently high tobacco tax rates while applying best practices in tax policy design and imple- mentation over the past decade (8, 10). For example, sustained and substantial tax increases have reduced tobacco use in LMICs such as Brazil (14), Turkey (15) and the Philippines (16). High-income countries also continued their leadership in this area, comprising 23 of the 38 countries judged to have sufficiently high tobacco taxes in 2018 (8). However, much remains to be done. The 2019 WHO report on the global tobacco epidemic (RGTE) shows that tobacco taxes are still the most underutilized tobacco control policy among the MPOWER measures (8), with only 14% of the world’s population being covered by sufficiently high tobacco taxes. Substantial progress has also been made in building the tools and evidence base for tobacco taxation. Volume 14 of the International Agency for Research on Cancer (IARC) handbooks of cancer prevention, Effectiveness of tax and price policies for tobacco control, published in 2011, is a key review of the literature published as of May 2010 on the effectiveness of tax and price policies in reducing tobacco use. The National Cancer Institute (NCI)-WHO Monograph on the economics of tobacco and tobacco control, published in 2016, details the evidence accumulated over the years from various countries, focusing not only on tax and price policies, but on all aspects of the economics of tobacco and tobacco control. In addition, numerous published studies from LMICs provide a comprehensive picture of the impact of tobacco taxation in different contexts. The updated Appendix 3 of the Global NCD Action Plan explains the cost-effectiveness of tobacco taxation (11), while the Global NCD Business Plan, Saving lives, spending less, built on this work by estimating a dollar figure for the return on investment expected from implementing the best-buy interventions for tobacco control, including taxation (12). The past decade has seen major steps forward for tobacco taxation in terms of global commitments, the number of countries implementing best practices and the 1 The Addis Ababa Action Agenda of the Third Conference on Financing for Development. Third Inter- national Conference, 13-16 July 2015, Addis Ababa, Ethiopia (https://sustainabledevelopment.un.org/ content/documents/2051AAAA_Outcome.pdf, accessed 17 February 2021). 6 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N expansion of the evidence base on tobacco taxes, particularly in LMICs. Within the development sphere, institutions such as the World Bank, the International Monetary Fund (IMF) and many other multilateral agencies are aligned with WHO on the importance of tobacco taxation and the need to improve its implementation (17–18). Considerable challenges remain; although it appears that the world is headed in the right direction, progress needs to accelerate at a much quicker pace in order to achieve the SDG targets by 2030. SHAPING A “NEW NORMAL” FOR TOBACCO TAXATION The global upheaval caused by the COVID-19 pandemic has cast an unprecedented spotlight on how well governments around the world prepared for and responded to the crisis. It has exposed glaring health systems vulnerabilities and highlighted the struggles of many countries to control the spread of the virus. But perhaps more than anything, the pandemic has demonstrated how the economy, trade, science, politics and many other aspects of our societies are very much interdependent and interconnected with the health of the population. It is clear that an individual’s state of health can significantly determine their susceptibility to disease and their ability to overcome it. People with NCDs are more vulnerable to becoming severely ill with a number of conditions, which also appears to be the case with COVID-19 (19). Tobacco use is a major risk factor for NCDs, and available research suggests that smokers are at higher risk of developing severe illness and dying from COVID-19 (20). Just as the different aspects of society are interconnected, so too are people’s health, the existence of health-promoting environments and the government policies and agencies that shape these environ- ments. This critical moment presents a unique opportunity and renewed motivation to discourage the use of harmful products such as tobacco and to further improve tobacco control measures, especially tobacco tax policy. Moving forward, a business-as-usual approach to tobacco taxation will not be sufficient. Responding to this new reality and preparing for the next pandemic entails implementing measures that promote healthier populations. Like the COVID-19 pandemic, any future pandemic will likely exacerbate health inequities, bring about more economic uncertainty and put pressure on governments’ fiscal capacities. Interventions such as higher tobacco taxes, which protect people’s health while generating more revenues and economic benefits, become even more important in such crisis situations. Given this context and the stakes involved, ministries of finance and tax authorities are in a unique and powerful position – one of saving not only livelihoods, but also lives. The importance of increasing tobacco taxes – one of the most effective public health tools available – cannot be overstated. The traditional approach of treating CHAP T ER 1. WHY T HIS M ANUAL? 7 tobacco tax exclusively as a revenue source has no place in the new normal. One cannot deny the scale of the tobacco epidemic, the necessity to correct market failures and the overwhelming evidence of tobacco taxation’s benefits to health and to the economy. The positive trend in the changing narrative around tobacco taxation needs to continue. Tobacco taxation should not be viewed in isolation from the rest of government policies, but rather as an important part of the whole, an essential piece in working towards our common goal of better health for all. OVERVIEW OF SUBSTANTIVE CHAPTERS This manual is primarily designed for policy-makers, finance officials, tax authorities and customs officials. It may also be useful for officials within health ministries or other government agencies, as well as nongovernmental organizations working in this area, including tobacco control advocates. Significant effort is made to present real-world examples and recent experiences from a wide range of countries to demonstrate success stories and lessons learned in raising tobacco taxes. A sub- stantial amount of evidence has been generated in LMICs over the past few years that supports and augments the existing evidence base, providing a much broader body of knowledge than was available when the first WHO technical manual on tobacco tax administration was released. Chapter 2 delves into the theory, practice and empirical evidence on tobacco excise tax policy, including current global trends. The chapter offers a detailed analysis of the various elements that constitute tax structure, aiming to provide policy-makers with a comprehensive understanding of the factors affecting prices, consumption and the market. It describes the key components to keep in mind when designing tobacco tax policy to maximize the impact of tax increases and improve the tax structure. The chapter also includes updated global price and tax data, specific examples from various countries and a discussion of tax base elasticity, automatic excise tax adjustments and pricing regulations, as well as descriptions of new and emerging nicotine and tobacco products, including electronic nicotine- and non- nicotine delivery systems (ENDS/ENNDS) and heated tobacco products (HTPs). Chapter 3 focuses on tobacco tax administration. It provides an in-depth discus- sion of the fundamental components that make tobacco tax collection effective and efficient, ensuring achievement of the health and revenue objectives of tax policy. It highlights the importance of cooperation among the various agencies involved in the implementation of tobacco taxes within countries and across borders. Building on country and regional experiences from previous decades, the chapter outlines specific measures and recommendations to maintain oversight of the whole tax compliance cycle. Also included are actions to facilitate control and enforcement, such as licensing, fiscal markings (e.g. tax stamps), tracking and tracing systems 8 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N and import and export controls. Finally, the discussion pivots towards the broader elements of a good tax system, such as proper resourcing of competent authorities, a strong judiciary and strict rules regarding corruption. Chapter 4 deals with the important challenges in the area of political economy when countries attempt to increase tobacco taxes or simplify the tax structure. The tobacco industry often relies on identified patterns of argumentation and tactics to obstruct such reforms, i.e. SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric (regressivity), (R) revenue reduc- tion and (E) employment impact – each of which is discussed thoroughly in this chapter. Also included is a detailed discussion of the measurement of illicit trade and a discussion of earmarking tobacco tax revenue for health purposes. Finally, Chapter 5 presents a comprehensive list of the best practices in tobacco tax policy and administration discussed throughout this manual. The list is intended to serve as a practical guide and quick reference to the salient points presented. CHAP T ER 1. WHY T HIS M ANUAL? 9 REFERENCES 1. Ranson K, Jha P, Chaloupka FJ, Nguyen SN. The effectiveness and cost-effectiveness of price and other tobacco control policies. In: Jha P, Chaloupka FJ, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000:427–447 (https://www.paho.org/hq/dmdocuments/2010/ Cost-effectiveness%20of%20price%20increases.pdf, accessed 4 February 2021). 2. Effectiveness of tax and price policies for tobacco control. Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications.iarc.fr/Book-And-Report-Series/Iarc- Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And-Price-Policies-For-Tobacco-Control-2011, accessed 4 February 2021. 3. Tobacco taxation in the United States. In: Lynch BS, Bonnie RJ, editors. Institute of Medicine (US) Committee on Preventing Nicotine Addiction in Children and Youths. Growing up tobacco free: preventing nicotine addiction in children and youths. Washington (DC): National Academies Press (US); 1994 (https://www.ncbi.nlm.nih.gov/books/NBK236771/, accessed 10 November 2020). 4. Sin tax reform. Manila: Department of Finance (Philippines); 2012 (https://www.dof.gov.ph/advocacies/ sin-tax-reform/, accessed 10 November 2020). 5. Goodchild M, Perucic AM, Nargis N. Modelling the impact of raising tobacco taxes on public health and finance. Bull World Health Organ. 2016; 94:250–7 (https://www.who.int/bulletin/ volumes/94/4/15-164707.pdf, accessed 4 February 2021). 6. The economics of tobacco and tobacco control. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute, NIH Publication No. 16-CA-8029A; 2016 (National Cancer Institute tobacco control monograph 21; https://cancercontrol.cancer.gov/ brp/tcrb/monographs/monograph-21, accessed 17 December 2020). 7. WHO report on the global tobacco epidemic, 2008: the MPOWER package. Geneva: World Health Organization; 2008 (https://www.who.int/tobacco/mpower/mpower_report_full_2008.pdf, accessed 10 November 2020). 8. WHO report on the global tobacco epidemic, 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/tobacco/global_report/en/, accessed 10 November 2020). 9. Goodchild M, Nargis N, Tursan d’Espaignet E. Global economic cost of smoking-attributable diseases. Tob Control. 2018;27:58–64 (https://tobaccocontrol.bmj.com/content/27/1/58, accessed 4 February 2021). 10. Guidelines for implementation of Article 6 of the WHO FCTC. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 4 February 2021). 11. Tackling NCDs: ‘best buys’ and other recommended interventions for the prevention and control of noncommunicable diseases. Geneva: World Health Organization; 2017 (https://apps.who.int/iris/ handle/10665/259232, accessed 10 November 2020). 12. Saving lives, spending less: a strategic response to noncommunicable diseases. Geneva: World Health Organization; 2018 (https://www.who.int/publications/i/item/WHO-NMH-NVI-18.8, accessed 4 February 2021). 13. Chaloupka FJ, Warner KE. The economics of smoking. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics. Elsevier; 2000;1(1):1539–1627. 14. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Rev Panam Salud Publica. 2016;40(4):243–9 (https://iris.paho.org/bitstream/ handle/10665.2/31306/v40n4a09_243-9.pdf?sequence=1&isAllowed=y, accessed 17 February 2021). 15. Cetinkaya V, Marquez PV. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 16. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington (DC): World Bank Group; 2016 (http://documents.worldbank.org/curated/en/638391468480878595/pdf/106777-PUB-PUBLIC- PUBDATE-7-26-2016.pdf, accessed 10 November 2020). 17. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 10 November 2020). 10 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 18. Irwin A, Marquez P, Jha P, Peto R, Moreno-Dodson B, Goodchild M, et al. Tobacco tax reform: at the crossroads of health and development – a multisectoral perspective. Washington (DC): World Bank Group; (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report- CrossRoads.pdf, accessed 2 February 2021). 19. Information note: COVID-19 and NCDs. Geneva: World Health Organization; 23 March 2020 (https:// www.who.int/publications/m/item/covid-19-and-ncds, accessed 10 November 2020). 20. WHO statement: tobacco use and COVID-19. Geneva: World Health Organization; 11 May 2020 (https:// www.who.int/news-room/detail/11-05-2020-who-statement-tobacco-use-and-covid-19, accessed 10 November 2020). 11 CHAPTER 2. Tobacco excise tax policy 2.1. GLOBAL OVERVIEW OF TOBACCO TAX PRACTICES A well-designed tax policy is key to having an effective tax policy. Any government that is planning to reform its tax policy must first understand the fundamental components of a good tax policy, as well as consider the strengths and weaknesses of different approaches to taxation, how they impact price and the requirements for tax administration. Understanding how the tobacco market operates in a country is equally important for policy-makers because of the inevitable interaction between the market and tax structures. Beyond the political considerations that strongly influence tobacco tax policy development, this chapter focuses on the technical aspects of tobacco taxation – excise tax in particular. Section 2.1 provides an overview of tobacco tax practices at the global level, focusing on the different ways countries structure excise tax. Section 2.2 emphasizes the importance of carefully designing excise tax policy, highlighting not only the significance of tax increases but also excise tax structure and its impact on prices, taking into account how market structure influences trends. This section also discusses the importance of measuring impact as another aspect of tax policy development, and it presents the crucial elements for performing measurement, as well as the relevant indicators available to monitor progress. Section 2.3 describes external policy considerations in the design phase to ensure that the goals of tobacco control and taxation are achieved. Intersectoral policy integration and coherence at the domestic level is discussed as a strategy to ensure that policies of other sectors do not inhibit or obstruct public health policy objectives. This section also reviews the current state of regional tax harmonization based on the experience of existing regional blocs and draws conclusions on the best policy approaches to preserve the public health interests of individual countries. Section 2.4 discusses new and emerging nicotine and tobacco products, in par- ticular HTPs and ENDS/ENNDS. It reviews the latest evidence on the health impacts of these products and current approaches to regulation. Key policy considerations are identified, and recommendations are provided for adopting an appropriate excise tax policy for these products. Section 2.5 summarizes the issues covered in the chapter and the key takeaways. 12 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.1.1 TAXES APPLIED ON TOBACCO PRODUCTS Taxes are classified as either direct or indirect. Direct taxes are imposed on the profit, income, property or wealth of persons or companies, whereas indirect taxes are imposed on the price of goods and services. Indirect taxes are most relevant to tobacco products taxation, because they directly influence price. A variety of types of indirect taxes can be applied to tobacco products. These include: • excise taxes – taxes that apply to a few selected commodities (they can also be applied to alcohol, fuel, sugar-sweetened beverages, etc.). • VAT or sales taxes – VAT is a multistage tax on all consumer goods and services that is applied proportionally to the price the consumer pays for a product. It is a tax on the amount by which the value of an article has been increased at each stage of its production or distribution. Some countries impose sales taxes instead of VAT. Unlike VAT, which is collected at every stage of the supply chain, sales taxes are generally levied at the point of retail on the total value of goods and services purchased. Ultimately, the consumer ends up paying the tax, whether it is a VAT or a sales tax. • import duties – taxes on selected goods imported into a country to be consumed in that country (i.e. goods that are not in transit to another country). In general, import duties are collected from the importer at the point of entry into the country. • other taxes – other indirect taxes, such as environmental taxes, that do not fall into any of the categories listed above. One of the most well-established and widely understood points in tax policy is that tobacco products should be subject to excise taxation. The focus of this chapter – and of this manual overall – is on excise taxes. They are the most important type of indirect taxes for tobacco control because they are applied directly to tobacco products and contribute the most to increasing the price of tobacco products relative to other goods and, subsequently, to reducing consumption. There are two basic types of excise taxes: • specific – levied as a monetary value per quantity of the product being taxed (e.g. 1 000 cigarettes, pack of 20 sticks, kilogram of tobacco); and • ad valorem – levied as a percentage of the value (e.g. retail price, or the producer/ex-factory price or the cost, insurance and freight [CIF] value1) of the product being taxed. These types of excise tax can be applied at a uniform or a differential (tiered) rate and on their own or in combination (i.e. a mixed system). 1 CIF is the value of an imported product as declared to customs upon entry into a territory. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 13 Given the widespread use of cigarettes – almost the only tobacco product used in some parts of the world – and the scarce availability of data for other tobacco products, this chapter focuses mainly on cigarettes. But there are a few examples and recommendations for other tobacco products, including those that are more prevalent in specific parts of the world (e.g. bidis or smokeless tobacco in South-East Asia and waterpipe tobacco in the Eastern Mediterranean region). 2.1.2 CIGARETTE TAXES AND RECENT TRENDS WORLDWIDE Tax and national income levels: the higher the income level, the higher the taxes and prices At the global level, cigarette price and tax levels correlate positively with a country’s income level: prices and taxes are higher in higher-income countries and lower as income level decreases. This trend has not changed over the years since 2008. Figure 2.1 presents the levels of price and tax by income groups for 2018, using the World Bank classification of income groups. Fig. 2.1 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by income group, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in purchasing power parity (PPP) adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Excise tax per pack Other taxes Retail price minus taxes Pr ic es a nd ta xa tio n pe r p ac k (P PP $ ) High-income Low-income 4.25 0.68 Middle-income 2.06 Total taxes: 5.30 (67.9% of pack price) Total taxes: 2.91 (58.3% of pack price) Total taxes: 1.18 (38.1% of pack price) PPP $ 7.80 PPP $ 4.99 PPP $ 3.09 Global 2.48 PPP $ 5.53 Total taxes: 3.36 (60.8% of pack price) 0 1 2 3 4 5 6 7 8 14 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tax and price levels: the higher the tax share, the higher the price Globally, cigarette prices correlate positively with tax percentage levels: as the total tax share (of which excise represents the largest part) as a percentage of retail price increases, the price of cigarettes generally also increases (see Fig. 2.2 below).2 This indicates that taxes do influence prices. Fig. 2.2 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by total tax levels, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Tax level and WHO regional classification: taxes and prices are highest in the European region, followed by South-East Asia, the Americas and the Western Pacific, with the lowest levels in the Eastern Mediterranean and African regions At the regional level (WHO regional classification), average levels of prices and taxes vary greatly. The highest level can be seen in the European region, which includes the European Union (EU) countries. The EU’s unified tax structure includes high levels of minimum taxes – which lead to high prices – and encourage member 2 This is a general trend and does not apply for every country; there are countries that have a large tax share but low prices for cigarettes. Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax ≥ 75% Total tax ≤ 25% 0 7 4 1 2 3 6 5 50% ≤ Total tax < 75% 25% ≤ Total tax < 50% Excise tax Other taxes Retail price minus taxes 4.44 2.24 1.57 PPP $ 5.07 PPP $ 5.33 PPP $ 7.07 PPP $ 2.60 0.17 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 15 countries to regularly increase their taxes to meet their obligations. However, with the current minimum level now being reached by all EU member countries, the motivation to increase excise taxes may wane. Indeed, the minimum cigarette excise amount in the EU has not been adjusted since it went into effect on 1 January 2014, and it is suffering from inflation erosion. Member States of the EU acknowledged this in June 2020 by stating that action at the EU level is required to ensure that minimum excise duty rates regain traction to effectively reduce the consumption of tobacco products and that the minimum rates of excise duties on a number of tobacco products would be increased (2). Excise taxes are lowest in the African and Eastern Mediterranean regions. And China – reported separately due to its size – has lower tax rates than the Western Pacific region (see Fig. 2.3). Fig. 2.3 Weighted average retail prices and taxation (excise and total) of most-sold brand of cigarettes, by region, 2018 Notes: China is represented separately from the Western Pacific Regional Office (WPRO) average because of its exceptionally large number of smokers compared with the number in other countries in the region. AFRO is the African Region, AMRO is the Region of the Americas, EMRO is the Eastern Mediterranean Region, EURO is the European Region, SEARO is the South-East Asia Region. Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Retail price Excise tax Total tax share % AFRO AMRO EMRO EURO SEARO WPRO w/o China China All 3. 80 4. 02 4. 24 4. 89 4. 02 1. 45 2. 89 2. 65 2. 48 7. 53 7. 27 5. 61 2. 51 1. 95 1. 05 42.2% 55.8% 61.2% 72.9% 63.2% 55.7% 60.8% 56.9% Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax share % 5. 53 0 1 2 3 4 5 6 7 8 9 10 40 30 50 60 70 80 20 16 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Global tax structures trend: more countries are adopting specific excise taxes or mixed excise systems that rely more on the specific component Since 2008, the number of countries that rely solely on ad valorem taxes or apply no excise taxes at all has decreased as more countries have adopted specific or mixed systems. More of the countries that have implemented a mixed system have increased the specific component of the tax structure relative to the ad valorem component (see Figs. 2.4 and 2.5).3 Fig. 2.4 Changes in excise tax structure, 2008–2018 Fig. 2.5 Changes in reliance on specific versus ad valorem component in mixed systems, 2008–2018 Source: (1). 3 For information about countries that applied each type of excise tax structure in 2018, see Annex 2.1. N um be r o f c ou nt ri es Specic excise Ad valorem Mixed excise No excise 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 54 56 57 63 50 45 24 23 24 21 19 15 49 47 44 42 41 55 57 59 56 60 63 62 Mixed excise Relying more on specic Relying more on ad valorem N um be r o f c ou nt ri es 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 45 22 23 23 24 22 26 27 27 27 32 35 37 50 54 56 57 63 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 17 The imposition of a minimum specific excise tax: half of the countries that have a mixed or ad valorem structure impose an excise tax floor A minimum specific excise tax ensures that at least a certain minimum amount of tax is paid, irrespective of price level. Almost half of the 101 countries that impose either ad valorem or a mixed excise for which data on minimum excise are available (47 countries) set a minimum specific excise tax.4 Nearly two thirds of those that set a minimum specific excise tax (29 countries) are high-income countries; most of them are in the EU, which requires its members to impose a minimum specific excise tax. The choice of tax base worldwide: almost half of the countries that apply a mixed or ad valorem excise system use retail price as the base Setting the base for applying a specific excise is relatively easy: most countries use a defined quantity of sticks for cigarettes, the weight in kilograms for tobacco and the weight in grams for other tobacco products (1). Different bases for ad valorem excises are applied in different countries. Nearly half of the 105 countries that implement either ad valorem or a mixed excise for which data are available (47 countries) use the retail price5 as the tax base for the ad valorem part, and most of those (28 countries) are high-income countries. Using the retail price as the base for the excise ad valorem tax is more effective than using the producer price or the CIF value. Unlike retail prices, which are easy for tax administrators to ascertain by monitoring the market, the producer price or CIF value is prone to undervaluation by producers or importers, who may pass on their margins to related parties further down the supply chain and successfully reduce their tax burden. This tactic is also known as transfer pricing. Additionally, global-level data show that the excise ad valorem on the retail price seems to lead to higher retail prices on average compared with an excise ad valorem applied on other bases, such as the producer price or CIF value (see Fig. 2.8 below). On complex tiered structures: 31 countries still apply complex, multitiered excise taxes on tobacco products As of 2018, 31 countries imposed excise taxes that varied according to defined char- acteristics of cigarettes, including price level, type of production, type of package and length of cigarette (Table 2.1). Some countries use more than one criterion to differen- tiate the tax rates. Indonesia, for example, imposes differential rates based on volume 4 This means that countries with a mixed system impose an overall minimum specific excise tax (where the yield of the specific plus the ad valorem excise cannot be below the set minimum specific excise tax), in addition to the excise on a specific component. 5 Countries that impose ad valorem on retail price exclusive of VAT are also included, since retail prices are easy to determine and VAT rates are known variables. 18 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N produced, type of cigarette and price level. In other countries, such as Member States of the EU, differential rates for cigarettes are prohibited by law, and the rate of the ad va- lorem tax and the amount of specific excise duty must be the same for all cigarettes (3). Table 2.1 Criteria used by countries for tiered excise taxes, 2018 BASE OF TIERS COUNTRY Retail price Bangladesh, Belarus, Indonesia, Jordan, Mozambique, Myanmar, Pakistan, Thailand Cigarette grade (e.g. premium, mid-grade, economy) Egypt, Japana, Mali Producer price China, Lao People’s Democratic Republic Production volume Indonesia Type filter/non-filter Belarus, Georgiaa, India, Kenya, Republic of Moldova, Nepal, Papua New Guinea hand/machine made India, Indonesia kretek/white cigarette Indonesia tobacco content (dark/ blonde or dark/light) Algeria, Bolivia (Plurinational State of ) Packaging soft/hard Mozambique, Uganda Cigarette length India, Nepal, Sri Lanka Trade (domestic/imported) Iran (Islamic Republic of ), Lebanon, Myanmar, Solomon Islands, Tonga, Uzbekistan Leaf content (domestic/imported) Fiji, United Republic of Tanzania a Japan and Georgia were using a tiered excise tax structure when these data were collected in 2018, but as of 2020, that is no longer the case. Source: (1). 2.2 DESIGNING EXCISE TAX POLICY Significantly increasing the taxes on and prices of tobacco products is the most effective and most cost-effective policy to control tobacco use (4). Increased taxes that are passed on to tobacco users as higher prices reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face challenges ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues such as addressing the SCARE6 tactics of the tobacco industry. This section provides guidance for policy-makers regarding the best tax structure to use from a health perspective, taking into consideration all the appropriate tax designs. It also proposes recommended indicators to consider when formulating policy change. 6 SCARE tactics are the tactics most commonly used by the tobacco industry when countries plan to increase tobacco taxes. They are described, and refuted, in detail in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 19 2.2.1 THE IMPORTANCE OF THE TYPE OF EXCISE TAX STRUCTURE The existing theoretical and empirical evidence on approaches to the choice of (uniform) specific and ad valorem excises is reviewed below, along with their effects on price, consumption, perceived quality and variety of tobacco products, govern- ment revenue and tax administration. The use of the word “quality” in this chapter does not refer in any way to the health impact of a tobacco product. It refers rather to the consumers’ perceptions of quality and their decision to buy a product, which they may evaluate based on the packaging, the blend used for the cigarette or anything that makes the product more appealing to them. Just to be clear, from a public health perspective, all cigarettes are equally harmful even if perceived by consumers as having higher or lower quality. The choice between ad valorem and specific taxation is influenced by the market structure, i.e. the nature and degree of competition in the market for goods and services. Although each country has its own specific characteristics, the tobacco market structure is typically a monopoly or an oligopoly where firms have the power to control prices – and hence exploit the tax structure – to their benefit. For example, China, the largest producer and consumer of tobacco products in the world (5), has a state monopoly. In Viet Nam, foreign brands are produced under licence by the state monopoly. In Thailand and Egypt, despite the presence of foreign companies, the market is dominated by the state-owned company. In Uruguay, the oligopoly is led by a domestically owned company. In Bangladesh, the oligopoly consists of domestically owned companies competing with foreign companies (6). In most of Africa, the market consists of transnational tobacco companies (7). The impact of tax structure on final price: uniform specific versus uniform ad valorem The choice between specific and ad valorem taxes is a long-standing issue in tax policy, as the level and structure of excises have different implications for the interests and goals of various groups. Given the market structure of the tobacco industry – typically a monopoly or oligopoly for most products in most countries – different excises may have different effects on government revenue, manufacturer profit, consumer price, perceived product quality and variety and tax administration (8–16). Consequently, the two types of excise taxes – specific and ad valorem – may have different implications for public health to the extent that they affect individual consumption via their impact on perceived product quality, variety and prices. Moreover, governments have the potential to influence tobacco excises to manage demand, raise revenue and promote public health. 20 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax on prices and in turn on consumption is also influenced by price and income elasticities, as well as consumer perceptions of quality (“perceived quality”) and the variety of available products, which, in turn, are closely linked to the type of tax structure adopted. Tax structure is affected by both the price elasticity of demand and the price elasticity of supply. The price elasticity of demand measures the responsiveness of consumer demand to changes in prices. The price elasticity of supply measures how sensitive producers are to changes in prices. Tobacco tax structure is also influenced by market structure. In a monopoly, the profit-maximizing firm sets the price, considering the price elasticity of demand: the lower the price elasticity (in absolute value) – i.e. the less sensitive the consumer is to price changes – the higher the price the monopolist can set. Profits are typically abnormal in a monopolist market structure, meaning total sales revenue is higher than total cost (where total cost includes a normal profit). A monopolist producer therefore receives more than the minimum reward required to invest its (physical and human) capital and undertake business risks. Economic theory predicts that in a private monopoly, prices are higher than in an oligopolistic market. This is not, however, necessarily true when the monopoly is owned by the state and the government’s objective is not straightforward profit maximization: the government might have other considerations, such as preserving jobs (e.g. in China) or keeping prices low for low-income consumers (e.g. in Egypt). Under a monopoly, an ad valorem taxation structure enables the monopolist producer to set prices lower than would be possible under a specific tax structure. This is feasible because under ad valorem taxation, when supply increases and price falls, the price reduction is not fully borne by the producer. Rather, the price reduc- tion is partly shared by the government since, as supply increases, the tax per unit of product sold falls. In other words, ad valorem taxation leads to lower prices and higher consumption relative to revenue-equivalent specific taxation. Technically, this means that the supply function is less elastic under ad valorem taxation. In contrast, under a specific taxation structure, any increase in the monopolist producer price will go to the producers as revenue, which incentivizes them to increase prices. The same logic also applies to an oligopolistic market structure, where profits again are, in general, abnormal. KEY TAKEAWAY 1 In a monopoly or an oligopoly, specific taxation incentivizes industry to set prices higher than it would with ad valorem taxation. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 21 Understanding oligopolies, however, is more complicated, since they are characterized by strategic interdependence among a few firms. This strategic interdependence extends to the relationship between industry and regulators. Industry anticipates the government’s regulatory policy – whether through tax or other interventions – and acts accordingly. For example, competitors may coordinate and lobby against a certain tax structure reform or tax rate increase. Under an oligopoly market structure, ad valorem taxation is a relatively more efficient tool for transferring part of the profits to the government as tax revenue, since it acts like both an excise and a profit tax. In contrast, a specific tax has a smaller (negative) effect on profits. This explains why we observe multinationals that are leaders in high-priced brands (e.g. Philip Morris International [PMI]) lobbying in favour of specific taxation (17). As an example, in the countries of the Cooperation Council for the Arab States of the Gulf (GCC), the tobacco industry has been trying for a long time to lobby governments to introduce a specific excise (18–21). After years of consideration and discussions on the possible introduction of excise taxes, the GCC adopted the Common Excise Tax Agreement of the States of the Gulf Cooperation Council in November 2016 (21), which introduced an ad valorem excise on tobacco products. Tobacco companies’ support for excise tax structures ultimately depends on the market segments they control in a particular country. A company selling mainly premium brands will favour specific excises, whereas a company that sells mid-priced or economy brands would favour ad valorem excise (17). When oligopolistic firms produce identical products, a specific tax has a stronger positive effect on price and is more likely to be overshifted to consumer prices than an ad valorem tax (13). Overshifting means that the price increases by more than the tax increase itself. Empirical evidence supports this (22–26). KEY TAKEAWAY 2 In an oligopoly, prices are likely to increase by more than the amount of the specific tax increase when demand is relatively inelastic. In general, demand for a product depends not only on prices but also on consumer perceptions of quality and preferences for variety. For example, the most popular brand in GCC countries is Marlboro, a premium brand (1). Consumers differ in their willingness to pay, depending on their respective perceptions of quality, which influence whether they ultimately purchase high- or low-priced brands. A tax-induced price increase can cause the following plausible responses from consumers or users of tobacco products: (1) a group of consum- ers will quit; (2) a group of consumers will reduce their overall consumption; 22 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N (3) another group, most likely high-income users, will switch to an upgraded version of the tobacco product if its relative price (compared to the cheaper brand) has been reduced, which is the case under a specific tax increase; there might also be a group of consumers in the lower income range who switch to lower-priced variants of the tobacco product if the price gap increases, as is the case under an ad valorem tax increase; and (4) another group might switch to the illegal market or buy products in a neighbouring country with a lower tax where possible. Consumers’ decisions to purchase are also affected by their preferences for variety – meaning preferences among products that consumers perceive as equal in quality but are given different characteristics by the producers to account for consumer taste preferences. Thus, it is possible that a tax increase that leads to an increase in average prices will lead to an increase in the total quantity demanded in the market because of an increase in the variety of product choices available to consumers. Variety enables new consumers to be captured, especially in an environment lacking certain regulations (e.g. without plain packaging and flavour bans). The tobacco industry was able to capture a new group of consumers when it introduced menthol cigarettes into the tobacco market. There is more than sufficient evidence that menthol cigarettes increased youth smoking initiation, increased nicotine dependence and reduced adult smoking cessation (27). To prevent this from happening in their countries, Member States of the EU have prohibited characterizing flavours other than tobacco in tobacco products (28). It is therefore important to consider the broader effects that the structure and level of an excise tax can have on average price, perceived quality and the variety of cigarette brands and other emerging substitutes. When consumers make choices based on dimensions other than quantity, the two types of tax structures are not equivalent, even in a perfectly competitive market where firms have no market power (29–30). To illustrate this point, consider a US$ 1 cost to improve consumer perceptions of quality for a tobacco product. This will lead to an equivalent price increase under specific taxation but not under ad valorem taxation. At an ad valorem rate of 20%, the price must increase by more than US$ 1, or by 1/(1 – 0.2) to cover the US$ 1 cost of improvement, due to the multiplier effect. A specific tax induces consumers to reduce the quantity demanded, but they might still choose to pay a higher price in exchange for a product that they perceive to be of better quality. An ad valorem tax, on the other hand, leads to a reduction in both quantity and perceived quality, not a substitution between them. An ad valorem tax has only an income effect and – unlike specific taxation – does not lead to substitution between perceived quality and quantity. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 23 KEY TAKEAWAY 3 Under specific taxation, the industry has incentives to create upgraded variants of tobacco products that attract new consumers and encourage consumption. When firms produce differentiated products, as the tobacco industry does, economic theory provides ambiguous results regarding specific versus ad valorem taxation. The relative effects of the two types of tax are not as straightforward as in the case of oligopolistic firms producing a homogeneous product. With differentiated products, the relative effects of the tax types depend on various assumptions: whether or not firms face symmetric costs, whether the number of firms is fixed or new firms can enter the market and the level of the tax revenue requirement. When firms face different costs, ad valorem taxes exacerbate the absolute differences in marginal costs between them. The high-cost brand is not considered a perfect substitute for the low-cost brand. A sufficiently high ad valorem tax rate may lead to a relative underproduction of the high-cost products (31). The effect of specific and ad va- lorem taxes on consumer perceptions of quality depends on market structure and the price and income elasticities of demand across various qualities. The relative price of the cheapest product does not necessarily remain unchanged or increase; it might fall (32). Empirical evidence showing that increases in the specific tax lead to a lower market share for the cheaper generic brands and an upward shift to premium brands (33–34) usually considers gradual tax increases and ignores income effects. Chaloupka et al. (35) found that in 21 EU countries that impose a mixed tax system, the price gap between premium and low-priced brands – while not reflecting the full distribution of cigarette prices – is smaller when the specific component of the mixed structure dominates.7 Although the price gaps are narrower under specific taxation, there is evidence that firms sometimes respond by introducing new, very cheap (subvalue) brands, or they exercise differential tax shifting. This practice has been evident in India for quite some time, with the Indian Tobacco Company launching a number of cheaper variants of its flagship cigarette brand, Gold Flake, to take advantage of a lower excise tax rate in the so-called microcigarette (< 60 mm length) market (36). Consequently, the cheapest end of India’s cigarette market has expanded significantly in recent years due in part to the marketing of new brand variants like Gold Flake Century. 7 The EU countries impose a mixed tax structure with a minimum tax floor. Some countries rely on the specific component more than others, but they remain within a given range (the specific component must be between 5% and 76.5% of total tax share of the weighted average price). 24 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As another example, transnational tobacco companies, which have sold ultra- low-priced brands in the United Kingdom since 2006, have managed to double their market share in a few years: their real price did not increase, since they absorbed part of the tax increases (37). The share of ultra-low-priced brands increased between 2001 and 2009 from 5% to 10%, while the market share of economy brands increased from 40% to 50% and the market share of premium brands and mid-priced brands decreased during the same period (from 35% to less than 25% for the former and from 15% to 5% for the latter). In order to keep the price of discount brands low and certain consumers in the market, firms may overshift the tax for premium products and undershift it for the lower-priced products (37–40). KEY TAKEAWAY 4 Evidence suggests that the price gap between brands is narrower under a specific tax structure. As the tobacco industry simultaneously consolidates producers and widens its portfolio of products, evidence is emerging that it is introducing cheaper brands while increasing the price of its expensive brands, therefore paradoxically widening the price gap within its products. The extent of the impact is still unclear, however, and this evidence does not negate the overall conclusion that a specific tax structure reduces price gaps. The impact of tax structure on final price: uniform specific, ad valorem and mixed systems Evidence from the 2019 RGTE (1) data suggests that the average price of the most- sold brand of cigarettes – weighted by the number of smokers – is the highest in countries implementing a mixed system that relies more on specific excise, followed by countries applying specific excise taxes only, followed by countries applying a mixed system that relies more on ad valorem and then by countries that apply ad valorem excise only (Fig. 2.6). The price is lowest in countries that have no excise at all. In past WHO reports on the global tobacco epidemic, countries that applied specific excise only had the highest price, on average. The trend may have changed partly because more countries are adopting mixed excise systems. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 25 Fig. 2.6 Weighted average price and excise for a pack of the most-sold brand of cigarettes, by excise tax structure, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Shang et al. (41) describe and compare price distributions, using data from 16 countries of the International Tobacco Control (ITC) Project that impose different cigarette tax structures. Specific uniform taxation tends to result in less variability in prices than all other structures (ad valorem tax, mixed tax, tiered tax). In general, structures other than uniform specific tax give rise to more opportunities for brand switching and tax avoidance. Reliance on complicated systems is likely to be as- sociated with wider price distribution, leading to greater tax avoidance, as there are more opportunities for substitution with cheaper brands when taxes rise. KEY TAKEAWAY 5 Evidence suggests that the tax structures most likely to lead to higher prices are uniform specific excise tax structures or mixed systems that rely more on specific excises. Mixed system relying more on specic excise Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Mixed system relying more on ad valorem excise Ad valorem excise No excise 2.21 Specic excise 2.62 3.26 0 7 6 5 4 3 2 1 1.66 Excise tax Other taxes Retail price minus taxes PPP $ 7.56 PPP $ 5.66 PPP $ 4.84 PPP $ 3.73 PPP $ 2.52 26 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax structure on final price: uniform excise versus tiered tax systems Another aspect of tax structure that impacts final price is the use of tiered taxation, i.e. tax rates that vary according to product characteristics. The characteristics can vary, from price level to the type of tobacco leaf contained in the cigarette, the size of production volume, the packaging, etc. Table 2.1 (earlier) lists the criteria used by 31 countries as the basis for different tax rates. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax (see Fig. 2.7). Fig. 2.7 Weighted average price and excise for a pack of the most-sold brand of cigarettes for countries with and without tiered taxation, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Uniform specific tax structures are likely to lead to relatively higher prices with less variability in price distribution. Compared with tiered tax structures that have differential rates based on brand characteristics, uniform taxation may reduce consumers’ incentive to switch to cheaper brands (leading to higher quit rates and lower prevalence), as well as decreasing manufacturers’ incentive to reduce their tax liabilities by changing their pricing strategies, production process or size (42–45). In Indonesia, for example, where small producers were taxed more fa- vourably, manufacturers had an incentive to reduce their scale of production but increase the number of affiliated small companies. The issue was resolved when tax authorities considered the aggregate production of all affiliated companies in the application of differential tax rates. By 2017, there were 786 active factories, while Excise uniform Price and taxation per pack of 20 sticks (PPP $) Excise tiers 3.28 1.94 Excise tax Other taxes Retail price minus taxes PPP $ 4.99 PPP $ 6.30 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 27 there had been 4 198 factories in 2006.8 Abolishing the differential tax rates would have been even more effective in removing the incentive for tax avoidance, as well as helping improve both public health and government finances. KEY TAKEAWAY 6 Evidence suggests that applying a uniform excise tax on cigarettes is not only easier to administer than tiered systems but also more likely to lead to higher cigarette prices. The impact of tax structure on final price: the significance of the choice of the tax base It is important for the excise tax to be applied to the base that leads to the greatest possible effect on price and revenue. For specific taxation, the tax base is the quantity of tobacco products. The quantity of cigarettes, cigars and bidis is measured in number of sticks; for other tobacco products, such as smokeless tobacco or roll-your-own (RYO), it is measured in the weight of the tobacco. When the tax is ad valorem, the choice of the tax base is important not only for health considerations – due to its effect on consumption – but also for tax revenue generation and industry profits. An ad valorem tax that is based on the ex-factory price (or CIF value) provides tobacco manufacturers with opportunities to reduce their tax liability, especially when they control the distribution system. Tobacco producers may sell cigarettes to distributors who are related parties at a reduced price, which then serves as the basis for calculating their ad valorem tax liability. Distributors, however, can then set high prices and share the extra profit with the producers (46). Because of the potential for such trade mispricing, the best practice is to use the retail price as the tax base and introduce a minimum excise tax per pack. Data in the 2019 WHO RGTE (1) show that, on average, the price level of a pack of cigarettes and the excise level are both much higher in countries that use retail price as the base for their ad valorem excise (Fig. 2.8). The maximum retail sales price, which includes all taxes, is used as the ad valorem tax base in the EU. That price also forms the tax base for ad valorem taxes in a growing number of LMICs, including Brazil, Egypt, Thailand, Turkey and Rwanda. 8 Indonesian Ministry of Finance, personal communication, 2017. 28 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.8 Weighted average price of the most-sold brand of cigarettes in countries that use retail price as the base for their ad valorem excise, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 7 The base on which the excise is applied is important. For specific excise, the base needs to be clearly defined (for cigarettes, cigars and bidis, it is the number of sticks; for other tobacco products, such as smokeless tobacco or RYO, it is the weight of tobacco). For ad valorem excise – where the base is typically either retail price, CIF value or producer price – evidence suggests that countries that apply the excise tax on the retail price of cigarettes tend to have higher prices than those that apply the tax on other bases. CIF and producer prices are difficult for government authorities to ascertain and are prone to undervaluation. The tax impact on final price: the significance of the minimum excise tax The use of a minimum excise tax in countries with ad valorem or mixed systems is another important factor in determining final price. On average, the price of a pack of cigarettes – as well as the excise level – is much higher in countries that impose a minimum specific excise than in those that do not (see Fig. 2.9). While more than half of the 47 countries that apply a minimum excise are members of the EU, removing EU countries from the average calculations produces the same conclusions. Ad valorem/mixed with retail price as base 3.72 PPP $ 5.01 PPP $ 6.41 Ad valorem/mixed with other base 1.79 Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 29 Fig. 2.9 Weighted average price of the most-sold brand of cigarettes in countries with and without a minimum specific excise tax, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 8 Among countries that apply an ad valorem or mixed excise tax on cigarettes, evidence suggests that those that impose a minimum specific excise tax tend to have higher prices than those that do not. The minimum excise tax also helps guarantee minimum excise revenues. Summarizing the advantages, disadvantages and impacts of the choice of excise tax structure for tobacco products Table 2.2 summarizes the characteristics of different types of tobacco excise taxes and the advantages and disadvantages of each type in relation to its impact on quantity demanded, perceived quality of brands offered, price, certainty and stability of revenue, administration and enforcement and opportunities for tax avoidance and tax evasion as they are predicted by the economic theory of imperfect competition and observed in real life. Ad valorem/mixed with minimum speci c 1.80 4.23 Ad valorem/mixed without minimum speci c Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes PPP $ 4.82 PPP $ 7.41 30 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 2.2 Characteristics of different types of tobacco excise taxes Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) TA X B A SE The unit of product (e.g. 1 000 cigarettes) The value of the product (e.g. retail, wholesale or manufacturer price) Excise is calculated on an ad valorem basis; however, if the calculated tax falls below a specified minimum amount, a specific tax rate applies Unit and value of product Both unit and value, unless the calculated tax falls below a specified minimum, in which case the tax base is the unit A D M IN IS TR AT IV E R EQ U IR EM EN TS The tax should be collected at the point of manufacturing or at the time of importation Low, as only the volume of the products needs to be ascertained Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as with a pure ad valorem regime Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises, as well as minimum specific excise tax compliance U N D ER VA LU AT IO N Not an issue Susceptible to undervaluation Provides an easy tool to prevent undervaluation of low-priced brands subject to the minimum specific excise The ad valorem part of the excise collection may be susceptible to undervalua- tion, depending on the choice of tax base The minimum specific excise prevents possible ad valorem tax base undervaluation of low-priced brands CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 31 Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IM PA C T O N P ER C EI V ED PR O D U C T Q U A LI TY Upgrading effect tends to reduce the relative tax on higher-priced brands Multiplier effect provides a disincentive to costly so-called quality improvement No incentive to upgrade higher- priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands IM PA C T O N P R IC E Tends to lead to relatively higher prices, particularly for low-priced cigarettes Tends to lead to relatively lower prices; price reductions will be subsidized if the multiplier effect is strong Tends to lead to relatively higher price increases for low-priced cigarettes An increase in the specific tax will to lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem payment if the base of the ad valorem includes excise An increase in the specific tax will lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem tax amount if the base of the ad valorem includes excise. Increases in the ad valorem and /or specific tax will raise the minimum tax paid if the minimum is a percentage of the total tax on, for example, weighted average price; they will reduce price gaps, given impact on perceived quality 32 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IN FL AT IO N The real value of the excise tax will be eroded unless the tax is adjusted in line with inflation The real value of the excise tax will be preserved as prices increase, at least to the extent that tobacco product prices follow inflation The real value of the minimum specific excise will be eroded over time unless the excise is adjusted in line with inflation The real value of the specific excise will be eroded unless the excise is adjusted in line with inflation The real value of the specific excise and the minimum specific excise will be eroded unless the excises are adjusted in line with inflation H EA LT H B EN EF IT S Will discourage consumption of tobacco products irrespective of the price band May encourage more trading down in favour of cheaper cigarettes, reducing the health benefit The minimum specific excise reduces incentives for trading down May reduce trading down Reduces trading down Source: (47). 2.2.2 OTHER TAX DESIGN CONSIDERATIONS The significance of automatic adjustments and indexation of specific tax to inflation Specific taxation does not depend on price and therefore, unlike the ad valorem tax, is not automatically adjusted for inflation. The real value of a specific tax is eroded over time as the price of the taxed product increases. Therefore, especially in countries with rapid growth in inflation, the nominal value of the specific tax must be increased regularly in order for the tax to maintain its real value. This is of great importance for both public health and public revenues, especially in countries where manufacturers do not increase prices regularly and/or low-priced tobacco products are the dominant products in the market. Table 2.3 lists countries that include automatic adjustments to their excise in order to avoid the erosion of the specific excise over time, using different units of adjustment and based on different frequencies. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 33 Table 2.3 Countries that include automatic adjustments to the specific excise COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Argentina Inflation (consumer price index [ CPI]), on a quarterly basis Armenia Minimum specific excise set to increase in the Tax Code by 15% on average between 2019 and 2021 Australia Wages – excise rates on tobacco and tobacco products increase in March and September each year, based on average weekly ordinary time earnings Bosnia and Herzegovina Specific excise rate is increased annually by at least 7.50 convertible marks per 1 000 cigarettes; minimum excise tax is increased annually to be at least 60% of the weighted average price Canada Inflation – federal tobacco tax rates are to be increased every five years, indexed to Canada’s CPI starting in 2019 Chile Inflation Colombia Specific tax set to 1 400 pesos, increased to 2100 pesos in 2018; starting in 2019, it will increase yearly by the CPI plus 4 points Costa Rica Inflation Dominican Republic Inflation, on a quarterly basis France Increase from 2017 to reach an average price for cigarettes of €10 per pack by 2020 Honduras Inflation, annually to December of the previous year Italy Minimum tax burden calculated every year in March on the basis of the weighted average price of cigarettes sold in the previous year New Zealand Inflation annually plus 10% annually from 2017 to 2020 Nicaragua Updated annually as of 1 January 2017, taking the highest among the annual devaluation of the official exchange rate of Cordoba with respect to the US dollar, published by the Central Bank of Nicaragua, and the annual inflation rate of the CPI published by the National Development Information Institute, observed in the last 12 months available North Macedonia Specific and minimum specific rate increase by 0.2 denars per cigarette on 1 July each year until 2023 Philippines Agreed tax increases and rates for specific excise tax between 2020 and 2023, with a 5% indexation thereafter Romania Inflation, annual (1 January) adjustment of the total excise according to inflation calculated on 1 October of the previous year Serbia Inflation, every six months Southern African Customs Union (SACU) – Botswana, Eswatini, Lesotho, Namibia and South Africa Inflation, on an annual basis9 Sweden Inflation 9 While the adjustment is not strictly automatic in the SACU, it is greatly informed by the inflation rate. The Treasury has some discretion. In recent years, the increases have typically been slightly above inflation. 34 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Turkey Producer Price Index Ukraine Agreement to increase the specific component 20% annually between 2019 and 2025 United Kingdom Increase by 2% above the retail price index (measure of inflation) for the 2015–2020 Parliament Source: (1). KEY TAKEAWAY 9 To avoid erosion of their specific excise tax, countries need to regularly – and, ideally, automatically – adjust the excise to inflation. The significance of automatic adjustments and indexation of specific tax to income growth In addition to the risk of erosion due to inflation, the effect of a (specific) tax can be significantly reduced if the tax is not adjusted for increases in consumer income. Income growth makes products more affordable – thereby encouraging consump- tion – especially in countries with rapid income growth. Australia is one of the rare countries that explicitly adjusts its specific excise rates according to wage growth (see Table 2.3). However, a number of countries have adopted automatic adjustments that are higher than inflation and sometimes largely cover income growth as well (see also Table 2.3). Adjusting tax for income growth contributes to increases in prices that make tobacco products less affordable (see section 2.2.3). KEY TAKEAWAY 10 The specific excise tax needs to be adjusted to reflect income growth so that tobacco products do not become more affordable over time. Measures for specific contexts: the role of pricing and other non-tax regulation Emerging evidence indicates that the tobacco industry finds ways to mitigate the impact of higher taxes on prices. For example, despite the heavy reliance on specific taxation in the United Kingdom, a price differential between premium and cheap cigarettes still exists. There is evidence that the tobacco industry does not always pass tax increases on to cheaper products (37, 48). Differential shifting among price categories is also observed in the EU (49–50), New Zealand (38) and the United States (51). Therefore, the public health community has suggested that pricing regulation could be considered as a method of eliminating inexpensive tobacco products that are often used by the young and the poor (52). Three types of pricing regulation are described below: minimum mark-up, price floor and price ceiling. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 35 Pricing regulation • Minimum mark-up – It has been argued that a minimum mark-up of wholesale tobacco prices could be a better strategy to raise prices of tobacco products than excise tax increases. Minimum mark-up laws aim to discourage the sale of products below an assessed cost by imposing a mark-up to the cost declared at different levels of the supply chain. Some studies in the United States have shown that minimum mark-ups do not increase average cigarette prices (53–54). However, a recent study of the impact of minimum mark-up/ price laws has shown that these laws are linked with higher prices, especially for the cheapest brands, and could be used as an effective tool to mitigate the impact of the industry’s price-reducing promotions (55). Another concern related to minimum mark-ups is that they can be manipulated by manufac- turers and are likely to lead to higher profits for the industry, as well as extra administrative costs for the government (56). • Price floor – A few studies suggest that setting a price floor, or a minimum price, is an alternative strategy for increasing tobacco taxes, particularly with respect to reducing health inequities (57–60). A price floor, imposed by the government or as a vertical restraint imposed by the supplier upon retailers, is a price that firms cannot legally undercut. Governments impose price floors to restrain unfair competition or, in the case of services, to increase quality. It is difficult, however, to find the right floor or to anticipate unintended conse- quences or an industry’s adjustments. A study in Malaysia, where a minimum price for cigarettes was imposed in 2010, found that the policy did not seem to have a meaningful impact on prices: licit brand prices remained well above the minimum price, while illicit brands remained well below it. This outcome may be a result of the floor being set too low or the proportion of illicit trade being high, either of which would reduce the effectiveness of the policy (52). In the EU, imposing minimum retail sale prices for cigarettes could be a breach of harmonized legislation concerning the internal market, as minimum prices would distort competition. Therefore, increasing minimum excise duties is recommended instead, to discourage consumption (61). Increasing the minimum excise duties would also result in the additional revenue going to the governments instead of contributing to industry profits. A price floor would probably lead to increased industry profits – giving the industry greater funds for its marketing strategies – and lower tax revenue for governments, reducing their ability to cover costs associated with tobacco use. By reducing price competition, the price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifica- tions). Competition among firms may prevent them from raising their prices, 36 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N but a government that imposes a price floor does this for them. Minimum pricing is likely to create windfall profits for manufacturers and retailers. It can even help manufacturers sustain a cartel. If the industry uses the money to increase promotions, advertising or grant proposals for related research, this could undo some of the potential benefits of the policy.10 Some recent evidence shows that, at least in the case of the United Kingdom, increased concentration of power among a handful of multinational corpora- tions is enabling them to undermine tax increases through increased price segmentation, and that requiring minimum prices might be a good way to address the problem. A longitudinal analysis of price data from the United Kingdom (48) has shown that despite regular excise tax increases over time, average real prices for cheaper segments of the tobacco market (in this case, cigarettes and RYO) did not increase – indicating an undershifting of the tax increases in those segments that resulted in increased sales volume. At the same time, average prices for more-expensive market segments increased, indicating overshifting of the tax increases that resulted in decreased sales volume. This industry strategy ensures that the most price-sensitive consumers remain addicted, while encouraging initiation and discouraging cessation. Furthermore, segmenting the market further by overshifting the tax increase on premium brands while undershifting it for cheaper brands mitigates the impact of declining consumption resulting from higher taxes while increasing overall industry margins and profitability. Another situation where setting minimum prices can be a useful policy is specific to the United States. Banning marketing and promotions11 is not possible under the freedom of expression protections of the Constitution of the United States (Amendment I), and it was estimated in 2008 that more than 82% of all advertising and promotional spending by the tobacco industry was focused on reducing the price of their products at the point of sale (62). This limitation on how government can set policy has paved the way for the implementation of minimum price policies in many states and cities to counter the detrimental impact of price promotions on consumption and on the tax policy itself. Huang et al. (55) found that the presence of minimum price laws was associated with higher cigarette prices. They also noted that cigarette prices were even higher than prices resulting from minimum price laws in states that also prohibit industry from engaging in other price-reduction strategies, 10 See, for example, the PMI strategy of setting up the Foundation for a Smoke-Free World and grant proposals for related research. 11 See section below on banning promotional discounts for tobacco products for further discussion about marketing and promotions. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 37 such as below-cost combination sales, using trade discounts to reduce the base cost of cigarettes and distributing below-cost coupons to consumers. In the contexts of both price segmentation and price promotions, the imposi- tion of minimum prices as a complementary policy to excise tax increases – not as a policy alternative – may help guarantee that taxes do indeed lead to the intended reduction in consumption. Nonetheless, more evidence is needed to support the effectiveness of this policy. • Price ceiling – Concerns about differential tax shifting have led to suggestions that a price cap may benefit public health by limiting the tobacco industry’s ability to reduce average prices by differentially shifting tax increases among various price segments (63–64). Because tobacco manufacturers operate across international markets, however, they could maintain low prices in one coun- try but maintain overall profitability by selling more premium products in another country. Additionally, limiting price increases does not fit the public health purpose of reducing consumption. It is worth noting that maximum retail prices are sometimes used as a base for calculating the ad valorem tax payments in countries with ad valorem or mixed tax systems. In a systematic review of the literature on non-tax policy approaches to raising prices, Golden et al. (63) hypothesized how such policies would influence price dispersion and average prices. Their study found that minimum price policies combined with promotion bans have the potential to increase average prices. This is, of course, relevant in a context where price promotions are present. From either a theoretical or a practical standpoint, however, it is clear that price policies cannot be used alone and should always be considered as complements to excise tax increases. Significantly increasing taxes is the most effective way to dissuade consumption, correcting whatever bias may exist. Significant tax increases also provide the added benefit of raising money for the government rather than profits for the tobacco industry. Nonetheless, a minimum price might help narrow the gap between cheap and pre- mium cigarettes when applied to all tobacco products to avoid product substitution. Other non-tax regulation • Banning promotional discounts for tobacco products – The sale of tobacco products at a discount rate – such as through reduced-price coupons or buy-one-get-one-free offers – encourages consumption and undermines tax increases. Such practices should be completely banned. They often exist outside the realm of the finance sector because they are considered a type of marketing – promotional discounts are usually addressed in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. 38 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N According to the 2019 RGTE (1), 118 countries out of 195 with all levels of income had such a provision implemented as of 31 December 2018. • Banning the sale of single sticks of cigarettes – Article 16 of the WHO FCTC, “Sales to and by minors,” paragraph 3, requires Parties to “prohibit the sale of cigarettes individually or in small packets which increase the affordability of such products to minors”. Some smokers opt for buying single sticks partly because of the lower im- mediate costs of buying cigarettes individually (65). Internal (unpublished) analysis of single-stick prices collected by WHO for the 2012, 2014 and 2016 editions of the WHO RGTE shows that, in fact, the aggregate price of 20 single sticks of cigarettes sold separately is generally higher than the price of a 20-cigarette pack sold in the market of a specific country. Despite this fact, single-stick sales – and sales of small-sized packs – make cigarettes accessible to consumers with limited disposable income. De Ojeda (66) found in a study conducted in Guatemala that single-cigarette sales are associated with increased cigarette accessibility for less-educated, lower-income populations and minors. Single-stick sales are also a feature of many markets in South-East Asia, including most notably Bangladesh and India, but also in other parts of the world, e.g. South Africa. Single-stick sales also reduce the impact of a tax increase, since the in- crease per stick is much smaller than the increase per pack (67). In a study investigating how smokers in New York City responded to a tax increase of US$ 1.25 per pack in 2008, Coady et al. (68) found that 15% of smokers bought more single cigarettes than they had previously.12 By allowing single stick sales, governments risk losing part of the ad valorem taxes if the tax base is the retail selling price; the retail price of single sticks is much more difficult to monitor than the retail price of packs of cigarettes, on which, for example, tax stamps with prices can be applied. An internal WHO analysis of the most recent tobacco control laws in 2018 in 195 countries found that 86 countries impose by law a ban on the sale of single sticks of cigarettes (36% of the countries are high-income, and 64% are LMICs). In addition to banning the sale of single sticks of cigarettes, 67 of the 86 countries specify a minimum size for packs of cigarettes. Most (52 countries) use the 20 cigarettes per pack standard, but minimum sizes 12 Before 2018, the use of single sticks was possible, but it has since been banned. See New York City Administrative Code. chapter 7: regulation of tobacco products, subchapter 1: Tobacco Product Regula- tion Act, §17-704.a-1. New York: New York Legal Publishing Corporation; 2020 (http://library.amlegal. com/nxt/gateway.dll/New%20York/admin/title17health/chapter7regulationoftobaccoproducts?f=tem plates$fn=default.htm$3.0$vid=amlegal:newyork_ny$anc=JD_T17C007, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 39 range from five sticks in Ghana up to 25 sticks in Papua New Guinea. Another 15 countries do not specifically ban the sale of single sticks but do specify the minimum size for packs of cigarettes. According to the WHO FCTC, in order to reduce affordability, single stick sales should be banned and a minimum number of cigarettes contained per pack should also be defined. KEY TAKEAWAY 11 A number of non-tax measures are closely connected to tax policies, including price regulations, bans on promotional discounts for tobacco products and bans on the sale of single sticks of cigarettes. The price policies discussed are (1) minimum mark-up, (2) price floors and (3) price ceilings. Current evidence does not yet demonstrate that minimum mark-ups and price floors lead to increases in average price. Nonetheless, they may be relevant in some specific contexts as complementary policies to excise tax increases. Price ceilings limit price increases, which can mitigate their impact on consumption. Price marketing strategies such as promotional discounts and the sale of single sticks undermine the effect of tax policies and should be banned. A minimum pack size should also be required by regulators. Tax increases and their possible impact on inflation At times, the inflationary impact of tax increases on cigarettes and other tobacco products is raised as an argument for not increasing these taxes. This may be a concern in countries where wages and/or a significant share of government spend- ing is indexed to inflation (e.g. for public pension payments) or where government policy is to keep inflation low. The extent to which tobacco product tax increases lead to increases in inflation depends on several factors, most notably the share of these taxes in prices and the weight tobacco prices are given in computing a price index. For example, if taxes account for 25% of tobacco product prices, a doubling of the tax (100% increase) will increase prices by 25%. If the weight given to tobacco products in the price index is 3%, the index will rise by 0.75% in response to the tax increase. As tobacco taxes account for a larger share of tobacco product prices, the inflationary impact of a tax increase will be greater. Similarly, as tobacco products are given more weight in computing a price index, a given tax increase will have a greater inflationary effect. In general, for most countries, the inflationary impact of tobacco product tax increases would be relatively small (47). 40 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Consumer price indexes have multiple purposes. They are an important economic indicator for most countries and are often a key determinant of monetary policy. Inflation rates have a direct impact on interest rates and exchange rates. In many countries, changes in wages, social security benefits and other payments are tied to inflation as measured by a price index. Price indexes are used to provide more accurate comparisons of changes in expenditures, incomes and prices for specific goods over time, as well as to allow comparisons across countries. Given the many uses of consumer price indexes and the potential inflationary impact of tobacco tax increases, some governments have developed alternatives that exclude tobacco (and sometimes other goods) for some uses. For example, since 1992, France has excluded tobacco products from the price index used for adjust- ing minimum wages (47). However, many countries continue to include tobacco product prices in their consumer price indexes. Excluding tobacco products from the basket of goods used in developing key price indexes would greatly reduce concerns about their impact on inflation. In addition, with declining consumption of tobacco products, the inclusion of their prices in key price indexes results in a distorted measure of price for many consumers. KEY TAKEAWAY 12 If governments are concerned about the potential inflationary impact of a tobacco tax increase because wages or some government spending may be tied to a price index, they can use a price index that excludes tobacco products. The importance of taxing cigarettes and other tobacco products in a comparable way While cigarettes are the most commonly used tobacco product globally, other tobacco products are as prevalent and sometimes more prevalent than cigarettes in some parts of the world. Bidis and smokeless tobacco are the main products consumed in some countries in South-East Asia – Bangladesh and India in particular – and waterpipes are widely used for smoking tobacco in the Eastern Mediterranean region (4). These products, as well as RYO, have historically been taxed much less than cigarettes (see, for example, Fig. 2.10 for Bangladesh and India, where the excise tax and prices of bidis and smokeless tobacco are much lower than those for cigarettes). This differential taxation undermines the health impact of excise taxes on tobacco products because (1) it encourages users to switch from cigarettes to the lower-taxed product (see the case of Thailand below); (2) it is not effective in reducing tobacco use in general, especially if the most widely used product in the country is not cigarettes; (3) it can encourage tax avoidance by companies that may CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 41 redefine products that are similar to cigarettes so that they fall within the lower- taxed product category (see the EU example below) and (4) it reduces the size of government revenues since those products could have been taxed at higher levels. Fig. 2.10 Price and tax of the most-sold brand of cigarettes, bidis and smokeless tobacco in Bangladesh and India, 2018 Source: (1). In Thailand, for example, the price of cigarettes has been raised quite successfully through taxation over a number of years, while taxes and prices of loose or RYO tobacco have until very recently remained unchanged. Indigenous tobacco used for RYO cigarettes has historically been exempt from excise, while foreign tobacco was taxed at a low level relative to that of manufactured cigarettes. Consequently, Thailand experienced growth in the RYO market even though cigarette consumption had been falling.13 The Thai government eventually took strong action to address this issue. First, the exemption for indigenous tobacco was removed in 2018. The Cabinet then approved an increase in the excise rate on small producers (of indigenous tobacco) from 0.005 baht per gram to 0.025 baht per gram in 2020, with another increase to 0.1 baht per gram scheduled for 2021 (69). In the EU, the minimum excise duty levels for cigars and cigarillos is significantly lower than that for cigarettes. The Member States of the EU are required to levy an 13 WHO Country Office for Thailand, personal communication, 2019. 12.8 0 190 54.9 64 24 Price Excise tax amount 80 12.8 3.8 44 Lo ca l C ur re nc y 0 50 100 150 200 Cigarettes (20 sticks) Bangladesh, taka India, rupees Bidis (20 sticks) Bidis (20 sticks) Cigarettes (20 sticks) Smokeless (20 g) 42 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N excise duty of at least €90 per 1 000 cigarettes, which should be 60% of the weighted average retail selling price of cigarettes released for consumption. For cigars and cigarillos, only €12 per 1 000 items, or an excise duty of 5% of the retail selling price, is required. As a result, the excise tax share on cigarettes is much higher in many EU countries than the share for cigars and cigarillos. In response, some companies started to market so-called borderline cigarillos. These products have characteristics similar to cigarettes but can be sold at a lower price because for excise purposes, they are considered as cigarillos. Although this issue seems to be largely solved by amendments to the definitions of these products at the EU level and a change in tax structures in some countries, it is important to be aware of the unintended incentives that can be created by large gaps in excise tax levels between product categories (70–71). For more details on industry tactics to undermine tax increases, see Box 2.1. KEY TAKEAWAY 13 To make excise tax on tobacco products more effective in reducing overall tobacco use and to avoid substitution between products, all tobacco products need to be taxed in a comparable way. The Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce the demand for tobacco) (73) recommend that all tobacco products should be taxed in a comparable way. Box 2.1 Industry tactics used to undermine tax increases Tax increases reduce the demand for tobacco products and present a threat to the tobacco industry’s high profits. The industry responds by using various strategies (17, 46, 48), including the following (46): Stockpiling (forestalling/front-loading) – Before the implementation of an an- nounced tax increase, manufacturers overproduce tobacco products, paying the pre-tax-increase rate. As a consequence, sales and tax revenue decline immediately but temporarily after the tax increase (while sales and revenues had increased sub- stantially just before the tax increase) and the industry attributes this drop in revenue to the emergence or increase of illicit trade. This practice results in tax avoidance if there is no law prohibiting it (see also the discussion on anti-forestalling in Chapter 3). Changing certain product characteristics (for example, weight or length) and/ or adjusting the production process – When tobacco products are taxed at different rates or are subject to different tax increases, the industry can, for example, re-label one type of tobacco product as another product that has a lower tax burden (as in the example of the EU above). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 43 Choosing the time of a price increase announcement strategically – The industry may raise prices in anticipation of a tax rate increase, generating extra profits in the period until the tax is actually implemented. When the tax increase is implemented, consumption and tax revenue will fall, but prices will not change, so the industry can claim that the tax policy was ineffective in reducing demand. Adopting price-discriminating strategies or price-related promotions – The industry may offer discounts, retailer rebates or added value (gifts) to tobacco purchases to minimize the loss of price-sensitive consumers. This, however, is not possible in countries where strict bans on tobacco advertising, promotions and sponsorship are implemented. Using brand proliferation (for example, launching a low-priced brand) and price segmentation – Manufacturers can choose to reduce prices of certain brands or introduce new, even cheaper ones to keep price-sensitive consumers in the market. There is evidence that firms introduce new cheaper products and use price-marking – printing the price directly on packs of tobacco products – to lock in their price (48). Such practices compromise both public health and revenue objectives. Differential shifting of tax increases across different price segments, depending on the market circumstances – The industry may increase the price of a product by more than the amount of the tax increase (tax overshifting) and blame the govern- ment for the total increase. Tax overshifting is profitable when demand is inelastic, that is, when the price increase more than offsets the reduction in sales. The industry may overshift the tax increase for higher-priced brands, which are expected to be more price inelastic than lower-priced brands. Additionally, to keep price-sensitive consumers in the market, the industry may temporarily absorb part (or all) of the tax increase on lower-priced brands. The differential tax shifting will lead to different responses in the demand for the different brands (37, 48). Lobbying government to distort interventions – Government policy might be influenced by tobacco industry lobbying, directly or indirectly. Policy-makers are not simply welfare or revenue maximizers; they also value political support. Industry lobbying might lead to adopting a favourable type of taxation, postponing tobacco tax increases or distorting the tax rate downwards (17). Article 5.3 of the WHO FCTC, “On the protection of public health policies with respect to tobacco control from commercial and other vested interests of the tobacco industry”, and its guidelines provide useful guidance on how to address tobacco industry interference. In fact, all 181 countries that are Parties to the WHO FCTC have a legal obligation to implement the requirements of Article 5.3. Having correct expectations about industry responses is important for estimating the impact of a tax increase on consumption and tax revenue. 4 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.2.3 MEASURING IMPACT AND RECOMMENDED INDICATORS Governments need to consider a number of indicators when formulating policy changes. Inappropriate assumptions about consumer behaviour, market structure and industry behaviour can lead to faulty policy analysis. Measuring impact on price and demand Consumption habits, local traditions and industry characteristics – such as the number of different brands offered, the possibilities of cross-border shopping and the presence and level of illicit trade – all affect the shape of the demand and supply of tobacco products, thereby determining the value of the price elasticities. Price elasticity, together with the industry’s pricing strategies – for example, the degree of tax shifting – and the tax share in the retail price, determine the elasticity of the tax base, regardless of whether the base is determined by quantity (for specific taxation) or transaction value (for ad valorem taxation). The importance of elasticity estimates Different types of elasticity should be considered: • price elasticity – own-price elasticity – measures the response of consumers’ demand for a product following a change in the price of the product. – cross-price elasticity – measures the response of consumers’ demand for a product when the price of another product changes. Cross-price elasticity can also occur between different brands or price segments for the same product. • income elasticity – the response of consumers’ demand for a product when their income level changes. Correct estimates of price and income elasticities are important for policy-makers who need to anticipate the impact of a tax increase on consumption and tax rev- enue. Estimates will vary depending on a number of factors, including whether responses are considered in the short run versus the long run, the functional form of the demand function used, whether factors such as addiction or tax evasion are accounted for and the way data are constructed. For example, details such as the degree of aggregation of data, whether gender- or age-specific data are used, the time span covered and which estimation procedures are used (e.g. ordinary least squares, two-stage least squares or generalized method of moments) will all affect the results of the estimate (72). Price elasticities may change over time, as well because of changes in any of the other factors affecting demand, such as income or tobacco control measures, and also because of changes in estimation techniques and the types or sources of data used. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 45 Moreover, what is of most interest is the price elasticity of total demand. A tax increase may reduce tax-paid retail sales but not necessarily total consumption. For example, smuggling can significantly bias price elasticities when the elasticities are estimated using legal sales data; not accounting for possible illicit trade might lead to overestimation. Similarly, when cross-border shopping is included, the price elasticity of demand is lower (in absolute value) (74). Estimating the total price elasticity of demand for legal and illegal consumption can be done by using cross- sectional data from nationally representative household surveys. However, this approach also has its weaknesses. For example, respondents tend to underreport their consumption of tobacco, which leads to bias in the size of demand. Price endogeneity14 is another technical problem that can be challenging to address. To comprehensively estimate the total effect of a tax increase on demand for all tobacco products as well as on tax revenue, the degree of substitutability between them needs to be estimated (55). Cross-price elasticity measures how the quantity demanded of a particular tobacco product changes when the price of another tobacco product increases. When this elasticity is positive, the products are substitutes; the higher the value of the elasticity, the closer substitutes the products are to one another. For example, positive cross-price elasticity between RYO and manufactured cigarettes implies that the demand for RYO increases as the price for cigarettes increases. Substitutability may also arise between different cigarette brands – when the relative price of economy brands increases, demand for premium brands may increase. This effect can be exacerbated when differential (tiered) taxation is ap- plied on different types of cigarettes, further widening the gap in prices between brands and segments and encouraging substitution. The substitutability between traditional and new and emerging tobacco and nicotine products is currently of great interest (see section 2.4 below). In some countries, different tobacco products can also be complementary rather than substitute goods. This means that when the price of a tobacco product increases, the demand for its complement drops because users are unlikely to use the complementary tobacco product alone. For example, a number of studies have found manufactured and indigenous bidi cigarettes to be complementary goods in India (75–76). The sign and magnitude of income elasticity vary across time, countries and demographic groups. For example, in the United States, a high-income country, income elasticity over time has changed from positive to negative, and cigarettes have switched from being a normal good to an inferior good (77–78). On the other hand, among LMICs, where prevalence of smoking tends to be relatively higher, 14 Price is endogenous because it is not an independent variable: it is estimated by dividing expenditure on tobacco by consumption of tobacco, with consumption being a dependent variable in the estimation of price elasticity. 46 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarettes might still be a normal good, with consumption increasing as income levels increase (positive income elasticity) (79–80). There are unobservable characteristics that differentiate higher-income smok- ers from lower-income smokers, such as differences in time and risk preferences, differences in associating a social stigma with smoking and differences in taste for smoking as a pleasurable activity. When these characteristics are ignored, estimates of the correlation between income and smoking-related outcomes are biased. Kenkel et al. (81), using techniques that estimate the causal effect of income on smoking among low-income adults, found that tobacco is a normal (even a luxury) good: higher income is associated with a higher probability of smoking participation and a lower probability of smoking cessation. These results are consistent with those regarding the impact of the business cycle – periods of expansion or recession in economic activity – on health be- haviour and outcomes. Ruhm (82–83), for example, found that smoking declines during temporary economic downturns and increases during economic expansions. Tarantilis et al. (84) found that estimates of income elasticities of demand in Greece were higher after the economic crisis of 2010 than before it. The financial crisis and the austerity measures shifted the demand for cigarettes downwards and turned cigarettes into a more income-elastic good. Interestingly, evidence from Germany suggests that the propensity to become a smoker significantly increases during an economic downturn. However, among those who are already smokers, cigarette consumption actually decreases (85). Ideally, when estimating price and income elasticities, the effect of non-price policies should also be accounted for. A recent study from South Africa shows that failing to take non-price policies into account will overstate the price effect (86). The NCI/WHO Monograph (4) suggests that price elasticity of demand for tobacco is on average -0.4 in high-income countries (ranging from -0.2 to -0.6). Estimates for LMICs are more variable, clustering around -0.5 (ranging from -0.2 to -0.8). A price elasticity of -0.5 means that a 10% increase in price would lead to a 5% reduction in consumption. KEY TAKEAWAY 14 Policy-makers need to know the elasticity of demand – including price elasticity (own-price and cross-price) and income elasticity – for tobacco products in their country in order to correctly assess the impacts of potential policy changes on consumption and subsequent revenues. These estimates need to be made on a regular basis to capture changes in demand over time. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 47 The importance of the tax base elasticity The tax base elasticity measures the sensitivity to a change in the tax rate of the base on which the tax is imposed – the base being tobacco consumption in the case of specific taxation and tobacco expenditure in the case of ad valorem taxation. The magnitude of the elasticity of the tax base depends on price elasticity of demand, the tax structure, the level of the tax rate and its share in price, along with the industry response through its decision to absorb, pass through or overshift the tax on to the retail price. Consumers’ preferences and income, the availability of substitutes and other non-price tobacco control measures also influence the tax base elasticity, essentially through the price elasticity of demand. The magnitude of the elasticity of the tax base also depends on social motivations, including price and tax expectations, which are ultimately impacted by successful tobacco control measures that affect consumers’ willingness to pay taxes or prices. In addition, the tax-base elasticity depends on smokers’ perceptions of the prob- ability of detection and tax enforcement when using illegal products, as well as the availability and accessibility of opportunities for tax evasion and avoidance. Finally, consumers’ willingness to pay taxes depends on their perceptions regarding the use of the tax revenue (87). Therefore, the tax base elasticity is largely influenced by government policy choices. Increasing the tax share in prices is recommended by WHO as a tool to achieve the public health objective of reduced tobacco use: a higher tax share in prices increases the tax base elasticity, all else remaining constant, and therefore increases the reduction in the tax base through the resulting reduction in smoking. However, manufacturers can be expected to attempt to manipulate the tax base elasticity through their pricing policies, such as tax shifting. As discussed earlier in this chapter, industry behaviour is itself affected by government tax policy and regulations. A number of factors need to be taken into account when considering tax pass- through. As discussed earlier, tax is more likely to be overshifted within a specific tax structure than within an ad valorem structure. There is also evidence of industry overshifting the tax for premium or expensive cigarette brands while undershifting the tax for cheaper brands. This indicates that within a given market, the industry’s decision on the extent of tax pass-through will vary based not only on the tax structure but also on the structure of the market. It will also vary by brand. But this does not give an indication about the impact of the tax increase on the average price of a tobacco product. In the context of the tax base elasticity and the impact of tax increases on revenues, it is important to assess how tax increases affect average prices. The example of South Africa is very useful here. Over the past two decades, South Africa has been consistently increasing its specific excise tax on cigarettes, which has led to large price increases. An analysis of the effect of excise tax increases on 48 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarette prices between 2001 and 2015 (26) shows that while there is evidence of tax overshifting, at least within a specified period of time, overall, the tax has been undershifted in real terms. This undershifting is due partly to increased competitive- ness in the market and partly to the introduction of low-priced brands. Of course, this encouraged some consumers to downshift their consumption to cheaper products, but it also pushed manufacturers of more expensive brands to absorb part of the tax increase to reduce the impact on price. A change in the level of the tax rate – with all other factors that influence con- sumption held constant – will result in a change in the tax revenue.15 Estimates of tax base elasticity help governments predict changes in tobacco tax revenues following a tax increase (see details in Annex 2.2). Under specific taxation: • tobacco consumption – the tax base – is expected to be price inelastic (17, 47); • prices increase by less than the tax increase, on average (there is no tax overshifting overall); and • consumption – the tax base – is also expected to be tax inelastic: the quantity of consumption falls less than proportionately to the tax increase, and the tax revenue increases. Under ad valorem taxation: • the tax base is the total consumer expenditure (or, equivalently, the industry sales revenue) on (legal) tobacco consumption – that is, the tax base under ad valorem taxation is determined by both price and quantity, which is itself a function of price; • the sign of the tax base elasticity – which can be either negative or positive – depends on the magnitude of the price elasticity of demand; • since evidence suggests that tobacco demand is price inelastic, the tax base elasticity is positive; • when the ad valorem tax rate increases, both price and quantity adjust, but quantity falls less than proportionately to the price increase, and tax revenue increases; and • a tax rate increase leads to both a higher level of revenue and a lower level of consumption; the value of the elasticity – and hence the tax revenue – increases with the degree of tax shifting. 15 This concept has been used by Laffer to argue that tax increases that are too high will reduce excise tax revenues (the so-called Laffer curve). For a detailed discussion on the Laffer curve, see section 4.4 in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 49 Taxation serves as an instrument for achieving both fiscal and public health objectives. If, after successful tobacco control interventions, prices reach levels where demand becomes elastic, the tax base is still most likely to be inelastic due to tax undershift- ing, since overshifting is not a good pricing policy when demand is elastic. In other words, a tax rate increase – in combination with non-price tobacco control measures that make consumers more sensitive to price (tax) increases – leads to decelerating but still positive marginal revenues. For an example of a tobacco taxation success story, see Box 2.2. For further details of countries’ experiences with tax increases and their impact on revenues, see section 4.4 in Chapter 4. KEY TAKEAWAY 15 Policy-makers’ key policy tool to control demand is tax. Therefore, it is essential they assess not only the impact of price on demand but, more appropriately, the impact of tax on demand: this is the tax base elasticity. The tax base elasticity is essentially determined by (1) the price elasticity of demand, (2) the degree to which the industry will pass the tax on to the retail price and (3) the tax as a share of the retail price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. Box 2.2 A tobacco taxation success story: Turkey Turkey is an example of a country that has been increasing taxes regularly and sig- nificantly over a relatively short period of time and has reaped the benefits of this policy. As shown in Fig. 2.11, the excise tax per pack of cigarettes more than doubled in real terms over 10 years, with the real price almost doubling as well. In parallel, tobacco excise revenues increased by 67% and cigarette sales decreased by 20%. Since the beginning of the country’s Health Transformation Program in 2003, Turkey has successfully increased public health spending and collected more tobacco tax revenue. According to the latest available figures, in 2015, tobacco tax revenue was equivalent to 42% of the country’s public health expenditure and 1.5% of GDP (88). 50 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.11 Tobacco excise revenue and consumption in Turkey (base year 2008), 2008–2018 Sources: Reference 1 for the price of the most-sold brand, Ministry of Finance for the sales and revenue data and IMF world economic outlook, April 2020. See https://www.imf.org/en/Publications/WEO/weo- database/2020/April for the adjustment for inflation. Impact on affordability While price increases clearly have an impact on consumption, when the effects of increasing per capita income of a population are not considered, the price impact may not be as strong as expected. Increases in a population’s income also increase its purchasing power. And, as indicated earlier, tobacco products generally behave like a normal good. Consequently, as income increases, it is expected that tobacco consumption will increase as well. To mitigate this effect, price increases (following tax increases) need to be greater than increases in income. This is where the concept of affordability comes in. Affordability examines the effects of both increasing prices and increasing incomes on consumer behaviour. A common and easy way to calculate affordability, made popular by Blecher and van Walbeek (89), is to use the percentage of GDP per capita required to buy 2000 cigarettes (or 100 packs of 20 cigarettes) in a given year. An increase in this proportion over time will indicate that cigarettes are becoming less affordable and should lead to reductions in consumption. Changes in trends in affordability of cigarettes over time help policy-makers understand how prices are evolving 0 1 2 3 4 5 6 Tu rk is h lir as p er p ac k N um ber of sticks, 10 000 000 Turkish liras 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real excise tax amount Real price Real tobacco excise revenues Cigarettes, per capita sales 1 100 500 1 700 2 300 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 51 relative to a population’s ability to purchase cigarettes and enable them to revise their policies accordingly. Recent studies in India, for example, highlight the wide price differential between manufactured cigarettes and indigenous tobacco products such as bidis and chewing tobacco, as well as the propensity for these indigenous products to become more affordable over time due to favourable or more lenient tax policies towards them (90–91). Figure 2.12 shows the change in affordability of a pack of the most-sold brand of cigarettes by country income group between 2008 and 2018. During this time period, affordability declined in almost 70% of high-income countries, while it declined in slightly more than 35% of middle-income countries and only 26% of low-income countries. Fig. 2.12 Number of countries that have experienced a change in affordability of cigarettes between 2008–2018, by income level Note: Change in affordability was computed as the least squares rate of change in the per capita GDP required to purchase 2000 cigarettes of the most-sold brand in local currency in a given year. The trend rate of growth was computed for countries with four or more years of data, including 2018. Affordability was assessed as not having changed if the least squares trend in the per capita GDP required to purchase 2000 cigarettes over the period 2008–2018 was not statistically significant at the 5% level. Source: (1). KEY TAKEAWAY 16 From a health perspective, in addition to examining the impact of a tax increase on the levels of price, demand and revenues, policy-makers should consider a tax hike that will lead to prices rising more than increases in their population’s income; a tax increase should make tobacco products less affordable to consumers so that demand will be effectively reduced. Cigarettes became less aordable Cigarettes became more aordable Aordability did not change Could not be assessed due to insucient data High income Middle income Low income 39 36 9 13 5 7 37 23 6 13 2 5 52 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Projecting impact on consumption, smoking prevalence and lives saved The WHO interactive smoking projection and target-setting tool (WHO ISPT) The WHO ISPT enables national policy-makers and tobacco control experts to explore the potential impact of proposed tobacco control policies. It uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. The WHO ISPT provides projections of (1) tobacco smoking rates and (2) tobacco-smoking-related deaths in a country under different policy settings and for different time periods. It was designed to promote multisec- toral collaboration within countries by enabling experts from various ministries (for example, health, education, finance, national statistics), civil society, academia and media to explore options for medium- and long-term tobacco control planning together with WHO experts. Use of the WHO ISPT enables strong partnerships for policy change advocacy, program development and evaluation.16 In particular, it can help policy-makers in the Ministry of Finance assess the specific contribution of tax policies – within overall tobacco control policies – towards achieving specific targets in tobacco prevalence reduction. Projecting impact on excise revenue The WHO tobacco tax simulation model (WHO TaXSiM) The WHO TaXSiM is a simple but data-intensive Excel-based tool that helps policy- makers analyse their tobacco tax policy and assess the impact of any excise tax increase or change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Using detailed data about the market – including the majority of brands found in the market, their market share and price levels and the applicable tax – and assumptions about price elasticity of demand, the WHO TaXSiM predicts the impact of tax changes on consumer prices, consumption volume and tax revenues generated by each brand and market segment for the following year. The exercise can be done for multiple years.17 By exploring market data in detail, in addition to assessing the potential rev- enue impact of changes in excise tax, the WHO TaXSiM is a useful instrument for highlighting weaknesses and opportunities in an existing tax system and market. It can also encourage policy-makers to create administrative databases that can be periodically updated to monitor the dynamics of the cigarette market. 16 The WHO ISPT is not available publicly, but WHO will work directly with interested countries upon request to use it to produce data-to-action-type plans. 17 For more information about the methodology, see https://www.who.int/tobacco/economics/tax- sim_background.pdf, accessed 29 September 2020. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 53 KEY TAKEAWAY 17 Policy-makers can use available tools to simulate the impact of tax increases on prices, consumption and revenues, as well as smoking prevalence and lives saved. Recommended indicators to monitor tobacco taxation progress MPOWER WHO publishes a biannual RGTE, which monitors global progress in tobacco con- trol. In particular, the report focuses on the implementation of the policy package MPOWER, a set of proven demand-reduction measures in line with the key provi- sions of the WHO FCTC (1). While raising taxes on tobacco (component R) is proven to be the most effective and cost-effective policy to reduce tobacco use (4), implementing the entire MPOWER package at the best practice level will reinforce the impact of R. For example, as mentioned earlier in this chapter, banning promotional discounts as part of the E measure (enforce bans on tobacco advertising, promotion and sponsorship) will favour price increases following a tax increase. If all the MPOWER tobacco control measures except R were implemented at the best practice level, all else remaining constant, revenues would be expected to decline. Thus, in order to maintain revenue levels, it is important to raise excise taxes on tobacco products regularly to compen- sate for the decline in tobacco use from the other four tobacco control measures. Tax share The main indicator in the R policy in the RGTE (1) is the total share of indirect taxes in the retail price of the most-sold brand of cigarettes.18 Countries whose most-sold brand of cigarettes has a total tax that is equal to or greater than 75% of the retail price are considered to be at the highest level of achievement. While total taxes include excise taxes, VAT (or sales taxes), import duties (when applicable) and other indirect taxes (where applicable), it is preferable to focus on excise taxes, since they are the component that most influences the relative price of tobacco. The share of excise tax in the retail price can be extracted from the RGTE database.19 The 2010 WHO technical manual on tobacco tax administration recommended making excise taxes account for at least a 70% share of excise taxes in the retail price of tobacco products (47). 18 For more details about how this indicator was compiled, see Technical Note III of the RGTE 2019 (https://www.who.int/tobacco/global_report/Technical-Note-III.pdf?ua=1). 19 See taxes and retail price for a pack of 20 cigarettes, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.1-Taxes-and-retail-price-for-a-pack- of-20-cigarette-most-sold-brand.xls?ua=1, accessed 29 September 2020). 54 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Affordability As discussed previously, the share of tax in the retail price is not enough to ensure that a policy will be successful in reducing demand. Any tax increase should lead to an increase in price that will effectively discourage consumption. While global trends indicate that a high tax share is positively correlated with a high price level (see Fig. 2.2 in section 2.1.2), this may not necessarily apply to a particular country; a tax share can be high, while at the same time tobacco products remain afford- able. For this reason, it is important to monitor not only tax increases but also whether those increases led to a price increase that is greater than income increases. As described in section 2.2.3 of this chapter, a common indicator is the percentage of GDP per capita required to buy 100 packs of 20 cigarettes in a given year.20 Other indicators As discussed in detail in sections 2.2.1 and 2.2.2, a good tax structure can make a tax policy more effective in increasing prices and decreasing affordability of tobacco products. Indicators can include whether a uniform excise is applied, whether it is a specific excise and whether it is adjusted regularly for inflation. A number of such indicators are also monitored through the RGTE and can be downloaded online.21 A tobacco tax indicator compiled in 2020 combines the various elements that form a good tobacco tax policy. The Tobacconomics Cigarette Tax Scorecard (92) rates a country’s tobacco tax policy performance based on best practices. The four components that determine the level of performance are (1) cigarette price (in PPP), (2) changes in the affordability of cigarettes over time, (3) the share of taxes (total and excise) in retail cigarette prices and (4) the structure of cigarette taxes (i.e. whether excise is applied; whether it is uniform or tiered; whether excise is specific, ad valorem or mixed; and, for the ad valorem component, if the tax is applied on the retail price and if there is a minimum specific excise and, for the specific component, if tax is automatically adjusted upwards). Each of the four components is given a score, using a five-point index, with the total score reflecting an average of the four component scores. The closer the total score is to 5, the bet- ter the tobacco tax policy performance is in a given country. While this published scorecard is currently applied only on cigarettes, it can be easily applied on other tobacco products, provided the needed data are available. 20 This indicator has also been compiled in the RGTE; see (https://www.who.int/tobacco/global_report/ Table-9.6-Affordability.xls?ua=1, accessed 29 September 2020). 21 See supplementary information on taxation, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.5-Supplementary-information-on-taxation. xls?ua=1, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 55 KEY TAKEAWAY 18 Tobacco taxation works best if implemented as part of a comprehensive MPOWER package. MPOWER is an overall indicator that incorporates all the key demand-side tobacco control measures. KEY TAKEAWAY 19 In addition to assessing the potential impact of a tax increase, policy-makers need to monitor progress over time. The share of the tax in the retail price is an indicator of progress. However, it is important to remember that an effective tax increase must translate into higher prices in order to make tobacco products less affordable. Combining all the components of a good tax policy into one scorecard can also be useful for assessing tobacco tax policy as a whole. 2.3 DOMESTIC AND REGIONAL POLICY INTEGRATION While it is essential to design tobacco tax policies with the utmost consideration of all the aforementioned factors, it is also important to consider how external factors can impact or even impede public health policy objectives. As Chapter 3 explains, cooperation among the various agencies that are directly involved in tax administration, collection and enforcement is important for effective and efficient tax policy implementation. But in the design phase, it is also essential to engage with agencies and other policy-makers that are not directly involved with taxation. Domestically, coordination is required to ensure that policies in non-health sectors do not negatively impact or even counteract tobacco control initiatives. For countries that are part of a regional bloc, harmonization of tobacco taxation is essential to protect the single market – as well as the health of the population – and to prevent tax revenue erosion, tax avoidance and tax evasion. 2.3.1 INTERSECTORAL COOPERATION ON DOMESTIC POLICY Domestic policies in agriculture, industry, trade, finance and labour have the po- tential to create or support incentives at different stages of tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. For example, subsidies provided to farmers or manufacturers involved in growing or processing tobacco can reduce prices and incentivize continued participation or even increase development in these areas, which is counterproductive to the goals of making tobacco products less affordable and reducing tobacco consumption. Multisectoral integration and policy coherence are needed at the country level to ensure that public policies and interventions in non-health sectors do not act against the intended public health impact of tobacco control and taxation. 56 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 20 Greater policy coherence in agriculture, industry, trade, finance and labour should be promoted to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. 2.3.2 REGIONAL TOBACCO TAX HARMONIZATION Policy integration is driven by the recognition that cooperation on domestic policies can substantially increase the gains from forming a regional bloc. Harmonization is desirable and may be necessary in certain areas with spillover effects, such as tax policy, the possibility of a so-called “race to the bottom” or threats to public health. Harmonization could be as simple as setting minimum standards and requirements based on global norms and best practices. Harmonization of tobacco taxation is required to ensure the establishment and proper functioning of a single market, prevent tax revenue erosion, prevent tax avoidance and tax evasion and protect people. When barriers to trade between countries are removed, harmonized tax rates support the single market because they improve the ability of consumers, producers and investors to make decisions that are not distorted by taxation but reflect real opportunity costs. Tax competition – where countries simply undercut each other’s tax rate – could prevent governments from raising sufficient funds to pursue social policy. To avoid such a race to the bottom, countries can establish minimum tax rates within the customs union (93). Even if tax competition is not present, when substantial tax differences exist in neighbouring countries, there is a clear incentive to trade across borders in order to reduce tax payments legally or illegally. The experiences of established regional economic communities offer important policy lessons, not only in terms of the general integration process but also for the process and extent of tax policy coordination. The EU implemented a successful regional tax harmonization scheme. Over the years, the focus in harmonization of tobacco taxes has broadened from the elimination of tax obstacles to the fight against harmful tax competition, tax avoid- ance and tax evasion and, more recently, to public health protection. Naturally, addressing these issues requires increasing convergence in fiscal policy and tax administration. Although price differentials still exist, setting a minimum on the share of taxes in the final price of tobacco products as well as a minimum excise tax has helped countries reach some level of harmonization. The EU experience confirms that both a declining tobacco consumption trend and stable revenues can be achieved with harmonized minimum excise rates (94). Moreover, the harmonization process has CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 57 offered opportunities for the enforcement authorities (tax or customs) to obtain information that can be used in the fight against fraud and tax evasion. On the other hand, the experience of harmonization efforts in the West African Economic and Monetary Union (WAEMU) shows how the absence of a supranational body (like the EU) or a hegemonic member state (see the SACU example below) can slow down policy integration that would benefit all member countries (95). The eight countries of the WAEMU are bound by a Tax Directive22 that requires them to impose an ad valorem excise on the CIF value or producer price of tobacco products, which is subject to under-declaration and is difficult to ascertain. Additionally, a maximum excise rate is imposed, and some members apply additional taxes to deal with this constraint. The Directive was revised in 2017 (96), but unfortunately the tax structure remains the same, and the maximum rate was not removed but rather has been increased. The Southern African Customs Union (SACU), which has five member countries, is the oldest existing customs union, established in 1910. Thanks to the hegemonic lead of South Africa, a country with a sophisticated administration system and an aggressive tobacco tax policy, SACU adopted a well-integrated tax policy that has benefited all its members (95). The GCC, established in 1981, is a regional intergovernmental political and economic union consisting of six states of the Persian Gulf. Home to one fifth of the global oil supply (97), the GCC has never relied on taxation as a source of revenue; no direct or indirect taxes were applicable in the region. Although there was no excise on tobacco products, as a customs union, the GCC countries have a common external tariff. This common tariff includes harmonized rates but also a harmonized structure. The import duty is 100% of the CIF value of tobacco products imported in the region, with a minimum tax per quantity imported. However, in recent years, to reduce their dependence on income from oil, GCC countries have considered diversifying their sources of income, including by de- veloping reliance on indirect taxes such as excise and VAT. In 2015, a decision was adopted at the 36th GCC summit meeting to implement selective taxes on all imported tobacco products and cultivated raw tobacco grown domestically (GCC Decision number 963/1). A follow-up decision in December 2016 formally agreed to the introduction of an excise tax on tobacco and other products such as sugary and energy drinks, as well as special goods (alcohol and pork meat), in all GCC countries. The decisions at the national level to implement this subregional decision came into force gradually in all GCC countries, starting with Saudi Arabia, which began 22 Directive No. 03/98/CM/WAEMU on the harmonization of Member States’ legislation of excise duties was adopted 22 December 1998. It was amended by Directive No. 03/2009/CM/WAEMU of 27 March 2009 with the objective of harmonizing excise duties within WAEMU. 58 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation in June 2017, followed by Bahrain and the United Arab Emirates (UAE) later that year, Qatar in 2018 and Oman in 2019 (1). Only Kuwait has yet to adapt its national laws accordingly. The excise introduced by the GCC countries has a structure somewhat similar to the import duty on tobacco products: the rate is 100%, but the base was changed from the CIF value to the retail price excluding taxes. The introduction of the excise led to large increases in the price of the most- sold brand of cigarettes in member countries between 2016 and 2018 – by 33% in Bahrain, more than 80% in the UAE and more than 100% in Saudi Arabia (1). In federations such as Canada and the United States – where the central govern- ment has real taxing power and some financial and regulatory control over the states or provinces – tobacco taxes are not harmonized (98–99). Even though there are significant interjurisdictional differences in taxes and prices, and tax harmonization holds great potential to reduce the scope of illicit transactions in the tobacco market, there is little evidence that Canadian provinces or individual states in the United States are interested in tobacco tax harmonization. Tax harmonization is most relevant in the context of further economic integration within a group of countries that are already part of a customs union, but it needs to be planned well to be effective. Discrepancies in law interpretation and a lack of standardization of tobacco product definitions and tax base lead to suboptimal situations. Tax rate alignment, or setting minimum rates, should come after tax structure alignment. It is important that governments support the move towards harmonization and are committed to dedicating enough financial resources and skilled personnel to oversee the entire process. KEY TAKEAWAY 21 In the context of regional economic integration and ongoing discussions regarding the possibility of harmonizing tobacco excise taxation among member countries, the experiences of existing groups can be instructive. So far, only the EU, SACU, WAEMU and, more recently, the GCC have effectively implemented a harmonized approach to excise taxation of tobacco products. Lessons learned indicate that harmonization should be planned well and should not come at the expense of tobacco control. Setting a common minimum specific excise tax, adjusted over time, is the best approach. This ensures that taxes and prices are above a minimum level, encouraging equalization of price levels and at the same time reducing affordability across countries. On the other hand, agreeing on maximum tax rates is a bad policy. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 59 Countries that wish to raise their taxes further for revenue purposes, health concerns or both should be given the space to do so. Concerns about illicit trade provoked by higher tax rates are best dealt with by strong cooperation in administration and enforcement, information sharing and adoption of new technology with common or interoperable information systems. 2.4 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In recent years, awareness of tobacco risks and harms, implementation of tobacco control provisions – especially under the WHO FCTC – and tightening of regulations have resulted in declining sales of cigarettes, primarily in high-income economies. This has changed the dynamics of the tobacco market. In response to these effective tobacco control measures, the tobacco industry has diversified its business by promot- ing a new portfolio of products, which they claim to be technological innovations that supposedly reduce the harms and risks associated with conventional tobacco products, particularly cigarettes. So-called novel tobacco products have been promoted by the tobacco industry as “cleaner alternatives,” “safer alternatives” and “reduced harm/risk products” with no smoke and no ash. On the basis of these claims, they negotiate for less-restrictive regulatory environments within countries. Some of the new products are also mar- keted or promoted for smoking cessation, despite the evidence of this outcome being inconclusive. Where these products are not banned, one of the debates in the global health community concerns the issue of their regulation and taxation. 2.4.1 HEATED TOBACCO PRODUCTS (HTPs) HTPs are tobacco products that produce aerosols containing nicotine and toxic chemicals upon heating of the tobacco or activation of a device containing the tobacco. These aerosols are inhaled by users sucking on or smoking the device. They contain the highly addictive substance nicotine (found in tobacco) as well as non-tobacco additives and are often flavoured. The tobacco in HTPs may be in the form of specially designed cigarettes (e.g. so-called heat sticks or Neo sticks) or pods or plugs. These products include IQOS from PMI, Ploom TECH from Japan Tobacco International (JTI), glo from British American Tobacco (BAT) and PAX from PAX Labs. HTPs differ not only from con- ventional cigarettes but also from ENDS – some of which are called e-cigarettes – as ENDS do not contain tobacco but rather a nicotine solution (see next subsection). However, the boundaries between the different products are becoming increasingly 60 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N difficult to define, given the emergence of so-called hybrid tobacco products that contain both nicotine solution and tobacco. HTPs are currently available in more than 40 countries and are banned in fewer than 10 countries. Even in countries where they are regulated, there is significant variation in the approaches taken to regulation. A variety of factors affect a country’s ability to control and regulate the use of HTPs, including national regulatory pow- ers, enforcement capacity regulatory frameworks, country capacity and tobacco industry interference (1). Most countries tax HTPs at a lower rate than cigarettes and on the kilogram of tobacco as a base when applying a specific or mixed excise (see Table 2.4). The use of such a base may be quite challenging for tax collection, especially because of the difficulty of checking the tobacco content in each stick. In the past, some countries taxed cigarettes per kilogram of tobacco, but today it is common practice to tax them per stick regardless of tobacco content. Table 2.4 Excise taxation of HTPs, first collected for July 2018 – updated for July 2020 OVERALL COMPARISON WITH CIGARETTES Type of excise Base unit is kg, overall rate lower than cigarettes Base unit is sticks, rate is the same as cigarettes Base unit is sticks, rate is lower than cigarettes Other Specific excise Albania, Austria, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Greece, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Montenegro, Netherlands, New Zealand, North Macedonia, Romania, Russian Federation, Slovakia, Slovenia, Sweden, United Kingdom Azerbaijan a, Japan, Ukraine b Armenia, Hungary, Jordan, Italy c, Philippines, Republic of Korea d Montenegro e, Republic of Moldova f , Serbia g Ad valorem excise (base is retail price unless specified other- wise between brackets) Spain, Switzerland Saudi Arabia and United Arab Emirates (base is retail price exclusive of excise and VAT) Indonesia h CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 61 Mixed system (ad valorem compo- nent based on retail price unless specified other- wise between brackets) France, Germany, Poland, Portugal Colombia, Georgia, Israel and West Bank and Gaza Strip (ad valorem excise base is wholesale price) a The specific excise rate applied is the same as for imported cigarettes, higher than the rate applied to domestically produced cigarettes. b The rate is the same as the minimum excise on cigarettes per 1 000 pieces. Rate and structure were effective as of 1 January 2021. c The specific excise rate is defined as 25% of the excise tax on cigarettes based on an equivalency used between cigarettes and HTPs. There are planned increases of this proportion to 40% by 2023. d In 2020 the specific excise rate was only 11% lower than cigarettes. e The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. f Specific excise rate is higher than for cigarettes but, unlike HTPs, cigarettes also face an ad valorem excise. Overall effect of excise is a slightly lower for HTPs. g The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. There is a planned phased increase of this proportion aiming equalization with cigarettes by 2025. h While cigarettes face a specific excise tax rate, HTPs face an ad valorem rate, the highest rate as defined by law, on the basis of a pre-defined minimum price. Sources: (1, 100, 101, WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021 and the Campaign for Tobacco-Free Kids website on Taxation and Price for Heated Tobacco Products https://www.tobaccofreekids.org/what-we-do/global/taxation-price/staging-tax-gap). A study by Liber (102) compared prices of HTPs and cigarettes in 34 countries and showed that while taxes have been systematically lower for HTPs than for cigarettes, prices were higher in half of the countries surveyed. KEY TAKEAWAY 22 HTPs, when taxed, are usually taxed lower than cigarettes, although they generally seem to be priced higher than cigarettes. It is important to remember that HTPs are tobacco products, and the same provisions that apply to tobacco products should apply to them as well. This is articulated in WHO’s information sheet on HTPs (103), which provides guidance on how these products should be regulated, as well as Decision FCTC/COP8(22) for novel and emerging tobacco products. Moreover, MPOWER measures, which help WHO Member States to implement the demand-reduction articles of the WHO FCTC, are applicable to HTPs, in particular, Article 6 for taxation. Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. Furthermore, HTPs contain chemicals not found in cigarette smoke, the health effects of which are not yet known. Independent assessment of industry data demonstrates that more than 20 harmful and potentially harmful chemicals 62 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N are significantly higher in HTP emissions than in cigarette smoke (104). Therefore, there is a need to learn more about these products and the health impacts of their emissions, as well as the impacts of exposure to these emissions. KEY TAKEAWAY 23 Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. From both public health and tax administration perspectives, HTPs should be taxed at the same level and in the same way as tobacco cigarettes. Some countries have already adopted this approach and are taxing HTPs at the same rate per stick as cigarettes (Azerbaijan, Colombia, Georgia, Israel, Japan, Ukraine and West Bank and Gaza Strip). Saudi Arabia and the UAE, which have recently introduced an excise tax on tobacco products as part of the GCC, are now applying the same import duty rate and excise tax structure for cigarettes and HTPs. Continuing developments in technology and changes in products have led to a recommendation to tax HTPs per unit. The definition of unit may vary by product within the HTP category. For example, one unit of IQOS is one heat stick, for Glo it is one Neo Stick and for Ploom TECH it is one tobacco pod. Governments will need to determine the exact definition of a unit for each product allowed on the market. The potential complexity of the market strongly supports limiting the types of HTPs allowed in a country and setting strict regulations to standardize the products as much as possible. Countries can also consider taxing the devices used to consume HTPs, i.e. the holder and the charger (see product description in Annex 3.1). KEY TAKEAWAY 24 HTPs are tobacco products, and they need to be treated as such. Where they are not banned, HTPs need to be strictly regulated and taxed. The recommendation is to tax them at the same level as cigarettes on a per-unit basis. Countries can also consider taxing the devices used for HTP consumption. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 63 2.4.2 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS)23 Products like ENDS and ENNDS have evolved rapidly over the past decade. ENDS heat a solution (e-liquid) containing nicotine, but not tobacco, and other chemicals that may be toxic to people’s health to create an aerosol, which is inhaled by the user. Examples of ENDS include Juul from Juul Labs, Vype from BAT and blu from Imperial Brands (1). Electronic non-nicotine delivery systems (ENNDS) are essentially the same as ENDS, but the e-liquid used generally does not contain nicotine. Upon testing, however, many so-called zero-nicotine solutions are found to contain nicotine (105–107). While generally considered as a single product class, ENDS products constitute a diverse group with potentially significant differences in the production of toxicants and delivery of nicotine. There are several coexisting types of devices for ENDS/ ENNDS on the market, including first-generation or so-called cigalikes, second- generation tank systems and even-larger third-generation or personal vaporizers. Collectively, they are also often referred to as e-cigarettes, vapes or vape pens. Other categories of ENDS include e-hookahs, e-pipes and e-cigars – hence, ENDS is an all-encompassing term for multiple product categories. Some of the products resemble their conventional tobacco counterparts – cigarettes, cigars, cigarillos, pipes or hookahs – while others are shaped more generically like pens, USB memory sticks or basic cylinders. Different forms of nicotine are also used in these ENDS, the most recent one being nicotine salts, which deliver high levels of nicotine (1). There are two types of ENDS/ENNDs products: open systems and closed systems. Open systems are devices that allow the user to buy e-liquids and fill their device with the mixtures they want (with no nicotine, different nicotine concentrations and/or flavours). Closed systems are products that come with a prefilled container (called a cartridge, pod or tank). For the past decade, divisive debates have been waged over the effectiveness of ENDS as smoking cessation aids – especially for tobacco users who are unable to give up the habit – as well as the possibility of ENDS playing a role in public health. However, the evidence remains inconclusive. Despite the tobacco industry and other related industries promoting these products as tools for quitting smoking, current evidence does not support their use as part of a population-based cessation strategy (108). Accordingly, the United States Surgeon General, in January 2020, concluded that 23 It is worth noting that ENDS are not tobacco products and not exactly new products – the technology has been around since the late 1980s (e.g. Premier, Eclipse and Accord). However, the recent generation of these products is new and has more or less piggybacked on the success of e-cigarettes. 64 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N E-cigarettes, a continually changing and heterogeneous group of products, are used in a variety of ways. Consequently, it is difficult to make generalizations about efficacy for cessation based on clinical trials involving a particular e- cigarette, and there is presently inadequate evidence to conclude that e-cigarettes, in general, increase smoking cessation (109). The evidence on the adverse health effects associated with use of ENDS is mounting, and when ENDS are used in combination with smoking – which is the practice of the majority of ENDS users (110) – the adverse health effects of two or more products are combined. However, there are insufficient data to understand the full breadth of these effects, as ENDS have not been on the market long enough for their long-term effects to be established. Nevertheless, the evidence is clear that the aerosols of the majority of ENDS and ENNDS, some of which are cancer causing chemicals. ENDS also contain nicotine, which is highly addictive. In addition, ENDS are associated with increased risk of cardiovascular diseases and lung disorders, as well as adverse effects on the developing fetus during pregnancy (108, 110). For adolescents, the use of nicotine can lead to dependence and may harm brain development. Use of ENDS could also lead to a new generation of nicotine and tobacco users, as seen in some countries, especially since these products are designed to appeal to young people. Although the specific level of risk associated with ENDS has not yet been determined conclusively, these products are undoubtedly harmful. Therefore they should be strictly regulated if allowed to be sold in domestic markets, and must be kept away from children. Taxation will be a key component of regulation, since it is an effective tool for influencing consumer behaviour. Some countries have taken the bold decision to completely ban these products. Approaches that have been taken range from partial to comprehensive bans, and ENDS/ENNDS products were banned in more than 30 countries in 2018.24 In other countries, they are regulated as, for example, consumer products, pharmaceutical products or tobacco products, or they are completely unregulated. WHO recom- mends that where ENDS/ENNDS are not banned, they should be regulated to achieve the following objectives: 1. prevent the initiation of ENDS/ENNDS by non-smokers and youth, with special attention to vulnerable groups; 2. minimize as much as possible potential health risks for ENDS/ENNDS users and protect non-users from exposure to their emissions; 3. prevent unproven health claims being made about ENDS/ENNDS; and 24 Data collected for the WHO RGTE 2019. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 65 4. protect tobacco control activities from all commercial and other vested interests related to ENDS/ENNDS, including the interests of the tobacco industry. ENDS/ENNDS present a risk to youth, who have taken up their use in high numbers in some countries, including Canada and the United States (1, 111). The Juul brand, for example, has quickly gained a significant e-cigarette market share in the United States (112–113). Its marketing and popularity have led the United States Food and Drug Administration (FDA) to raise serious concerns and to seek solutions to effectively prevent youth from taking up the use of ENDS/ENNDS (114). The city of San Francisco banned the sale of e-cigarettes in June 2019 (115). In addition to posing a risk for initiation by youth, ENDS can attract non- tobacco users or prevent current smokers from quitting. Taxation could play a role in preventing the uptake of these products, specifically among non-smokers, vulnerable groups, children and adolescents. KEY TAKEAWAY 25 The long-term health effects of ENDS/ENNDS products are still unknown, but they are clearly harmful to health. Furthermore, evidence on the effectiveness of ENDS products as a smoking- cessation aid remains inconclusive. Taxing these products could play a role in preventing their uptake, specifically among non-smokers, vulnerable groups, children and adolescents. Price elasticity of demand for ENDS products In the context of taxation, it is important to ask whether demand for ENDS is price-responsive. Preliminary evidence, although almost exclusively focused on e-cigarette data from the United States, indicates that this is the case: demand for e-cigarettes may be even more price-responsive than the demand for conventional cigarettes, so taxes can be used to deter initiation by never-users (116–123). Most of the studies of price elasticity of demand for ENDS products also demonstrate that e-cigarettes and conventional cigarettes are partial substitutes – that is, they show positive cross-price elasticity. The magnitude of the elasticity indicates the degree of substitutability between products: the higher its value, the closer the products are to being substitutes, with higher cigarette prices being associated with increased e-cigarette sales. Some of the studies also show a substitutability effect in the other direction, with increased prices for e-cigarettes leading to an increase in conven- tional cigarette use (117, 120). All of the studies show evidence of substitutability except for one (124), which differentiates between exclusive and dual users and shows no evidence of substitution between e-cigarettes and conventional cigarettes. 66 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The presence of concurrent (or dual) use – consumers using both conventional cigarettes and ENDS products – complicates results and highlights the need for more research in this area to better disentangle the different effects. Liber et al. (125) analysed sales prices in a sample of 45 countries and concluded that comparable units of conventional cigarettes cost less than disposable e-cigarettes. The units considered for pricing e-cigarettes included both the e-liquid and the rechargeable device. Taken alone, the price of e-liquids is on average much lower than that of cigarettes in high-income countries and the same in LMICs. The time needed to buy back a rechargeable device is estimated to be less than two weeks in most countries. One can argue that increasing price differentials by further increasing taxes on regular cigarettes could be effective in driving current smokers of regular cigarettes to e-cigarettes (126) as a potentially lower-risk alternative (127). However, the ef- fectiveness of ENDS as smoking cessation devices is still being debated; a study by Sweet et al. (128) shows that dual use of e-cigarettes as a potential tool for cessation was effective only in the short term. Moreover, significantly more smokers said they would quit if cigarette prices doubled and e-cigarettes were not available (122) or that they would never have become addicted to nicotine if e-cigarettes had not been so readily available (129). Once an e-cigarette user is addicted to nicotine, there is a risk of initiating traditional tobacco products use (130). In general, cessation can be better facilitated by governments via stronger implementation of the other tobacco control policies that have been proven effective at reducing use. KEY TAKEAWAY 26 Few studies are available on the price elasticity of ENDS products, and the available data come almost exclusively from the United States. These early studies indicate that demand for e-cigarettes will go down as the price of e-cigarettes increases. Generally, the results also show that cigarettes and e-cigarettes are partial substitutes, where an increase in cigarette price would increase the demand for e-cigarettes while reducing demand for cigarettes. But these results do not differentiate between people who are exclusive cigarettes or e-cigarette users and those who are users of both products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 67 Tax structure Different countries impose different tax structures on ENDS/ENNDS products25 (see details in Table 2.5). The Republic of Korea, for example, imposes a specific tax per millilitre of ENDS/ENNDS e-liquid (131), while Indonesia imposes an ad valorem tax on the retail price of the e-liquid; the maximum rate allowed by law for tobacco products (132). In the United States, there is no common way to tax e-cigarettes among the states that do tax them (133–134). The situation is similar in the EU, where new and emerging nicotine and tobacco products are not currently covered by the tobacco tax directive, and Member States may apply a national tax as they see fit under their own rules. All the EU countries that tax ENDS products apply a specific excise per millilitre of e-liquid. These different tax treatments have the potential to distort the functioning of the internal market. In February 2020, the European Commission concluded that the current provisions of the harmonized directive are no longer relevant for the taxation of ENDS and HTPs, and this is a source of concern from the internal market perspective (135). In June 2020, the Member States of the EU reiterated that it is urgent and necessary to upgrade the EU regulatory framework by harmonizing defini- tions and the tax treatment of novel products such as ENDS/ENNDS and HTPs (2). Table 2.5 Types of excises applied on ENDS/ENNDS products e-liquids globally and in individual states in the United States, as of July 2019 (updated as of July 2020 for all countries except the United States) TYPE OF EXCISE COUNTRIES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Albania, Kazakhstan, Kyrgyzstan, Pakistan, Portugal, Republic of Korea, Romania, Russian Federation, Slovenia, Sweden Azerbaijan, Cyprus, Estonia, Finland, Georgia, Greece, Hungary, Italyb, Latvia, Lithuania, Montenegro, Morocco, North Macedonia, Philippines, Serbia Ad valorem (% of retail price or import value) Bahraina Indonesia, Jordan, United Arab Emirates, Yemen TYPE OF EXCISE INDIVIDUAL STATES IN THE UNITED STATES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Delaware, Illinois (Chicago), Cook County, Louisiana, Ohio, Puerto Ricoc, Connecticutc Kansas, North Carolina, Washington, West Virginia, Wisconsin 25 The focus is on the e-liquid used for ENDS/ENNDS products. 68 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ad valorem (% of wholesale/ distributor price) Alaska (Juneau, Matanuska- Susitna Borough), California, Illinois, Maine, Maryland (Montgomery County), Minnesota, Nevada, Pennsylvania, Vermont, Washington DC, Virgin Islandsc New York Mixed New Jersey, New Mexico a Tax applied to e-shisha (or e-hookah) because e-cigarettes are banned in Bahrain. b Italy imposes differential rates for nicotine and non-nicotine containing liquids. c States in which it is unclear if only ENDS or both ENDS and ENNDS products are taxed with an excise. Sources: (1, 135, complementary data from Frank Chaloupka and WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021). Table 2.6 provides reference material on the pros and cons of different considerations for determining the tax structure and base of ENDS/ENNDS products e-liquids. Table 2.6 Excise tax options for ENDS/ENNDS products e-liquids TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Specific Volume of nicotine- containing e-liquid (regardless of concentration) 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only the volume needs to be determined 1. Difficult to compare if tax equivalencya with cigarettes is sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarettes and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed 5. May encourage more do-it- yourself (DIY) products where e-liquids are mixed by the users themselves, which increases the risk of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 69 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of e-liquid regardless of nicotine presence 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only volume needs to be determined 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Does not require laboratory capacity to detect the presence of nicotine in liquids 1. Difficult to compare if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed Volume of all e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL are affected the most; also reduces the price gap between different products 2. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 3. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 4. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 5. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to user 1. Difficult to compare with cigarettes if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed 70 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of nicotine- containing e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL affected the most; also reduces the price gap between different products 2. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as the nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 3. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 4. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to the user 1. Difficult to compare if tax equivalency with cigarettes is being sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids as there is no simple way currently available to determine whether the e-liquid contains nicotine; self-declarations by industry are not sufficient 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed Ad valorem Producer price/ CIF value of nicotine- containing e-liquid (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 71 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Producer price/ CIF value of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Retail price of nicotine- containing e-liquids (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 2. Requires capacity to monitor the market to assess market prices 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled in an e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risk of accidents, illness and death 72 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Retail price of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires capacity to monitor the market to assess retail prices 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Ad valorem with minimum specific or mixed Ad valorem + min specific 1. Volume of nicotine- containing e-liquids will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of nicotine- containing e-liquids will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up. b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products 1. Requires capacity to monitor the market to assess retail prices 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotin; (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Difficult to set a minimum specific excise amount/specific excise amount, especially if tax equivalency with cigarettes is sought 5. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 73 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Ad valorem + min specific 1. Volume of all e-liquids regardless of nicotine presence will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of all e-liquids regardless of nicotine presence will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 1. Requires capacity to monitor the market to assess retail prices 2. Difficult to set a minimum specific excise amount/specific excise amount, especially if equivalency with cigarettes is sought 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels a Tax equivalency can be measured in different ways: (1) in terms of tax burden (as % of the retail price) or (2) as the exact amount of tax for equivalent quantities (assuming an equivalency between a certain volume of e-liquid and a pack of cigarettes). b There is also the option to use the producer price/CIF value as a base for the ad valorem component, but it is a weaker option because the base is difficult to ascertain and therefore prone to undervaluation. Note: Table compiled following a WHO Expert Meeting on Taxation of Electronic Nicotine and Non-Nicotine Delivery Systems (ENDS/ENNDS), Geneva, Switzerland, 2–4 September 2019. There is currently a lack of evidence on the practical challenges being faced by countries favouring one approach over the other. Furthermore, such data are difficult to obtain because the nature of the market is constantly changing. However, a clear recommendation can be made with regard to which e-liquids to tax. As indicated in Table 2.5, some countries tax all e-liquids – whether or not they contain nicotine (ENDS and ENNDS products) – while some tax only nicotine- containing e-liquids (ENDS products). As shown in Table 2.6, there is evidence that in a number of instances, ENNDS products do contain some nicotine. Additionally, ENNDS products are not harmless (136–137). It is therefore recommended that all e-liquids be taxed for both ENDS and ENNDs products. The question of whether to employ differential taxation based on nicotine content seems reasonable from a health perspective, since nicotine is a toxic substance. However, this would likely create an additional burden for tax administrators as 74 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N they would need to determine the nicotine concentration of e-liquids on the market. Additionally, this may no longer be relevant, as advancements in technology indicate that other features of the product can influence nicotine delivery beyond the actual concentration of the e-liquid. It is now possible to increase nicotine delivery at low nicotine concentrations by increasing battery power (by reducing resistance or increasing voltage) (138). In terms of implementation, while most countries seem to have adopted a specific excise tax on ENDS/ENNDS e-liquids per millilitre, one benefit of implementing ad valorem taxation is that it seems relatively easier to regulate in the context of a large heterogeneity of products. However, it is essential that the tax be applied on the retail price value of the products, as this base is easier to ascertain than any other value that could be declared by the manufacturer. It is also important to add that regulation of the characteristics of ENDS/ENNDS products is essential, and it should be implemented along with any tax policy adopted. Regulations should include: 1. setting a maximum nicotine concentration per millilitre to safeguard public health, including reducing the risk of dependence, especially among youth; 2. setting a maximum volume for cartridges to reduce toxicants exposure and possibly limit use; 3. setting a maximum capacity for refill containers to reduce toxicants exposure and possibly limit use; 4. setting a maximum battery power to reduce the possibility of influencing nicotine and toxicant delivery; and 5. taxing nicotine regardless of its source (e.g. tobacco, eggplant, synthetic). Countries may choose to impose an excise tax on ENDS and ENNDS devices26 as well. The easiest type of tax would be an ad valorem tax based on the declared retail price. If countries choose not to impose an excise tax on these products, they should at least impose the regular VAT or sales tax rate. Imposing an excise tax on devices can be challenging from an administrative perspective, as all components need to be clearly defined and classified as devices for ENDS/ENNDS consumption. For example, if the device is assembled after importation and some parts may be used for other purposes than ENDS/ENNDS consumption, authorities may face a challenge in detecting and differentiating which component parts would be subject to excise tax and which would not. 26 See Annex 2.3 for an overview of elements of devices used in ENDS/ENNDS products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 75 KEY TAKEAWAY 27 There is currently not enough evidence to recommend one tax structure over another for ENDS/ENNDS products. However, it is clear that taxing the e-liquids used for consumption is key. The excise tax should be applied on all e-liquids, whether or not they contain nicotine. If the preferred type of excise tax is ad valorem, it should be applied to the retail price. Countries can consider taxing devices as well, but they need to adequately assess their administrative capacity to do so. Policy-makers need to be mindful of the diversity and rapid evolution of ENDS/ ENNDS products and adjust accordingly. Regulation must reflect this reality so that loopholes will not be exploited by the industry. For example, ENDS/ENNDs products include not only e-cigarettes, vapes and vape pens but also other categories such as e-hookahs, e-pipes and e-cigars. Lawmakers need to be clear about how ENDS/ENNDS products are defined so that subcategories do not fall under the radar when regulation comes into effect. Definitions will also be relevant when it comes to taxation. An unclear definition can lead to a seemingly contradictory situation, such as in Bahrain, where e-cigarettes are banned but e-hookahs are not.27 Finally, while policy-makers need to be mindful of the emergence of new products and must take appropriate actions to protect the health of their citizens, it is important to remember that the overwhelming share of nicotine consumption remains that of tobacco products, especially cigarettes. The total market value of ENDS/ENNDS and HTPs sales in 2018 was less than 2.2% of the total market value, while cigarette sales alone accounted for 91% of the same total market value (139–140). 2.5 CONCLUSIONS An overview of excise tax application globally reveals a broad variety of price and tax levels, as well as structures used for taxing tobacco products, in particular, cigarettes. Some trends, however, indicate that tax and price levels are higher among higher- income countries. The rate of taxes also matters: higher tax rates are correlated with higher prices, and higher prices change behaviour, which leads to a reduction in consumption. More countries are moving away from ad valorem taxes and towards either mixed or specific excise systems, and there are few countries that do not impose any excise tax on cigarettes. 27 In Bahrain, the Ministry of Production and Trade Decision 38 of 2013 banned e-cigarettes, while the official list of excisable products from the Ministry of Finance includes e-shishas (or e-hookahs), making them apt to be taxed and therefore considered legal. 76 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Significantly increasing the taxes and prices of tobacco products is the most effective and cost-effective policy to control tobacco use. Increased taxes – which are passed on to smokers as higher prices – reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face several challenges, ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues, as well as the perceived contribution of the tobacco sector to economic development. In designing tax policy, the tax structure adopted not only affects consumption overall, it also shapes the market structure. Ad valorem taxation incentivizes industry to set prices lower than specific taxation does. Evidence suggests that under a specific tax, the price gap between premium and lower-priced products is narrower, therefore reducing incentives for substitution to lower-priced products following a tax increase. However, as industry consolidates producers and widens its portfolio of products, new evidence indicates that the industry is introducing cheaper brands while increasing the price of its expensive brands, therefore, paradoxically, widening the price gap between its products. Evidence also suggests that prices are higher under a specific excise tax structure. Additionally, from a tax administration perspective, a specific tax is easier to imple- ment, since only the quantity produced needs to be ascertained rather than the value of the product. Another aspect of tax structure is the use of tiered taxation – that is, tax rates that vary on the basis of different product characteristics. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax. Tiered taxation encourages substitution from premium to cheaper brands, maintaining smoking prevalence and reducing the health im- pact of tax rate increases. In addition to leading to lower prices, tiered taxation is difficult to administer and creates opportunities for the tobacco industry to avoid and evade taxes. The design of a tax structure must also consider the base on which tax is applied. The choice of base should lead to the highest possible effect on price and revenue. For specific taxation, the tax base is the quantity. When the tax is ad valorem, the CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 77 choice of the tax base is important not only for health considerations, through its effect on consumption, but also for tax revenue generation, as well as industry profits. An ad valorem tax based on the producer price, or CIF value, gives tobacco manufacturers opportunities to reduce their tax liability, especially when they control the distribution system through related parties. The best practice in an ad valorem (or mixed) excise structure is to use the retail price as the tax base and introduce a minimum excise tax per pack. Other tax design considerations include the importance of using automatic adjust- ments and indexation to inflation and income growth for the specific excise tax in order to avoid erosion of the tax over time. Emerging evidence indicates that tobacco taxation does not always achieve the intended results, because the tobacco industry finds ways to circumvent it. Non- tax policies such as pricing regulation (in particular, minimum mark-ups or price floors/minimum prices) may be seen as a complementary approach to ensuring a high price level and discouraging consumption of tobacco products. So far, these policies have not proven to increase average prices. A price floor is likely to lead to increased industry profits, giving the industry greater funds for its marketing strategies (such as the introduction of new products), and lower tax revenues for governments. By reducing price competition, a price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifications). However, where powerful multinationals are operating in certain markets with presence in all market segments and with the capability to overshift a tax on some brands while undershifting the tax on others, or where price promotions cannot be banned, minimum price policies may help increase the effectiveness of tax increases. Other non-tax policies affecting price levels are those relating to promotional dis- counts for tobacco products and the sale of single sticks of cigarettes. Both should be completely banned. The ban of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Higher taxes are the most effective way to dissuade consumption, with the added benefit of raising money for the government – money that can be earmarked for health and education programs, rather than going as profits to the tobacco industry. Additionally, in order to make excise tax on tobacco products more effective in reduc- ing overall tobacco use and in line with the recommendation of the Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce 78 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N the demand for tobacco), all tobacco products need to be taxed in a comparable way; the focus should not be on cigarettes only. Tax choices and reforms have various and sometimes conflicting consequences for the market. For example, there might be a trade-off between quantity and variety or perceived quality implications. It is important for the government to recognize that firms respond strategically to changes in tax policy. Close monitoring of the market is necessary to form correct expectations about industry responses and enable estimates of the impact of a tax increase on consumption and tax revenue. To estimate the total effect of a tax increase on demand for tobacco products and tax revenue, it is important to use correct estimates of the own-price elasticity of demand, the cross-price elasticity and the income elasticity of demand. It is also important to use updated estimates of demand elasticities, as the environment within which consumers make decisions continues to change. For example, financial crises or successful tobacco control interventions can be expected to shift demand and change elasticity. Another key measure of the impact of tax policy is the tax base elasticity. Policy- makers need to be mindful of the three key components of tax base elasticity: (1) the price elasticity of demand of tobacco, (2) the share of the tax in the consumer price and (3) the degree of pass-through of the excise tax rate increase to consumer price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. It is important to acknowledge that if tax increases lead to increases in prices be- low concurrent increases in income levels, they will not be effective in reducing consumption, as tobacco remains a normal good in most countries. Policy-makers need to account for affordability when considering tax increases. They should ensure that tax increases are high enough to increase prices above income growth so that consumption goes down effectively. When designing tax policy and deciding on the right level to impose, policy-makers need to assess and project the impact of their policy decisions. Monitoring and evaluation are important. Tools for measuring impact can be very helpful, and several such tools exist. The WHO ISPT, for example, looks not only at the impact of tax policy but also at a set of tobacco control policies, and this enables national policy-makers and other tobacco control experts to explore the potential impact of CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 79 future tobacco control policies. The tool uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. More specific to tobacco tax policy, the WHO TaXSiM assesses the impact of any excise tax increase and change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the effectiveness of their policies and whether those policies have an impact on tobacco use over time. The implementation of the MPOWER package is one useful indicator for assessing tobacco control overall. Tobacco taxation works best if it is implemented as part of a comprehensive MPOWER package. The tax share in the retail price of a selected tobacco product is one indicator of the effectiveness of tax policy, but a more important one is affordability, that is, whether tax increases do lead to price increases that are above income and general price increases. A useful indicator to assess the performance of the tax policy overall is the Tobacconomics Cigarette Tax Scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). Domestic policies in agriculture, industry, trade, finance and labour all have the potential to create or support incentives at different stages in tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. Greater domestic policy coherence should be pro- moted across different sectors of the government to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. Differential tax structures and rates have the potential to distort the functioning of the internal market. Harmonization of tobacco taxation ensures the establishment and proper functioning of a single market; prevents tax revenue erosion, tax avoidance and tax evasion; and protects people’s health. In this context, tax competition, where countries simply undercut each other’s tax rate, might prevent governments from achieving their tobacco control objectives and raising sufficient funds to pursue public health policies. To avoid such a race to the bottom, countries can establish minimum tax rates on all tobacco products. A common high minimum specific excise tax is the best approach to ensure that taxes and prices are above a minimal level. 80 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N In recent years, the world has experienced the rise of new and emerging tobacco and nicotine products, including ENDS/ENNDS and HTPs, which the industry claims are safer than traditional tobacco products. The evidence so far suggests that these products could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of these products as well as initiation, cessation and switching of tobacco use behaviours among different socioeconomic groups, are not yet clear. Until more evidence for the claimed benefits of these tobacco products is available, caution should be taken in developing tax policy. Therefore, the current recommendation is for HTPs to be taxed at the same level as cigarettes on a per-unit basis regardless of tobacco content. In countries where they are not banned, ENDS/ENNDS products must be regulated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS products taxation. Nicotine- and non-nicotine-containing e-liquids should be taxed equally. If ad valorem excise is chosen as the structure, the base should be applied on the retail price. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While the evolution of new and emerging tobacco and nicotine delivery systems merits the attention of administrators and regulators, it is worth remembering that the bulk of the world’s tobacco consumption is still in conventional cigarettes. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 81 REFERENCES 1. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 2. Outcome of proceedings: Council conclusions concerning the structure and rates of excise duty applied to manufactured tobacco. Brussels: Council of the European Union; 2 June 2020 (8483/20; https://www.consilium.europa.eu/media/44235/st08483-en20.pdf, accessed 28 September 2020). 3. Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco: Article 7, paragraph 2. Brussels: OJEU. 2011;176:24–36 (https://eur-lex.europa. eu/LexUriServ/LexUriServ.do?uri=OJ:L:2011:176:0024:0036:EN:PDF, accessed 28 September 2020). 4. U.S. National Cancer Institute and World Health Organization.The economics of tobacco and tobacco control. NIH Publication No. 16-CA-8029A. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute; 2016 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/brp/tcrb/monographs/monograph-21, accessed 28 September 2020). 5. Drope J, Schluger N, Cahn Z, Drope J, Hamill S, Islami F, et al. The tobacco atlas. Atlanta: American Cancer Society and Vital Strategies; 2018 (https://tobaccoatlas.org/wp-content/uploads/2018/03/ TobaccoAtlas_6thEdition_LoRes_Rev0318.pdf, accessed 24 January 2021). 6. Sector reports, cigarettes. In: GlobalData [website]. London: John Carpenter House; 2019 (restricted access). 7. Vellios N, Ross H, Perucic AM. 2018 trends in cigarette demand and supply in Africa. PLOS ONE. 2018;13(8):e0202467 (https://doi.org/10.1371/journal.pone.0202467, accessed 28 September 2020). 8. Keen MJ. The balance between specific and ad valorem taxation. Fiscal Studies. 1998;19(1):1–37 (https://www.ifs.org.uk/fs/articles/fsmick1.pdf, accessed 24 January 2021). 9. Kay JA, Keen MJ. The structure of tobacco taxes in the European community. IFS report series. Institute for Fiscal Studies; 1982. 10. Kay JA, Keen MJ. How should commodities be taxed?: Market structure, product heterogeneity and the optimal structure of commodity taxes. Euro Econ Rev. 1983;23(3):339–58. 11. Kay JA, Keen MJ. Commodity taxation for maximum revenue. Public Finan Q. 1987;15(4):371–85. 12. Kay JA, Keen MJ. Product quality under specific and ad valorem taxation. Public Finan Q. 1991;19(2):238–47. 13. Delipalla S, Keen MJ. The comparison between ad valorem and specific taxation under imperfect competition. J Public Econ. 1992;49(3):351–67. 14. Suits DB, Musgrave RA. Ad valorem and unit taxes compared. Q J Econ. 1953;67(4):598–604. 15. Skeath SE, Trandel GA. A pareto comparison of ad valorem and unit taxes in noncompetitive environments. J Public Econ. 1994;53(1):53–71. 16. Myles G. Imperfect competition and the optimal combination of ad valorem and specific taxation. Int Tax Public Finance. 1994;3:29–44. 17. Effectiveness of tax and price policies for tobacco control. Geneva: WHO Press, World Health Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications. iarc.fr/Book-And-Report-Series/Iarc-Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And- Price-Policies-For-Tobacco-Control-2011, accessed 28 January 2021) 18. Bevan J. GCC taxation. Aylesbury (UK): Rothmans Exports Limited; 1992 (https://www.industrydocuments. ucsf.edu/docs/yxpy0198, accessed 29 September 2020). 19. Morton IT. Note from IT Morton to I Birks regarding agreement on specific taxation; 8 February 1992 (https://www.industrydocuments.ucsf.edu/docs/#id=nxpy0198, accessed 29 September 2020). 20. Galgut R. Oman. BAT (UK and Export) Limited; 16 April 1992 (https://www.industrydocuments.ucsf. edu/docs/xxpy0198, accessed 29 September 2020). 21. Common excise tax agreement of the states of the Gulf Cooperation Council [website]. Abu Dhabi: UAE Ministry of Finance; 2016 (https://www.mof.gov.ae/En/lawsAndPolitics/govLaws/Documents/ Common%20Excise%20Tax%20Agreement.pdf, accessed 29 September 2020). 82 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 22. Brock B, Choi K, Boyle RG, Moilanen M, Schillo B. Tobacco product prices before and after a statewide tobacco tax increase. Tob Control. 2016;25(2):166–73. 23. Hanson A, Sullivan R. The incidence of tobacco taxation: evidence from geographic micro-level data. Natl Tax J. 2009;62(4):677–98. 24. Delipalla S, O’Donnell O. Estimating tax incidence, market power and market conduct: the European cigarette industry. Int J Ind Organ. 2001;19(6):885–908. 25. Keeler TE, Hu TW, Barnett PG, Manning WG, Hai-Yen S. Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxation. J Health Econ. 1996;15:499–512. 26. Linegar DJ, van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2017;0:1–7 (https://tobaccocontrol.bmj.com/content/27/1/65, accessed17 February 2021). 27. Villanti AC, Collins LK, Niaura RS, Gagosian SY, Abrams DB. Menthol cigarettes and the public health standard: a systematic review. Public Health. 2017;17:983 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5747135/, accessed 24 January 2021). 28. Directive 2014/40/EU of the European Parliament and of the Council of 3 April 2014 on the approximation of the laws, regulations and administrative provisions of the member states concerning the manufacture, presentation and sale of tobacco and related products and repealing Directive 2001/37/EC: Article 7. Brussels: OJEU. 2014;57:1–38 (https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=OJ:L:2014:127:FULL&from=EN, accessed 24 January 2021). 29. Delipalla S, Keen MJ. Product quality and the optimal structure of commodity taxes. J Public Econ Theory. 2006;8:547–54. 30. Obara T, Tsugawa S. The welfare comparison of ad-valorem tax and specific tax with both quality and quantity choice of a consumer. Theor Econ Lett. 2017;7:1797–1813 (https://www.scirp.org/journal/ paperinformation.aspx?paperid=79862, accessed 24 January 2021). 31. Anderson S, de Palma A, Kreider B. Tax incidence in differentiated product oligopoly. Journal of Public Econ. 2001;81(2):173–92. 32. Razzolini L, Shughart WF, Tollison RD. On the third law of demand. Econ Inq. 2003;41:2. 33. Sobel RS, Garrett TA. Taxation and product quality: new evidence from generic cigarettes. J Polit Econ. 1997;105:880–7 (http://faculty.citadel.edu/sobel/All%20Pubs%20PDF/Generic%20Cigarettes. pdf, accessed 24 January 2021). 34. Nargis N, Fong GT, Chaloupka FJ, Li Q. The choice of discount brand cigarettes: a comparative analysis of International Tobacco Control (ITC) surveys in Canada and the United States (2002-2005). Tob Control. 2014;23:i86–i96 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/suppl_1/i86. full.pdf, accessed 24 January 2021). 35. Chaloupka FJ, Peck R, Tauras JA, Xu X, Yurekli A. Cigarette excise taxation: the impact of tax structure on prices, revenues and cigarette smoking. Cambridge: National Bureau of Economic Research; 2010 (NBER Working Paper 16287; https://www.nber.org/system/files/working_papers/w16287/w16287. pdf, accessed 24 January 2021). 36. Cigarettes in India, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 37. Gilmore AB, Tavakoly B, Taylor G, Reed H. Understanding tobacco industry pricing strategy and whether it undermines tobacco tax policy: the example of the UK cigarette market. Addiction. 2013;108(7):1317–26 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3746125/, accessed 24 January 2021). 38. Marsh L, Cameron C, Quigg R, Hoek J, Doscher C, McGree R, et al. The impact of an increase in excise tax on the retail price of tobacco in New Zealand. Tob Control. 2016;25(4):458–634. 39. Brock B, Schillo BA, Moilanen M. Tobacco industry marketing: an analysis of direct mail coupons and giveaways. Tob Control. 2015;24:505–8. 40. Chaloupka FJ, Warner KE. The economics of smoking. Handbook Health Econ. 2000;1:1539–1627. 41. Shang C, Chaloupka FJ, Fong G, Thompson M, O’Connor RJ. The association between tax structure and cigarette price variability: findings from the ITC Project. Tob Control. 2015;24:iii88–iii93 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/24/Suppl_3/iii88.full.pdf, accessed 24 January 2021) 42. Li Q, Hyland A, Fong GT, Jiang Y, Elton-Marshall T. Use of less expensive cigarettes in six cities in China: findings from the International Tobacco Control (ITC) China Survey. Tob Control. 2010;19(suppl CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 83 2):i63–i68 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/19/Suppl_2/i63.full.pdf, accessed 24 January 2021). 43. Hyland A, Laux FL, Higbee C, Hastings G, Ross H, Chaloupka FJ, et al. Cigarette purchase patterns in four countries and the relationship with cessation: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15(suppl III):iii59–iii64 (https://tobaccocontrol.bmj.com/ content/tobaccocontrol/15/suppl_3/iii59.full.pdf, accessed 24 January 2021). 44. Tsai YW, Yang CL, Chen CS, Liu, TC, Chen PF. The effect of Taiwan’s tax induced increases in cigarette prices and brand switching and the consumption of cigarettes. Health Econ. 2005;14:627–41. 45. Cummings KM, Hyland A, Lewit E, Shopland D. Use of discount cigarettes by smokers in 20 communities in the United States, 1988-1993. Tob Control. 1997;6(suppl 2):S25–S30 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/6/suppl_2/S25.full.pdf, accessed 24 January 2021). 46. Ross H, Tesche J. Undermining government tax policies: common strategies employed by the tobacco industry in response to increases in tobacco taxes. Economics of Tobacco Control Project, School of Economics, University of Cape Town and Tobacconomics, Health Policy Center, Institute for Health Research and Policy. Chicago: University of Illinois at Chicago; 2016 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC5747349/, accessed 26 January 2021). 47. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 48. Hiscock R, Branston JR, McNeill A, Hitchman SC, Partos TR, Gilmore AB. Tobacco industry strategies undermine government tax policy: evidence from commercial data. Tob Control. 2017;0:1–10 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 49. Filippidis F, Laverty A, Hone T, Been JV, Millett C. Association of cigarette price differentials with infant mortality in 23 European Union countries. JAMA Pediatr. 2017;171(11):1100–1106 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5710372/#:~:text=Cigarette%20price%20increases%20across%20 23,priced%20cigarettes%20during%20this%20period, accessed 26 January 2021). 50. López-Nicolás A, Belén Cobacho M, Fernández E. The Spanish tobacco tax loopholes and their consequences. Tob Control. 2013;22(E1):e21. 51. Ballester L, Auchincloss A, Robinson L, Mayne S. Exploring impacts of taxes and hospitality bans on cigarette prices and smoking prevalence using a large dataset of cigarette prices at stores 2001–2011, USA. Int J Environ Res Public Health. 2017;14(3):e318 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5369154/, accessed 24 January 2021). 52. Liber A, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia survey. Tob Control. 2015;24:iii83-iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed on 26 January 2021). 53. Feighery E, Ribisl K, Schleicher N, Zellers L, Wellington N. How do minimum cigarette price laws affect cigarette prices at the retail level? Tob Control. 2005;14:80–85 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/14/2/80.full.pdf, accessed 26 January 2021). 54. Tynan MA, Ribisi KM, Loomis BR. Impact of cigarette minimum price laws on the retail price of cigarettes in the USA. Tob Control. 2012;22:e1. 55. Huang J, Chriqui JF, DeLong H, Mirza M, Diaz MC, Chaloupka FJ. Do state minimum markup/price laws work? Evidence from retail scanner data and TUS-CPS. Tob Control. 2016;25:i52–i59 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i52.full.pdf, accessed 26 Jan 2021). 56. Whitehead R, Brown L, Riches E, Rennick L, Armour G, McAteer J, et al. Strengths and limitations of tobacco taxation and pricing strategies. Edinburgh: NHS Health Scotland; 2018 (http://www. healthscotland.scot/media/1829/rapid-evidence-review-strengths-and-limitations-of-tobacco- taxation-and-pricing-strategies.pdf, accessed 27 January 2021). 57. Golden S, Farrelly M, Luke D, Ribisl K. Comparing projected impacts of cigarette floor price and excise tax policies on socioeconomic disparities in smoking. Tob Control. 2016;25:i60-i66 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i60.full.pdf, accessed 27 January 2021). 58. Doogan NJ, Wewers ME, Berman M. The impact of a federal cigarette minimum pack price policy on cigarette use in the USA. Tob Control. 2018;27(2):203–08. 84 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 59. Marynak K, Xu X, Wang X, Holmes CB, Tynan MA, Pechacek T. Estimating the impact of raising prices and eliminating discounts on cigarette smoking prevalence in the United States. Public Health Rep. 2016;131:536–43 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4937114/#:~:text=We%20found%20 that%20each%20price,smokers%20aged%20%E2%89%A526%20years., accessed 27 January 2021). 60. McLaughlin I, Pearson A, Laird-Metke E, Ribisl K. Reducing tobacco use and access through strengthened minimum price laws. Am J Public Health. 2014;104(10):1844–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4937114/#:~:text=We%20found%20that%20each%20price,smokers%20aged%20 %E2%89%A526%20years., accessed 27 January 2021). 61. European Commission v. French Republic, C-197/08, European Court of Justice 4 March 2019 (https:// eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A62008CC0197, accessed on 27 January 2021). 62. Tobacco price promotion: policy responses to industry price manipulation. Boston: Public Health and Tobacco Policy Center; October 2011 (http://tobaccopolicycenter.org/documents/Price%20 Promotion%20Overview%20FINAL.pdf, accessed 28 September 2020). 63. Golden SD, Holt Smith M, Feighery EC, Roeseler A, Rogers T, Ribisl KM. Beyond excise taxes: a systematic review of literature on non-tax policy approaches to raising tobacco product prices. Tob Control. 2016;25:377–85. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/4/377.full.pdf, accessed 27 January 2021). 64. Branston J, Gilmore S. The case for Ofsmoke: the potential for price cap regulation of tobacco to raise ₤500 million per year in the UK. Tob Control. 2014;23(1):45–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3812939/, accesed 27 January 2021). 65. von Lampe K, Kurti M, Johnson J. “I’m gonna get me a loosie” understanding single cigarette purchases by adult smokers in a disadvantaged section of New York City. Prev Med Rep. 2018;12:182–5. (https:// www.sciencedirect.com/science/article/pii/S2211335518302195, accessed 27 January 2021). 66. de Ojeda A, Barnoya J, Thrasher JF. Availability and costs of single cigarettes in Guatemala. Nicotine Tob Res. 2013;15(1):83–7 . 67. Lal P, Kumar R, Ray S, Sharma N, Bhattarcharya B, Mishra D, et al. The single cigarette economy in India - a back of the envelope survey to estimate its magnitude. Asian Pac J Cancer Prev. 2015;16(13):5579-82 (http://journal.waocp.org/article_31295_9ebd3981652934bbf5d78ec066b6a39d.pdf, accessed 27 January 2021). 68. Coady MH, Chan CA, Sacks R, Mbamalu IG, Kansagra SM. The impact of tax on purchasing behaviors among New York City smokers. Am J Public Health. 2013;103(6):e54-e60 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC3698745/, accessed 27 January 2021). 69. Cigarettes in Thailand, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 70. Cigars in Germany, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 71. Report from the Commission to the Council on Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2018 (COM(2018) 17 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/report_excise_duty_manufactured_ tobacco_12012018_en.pdf, accessed 28 September 2020). 72. Gallet CA, List JA. Cigarette demand: a meta-analysis of elasticities. Health Econ. 2003;12(10):821–35. 73. Guidelines for implementation of article 6 of the WHO FCTC. Geneva: World Health Organization; 2016 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 29 September 2020). 74. DeCicca P, Kenkel D, Liu F. Excise tax avoidance: the case of state cigarette taxes. J Health Econ. 2013;32(6):1130–41 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3855906/, accessed 27 January 2021). 75. Jawad M, Lee JT, Glantz S, Millett C. Price elasticity of demand of non-cigarette tobacco products: a systematic review and meta-analysis. Tob Control; 2018;27(6):689–95 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6056338/, accessed 27 January 2021). 76. John RM. Price elasticity estimates for tobacco products in India. Health Policy Plan. 2008;23(3):200–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2800993/, accessed on 27 January 2021). 77. Wasserman J, Manning WG, Newhouse JP, Winkler JD. The effects of excise taxes and regulations on CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 85 cigarette smoking. Journal of Health Econ, Elsevier. 1991;10(1):43–64. 78. Cheng K, Kenkel D. U.S. cigarette demand: 1944-2004. B.E. Journal of Econ Analysis & Policy. 2010;10(1):10.2202/1935-1682.2438 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3820165/, accessed 28 January 2021). 79. Bobak M, Jha P, Nguyen S, Jarvis M. Poverty and smoking. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000. 80. Peck, RM. Equity issues, tobacco, and the poor. Yurekli A, de Beyer J, editors. Washington (DC): World Bank Group; 2013. (Economics of Tobacco Toolkit, Tool 6; https://openknowledge.worldbank.org/ handle/10986/16265, accessed 28 January 2021). 81. Kenkel DS, Schmeiser MD, Urban C. Is smoking inferior?: evidence from variation in the earned income tax credit. J Hum Res. 2014;49:1094–1120. 82. Ruhm, CJ. Healthy living in hard times. Journal of Health Econ. 2005;24(2):341–63. 83. Ruhm, CJ. Are recessions good for health? The Quarterly Journal of Econ. 2000;115(2):617-50. 84. Tarantilis F, Athanasakis K, Zavras D, Vozikis A, Kyriopoulos I. Estimates of price and income elasticity in Greece. Greek debt crisis transforming cigarettes into a luxury good: an econometric approach. BMJ Open. 2015;5(1):e004748. (https://bmjopen.bmj.com/content/bmjopen/5/1/e004748.full.pdf, accessed 28 January 2021). 85. Kaiser M, Reutter M, Sousa-Poza A, Strohmayer K. Smoking and the business cycle: evidence from Germany. Discussion Paper Series. Bonn, Germany: IZA Institute of Labour Economics; 2017 (http:// ftp.iza.org/dp10953.pdf, accessed 28 September 2020). 86. Tingum EN, Mukong AK, Mdege N. The effects of price and non-price policies on cigarette consumption in South Africa. Tob Induc Dis. 2020;18(July):62 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC7398599/, accessed 28 January 2021). 87. Vardavas CI, Filippidis F, Agak, I, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the importance of earmarking revenue to health care and tobacco control. Tob Induc Dis. 2012;10(1):21 (http://www. tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the-revenue-to- health-care-and-tobacco,66110,0,2.html, accessed 28 January 2021). 88. Marquez P, Catinkaya V. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 89. Blecher EH, van Walbeek CP. An international analysis of cigarette affordability. Tob Control. 2004;13:339– 46 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/13/4/339.full.pdf, accessed 28 January 2021). 90. Goodchild M, Sinha P, Munish VG, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/handle/10665/331836, accessed 28 January 2021). 91. John RM, Dauchy E. Trends in affordability of tobacco products before and after the transition to GST in India. Tob Control Online First. 20 March 2020. 92. Chaloupka FJ, Drope J, Siu E, Vulovic V, Stoklosa M, Mirza M, et al. Tobacconomics cigarette tax scorecard. Chicago: Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/636/uic-tobacco-scorecard-report-final.pdf, accessed 28 January 2021). 93. Kanbur R, Keen MJ. Jeux sans frontieres: tax competition and tax coordination when countries differ in size. Am Econ Rev. 1993;83(4):877–92 (https://www.researchgate.net/publication/4901220_Jeux_ Sans_Frontieres_Tax_Competition_and_Tax_Coordination_When_Countries_Differ_in_Size, accessed 28 January 2021). 94. Bouw A. Tobacco taxation in the European Union: an overview (English). Washington (DC): World Bank Group; 2017 (http://documents.worldbank.org/curated/en/493581492415549898/Tobacco- taxation-in-the-European-Union-an-overview, accessed on 28 September 2020). 95. Blecher E, Drope J. The rewards, risks and challenges of regional tobacco tax harmonization. Tob Control. 2014;0:1–5. 96. Directive No.n° 01/2017/CM/UEMOA of 22 December 2017. Paris: Droit-Afrique; 2017 (https://www. 86 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N droit-afrique.com/uploads/UEMOA-Directive-2017-01-droits-accise-tabacs.pdf, accessed 29 September 2020). 97. Mirzoev T, Zhu L, YangY, Pescatori A, Matsumoto A, Callen T. The future of oil and fiscal sustainability in the GCC region. Middle East and Central Asia Research Department. Washington (DC): International Monetary Fund; 2020 (https://www.imf.org/-/media/Files/Publications/DP/2020/English/FOFSGCCEA. ashx, accessed 28 January 2021). 98. State cigarette excise tax rates & rankings. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/factsheets/0097.pdf, accessed 28 September 2020). 99. A map and table comparing cigarette prices in Canada (March 2018). Toronto: Non-Smokers’ Rights Association/Smoking and Health Action Foundation; 2018 (https://nsra-adnf.ca/wp-content/ uploads/2018/03/180302_map_and_table.pdf, accessed 4 February 2021). 100. European Commission excise duty rates on manufactured tobacco. Brussels: European Commission; 31 December 2017 (https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco- energy/excise-duties-tobacco_en, accessed 1 July 2018). 101. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2019 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study-on-the-tobacco-taxation-directive-2019_en.pdf, accessed 28 September 2020). 102. Liber AC. Heated tobacco products and combusted cigarettes: comparing global prices and taxes. Tob Control; 2018;28(6). 103. Heated tobacco products: information sheet - 2nd edition. Geneva: World Health Organization; 2020 (https://www.who.int/publications/i/item/WHO-HEP-HPR-2020.2, accessed 28 January 2021). 104. Kass Lempert L, Glantz S. Analysis of FDA’s IQOS marketing authorisation and its policy impacts. Tob Control 2020;0:1–9. 105. Chivers E, Janka M, Franklin P, Mullins M, Larcombe A. Nicotine and other potentially harmful compounds in “nicotine-free” e-cigarette liquids in Australia. Med J Aust. 2019;210(3):118–9 (https://onlinelibrary. wiley.com/doi/epdf/10.5694/mja2.12059, accessed 28 September 2020). 106. Wenzl T, Zelinkova Z, Regueiro J, Giri A. Scientific-technical support activities to DG TAXUD-C-2 on the option to include e-cigarettes within the scope of excisable goods for the Impact Assessment on a possible revision of Directive 2011/64/EU. Administrative Arrangement TAXUD/2015/DE/329. Luxembourg: Publications Office of the European Union; 2017 (https://publications.jrc.ec.europa.eu/ repository/bitstream/JRC105828/kjna29062enn.pdf, accessed 28 September 2020). 107. Lisko JG, Tran H, Stanfill SB, Blount BC, Watson CH. Chemical composition and evaluation of nicotine, tobacco alkaloids, pH, and selected flavors in e-cigarette cartridges and refill solutions. Nicotine Tob Res. 2015;17(10):1270–78. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573955/, accessed 28 January 2021). 108. Ghebreyesus TA. Progress in beating the tobacco epidemic. Lancet. 17 August 2019;394:548–49 (https:// www.thelancet.com/pb-assets/Lancet/pdfs/S0140673619317301.pdf?hss_channel=tw-27013292, accessed 29 September 2020). 109. Smoking cessation, a report of the Surgeon General. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2020. (https://www.hhs.gov/sites/default/ files/2020-cessation-sgr-executive-summary.pdf, accessed 8 October 2020). 110. E-cigarettes are harmful to health. Geneva: World Health Organization. 5 February 2020 (https://www. who.int/news-room/detail/05-02-2020-e-cigarettes-are-harmful-to-health, accessed 29 September 2020). 111. Shan L, Manzione LC, Azagba S. Psychological well-being and dual-use of cigarettes and e-cigarettes among high school students in Canada. J Affect Disord. 2020;265:357–63 112. King BA, Gammon DG, Marynak KL, Rogers T. Electronic cigarettes sales in the United States, 2013- 2017. JAMA. 2018;320(13):1379-80 (https://jamanetwork.com/journals/jama/fullarticle/2705175, accessed 28 January 2021). 113. LaVito A. Popular e-cigarette Juul’s sales have surged almost 800 percent over the past year [website]. CNBC. 2 July 2018 (https://www.cnbc.com/2018/07/02/juul-e-cigarette-sales-have-surged-over-the- past-year.html, accessed 28 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 87 114. Statement from FDA Commissioner Scott Gottlieb, M.D., on new enforcement actions and a youth tobacco prevention plan to stop youth use of, and access to, JUUL and other e-cigarettes. Washington (DC): US Food and Drug Administration; 23 April 2018 (www.fda.gov/NewsEvents/Newsroom/ PressAnnouncements/ucm605432.htm, accessed 29 September 2020). 115. Paul K. San Francisco becomes first US city to ban sale of e-cigarettes. The Guardian. 25 June 2019 (https://www.theguardian.com/us-news/2019/jun/25/san-francisco-e-cigarettes-vaping-juul-ban, accessed 29 September 2020). 116. Pesko MF, Courtemanche CJ, Maclean JC. The effects of traditional cigarette and e-cigarette tax rates on adult tobacco product use. J Risk Uncertain. 2020;60:229–258. 117. Cotti CD, Courtemanche CJ, Maclean JC, Nesson ET, Pesko MJ, Tefft N. The effects of e-cigarettes taxes on e-cigarettes prices and tobacco product sales: evidence from retail panel data. Cambridge: National Bureau of Economic Research; 2020 (NBER Working Paper 26724; https://www.nber.org/ system/files/working_papers/w26724/w26724.pdf, accessed 28 January 2021). 118. Saffer H, Dench DL, Grossman M, Dave DM. E-cigarettes and adult smoking: evidence from Minnesota. Cambridge: National Bureau of Economic Research; 2019 (NBER Working Paper 26589; https://www. nber.org/system/files/working_papers/w26589/w26589.pdf, accessed 28 January 2021). 119. Huang J, Gwarnicki C, Xu X, Caraballo RS, Wada R, Chaloupka FJ. A comprehensive examination of own- and cross-price elasticities of tobacco and nicotine replacement products in the U.S. Prev Med. 2018;117:107–14 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6195827/, accessed 28 September 2020). 120. Zheng Y, Zhen C, Dench D, Nonnemaker JM. U.S. demand for tobacco products in a system framework. Health Econ. 2017;26:1067–86. 121. Stoklosa M, Drope JM, Chaloupka FJ. Prices and e-cigarette demand: evidence from the European Union. Nicotine Tob Res. 2016;18(10):1973–80. 122. Grace RC, Kivell BM, Laugesen M. Estimating cross-price elasticity of e-cigarettes using a simulated demand procedure. Nicotine Tob Res. 2015;17(5):592–8. (https://www.researchgate.net/ publication/270291421_Estimating_Cross-Price_Elasticity_of_E-Cigarettes_Using_a_Simulated_ Demand_Procedure, accessed 28 January 2021). 123. Huang J, Tauras J, Chaloupka FJ. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob Control. 2014;23:iii41–iii47 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/23/suppl_3/iii41.full.pdf, accessed 17 February 2021). 124. Cheng KW, Chaloupka FJ, Shang C, Ngo A, Fong GF, Borland R, et al. Prices, use restrictions and electronic cigarette use—evidence from wave 1 (2016) US data of the ITC Four Country Smoking and Vaping Survey. Addiction. 2019;114(suppl 1):115–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC6661221/, accessed 28 January 2021). 125. Liber AC, Drope JM, Stoklosa M. Combustible cigarettes cost less to use than e-cigarettes: global evidence and tax policy implications. Tob Control. 2016;0:1–6 (https://www.researchgate.net/ publication/299473181_Combustible_cigarettes_cost_less_to_use_than_e-cigarettes_Global_ evidence_and_tax_policy_implications, accessed 28 January 2021). 126. Chaloupka FJ, Sweanor D, Warner KE. Differential taxes for differential risks — toward reduced harm from nicotine yielding products. N Engl Journal Med. 2016;373(7):594-7. 127. Goniewicz ML, Knysak J, Gawron M, Kosmider L, Sobczak A, Kurek J, et al. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob Control. 2014;23:133–9. 128. Sweet L, Brasky TM, Cooper S, Doogan N, Hinton A, Klein EG, et al. Quitting behaviors among dual cigarette and e-cigarette users and cigarette smokers enrolled in the tobacco user adult cohort. Nicotine Tob Res. 2019;21(3):278–84 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6379027/, accessed 28 September 2020). 129. Wills TA, Knight R, Williams RJ, Pagano I, Sargent JD. Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics. 2014;135(1):e43–e51 (https://www. ncbi.nlm.nih.gov/pmc/articles/PMC4279062/, accessed 28 January 2021). 130. Electronic Nicotine Delivery Systems: report by WHO. Geneva: World Health Organization; 2014. (FCTC/ COP/6/10 Rev.1; https://apps.who.int/gb/fctc/PDF/cop6/FCTC_COP6_10Rev1-en.pdf, accessed 28 January 2021) 88 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 131. WHO report on the global tobacco epidemic 2017: monitoring tobacco use and prevention policies. Geneva: World Health Organization; 2017 (https://www.who.int/tobacco/global_report/2017/en/, accessed 28 January 2021). 132. Regulation of the Minister of Finance of the Republic of Indonesia, number 146, PMK.010/2017 on excise tax rates of tobacco products. Jakarta: Ministry of Finance Indonesia; 2017. 133. Gourdet CK, Chriqui JF, Chaloupka FJ. A baseline understanding of state laws governing e-cigarettes. Tob Control. 2014;23:iii37-iii40 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4078672/, accessed 28 January 2021). 134. An act to ensure that nicotine products are taxed equally. 127th Maine Legislature: first regular session-2015. H.P. 670, Legislative Document No. 973. 2015 (https://www.mainelegislature.org/legis/ bills/getPDF.asp?paper=HP0670&item=1&snum=127&PID=, accessed 28 January 2021). 135. Evaluation of the Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco. Commission Staff Working Document. Brussels: European Commission; 2020 (SWD(2020) 32 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/10- 02-2020-tobacco-taxation-report.pdf, accessed 28 September 2020). 136. Electronic nicotine delivery systems and electronic non-nicotine delivery systems (ENDS/ENNDS): report by WHO. Geneva: World Health Organization; 2016. (FCTC/COP/7/11; https://www.who.int/ fctc/cop/cop7/FCTC_COP_7_11_EN.pdf, accessed 28 September 2020). 137. E-cigarette use among youth and young adults: a report of the Surgeon General. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2016 (https://e-cigarettes. surgeongeneral.gov/documents/2016_SGR_Full_Report_non-508.pdf, accessed 28 September 2020). 138. Wagener TL, Floyd EL, Stepanov I, Driskill LM, Frank SG, Meier E, et al. Have combustible cigarettes met their match? The nicotine delivery profiles and harmful constituent exposures of second-generation and third-generation electronic cigarette user. Tob Control 2017;26:e23–e28 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/26/e1/e23.full.pdf, accessed 28 September 2020). 139. Vapour products data. In: Euromonitor International [website]. London: Euromonitor International; 2019 (restricted access). 140. Global market data: segment insights. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 89 ANNEX 2.1 Table A2.1 Countries that apply different types of cigarette excise tax structures, 2018 SPECIFIC EXCISE (65 COUNTRIES) AD VALOREM EXCISE (42 COUNTRIES) MIXED EXCISE (63 COUNTRIES) NO EXCISE (15 COUNTRIES) Albania, Andorra, Australia, Azerbaijan, Barbados, Belarus, Belize, Bolivia (Plurinational State of ), Burundi, Canada, Cook Islands, Dominica, Ecuador, Eswatini, Fiji, Gambia, Honduras, Iceland, India, Indonesia, Jamaica, Japan, Jordan, Kazakhstan, Kenya, Kiribati, Kyrgyzstan, Lesotho, Malaysia, Mauritius, Mongolia, Mozambique, Myanmar, Namibia, Nepal, New Zealand, Nicaragua, Norway, Pakistan, Palau, Papua New Guinea, Peru, Philippines, Republic of Korea, Saint Lucia, Saint Vincent and the Grenadines, Samoa, Seychelles, Singapore, Solomon Islands, South Africa, Sri Lanka, Suriname, Tajikistan, Timor-Leste, Tonga, Trinidad and Tobago, Uganda, United Republic of Tanzania, USA, Uruguay, Uzbekistan, Vanuatu, Yemen, Zimbabwe Argentina, Armenia, Bahrain, Bangladesh, Benin, Burkina Faso, Cabo Verde, Cambodia, Cameroon, Chad, Comoros, Côte d’Ivoire, Cuba, Democratic Republic of the Congo, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Ghana, Grenada, Guatemala, Guinea-Bissau, Liberia, Madagascar, Mali, Mauritania, Niger, Panama, Paraguay, Saint Kitts and Nevis, Saudi Arabia, Senegal, Sierra Leone, Sudan, Syrian Arab Republic, Togo, Turkmenistan, Tuvalu, United Arab Emirates, Venezuela (Bolivarian Republic of ), Viet Nam, Zambia Algeria, Austria, Belgium, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Central African Republic, Chile, China, Colombia, Congo, Costa Rica, Croatia, Cyprus, Czechia, Denmark, Dominican Republic, Egypt, El Salvador, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iran (Islamic Republic of ), Ireland, Israel, Italy, Lao People’s Democratic Republic, Latvia, Lebanon, Lithuania, Luxembourg, Malta, Mexico, Montenegro, Morocco, Netherlands, Nigeria, North Macedonia, Poland, Portugal, Republic of Moldova, Romania, Russian Federation, Rwanda, Sao Tome and Principe, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Thailand, Tunisia, Turkey, Ukraine, United Kingdom, West Bank and Gaza Strip Afghanistan, Angola, Antigua and Barbuda, Democratic People’s Republic of Korea, Iraq, Kuwait, Libya, Maldives, Marshall Islands, Micronesia (Federated States of ), Nauru, Niue, Oman,a Qatar,a Somalia a This table shows the status of cigarette excise tax structures as of July 2018 and does not account for changes occurring after that date, in particular for the cases of Qatar and Oman, which introduced excise on tobacco in January 2019 and June 2019, respectively. Source: WHO RGTE. 90 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 2.2 THE ANALYTICS OF THE TAX BASE ELASTICITY Assume tax revenue R = tsQ or R = tvPQ , where Q is the quantity consumed, ts is the specific tax, tv is the ad valorem tax and P is the consumer price. The following equations can help to illustrate the different components of the tax base elasticity. Under a specific excise regime, change in revenue depends essentially on the change in consumption: where R is the tobacco tax revenue, is the specific excise tax and is the tobacco tax base elasticity. The tax base elasticity is made of: where ε, the price elasticity = , is the degree of pass-through of the specific excise tax rate increase on consumer price and is the tax-price ratio. Under an ad valorem excise regime, change in revenue depends essentially on the change in tobacco expenditure: where R is the tobacco tax revenue, tav is ad valorem excise tax and ηav is the tobacco tax base elasticity. The tax base elasticity here is made of: where is the degree of pass-through of the ad valorem excise tax rate increase on consumer price, is the tax-price ratio and ε the price elasticity = . ∂R = Q (1+ηs)∂ ts ηs = ε ts P ∂P ∂ts tav P ηav = (1 + ε) tav P ∂P ∂tav∂P ∂tav ∂P ∂ts ΔQ ΔP P Q ΔQ ΔP P Q ts P ∂R = P Q (1+ηav)∂tav CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 91 ANNEX 2.3 ELEMENTS OF THE DEVICES THAT MAKE UP ENDS/ENNDS PRODUCTS The main components of any ENDS/ENNDS kit include essentially: • USB charger (not a car charger) • Inbuilt battery Additionally, For open systems • Tanks (refillable containers) with removable atomizer (often sold bundled with atomizers) • Clearomizers/refillable pods (no removable atomizer) • E-liquid For closed systems • Disposable e-cigarettes: not rechargeable, thrown away after e-liquid is finished • Nondisposable e-cigarettes: – Pre-filled cartomizers (cartridges designed to go with the cigalike kit) – Pre-filled tank refills/pods (pods or cartridges designed to go with the prefilled tank/pod kits) Some definitions: • Atomizer: uses a heating element to vaporize the e-liquid • Cartomizer: combines the cartridge/tank and the atomizer • Clearomizer: same as cartomizer, uses different technology • Cartridge/tank/pod: container that includes the e-liquid In summary, ENDS/ENNDS product devices include the following: • USB charger (not a car charger) • Inbuilt battery • Disposable e-cigarettes • Atomizer • Cartomizer/clearomizer • Cartridge/tank/pod with or without atomizer • Pre-filled cartridge/tank/pod (for closed systems, includes e-liquid) • E-liquid (added in the cartridge/tank/pod in open systems) Source: ECigIntelligence, 2020. Information also obtained from vaping websites, including https://www. misthub.com/blogs/vape-tutorials/76788357-tutorial-atomizer-vs-cartomizer-vs-clearomizer, http:// www.bestclearomizer.com/clearomizer-vs-cartomizer-vs-atomizer/, https://wayofleaf.com/accessories/ vapes/atomizer-vs-clearomizer-vs-cartomizer, https://wayofleaf.com/accessories/vapes/atomizer-vs- clearomizer-vs-cartomizer, accessed 15 July 2020. 92 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 93 CHAPTER 3. Tobacco tax administration 3.1 INTRODUCTION Imposing excise taxes on tobacco products usually serves more than one purpose. Governments often find themselves balancing interests between financial and public health objectives. Both objectives can best be achieved by an efficient and effective competent authority with strong technical capacity to enforce and collect taxes. A competent authority is the agency, organization or department that is legally as- signed to complete a particular activity; in the case of administering tobacco taxes, the competent authority is often a tax administration, revenue authority, customs department or ministry of finance. Article 6 of the WHO FCTC (1), along with its guidelines (2), provides a solid foundation for sound tax administration. As stated under section 1.5 in the guidelines, tobacco tax systems should be efficient and effective. They should be structured to minimize the costs of compliance and administration, while ensuring that the desired level of tax revenue is raised and health objec- tives are achieved. Efficient and effective administration of tobacco tax systems enhances tax compliance and collection of tax revenue while reducing tax evasion and the risk of illicit trade. Efficiency in tax administration refers to minimizing the costs per unit of tax revenue collected. It is measured by comparing the resources used with the revenues gener- ated. Effectiveness in tax administration refers to a high level of compliance – also described as taxpayers meeting their obligations. Thus, an efficient and effective competent authority collects the tax at a minimum cost while ensuring conformity to the rules. Tobacco taxation is the single most effective tobacco control measure for re- ducing tobacco use and is best implemented as part of a comprehensive tobacco control plan (3). Illicit trade – including smuggling and illicit manufacturing – and tax avoidance undermine the effectiveness of tax policies and their objectives (4). The impact on illicit trade is often cited by opponents of tax increases, who argue that increasing taxes increases illicit trade. They contend that illicit trade can lead 94 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N to lower revenues for governments and lower prices. The challenge faced by the competent authorities is to ensure that due taxes are declared and collected on all tobacco products that are manufactured in and/or imported into its jurisdiction, while at the same time detecting tobacco products that are illegally manufactured in and/or imported into its jurisdiction, stopping such activity and prosecuting the responsible parties. This chapter describes the shared characteristics of good tax administrations, including best practices based on country experiences. It regularly refers to the WHO FCTC, and – given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade – draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (5). Any practice or conduct prohibited by law and related to production, shipment, receipt, possession, distribution, sale or purchase of tobacco products – including any practice or conduct intended to facilitate such activity – is considered as illicit trade (Article 1). The objective of the Protocol is to eliminate and prevent all forms of illicit trade in tobacco products. At the same time, the Protocol includes measures for tobacco tax administration based on international best practices, which makes it relevant for all countries, even those that are not Parties to it. The Protocol was adopted at the fifth session of the COP to the WHO FCTC in 2012 and entered into force on 25 September 2018. As indicated in the Preamble, it was developed in response to the increasing international illicit trade in tobacco products (5). The Protocol covers three main areas: (1) measures to control the supply chain (Part III); (2) measures dealing with offences, including sanctions (Part IV) and (3) international cooperation (Part V). Different provisions of the Protocol are discussed in detail throughout this chapter, and section 3.4 is devoted specifically to control and enforcement. 3.2 INSTITUTIONAL ARRANGEMENTS Competent authorities that collect taxes effectively in an efficient way share a number of attributes. The organizational structures of these authorities contain clearly defined roles, responsibilities and rules for coordination among relevant bodies. Moreover, competent authorities collect data regularly and manage information needed for assessing risks. The key to successful risk management is to share this information among relevant authorities both within a country and between countries. Effective and efficient competent authorities also regularly evaluate their performance and accountability according to key performance indicators to identify areas for improve- ment. These characteristics are discussed in greater detail in the following sections. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 95 3.2.1. CLEARLY DEFINED ROLES AND RESPONSIBILITIES OF COMPETENT AUTHORITIES The designation of competent authorities for the implementation and enforcement of tax laws – including clear definitions of the boundaries of authority among numerous agencies within a country – is essential for efficient collection of taxes. Areas where different agencies need to cooperate and share data must also be defined. Overlap of activities by different authorities leads to inefficient use of resources, whereas gaps create opportunities for fraud, leading to ineffective tax laws. The importance of clearly defined roles and responsibilities applies not only to tax authorities and customs but also to law enforcement agencies, including police and border control forces. The implementation and enforcement of taxation is organized differently in various countries. The most common structure separates customs and tax administration. The trend since the 1990s, however, has been to combine these functions into one agency, such as Her Majesty’s Revenue and Customs (HMRC) in the United King- dom, SUNAT in Peru1 and AFIP in Argentina.2 Several countries have increased coordination between tax and customs by creating a revenue secretariat and also implementing systems to share tax records as a single taxpayer account. Coordina- tion between tax policy and tax administration authorities has also increased. One can think of combining both into one department within the ministry of finance or ensure that tax administration authorities are consulted during the tax policy process. Some tasks, such as licensing, may be handled by other ministries such as the ministries of health, agriculture or trade. For example, the Ministry of Health of Brunei and the Health Science Authority of Singapore are responsible for the licensing of importers of tobacco products (6). In some federal countries, including Colombia and the United States, excise taxes – including tobacco taxes – are collected and enforced by local or state tax administrations. Other countries have organized the administration of national taxes by establishing a single unified revenue body. Particularly in larger economies, that body is often responsible for both direct and indirect taxes, including excise taxes, and reports to the ministry of finance. All the functions needed for effective and efficient tax administration are established within these bodies (7). No matter what the institutional arrangements may be, it is vital that the agencies cooperate and exchange information and that their competencies find their basis in law. More information on this topic is provided in section 3.2.2. 1 Law Decreto Supremo 061-2002-PCM - Disponen fusión por absorción de la Superintendencia Na- cional de Administración Tributaria – SUNAT con la Superintendencia Nacional de Aduanas - Aduanas [Supreme decree year 2002 about the merger between Tax and Customs Administration]. Lima: El Peruano, 12, July 2002 (in Spanish) (http://www.sunat.gob.pe/legislacion/sunat/ds061-2002-PCM.pdf, accessed 13 November 2020). 2 Administracion Federal de Ingresos Publicos, Decreto 618/1997 [Federal Administration of Public Revenue, Decree 618] (in Spanish) (http://servicios.infoleg.gob.ar/infolegInternet/an- exos/40000-44999/44432/norma.htm, accessed 13 November 2020). 96 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Many countries, however, have separate bodies for the collection of taxes and customs duties. A 2015 survey of 135 tax administrations worldwide found that only 36% of them were responsible for both tax administration and customs ad- ministration (8). In most countries, customs authorities are more likely to collect excise duties on imports, and in many countries, VAT or sales tax is collected jointly with tobacco tax, particularly for imported products. This simplifies controls and creates synergy by unifying common processes and procedures, resulting in cost savings for tax administrations and taxpayers. The involvement of multiple bodies in tax collection requires especially good collaboration and information-sharing to ensure efficient and effective collection of taxes and duties. KEY TAKEAWAY 1 Institutional arrangements with clearly defined roles and responsibilities – designed to prevent overlaps and voids – contribute to effective and efficient tax administration. 3.2.2. EFFECTIVE COORDINATION AMONG RELEVANT BODIES Coordination at the national level Coordination among relevant bodies is key to effective tobacco tax administration. This means not only clearly defined roles and responsibilities, as described in the previous section, but also coordination among the competent authority, customs and those responsible for formulating, analysing and implementing tax policy. Regardless of the institutional arrangements – whether the responsible parties are all within the ministry of finance or in separate government agencies – all parties need to cooperate and exchange information to optimize tax collection and enforcement of tax policy. In practice, this means that information should be shared among, for example, customs, local government units that issue licences and health authori- ties – particularly those that regulate the sale of tobacco products. For tax authorities, the most relevant information concerning excise taxes in- cludes the identity of taxpayers and those involved in the trade of tobacco (import and export data, licences, criminal records, tax returns, bank statements, etc.); the category, quantity, value and location of manufactured goods; and the movement of those goods until all taxes are paid. Legal impediments to obtaining this informa- tion – such as bank secrecy or privacy regulations – should be kept in mind, and where needed, exceptions for fiscal procedures should be incorporated into law. Seizure data are also a valuable source of information; more details on this are provided in section 3.4. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 97 Tax authorities should regularly coordinate with law enforcement agencies – such as the police and border control forces, depending on a country’s laws – to properly monitor tobacco-related activities and enforce the tax laws. Often, the competent tax authority and customs authorities work in close cooperation with anti-fraud teams.3 Coordination and sharing of information can be required in legislation or regulations to ensure a streamlined process and avoid confusion. This can be done on an ad hoc basis as needed or with formal planned exchanges of information and regular meetings. It is recommended that at least a legal basis for exchange or access to information among government bodies be established to prevent claims during legal procedures that evidence was obtained unlawfully. Some countries go beyond exchanging information and cooperation. In the Neth- erlands, for example, customs authorities not only carry out work for the Ministry of Finance, they also carry out non-fiscal tasks for seven other departments, including the Ministry of Agriculture, Nature and Food Quality; the Ministry of Justice and Security; and the Ministry of Foreign Affairs (9). These activities are often based on bilateral agreements between the Ministry of Finance and the other departments. In other countries, such as the United States and Canada, Customs and Border Protection are not part of the Ministry of Finance; they are part of the Department of Homeland Security in the United States and the Ministry of Public Safety and Emer- gency Preparedness in Canada. These agencies also carry out many non-fiscal tasks. Along with the implementation of new tobacco control and tax laws, several countries have also created high-level committees to ensure good coordination and implementation of the laws. Led by health and finance ministries, committees ensure coordination and fine-tuning to achieve desired results. Botswana, Chile, Colombia, Indonesia and Senegal, among other countries, have successfully started with coordina- tion, planning and monitoring of tobacco laws’ implementation through periodic com- mittee meetings. The committees usually include representatives from the ministries of health, finance, tax and customs, police, transport and, in some cases, education. Coordination across borders Effective approaches to control smuggling in tobacco products require interventions at the borders of jurisdictions and therefore must involve the border agencies. However, with the globalization of trade, there is a need for close coordination not only between tax and border control authorities but also between different jurisdictions. Recent cases have demonstrated that an absence of formal cooperation frameworks may expose a market to financial crime, including money-laundering and financing of terrorism (10). 3 See, for example, Focus on tax fraud. Customs administration of the Netherlands, tax and customs administration. 2017;2 (https://customsnl-insight.nl/article/309563676, accessed 3 October 2020). 98 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Accession to international cooperation agreements such as the Protocol, the Organisation for Economic Co-operation and Development (OECD) Multilateral Convention on Mutual Administrative Assistance in Tax Matters and other regional arrangements will contribute greatly to the effective exchange of information and cooperation among enforcement agencies. An effective exchange of market data and information from participating jurisdictions can prevent potential cross-border crimes and loss of domestic revenue. International cooperation reinforces domestic measures to stop illicit trade and raise much-needed revenues. Parties to the Protocol have a commitment to cooperate with one another and to share information to meet their obligations under the Protocol (Article 20). The Protocol itself is the legal instrument that allows Parties to cooperate and share information across borders. Authorities of governments that are not Parties to the Protocol or another coopera- tion agreement that represents a legal instrument to exchange information could conclude a mutual assistance agreement or exchange of information agreement to guide the procedures under which information exchange can take place effectively. The Revised Kyoto Convention of 2010 promulgated by the World Customs Organization (WCO) recommends that jurisdictions that enter into bilateral agree- ments require the other jurisdiction to provide pre-arrival information on goods bound for their customs territory. A survey of 87 WCO members in 2013 found that the vast majority of customs administrations had the legal authority to share information related to the supply chain of tobacco products with other administrations (11). Some economic blocs have also established harmonized legislation applying to all of their Member States to provide administrative cooperation to efficiently cooperate on tax matters (12). Coordination can include the establishment of a special agency to ensure the safety and proper functioning of external borders, such as the European Border and Coast Guard Agency, also known as Frontex (from the French frontières extérieures, “external borders”). In some of the Frontex-led operations, EU and non-EU countries cooperate together with international organizations to target cross-border crime, including the smuggling of cigarettes and raw tobacco (13). Criminals who engage in illicit trade of tobacco products are usually also en- gaged in related criminal activities such as bribery, money laundering, corruption, obstruction of justice and even financing of terrorist organizations (14). A number of international treaties provide the legal framework for addressing such conduct through mechanisms that tackle illicit trade from a criminal justice perspective, such as the United Nations Convention against Transnational Organized Crime, the United Nations Convention against Corruption and the International Convention for the Suppression of the Financing of Terrorism. Table 3.1 summarizes the types of structures available for such coordination. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 99 Table 3.1 Structures for coordinating mechanisms TYPE OF COORDINATION BASIS INVOLVED ACTORS National coordination Agreements with a basis in law between national agencies Customs authorities, ministries of finance and those responsible for formulating, analysing and implementing tax policy; law enforcement agencies, such as police and border control forces; and anti-fraud teams Agreements between ministries or a basis in law or regulation on the establishment of high- level committees Ministry of health, finance, revenue, justice, transport and sometimes education and enforcement entities such as customs and police Bilateral coordination Bilateral cooperation agreements National governments Regional coordination Regional arrangements such as • Harmonized legislation applying to all Member States of an economic bloc to provide administrative cooperation in taxation to efficiently cooperate on tax matters • Regulation to jointly establish a special agency to ensure the safety and functioning of external borders EU Member States, the European Border and Coast Guard Agency (Frontex), customs, law and border enforcement agencies International coordination International treaties or conventions such as • The Protocol • OECD multilateral Convention on Mutual Administrative Assistance in Tax Matters • United Nations Convention against Transnational Organized Crime • United Nations Convention against Corruption • International Convention for the Suppression of the Financing of Terrorism Parties to international treaties and conventions, law and border enforcement agencies KEY TAKEAWAY 2 Regardless of differing institutional arrangements, coordination and cooperation within a country and across jurisdictions are essential to optimize tax collection and enforcement of tax policy. 100 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 3.2.3. EVALUATION OF PERFORMANCE AND ACCOUNTABILITY Key strategic indicators are useful for assessing the performance of a competent authority. Performance indicators can include measures such as net revenue col- lected, total expenditures compared with budgeted amounts, the ratio of costs to collection, measures of filing and payment compliance and taxpayer satisfaction (15). Several international organizations, including the IMF, the World Bank, the Inter-American Development Bank and OECD have developed tools to evaluate tax and customs with key performance indicators. This section provides information on some of the indicators that are particularly useful for measuring performance related to tobacco taxes, including the cost of collection ratio, tax gap analysis and tax revenue targets. Cost of collection ratio Collection costs vary among countries. The cost of collection ratio is the total ex- penditure as a percentage of the total net taxes collected. This ratio is often used as a measure of efficiency and effectiveness of competent authorities. In Table 3.2, the cost of collection ratio is calculated for country groups by income level, based on an annual IMF survey. The numbers in the table give an indication of resources used and revenues collected for taxes in general. The same definition of cost of collection was used for all countries. The tax revenue excludes VAT and excise taxes on imported products, so it reflects internal taxes only: personal and corporate income taxes, VAT and excise on domestic production. Customs duties are also not included. The results show the differences among countries at various income levels. Other contributing factors include differences between tax systems, economic situations and compliance levels. Table 3.2 Cost of collection ratio in 2015 per 100 units (ratio of average recurrent budget to revenue collecteda) GROUP (SAMPLE SIZE) 2015 Low-income countries (6) 1.3 Lower-middle-income countries (15) 1.6 Upper-middle income countries (18) 0.9 High-income countries (36) 0.9 All (76) 1.1 a Does not include VAT or excise on imports Source: (Reference 8, Appendix Table 12). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 101 As one would expect, given lower levels of automation and resources, the ratio is higher for low- and lower-middle-income countries, greater than 1.0 (more than 1.0 currency unit needed to collect 100 currency units). The ratios for upper-middle income and high-income countries are below 1.0, indicating more efficient and/or effective collection systems. The cost of collection might be less relevant for taxes that are introduced with other than solely financial objectives, such as influencing a change in behaviour. In particular, in the case of excise taxes applied on tobacco products, the cost of tax collection does not reveal the full picture. If excise tax rates are increased substan- tially – or at least increased above inflation and income growth – consumption will be reduced. As a result, health care costs will be reduced due to reduced tobacco- related mortality and morbidity and increased productivity. These savings are not factored into the ratio of cost of collection to revenue, but governments do benefit from these lower expenses overall. Nevertheless, the cost of collection can be used as an indicator of the efficiency of a competent authority. Tax gap analysis Tax gap analysis is another method of determining how effectively taxes on tobacco products are collected. The tax gap is the difference between the tax due and the tax that is collected. For example, the theoretical tax due under an ad valorem tax on the retail price of cigarettes would be the average price of a pack of cigarettes multiplied by the number of packs sold (estimated from household expenditure surveys, for example) multiplied by the tax rate. This outcome can then be compared to the actual revenues collected (16). The effectiveness of tax collection can also be determined by using the macro- economic input-output matrix, measuring the added value of the economic sector – tobacco in this case – and the theoretical VAT due and then comparing the result with the real VAT collection. This methodology is valid for measuring domestic tax evasion (more information on the use of this method to estimate illicit trade is provided in Chapter 4, section 4.1). Tax revenue target The performance of a competent authority can also be evaluated by determining whether the tax revenue target has been met, if mandated, for a given tax period. Although revenue forecasts are often used as targets, caution is advised. Forecast revenues could include assumptions such as economic growth, inflation and amount collected. Forecasting is a good practice, however, and competent authorities should provide input to the government for the forecasting of revenues to improve the quality of the estimates. Competent authorities should monitor the actual collections 102 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N in comparison with the forecasted revenues, but the theoretical base may not be attainable for a variety of reasons. In addition, a revenue target could provide an incentive for some customs and competent authorities to simply aim to reach the target amount, rather than making efforts to collect the maximum amount possible with the available resources. 3.3 THE TAX COMPLIANCE CYCLE For any tax, there are associated compliance, control and enforcement processes. The compliance cycle usually includes registration and licensing, tax declarations, recordkeeping, storage in warehouses, duty suspension, collection of tax and tax refunds. Figure 3.1 illustrates the typical stages of the tax compliance cycle. Fig. 3.1 Tax compliance cycle 3.3.1 REGISTRATION AND LICENSING Along with regulating and ensuring the integrity of those who deal with controlled substances or goods, the main objective of licensing is to regulate the supply chain. Licensing is a powerful tool for obtaining more information and securing the supply chain of tobacco products. Parties to the Protocol are committed to licensing the manufacturing, import and export of tobacco products and manufacturing equipment (Article 6). In addition, Parties are committed to endeavouring to license – as considered appropriate – the persons involved in the growing of tobacco and the retailing, transporting, wholesal- ing, brokering, warehousing and distribution of tobacco products or manufacturing Registration and licensing (renewal) Tax declaration Authorities: audit and control Taxpayers: recordkeeping Audit and control of information provided in tax declaration Payment and collection of tax Refund Licensed activities, for example: import, production, transport, storage, export, etc. 1 2 3 456 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 103 equipment (Article 6). To ensure an effective licensing system, Parties shall monitor and collect, where applicable, any licence fees that may be levied and consider using them in effective administration and enforcement of the licensing system, for public health or for any other related activity in accordance with national law. If feasible, each Party shall require that retailers and tobacco growers – except for traditional growers working on a noncommercial basis – maintain complete and accurate records of all relevant transactions in which they engage, in accordance with its national law (Article 9.4). Article 6.3(b) of the Protocol provides a list of information to be requested from the applicant of the licence, including: • relevant identity information on the applicant • business location of the manufacturing unit or warehouse and production capacity • detailed list of tobacco products and equipment used • description of where the manufacturing equipment will be installed and used • documentation or declaration of any criminal records • information on bank accounts to be used for transactions and payments • description of intended use and intended market of sale of the tobacco products. To make it easier for authorities to collect all the information they need, rules of confidentiality could be exempted in the licensing process. Licences can be general – covering all activities requiring a licence – or issued for each activity separately, such as different licences for manufacturing, importing and retail. A general licence is less burdensome for the licensing authority, whereas licences for each type of activity offer greater control but at the cost of more adminis- tration (17). The cost of implementing the licensing system should be proportionate to the potential impact of the system. Not only should the type of licences be taken into consideration, the process and information needed to obtain a licence should be carefully considered to ensure proportionality. The more stringent the process is – in terms of the information required and the obligations the system imposes on licensees – the more burdensome the regime will be on both businesses and the authorities who must administer and enforce it. The more information is collected, the higher the compliance and administrative burden will be. It is recommended that the added value of the information be balanced with the additional compliance, administrative and/or enforcement burden. The level of stringency should be decided with consideration of factors such as the level of risk of the activity and the availability of enforcement capacity. A more stringent regime might be justifiable for activities that pose a higher risk for the 104 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government in terms of potential loss of tax revenues – such as the import, production and handling of excisable products on which the excise taxes have not yet been paid. Authorities could consider setting licence fees at a high enough level to cover the costs of administering and enforcing the system. For an example of a system that relies mainly on licensing and permissions, see the case study of Australia in Box 3.1. Wholesalers, distributors and retailers of tobacco products could also be required to obtain a licence before they can engage in the trade of those products. This would enable the competent authority to require reports on, for example, transactions relating to the purchase and sale of tobacco products. Moreover, it would allow the authorities to complete the audit trail of the entire supply chain and to obtain data that will help tax and health policy-makers properly and effectively monitor tobacco products. Governments could also require a licence for entities dealing with raw materials or growing tobacco, including farmers. If licensing of tobacco farmers is deemed appropriate and subsequently required in a country, the farmers have to identify and register their farm areas and location to obtain a licence. The benefit of requiring licences for farmers is that the control of the legitimate supply chain is extended to the identification of the source of the raw material for tobacco products. It also makes it more difficult to divert raw tobacco from the licit to the illicit supply chain. Licences are issued by different agencies across the world. In Brazil, for example, the Health Surveillance Agency is responsible for providing licences. Operators need to obtain approval of the layout of manufacturing and warehousing facilities before they can operate. In addition, they must demonstrate how they will comply with other laws and regulations – for example, by showing the design of product packaging, including the pack, carton and master case. The factory location must be identified before manufacturers can obtain a licence. Finally, a licence is required for the importation of machinery to produce tobacco products (18). Licences can be a source of useful information if authorities establish the informa- tion that applicants must supply in order to obtain the licence. Such information could include the quantity, price and how the tobacco harvests are disposed, as well as the identity of the buyers. It is recommended that an effective licensing regime collect information to establish both the identity and characteristics of applicants by requiring criminal records on relevant offences, such as previous noncompliance with tobacco licences or fraud. To avoid loopholes for monitoring raw tobacco, importers of tobacco leaf could also be licensed or at least required to register and report information on quanti- ties, sources and sales. In some countries, this information is already collected by a government agency other than tax authorities, for example, by the ministry of agriculture. Duplication of requirements and reporting should be avoided through CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 105 legislation and coordination among agencies. If licensing of (small-scale) farmers is difficult to implement, subsequent purchasers (first processors) in the supply chain could be licensed and regulated instead. Licensing first processors is often less burdensome to enforce for competent authorities because, in general, there are far fewer first processors than there are growers. For example, in the EU, between 50 and 100 first processors have been identified, compared with 55 000 farmers (19). Countries could also consider requiring registration of persons or entities engaged in the manufacture and import or sale of materials used for the manufacturing of tobacco products, such as cigarette papers, tobacco leaves, additives, adhesives, acetate or any other type of filters used for cigarettes, tipping paper and cellophane or plastic wraps, as well as materials for packing the cigarettes into packs, reams and master cases. In addition, tobacco manufacturers could be required to obtain a licence before they can purchase these materials. The Parties to the Protocol should decide on appropriate measures, depending on research as to whether key inputs that are essential for manufacturing of tobacco products exist and can be identified and subject to effective controls. KEY TAKEAWAY 3 The objective of licensing is to regulate and secure the supply chain. It is a powerful tool for obtaining information for verification, further investigation and audits. Ideally, all persons involved in the growing of tobacco and retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. Licensing requisites Based on case studies and best practices – including experiences from managing bonded warehouses where the value of merchandise or suspended duties or taxes is high – the following kinds of information could be required to obtain a licence, in particular, for producers, warehouses and distributors of tobacco products: • certification of safety of installations, perimeter security for production and storage (may include CCTV [closed-circuit television]4 access for tax administration) • certification of financial solvency • detailed online, real-time inventory of tobacco products and main raw materi- als, accessible by tax administration • electronic accounting systems 4 The term “closed-circuit television” is used generically to describe surveillance camera systems. 106 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • detailed lists of owners and managers • banking and other financial records • periodic electronic reports of transactions for tobacco products • anytime tax administration right of entry for inventories • mandatory electronic tax returns and payments • mandatory prior-to-arrival customs declarations for tobacco products • declarations of compliance with the tax stamp system (if applicable) • for those involved in import or export, authorized economic operator (AEO) certification • proof of compliance with the bond or guarantee regime • agreement to finance reasonable cost of inspections and tracking and tracing. Box 3.1 Case study of licensing in Australia Australia has taken an approach to controlling tobacco taxes that differs from that in many other countries. It has not used fiscal marks or tracking and tracing.5 Instead, it administers tobacco taxes through licensing and permission-based systems aimed at facilitating operations by lower-risk entities while preventing or tightly controlling commerce involving higher-risk entities. The domestic tax agency, the Australian Taxation Office (ATO), is responsible for most of the controls. These controls cover tobacco that is grown or manufactured and imported as finished goods or as leaf for manufacturing in Australia. In fact, the legal tobacco market in Australia consists only of imported finished tobacco products. In 2006, all tobacco-growing licences were cancelled by the ATO because manufacturers switched to cheaper leaf from external suppliers. The last domestic cigarette manufacturers closed in 2015 and 2016, and there has been no legal domestic tobacco growing or manufacture since then. The ATO administers all other functions relating to the import of tobacco and tobacco products, including licensing of bonded warehouses used to store imported products and issuance of permissions to undertake movement of bonded tobacco products between licensed bonded warehouses or to places of export. Importers must apply for a licence for a bonded warehouse to store imported tobac- co. The applicant must meet general criteria such as fitness, recordkeeping and security.6 These criteria are designed to ensure that only low-risk entities are able to enter the ex- cise tax system. Risk levels are also kept at an acceptable level through provisions allow- ing the suspension or cancellation of licences, subject to appeal. Licences are valid for a 5 See sections 3.4.4 and 3.4.5 for detailed discussions on fiscal marks and tracking and tracing. 6 The entity must not have been charged with an offence under the Excise Act or any Commonwealth, State or Territory Act that carries a penalty in excess of US$ 105 000 in the previous 12 months (or convicted in the previous 10 years), has shown a history of compliance with tax law in the previous four years, has had no previous cancellation of a licence, has adequate financial resources and is not in receivership. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 107 three-year period, with automatic renewal for licensees with demonstrated compliance. The permission system relies on post-transaction audits of commercial records. Criteria used to assess risk include the size of the duty liability, the compliance record of both parties and the possibility of diversion into the market. When there is a perceived risk of revenue loss, the application can be denied or a financial security deposit can be required. Exports of tobacco products are also subject to an export declaration process with the Australian Border Force. An approved export declaration is required for the products to be able to leave the country. Following recommendations from a government task force in 2017, the status of tax-suspended, bonded tobacco was eliminated as of 1 July 2019. In addition, an import licensing regime was introduced, and commercial tobacco imports without a licence are banned. Importers are required to identify their duty liabilities at import and make immediate payment; there are no credit terms available. Full payment of duties and taxes to the Australian Border Force are required prior to a release of tobacco products into the country. Sources: (20–21). As mentioned above, certification as an AEO could be requested as part of the licensing process. Most customs authorities are familiar with the concept of AEOs. Created by the WCO, AEO principles were initially focused on security concerns (22). Having a special licensing regime for operators of the tobacco supply chain is recom- mended due to the special nature of the product. For countries that have no system in place, AEO certification could be a starting point for setting up such a regime. An AEO is defined by the WCO SAFE Framework of Standards (22) as a party involved in the international movement of goods – in whatever function – that has been approved by, or on behalf of, a national customs administration as complying with WCO or equivalent supply chain security standards. AEOs include, inter alia, manufacturers, importers, exporters, brokers, carriers, consolidators, intermediaries, ports, airports, terminal operators, integrated operators, warehousers and distributors. For many years – in some cases, even since the 1970s – customs administrations have been increasingly involved in the security of the international trade supply chain. More recently, customs administrations have developed security programmes in a global context. The AEO is part of these programmes, and in 2005, the WCO adopted the SAFE Framework of Standards. Since then, a number of traders have been required to make substantial investments in order to obtain AEO status and must continue to invest to maintain that status. The AEO program is also recognized by the Trade Facilitation Agreement, a multilateral agreement signed by 174 countries (23). 108 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Some regional blocs have further specified the standards for AEOs and provide clear and well-structured information on their websites to guide and encourage operators to apply for AEO status. A good example of this practice is the website of the Revenue Commissioners of the Republic of Ireland, which contains the in- formation shown in Box 3.2. Box 3.2 AEO: Republic of Ireland Tax and Customs What are AEOs? AEO status is a certified standard authorization issued by customs administrations in the European Union (EU). It certifies that an economic operator has met certain standards in relation to: • safety and security • systems to manage commercial records • compliance with customs rules • financial solvency • practical standards of competence or professional qualifications. This is primarily a trade facilitation measure that recognizes reliable operators and encourages best practices in the international supply chain. As an AEO, an operator could benefit from: • recognition worldwide as a safe, secure and compliant business partner in international trade; • lower risk scores in risk analysis systems when profiling; • priority treatment if physical controls are conducted; • mutual recognition of AEO programmes under Joint Customs Cooperation Agreements, which could result in faster movement of goods through third- country borders; • reduced data sets for entry and exit summary declarations (this applies only to AEO safety and security); • easier access to simplified procedures; • reduction or waiver of comprehensive guarantees. The conditions for AEO status apply to all businesses regardless of size. Manufacturers, exporters, freight forwarders, warehouse keepers, clearance agents, carriers and importers may all apply for AEO status. Source: (24). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 109 3.3.2 DATA COLLECTION, DECLARATIONS AND ACCOUNTING The effectiveness of risk analysis depends on the quality and reliability of the available data. This is also the case for risk analysis in relation to tobacco taxes. Obtaining reliable data can be a challenge in many countries, but the use of electronic sys- tems to collect and manage data is increasing in most competent authorities. The introduction of VAT in many countries around the world has greatly improved the availability of data that can be used for tobacco tax analysis, since reporting is done along the supply chain on, for example, the value, quantity of goods and transaction date. Most countries applying excise duties also have a VAT system in place. In addition, more countries are becoming Parties to the Protocol. With the implementation of the Protocol, more data will become available because countries will be obliged to implement, among other measures, licensing systems with report- ing requirements and tracking and tracing systems. More information on tracking and tracing systems is provided in section 3.4. The obligations of the Protocol will also assist in monitoring the stock of tobacco products. Ideally, all entities involved in the tobacco product supply and distribution chains should be licensed and required to record every transaction that occurs. As this might be burdensome for both tax authorities and taxpayers, the use of automated and electronic systems is recommended in order to decrease the costs of compliance. An accurate inventory system for all raw materials, machinery, goods in process and finished products can be required. It is even more important to have good recordkeeping of the required data. As the volume of reported data increases, a good information technology (IT) system will be needed. The use of IT for periodic tax declarations, accounting, inventory and financial data is critical for obtaining accurate information and decreasing costs for the entire reporting system. Most countries now have some level of automation that can facilitate data analysis. An emerging trend is the use electronic invoices, issued by traders, as part of online real-time information for tax administration. Countries generally start by using electronic invoices at public utility companies and then later expand the use to large companies. Electronic invoices minimize the use of paper, contribute to automated recordkeeping and give accurate and timely information about transactions for tax administration. Several countries began using electronic invoices for companies on a voluntary basis and later made their use mandatory, especially for large companies with a high number of transactions – including the tobacco industry. Electronic invoices have been implemented successfully in EU countries and almost all Latin American countries, as well as several Asian countries.7 7 Electronic Invoicing in Latin America: English Summary of the Spanish Document; Inter-American Development Bank, Inter-American Center of Tax Administrations, 2018 (https://publications.iadb.org/ publications/english/document/Electronic-Invoicing-in-Latin-America.pdf ). 110 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N To verify that information is accurate, competent authorities could systematically cross-check declared information against third-party information (e.g. from banks, financial institutions, employers) or match the data with the information in registers of other government agencies. Processes of cross-checking and data matching could also be automated to minimize the administrative burden (25). KEY TAKEAWAY 4 Reliable data are essential for effective risk analysis. While obtaining these data can be challenging, electronic systems can help reduce the burden by automating procedures of data collection and cross-checking of information with different sources. 3.3.3 RECORDKEEPING Parties to the Protocol are committed to requiring, as appropriate, that all persons or entities engaged in the supply chain of tobacco, tobacco products and manu- facturing equipment keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of tobacco products (Article 9). Relevant information includes market volumes, trends, forecasts of tobacco products and quantities of tobacco products and manufacturing equipment kept in stock in tax and customs warehouses in transit, transhipment and under duty suspension. This information should be required from the persons and entities engaged in the supply chain and submitted to the competent authority on a regular basis, as provided for in the law. The competent authority can use the submitted information to monitor compliance with tobacco regulations and payment of taxes. A registry with this level of detail can realistically be kept only in electronic form. Records must provide full accountability for materials used in the production of tobacco products. The intention is that tax authorities and manufacturers should be able to reconcile the production quantities with the inputs used in production – thereby providing confidence that no unrecorded or illicit production has occurred. Obligations should also be imposed on suppliers of key inputs to show that supply is commensurate with demand (17). KEY TAKEAWAY 5 To monitor compliance and payment of taxes, all persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions, as well as details of materials used in the production of tobacco products. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 111 3.3.4 WAREHOUSING, STORAGE AND DISTRIBUTION According to Article 6.2 of the Protocol, all Parties shall endeavour to license persons involved in any wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Maintaining a system of authorization allows the authorities to carry out controls in production and storage facilities to ensure that taxes are paid (2). The approval process to obtain an authorization could include an evaluation of the layout of the plant or warehouse, the machinery that will be used and the flow of production, warehousing and shipping, including the points of entry and exit of raw materials and finished products. The basic method of monitoring production and ensuring that only tax-paid products are released to the market from the premises is to identify the production facilities and to control the entry and exit points. From time to time, the competent authority should conduct a physical inventory of the goods contained therein to check whether all documentation was duly prepared and approved and to determine the accuracy and completeness of the records kept. If the jurisdiction requires tax stamps to be placed on the tobacco products, only products with the proper stamps affixed can be withdrawn. Generally, tobacco products for which the required taxes have not been paid and, if required, fiscal marks have not been affixed should not be allowed into warehouses. For practical reasons, many countries allow suspension of excise duties, meaning that prior authorized persons can produce, send, receive and store tobacco products on which the excise duty has not yet been paid. The relevant authorities could also require that products on which the taxes have been paid should not be stored in the same areas as the products under duty suspension. Obviously, products under suspension of payment of excise duties are at high risk, which could justify stricter requirements for production, trade, storage and handling. Australia, which has a strict system of licensing and requirements for permission to move tobacco products, has migrated to a new system that eliminates bonded warehouses from the supply chain as of 1 July 2019. Importers are required to have an import licence and to pay excise taxes on cigarettes immediately upon import (see Box 3.1 in section 3.3.1). 3.3.5 DUTY SUSPENSION Many countries require authorization of natural or legal persons (as authorized warehouse keepers) to produce, process, hold, receive and dispatch products sub- ject to excise duty during their business. Producing, processing, holding, receiving and dispatching excise goods often take place under suspension of the excise duty. Guarantees can be requested from authorized persons to secure the payment of taxes. Features of such a system may include strict criteria for granting authorization, 112 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N warehouse pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. The use of a computerized system for monitoring movements of excise goods under suspension of excise duty can be a control as well. Different licences for products under duty suspension could also be considered. This would make enforcement easier and less burdensome for both authorities and operators. In general, it is recommended to allow the handling of excise goods under suspension of duties only if strict criteria are met. Such criteria could include pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. In principle, the movements of tobacco products should also be covered by the tracking and tracing system. Considering the high risk related to these products, additional monitoring could be considered appropriate, such as a computerized system monitoring the movements of excise goods under suspension of excise duty. In the design of such a system, it is recommended that close attention be paid to customs procedures for import and export to ensure alignment and avoid a vacuum in monitoring. An example of a computerized system is the EU’s Excise Movement and Control System, which follows the movement of all excise products – including manufactured tobacco products – for which excise taxes have not been paid. The system records the movement in real time and is thereby an important tool for combatting fraud. In addition, this system is indispensable for the exchange of information and co- operation between the relevant authorities of Member States of the EU (26). Finally, authorization is required before tobacco products can be produced, imported or stored under suspension of excise duties (27). KEY TAKEAWAY 6 Products under duty suspension of excise taxes are at a higher risk of tax evasion, which can justify stringent measures such as requesting guarantees to ensure the payment of taxes, additional licensing requirements, compliance with computerized systems to monitor the movement of excise goods under suspension and on-site authorization and audits. 3.3.6 COLLECTION OF TAXES To reduce the complexity of tax collection systems, it is recommended that excise taxes be imposed at the point of manufacture, import or release from storage or production warehouses for consumption. This is common practice in the majority of countries that impose excise taxes. Collecting taxes at this level of the supply chain CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 113 greatly limits the number of taxpayers and thus the resources needed to control them. Encouraging taxpayers to use electronic payment methods can also increase the chances of collecting all taxes. The same applies to requiring guarantees for certain high-risk activities, such as the handling of goods under duty suspension. Many countries decide on a case- by-case basis the level of the guarantee, depending on the situation of the requestor and the level of risk (quantity or value and potentially due excise taxes) that the regular business activities represent in a given time frame. Some countries allow a reduction of guarantees for operators with a track record of good compliance. It should be noted that a guarantee is not a limitation of the liability; taxpayers can still be requested and liable to pay an amount far above the level of the guarantee. Tax payments should be required by law to be remitted at fixed intervals after sales or on a fixed date each month (2). Many countries have a specialized collection enforcement unit that works full-time on the collection of taxes. It is important to have a stop-filer or payment control that can act immediately when noncompliance occurs, by sending a message and phone call of late declaration or late payment to the taxpayer. This increases the likelihood of keeping taxpayers compliant. If nondeclaration or nonpayment persists, the bond or guarantee could be executed. Another reason for collecting excise taxes around the time of production or import is that quantities can be monitored more effectively at these points. There are different options for monitoring the supply chain of tobacco products. The decision about what kind of monitoring system to use depends on the country’s financial, technical and human resources. The weakest form of monitoring is in- dustry self-declaration. Activities to verify compliance and ensure the collection of the full amount of taxes due can include, for example, physical checks, audits, cross-checking of declared information with third-party data and inspection of administration and recordkeeping. In general, in countries with poor administration systems, enforced compliance is carried out by imposing physical control over the production or manufacturing process. The cost of physical control increases when there is a potential for fraud by excise officers. However, fraud can be diminished significantly when excise officers are rotated frequently among different locations and supervisors make surprise visits. Historically, some countries (e.g. India) have posted tax administration staff at production facilities to monitor production and removals. In India, a staff member of the competent authority is placed in cigarette and large bidi manufacturing facili- ties around the clock. Each officer records the daily production and the quantity of cigarettes/bidis that leaves the factory and reports to the next officer.8 8 Ministry of Finance India, personal communication, 2009. 114 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A better option is to monitor production remotely. The competent authority can require the installation of CCTV cameras in strategic places throughout the manufacturing and warehousing facilities. With these, the authority can establish a central command post from which the facilities and activities can be continu- ously monitored and documented. In addition, the competent authority can carry out physical inventory controls from time to time and – if electronic invoices are implemented – cross-checking between invoices and declared inventory. This is also an effective way to prevent collusion between staff of a competent authority and manufacturers or importers. For example, in 2015, the Bureau of Internal Revenue of the Philippines required all tobacco companies to install CCTV cameras in their production lines and warehouses. This decision was taken in response to large seizures of untaxed cigarettes, with the objective of monitoring production to ensure the payment of all taxes. The collection process must also be supported by IT systems. These systems must provide for transparency and accuracy to ensure a safe process for the flow of payments from taxpayers to the tax treasury. Most countries have implemented automated electronic systems for tax payments linked to each declaration, for both domestic and import. It is key for tax administrations to have a comprehensive agreement with the banking system in order to obtain lower transaction costs, if applicable. Some countries have implemented a state payment web portal that allows citizens to pay their taxes and other fees such as county fees, fees for car permits and licences and agricultural, health and environmental fees online. KEY TAKEAWAY 7 Excise taxes should be imposed at the point of manufacture, import or release from storage or production warehouses for consumption, to ensure that quantities can be monitored effectively. This also reduces the complexity of tax collection systems by limiting the number of taxpayers and thus the resources needed to control them. 3.3.7 TAX REFUNDS Refunds for VAT, excise taxes and customs duties are a common process in most countries, under the principle that consumption taxes are not exported. Frequency and methods of refund vary by country. It is common to have monthly refunds (if there are exports during the period), and the reimbursements may be sent directly to the exporter or reserved as a credit to pay other taxes. An alternative used by some countries that have a high volume of exports is a so-called zero rate, or suspension, meaning that indirect taxes (VAT, excise taxes and customs duties) are suspended for the whole chain – from import of raw materials to production and packing until CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 115 export. This regime requires a special licensing process. Since the tobacco industry has an export component, the refund process for this sector requires special atten- tion for tax administration. 3.4 CONTROL AND ENFORCEMENT Control and enforcement are the main functions of tax administration. In fact, most tax laws include the objective “to control and enforce tax compliance” and, for customs, “to control and enforce tax and duty payments at the border” or similar phrases. The Protocol provides guidance for control and enforcement of tobacco taxes. Efficient and effective competent authorities often have a strategic plan to ensure compliance, a risk-based approach to identify the problematic points in the chain and the ability to direct resources accordingly to high-risk or high-value areas. Tasks that can play a role in control and enforcement include controlling the registration and licensing process, due diligence, verifying declarations and collec- tion of taxes. Production and distribution controls including tracking and tracing, fiscal markings, audits and import and export controls all play a role in control and enforcement. This section describes the main activities for improving control and enforcement, focusing on the tobacco supply chain. The procedures and penalties that can be enacted once illicit trade in tobacco has been detected are also discussed. 3.4.1 CONTROL AND ENFORCEMENT PLANNING Strategic plan In modern tax administrations, it is common to have a strategic plan, with control and enforcement as pillars. Appropriate control of the compliance cycle is key to keeping taxpayers in compliance and preventing illicit trade and tax avoidance. For this reason, most tax administrations focus a majority of their resources on preventive policy. Some examples of this can be found in the strategic plans of the United Kingdom and the United States’ Internal Revenue Service (IRS). In the United Kingdom, HMRC has had a well-developed strategic plan for years. A key pillar of the plan focuses on keeping taxpayers compliant. This is the concept of prevention: controlling initial minor noncompliant behaviour for the majority of taxpayers, while using strong enforcement for the minority on the noncompliant side (28). The strategic plan of the IRS has a similar approach, with a focus on control. If noncompliance is detected, data analysis and behavioural insights are used to identify the best way to address noncompliance. Early intervention or self-correction are examples of ways to address detected noncompliance. The IRS also highlights the importance of resolving noncompliance to ensure taxpayer confidence in the tax system and protecting the integrity of the system (29). 116 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Risk-based approach Following the establishment of a strategic plan, an enforcement and control plan must be drafted. This plan should include definitions of the activities that will be enforced, the taxpayers upon whom they will be enforced and the circumstances under which they will be enforced, as well as allocating resources for staffing, audit- ing, infrastructure and IT. Targets must also be defined, including the number of interventions and the amount of additional collected revenue or reduction of tax evasion. Several tax administrations elaborate annual plans with periodic perfor- mance reviews aimed at improving results, and they correct allocations and targets as needed. Clear targeting of interventions is needed for better results, more efficient use of resources, lower costs for taxpayers and more effective collection. In other words, the point is to focus interventions on those who have a higher probability of noncompliance. Using a risk-based approach can be particularly beneficial. Tax risk management is a key element of control strategy in modern tax admin- istration. A risk is a possible threat to reaching objectives such as collecting taxes in an effective and efficient way for competent authorities. Risk assessment is the process of analysing risks and deciding on the best way to manage an identified risk. The responses can vary from acceptance to mitigation to avoidance. Proper risk assessment allows competent authorities to use their available resources most efficiently and to become more effective in dealing with risks. It can be used to improve compliance by identifying taxpayers or types of activities with a high risk of noncom- pliance. Groups of taxpayers with the same characteristics often have similar risks. Groups with a high risk of noncompliance could then be subject to greater review. Areas of potentially greater risk of noncompliance in the tobacco supply chain include import, export and transfers to and from warehouses, particularly when they take place under duty suspension. Gathering risk-related information from internal and external sources is a best practice in compliance risk management. Such sources could include third-party information (e.g. from banks, credit card companies, transport companies), studies on taxpayer behaviour and research on compliance issues, tax gap analysis, tax audits and declarations (30). Risk management uses these different sources of data along with algorithms to find patterns of high noncompliance. Risk analysis can indicate reduced risk as well. Lower-risk areas are likely to need less governance to ensure compliance, which allows for resources to be directed elsewhere. Risk assessment can therefore help with strategic allocation of limited resources to the areas of greatest risk while at the same time reducing the burden on lower-risk taxpayers. Risk management has always been done by competent authorities, but data availability and statistical methods to identify patterns have changed the way risks CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 117 can be assessed. Although many risk assessment systems are still done manually or include manual elements, the use of intensive data techniques allows systematic, deeper and more targeted analysis (31). Modern risk assessment makes use of electronic data on taxpayers, tax payments, declarations from other taxes, such as VAT, and third-party information. With these data, tax authorities can identify indicators that suggest where further activities might be required to ensure compliance. For example, VAT invoices can be used to match reported purchases of inputs of tobacco leaf to sales invoices of tobacco leaf wholesalers. For taxpayers (i.e. those who are licensed and provide required reports), competent authorities can create a business analytics program to determine whether the data reported are consistent on each side of the transactions. Moreover, in countries with a VAT system, competent authorities can compare the data reported by taxpayers under the VAT system with data reported under the tobacco excise tax system to detect any inconsistency. VAT invoices can also be used to verify inputs and sales data. If VAT is collected at all levels of the supply chain, it is easier for govern- ments to monitor the supply chain for the enforcement of excise duty obligations. Regular surveys on tobacco consumption that use the same methodology can also provide indications about the level of compliance with excise tax policy. A sudden drop in revenue that is not reflected in consumption data could be an indication of illicit manufacturing, illegal imports, cross-border shopping or forestalling. In addition, seizure data can provide valuable information on areas and activities at high risk of noncompliance. The structure of tobacco tax policies should also be taken into account when con- ducting risk analyses. If excise tax rates are increased, there might be a greater risk of forestalling or front-loading (see the discussion on anti-forestalling later in this section). Differentiated excise tax rates based on product or packaging characteristics – such as distinctions between soft and hard packs or filter and nonfilter cigarettes – are also prone to manipulation by operators, which could affect tax revenues. One of the options to mitigate these risks is to amend the excise tax policy and apply a uniform tax rate. For customs transactions, the use of risk management is a key element in target- ing merchandise and support declarations to be inspected. Before the 1990s, most customs agencies used random criteria for selecting targets for inspection. Since that time, many countries have implemented risk-based approaches for selecting inspections. Historical data on importers and trade communities, complemented by artificial intelligence technology, show that risk management tools dramatically increased the effectiveness of physical inspections. Most modern customs agencies have implemented such techniques, allowing for more effective control processes while facilitating smoother processes for those transactions that are in compliance. Box 3.3 details some of the recent changes in risk management processes. 118 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.3 Changes in risk management The OECD developed Fig. 3.2 to show the framework and key steps for understanding compliance risks in 2004 (32). The same approach is still used to identify, assess and prioritize risk. However, many competent authorities now use new technologies and advanced data analytics, along with more information sources, including external data from banks, employers and sales invoices for VAT, for example. Fig. 3.2 Compliance risk management process Source: (32). The methods of identifying risks and the analysis of compliance behaviour have also changed. Traditionally, competent authorities used audits to identify high-risk cases. With more diverse and better data, competent authorities can now use more evidence-based approaches to examine risk patterns. Success of compliance activi- ties is now more often measured in terms of their impact on the overall compliance environment, rather than only on increased revenues.9 9 For more information on effective risk management with several indicators and a checklist of questions, see the Tax Administration Diagnostic Assessment Tool (68). OPERATING CONTEXT Assess and prioritize risks Evaluate compliance outcomes: - Registration - Filing - Reporting - Payment Monitor performance against plan Analyse compliance behaviour (causes, options for treatment) Determine treatment strategies Identify risks Plan and implement strategies CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 119 In Indonesia, use of the compliance risk management process reduced the share of illicit trade in total consumption of cigarettes from 12% to 3%. More information on this can be found in the case study of Indonesia in Box 3.13 later in this chapter. Understanding the products – as well as the supply and distribution chains – al- lows competent authorities to identify which areas along a chain pose the greatest risk and therefore require more resources. Detailed information on the composition of selected tobacco products is given in Annex 3.1. KEY TAKEAWAY 8 Risk analysis helps identify the points of intervention that have higher probabilities of noncompliance. A risk-based approach with targeted interventions allows for better results and more efficient use of resources to ensure effectiveness of tax collection. 3.4.2 CONTROLS OVER THE TOBACCO SUPPLY CHAIN As defined in Article 1 of the Protocol, the supply chain covers the manufacture of tobacco products and manufacturing equipment – as well as their import or export – and may be extended, where relevant, to one or more of the following activities when so decided by a Party: 1. retailing of tobacco products 2. growing of tobacco, with the exception of traditional small-scale growers, farmers and producers 3. transporting of commercial quantities of tobacco products or manufacturing equipment 4. wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Article 4.1 of the Protocol requires parties to “adopt and implement effective mea- sures to control or regulate the supply chain of tobacco products to prevent, deter, detect, investigate and prosecute illicit trade in such goods and to cooperate with one another to this end”. Concrete measures to regulate the supply chain, as well as best practices in this regard, are discussed further below. Figure 3.3 shows the main places for reporting and monitoring along the supply chain: import, ex-factory and removals from warehouses. Manufacturers could be required to report imported inputs at the border, as importers of finished products do. If components are subject to licensing, information can be required as part of the licensing process. The arrows in Fig. 3.3 represent transporting, which is also part of the supply chain. 120 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 3.3 Cigarette supply chain from manufacture or import to retail sale 3.4.3 LICENSING10 AND DUE DILIGENCE A licensing system is effective only if it is properly controlled. Most tax administra- tions have experience with licensing processes for excise taxes on products such as alcoholic beverages and energy products. It is strongly recommended that lessons learned with the licensing process of such products be applied when implementing and enforcing tobacco-related licensing. Licensing provides timely and accurate data that can serve as the basis for audits because it identifies and controls legitimate operators. For new operators, the process to obtain a licence could include visits and verification of production factories, storage facilities and distribution premises. Countries that have no licensing system in place and would like to start applying licences could allow a transitional period for existing operators to comply with the new licensing requirements. The process of licensing control must be carried out and updated periodically, in particular by controlling the validity of bonds or guarantees, the proper functioning of the required systems (CCTV, for example) and recordkeeping. 10 Licensing is discussed here in the context of due diligence and enforcement. Details about how licensing can be set up and what information could be requested are presented in section 3.3.1. Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blend d. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: wher he tobac o will be roll d into st cks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 121 Where licences are required, the law should include a provision specifying that purchases from unlicensed suppliers – or sales to unlicensed purchasers – are not allowed. This means that both suppliers and purchasers would need to verify those with whom they are doing business. This requirement provides enforcement au- thorities with an entry point to enforce the licensing system at both ends. Also, a licensing requirement for manufacturing equipment assists authorities in identifying and prosecuting illegal manufacturing of tobacco products, reducing the burden of proof substantially. In many countries, the presence of manufacturing equipment is not sufficient proof that illegal manufacturing is taking place; the machinery has to be in operation and producing illegal tobacco products when authorities inspect the location. With a licensing requirement, however, the presence of machinery without a licence is sufficient for authorities to act. The validity of licences should be time-limited, requiring renewals or reapplica- tion, to maintain a high level of control. Adherence to the conditions required for a licence should be controlled by the authorities, and penalties for noncompliance – for example, suspension or withdrawal of a licence – should be severe enough to act as a deterrent (33). Regulations for licensing should provide for inspection of the licensee’s products and premises, with penalties for noncompliance, which could include criminal and civil prosecution for serious or repeated offences. As stated in Article 6.3(a) of the Protocol, Parties need to establish or designate a single authority or multiple authorities to issue, renew, suspend, revoke and/or cancel licences. In accordance with Article 7 of the Protocol, persons engaged in the supply chain are required by law to conduct due diligence before and during business relationships. They also must report to the competent authorities any evidence that a customer is engaged in activities in contravention of its obligations arising from the Protocol. This requirement includes customer identification, monitoring of sales to ensure that the quantities are commensurate with demand for such products within the intended market and taking measures to ensure compliance. Knowledge of the demand of a market is indispensable for determining if there is a case of oversupplying. If the supply of tobacco products to a lower-taxing foreign market exceeds the demand, it creates a higher risk that these products will be smuggled back into a higher-taxing country, undermining the objectives and effectiveness of the higher-taxing jurisdiction. In the past, some governments decided to impose a fine on tobacco companies if the quantities supplied were significantly higher than the demand and the risk of being smuggled back into their jurisdiction was judged to be high (34,35).11 11 Excise duty rates applied in all the EU countries can be found on the European Commission’s webpage: https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco-energy/ excise-duties-tobacco_en. 122 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 9 Licensing helps to identify and control legitimate operators. The data obtained from licensing can serve as a basis for audits. Licences should be controlled on a regular basis and updated periodically to ensure their validity. 3.4.4 FISCAL MARKINGS (E.G. TAX STAMPS) Fiscal markings are another important tool for controlling and monitoring pro- duction and import of tobacco products. Their use is generally considered to be appropriate for increasing compliance with tax laws. Fiscal markings can also be of help in distinguishing between genuine and illicit tobacco products. Tax stamps or other fiscal markings affixed to packs of cigarettes or tobacco products facilitate the collection of excise taxes, as well as audits and enforcement actions. The pres- ence of fiscal markings enables both the competent authority and the public to monitor whether the taxes on tobacco products were properly paid. It thus assists the competent authority in investigating illicit trade and prosecuting violations. Fiscal markings include tax stamps, enhanced tax stamps (banderols) and digital tax stamps. Examples of fiscal marks are tobacco stamps, tax stamps, excise stamps, tax stickers and banderols. Box 3.4 presents details on the different types and features of tax stamps. Tobacco products for export are often required to be marked that they are for export. Box 3.5 provides useful information regarding the International Organization for Standardization (ISO) standard for excise tax stamps. The terminology “fiscal mark” holds no indication of the characteristics of the mark. A fiscal marking is affixed to each pack of tobacco product. Requiring a standard package size can facilitate the application of the markings (2). Fraudsters can be deterred from attempting to re-use fiscal markings (in particular stamps) by having the marking affixed to each pack of cigarettes (or other tobacco product) before the pack is wrapped with cellophane (36). In most cases, tax stamps are purchased by the producer or importer and applied to each product sold as proof of excise tax payment (33). Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Box 3.4 Types and features of tax stamps Over time, tax stamps and markings have become more sophisticated. In the past, tax stamps were often paper-based and easy to counterfeit. New tax stamps use additional security features to make them more difficult to counterfeit. Authentica- tion solutions against counterfeiting can utilize various security features, including: • overt features – features that can be verified by the naked eye; CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 123 • covert features – features that can be authenticated only by using dedicated and specialized electronic readers; • semi-covert features – features requiring a simple tool that does not involve extensive training; and • forensic features – features that can be identified through laboratory analysis. Tax markings can be either physical or digital: • physical markings – the information is contained in the document or device attached to the package. • digital markings – information is obtained through a link with a database and by decrypting with the tools and keys used for creation of the data. The term “digital tax stamp” sometimes leads to confusion, as some paper-based stamps with digital components are also described as digital tax stamps. Tax markings that are fully digital do not contain information in the document or device attached to the package. It is probably too simplistic to say that digital tax stamps are more secure than paper- based tax stamps. Both types have advantages and weaknesses. For instance, both physical and digital tax stamps can be weak or strong on security features. Neverthe- less, the management, production, sales, transport and monitoring of physical tax stamps require increased attention. Tax stamps have the same value as banknotes and are a possible attraction for theft, loss and fraud. In Belgium, for example, the Court of Audit severely criticized the lack of control of the production and stock management of tax stamps in 2015. The Court of Au- dit concluded that tax stamps issuance should operate under recognized security practices and procedures relative to the security risk associated with the various production, distribution and issuance processes. Moreover, it was noted that new printing technology of digital tax stamps on packs may facilitate stock management and lead to less fraud. In March 2016, Belgium changed its stamps. The printing became an in-house process by the financial federal government department. The new stamps are still printed with a watermark, but they also have a digital component. The change resulted in a cost reduction by standardizing the sizes and optimizing the production process. Some of the more advanced fiscal marking technologies include embedded threads and watermarks; special inks and coatings, such as so-called invisible inks, holograms and foils; and calculated or changeable content. Because of their enhanced security features, these stamps can be more expensive than traditional stamps. In the state of California in the United States, the traditional stamps cost US$ 0.42 per 1 000 stamps. The cost of the first generation of high-tech stamps was 10 times higher, at US$ 4.77 per 1 000 stamps. This price nearly doubled for the second-generation encrypted stamp, to US$ 8.20 per 1 000 stamps. Nevertheless, California collected 124 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N about US$ 450 million of additional tax revenue in the first decade following the implementation of encrypted tax stamps. This additional revenue was far greater than the costs of implementation and enforcement. Other jurisdictions have also revised their tax stamps to incorporate new technolo- gies. The state of Michigan, for instance, replaced heat-applied cigarette tax stamps with digital pressure-applied stamps in 2015. Michigan deployed a tax stamp with several overt and covert security features and a unique quick response (QR) code and serial number. QR codes (machine-readable codes consisting of an array of black and white squares, typically used for storing URLs or other information) can have purposes beyond tracking and tracing. The QR code can be read by consumers with a smartphone or tablet application to access information on smoking-cessation programs, report violations of the state’s youth access policies, connect to a tip line to report noncompliant packs and learn about the harms from illicit tobacco sales and purchases. Enforcement authorities can validate stamps using the smartphone- based eTRACS (Electronic Tax Reporting and Audit Compliance System). As part of the system’s implementation, the Michigan State Police department created teams of enforcement officers in each of the state’s seven districts and the state Department of Treasury created its own enforcement team. Sources: (37–42). Box 3.5 The ISO standard on excise tax stamps The ISO published its excise tax stamp standard (ISO/TC 292/SC) in October 2018. The purpose of the ISO standard is to assist tax and finance authorities in enhancing compliance with excise tax regulations. A tax stamp is defined as a visible tax stamp, label or mark placed on certain types of consumer goods to show that the applicable excise tax has been paid. The ISO standard applies to tax stamps that are physical in nature – not to digital markings, which are directly printed on to packs without a physical component. “Authentication” in this standard refers to the authentication of the tax stamp, not the product on which the tax stamp is affixed. In other words, authentication of a tax stamp on a cigarette pack means that the tax stamp is authentic but does not guarantee that the pack is authentic. In addition, control measures are needed at the time of the application of the stamp to verify the conformity of the tax stamp with the corresponding product. The standard provides guidance on the content, security, issuance and examination of physical tax stamps used to indicate that the required taxes have been paid and that the tax stamp is authentic. The use of stamps to facilitate tracking and tracing within the supply chain is not described. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 125 Specifically, the ISO standard deals with the following issues: • defining the functions of a tax stamp • identifying and consulting with stakeholders • planning the procurement process and selection of suppliers • the design and construction of tax stamps • the overt and covert security features that provide protection of the tax stamp • the finishing and application processes for the tax stamp • security of the tax stamp supply chain • serialization and unique identifier codes for tax stamps • examination of tax stamps • monitoring and assessing tax stamp performance. A stamp may fulfil many functions, but the core business of tax stamps is to ensure and facilitate the collection of revenue. The tax stamp must use a combination of security features. The tax authority should ensure that the tax stamp can be authen- ticated and that counterfeit, altered, tampered or otherwise fraudulent tax stamps can be detected. The standard provides detailed information on the different components of the tax stamp such as the substrate, inks, adhesives, laminate, authentication or security features and the unique identifier that should enable checks on the payments of the required tax. The process of procurement is discussed in detail in the standard. The tax author- ity should ensure that the procurement process is open, transparent and meets the sustainability objectives. The tax authority should set out the goals and requirements to give tendering organizations more leeway in proposing optimum solutions that might be different from those the authority would specify. The standard is not prescriptive; rather, it provides a catalogue of options. It does not, for instance, recommend specific security features, but it does describe the different types of features that are necessary for a tax stamp to be secure. Tax officials still need to make decisions and choose the option that suits them best, but the standard remains recommended reading for those who would like to introduce tax stamp programmes in their jurisdiction. ISO standards are not freely available but can be purchased at the ISO Store (www.ISO.org) or from an ISO national member body. Source: (43). According to Article 8 of the Protocol, each Party shall require that unique, secure and nonremovable identification markings – such as codes or stamps – are affixed to or form part of all unit packets, packages and any outside packaging of cigarettes for the implementation of the tracking and tracing system within a period of five 126 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N years, and of other tobacco products within a period of 10 years, of entry into force of the Protocol for that Party. The Protocol specifies that at least the following information shall form part of the unique marking: • date and location of manufacture • manufacturing facility • product description • where available, the intended market of retail sale. In several countries, QR codes are used as fiscal markings for tobacco and alcohol tax control. Each stamp has a unique identifier code and a QR code. The data stored in the QR code provide the following product information: • manufacturer • production location • stamp order date • tax status and class • brand • intended market • unique identifier (serial number). KEY TAKEAWAY 10 The use of fiscal markings is generally considered to be an appropriate tool for increasing compliance with tax laws. Fiscal markings can also be helpful for distinguishing between genuine and illicit tobacco products. 3.4.5 TRACKING AND TRACING A tracking and tracing system assists authorities in determining the origin of tobacco products – and the point of diversion, if applicable – as well as monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to enable authorities to have information on all transactions through the entire tobacco product supply chain until duties are paid or other obligations are discharged. Traceability is not used only for tobacco products. It is also used to improve the supply chain function, as in the case of parcel services, as well as for product safety reasons, to manage potential product recalls and for regulatory reasons. Tracking is the process that monitors where a product is at all times while also creating a time and location record for all movements. Tracing is the ability to identify the past locations of a product, so that the product’s route can be followed CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 127 back to its origin (44). In other words, traceability is “the ability to trace the history, application or location of an object” (45). A tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it is possible to unambiguously register that product’s movements. Other necessary characteristics include the ability to share the registered move- ment information and to authenticate products. This enables a product’s status to be captured through the supply chain and its history to be identified and verified retrospectively. According to Article 8.4.1 of the Protocol, Parties should require the following information to be available: • the date and location of manufacture • the manufacturing facility • the machine used • the production shift or time of manufacture • the name, invoice, order number and payment records of the first customer not affiliated with the manufacturer • the product description and intended market of retail sale • any warehousing and shipping • the identity of any known subsequent purchaser • the intended shipment route, date, destination, point of departure and consignee. A good tracking and tracing system enables the government to properly monitor the supply chain, improves its ability to ensure collection of the proper duties and taxes, provides it with the ability to authenticate whether the identification marking is genuine and matches the product and improves its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. The following elements are required for an effective tracking and tracing system (46): • A serialized unique identification marking for each package of product. These identifiers are a distinctive combination of numbers, letters or both. They cannot be predictable or used more than once. The representation of the identifier on the package can be human-readable (letters or numbers) or machine-readable (barcodes). Generation of codes and encryption that are part of a tobacco industry patent should be excluded. • A data carrier with the serialized unique identifier and other information such as date and location of manufacture, manufacturing facility, product description and, where available, the intended retail market. This informa- tion should be readable by authorized agencies of any Party to the Protocol. The data carrier should comply with quality standards and be suitable for 128 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N high-speed production lines. Two-dimensional barcodes, for example, meet these standards and are readable with inexpensive equipment. • A link and parent-child relationship (called aggregation) between different packaging units that offers the option to trace a pallet without the need to scan all the packs and master cases of that pallet. • Recordkeeping of all shipping and receiving events along the supply chain. This includes, for example, the departure location and the arrival location, as well as the involved operators. International standards from the ISO are recommended for the capture and exchange of data and events. • The use of international standards for key information that is encoded in the data carrier (5). An example of a unique and internationally recognized identifier for products is a Global Trade Item Number. The following details on information storage and sharing are drawn from various sections of the Protocol. Data and events along the supply chain must be stored in an independent database that is controlled by competent government authorities. At the global level, national and/or regional databases can be interconnected to facilitate international inquiries by competent authorities. Parties to the Protocol agree to establish a global information-sharing focal point located at the Conven- tion Secretariat of the WHO FCTC, accessible to all Parties, enabling them to make enquiries and receive relevant information. Each Party shall ensure that the information recorded under paragraph 5 of Article 8 of the Protocol is accessible to the global information-sharing focal point on request, subject to paragraph 9, through a standard electronic secure interface with its national and/or regional central point. The global information-sharing focal point shall compile a list of the competent authorities of Parties and make the list available to all Parties. The cost of tracking and tracing systems is a concern for many countries, but as indicated in paragraph 14 of Article 8 of the Protocol, jurisdictions may require the tobacco industry to bear any costs associated with putting in place the tracking and tracing system in a country (46). In Brazil, the cost for cigarette manufacturers was US$ 0.0185 per pack (42). In Kenya, the cost for manufacturers was US$ 0.024 per pack (42). Along with considering the characteristics of a tracking and tracing system in selecting a particular one, it is important to avoid conflicts of interest, ensure fair and transparent dealing with suppliers, implement a zero-tolerance policy for corruption or anti-competitive behaviour and ensure compliance. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, which deals with industry interference, and Article 8 of the Protocol. Article 8.13, which states that “each Party shall ensure that its competent authorities, in participating CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 129 in the tracking and tracing regime, interact with the tobacco industry and those representing the interests of the tobacco industry only to the extent strictly necessary in the implementation of this Article”. Box 3.8 provides a cautionary example of a tracking and tracing system that is not compliant with Article 5.3 of the WHO FCTC. While the Protocol contains a great deal of information on the requirements that a tracking and tracing system should meet, questions come up in relation to the implementation of such systems. To achieve the objectives of the Protocol, the Meeting of the Parties (MOP) to the Protocol, as the governing body of the treaty, has the prerogative to establish subsidiary bodies, such as expert groups and working groups. In decision FCTC/MOP1(6), the MOP established a working group for the development and implementation of tracking and tracing systems in accordance with Article 8 of the Protocol, including the global information-sharing focal point (Article 8.1) and unique identification markings for cigarette packets and pack- ages (Article 8.3), to further elaborate on the next steps. The working group will produce a comprehensive report compiling good practices and experiences on the implementation of tracking and tracing systems, as well as unique identification markings for cigarette packets and packages at national or regional levels. The working group was also given a mandate to prepare a conceptual analysis of how a global information-sharing focal point could be set up. Implementing a complete tracking and tracing system with fiscal markings takes time. In most of the countries that have already implemented tracking and tracing, it took several years from starting with the legal framework to final implementation. Several hurdles need to be overcome: • Legal framework approval is usually delayed by the tobacco industry. • Knowledge of tracking and tracing and associated technologies is scarce at tax administrations. • Tender and bidding processes are complex. • Coordination between domestic tax authorities and customs is weak. Even though the process might be lengthy, the investment in a tracking and tracing system will be repaid with the amount of tobacco taxes that are not lost due to evasion. When implementing a new tracking and tracing system, tax administration should ask for collaboration and technical assistance from intergovernmental organizations and countries that have successfully implemented such systems, in order to speed up and ensure success of the process. Examples of tracking and tracing systems implementation in Chile, Kenya and the EU are detailed in Boxes 3.8, 3.9 and 3.10. For countries that already have fiscal markings in place, the potential interaction between the markings and the implementation of a tracking and tracing system should be taken into account. Further information on this interaction is presented in Box 3.6. 130 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.6 Tracking and tracing and fiscal markings It is becoming increasingly common for stamps to contain some tracking and trac- ing features, such as unique identification markings and basic information on the product that describes the company, tax status or the product itself. The intention is to mark each pack with a unique identification marking so it can be monitored from the point of production to the retailer, including each step in between, thereby creating a complete time and location history. Although a tax stamp could meet the requirements of Article 8 of the Protocol and have tracking and tracing features, in general, the focus of tax stamp systems differs from that of tracking and tracing systems. Tracking and tracing is more than the unique, secure and nonremovable identification markings on the packages of tobacco products. It implies reading or scanning the codes; linking the codes between packs, cartons, master cases and pallets; uploading the information to a database; recording of any shipping and receiving events along the supply chain; and interconnecting the different databases. While new tax stamp programs contain tracking and tracing features, they are primarily intended to facilitate tax collection on the domestic market and not to track duty-suspended cross-border trade or the export of products. The focus of tax stamp systems is on authenticity and the proof that taxes are paid. The focus of tracking and tracing systems is on unique identification and on control of the movements in the supply chain by monitoring and investigating the past and future location of products. Tax stamp programs focus on stock management, verification (that the stamps correspond to the product) and authentication (that the stamps are genuine), while the focus of tracking and tracing systems is on the origin, intended route, first customer and final destination. The focus of tax stamps is primarily on individual packs intended for the duty-paid domestic market, while the focus of tracking and tracing systems is on all packaging (packs, cartons, master cases, pallets) and certainly – but not exclusively – for the duty-suspended export market. Nevertheless, sometimes there can be synergies. For example, the EU countries that require a tax stamp or national identification mark for fiscal purposes have the option to use it as the security feature for tracking and tracing purposes, provided that the requirements are met. In summary, tax stamps can be converted to or be part of a tracking and tracing system when the converted system provides aggregation between packs, cartons and master cases and records all movement along the sup- ply chain. For the export market, a unique identification marking should be added. Sources: (39–42). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 131 Box 3.7 What not to do: use the industry solution to tracking and tracing Codentify is a serialization system used to produce cigarette pack markers for the purpose of verifying whether cigarette packs are legal. It was patented by PMI but subsequently licensed at no cost to other major cigarette manufacturers. In 2016, Codentify was transferred to Inexto, which is an affiliate of the French group Impala. What is the problem with Codentify/Inexto? The main issue is that Codentify/Inexto’s links to the tobacco industry make it incom- patible with the Protocol, which came into force in September 2018. The Protocol specifies that obligations assigned to a Party shall not be performed by or delegated to the tobacco industry. Additionally, many elements indicate it is an ineffective means of authentica- tion. For example, the 12-character digital codes generated by Codentify can be easily duplicated or cloned and used as originals on either a counterfeit or genuine pack, which can then pass the system’s basic verification test. The codes are also produced by relatively unsecured, commercially available equipment and do not include high-security features capable of protecting the authenticity of identifier numbers. Systems that use multilayered, advanced security solutions that enable distributors, retailers, customers and authorities to identify noncompliant products are more secure. Another problem is that Codentify/Inexto cannot track products as efficiently as other available systems. It requires a much larger enforcement capacity to achieve the same detection rates as other systems that are not linked to the tobacco indus- try. Authorities would have to inspect significantly more packs marked under the Codentify system than is necessary under some other systems to achieve the same certainty of not missing a fraudulent pack. In addition, not all stakeholders will be able to verify that a pack marked under the Codentify system is genuine, while other available systems do offer this possibility. Source: (47). 132 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.8 A successful tobacco traceability system: SITRAF, Chile The Servicio de Impuestos Internos (SII, Internal Revenue Service of Chile) has success- fully coordinated and implemented a tracking and tracing system for tobacco products. Application of a compliance management model In the framework of a compliance management model, a traceability system for tobacco products utilizes a structural measure to reduce tax evasion. It is estimated that evasion of taxes (VAT and excise taxes) in the cigarette market in Chile amounts to 16.6% of the country’s total market annually – approximately US$ 300 million. The traceability system implemented in Chile (SITRAF (TAB2)) allows authorities to know in a certain and timely manner the quantities of cigarettes produced or imported into the country. Moreover, it helps authorities to distinguish between counterfeit products and original products that did not comply with payment of the tax. In 2018, the implementation of the traceability system was awarded to a company through public bidding. The company is in charge of the implementation and operation of the system for five years, according to a contract signed with the SII, and must maintain a team of 20 people available for the project. Direct markings are applied to items produced in Chile for national consump- tion, and stamps are used for imported products. For both types of product, the marking is based on a data matrix code, which is printed using security ink that is distinguishable from any other type of ink with specific devices that are provided by the awarded company. Although products for export are not subject to marking, they are controlled and accounted for by the traceability system. For national production, devices are installed on each production line that rec- ognize the type of pack being produced, print a unique code on each pack and then read it (activation) to save all the information on servers located in the production plant. This information is transmitted to the central servers of the system and then to the SII. It is also available for on-site inspection. In the case of imported products, the stamps must be acquired in Chile by each importer and then sent to its producer abroad, which is responsible for adhering them to each pack of cigarettes prior to wrapping the packs with cellophane, us- ing applicators on the production lines. Once the cigarettes enter Chile, the tax determination process has been completed in the service (Provisional Free Transit Guide) and the corresponding taxes have been paid to customs (Import Declara- tion), the importer must enter the data on the stamps used by the importer on the platform of the traceability system. After validation, the stamps can be activated in the system – that is, they are recognized as valid for commercialization. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 133 In addition to the devices provided by the company, SII has developed a smart- phone application for verification by citizens. Although the application cannot verify the authenticity of the ink used, it is able to verify whether a code is correctly gener- ated and display the information contained in the traceability system for the code (brand, variety, quantity of cigarettes, products/importer) so that the taxpayer can verify its consistency. Progress of the compliance management model Implementation of the traceability system has required a coordinated effort both within the service – for the generation of instructions, procedures and computer developments – and with other institutions, such as the National Customs Service and the Ministry of Health. It is an unprecedented project in terms of coordinating the implementation of the system in the production lines of the different tobacco companies in the country and the provider company. Some of the main milestones of the project are: September 2014: Law 20,780 on Tax Reform establishes an obligation to implement the system within a term of six months, after the publication of the resolution determining the obligated tax payers. May 2015: Resolution No. 47 determines obligated taxpayers. June 2015: Circular No. 47 describes obligation to incorporate stamps or distinctive marks as a traceability mechanism. February 2016: Law 20,899 on Tax Reform simplifies the definition of the system, allowing the system to be outsourced or provided by the SII, in addition to making the type of traceability more flexible. August 2016: Traceability system regulation D.S. 1,027 is issued (published on 28 December 2016). March 2017: Exempt Resolution No. 49 of the Ministry of Finance authorizes the SII to outsource all or part of the traceability system. June 2017: Bidding bases in public market are published. February 2018: Tender is awarded to selected company. June 2018: Decision of contract is made by General Comptroller of the Republic. August 2018: Resolution No. 61 determines taxpayers obliged to apply the trace- ability system. August 2018: Holding of first workshops for detailed definitions of the project, with the participation of Customs, Ministry of Health, provider and SII. September – October 2018: Visits of plants to coordinate with producers and define adaptations to production lines for system implementation. January 2019: Resolutions No. 6, 7 and 8 passed, with response to a request for an 134 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N extension of producers; start of system implementation in all production lines in the country. February 2019: Resolution No. 16 establishes a term to commercialize the remaining stock without marking. 2019: Resolution No. 24, with request for extension to importer, includes training of customs staff. March 2019: The traceability system is started up. June 2019: Stamp process begins for all imported cigarettes. Currently, the system is installed and operating in all production lines in the coun- try, placing traceability markings on virtually 100% of the cigarettes produced and imported. During 2019, the traceability system enabled controlling approximately 1 175 million packs: 744 million produced for national consumption, 409 million produced for export and 22 million imported yearly. Source: (48). Box 3.9 Case study of Kenya’s implementation of a tracking and tracing system Kenya’s current tracking and tracing system was preceded by a series of reforms in both tax structure and administration of excise taxes. The reforms included electronic cargo monitoring of exports, which allowed for automatic monitoring and reporting. The system appears to be highly effective because it requires less capacity and is less prone to manipulation than earlier systems. The experience of Kenya shows that a lower-middle-income country can successfully implement a sophisticated system capable of decreasing illicit trade. It also shows the importance of other measures such as strengthening enforcement, increasing cooperation and communication among different agencies and increasing penalties for noncompliance. Illicit trade, as measured by the Kenya Revenue Authority (KRA), was estimated to be around 15% of total consumption in the market during the initial reform period. After the introduction of the new system in 2015, it dropped to 5%. Timeline of the major reforms: 2003: In this period, the paper tax stamps used had a unique identifier and were colour-coded to indicate the type of product. Regular compliance checks were in- troduced. In 2007, the cost of a stamp was 2.124 Kenyan shillings or US$ 0.023 per pack. However, the stamps were found to be easily counterfeited and could not be linked to specific brands. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 135 2008: The KRA proposed a tracking and tracing system and increasing tax rates. The new system was introduced gradually. 2010: Enhanced security features, including ultraviolet markings, were added to the paper stamps. The stamps were to be clearly visible when packs were displayed for sale and placed so that opening a pack would destroy the stamp. The stamps were verified at four different points along the supply chain. The costs were just slightly higher than those of the previous stamps at US$ 0.024 per pack. Licensing was introduced for domestic manufacturers, subject to annual renewal. Importers were required to register with the KRA. Licences required submission of details on the company directors, inventories and equipment, accounting systems, input-to-production ratios and brands produced. Penalties for noncompliance were increased and included up to three years in prison. An electronic cargo tracking system was launched. Electronic seals were affixed on containers or trucks, and GPS technology was used for tracking. A bond was payable on exports to cover excise and VAT taxes. The bond was released only when the goods reached the final destination and taxes were paid. Verification involving both countries of the business deal takes place at the bor- ders. The electronic system provides information about the departure and arrival of the goods and the disarming of the seals. Authorities in the importing country are notified before the shipment leaves the domestic production facility. The system reduces the number of checkpoints and staff needed and generates arrival reports that can be verified with VAT refund requests. As a result of these changes, three factories and seven of the 10 importers were shut down due to noncompliance. Exports to Côte d’Ivoire, Eritrea, Mali and Sudan stopped because companies could not provide evidence that the goods reached the final destination and taxes were paid. More than US$ 11 million in excise tax losses was recovered in 2011. The KRA estimated that illicit trade dropped to 8%. 2013: A contract was signed to introduce a tracking and tracing system for tobacco and alcohol, the Excisable Goods Management System, in April. The system added production counting, tracking and tracing, stock control, processing and other data collection to the existing system. Infrastructure requirements included high-speed broadband internet at production facilities, warehouses, the KRA and ports, along with reliable power or backup generators at those points. Implementation was planned in three stages: • Stage 1 – A new electronic digital stamp with a unique identifier was introduced. It included a data matrix code plus overt markings (holograms, fluorescent fibres, a security link for KRA authentication and visible two-dimensional codes for verification and activation), semi-covert markings (UV features, fluorescent prints detectable by specialized devices, mini text printing for retailers and 136 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N distributors) and forensic taggants for use in prosecutions. The stamp also included human-readable codes for verification by short message service using the KRA web portal. • Stage 2 – Control and monitoring systems were automated in February 2014. Manufacturers had to install photosensitive readers on production lines, with data automatically sent to the KRA in real time. Each stamp was activated and associated with a brand and package size on the line. The KRA database is automatically updated every 15 minutes. • Stage 3 – Market surveillance began, with 83 officers given powers to seize illicit cigarettes and make arrests. The officers were equipped with hand-held devices that transmitted data to the KRA for authentication. Distributors and retailers became liable for selling products without an excise stamp and were subject to fines plus prison sentences of up to three years for noncompliance. In 2016, a smartphone application became available with which the public could authenticate cigarette packs. Importers must now buy digital stamps and send them to export facilities in other countries to be affixed. Tax liability is due at removal from a factory or at import. The electronic cargo monitoring system is still in effect. 2016: The Excise Duty and Tax Procedures Acts clarified new obligations and penalties. 2017: A new integrated customs management system was launched. The KRA estimates that illicit trade levels are now around 5%. The current, more comprehensive digital system is cheaper than the previous paper tax stamp system. Manufacturers pay for the production monitoring system, but it counts as a busi- ness expense on corporate tax returns. In 2018, two manufacturers and 10 licensed importers were operating in Kenya. In 2018, aggregation between the markings of packs, cartons and master cases had not yet been implemented but was expected to be forthcoming. Source: (36) Box 3.10 Case study of the new EU tracking and tracing system, May 2019 Cigarette smuggling and other forms of illicit trade in the EU is estimated to cause a loss of €10 billion in revenue annually. In 2018, 4.2 million packs (20 sticks per pack) of illegal cigarettes were seized by customs in the EU. Illicit tobacco production was also increasing: an illegal factory in Ireland, dismantled in 2018, was capable of producing 250 000 cigarettes per hour. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 137 The EU tobacco control policy is described in the Tobacco Products Directive and is influenced by the Protocol. Article 15 of the Directive calls for the traceability of cigarettes and RYO tobacco products by May 2019 and of other tobacco products by May 2024. The EU tracking and tracing system is sufficiently flexible to be imple- mented at both the regional and the single-country level. Countries can choose among providers as long as the basic requirements are met. The policy provides a high level of protection against any attempts at manipulating the data. The report- ing obligations cover all the economic operators involved in the manufacture and distribution of tobacco products. The EU system requires all unit packets of tobacco products to be marked with a unique identifier Information on the movements of those products is to be stored by third-party data storage providers. The suppliers of the unique identifiers and data stor- age are to be financially and legally independent from the tobacco industry. The data are to be fully accessible to authorities of EU Member States for enforcement purposes. The generation of unique identifiers, as well as all other codes required for pre- registration of economic operators, facilities and machines, will be done at the Member State level by designated identifier issuers. Manufacturers and importers are required to supply information relating to the product and production lines when requesting unique identifiers from the issuers. The issuers will then generate and deliver batches of unique identifiers. On the production line, manufacturers of tobacco products will complete each unique identifier with a marking indicating the date and time. The unique identifier will be a machine-readable, optical, one- or two-dimensional barcode. An anti-tampering device, capable of creating an unalterable independent record of the verification process, must have been installed previously. This additional record will be accessible to public authorities for potential investigation and inspection. Unit packets, as well as aggregated packages such as cartons, master cases or pallets, can be tracked and traced throughout the supply chain. Tracking is also allowed at an aggregated packaging level as long as unit packets remain traceable. During transport, each dispatch and arrival up to the final dispatch to the first retail outlet must be recorded and reported. All recorded information must be submitted to the independent third-party data storage facility, generally within three hours, and 24 hours before dispatch and transloading. Costs, including operational costs, are shifted to the tobacco industry, in line with Article 8 of the Protocol. The EU system of tobacco traceability and security features became operational on 20 May 2019. Sources: (36, 49–50). 138 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 11 A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as monitoring and controlling the movement of tobacco products and their legal status. 3.4.6 ANTI-FORESTALLING “Forestalling” is a term that describes increases in the production or stock of products in anticipation of a tax increase (2). Other terms referring to this practice include “stockpiling” and “front-loading”. Forestalling occurs when manufacturers or importers increase their tax-paid stock or oversupply the market by increasing production or imports in order to pay the previous lower rate. It reduces and delays the effective- ness of tax measures. The effective starting date of the new tax rate will be delayed, revenues will be lower and the possible effect on prices and thus consumer behaviour will also be postponed. To illustrate how anti-forestalling measures function, an example is presented in Annex 3.2. A legal basis must exist for anti-forestalling measures; otherwise, the govern- ment cannot prevent the industry from forestalling. Legal measures to deal with forestalling include (51–52): 1. Limiting the amount of tobacco products that can be released subject to the old tax rate and levying the new tax on the products exceeding that limit. 2. Levying the new tax rate on all goods that are still in stock and not yet sup- plied to the final consumer. 3. Limiting the number of tax stamps issued at the rate that was in effect before the increase or limiting the time that products with a tax stamp with the old rate can be sold. 4. Requiring producers and importers to buy new tax stamps annually or after a tax increase. Under the first three measures, the competent authority determines the limit for taxation at the previous (lower) rate. The quantity allowed may be based on the shelf life of tobacco products – around six months for cigarettes – or normal inventory levels, such as an average over the previous three years. The first measure, limiting the amount of tobacco products that can be released, requires resources from the competent authority for enforcement. Authorities may decide to post inspectors in each production facility, but even without posting inspectors, procedures are necessary for determining when the allowed quantity has been exceeded and what subsequent actions to take. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 139 The second measure, levying the new tax on goods in stock and not yet supplied to the final consumer, might be difficult to implement. The competent authority is required to monitor the manufacturing process and, at the very least, to conduct a stock-taking of all players in the supply chain – including the various manufacturers, importers, wholesalers and retailers – in a very short window of time. If monitoring covers only the stock of manufacturers and importers, this measure could easily be circumvented by ensuring the stock is sold to others in the supply chain or by setting up separate distribution companies to purchase the stock. Controlling stock at the retail level is burdensome and might not be administratively feasible given the large number of cigarette retailers. It becomes even more burdensome if there is no licensing requirement for retailers because they will first have to be identified. The third measure accomplishes the same thing as the first if a country uses tax stamps. The fourth also requires tax stamps and is simpler for the competent authority but somewhat more burdensome for the tobacco companies, since stamps must be purchased every year. Box 3.11 provides examples of anti-forestalling measures in EU countries. Box 3.11 Examples of anti-forestalling measures in EU countries Several EU Member States have taken measures to limit forestalling. A cautious ap- proach seems to be required for designing such measures to ensure that they comply with EU legislation and the general principles of EU law – in particular, the principle of proportionality. No disputes have occurred concerning the right of initiative of EU Member States to implement anti-forestalling measures. Nevertheless, several EU Member States had to defend their measures in front of the Court of Justice of the European Union, the institution that ensures all national legislation is in line with EU law and a consistent application of that law (53). The Court acknowledged that anti-forestalling measures are appropriate to combat tax evasion and tax avoidance. Moreover, the Court emphasized that fiscal legislation is an important and effective instrument for discouraging consumption of tobacco products and therefore for protecting public health (54–55). However, the measures taken should be proportion- ate to the objectives. The principle of proportionality means that only the action needed to achieve the objective should be taken, and it should not exceed what is necessary. This principle regulates the measures taken within the EU and is included in the Treaty on the European Union. The Court demanded that Portugal amend its legislation to ensure compliance with the principle of proportionality. Belgium, Estonia and Hungary were also urged to change their anti-forestalling measures to bring them in line with EU legislation (56–57). 140 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The majority of the EU Member States have anti-forestalling measures in place, but there is no harmonization of these measures. The following are some examples:12 • Portugal limits the quantity of cigarettes that can be released in the last four months of a year to the average of the previous 12 months plus 10%. In addition, manufacturers and importers must sell cigarette packages with a tax marking of the preceding calendar year within three months. For other tobacco products, longer limits apply. • Denmark limits the number of tax stamps issued before a tax increase at the old rate to 20% more than are usually purchased in the two months before the end of the year. • In Poland, tax stamps are valid only for the current calendar year, and cigarettes with the old stamp can be sold only through February of the following year. • In Romania, companies must apply for approval to release for consumption from the customs office. Source: (58). KEY TAKEAWAY 12 Forestalling reduces and delays the effectiveness of tax measures. Implementing anti-forestalling measures can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. 3.4.7 ADDITIONAL NATIONAL AUDITS AND CONTROLS In addition to the previously described measures, several periodic audits and controls could be implemented to increase compliance with tax laws. The most common audits and controls are the following: • Cost audit – The cost audit method provides expected VAT and tobacco tax collection by simulating the intermediate and final cost of cigarettes. It starts with the inventories of raw materials and estimates added values and final cost, then matches the results with real collection from the tobacco supply chain. Annex 3.1 provides more information about the components that make up some selected (tobacco) products. • Transfer pricing audit – To ensure companies pay their fair share of tax, prices of transactions between related companies should be assessed, and when prices are not in line with the market conditions, they should be corrected. Companies that operate at the international level (transnational companies), including many tobacco companies, can manipulate import or export prices 12 Considering the frequency of court cases concerning anti-forestalling incidents, these measures might have been replaced or amended. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 141 of merchandise or raw material to related companies or branches in other countries, with the objective of lowering profits in countries with higher tax rates, and can transfer those profits to countries with lower taxes. • Price and market monitoring – Retail price surveys can provide information about variance from the market price in certain locations, highlighting areas of potential tax avoidance or illicit trade. Physical control of such locations requires rapid response teams, as implemented in the Philippines. To monitor the tax compliance of its taxpayers, the competent authority in the Philippines needs to understand the tobacco market; it must have information on brands, market segments and prices of products. This information enables authori- ties to estimate the impact of tax and price changes on consumer behaviour and revenues. Market data can be analysed as part of risk management and anti-fraud analysis to determine whom to investigate for noncompliance and when to do so. Sales data can be triangulated to validate other data sources, such as household surveys on prevalence. Market data and trends are also useful indicators for determining whether there is a case of oversupplying. See also section 3.4.3. • Consumer control – Involving the public via awareness campaigns has also been shown to be effective. Consumers have the right to be assured that the products available in the market are authentic and come from legitimate sources. Thus, it is in the consumers’ interest to understand and be able to verify that they are buying genuine products. The features of the fiscal marks on tobacco products should help consumers distinguish between genuine and illicit products. Some countries – Kenya, for example (see the case study in Box 3.9) – use a smartphone application to allow anyone to check both covert and overt features and to report any cigarettes with incorrect markings. Other countries, such as the Netherlands, have developed a smartphone application that allows anyone to report a suspected case of excise tax fraud. • Cross-check controls – Competent authorities should consider using multiple sources to obtain market data and determine if these data are consistent with tax declarations. VAT declarations can be used to verify that suppliers and purchasers of raw materials and final products are reporting the same amounts. Bank information can be used to verify both sides of transactions along the supply chain. Any discrepancy can alert the competent authority to conduct further investigation for possible illicit trade of tobacco or tax evasion. 142 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 13 Several different types of periodic audits and controls that can be carried out to increase compliance, including cost audits, transfer pricing audits, price and market monitoring, consumer controls and cross-check controls. 3.4.8 IMPORT AND EXPORT CONTROLS Parties to the Protocol should allow import and export of tobacco products and manufacturing equipment only by duly licensed natural persons or legal entities (Article 6.1). A well-known strategy used by fraudsters is to declare products for export so that no duties are due to the country of export. These products are subse- quently transported through other countries, using the in-transit regime that allows temporary suspension of duties until the goods arrive at their final destination. Before arriving at their final destination – where the excise duties would be due – the goods disappear or are lost while being diverted to the illegal supply chain. The goods may never leave the country, or they may be smuggled back into the country from which they were exported without declaring or paying duties. This risk of loss of revenue can be mitigated by requiring a guarantee or bond, which will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless the required fiscal marking (such as a tax stamp or export label) is affixed on the pack, according to the national law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Illicit tobacco trade could be decreased significantly if the various competent authorities that have jurisdiction over manufacturers and exporters of tobacco prod- ucts and manufacturing equipment would provide the competent authority at the destination with prior information when a shipment has been authorized and is about to take place. The information could include the name of the consignee, a description of the item shipped and the quantity. Also, the competent authority at the destination should inform the competent authority having jurisdiction over the shipper that a shipment was received along with the pertinent information relating to the shipment. A good IT system is also required for import and export. Electronic processing of prior-to-arrival manifest and import declarations is recommended. Most countries have implemented an online customs system to process import and export declara- tions, including all required data such as country of origin or export, description of merchandise, value, weight, cargo insurance, carrier, importer or exporter and broker identification, detailed tax duties to pay and final destination. The World Trade Organization (WTO) Trade Facilitation Agreement provides sev- eral tools for better controls, including collaboration between customs administrations, CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 143 exchange of information, use of non-invasive devices and prior-to-arrival import declaration (23). Non-invasive detection equipment at customs posts is highly capable of detect- ing contraband merchandise. The most common tools are X-ray scanners that are used for small parcels, containers, trucks and trains. Most modern ports have also implemented the use of X-ray scanners, and such technology is improving the speed of controls to as little as two minutes per container. Although the cost of scan- ners is declining, it remains inaccessibly high for countries with limited resources. Fortunately, scanners are often available for lease, making them accessible for tax administrations in those countries. Less sophisticated and less costly detection equipment includes endoscopes, mirrors, night vision equipment, cameras and automatic licence plate readers (33). A still less expensive alternative is the use of dogs, which can be trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied in order to reduce contra- band. These include the separation of processing operations from the sealed storage of taxed and untaxed products. Physical and direct control by officials of the excise authority during a part or the whole of an operation can be applied (for example, physical escort of the transit consignment from border to border by individual trucks or in a convoy, or application of radio or satellite tracking systems such as GPS-enabled devices to goods, conveyances, vehicles or containers). Control at borders is essential and should include integrated technology and cooperation with agencies at the border station. Frontline officers should be sup- ported by appropriate intelligence, guidance and supervision from management, as well as technical aids to enforcement. Within a country, mobile excise control units are helpful for verifying excisable goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. These operations require close coordination between police, border guards and other public services. Exports also require special attention, in particular if VAT and tax refunds are granted to the export of tobacco products. Validating the real exit in the declared amount is essential to avoid illegal re-entry to a territory and the improper refund of taxes. For any tax refund, an audit including tax credit information must be carried out. The audit may include the invoices for the whole chain involved in the export, including tobacco farmers, first processors, manufacturers, wholesalers, storage and transport. 14 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 14 To ensure control of import and export, it is recommended that only duly licensed persons and entities be allowed to import and export tobacco products and manufacturing equipment. 3.4.9 FREE ZONES AND TRANSHIPMENT POINTS The term “free zone” is very broad and can refer to a number of different types of areas. The Financial Action Task Force listed the following types in its 2010 report (59): free trade, export processing, enterprise, free ports, foreign trade, special economic zones and bonded warehouses. A number of these areas can include tobacco manufacturing and trade. By definition, controls such as regulation and oversight within free zones are less strict than in other areas. This can make them appealing to persons involved in illegal cigarette manufacturing or trade (17). In fact, illicit activities related to free zones (not limited to tobacco) are regularly documented by organizations that recognize the linkage. These activities include money laundering, tax evasion and trade in counterfeit goods or other illicit goods (60). A report from the European Parliament (61) referring to free ports in particular, mentioned that the motivation for using them included a “high degree of secrecy and deferral of import duty and indirect taxes”. The report even proposed the “urgent phasing out of free ports”. In the European Parliament report, free ports are free zones that function as (semi-) permanent storage areas for high-value goods. The Protocol includes a time-bound provision of effective controls on all manu- facturing and transactions of tobacco products in free zones (Article 12). Free zones are defined as a part of the territory of a Party where goods are considered to be outside the customs territory for import duties and taxes (Article 1.5). This is the same definition used in the International Convention on the Simplification and Harmonization of Customs Procedures (Revised Kyoto Convention) (62). Parties to the Protocol must implement effective controls in free zones within three years of entry into force of the Protocol. For countries not yet Parties to the Protocol, stringent controls of manufacturing and transactions involving tobacco products in free zones are an important component of an effective and efficient tax administration. One of the measures for dealing with free zones within the Protocol includes implementing “effective controls on all manufacturing of, and transactions in, tobacco and tobacco products, in free zones, by use of all relevant measures as provided in this Protocol”. As indicated in an Interpol report (17), a significant vulnerability of free zones is the fact that different economic operations (e.g. manufacturing, assembly, re-packaging and warehousing) take place outside the control of authorities. It is therefore essential for customs administrations to exercise their authority in free zones CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 145 to effectively identify and fight illicit trade in tobacco products. Relevant measures listed in the Protocol should be applied. These include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a track- ing and tracing regime. Removing exemptions on excise taxes is an additional way to increase control and remove incentives for using free zones as a means for tax evasion. Indonesia, for example, imposes excise taxes on cigarette manufacturing in its free trade zone.13 Parties to the Protocol shall also prohibit the intermingling of tobacco products with non-tobacco products in a single container or any other such similar transporta- tion unit when removed from free zones. Finally, each Party is to “adopt and apply control and verification measures to the international transit or transhipment of tobacco products and manufacturing equipment in conformity with the provisions of the Protocol”. Article 13 of the Protocol, which covers all duty-free sales of tobacco products, requires Parties to the WHO FCTC to consider prohibiting or restricting the sale to or import by international travellers of tax-free or duty-free tobacco products, as mentioned in Article 6 of the WHO FCTC. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect government revenues by creating a loophole in the tax structure (2). KEY TAKEAWAY 15 Customs administrations should exercise their authority in free zones to prevent different economic operations from taking place outside the control of authorities. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a tracking and tracing regime. 3.4.10 PROCEDURES AFTER DETECTING ILLICIT TRADE OF TOBACCO The procedures described in previous sections are intended to increase compliance and to prevent illicit trade. When smuggling or illicit trade is detected – through, for example, audits, tracking and tracing systems, verification of declarations or border control – actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. To deter further illegal behaviour, a comprehensive audit of everyone and everything involved in the illicit acts must also be carried out. Assets and vessels involved in the illicit activity can be seized, and financial accounts can be frozen. Some countries, including the United Kingdom, Canada and Chile, have also adopted a strategy known as “follow the 13 Indonesian Ministry of Finance, personal communication, January 2020. 146 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N money” to obtain more information on those who finance illicit trade. This strategy is intended to have a further-reaching effect by targeting those who finance the transport, production and storage of illicit products. The United Kingdom, Canada and Chile all created special teams tasked with identifying and targeting the offenders. Furthermore, Article 18 of the Protocol provides for the confiscation and destruction of tobacco, tobacco products and manufacturing equipment. One of the difficulties faced by competent authorities in exercising the authority to seize and forfeit products and/or equipment used in the manufacture or distribution of tobacco products is the cost of keeping or storing the goods and/or machinery before destruction. Thus, the law should also provide for a mechanism and timetable for the disposal and/or destruction of seized and forfeited goods or machinery, while prescribing a mechanism by which these properties can still be presented as admissible evidence in a judicial proceeding. Boxes 3.12 and 3.13 provide examples of successful efforts to combat illicit trade in the United Kingdom and Indonesia. KEY TAKEAWAY 16 As soon as smuggling or illicit trade in tobacco products is detected, actions such as collecting taxes and seizing and destroying smuggled and/or illicit tobacco must be taken. Box 3.12 The United Kingdom’s experience in fighting illicit trade in tobacco products In 2000, illicit cigarettes accounted for 22% of the cigarette market in the Unit- ed Kingdom. To deal with the problem, Her Majesty’s Customs and Excise14 implemented a major anti-smuggling effort. The strategy was refreshed with ad- ditional resources and measures in 2011 and reviewed in 2015. The result was a steady decline in the illicit cigarette market to 10% by 2013/2014. The measures taken were comprehensive and included hiring 1 000 new customs officers and investigators. In addition, tobacco supply chain legislation was introduced, aimed at discouraging tobacco manufacturers from facilitating smuggling. Tougher sanctions included increased fines of up to £5 million levied on a manu- facturer, criminal prosecution with sentences up to seven years, confiscation of assets as part of the proceeds of the crime, payment of duty on the confiscated goods plus penalties up to 100% of the duty, prohibition of the sale of tobacco products for 14 By the time of the renewed strategy, the respective bodies were the HMRC and the United Kingdom Border Agency. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 147 up to six months, unlimited fines for selling tobacco without the United Kingdom duty-paid fiscal mark after 13 March 2015, referral for withdrawal of the transporter’s licence, use of immigration sanctions to refuse entry to the United Kingdom for tobacco smugglers and civil action, including bankruptcy. The cost of these measures was £209 million over the first three years of the program and around £100 million annually by 2008/2009. This figure covers only HMRC and excludes any costing of the United Kingdom Border Agency. In 2013/2014, tobacco tax revenues were £9.5 billion. After a review of the strategy in 2015, the controls on raw tobacco were strength- ened by the introduction of an approval system in 2017. Anyone who manufactures, purchases, acquires, owns or is in the possession of a tobacco products manufacturing machine must be licensed with customs as of 1 August 2018 (63). The United Kingdom ratified the Protocol on 27 June 2018. It was the 40th country to ratify, which was the trigger point for the Protocol to enter into force. Fig. 3.4 Estimate of the illicit cigarette market and United Kingdom tax-paid consumption Sources: (16, 64). Bi lli on c ig ar et te s UK tax paid consumption Illicit market 20 00 -01 20 01 -02 20 02 -03 20 03 -04 20 04 -05 20 05 -06 20 06 -07 20 07 -08 20 08 -09 20 09 -10 20 10 -11 20 11 -12 20 12 -13 20 13 -14 20 14 -15 20 15 -16 20 16 -17 20 17 -18 0 10 20 30 40 50 60 148 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.13 A success story: Indonesia reduced illegal cigarettes from 12% to 3% of the market In some countries, most of the illegal cigarettes are imported, but in Indonesia most of them are produced within the country by unregistered manufacturers that are usually home-based and relatively small. In several territories, specifically on the Island of Java, illegal cigarettes have been produced for generations. This practice is supported by the availability of raw tobacco materials and cloves, as well as cheap labour costs, especially for female workers. Indonesia also produces cigarette products that are not available in most other countries. These products – known as handmade clove cigarettes (sigaret kretek tangan) – contain cloves, and the production process covers blending, rolling and packing. Parts of the process are done by hand, and 99% of the labourers are women. The strategy for combatting the illegal cigarette trade in Indonesia is divided into two main parts: preventive actions and responsive actions. Indonesia’s success in effectively tackling illicit trade is attributed to the following key factors. Monitoring and surveillance Preventive actions consist of administrative measures – such as issuance of permits and the excise stamp purchasing mechanism – that use risk management by optimizing the Excise Service Information System (ExSis). With this IT system, the Directorate General of Customs and Excise (DGCE) can oversee both daily transactions and daily production from factories. When information of suspicious activities is obtained, DGCE can suspend the purchase of excise stamps. The efforts to fight the illegal cigarette trade also invite stakeholders to be involved by supplying information regarding high-risk areas and regional governments. Strategic communications and community involvement DGCE continuously disseminates information and conducts public education to fight illegal cigarettes. These efforts are conducted every year, using a special campaign slogan. In 2019, the slogan was “Gempur Rokok Ilegal” (“Fight Illegal Cigarettes”). Key performance indicators for DGCE units and offices Parallel with the above-mentioned preventive actions, DGCE also continuously conducts responsive actions: enforcement, investigation and audit activities in cigarette factories. Enforcement activities are planned and measured by consider- ing the limited human resources and the large scale of the monitored territories. To demonstrate the effectiveness of administrative and enforcement measures in curbing the trade of illegal cigarettes, both types of activities are translated into key CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 149 performance indicators for all DGCE working units and offices, including regional offices and personnel. Use of technology and intelligence Coordination between different DGCE offices responsible for monitoring the produc- tion and marketing of cigarettes is maintained by sophisticated IT applications that enable efficient distribution of information and investigation activities. The applica- tions are the Customs Intelligence and Tactical Centre for data and analysis and the Centre for Command and Control (Pusat Komando dan Pengendalian/Puskodal) for ensuring that sea patrols work effectively and efficiently. Independent evaluation To evaluate the efforts and activities to reduce the circulation of illegal cigarettes in Indonesia (e.g. cigarettes without stamps, with fake stamps or with used stamps), in 2016, the government commissioned the University of Gadjah Mada in Yogyakarta to conduct a survey using a stratified random sampling method. To maintain objectiv- ity and independence, an independent body from this well-known university was appointed to conduct the survey. The survey results showed that the level of illegal cigarette circulation in Indonesia was 12.1% of total consumption. In 2018, the DGCE commissioned the University of Gadjah Mada to conduct another survey. Results showed that circulation of illegal cigarettes had been reduced to 7.0%. In 2019, using the same method the university used, DGCE conducted a survey that showed a reduction to 3.0%. Fig. 3.5 shows the results of the surveys. Fig. 3.5 Share of illicit trade in total cigarette consumption in Indonesia, 2016–2019 Source: Customs and excise department, Ministry of Finance, Indonesia, personal communication, 2020. 2016 2017 2018 2019 12.1% 10. 9% 7% 3% 150 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The results of the actions taken can be used as feedback for DGCE in developing excise policies for both service and enforcement, including policies for excise tariffs. DGCE will provide recommendations for an optimum policy format for controlling consumption, maintaining labour protections, optimizing revenue and, most impor- tantly, constantly reducing the consumption of illegal cigarettes on a national level. 3.4.11 PENALTIES Penalties and sanctions must be sufficient to deter illegal activities. Otherwise, finan- cial penalties may simply be paid as a cost of doing business while the illegal activity continues. The Protocol specifies commitments for Parties and provides information on best practices for non-Parties. Article 14.1 in Part IV of the Protocol requires each Party to establish unlawful activities, including manufacturing, wholesaling, broker- ing, selling, transporting, distributing, storing, shipping and importing or exporting tobacco products or manufacturing equipment without the payment of applicable duties or taxes or without using fiscal stamps or other required markings or labels. Articles 14.2 and 15 of the Protocol mandate Parties to determine which of the types of unlawful conduct set out in Article 14.1 shall be criminal offences. Parties must adopt legislative and other measures to give effect to such determinations, as well as to define whether the liability for committing illicit trade in tobacco is a criminal, civil or administrative offence. Article 17 further provides that the Parties shall consider adopting measures as needed to authorize competent authorities to levy penalties in an amount proportionate to lost taxes and duties resulting from the commission of illicit trade. Box 3.14 provides a case study of how Colombia used penalties to fight illicit trade. Box 3.14 The use of penalties to combat illicit trade in Colombia In 2017, the specific tax on cigarettes in Colombia was doubled, increasing from COL$ 700 per pack in 2016 to COL$ 1 400 in 2017. The tax rate was tripled from 2016 levels in 2018, reaching COL$ 2 100 per pack. A provision was added to increase taxes annually after 2018 at the rate of inflation plus 4%. In 2015, before the tax increase, Law 1762 introduced a number of measures to fight illicit trade more effectively. The length of imprisonment for dealing in contraband cigarettes was increased from 3–5 years to 4–12 years. Moreover, government officials who facilitate illicit trade – or anyone involved in transporting or retail sales of illicit tobacco – face similar prison terms. The law allows vehicles used for smuggling to be confiscated, and penalties were increased for illicit trade that is conducted through areas such as special economic zones. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 151 Under the law, illicit trade is considered to be a source of money laundering, which means that the Financial Intelligence Unit can use the same methods it uses to investigate other illegal financial activities. This practice is not common. The law specifically created new sanctions related to alcohol and tobacco excise tax evasion, including the seizure of goods, fines, closure of retail outlets and the suspension or cancellation of licences, authorizations or registries. Arrests and seizures have increased under the new law. Indeed, since its enactment, law authorities reported that between 2016 and 2018, five criminal organizations were dismantled, 53 individuals were apprehended and 72 assets were confiscated. In addition, 2 236 individuals were apprehended and 503 vehicles transporting smuggled goods were confiscated, as transport of such goods is now also considered a crime under the law. More importantly, thanks to the large tax increases, consumption decreased while revenues increased substantially in 2017 and 2018 (see Fig. 3.6). It is estimated that illicit trade in cigarettes in five Colombian cities in 2016 constituted 3.5% of total consumption, a much lower estimate than the industry data suggest. In 2017, after nine months of the tax increase implementation, a similar study found that illicit cigarettes remained low, at 6.4% of total consumption. Fig. 3.6 Packs sold and tobacco tax revenue before and after the tax increase in Colombia, 2016–2018 Sources: (65–67 and Ministry of Finance, Colombia (Direccion de Apoyo Fiscal), personal communication, 2020). 2016 2017 2018 Cigarette sales, million packs Cigarette excise revenue, million US$ current 0 100 200 300 400 500 600 700 800 673.7 521 446.4 195.1 301.2 386.2 33.7% reduction 97.9% increase 152 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N For consumers in possession of illicit tobacco, the minimum penalty should be confiscation and destruction of the illicit tobacco products found in their possession, with payment required for the unpaid tax and duties on those products. In the state of California in the United States, it is illegal to possess a tobacco product on which taxes are due and not yet paid. The burden of proving that taxes have been paid is on those who have the products in their possession. The provi- sion is enforced by the California Department of Tax and Fee Administration and local law enforcement agencies. A violation is a misdemeanour, with a maximum fine of US$ 5 000 and/or up to one year in prison. Illegal packages are subject to seizure and forfeiture. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are using this type of legislation to address illicit trade of tobacco. Asset confiscation and increased penalties for involvement in illicit trade are becoming more common as well. Withholding or even confiscation of trucks involved in smuggling is also common in several countries. KEY TAKEAWAY 17 Penalties and sanctions imposed should be sufficient to deter illegal tobacco trade activities. Penalties should be levied in amounts proportionate to lost taxes and duties resulting from illicit trade. 3.5 TAX ADMINISTRATION OF OTHER TOBACCO PRODUCTS In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. However, there is a lack of standard- ization of other products and sometimes large informal markets. For example, it is estimated that two thirds of waterpipe tobacco in the EU is non-duty-paid (19). Other tobacco products – such as bidis in South-East Asia, waterpipe tobacco in the Eastern Mediterranean region and snus in Sweden – are considered part of a country’s traditions. This sometimes leads to situations where governments are hesitant to strongly regulate and tax these products. Some products, such as kreteks (clove cigarettes) in Indonesia and bidis in India and Bangladesh, are mainly sold in one market. Other tobacco products are more likely to be produced by hand on a small scale, making it difficult to detect and collect taxes on them. The same applies to RYO tobacco, which can be produced on a small scale by hand or with the use of small machinery. The trade in raw tobacco and small-scale home production of tobacco often take place outside of monitoring and control systems (19). As mentioned in section 3.3.1, countries have found various solutions to address this problem, including prior approval for purchase or sale of raw materials and reg- istering, authorizing or licensing of all operators and growers that handle raw tobacco. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 153 3.5.1 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In principle, adding a product to an existing tax framework is not likely to impose significant costs. It is reasonable to expect that challenges similar to those faced in dealing with conventional tobacco products will be faced in the collection of taxes on new products, as market players will attempt to use loopholes in tax regulation to avoid or evade taxes whenever possible. However, new challenges are expected to arise when those new products involve rapidly changing technology and where their market dynamics are widely unknown. Furthermore, taxation of new tobacco products may require additional capacity, as a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. 3.5.2 HEATED TOBACCO PRODUCTS (HTPs) Many countries apply a specific excise on tobacco products according to tobacco weight (see Table 2.4 in Chapter 2). With HTPs, this is likely to impose a challenge, since assessing the content of tobacco in a heated tobacco stick will be an additional burden. From a tax administration perspective, it will be easier for authorities to apply taxes per stick or per unit, as is done for cigarettes. 3.5.3 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS) PRODUCTS Some countries tax only nicotine-containing e-liquids while others tax both nicotine- and non-nicotine-containing e-liquids. Taxing only nicotine-containing e-liquids re- quires laboratory capacity to detect the presence of nicotine (see Table 2.5 of Chapter 2). Self-declarations by industry are not sufficient, since some e-liquids labelled as nicotine-free have been found to contain nicotine (see section 2.4.2). Therefore, it is simpler to tax both nicotine- and non-nicotine-containing e-liquids. One likely challenge of taxing all e-liquids will be the capacity to detect and differentiate whether the e-liquids used in ENDS/ENNDS are falsely declared as being for other purposes at the import and manufacturing levels. More information on advantages and disadvantages of different excise tax policies is given in Table 2.6 of Chapter 2. The challenge in taxing the other components of ENDS/ENNDS products is their diversity (see section 2.4.2 of Chapter 2) and the possibility that some parts may be used for other purposes (e.g. in batteries). As indicated earlier, rapidly changing technology and the lack of control and knowledge of the market make taxation of ENDS/ENNDS devices challenging. It may be for this reason that the majority of countries that tax those products address only the e-liquids. When applying a tax on these newer products, countries should be aware that many customers buy their products online. It is therefore recommended that countries 154 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N deciding to tax these products draw up a proper implementation plan, including how taxes will be collected on imported products and online sales. Online cross- border sales of tobacco products are not permitted in all countries. Several EU countries have banned such sales, which include online sales from retail outlets to consumers in another country. Of course, such bans makes sense only if there is also capacity to enforce them. As in the case of taxing tobacco products, the following actions will be important to more effectively impose taxes on these products: 1. implementing strong enforcement mechanisms such as licensing, recordkeep- ing and control of the supply chain, which can include but is not limited to: a) imposing strict licensing of retailers, importers and manufacturers; ideally, licensing of all those involved in the supply chain and developing a tracking and tracing regime for ENDS/ENNDS products and HTPs (to share cost, this can be done in tandem with the system developed for cigarettes); b) exercising the right to set the frequency and type of audits or controls; c) exercising the right to confiscate goods; and d) imposing sanctions such as penalties, fines and/or withdrawal of licences (if applicable) if legislation is not respected. Specific to ENDS/ENNDS products: 2. implementing highly consequential sanctions for producers who declare nicotine-containing e-liquids as “non-nicotine-containing”; and 3. requiring a fee (contribution to the costs) for laboratory tests when a new product is brought on the market or when there are significant modifications to an existing one. More information on the policy options to apply excise taxes on ENDS/ENNDS is given in Chapter 2, section 2.4.2, Table 2.6. KEY TAKEAWAY 18 In principle, the administration of taxes on new and emerging nicotine products and tobacco products should be similar to that for cigarettes. Due to the lack of standardization of these products, however, a rapid and constantly evolving understanding of them and their supply chain will be required to achieve effective and efficient administration of taxes. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 155 3.6 THE BROADER ELEMENTS OF A GOOD TAX SYSTEM 3.6.1 PROPER RESOURCING OF COMPETENT AUTHORITIES In addition to legal tools and a legal basis on which to act and enforce, the authori- ties in charge of implementing excise tax laws should be provided with sufficient resources to hire the necessary staff to properly implement and enforce them. The necessary staffing could encompass multiple agencies and will often require coop- eration between agencies, since some aspects – such as regulation, licensing and border control – may be performed by agencies other than the competent authority. Staff of competent authorities need the necessary tools, equipment, training and supplies to carry out their functions. This requirement includes the means to build or purchase and maintain a software system that will allow taxpayers to submit required information electronically. Electronic filing has benefits for both taxpayers and authorities. It minimizes the compliance cost for taxpayers and can therefore support voluntary compliance (68). To identify risks of noncompliance, the software system should offer competent authorities the ability to analyse the data submitted by taxpayers and cross-check it with data from other taxes – such as VAT – and third-party sources, such as banks and household surveys. Another option for authorities is to make more efficient use of existing resources. For example, an authority could optimize the risk management system by switching to a risk-based approach: resources could be saved by auditing taxpayers who are more likely to be noncompliant based on risk analysis rather than auditing all of them. Other problems that challenge the effective functioning of a competent authority are lack of a coherent strategy and problems with professionalism related to lack of training or corruption (69). Having a strategy avoids directing resources towards less- important areas. The strategy should always be aligned with the objectives, so that com- petent authorities can identify which steps they should take and in which order they should take them to reach these objectives. A strategy is indispensable to prioritizing and organizing resources so that identified issues or risks can be addressed efficiently. 3.6.2 CORRUPTION Competent authorities should implement tax laws with integrity and have strict rules and regulations for detecting corruption. Strict rules and regulations should also be in place for the punishment of both agency personnel and taxpayers who engage in corrupt practices. Corruption within a competent authority results in the improper monitoring of tax compliance and is one of the causes of the proliferation of illicit trade in tobacco products. It also erodes confidence in competent authorities and ultimately in governments overall. In addition to effective laws and regulation, strong internal audits covering prevention, investigation and sanctions should be 156 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implemented. To improve prevention, a risk map should be created that highlights areas of misconduct and possible leakages. An action plan to update controls should be established to improve areas of weakness detected in procedures and systems. The audits should also be scheduled regularly. Prepared internal auditors with pow- ers to conduct investigations are necessary. Sanctions for corruption, including administrative sanctions and criminal prosecution, must be strong. 3.6.3 A STRONG JUDICIARY The judicial system should be honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. The appeals process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially in the context of illicit trade. KEY TAKEAWAY 19 Broader elements of a good tax system include (1) proper resourcing of competent authorities to hire staff and obtain necessary equipment and systems, (2) strict rules and regulations to detect and punish corruption among both agency personnel and taxpayers and (3) ensuring that the judicial system is honest and independent, with disputes being solved as quickly as possible. 3.7 CONCLUSIONS Policies are more effective if they are properly implemented and enforced. Com- petent authorities have a key role in the achievement of financial and public health objectives of excise taxes. Given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade, this chapter draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (i.e. the Protocol). The Protocol provides a blueprint of measures to address the problem of illicit trade and can be used as a model even by countries that are not Parties to it. Qualities of an effective and efficient tax administration include institutional ar- rangements where roles and responsibilities of competent authorities are clearly defined to avoid overlap and voids. Additionally, effective collaboration among relevant bodies must be facilitated. At the national level, within any organizational arrangement, it is vital that agencies cooperate and exchange information and that their competences find their basis in law. A legal basis for exchange or access to information between government bodies should be ensured. At the international level, especially for border control, the role of customs is key, and access to international CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 157 cooperation agreements such as the Protocol is very useful. An organizational tax administration structure must include a system of performance evaluation and accountability through pre-defined key indicators. To ensure compliance, the accuracy of information for the tax compliance cycle is key, including clear and straightforward taxpayer registration and licensing, declara- tion, recordkeeping, warehousing, distribution, collection and tax refund processes. • Licensing is a powerful tool for obtaining information and securing the supply chain of tobacco products. Ideally, all persons involved in the growing of tobacco and the retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. • Collecting as much information as possible on the business of tobacco and recording all transactions are key to reducing tax evasion, but this may be burdensome for authorities. The use of IT for periodic tax declarations, ac- counting, inventory and financial information is critical for obtaining accurate information and can help decrease the cost of the whole reporting system. • Recordkeeping should be ensured. All persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of those products. • Maintaining a system of authorization for warehousing allows the authori- ties to carry out controls in production and storage facilities to ensure that taxes are paid. Ideally, bonded warehouses should be eliminated from the supply chain. • Duty suspension – which is often applied during the producing, processing, holding, receiving and dispatching of excise goods – should be granted only if strict criteria are met (e.g. for granting authorization, warehouse pre-authoriza- tion visits, adequate stock control measures, checking the origin of excise prod- ucts and the entire production process and coding and marking of products). • To limit the number of taxpayers a competent authority has to manage, tax collection should take place close to the point of production and import. • Refunds for VAT, excise taxes and customs duties are common in most coun- tries, under the principle that taxes are not exported. The refund process must be closely monitored to avoid opportunities for tax evasion. Control and enforcement – key components of tax administration – include a number of measures to secure the supply chain: licensing and due diligence, fiscal markings, 158 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N track and trace, anti-forestalling measures, audits and controls, import and export control and attention to free zones and transhipment points. Control and enforce- ment need to be included as pillars in the strategic plan of the tax administration. Enforcement and control plans must be designed to define the activities and taxpayers that are subject to enforcement and to allocate staffing, auditing, infrastructure and IT resources. Targets must be defined, including the number of interventions and any additional collection or reduction of tax evasion. This includes choosing interventions for those who have a higher probability of noncompliance (the risk- based approach). In the tobacco supply chain, import, export and transfers to and from warehouses may be areas at greater risk of noncompliance. • Licensing provides timely and accurate data that can serve as the basis for audits, since it identifies and controls legitimate operators. The process of licensing control must be carried out and updated periodically – in particular, by controlling the validity of bonds or guarantees and the proper functioning of the required systems and recordkeeping. Where licences are required, the law should include a provision that disallows purchases from unlicensed suppliers or sales to unlicensed purchasers. This means that both suppliers and purchasers will need to verify those with whom they are doing business. This will substantially help to reduce the burden of proof for authorities. In addition, to maintain a high level of control, the validity of licences should be limited in time, making renewals or reapplication required. • Another important measure for controlling and monitoring production and import of tobacco products is the use of fiscal markings (e.g. tax stamps). In addition to increasing compliance with tax laws, fiscal markings can help distinguish between genuine and illicit tobacco products. The use of fis- cal marks enables both the competent authority and the public to monitor whether the taxes on tobacco products have been properly paid. In addition to locally produced and imported products, tobacco products for export should also be required to be marked, but with an indication that they are for export. Requiring a standard package size can facilitate the application of fiscal markings. To lower the chance that fraudsters attempt to re-use fis- cal markings (in particular, stamps) the marking should be applied to each pack of cigarettes (and other tobacco products) before the pack is wrapped with cellophane. Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Fiscal markings should include several security features to make them more difficult to counterfeit. These can include overt, covert, semi-covert and/or forensic features. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 159 • Tracking and tracing systems assist authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as in monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to provide authorities with information on all transactions throughout the entire tobacco product supply chain until duties are paid or other obligations are discharged. Any tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it becomes possible to unambiguously register that product’s movements. A good tracking and tracing system enables the government to properly monitor the supply chain, improve its ability to ensure collection of the proper duties and taxes, authenticate whether the identification marking is genuine and matches the product, improve its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. To reduce the financial burden of implementing such a system, jurisdictions could require the tobacco industry to bear the cost. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, and governments should ensure that the system is independent from the tobacco industry. While the objectives of fiscal markings and tracking and tracing systems are different, stamps increasingly contain tracking and tracing features. • Implementing legal measures to prevent forestalling can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. Forestalling, stockpiling or front-loading occur when manufacturers or im- porters increase their tax-paid stock or oversupply the market by increasing production or imports before a tax increase in order to pay the previous lower rate. • Periodic audits and controls can be implemented to increase compliance. These include cost audits, transfer price audits, price and market monitoring, consumer controls and cross-check controls. • Import and export of tobacco products and manufacturing equipment should be allowed only for duly licensed natural persons or legal entities. The risk of loss of revenue can be mitigated by requiring a guarantee or bond that will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless required fiscal markings (such as tax stamps or export labels) are affixed on the pack, according to the law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Exchange of information between jurisdictions on the movement of goods can also reduce the risk of evasion. 160 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Non-invasive detection equipment (such as X-ray scanners) can be used at customs posts to detect contraband merchandise. A cheaper alternative is the use of dogs that are trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied to reduce contraband. Such measures include the separation of processing operations from the sealed storage of taxed and untaxed products. Within a country, mobile excise control units are helpful in verifying excis- able goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. Physical control operations require close coordination between police, border guards and other public services. • Controls such as regulation and oversight are usually less strict in free zones and transhipment points. This can make free zones appealing to persons involved in illegal cigarette manufacturing or trade. Customs administrations should exercise their authority in free zones to effectively identify and fight illicit trade in tobacco products. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing tracking and tracing regimes and removing exemptions from excise taxes. Other actions include the prohibition of intermingling of tobacco products with non-tobacco products in a single container or other similar transportation unit when the products are removed from free zones. Sale of tax-free or duty-free tobacco products to international travellers should be prohibited, as these sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect govern- ment revenues by creating a loophole in the tax structure. Procedures after detection of illicit trade of tobacco products should be clearly defined. If smuggling or illicit trade is detected through audits, tracking and tracing systems, verification of declarations or border control, actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. It is also important that penalties and sanctions be sufficient to deter illegal activities. Low financial penalties may simply be paid as a cost of doing business while the illegal activity continues. The minimum penalty for consumers in possession of illicit tobacco products should be confiscation and destruction of the products found in their possession and required payment for the unpaid tax and duties on those products. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are taking advantage of this type of legislation and using it to address illicit trade of tobacco as well. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 161 In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. The challenge for taxation of other products includes the lack of standardization of those products and sometimes large informal markets. Knowledge of the product and the supply chain greatly help to facilitate effective tax administration. The trade in raw tobacco and small-scale home production of RYO and other products such as bidis often takes place outside of monitoring and control systems. The best way to address this challenge is to enforce prior approval for purchase or sale of raw materials and a requirement to register, obtain an authorization or license all operators and growers handling raw tobacco. In principle, adding new and emerging nicotine and tobacco products to an exist- ing tax framework is not expected to impose significant costs. It is reasonable to expect that similar challenges will be faced in the collection of taxes on these newer products, as market players will attempt to use the current loopholes in tax regulation to avoid or evade taxes on these products whenever possible. However, challenges are expected to arise, as newer products involve rapidly changing technology, and their market dynamics are widely unknown. Furthermore, additional capacity may be required, since a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. Because the newer nicotine and tobacco products are widely purchased online, countries deciding to tax these products should draw up a proper implementation plan that includes rules on how taxes will be collected on imported products and online sales. Online cross-border sales are not permitted in some countries. The elements of a good tax system include (1) proper resourcing of competent authorities sufficient for hiring the necessary staff to properly implement and en- force excise tax laws; (2) implementation of tax laws with integrity and with strict rules and regulations to detect corruption and for the punishment of both agency personnel and taxpayers who are engaged in corrupt practices; and (3) ensuring that the judicial system is honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. 162 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N REFERENCES 1. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 2. WHO FCTC guidelines for implementation of Article 6 Price and tax measures to reduce the demand for tobacco. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/treaty_instruments/ Guidelines_Article_6_English.pdf?ua=1, accessed 5 October 2020). 3. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 4. Joossens L, Raw M. From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–234 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 5 October 2020). 5. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 6. Reyes IP. Measures to control the tobacco supply chain in the ASEAN: a review of excise tax stamps, track and trace mechanisms, and licensing systems in the ASEAN region. Bangkok: Southeast Asia Tobacco Control Alliance; 2014 (https://seatca.org/dmdocuments/SEATCA%20MEASURE%20BOOK%20. pdf, accessed 5 October 2020). 7. Italy’s tax administration: a review of institutional and governance aspects. Paris: Organisation for Economic Co-operation and Development; 2016 (https://www.oecd.org/tax/administration/italy- tax-administration-a-review-of-institutional-and-governance-aspects.pdf, accessed 5 October 2020). 8. Crandall W, Gavin E, Masters A. ISORA 2016: understanding revenue administration. Washington (DC): International Monetary Fund; 2019 (Departmental Paper Series No. 19/05; (https://www.imf.org/en/ Publications/Departmental-Papers-Policy-Papers/Issues/2019/03/07/ISORA-2016-Understanding- Revenue-Administration-46337, accessed 5 October 2020). 9. Dutch customs in 2018. Heerlen: Customs Administration of the Netherlands; 2019 (https://over-ons. belastingdienst.nl/wp-content/uploads/2019/04/Dutch-Customs-in-2018.pdf, accessed 8 October 2020). 10. ESAs transform the way competent authorities cooperate with each other on AML/CFT matters [press statement]. European Banking Authority; 16 December 2019 (https://eba.europa.eu/esas-transform- way-competent-authorities-cooperate-each-other-amlcft-matters, accessed 5 October 2020). 11. Illicit trade report 2013. Brussels: World Customs Organization; 2014 (http://www.wcoomd.org/-/ media/wco/public/global/pdf/topics/enforcement-and-compliance/activities-and-programmes/ illicit-trade-report/itr_2013_en.pdf?la=en, accessed 5 October 2020). 12. Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation and repealing Directive 77/799/EEC. Brussels: OJEU; 2011. (L 64/1; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:32011L0016&from=en, accessed 29 January 2021). 13. 439 stolen cars seized in 12 days as part of Frontex-led operation. Frontex News Release; 14 October 2019 (https://frontex.europa.eu/media-centre/news-release/439-stolen-cars-seized-in-12-days-as- part-of-frontex-led-operation-vvvirI, accessed 5 October 2020). 14. Focus on: the illicit trafficking of counterfeit goods and transnational organized crime. Vienna: United Nations Office on Drugs and Crime; 2014 (https://www.unodc.org/documents/counterfeit/FocusSheet/ Counterfeit_focussheet_EN_HIRES.pdf, accessed 5 October 2020). 15. Crandall W. Revenue administration: performance measurement in tax administration. Washington (DC): International Monetary Fund; 2010 (Technical Notes and Manuals 10/11; https://www.imf.org/ external/pubs/ft/tnm/2010/tnm1011.pdf, accessed 7 October 2020). 16. Measuring tax gaps 2019 edition – tax gap estimates for 2017-18. London: HM Revenue and Customs; 2019 (https://webarchive.nationalarchives.gov.uk/20200701215139/https://www.gov.uk/government/ statistics/measuring-tax-gaps, accessed 29 January 2021). 17. Countering illicit trade in tobacco products: a guide for policy-makers. Lyon: INTERPOL; 2014 (Legal handbook series; https://www.aph.gov.au/DocumentStore.ashx?id=704d9168-2aa7-4df4-bb81- CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 163 0fe1f357dfa6&subId=408832, accessed 29 January 2021). 18. Resolução- RDC No. 226 de 30 de Abril de 2018 [Resolution of April 30, 2018]. Ministério da Saúde/ Agência Nacional de Vigilância Sanitária/Diretoria Colegiada; 2018 (in Portuguese) (https://www.in.gov. br/materia/-/asset_publisher/Kujrw0TZC2Mb/content/id/12527411/do1-2018-05-02-resolucao-rdc- n-226-de-30-de-abril-de-2018-12527407, accessed 17 February 2021). 19. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 20. Preece R. Addressing the illicit flow of tobacco products in Australia. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 21. Excise on tobacco [website]. Canberra: Australian taxation office; 2019 (https://www.ato.gov.au/ business/excise-on-tobacco/, accessed 5 February 2021). 22. Safe Framework of Standards. Brussels: World Customs Organization; 2018 (http://www.wcoomd. org/-/media/wco/public/global/pdf/topics/facilitation/instruments-and-tools/tools/safe-package/ safe-framework-of-standards.pdf?la=en, accessed 29 January 2021). 23. The Trade Facilitation Agreement. Geneva: World Trade Organization; 2017 (https://www.wto.org/ english/docs_e/legal_e/tfa-nov14_e.htm, accessed 29 January 2021). 24. Authorised economic operator (AEO) [website]. Kildare: Irish Tax and Customs; 2020 (https://www. revenue.ie/en/customs-traders-and-agents/authorised-economic-operators/index.aspx, accessed 7 October 2020). 25. Performance outcome area 1: integrity of the registered taxpayer base. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www. tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 26. Commission implementing regulation (EU) 2016/323 of 24 February 2016 laying down detailed rules on cooperation and exchange of information between member states regarding goods under excise duty suspension pursuant to Council Regulation (EU) No 389/2012. Brussels: OJEU; 2018 (http://data. europa.eu/eli/reg_impl/2016/323/2018-02-15, accessed 7 October 2020). 27. Council Directive 2008/118/EC concerning the general arrangement for excise duty repealing Directive 92/12/EEC. Brussels: OJEU; 2009 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:3 2008L0118&from=en, accessed 29 January 2021). 28. Our strategy. London: HM Revenue & Customs, 2017 (https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/629941/HMRC-Strategy.pdf, accessed 31 January 2021). 29. Strategic plan FY2018-2019. Washington (DC): Internal Revenue Service, 2017 (https://www.irs.gov/ pub/irs-pdf/p3744.pdf, accessed 31 January 2021). 30. Performance outcome area 2: effective risk management. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/ assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 31. Tax administration 2017: comparative information on OECD and other advanced and emerging economies. Paris: Organisation for Economic Co-Operation and Development; 2017. 32. Compliance risk management: managing and improving tax compliance. Paris: Organisation for Economic Co-Operation and Development; 2017 (https://www.oecd.org/tax/administration/33818656. pdf, accessed 7 October 2020). 33. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 34. Evans P. U.K. Fines British American Tobacco for Oversupplying Cigarettes in Belgium. The Wall Street Journal. 13 November 2014 (https://www.wsj.com/articles/u-k-fines-british-american-tobacco-for- oversupplying-cigarettes-in-belgium-1415873470, accessed 31 January 2021). 35. Doward J. BAT fined for oversupplying tobacco in low-tax European jurisdictions. The Guardian. 164 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 16 November 2014 (https://www.theguardian.com/business/2014/nov/16/bat-fined-for-oversupplying- tobacco-in-low-tax-european-jurisdictions, accessed 31 January 2021). 36. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021) 37. Cour des comptes. Accises sur les produits du tabac, Bruxelles [Court of Auditors, Excise duties on tobacco products, Brussels]. Bruxelles: Cour des comptes; 2015 (in French) (https://www.ccrek.be/ docs/2015_27_AccisesProduitsDuTabac.pdf, accessed 7 October 2020). 38. Chaloupka FJ, Edwards SM, Ross H, Diaz M, Kurti M, Xu X, et al. Preventing and reducing illicit tobacco trade in the United States. Atlanta: Centers for Disease Control and Prevention; 2015 (https://www. cdc.gov/tobacco/stateandcommunity/pdfs/illicit-trade-report-121815-508tagged.pdf, accessed 29 January 2021). 39. The Tax Stamp Forum 2015. Hexham: Reconnaissance International (https://10times.com/tax-stamp- forum-miami, accessed 10 September 2020). 40. Tax stamps: a technical study and market report. London: Reconnaissance international; 2012. 41. ЕАНС/2013/Health/ll concerning the provision of an analysis and feasibility assessment regarding EU systems for tracking and tracing of tobacco products and for security features. Brussels: European Commission, 2015 (https://ec.europa.eu/health/sites/health/files/tobacco/docs/2015_tpd_tracking_ tracing_frep_en.pdf, accessed 29 January 2021). 42. Ross H. Controlling illicit tobacco trade: international experience. Economic research informing tobacco control policy. Cape Town: University of Cape Town; 2015 (https://tobacconomics.org/ uploads/misc/2015/05/Ross_International_experience_05.28.15.pdf, accessed 18 December 2020). 43. Security and resilience — authenticity, integrity and trust for products and documents — guidelines for the content, security, issuance and examination of excise tax stamps. Geneva: International Organization for Standardization; 2018. 44. Borkowski F, Twomey C. European Union: an update on EU policies. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 45. Quality management systems — Fundamentals and vocabulary. ISO 9001:2015(en). Geneva: International Organization for Standardization; 2015 (https://www.iso.org/obp/ui/#iso:std:iso:9001:ed- 5:v1:en, accessed 16 February 2021). 46. Protocol to Eliminate Illicit Trade in Tobacco Products: questions & answers. Geneva: World Health Organization; 2019 (https://www.who.int/fctc/protocol/faq/en/, accessed 7 October 2020). 47. Ross H, Eads M, Yates M. Why governments cannot afford Codentify to support their tracking and tracing solutions. Tob Control. 2018;27(6):706–08. 48. Ser vicio de Impuestos Internos, Chi le, Noticias, 28 May 2019. (http://www.si i .c l/ noticias/2019/280519noti01er.htm, accessed 20 October 2020). 49. Directive 2014/40/EU of the European Parliament and of the Council. OJEU. 2014;127:1–38 (https:// ec.europa.eu/health/sites/health/files/tobacco/docs/dir_201440_en.pdf, accessed 7 October 2020). 50. Report on the EU customs enforcement of intellectual property rights: results at the EU border. Luxembourg: Publications Office of the European Union, 2019 (https://ec.europa.eu/taxation_customs/ sites/taxation/files/2019-ipr-report.pdf, accessed 7 October 2020). 51. Matthews P. Forestalling ahead of property tax changes. London: Office for Budget Responsibility; 2016 (OBR Working Paper No. 10; https://obr.uk/docs/dlm_uploads/Working-paper-No.10-1.pdf, accessed 29 January 2021). 52. Ross H, Tesche J, Vellios N. Undermining government tax policies: common strategies employed by the tobacco industry in response to tobacco tax increases. Prev Med. 2017;105S:S19–S22 (https://www. sciencedirect.com/science/article/pii/S0091743517302165?via%3Dihub, accessed 17 February 2021). 53. European Commission v Portugal supported by Belgium, Estonia and Poland, Case C-126/15. Judgement of the Court (Fifth Chamber); 29 June 2017 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?u ri=CELEX:62015CJ0126&from=EN, accessed 31 January 2021). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 165 54. Amalia Valesko v Zollamt Klagenfurt, Case C-140/05. Judgement of the Court (Second Chamber); 5 October 2006 (http://curia.europa.eu/juris/showPdf.jsf;jsessionid=02DEE17CC11845BAE0F42FC7 CFD24C85?text=&docid=65599&pageIndex=0&doclang=en&mode=lst&dir=&occ=first&part=1&c id=2591459, accessed 29 January 2021). 55. European Commission v French Republic, Case C-197/08. Judgement of the Court (Third Chamber); 4 March 2010 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:62008CJ0197&qid =1612223322830&from=EN, accessed 31 January 2021). 56. Commission requests that Belgium amend its legislation on tobacco: letter of formal notice. Brussels: European Commission; 8 November 2018. (Memo 18/6247; https://ec.europa.eu/commission/ presscorner/detail/EN/MEMO_18_6247, accessed 29 January 2021). 57. Commission requests that Belgium amend its legislation on tobacco, MEMO/18/6247 of 8 November 2018. 58. Pederson H, Floristean A, Iseppi L, Dawkins R, Smith C, Mørup C, et al. Study on the measuring and reducing of administrative cost for economic operators and tax authorities and obtaining in parallel a higher level of compliance and security in imposing excise duties on tobacco products. Brussels: European Commission; 2014 (https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/ ramboll-tobacco-study.pdf, accessed 3 October 2020). 59. Money laundering vulnerabilities of free trade zones. Paris: Financial Action Task Force; 2010. https:// www.fatf-gafi.org/media/fatf/documents/reports/ML%20vulnerabilities%20of%20Free%20Trade%20 Zones.pdf, accessed 7 October 2020). 60. Omi K. ‘Extraterritoriality’ of free zones: the necessity for enhanced customs involvement. World Customs Organization; 2019 (WCO research paper no. 47; http://www.wcoomd.org/-/media/wco/public/global/ pdf/topics/facilitation/ressources/permanent-technical-committee/225-226/itemixb_fzs_wco_e. pdf?la=en, accessed 29 January 2021). 61. European Parliament resolution of 26 March 2019 on financial crimes, tax evasion and tax avoidance. European Parliament; 2019 (2018/2121(INI)); http://www.europarl.europa.eu/doceo/document/TA- 8-2019-0240_EN.html, accessed 7 October 2020). 62. International Convention on the Simplification and Harmonization of Customs Procedures. Brussels: World Customs Organization; 2008 (http://www.wcoomd.org/Topics/Facilitation/Instrument%20 and%20Tools/Conventions/pf_revised_kyoto_conv/Kyoto_New, accessed 7 October 2020). 63. Evidence Notice 2004: tobacco duty – tobacco products manufacturing machine licensing scheme. London: HM Revenue & Customs, 2018 (https://www.gov.uk/government/publications/excise-notice- 2004-tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme/excise-notice-2004- tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme, accessed 31 January 2021). 64. Tackling illicit tobacco: from leaf to light: the HM Revenue & Customs and Border Force strategy to tackle tobacco smuggling. London: HM Revenue and Customs; 2015 (https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/418732/Tackling_illicit_ tobacco_-_From_leaf_to_light__2015_.pdf, accessed 7 October 2020). 65. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 66. Cardenas M. Case studies in illicit tobacco trade: Colombia. Chicago: University of Illinois at Chicago; 2020 (Tobacconomics Fact Sheet; https://tobacconomics.org/files/research/606/UIC_Colombia-Illicit- Trade-Fact-Sheet_v1.4.pdf, accessed 29 January 2021). 67. Maldonado N, Llorente BA, Escobar D, Iglesias RM. Smoke signals: monitoring illicit cigarettes and smoking behaviour in Colombia to support tobacco taxes. Tob Control. 2019;29:s243-s248 ( https:// tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/s243.full.pdf, accessed 29 January 2021). 68. Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 69. Pellechio A, Tanzi V. The reform of tax administration. Washington (DC): International Monetary Fund; 1995 (Working Paper no. 95/22; https://www.elibrary.imf.org/view/IMF001/07242- 9781451843941/07242-9781451843941/07242-9781451843941_A001.xml?language=en&redirect=true, accessed 29 January 2021). 166 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.1 COMPOSITION OF TOBACCO PRODUCTS To implement and enforce tobacco taxes in the most efficient way, competent authori- ties should be familiar with all of the components of tobacco products, including each of the raw materials used in their manufacture, production inputs and tobacco manufacturing machinery. Knowledge of the components of excisable products and machinery provides valuable information to identify activities at high risk for non- compliance, implement measures to ensure all taxes are paid and prevent illicit trade. Nicotine, non-nicotine and tobacco products and their component parts In most countries, the ministry of finance determines tax policy, including which tobacco products are taxed, while the ministry of health is responsible for product and use regulation. This could lead to different definitions of the same product, depending on which ministry is responsible for a given law or regulation. Wherever possible, a clear and common definition should be developed to simplify procedures and avoid confusion. Tobacco products take various forms, and not all may be regulated or subject to excise tax in a specific jurisdiction. In addition to cigarettes, other traditional tobacco products include smokeless tobacco – such as chewing tobacco, snuff and snus – as well as bidis and kreteks (clove cigarettes), which can be hand-rolled or manufactured, pipes, hookah or waterpipe and cigars. Cigarettes It is important to understand the materials and component parts of the tobacco products most commonly used in a particular country. Cigarettes are the most common and significant tobacco products in terms of volume and tax revenues in most jurisdictions. A cigarette stick is composed of: • the tobacco blend of various types of tobacco plant (leaves and stem and other plant parts) and additives (including flavours); • the cigarette paper used to wrap the tobacco blend to make up the tobacco rod; • the acetate filter that forms the white portion at the tip of a filtered cigarette, which is in direct contact with the smoker’s mouth; • the tipping paper or wraps around the filter; and • the adhesive that secures the cigarette paper around the tobacco blend and the tipping (1). Each manufacturer follows a specific process to produce cigarettes. Aside from the tobacco blend, manufacturers also vary the size of cigarette paper and tipping paper and the length of acetate filter used per stick (1). In some countries, these elements are standardized. In an ideal regulatory framework, a manufacturer would CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 167 be required to submit information on the specific process for each brand and variant of the tobacco product that it manufactures to the competent authorities as part of the licensing requirement (see section 3.4.3). Authorities could, for example, require manufacturers to submit this information in order to obtain a licence. The minimum requirements of the administration and the information that should be included could be laid down in law or lower regulation to ensure that authorities have the information in their possession for all licensed manufacturers. This information contributes to verifying whether a company is reporting the actual quantity of cigarettes manufactured for sale and sold cigarettes by comparing the amount of materials used for production and the quantities used per cigarette with the total number of manufactured cigarettes. Eventually, this information also contributes to validating whether the taxes are properly paid. Figure A3.1 shows the component parts of a typical machine-made traditional cigarette. Fig. A3.1 Component parts of a machine-made cigarette Source: Author’s compilation. Photo by Walter Klerx. Not all parts of tobacco products are subject to the same level of control. According to Article 6.5 of the Protocol, five years following the entry into force of this Protocol, the MOP shall ensure at its next session that evidence-based research is conducted to ascertain whether any key inputs exist that are essential to the manufacture of tobacco products, are identifiable and can be subject to an effective control mechanism. On the basis of such research, the MOP shall consider appropriate action. In addition to the component parts of tobacco products, materials needed for packaging a specific number of sticks into a pack of cigarettes, usually 20 per pack, can be monitored. These materials include the foil paper, the package paper (which could bear the brand name, design and health warnings), the fiscal marking (if required) and the plastic or cellophane wrap. A fixed number of packs of cigarettes, Filter Tipping paper Cigarette paper Tobacco Tobacco rodFiltration zone 168 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N normally 10 packs, are packed into cartons, also called reams. These cartons are usu- ally made of soft paperboard or cardboard, possibly with branding, and are wrapped in plastic wrap or cellophane. Fifty cartons are packed in master cases, which are made of sturdier and thicker paperboard and stacked on pallets (usually 50 master cases to a pallet). An effective regulatory framework would require manufacturers and importers to provide information to the competent authorities on packaging and design, as well as the number of sticks per pack, carton and master case. Prior approval for purchase or sale of materials used in the cigarette production process can also be required. In the Philippines, suppliers of such raw materials, including those providing tobacco papers and filter components, are required to have a licence (2). In some of the Member States of the EU, raw tobacco is also subject to fiscal and legal requirements. For example, in Slovakia and Poland, raw tobacco can be handled only by authorized operators. While authorized operators do not have to pay excise duties on raw tobacco, if raw tobacco is detected by an unauthorized operator, excise duties will be due. Hungary, Italy and the United Kingdom require registration or authorization for all operators and growers handling raw tobacco (3). In addition to knowing the quantities of inputs required to produce a specific amount of a regulated product (e.g. cigarettes), the competent authority also needs to understand the supply, manufacturing and distribution chains to be able to properly monitor, regulate and determine whether taxes have been paid (see also Fig. 3.3). Novel and emerging nicotine, non-nicotine and tobacco products In recent times, new products have been introduced to several markets, namely, ENDS, ENNDS and HTPs. ENDS usually comprise a nicotine-containing e-liquid but do not contain to- bacco. ENNDS are essentially the same but do not (ostensibly) contain nicotine. The WHO COP requested the Convention Secretariat to invite Parties to monitor and report on scientific, regulatory and market developments such as initiation, cessation, advertising and promotion of ENDS and ENNDS. Furthermore, the COP requested WHO to report on the development of methods by regional and international standards-development organizations for the testing and measuring of contents and emissions of these products (4). There are different types of e-cigarettes – the most common type of ENDS and ENNDS – and currently there are four generations of products. However, they can be divided into two broad categories: open systems and closed systems. Both types of e-cigarette use a wick and a heat source to generate an aerosol. The wick is saturated with e-liquid, and a microprocessor is used to control operations (not all include this). Some e-cigarettes also have an LED light to imitate the burning end of a conventional cigarette (5). Fig. A3.2 presents examples of open and closed systems. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 169 E-liquid pod cover Rechargeable battery Heating element (heats solution, aerosolizing nicotine) Mouthpiece E-liquid pod Fig. A3.2 Examples of open and closed systems of ENDS/ENNDs products Open ENDS/ENNDS system (e-cigarette) Closed ENDS system (e-cigarette) Source: (6). Unlike ENDS/ENNDS, HTPs do contain tobacco. HTPs produce aerosols containing nicotine and toxic chemicals when tobacco is heated or when a device containing tobacco is activated (7). HTPs are composed of two elements: the sticks or pods that contain the tobacco and the device used to heat the tobacco. Both are necessary for the product to be used. Fig. A3.3 shows an example of a heated tobacco product. HTPs are tobacco products and are therefore subject to the regulatory measures contained in the WHO FCTC. Rechargeable battery Power button (to start vaping) Mouthpiece Atomizer / Heating element (heats solution, aerolizing nicotine) E-liquid tank (refillable e-liquid nicotine tank) 170 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. A3.3 Components of an HTP Holder: Tobacco stick: Note: PLA: polyactic acid, MPF: mouthpiece filter. Sources: (8-9). More information on tax administration of other tobacco products is presented in section 3.5. Tobacco stick Holder Charger Casing Control Electronics Battery Heating Blade Heatstick total lenght: 45mm MPF (7 mm) PLA (18 mm) Tobacco plug (12 mm) Hollow acetate tube (8 mm) Outer paper Diameter max 7.42 mm Tipping paper CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 171 REFERENCES 1. Memorandum by Gallaher Group Plc: the tobacco industry and the health risks of smoking. Select Committee on Health Minutes of Evidence. London: UK Parliament, 2000 (https://publications. parliament.uk/pa/cm199900/cmselect/cmhealth/27/0011323.htm, accessed 7 October 2020). 2. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 3. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco.. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 4. Electronic nicotine delivery systems and electronic non-nicotine delivery systems. Geneva: World Health Organization; 2016 (Decision FCTC/COP7(9) of the Conference of the Parties to the WHO Framework Convention on Tobacco Control; https://www.who.int/fctc/cop/cop7/FCTC_COP7_9_EN.pdf?ua=1, accessed 7 October 2020). 5. Brown CJ, Cheng JM. Electronic cigarettes: product characterization and design considerations. Tob Control. 2014;23:ii4-ii10 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3995271/pdf/ tobaccocontrol-2013-051476.pdf, accessed 31 January 2021). 6. E-cigarettes and vapor products [webpage]. King County, Washington (USA); 2019 (https://www. kingcounty.gov/depts/health/tobacco/data/e-cigarettes.aspx, accessed 9 October 2020). 7. Heated tobacco products. Geneva: World Health Organization; 2020 (WHO/HEP/HPR/2020.2 Information sheet; https://apps.who.int/iris/bitstream/handle/10665/331297/WHO-HEP-HPR-2020.2- eng.pdf?sequence=1&isAllowed=y, accessed 7 October 2020). 8. Tobacco heating system (IQOS) briefing document. Silver Spring: US Federal Drug Administration; 2018 (https://www.fda.gov/media/110377/download, accessed 31 January 2021). 9. Premarket tobacco product application: technical project lead review. Silver Spring: US Federal Drug Administration; 2017 (https://www.fda.gov/media/124247/download, accessed 7 October 2020). 172 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.2 EXAMPLE OF FORESTALLING AND COUNTERMEASURES This example is hypothetical but inspired from the situation in the Philippines. The amounts and prices have been altered, however, and it is assumed that the normal inventory kept by a tobacco manufacturer is two months.  The excise tax imposed on a pack of cigarettes in the current year is US$ 3.00; it will be increased to US$ 3.30 at the beginning of the new fiscal year (January in this example). The monthly production of Brand Y cigarettes of company X, which it declares for tax purposes, is as follows: MONTH OF THE CURRENT YEAR PACKS OF CIGARETTES January 10 000 000 February 10 500 000 March 9 900 000 April 11 000 000 May 10 200 000 June 10 600 000 July 9 700 000 August 10 100 000 September 9 900 000 October 10 100 000 November 20 000 000 December 25 000 000 Since the normal inventory is two months, the quantity in the two months prior to the implementation of the new excise tax rate is disregarded. The shelf life of tobacco products is approximately six months. The average of the six months prior to November is computed to obtain the quantity presumed to be produced or imported if there was no tax increase. The quantity from May to October (inclusive) divided by 6 is 10 100 000 packs. Thus, any quantity produced beyond 10 100 000 packs for the months of November and December (the months prior to the implementation of the new tax rate) is assessed using the new tax rate. In this example, 10 100 000 of the packs produced in November will be taxed at the old rate of $3.00, and 9 900 000 packs will be taxed at the new rate of $3.30. For December, 10 100 000 packs will be taxed at $3.00, while 14 900 000 packs will be taxed at $3.30. Without imposing these measures, the government would have been deprived of the excise tax increase on 24 800 000 packs. In addition, the effect of the increase on prices and consumers would have been delayed by approximately two months. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 173 In countries using tax stamps, the withholding of the issuance of stamps is a well-known approach to counter forestalling. Another practical solution is to allow the competent authorities to request advances from the industry to cover revenue shortfalls, provided there is a legal basis for such requests. 174 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 175 CHAPTER 4. Political economy As with any proposed government action, policy-makers need to navigate the political environment of tobacco taxation at every stage of policy development, implementa- tion and administration. While every country’s distinct history, culture, systems and structural forces shape its unique political landscape, there are some universal themes when it comes to tobacco control and particularly tobacco taxation. These themes boil down to the distribution of money, power and resources. The tobacco industry, as both a political and economic player, understands these themes well. The industry has been effective in using principles of the political economy of tobacco taxation in its efforts to block important advancements in tobacco control. Nevertheless, the savvy policy-maker can see through the industry arguments by considering who benefits from industry-favoured policy measures and interven- tions. The industry’s challenges to tobacco tax policies can be organized into the five categories of SCARE tactics. This chapter provides a road map to help policy-makers navigate the political economy of tobacco taxation through each of these themes. The first five sections dissect the tobacco industry framing of each issue, pinpoint- ing the flaws in each argument, identifying the extent to which each concern has merit and suggesting how a responsible government can address each one. These discussions are supported by unbiased evidence from independent, peer-reviewed research, as well as specific examples from country experiences. Sections 4.1 through 4.5 on SCARE tactics will equip policy-makers with the tools they need to proceed with confidence that their tobacco tax policy – developed and implemented fol- lowing the guidelines spelled out in this technical manual – will bring about the greatest health and economic benefits for their constituents, regardless of industry attempts to thwart them. Section 4.6 further buoys policy-makers’ efforts to ensure the beneficial impacts of their policies, as it describes how earmarking can improve the political economy of tobacco taxation by funding programmes and initiatives that promote and support the health and well-being of the population. 176 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.1 SCARE TACTIC S: SMUGGLING AND ILLICIT TRADE 4.1.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases will inevitably result in an increase in the illicit trade in tobacco products (1–2). They claim that higher tax rates and higher prices strengthen the financial incentives for criminal enterprises to supply cigarettes from lower-tax jurisdictions, boost domestic tax evasion and encourage smokers to seek cheaper illegal cigarettes. The industry also challenges the argument that tobacco tax hikes increase government revenue by claiming that the presence of an illicit tobacco market will actually reduce revenue collection following a tax increase. More recent versions of this argument – adapted to address public health concerns about tobacco use – claim that illicit market growth also offsets reductions in smoking prevalence that would otherwise be brought about by tobacco tax increases. In summary, the tobacco industry and its allies claim that raising tobacco taxes are ineffective – and even counterproductive – because they are circumvented by illicit markets, which prevents the government from achieving its public health objective and reduces rather than increases tax revenues. When a country considers a proposal to increase tobacco excises, the tobacco industry and its allies frequently make exaggerated claims about the size and scope of illicit tobacco trade in that country. Opponents of tobacco tax increases argue that price differentials are the exclusive – or at least the dominant – cause of illicit trade. Influenced by this fear-inducing faulty diagnosis, tax authorities frequently find it difficult to make decisions about tobacco taxes. However, the industry diagnosis always contains the same erroneous elements. First, the illicit trade in a country is frequently less than the industry portrays it to be, and the country’s tax enforcement policy towards tobacco products is rarely unique or in any way different from the norm in the country (3). Second, the scale of illicit trade in tobacco is not exclusively or even primarily determined by tax or price differentials. Typically, it results from a set of governance problems characterized by government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, lack of dissuasive sanctions and/or weak norms regarding participation in illegal and informal markets (4–7). This section provides guidance for tax and other relevant authorities on how to respond to the tobacco industry SCARE tactic that increasing tobacco taxes will lead to smuggling and illicit trade in their countries. Tax authorities need to know the nature, causes and extent of illicit trade so that they can define the problem properly and formulate an appropriate response. This chapter addresses the available tools for better defining and understanding specific illicit trade problems, particularly tools that facilitate independent assessment of the magnitude of that trade. Improvements CHAP T ER 4. PO LI T I C AL ECO N OMY 177 to governance within the realm of tax authorities – such as best practices in tobacco tax administration and policies to improve the effectiveness of fiscal regulations or norms regarding participation in informal and illegal markets – are discussed in Chapter 3. This section first describes the nature of the illicit tobacco trade to highlight some of its complexities and identify complementary policies for tackling the problem. Next, evidence that calls into question the link between illicit trade and high prices or tax rate changes is discussed. Finally, to help tax authorities assess their own situation, several different methodologies are presented to estimate the scope of the illicit tobacco trade and to evaluate estimates of that trade for a particular country or tax jurisdiction. 4.1.2 THE NATURE AND EXTENT OF THE ILLICIT TOBACCO TRADE The WHO FCTC defines illicit trade as: any practice or conduct prohibited by law and which relates to production, ship- ment, receipt, possession, distribution, sale or purchase, including any practice or conduct intended to facilitate such activity (8). Non-duty-paid tobacco products found in a jurisdiction (i.e. through littered-pack surveys) could be the result of either of two related but distinct activities: tax evasion and tax avoidance. Tax evasion is a set of unlawful actions seeking the non-payment of tobacco taxes and duties, whereas tax avoidance comprises legal actions with the purpose of avoiding payment of some or all taxes, such as bringing an amount of cigarettes up to the legal allowance from a lower- into a higher-tax jurisdiction. Tax avoidance is not illegal and is therefore not considered part of illicit trade in tobacco products.1 The focus of this section is on tax evasion activities, which can occur in the movement across borders or in domestic production and distribution. When tax evasion happens across borders, it is known as smuggling (9) and can be done on a large scale or a small scale. Tax evasion in the domestic market can be partial, when licensed and authorized producers or distributors comply with only part of their tax obligations, or total, when the whole production and distribution system is illegal and out of sight of tax administrators (5, 10–14).2 Large-scale tax evasion schemes can be run by different types of producers and their associated distributors, such as the transnational tobacco companies (TTCs) and their national subsidiaries, 1 Tax avoidance practices –common among states in the United States and countries in the EU – are not analysed in this section. 2 Tax evasion is normally considered as illicit manufacturing in the literature (14). 178 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N other local tobacco companies producing their own brands and illegal factories that normally counterfeit other brands or produce illicit brands. Large-scale smuggling involves, for example, taking advantage of tax-free zones and mislabelling shipping products prior to or during transit (11) or using so- phisticated clandestine networks. This form of tax evasion is systemic and can be carried out by TTCs (12), by local companies producing in countries with low tax enforcement, e.g. Paraguay (15) that feed neighbouring countries and regional illicit hubs through a network of clandestine distributors (15) or by companies located in tax-free zones like Jebel Ali and Dubai in the UAE (16) or such zones in Russia or Cyprus (17). The origins, routes and quantities of large-scale smuggling frequently change as affected countries or markets react by strengthening enforcement and seeking bilateral cooperation with the jurisdictions of origin. For example, the 2013 comprehensive strategy of the EU (10) enhanced bilateral cooperation with major source and transit countries of illicit cigarettes coming to Europe, including Russia, Belarus and Ukraine. These agreements improved day-to-day cross-border coopera- tion, reduced illegal flows and introduced gradual tobacco excise rate adjustments in those three countries to bring them to European levels. Small-scale smuggling (also known as ant smuggling or bootlegging) is the cross-border trafficking of cigarettes in quantities that are larger than the allowable limits (e.g. two cartons) but smaller than large shipments (e.g. truckloads, cargo containers), normally for the purpose of selling at a profit (11). This type of illicit trade may exist in places where there are opportunities within neighbouring tax jurisdictions. For example, small-scale smuggling is commonly done by individuals living in French and German provinces near lower-taxed countries (e.g. Belgium, Luxembourg, Switzerland, Spain, Poland and Czechia) (18). Counterfeiting is a form of illicit manufacturing that involves the production of tobacco products (including packaging and tobacco filler) without the approval of the trademark holder (13). Another product of illicit manufacturing is so-called cheap or illicit whites. Cheap whites are branded (e.g. Jin Ling) or unbranded cigarettes that are legally or illegally produced3 and knowingly sold in the illicit market (17). Cheap whites are not usually produced by TTCs (17, 19).4 They are produced by small tobacco produc- tion companies in one country and often sold in illegal markets of neighbouring 3 Ross et al. (17) analysed this issue and found that the sale to the first purchaser is usually legal. Their analysis covers the production in free zones (i.e. in the UAE, Russia and Cyprus) and production exported from Viet Nam, Indonesia and China. In those cases, there is no need to make the first sale illegally. However, cheap white production in Paraguay is sold to domestic distributors, and most of those sales are completely illegal. 4 Ross et al. (17) and Gilmore et al. (19) identify some cheap white brands sold by TTCs, such as President (PMI), produced in Ukraine, and Esse (Korea Tobacco & Ginseng Company, KT&G), produced in Indonesia. CHAP T ER 4. PO LI T I C AL ECO N OMY 179 countries. For example, in Paraguay, cheap whites are produced on a large scale by a few companies under the guise that they are marketed domestically, but a large share is smuggled into Uruguay and Brazil (4). Iglesias et al. (20) showed how TTCs’ cheap brands were illicitly shipped through Paraguay to be sold in the Brazilian and Argentine markets in the 1990s. This contributed to increased production of cheap whites in Paraguayan firms, which continued the illicit business even after Brazilian legislation obstructed the illegal activity of the TTCs. Domestic tax evasion is a pervasive phenomenon, particularly in LMICs. Partial tax evasion in tobacco products can be found at any level of tax rates or prices and is generally the result of defective legislation or weak tax enforcement.5 Complete or total tax evasion occurs when producers and distributors are clandestine or when there are serious institutional challenges to tax enforcement between two tax jurisdictions, such as between the United States and Native American Reservations. Evidence of illicit manufacturing has increased in recent years in several places in the world, including the EU (10) and Brazil (21). TTCs were predominant in illicit trade activity until the end of the 20th century, and even with the entrance of new actors into the illicit business, TTCs have not entirely exited. Gilmore et al. (22) analysed industry-funded data and seizure data and concluded that TTCs are still involved in illicit trade in Europe, despite the Anti-Contraband and Anti-Counterfeit Agreements (the “Agreements”) signed between the four TTCs and the EU (23).6 Using industry-funded data, Gilmore et al. show that 58% of illicit EU cigarettes can be attributed to the four main TTCs. When seizure data are used, 69% to 73% of illicit EU cigarettes can be attributed to these firms (22). It is always difficult to assess the extent of the global illicit tobacco trade because of its illegality, its global and changing nature and problems with data collection (24). Before the 21st century, when TTCs were almost unique actors in the large-scale smuggling of well-known cigarette brands, the difference between global exports and imports of cigarettes could provide a good approximation of the size of this problem globally (7). However, with the growth of illicit manufacturing in general, the manufacturing of cheap whites and the illegal movements of those products 5 This occurs when licensed and authorized producers underreport actual quantities and sell the non- duty-paid produced quantities through illegal channels. It can also include instances when producers do not report quantities at all, as in many ad valorem systems of LMICs. 6 From 2000 on, the European Commission and 10 Member States launched court cases regarding smuggling and money laundering against several TTCs. To end the court cases, the Anti-Contraband and Anti-Counterfeit Agreements were signed, which required the TTCs to exercise stringent control over their supply chain (through tracking and tracing, due diligence and anti-money-laundering and reporting obligations), share operational intelligence with Member States and the EU and pay penalties for seizures, as well as annual payments over a period of 12 years. The agreement with PMI has ended, the one with JTI will end in 2022 and the others with Imperial Tobacco and BAT will run until 2030 (23). 180 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N over the past two decades, trade statistics are no longer as useful as they were in the past. Joossens et al. (25) tried to estimate the size of the global illicit cigarette market by adding different types of national estimations prepared around 2007. They found that the estimated size was 657 billion cigarettes per year, or 11.6% of the global cigarette market. According to Joossens et al., illicit trade costs govern- ments US$ 40.5 billion in tax revenue worldwide, and eliminating illicit tobacco trade would recover US$ 13 billion in immediate revenue in high-income countries and US$ 18.3 billion in LMICs.7 Descriptions of the types of illicit trade are useful for developing the first com- ponent of a strategy to fight it: assess the nature and size of the problem. Table 4.1 presents all the main components of a strategy to fight illicit trade. To make progress in this first component – knowing the problem – authorities could use and adapt existing instruments of health surveillance or seek partnerships with academia and independent specialists to investigate the issues involved, using different methodolo- gies (see subsection 4.1.4 and Annex 4.1 on methodologies to assess the nature and size of the problem). Knowing the nature of the problem requires the cooperation of different government actors – for example, to investigate both the financial and criminal operations of organized crime behind the illicit trade. The gathering of qualitative information on the nature of the illicit trade should start simultaneously with the statistical work of measuring the magnitude of the problem. Table 4.1 Components of a strategy to fight the illicit tobacco trade 1. Assess the nature and size of the problem Use and adapt existing health surveillance and other existing national surveys to assess the problem Seek partnerships with academia and independent specialists to find ways to rigorously study illicit trade Use financial and police investigations to identify and fight organized crime operating in illicit trade 2. Start identifying and implementing appropriate country-specific policies and strategies to address illicit trade Improve tax and customs administration to close the legal and administrative loopholes facilitating illicit trade Implement other appropriate policies to deal with country-specific problems 3. Become a Party and/or implement the Protocol to Eliminate Illicit Trade in Tobacco Products Adapt the Protocol supply-chain control obligations Adjust national penalties for illicit trade offences Seek and build international cooperation 7 The WCO publishes an Illicit Trade Report annually, with the main characteristics and trends of illicit flows in key products, including tobacco, using data based on customs seizures. CHAP T ER 4. PO LI T I C AL ECO N OMY 181 Methodologies available to estimate the nature and size of illicit trade are discussed in subsection 4.1.4. This is the first step for dealing with SCARE tactic S. Chapter 3 discusses at length the relevant tax administration measures and best practices to minimize opportunities for illicit trade in tobacco products. Table 4.2 presents examples of appropriate policies and strategies targeted to address specific types of illicit trade in addition to the best practices described in Chapter 3. After completing the first step of this strategy, tax, health and justice authorities should discuss how to face country-specific problems, considering not only tax and customs administration measures but also social, law enforcement and international cooperation policies and strategies. Table 4.2 Suggested policies and strategies to address country-specific illicit trade problems MAIN TYPE OF ILLICIT TRADE IN THE JURISDICTION PROBLEMS POLICIES/STRATEGIES TO USE Bootlegging Neighbouring low-tax jurisdiction Bilateral negotiations to harmonize tobacco tax systems Difficulty of controlling people’s movements in countries with extensive land borders Identify and establish suitable social protection or employment policies for targeted populations in border regions Extensive land border with multiple accesses Bilateral cooperation with law enforcement and border control forces, monitoring of access routes to main consumption markets Large-scale smuggling from neighbouring jurisdiction Neighbouring low-tax jurisdiction and difficulties in controlling borders Bilateral negotiations to harmonize tobacco tax systems and bilateral law enforcement cooperation Producers and distributors in the lower-tax jurisdiction aiming to supply the high-tax jurisdiction Bilateral cooperation to harmonize tax systems and control producers and distributors in the origin country, create conditions for legal exports and taxed imports Large-scale smuggling from a third country or tax-free zones Producers and distributors aiming to supply non-duty-paid tobacco products wherever possible Customs and other forms of international cooperation to control and monitor exports from identified areas Domestic tax evasion Existence of many small informal or semi-formal producers Encourage business concentration through producer associations and cooperatives, create incentives for formalization and establish licensing rules and basic electronic information systems for raw material and production 182 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Underreporting from formal producers Improve tax administration with policies such as basic electronic information systems for inputs and production, establish neutral procedures to verify production, improve audit systems, increase third-party information on inputs and production of tobacco products Clandestine factories Law enforcement investigation of commercial associations with raw- material and machine producers and distributors Governments should try to identify the incentives and governance problems that encourage and allow illicit trade movement inside their country. As seen in Table 4.2, the design and implementation of policies to deal with those problems do not depend exclusively on tax and customs authorities; they also depend on the efforts of the police and law enforcement, the Justice Department and the judicial apparatus. In other words, a great deal of coordination and consultation among different types of government bodies and expertise is needed to produce an adequate response. It is also clear from Table 4.2 that domestic tax evasion by formal producers can be tackled by tax authorities and is mainly related to the supply-chain-control provisions of the Protocol. The Protocol (26) builds upon and complements Article 15 of the WHO FCTC, which addresses means of countering illicit trade in tobacco products as a key aspect of a comprehensive tobacco control strategy. The Protocol is a blueprint of measures to deal with this problem, and its provisions should be part of any strategy for fighting the illicit market. It is a legally binding treaty in its own right that entered into force on 25 September 2018. As described in detail in Chapter 3, the Protocol has three main lines of action: supply-chain controls, recommendations on how to treat unlawful conduct related to the illicit tobacco trade and suggested mechanisms to seek and build international cooperation to fight that trade. Countries can start implementing Article 15 of the WHO FCTC and the appropriate polices or strate- gies recommended by the Protocol even before acceding to it, selecting those most suitable to the nature and extent of their particular problem. Such transitional work will facilitate the eventual implementation of the Protocol, because any plan to correct loopholes in tax and customs practices will bring government authorities closer to the best practices recommended in the Protocol. 4.1.3 DETERMINANTS OF TAX EVASION: THE ROLE OF PRICE LEVELS The argument that price and tax rates are the main determinants of the illicit tobacco trade has persuaded some governments (e.g. Uruguay and Georgia in the past) to avoid policies that may lead to cigarette price increases (e.g. excise tax rate increases) (4). CHAP T ER 4. PO LI T I C AL ECO N OMY 183 Some governments (e.g. Canada in 1994, Brazil in 1999 and Pakistan in 2017) (20, 27) have even reduced tax rates in attempt to reduce the illicit trade. The wider scholarly literature demonstrates that illicit trade is not a monocausal phenomenon (7) but is the result of many factors, most of them related to gover- nance issues. Government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, weak norms regarding participation in illegal and informal markets and conflicts between neighbouring countries (5) all contribute to the existence and growth of the illicit tobacco trade. It is difficult to isolate the role of price from each of the other factors because (1) obtaining prices and quantity measures of illicit trade is inherently challenging; (2) in most countries, there are many cigarette brands, and prices vary between and even among brands; and (3) there is a lack of good measures to deal with nonprice factors affecting illicit trade, such as government corruption and ineffective criminal justice. These constraints make it challenging to develop rigorous empirical evidence about how price and other factors affect illicit trade. Despite these fundamental challenges, the economic literature has produced credible evidence that price is always only one factor – and often not the most important factor – determining the extent of illicit trade. Many econometric studies about the influence of price and other factors have focused on cross-border shopping (or small-scale bootlegging from low- to high-tax jurisdictions), given the availability in the United States and Europe8 of sales data for low- and high-tax jurisdictions, classified in a convenient way by geographical zones – i.e. close to or far from the borders. Those studies attempted to explain the illicit trade flows or the relatively higher sales in low-tax jurisdictions as a function of price and tax differentials between the lower-tax and surrounding higher-tax jurisdictions, after controlling for other important factors affecting cross-border sales such as proximity to borders and levels of corruption (6, 7, 11, 28).9 The main conclusion of the studies is that illicit trade flows are not linked solely to price (29). Some show a significant effect of price differentials together with other factors, but others do not find significant price differential effects. The important policy implication of these analyses is that decreasing tobacco tax rates and real prices in higher-tax jurisdictions could have minimal or no effect on illicit market shares.10 8 This was a traditional strand of the literature in the United States on trade among states, and to a lesser extent in European countries, most of which used conventional but inaccurate illicit trade measurements. 9 Recently, PMI-Altria financed some studies of factors affecting cross-border sales. One of those studies, Prieger and Kulic (28), criticized Merriman et al. (2000) (9) and arrived at the conclusion that in cross-border shopping, price differentials are important for determining the magnitude of illicit trade. 10 Brazil decreased tax rates and real prices at the beginning of the 21st century to fight illicit trade coming from lower-tax jurisdictions. After this action, however, the government lost revenues, and the size and scope of illicit trade remained unaltered, according to industry sources (20). 184 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Observational and case studies provide information that may improve public policy even when they are unable to produce compelling evidence of causal relation- ships. Some observational studies have correlated price levels with illicit market shares, using large samples of countries. Joossens et al. (25) found that countries with high taxes and prices normally have lower shares of illicit trade than countries with lower tax shares and prices. In their sample, high-income countries generally have relatively high cigarette prices and tax shares, but their favourable results (i.e. lower levels of illicit trade) are related to effective tax administration and lower corruption levels. In contrast, LMICs generally have lower prices and tax shares, along with significant illegal market shares. Joossens et al. attribute difficulties in fighting illicit trade to weak tax and customs administrations and, in most cases, institutional and legal challenges (25). Figure 4.1.1 illustrates the relationship between price and illicit trade, using the price (in US$) per pack of the most-sold brand of cigarettes and the estimated level of illicit trade for 94 countries in 2018.11 There is no apparent unique association between the two variables. Running a linear regression with retail price as the explana- tory variable and share of illicit trade as the dependent variable shows an inverse, but not statistically significant, relationship between price and illicit market share.12 Figure 4.1.1 illustrates some particular cases: • Many countries with low prices (i.e. lower than US$ 2 per pack) have the highest levels of illicit trade in the sample, e.g. Brazil (BRA) ($1.33 and 46.3% illicit share), Pakistan (PAK) ($0.39 and 40%), Ethiopia (ETH) ($0.55 and 32.9%), Ghana (GHA) ($1.06 and 29%) and Cameroon (CMR) ($0.89 and 25%). • In contrast, many of the countries with prices between US$ 4 and $8 – which could be considered high enough for financial incentives to operate – have illicit trade shares of less than 10% of total consumption. These countries include the Republic of Korea (KOR) ($4.02 and 0.8%), Czechia (CZE) ($4.31 and 2.9%) and Sri Lanka (LKA) ($6.89 and 1.6%). • All countries that have very high prices – higher than US$ 8 – except for Ireland, register illicit trade shares below 20%. These countries include France (FRA) ($9.39 and 17.8%), Switzerland (CHE) ($8.71 and 5.5%), Singapore (SGP) ($10.35 and 3.7%) and Norway (NOR) ($14.51 and 9.6%). 11 National estimates of the magnitude of illicit trade are controversial. The tobacco industry’s numbers overestimate the problem and are based on questionable methodologies. Estimates with a rigorous and transparent methodology are not available for a large sample of countries for the same year. In order to compare price levels with illicit market shares, Euromonitor’s estimations of illicit market share were selected, for two reasons: they are comparable estimates for a large sample of countries in a given year, and no one could argue that they are biased towards tobacco control’s points of view. The use of Euromonitor data does not imply that WHO fully agrees with all the details and methodologies used to obtain them. 12 Other factors must be taken into account to transform this observational analysis into a rigorous analysis of cause and effect. CHAP T ER 4. PO LI T I C AL ECO N OMY 185 Fig. 4.1.1 Share of illicit trade versus retail price of the most-sold brand of cigarettes in US$, by country, 2018 Note: The extent of illicit trade in cigarettes is measured by Euromonitor as the estimated quantity of illegal cigarettes consumed in a country divided by the estimated total consumption of cigarettes in that country. Sources: (27, 30). 51 2 3 4 6 7 8 9 10 11 12 13 14 1615 0 10 20 25 15 5 30 35 40 45 50 MYS BRA ECU ECUPAK ETH PAN CRI URY GRCIND GTM SLV ARE LVAHND NGA ZAF CAN FRA LBNAGO IRL DOM GHA CMR Retail price, USD per pack Ili ci t t ra de % IRQ VNM COL MMR AUT SWE EST PER ESP ROM POL CHL ISR GBR KEN LAO KHM EGY MKD TZA THA TUN BGD UZB UKR GEO KAZ CHN KWT HRVJOR CZE SVK KOR JPN BLR SRB TUR OMN FIN NLD CHE SGP SAU USA LKA DNK DEU ITA HUN SVN PRT BOL DZAIDN CIV BEL MAR ARG AUS NOR NZL LTU BIH MEX BGRPHLAZE RUS 186 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As indicated in numerous studies and analyses of illicit trade in tobacco products (4–5), the most effective way to tackle the problem is not to forgo tax increases but rather to strengthen the capacity to fight the trade. Therefore, it is important to consider the relationship between good governance and illicit trade. The more capacity a country has to counter illicit trade in general, the lower the level of that trade will be. An index compiled by the Economist Intelligence Unit (EIU), the Global Illicit Trade Environment Index, measures countries’ structural capacity to fight illicit trade overall. The EIU indicator is a combination of four indicators or categories designed to assess countries’ performance in those areas; the closer the overall indicator is to 100, the better the country’s capacity to fight illicit trade. The four categories are:13 1. government policy, which measures the government’s commitment to pro- actively monitoring and preventing illicit trade; 2. supply and demand, which measures the extent to which the domestic en- vironment discourages or encourages supply and demand for illicit goods; 3. transparency and trade, which measures transparency and the degree of governance applicable to free-trade zones and transhipments; and 4. customs environment, which measures how effectively customs services facilitate legitimate trade while at the same time preventing illicit trade. Figure 4.1.2 illustrates the relationship between the EIU indicator and the estimated level of illicit trade in cigarettes in a set of countries. There is an inverse and statisti- cally significant relationship between the indicator and the estimated level of illicit trade in cigarettes. This suggests that as the capacity to fight illicit trade in general increases, the illicit trade in cigarettes falls.14 13 For more details about this indicator, visit http://illicittradeindex.eiu.com/. 14 The association was significantly different from zero at a 90% confidence level, using a linear regression between the two variables. CHAP T ER 4. PO LI T I C AL ECO N OMY 187 Fig. 4.1.2 Share of illicit trade versus the EIU indicator in 70 countries, by country, 2018 Sources: (30–31). % Il ic it tr ad e ci ga re tt es MYS 0 10 15 5 20 25 30 35 40 45 50 55 60 EIU GIT indicator 20 3010 40 50 60 70 80 90 JPN KOR HRV SAU UKR BLR KHM LAO MMR IRQ GTM DOM VNM MAR SRB PER BGR MEX ROM COL ZAF BIH PHL IDN ITA PRT HUN BEL DEU ESP ARG AUS SWECHL TW POL ISR AUTLTU FRA CAN IRL LVA ARE GRC IND URY CRI PAN PAK ECU BRA EST GBR FIN SVN NLD SVK TUN RUS DZA TUR CHNKAZ THA CZE USA SGP NZL DNK 188 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Rigorous independent research has established that despite the challenges of illicit trade, taxation of tobacco products is an effective public health intervention that substantially reduces tobacco use and generates government revenue (5). Further, when cigarette taxes increase, governments generate higher revenue and consumption is reduced (32–33). However, ineffective tax administration can allow illicit trade to grow and can undermine some of the benefits of tobacco taxation by making cheaper cigarettes available. For example, the average street price of smuggled cigarettes in Malaysia is 55% lower than its legal tax-paid equivalent (34). Illicit tobacco trade also reduces government tax revenue and may increase health costs associated with smoking and costs associated with policing. 4.1.4 MEASURING ILLICIT TRADE IN TOBACCO PRODUCTS The magnitude of illicit trade is a powerful argument in tax policy discussions, and for this reason the tobacco industry funds estimation of illicit trade in countries or regions of particular interest to itself (i.e. Project Sun and Project Star in the EU and Oxford Economics in East Asia). However, a recent systematic review of industry data on illicit trade finds substantial methodological weaknesses in industry-commissioned reports (24). Furthermore, Blecher et al. (35) argue that industry-funded studies tend to systematically overestimate the size of illicit trade to persuade authorities to abandon tobacco tax reforms. Independent researchers have also uncovered inconsistencies in tobacco-industry-funded estimates (36). Some examples of inflated industry-linked illicit trade estimates are given in Table 4.3, which compares peer-reviewed and independent studies with estimates funded by the tobacco industry. Because some countries have several industry estimates from different sources or years, Table 4.3 presents the estimate included in the article that published the independent study, because it was considered as representative and adequate to illustrate the overestimation. In all cases, the industry estimates exceed those of the independent studies. Measuring the scale of illicit trade can be a daunting task for governments because different methods are employed by independent researchers, governments and the tobacco industry. Nonetheless, it is worth investing in these studies because they drive policy discussions and can be used to evaluate the impact of policies (e.g. tax increases, plain packaging and health warnings). CHAP T ER 4. PO LI T I C AL ECO N OMY 189 Table 4.3 Illicit market share estimated in independent studies compared with estimates in tobacco-industry-funded studies COUNTRY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDEPENDENT STUDIES SOURCE OF THE INDEPENDENT STUDY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDUSTRY- FUNDED STUDIES INSTITUTION RESPONSIBLE FOR THE INDUSTRY- FUNDED STUDIES Colombia 3.5% of the total market in five cities, 2016 Maldonado et al., 2018 (37) 13% of the total market, 2014 FND and INVAMER, 2015 Chile 16.3% of the total market in the Metropolitan Region of Santiago, 2017 Paraje et al., 2020 (38) 24.3% of the total market, 2017 Observatorio del Comercio Ilícito BATC, 2017 Brazil 28.8% of the total market, 2014 Iglesias et al, 2017 (39) 34%, of the total market, 2014 BAT public statement, 2015 Mexico 8.8% of the total market in eight major cities, 2017 Saenz de Miera Juarez et al., 2020 (40) 16.6% of the total market, 2012 Confederación de Cámaras Industriales, 2012 As shown in Table 4.4, methodologies to measure illicit trade can be grouped into three types: (1) direct measurement; (2) residual methods and (3) expert opinion (12). Direct measurements rely on evidence directly linked to actual illicit behaviour and pack observation; residual methods infer evasion based on theory and evidence about consumption and legal sales; and expert opinion distills information garnered from talking to individuals with the most direct knowledge of the tobacco market. Each method has advantages and disadvantages. No single method is unambigu- ously superior to others, but direct measurement and residual methods are more conducive to determining the size of the illicit market, whereas expert opinion could provide insight into the details of the market’s operations.15 Table 4.4 presents the relative amount of resources and the degree of expertise required to implement each main measurement method, as well as the primary purpose, data collection characteristics, sampling features and unit of analysis. A brief description of each of the methods is presented in Annex 4.1. Merriman (11) and Ross (9) provide more expansive details. There is no simple selection rule for deciding what measurement method to use. The major factors to consider when selecting a method or methods include (1) the nature and characteristics of the illicit trade problem (i.e. where and how the 15 In interviews with experts from the tobacco industry, provisions of Article 5.3 of the FCTC and its Guidelines need to be followed. 190 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N problem manifests and whether domestic tax evasion or illegal inflows of foreign brands or a combination of both predominates), (2) previously collected data, (3) available budget and (4) expertise of available analysts. Available budget and staff skills are often the main restrictions that governments face. Therefore, Table 4.4 orders the measurement methods according to resources needed and available expertise. For example, residual methods and expert opinion can provide crude but useful estimates at low cost and require the lowest levels of technical sophistication. Another low-cost option for countries that employ popula- tion health surveillance surveys is to add questions to measure illicit trade, such as brand name, value and quantities of the last purchase. In contrast, the direct measurement approach often requires sophisticated research designs and expensive (and time-consuming) field research. Table 4.4 Overview of resources and expertise needed and main purpose of measurement methods METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Seizures (D) $ Low Identify trends in types of products, transporta- tion methods, points of entry and brand names Secondary data use Non- probability Shipments Use of existing health sur- veillance surveys – self- reported consump- tion (D) $ Low, only additional questions Size of illicit trade, adding or improving questions on brands, value and quantities of the last purchase Additional primary data collection Probability Individuals Gap analysis (R) $ Medium Provides a measure of changes in illicit trade Secondary data use Universe Nations Econo- metric modelling (R) $ High Estimation of price elasticity of substitution from tax- paid to illicit products Secondary data use Universe Geography Expert interviews (E) $ Low Characteristics of the illicit trade Primary data collection Non- probability Individuals Smoker intercepts and pack observa- tion surveys (D) $$$ Medium Size and characteristics of illicit trade, probability-based sample to be representative of population Primary data collection Probability Individuals CHAP T ER 4. PO LI T I C AL ECO N OMY 191 METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Pack return and swap surveys (D) $$$ Medium Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Littered- pack surveys (D) $$$ Medium Size and characteristics of illicit trade, compa- rability with industry estimation using empty-pack surveys Primary data collection Probability Individuals Covert purchases (D) $$$ Medium Type of products and trade channels of illicit trade Primary data collection Probability Geography Self-report consumer surveys (D) $$$ High Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Notes: Universe includes total population; D = direct measurement, R = residual method, E = expert opinion. Scale for resource costs assuming a moderately sized study (e.g. a representative study of a region of several million): $ (cheapest) – weeks of skilled labour hours; $$ (moderately expensive) – 1 to 2 months of skilled labour hours; and $$$ (most expensive) – 6 to 12 months of skilled and unskilled labour hours. A more detailed description of the different measurement methods is given in Annex 4.1. To further assist responsible authorities in deciding which method to select, Table 4.5 presents the key characteristics of each of the measurement methods, along with the main advantages and disadvantages of each. Countries may begin with methods that require fewer resources and less skills to obtain an overview of the problem. Seizures – which are a by-product of law enforcement efforts – pro- vide a first step, and countries can analyse the information obtained (origin of the products, brands, location, etc.) and report the results to increase public awareness of the problem.16 Alternatively, countries can add questions related to illicit trade to existing and funded health surveillance surveys conducted regularly by health surveillance authorities and statistical authorities. In that way, cooperation in using existing measurement methods between health authorities – the tobacco control office and health surveillance unit – tax and customs authorities and the national 16 Seizures are useful for obtaining qualitative information about the illegal activity, but they have to be treated very cautiously in projecting the size of the problem. Countries may think they have a very large problem because they have competent authorities doing an extraordinary job at finding illicit goods. On the other hand, countries can have less-efficient authorities making few seizures, and in these environments, seizures tell nothing about the size and nature of the problem. 192 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N statistical office could be a starting point for identifying the nature and size of the illicit trade problem in the country. Direct observation of packs has been increasingly implemented in many LMICs, through different types of surveys such as intercepts of smokers or retailers, pack return, littered-pack inspections and covert purchases of cigarettes. These activities have expanded the skills of independent researchers and academia and increased knowledge of these methods. Also, increasingly cheaper digital technologies allow interviewers to take pictures and record pack characteristics in direct observation surveys or in larger national self-report consumer surveys. Table 4.5 Key characteristics, advantages and disadvantages of illicit trade measurement methods METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Seizures (D) Statistics of tobacco products confiscated by local and national authorities Readily available from law enforcement agencies May not provide a representative picture of the size and/or nature of illicit trade Using existing health surveillance surveys to obtain self-reported consumption (D) Adding or improving questions about brand names, quantities, prices, locale of purchase and other factors Produces good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Gap analysis (R) Compare self-reported consumption data with observed (usually administrative) data about tax-paid sales When quality data are available, is simple and easily reproduced (providing for measurements over time) and explainable Data on tax-paid sales and/or consumption are frequently inaccurate and in many cases do not provide information on the size of the illicit market, but only on changes over time Econometric modelling (R) Estimated according to the difference between tax-paid sales and predicted consumption given by the model Because it is consistent with a long tradition of economic theory, empirical estimates can be evaluated Requires high-quality data on a variety of important variables over a period of time and advanced econometric modelling expertise CHAP T ER 4. PO LI T I C AL ECO N OMY 193 METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Expert interviews (E) Experts include researchers (e.g. in economics, criminal justice and public health), journalists, tax and enforcement specialists, product manufacturers and wholesalers Useful for identifying the nature of and trends in the marketplace (e.g. venues where illicit cigarettes are sold, modes of entry), and the interviews can be useful for defining the method to assess the size of the illicit trade Information obtained may not be generalizable, and expert knowledge may be outdated or limited by the experts’ experience; also, experts often have strong biases Smoker/retailer intercepts and pack observation surveys (D) Examining the packs of smokers and cigarette retailers, convenience or probability-based sample Is direct and objective, and smokers do not suffer from any value judgements The difficulty of identifying areas representative of the tobacco use population and sampling important subpopulations such as elderly and immobile smokers, but household surveys could overcome sampling issues Pack return and swap surveys (D) Also a pack observation study using survey sampling techniques to examine smokers’ pack characteristics May decrease the stigma associated with traditional smoking surveys In LMICs, survey distribution may be unreliable because of the mail delivery system Littered pack surveys (D) Also known as empty-discarded-pack surveys; publicly discarded packs bear characteristics (e.g. tax stamps, public health warnings) that indicate whether they are tax compliant Yields estimates that are less likely to be biased from issues of social desirability, recall error and confidentiality Significant budgets could be needed to employ field researchers to collect, code and analyse the data; surveys do not provide information about the smoker and the price paid Covert purchases (D) Uses covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes Directly identifies sources of illicit cigarettes It is difficult to create a sampling frame of retailers for illicit sources or to know what smokers are actually buying and how much Self-report consumer surveys (D) Surveys can be distributed to individuals or households, using various modes of distribution Good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Notes: D = direct measurement, R = residual method, E = expert opinion. 194 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ultimately, when capacity allows, more solid estimates will need to be made using more than one methodology at a given point in time. Ideally, estimates will be made on a regular basis in order to assess the evolution of illicit trade over time and its possible connection to policy changes. 4.1.5 CRITIQUING STUDIES THAT MEASURE THE SIZE OF THE ILLICIT MARKET Measuring illicit trade is a challenge for researchers, industry and governments, because the trade is, by definition, hidden from plain sight. Buyers and dealers sometimes go to great lengths to ensure that their participation in illicit activity is concealed. Over the years, researchers and government agencies have been increas- ingly interested in estimating the size of illicit markets and identifying effective interventions. As consumers of research, governments should critically examine available studies and evaluate them on their scientific rigor and methodological transparency. Measurement issues are particularly acute with respect to the illicit tobacco trade because it is a politicized topic. High estimates may raise questions about the tobacco industry’s ability to control the supply chain, its involvement in illicit diversion, the impact of taxation policies and the effectiveness of enforcement strategies. While the industry has portrayed itself as taking an active stance in measuring and fighting illicit trade (e.g. Project Star, conducted by KPMG LLC but paid for by PMI, later followed by Project Sun), in the past it has used smuggling as a strategy to enter closed markets – for example, in China and Russia (19, 41). Govern- ments should carefully scrutinize evidence about the illicit tobacco trade produced by industry or quasi-industry sources and are advised to seek alternative evidence. Quasi-industry reports are studies commissioned by the industry but published by private research companies (e.g. Ernst and Young, Oxford Economics) (11). Characteristics of good analyses One of the main characteristics of a good analysis is scientific rigor, which involves the use of relevant theoretical frameworks, sound statistical methods and examination of the robustness of findings (e.g. sensitivity analyses). High-quality research reports provide transparent explanations about their methodology and statistical analysis steps undertaken, as well as supplementary analyses that established the robustness of the findings. For example, Joossens et al. (25) clearly describe the data sources used (limitations and advantages and where they can be found) and calculations performed on the number of lives that would be saved if the global market share of illicit cigarettes was eliminated. Explanations should be detailed enough to allow future researchers to scrutinize the analysis and replicate the findings. Replicability is another hallmark of good science. For example, littered-pack studies should detail where and when data collection took place, how many packs CHAP T ER 4. PO LI T I C AL ECO N OMY 195 were collected per geographical unit, the protocol of identifying the illicit packs (e.g. characteristics of the warning labels, brands, tax stamps, etc.) and details of statistical analyses. There should also be explanations of the representativeness of the selected geographical areas. Failure to provide this depth of information may call into question the generalizability of a study and whether there are faults with the chosen method. In the context of policy decisions regarding illicit trade, the most useful data provide information about a representative sample of individuals and geographies. Studies that are limited to, for example, one group of individuals based on specific characteristics or a given geography may yield biased information. Research reports also should be clear about the study’s limitations. For example, studies that measure illicit trade often do not measure product counterfeiting and do not include non- cigarette tobacco products in their estimates (42). When statistical estimates are included, they should provide confidence intervals as well as point estimates to account for uncertainty resulting from simple random chance (11). Characteristics of flawed analyses Flawed analyses can convolute and distort scientific knowledge about illicit trade. Flaws usually manifest in the data, methodology, statistical analysis and/or interpreta- tion of the results (11). Studies may be purposefully designed with methodological flaws to yield high or low estimates of the trade. For example, research showing that illicit trade constitutes a large share of the total market may be used to support arguments that taxes cause sharp increases in illicit trade, whereas lower estimates may be used to support arguments that certain governmental interventions (e.g. increased retail inspections) are effective. Pressures to skew data may also be tied to funding. For example, high estimates can sway governments to provide more resources for law enforcement activities. Analyses can be purposefully skewed by using data sources or data collection methods that will provide biased estimates. Flawed studies sometimes provide incomplete or inaccurate descriptions of their methodology. They may lack detail regarding the quality of the data used or information about how the data were collected and analysed. For example, a common weakness in industry-funded research on discarded packs is that the methods of collection and forensic analysis are not reported, ostensibly because they are “proprietary” information (43). However, these methodological details are key to assessing whether a study’s findings are biased by sampling error, model misspecification, measurement error, non-response or other flaws. It may be impossible to assess measurement error if researchers fail to disclose questions included in a survey instrument. Survey items used to measure the illicit tobacco trade may be imprecise. For example, asking respondents the frequency with 196 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N which they purchase “cheap” cigarettes may yield biased estimates, particularly if consumers can purchase cigarettes at discounted prices by using coupons. To more ac- curately measure tax evasion, surveys must include questions about the location of last purchase, purchase price, presence of public health warning labels and brand names. There are other ways that flawed studies can inadvertently or purposefully distort estimates of illicit trade. For example, data collectors can intentionally oversample areas known to be hot spots of illicit sales or sites that residents from lower-tax jurisdictions visit. Researchers can collect discarded cigarette packs close to the borders of countries with lower taxes to (inadvertently or purposefully) demonstrate the undesired side effects of tax policies. Studies published in non-peer-reviewed or lightly peer-reviewed outlets such as edited book volumes or policy briefs should be viewed with more scepticism than those published in highly regarded peer- reviewed outlets. 4.1.6 CONCLUSIONS Globally, the illicit tobacco trade continues to be a major concern for tax admin- istrators because of the challenges it generates to collecting higher revenues as well as the challenges to accurate and independent measurement. Industry figures provide distorted conclusions regarding the extent of the problem – frequently with a monocausal explanation of the link between illicit trade and tobacco taxation. Illicit trade comprises multisystemic issues and requires multiple strategies. Worldwide, countries at different levels of economic development have implemented a variety of effective measures to combat the illicit trade in tobacco products. The Philippines and the United Kingdom, for example, have addressed illicit trade as part of their overall tobacco tax reform (4). Price (and tax) levels are not a key determinant of illicit trade, the presence of which is exacerbated by the lack of tax administration capacity. Refraining from increasing taxes is not the solution; countries should instead respond with a com- prehensive strategy that includes at least these three main components: 1. It should identify – independently from the industry – the nature and dimen- sions of the problem. It is necessary to assess scientifically and with the best statistical practices the size of the illicit trade to understand the characteristics and scope of the problem. 2. It should identify and implement appropriate policies and strategies targeted at addressing the specific type of illicit trade the country is experiencing. It should address directly the country-specific institutional and/or governance challenges – as well as the lack of multilateral coordination that can exacer- bate illicit trade – and improve tax and customs administration practices as described in Chapter 3. CHAP T ER 4. PO LI T I C AL ECO N OMY 197 3. It should implement best practices contained in the WHO FCTC Protocol to Eliminate the Illicit Trade in Tobacco Products and accede to the Protocol if the country is not yet a Party. There are proper methods and policies with which to address the illicit tobacco trade. If countries start implementing the appropriate policies, they can raise tobacco taxes and reap health and revenue benefits even in the presence of illicit trade. 198 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.2 SCARE TACTIC C: COURT AND LEGAL CHALLENGES 4.2.1 INTRODUCTION The tobacco industry views well-designed and significant tax increases as a threat to the profit, growth and long-term sustainability of its business. As noted by PMI in 1985: Of all the concerns there is one – taxation – that alarms us the most. While [other restrictions] … do depress volume, in our experience taxation depresses it much more severely (44). The industry is, however, less likely to launch direct legal challenges to excise taxes than to other tobacco control measures (see Box 4.2.1 for details), because taxation – and excise tax in particular – is a comparatively well-established regulatory measure; in many jurisdictions, taxes have been levied on tobacco products for more than a century. There is also less unanimity in opposition to taxes among tobacco industry actors, because differences in the market position of different tobacco companies affect their interests in tax policy. This, in turn, decreases the likelihood that they will act collectively on the issue (45). BAT’s stated strategy in the early 1990s was to influence governments with regard to the level and structure of tobacco taxation in order to promote market growth and to secure competitive advantage (46). Nevertheless, tobacco industry actors will still legally challenge, or at least legally threaten, significant tax measures when vulnerabilities in their design, adoption or implementation are apparent. Box 4.2.1 Court and legal challenges to tobacco tax measures Evidence suggests that the tobacco industry and its allies instigate fewer legal actions against tax measures than against other tobacco control measures: 1. The Campaign for Tobacco-Free Kids’ tobacco control laws database contains only a handful of cases concerning tobacco tax measures, but hundreds on other tobacco control topics. This pattern can also be seen in a 2018 review of tobacco control legal challenges that examined this and two other databases to select 96 cases relevant to the question of the WHO FCTC’s usefulness in litigation (47). Only 6 of these 96 cases were challenges related to tax measures. CHAP T ER 4. PO LI T I C AL ECO N OMY 199 2. A 2013 systematic review of empirical studies on tobacco industry interference with tobacco tax policy found that only 9 of 36 relevant articles reported the specific use of litigation as a tobacco industry tactic (1). All 9 concerned constitutional challenges to earmarking provisions for tobacco tax initiatives in the United States (1). 3. A 2015 study on industry interference in LMICs cited legal challenges to tobacco control measures in 15 countries as examples of industry interference, but none of the challenges concerned a tobacco tax measure (48). 4. A 2016 analysis of papers published in systematic reviews of industry inter- ference with tax and marketing measures found that only 5 of 65 papers concerning tobacco tax related to the use of litigation or threats of litigation to interfere with tobacco tax measures (49). The tobacco industry makes extensive use of legal experts (1, 50–52) who study all relevant laws and regulations closely to determine their likely and arguable boundar- ies for the purpose of manipulating regulations and regulators (1, 50–52). Based on this expert advice, tobacco companies know when regulations remain within the bounds of both international and domestic obligations but can still argue that legally permissible tobacco control measures would be defeated in litigation if passed (48, 51–52). As the threat of a legal challenge alone can be used to the industry’s advantage, recourse to litigation is seldom needed or desirable (1, 45, 48, 51, 53–56). Even when litigation is launched, the objective may be to delay or weaken a measure rather than to win on the merits of the case (1, 45, 48, 53). To counter actual and threatened legal challenges, policy-makers need to be aware of relevant legal obliga- tions when preparing and implementing tobacco control measures. Fortunately, the tobacco industry playbook is relatively predictable. Tax and other tobacco control measures can thus be designed to strengthen the regulators’ legal position against genuine threats and enable them to dismiss baseless industry threats. 4.2.2 COUNTRY EXPERIENCES WITH LEGAL CHALLENGES TO TOBACCO TAXATION Legal obligations that are relevant to tobacco taxation include those under do- mestic law and international instruments such as international trade agreements and international investment agreements (IIAs).17 Some of the legal issues that a tax measure may encounter are outlined in Table 4.6. Case studies from various countries illustrate how these legal issues have and have not been avoided in the 17 Relevant international trade agreements include the WTO Agreement and custom unions such as the EU, the East African Customs Union and Mercosur. Relevant IIAs include bilateral investment treaties and the investment chapters in free trade agreements and within custom unions. 200 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N passage, design and implementation of tobacco taxes. These issues are not the norm, however, and should not give rise to undue apprehension. The case studies are rated as positive, mixed or negative based on the extent to which the legal decision upheld the taxation measure in question. Table 4.6 Potential legal issues for tobacco tax measures VULNERABILITIES LEGAL OBLIGATIONS CASE STUDIES Inadequate consultation and other procedural vulnerabilities Domestic procedural law 1, 2 Due process protections for investors under IIAs None Procedural requirements under WTO Agreements and Custom Unions 3 Discrimination against imports or investors Nondiscrimination obligations under WTO Agreements and Customs Unions 8, 9, 10 Nondiscrimination obligations under IIAs 11 Investment incentives or inducements Arbitration mechanisms under investor-state contracts 12 Fair and equitable treatment clauses of IIAs None Other substantive breaches Constitutional rights and restrictions on taxation 4 Statutory restrictions on the imposition of taxation 6 Expropriation clauses of IIAs 5 Ultra vires (the scope of legal authority) 7 Avoiding procedural vulnerabilities in tax laws Procedural defects can be avoided by taking great care in progressing and imple- menting regulatory or legislative provisions. Procedural concerns pose a dilemma for tobacco control regulators. Article 5.3 of the WHO FCTC and the COP guide- lines for its implementation state that policy-makers and regulators should interact with the tobacco industry only when and to the extent strictly necessary (57). For taxation measures, interaction might be necessary because consultative and de- liberative processes could be prescribed under domestic constitutional provisions and procedures for good governance, due process requirements of IIAs and some international trade agreements. The tobacco industry may use these requirements as leverage to delay, distort or hijack the rule-making process in contravention of Article 5.3. Accordingly, interactions with the tobacco industry should be limited to strictly necessary consultation conducted in a transparent or public manner but with care that this does not come at the expense of a measure’s defensibility. The proper balance will depend on the jurisdiction in question, since constitutional, statutory and applicable international legal obligations vary. CHAP T ER 4. PO LI T I C AL ECO N OMY 201 CASE STUDY 1 (MIXED): Industry manipulation of legislative procedures In 2012, a bill stipulating, among other things, the creation of a new specific excise tax on cigarettes passed its final reading in Costa Rica’s Legislative Assembly. Passage of the bill had, however, proceeded under “urgency” and notwithstanding a pending constitutional enquiry (a constitutional query is meant to prevent passage of a bill).18 ISSUE MAJORITY DECISION MINORITY DECISION LESSON Whether the court could consider the enquiry despite passage of the bill and the effect the bill’s passage could have despite the enquiry. The enquiry was taken up by the Supreme Court’s Constitutional Division’s majority (58). The signing and publication of the bill by the executive was suspended by the Constitutional Division pending their decision on the merits of the case – which, in the end, found any question of the bill’s constitutionality baseless (58). The enquiry was inadmissible by reason of having been filed too late and notice of its filing having not been received by the legislature prior to the reading of the bill (58). In disagreement with the majority, the minority held that the court could not consider the enquiry or suspend the bill’s signing by the executive – the final step in becoming law. This challenge demonstrates how the tobacco industry’s defenders may attempt to frustrate and impede a tax measure’s passage. In this case, the challenge seemed to have been a delaying tactic, as it was posted on the same day as the final reading of the bill. Its authors may have either wanted its pending nature to cause the legislature to delay or, as occurred, to create conditions for a procedural and constitutional challenge in the absence of delay. All the grounds of the challenge itself were found to be without merit. Although such frivolous challenges cannot be prevented, they can and should be anticipated to ensure that they do not lead to a tax measure’s defeat. 18 “Urgency” is a procedure under which a bill is progressed through a legislature in an expedited fashion. 202 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N CASE STUDY 2 (POSITIVE): Adhering to domestic procedural requirements Kenya’s tobacco control regulations required the tobacco industry to pay a levy to compensate the state for health care and other negative externalities of smoking. In a 2016 challenge brought against these regulations, the plaintiff, BAT, was unsuccessful on every count (59–60). Even though the levy was not considered a tax measure by the court, the case study is instructive on how regulators may safeguard tax measures against procedural challenges. ISSUES LAWS AND ARGUMENTS DECISIONS LESSONS Whether the government’s consultations on the measure were adequate. Asserting that the Constitution and the Statutory Instruments Act together meant that “appropriate consultations with persons who are likely to be affected” were required because of the measure’s likely substantial effect on business. BAT claimed that this standard was not met. Kenya’s government claimed that it was under no obligation to undertake special or extensive consultation with the tobacco industry. The judge found in favour of Kenya’s government, noting that (1) the requirement to consult does not imply that any particular view needs to prevail; (2) dissatisfaction with the level of consultation is not decisive; (3) on the facts, industry was allowed, and often invited, to send representatives to all relevant public consultative meetings and parliamentary committee hearings; and (4) consultation on the regulations was adequate (59). The tobacco industry carefully scrutinizes legislative and regulatory processes for defects. In this case, Kenyan government officials appropriately distanced themselves from the tobacco industry by not permitting its representatives special consideration but did permit their attendance at public meetings and the ability to submit their views under usual procedures. In this way, both the principles behind WHO FCTC Article 5.3 and the requirement for consultation under Kenyan law were observed. CHAP T ER 4. PO LI T I C AL ECO N OMY 203 Avoiding procedural issues in tax administration CASE STUDY 3 (NEGATIVE): Contravening procedural requirements in international obligations In 2010, a WTO panel held that Thailand violated the Customs Valuation Agreement (CVA) by the process it used to value cigarettes that Phillip Morris (PM) Thailand imported into the country from a related party, PM Philippines. Customs values are important as they are the tax base for tariffs and can feed into the base for other taxes levied against the value of the good, such as ad valorem excise taxes and VAT. Transaction values declared by PM Thailand were rejected by Thai tax authorities as influenced by the relationship between the parties and a customs value determined by deduction was substituted (61). ISSUE LEGAL OBLIGATION DECISION LESSONS Whether Thailand adequately consulted with PM Philippines before rejecting its declared transaction value (61). The CVA requires good faith exchange of reasons and information, with opportunities for response (61). Thailand had failed to properly explain its reasons for rejecting the transaction value, as well as its belief that price was influenced by the relationship between the two parties (61). This was a violation of the CVA. Thailand did not appeal these findings. Thailand’s authorities needed to take greater care in their dealings with the tobacco industry to ensure they met the pertinent procedural obligations. In this instance, a specific and high standard of consultation – the provision of detailed reasons and an opportunity for response – was prescribed by the CVA and Thailand failed to meet it. 204 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ensuring compliance with substantive requirements Rules found in domestic and international law also establish substantive obliga- tions. This subsection focuses on substantive obligations found in domestic law. International obligations concerning discrimination and investment incentives are considered in the next subsections. CASE STUDY 4 (POSITIVE): Tax measure found to be consistent with the Constitution The Chilean government introduced a substantial increase in tobacco and fuel excise, and in 1995, a coalition of taxpayers brought a Constitutional challenge to the measure (62). ISSUE DECISION LESSON Whether the tax was, per article 19 of the Chilean Constitution, “obviously disproportionate or unjust” (62). The excise tax increase did not violate the Constitution, as it was neither confiscatory nor manifestly irrational. Generally applicable excise taxes are not vulnerable to challenges for being excessive, unfair or disproportionate. CASE STUDY 5 (POSITIVE): Failure to grant tax rebates not an expropriation under an IIA This case study is an example of a claim for breach of an expropriation clause in an investment treaty. Such clauses protect foreign investors against measures that can be construed as directly or indirectly seizing an investment or depriving it of its value (63). In the case, an investor was, for more than a decade, denied tax rebates by the Mexican government. This affected the profitability of the business of purchasing and reselling Mexican cigarettes abroad, and the investor brought the claim to an investment agreement arbitral tribunal in 2002 (64). ISSUE DECISION LESSONS Whether Mexico’s failure to grant rebates to the investor exceeded the bounds of valid regulation to constitute indirect expropriation of the investor’s investment (65). There was no expropriation. The arbitral tribunal noted that not all business problems are violations: the investor had no right to participate in the “grey market” export of cigarettes and there were sound reasons to restrict that market (65). Further, the investor was able to participate in other business ventures and actually continued to have business success (65). Claims of indirect expropriation made under IIAs are unlikely to be successful, as generally applicable tax measures are a legitimate form of regulation. A mere loss of profit will not suffice. Claims of expropriation will not succeed unless a substantial or significant deprivation of the investment results. CHAP T ER 4. PO LI T I C AL ECO N OMY 205 CASE STUDY 6 (NEGATIVE): A regulation contrary to superior domestic legislation In 2011, an Indonesian tobacco industry association group, FORMASI, challenged a new excise regulation. Since 2009, the government had been implementing a tiered specific excise tax system based on a set of characteristics (size of production, type of cigarettes and price levels). In 2011, excise rates were increased in nearly all of the 19 tiers, but the reference prices were not accordingly adjusted. This gave rise to a legal issue. ISSUE DECISION LESSONS Whether new excise regulations breached a 57% ceiling for the rate of excise on the retail sale price of tobacco products under the superior Excise Law (66–70). The challenge specified that excise exceeded this ceiling for hand-rolled domestic clove cigarettes (kreteks) (68, 71). The Court found in favour of the tobacco industry association, and the government was required to immediately revoke the 2011 regulation. (69–70). It is advisable to stay within the rules and be aware of legal hierarchies – including superior domestic legislation. The tobacco industry scrutinizes all increases in tobacco taxes. In this case, a breach of a legislative requirement for a single category of tobacco product resulted in Indonesia suffering lost revenue and a setback in its efforts to reduce tobacco consumption. Ensuring a tax measure is within an authority’s legal power A tax measure is ultra vires when it goes beyond the legal power of the enacting body. As with case study 6, this is a legal issue that involves legal hierarchies. In ultra vires cases, however, instead of centring on conflict between inferior and superior law, the issue is whether an authority that enacts a tax measure is authorized to do so. This issue may arise when a tax measure is enacted by a subnational jurisdiction or by an executive acting under a statutory delegation. CASE STUDY 7 (NEGATIVE): Tobacco taxation contrary to the Australian Constitution ISSUE DECISION LESSONS Whether New South Wales’ licensing and penalty fees regime constituted an excise tax by other means contrary to the Australian Constitution’s exclusive grant of that power to the federal government (72). The court found that state licensing fees were excise taxes and that this was contrary to the Australian Constitution (72). Authorities enacting tobacco tax measures must act within the scope of their legal power. 206 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Avoiding discrimination against imports and foreign investments Although inherently discriminatory, customs duties may be used subject to the agreed upper limits in a country’s trade agreements. Excise and other taxes designed with the aim of raising tobacco prices to reduce demand and advance human health should be origin-neutral: they should not seek to tax local products less than foreign products or aim to treat foreign products differently from one another. Tobacco tax measures are pursuing objectives other than health when they aim to raise the price of imports more than that of local products or seek to burden favoured market participants less than others. Solely health-protective tobacco taxes will not ordinarily violate Articles III:2 and I:1 of the GATT (the WTO’s General Agreement on Tariffs and Trade), which prohibit discriminatory taxation (in light of general exceptions). Nor will solely health-protective tobacco taxes directly violate anti-discrimination protections for investors found in the national-treatment (NT), most-favoured-nation (MFN), expropriation and fair-and-equitable-treatment (FET) clauses of IIAs (63, 73–74). It is possible to make claims for breach of international obligations on grounds other than discrimination, but such claims are generally highly unlikely to succeed. CASE STUDY 8 (NEGATIVE): BAT v Uganda (2017 East African Court of Justice) DISCRIMINATION LESSON Uganda established a higher level of excise taxes on imported cigarettes – including those from Partner states of the East African Customs Union (75) – than on local cigarettes. Its implementation was discrimination contrary to Article 15 of the Customs Union Protocol (75). Differential taxation explicitly based on origin can be construed as protectionist discrimination in violation of international obligations. The tobacco industry can also turn to international trade agreements outside of the WTO – in particular, customs union mechanisms. Difficulties arise when ostensibly origin-neutral and health-protective tobacco taxes result in dissimilar taxation of tobacco products (73). Discrimination does not exist simply because there is dissimilar taxation – the taxation must adversely impact imported goods more than local products, the imports of one nation more than another or a particular investor’s products more than comparable products. Where dissimilar taxation between product categories results in discrimination, the tax will ordinarily still be lawful if the dissimilar taxation is based solely on a legitimate regulatory distinction between the product categories in question.19 19 The precise applicable rules vary depending on the nature of the legal obligations in question. Under the GATT, dissimilar taxation of like or directly competitive products can be justified based on scientifically grounded distinctions between products under Article III:2 and, in the alternative, discrimination that is necessary under the explicit carve-out for health-protective measures, Article XX(b) (73). For the MFN and NT clauses of IIAs, differential taxation can be argued as nondiscriminatory on the basis that difference in harm means the products are not “alike” or, in the alternative, discrimination is justified based on scientific evidence of differences in harm and rational reasons for the health-protective role of differential taxation (63). CHAP T ER 4. PO LI T I C AL ECO N OMY 207 Where discrimination is inadvertent, lack of an intention to discriminate is not sufficient as a defence for breach of obligations under IIAs or the GATT (73, 76). Policy-makers should carefully scrutinize measures to determine: 1. whether an aspect of a tax measure’s design or implementation may be more to the detriment of imports or foreign investors than of local products or domestic investors; 2. whether the potentially discriminatory aspect of the tax measure serves any useful purpose in supporting the tax measure (i.e. it is needed to achieve the health goal); 3. whether there is any reasonable alternative that could achieve the same effect without the potential for discrimination; (i.e. it is indispensable) and 4. when it is needed and indispensable there is a good chance that it will be defensible. The case studies below provide examples of discrimination arising in connection with a tobacco tax measure. CASE STUDY 9 (NEGATIVE): Thailand – Customs and Fiscal Measures on Cigarettes from the Philippines (2010 WTO panel) The facts of this case are presented in case study 3. This case study examines claims of discrimination rather than the procedural issues. DISCRIMINATION EXPLANATION LESSONS Thailand implemented its policy for determining the tax base for VAT on cigarettes inconsistently (61). Thailand applied a methodology in fixing the tax base, in particular a marketing cost component, of imported cigarettes that differed from that for local products (61). This resulted in the marketing cost component for the imported cigarettes being higher than it would have been under the general methodology. This difference in treatment was insufficiently justified and therefore considered discriminatory. As there is potential for inadvertent discrimination when the base for an ad valorem tax is fixed, tax base determinations must be consistent and well- reasoned (61). This case study demonstrates how policy- makers need to take care in designing and implementing ad valorem taxes to ensure they are nondiscriminatory and legally defensible. 208 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Thailand’s VAT rebate policy imposed a potentially higher tax burden and also created more burdensome administrative requirements for imported cigarettes (61). Resellers of cigarettes produced by a government entity were granted an exemption from VAT (61). Although resellers of imported cigarettes would be eligible for a tax credit on their VAT, this was not an automatic process (61). The distinct treatment of resellers of imported cigarettes and those of local cigarettes resulted in the risk that there would be a higher VAT burden for the former (61). The distinct treatment also imposed an additional administrative burden on resellers of imported cigarettes and altered conditions of competition (61). Rules for the collection and enforcement of tax obligations should be the same, or as similar as practicable, in both form and effect for domestic and imported tobacco products. CASE STUDY 10 (NEGATIVE): Dominican Republic – Measures Affecting the Importation and Internal Sale of Cigarettes (2004 WTO panel; 2005 WTO Appellate Body) Under article XX(d) of the GATT, discrimination that is necessary to secure com- pliance with a legitimate tax measure will be justified provided there is no less- discriminatory alternative. In this case, this justification was used unsuccessfully. DISCRIMINATION EXPLANATION LESSONS The Dominican Republic’s tax stamp regulations were discriminatory towards imported goods (77). Under the regulations, all cigarette packs had to be affixed with tax stamps, but imported cigarettes were to be affixed with tax stamps under the supervision of local tax authorities following importation, while locally manufactured cigarettes could be affixed with a tax stamp in the course of production. This de facto distinction between local and imported products modified the conditions of competition to the detriment of imported cigarettes by (1) increasing costs for importers and (2) impairing the aesthetics of imported products (77). The panel did not consider this discrimination justified: it was not necessary for the enforcement of tax measures, because less restrictive alternatives were available such as permitting importers to affix tax stamps during the course of production (Dominican Republic – measures affecting) (77). The panel’s findings were upheld on appeal (78). Policies crafted to ensure compliance with tax measures need to also be nondiscriminatory. Discrimination claims can arise when compliance costs are higher for imports than for local products and this de facto distinction is avoidable. It is important to consider whether less burdensome alternatives may achieve the same objective. CHAP T ER 4. PO LI T I C AL ECO N OMY 209 CASE STUDY 11 (NEGATIVE): Feldman Karpa v Mexico (2002 ICSID [International Centre for Settlement of Investment Disputes] Arbitral Tribunal) Arbitral tribunals have accepted differences in treatment accorded to investors protected by IIAs when there is a legitimate connection between the distinctions drawn and public welfare objectives (76). The facts of this case are presented in case study 5. This case study examines aspects of the case involving the investor’s claim of discrimination, rather than the substantive issue of expropriation. Claims of discrimination are made on different grounds than claims for expropriation, which is why the case was decided differently on this claim. DISCRIMINATION LESSONS Denial of foreign investors’ claims for tax rebates. Tax rebate claims were granted to similar local investors (65), which was a violation of an IIA’s national treatment clause (65). Foreign and local investors must be treated similarly, and consistent and well-documented policies must be used to guide administrative decisions. The denial of the rebates may have been justified, but the government was unable to establish this due to a lack of documentation. Avoiding the investment incentives trap Investor-state contracts between the tobacco industry and governments should be avoided. They are not merely “contractual” in the domestic law sense, as even in the absence of an applicable IIA, they can be internationalized to provide inves- tors the right to (1) remove dispute settlement from the state’s court in favour of independent arbitration and (2) remove the dispute from the state’s legal framework in favour of general principles of law (63, 76). Commitments under these clauses cannot, therefore, be legislatively moderated or extinguished, nor can liability be limited within domestic courts that may be more likely to favour the state’s right to regulate in favour of public health (76). Investor-state contracts and other noncontractual inducements can be further internationalized by umbrella clauses within IIAs. Such clauses make reneging on undertakings assumed towards investors a breach of the IIA (76). Moreover, even in the absence of an umbrella clause, contracts and inducement can underpin a claim for legitimate expectation and breach of fair and equitable treatment and can also strengthen an investor’s claim for indirect expropriation (63). Arbitral awards make clear that although taxes can be expected to vary and tobacco will be regulated, investors can have the legitimate expectation that states will abide by formal inducements and written contractual undertakings. A common clause within investor-state contracts, the stabilization clause, is ruinous to evidence-based tobacco control’s most effective measure: excise tax in- creases. Stabilization clauses purport to freeze specific domestic law from the time 210 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of investment (63). Seemingly less onerous, economic equilibrium clauses require contracting states to compensate for regulatory changes that negatively affect an investment’s value (63). There is little difference in effect between these two types of clauses: liability for the cost of breaching an equilibrium economic clause can be onerous enough to make it fiscally challenging and politically unpalatable. CASE STUDY 12: An investor-state contract A state entered into an investment agreement with a TTC in 2001 on the privatiza- tion of its state-owned tobacco enterprise and creation of a joint venture. This investment was to provide economic benefits under the agreement: the joint venture would increase exports and profit using the TTC’s cash and expertise while also ensuring prioritization of local employment, manufacturing and resources. The final investor-state contract included a form of economic equilibrium clause under which any increase in the excise tax rates applied to the company’s tobacco products before a set date would be compensable. While the agreement was not removed from the state’s law, it provided for independent arbitration in case of a dispute over its compensation. In addition, there is a bilateral investment treaty between the host state and another state in which the TTC’s subsidiary has residence that includes a FET clause – this could buttress, if needed, the protection provided by the stand-alone arrangements of the investor-state contract. There were similar less-formal inducements offered to a separate TTC. The extent to which incentives have been granted to the tobacco industry is unknown, but contracts and inducements are likely to be offered in the context of the privatization of state-owned tobacco interests and in dealings between investors and state-owned tobacco enterprises (63). Although countries have been entrapped by their incentives to industry, the investor-state contract provides the clearest example of how undertakings and inducements with the tobacco industry under- mine tobacco control (56, 79–80). States should avoid offering industry incentives and, in particular, entering into contractual undertakings with the industry. More systematically, government should consider avoiding IIAs that elevate incentives and inducements above sensible and reasonable regulation. 4.2.3 CONCLUSIONS Health-protective and origin-neutral tobacco excise taxes are legally defensible, and industry threats are usually baseless. There are, however, certain rules governing procedure, design and consultation that governments may need to consider: 1. Governments should be aware of the standard of consultation required under do- mestic law and any applicable international obligations (case studies 1, 2 and 3). CHAP T ER 4. PO LI T I C AL ECO N OMY 211 It is important to distance the tobacco industry from the policy-making process to the extent that this is permissible. Do not grant the industry special consideration, but do ensure that it is consulted with as required – for example, by providing public meetings, timely information and the ability to submit industry views – while being aware of potential procedural manipulation (case studies 1, 2 and 3). 2. Excise tax is generally safe from challenges that claim it is confiscation or expropriation under domestic or international law (case studies 4 and 5). But express limits on taxation can be found in other laws or a country’s constitution or in the limits of the power to tax granted to an authority (case studies 6 and 7). 3. Explicit and de facto discrimination against foreign tobacco products or investors must be avoided in the design, implementation or enforcement of tax measures (case studies 8, 9, 10 and 11). Legal issues may arise not from the tax measure itself, but rather from ancillary measures that support its implementation (case studies 9 and 10). 4. Explicit differentiation between products based on their effect on health may be challenged as discrimination if it falls heaviest on imported products and has to be justified on the basis of evidence of impact on health and a lack of alternatives. 5. Investment incentives in the form of inducements or contractual undertak- ings should not be offered, as these may be binding (case study 12) or may ground a challenge under an IIA; they are also contrary to the WHO FCTC Article 5.3 Guidelines. 212 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.3 SCARE TACTIC A: ANTI-POOR RHETORIC (REGRESSIVITY) 4.3.1 INTRODUCTION In their efforts to lobby against tax increases, the tobacco industry and its affiliates often claim that increases in tobacco taxation will hurt the poor (81–82). This argu- ment is based on the concept of regressivity in relation to taxation. Conceptually, a tax can be regressive if it means that lower-income people pay a greater proportion of their household income to meet the tax burden than do wealthy people. In other words, the tax burden tends to be relatively higher for lower-income households than for middle- and high-income households. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use. Second, higher tobacco taxes and prices can induce behaviour change among the population, as reflected in the price elasticity of demand (83–84). In combination, these broader considerations effectively make tobacco taxation a progressive – rather than regressive – public health intervention. 4.3.2 REGRESSIVITY AND THE BROADER PERSPECTIVE In a narrow sense, tobacco taxation can be seen as regressive because lower-income people must allocate a relatively greater proportion of their household income than wealthy people to pay for tobacco products when those products become more expensive following a tax increase. In many countries, people from lower-income groups use tobacco more than other people (85). A systematic literature review by WHO found a robust association between lower income and a higher prevalence of current smoking among adults, both men and women (86). This finding was consistent across three decades of studies, across most geographic regions and across countries of different income classifications. For example, in India, high rates of tobacco use – i.e. use by more than 30% of the adult population – are found only in lower-income states such as Assam and Odisha, where net state domestic product is still below 100 000 rupees per capita (see Fig. 4.3.1) (87). CHAP T ER 4. PO LI T I C AL ECO N OMY 213 Fig. 4.3.1 Relationship between adult tobacco use and net state domestic product per capita in states and union territories of India, 2016–2017 Source: (87). However, this finding does not account for broader health and economic factors that determine the full impact on households. Tobacco taxation can in fact be viewed as a progressive – or pro-poor – policy when these wider considerations are properly ac- counted for and explained. In terms of health concerns, the relatively high use of tobac- co among low-income populations translates into a much greater burden of tobacco- attributable diseases for these populations, including higher morbidity and mortality. Low-income groups are also less able to afford medical care to treat tobacco- attributable diseases, and large out-of-pocket medical expenditures can further impoverish many families. Consequently, many poor individuals do not get or even seek the medical care they need. One study found that in Bangladesh, 55% of patients diagnosed with a tobacco-attributable illness did not seek further medical care. This lack of health care utilization was attributed in part to prohibitively high out-of-pocket treatment costs (88). The combination of high rates of tobacco use and lack of access to affordable medical care means that tobacco use measurably contributes to the poverty rate in a number of high-tobacco-burden countries, including China and India (89–90). A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India 214 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The poor are also known to be more price-sensitive than the wealthy; lower-income smokers exhibit higher price elasticities than their higher-income counterparts. This is demonstrated in recent studies by the World Bank, findings of which are shown in Fig. 4.3.2 (91). The poor respond more strongly to higher tobacco taxes and prices by reducing their use of tobacco products more than others, and thus they benefit disproportionately in terms of avoiding tobacco-related deaths, diseases and associated medical costs. A similar conclusion was drawn in a systematic review of the population impact of tobacco control policies on socioeconomic inequities in high-income countries at the late stage of the tobacco epidemic (92). The review found 16 relevant studies relating to taxation, only one of which found a regressive association between tax and the social economic gradient (seven found a progressive impact, while the others produced mixed results). Fig. 4.3.2 Price elasticity of tobacco consumption, medium estimate, by decile Source: (91). This wider economic perspective is explained in the World Bank’s Extended Cost- Benefit Analysis (ECBA) framework, which assesses the distributional impact of tobacco tax increases on health, among other factors (82, 83). That is, the ECBA framework looks beyond the simple or partial definition of regressivity (i.e. impact El as tic it y Deciles -1.2 1 2 3 4 5 6 7 8 9 10 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina CHAP T ER 4. PO LI T I C AL ECO N OMY 215 on household expenditure by income levels) to capture the full distribution of benefits, including improved health and income. The ECBA framework has been applied in studies of various countries, including Bangladesh, Bosnia and Herzegovina, Chile, Indonesia, Republic of Moldova, South Africa, the Russian Federation, Ukraine and Viet Nam. The evidence from these studies supports the view that effective tobacco tax policies can generate pro-poor and welfare-improving outcomes. When reductions in medical expenditures and additional years of working life that result from lower smoking-related mortality are taken into account, the overall policy of tobacco tax increases becomes progressive rather than regressive (see Fig. 4.3.3) (84). A similar conclusion has been reached in studies of high-income countries, such as the United States, where a tobacco tax increase was enacted in 2009 (93). Fig. 4.3.3 Impact of a 100% price increase, with medium elasticities, by deciles Source: (91). Tobacco tax increases will also often lead wealthier smokers to contribute relatively more than poorer smokers to the overall amount of tax revenue collected. This is because poorer smokers reduce their consumption the most, since they are more price- sensitive and wealthier smokers also tend to purchase premium (higher-priced and In co m e ga in s (% ) Deciles -1 2 3 54 6 7 8 9 10 0 1 2 3 4 5 1 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina 216 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N taxed) tobacco products (94). Hence, higher tobacco taxes can be seen as progressive in terms of additional revenue collection and health gains resulting from cessation, as well as from preventing the young from taking up smoking in the first place. One study from China suggests that a 50% tax increase would raise US$ 703 billion over 50 years, with just 14% of this increase being borne by smokers in the lowest income quintile (95). In addition, the tax increase would yield a savings of US$ 24 billion in expenditures on tobacco-related diseases, with about 28% of these savings being enjoyed by smokers in the lowest income quintile. The ECBA framework actually presents a rather conservative profile of the net benefits of raising tobacco taxes, since it does not include other sources of gain, such as reduced harm from exposure to second-hand smoke, increased productivity and the potential for poor households to benefit from social programmes funded through increased tax revenues (96). Assessments of the distributive impact of the 2009 tobacco tax increase in the United States found that the overall progressivity of the increase was enhanced by the tax rev- enue being used to expand health insurance coverage for children of low- and middle- income families (97). Accounting for this expanded coverage added to the progressiv- ity of the overall legislative package, the bottom line being that the impacts are positive for lower-income quintiles and greatest, on average, for low-income households (93). Similarly, a large proportion of the tobacco tax revenues from the Philippines’ so-called Sin Tax Reform was used to subsidize universal health coverage (UHC) for poor and near-poor families. Globally, 37 countries are known to earmark some tobacco tax revenues for health programs, with many of these programs indirectly benefiting the poor and less-advantaged disproportionately more than other groups (27) (for details on earmarking, see section 4.6). 4.3.3 CONCLUSIONS Contrary to the perception of tobacco taxation being regressive, it is a strong pro- poor policy when the broader economic impacts are taken into consideration. The tax burden is not a complete indicator of regressivity, since it does not include the negative health and economic impacts of tobacco-attributable diseases or the positive impacts of behaviour change in response to tax and price increases. The health and economic burdens of tobacco-attributable diseases fall dispropor- tionately on the poor, who tend to have higher tobacco use and are also the least able to afford the necessary medical care. Because the poor tend to be more price-sensitive, they curtail their use and consumption more significantly than wealthier smokers in response to tax increases, which in turn reduces their downstream health and economic costs. Tobacco taxation can be made even more progressive by earmarking or allocating tobacco tax revenues for social goods and services that benefit the poor (see section 4.6). CHAP T ER 4. PO LI T I C AL ECO N OMY 217 4.4 SCARE TACTIC R: REVENUE REDUCTION 4.4.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases result in reduced tax revenues for the government. According to them, the reduction in revenues is caused either by substitution to cheaper, lower-taxed or smuggled tobacco products or by reductions in consumption overall (98–99). The tobacco industry often refers to the Laffer curve to make this argument. According to this curve, revenues increase along with tax rates up to a certain point, after which further increasing tax rates leads to declining revenues. When considering tobacco taxes, the tobacco industry assumes that countries are already approaching or are even beyond the critical tax rate level (98). However, the argument rests on a narrow theoretical and empirically unsubstanti- ated foundation (98–100). The price inelastic demand for tobacco and the relatively low tax share in prices in many countries explain the win-win for public health and finance, i.e. that declines in consumption and increases in revenues can occur simultaneously (98, 101). Furthermore, many country examples (see case studies below) demonstrate that well-designed and well-implemented tobacco tax increases lead to increases in revenue, at least in the short to medium term (98, 100). Although consumption will diminish with a tobacco tax increase, the percentage increase in excise tax per unit is greater than the percentage decrease in tobacco consumption, cancelling out at least some of the effect of reduced consumption on revenue (98–99). A change in the tax rate, with all other factors influencing consumption kept constant, corresponds to a change in the tax revenue and is represented by a move- ment along the Laffer curve. As the tax rate changes, so does the elasticity of the tax base; each point on the Laffer curve corresponds to a different tax base elasticity. When one or more of the other factors changes, this affects the position of the curve, and the tax base elasticity changes at a given tax rate. For example, a successful smoke-free policy or advertising ban that reduces the demand for tobacco shifts the curve down, reducing the tax revenue potential for each tax rate. To demonstrate that few, if any, countries are beyond the revenue-maximizing point on the Laffer curve, Table 4.7 shows the revenue impact of increasing excise taxes under different scenarios, using different price elasticities of demand, different levels of tax increases and different starting tax shares, depending on country income levels. This is the tax base elasticity approach from which the Laffer curve is derived (for more details, see section 2.2.3 and Annex 2.2). The total and excise tax shares shown are weighted averages for each country income group, calculated from the RGTE dataset. The revenue gains were simulated using progressive levels of excise tax increases (25%, 50%, 75% and 100%) and varying price elasticities of demand (-0.4 to -1.2). 218 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.7 Percent increase in excise revenues under different scenarios of tax levels, tax increases and price elasticities20 Income group Total tax as % of retail price Excise tax as % of retail price Increase in excise tax Increase in excise revenue when price elasticity of demand is: -0.4 -0.6 -0.8 -1.0 -1.2 LOW INCOME 38% 22% 25% 22% 20% 19% 17% 16% 50% 43% 39% 36% 33% 29% 75% 63% 57% 52% 46% 41% 100% 82% 74% 66% 59% 51% MIDDLE INCOME 58% 41% 25% 19% 17% 14% 11% 9% 50% 37% 31% 26% 20% 15% 75% 54% 45% 36% 27% 19% 100% 71% 57% 45% 34% 23% HIGH INCOME 68% 55% 25% 18% 15% 11% 8% 5% 50% 35% 27% 21% 14% 8% 75% 50% 39% 29% 19% 10% 100% 65% 50% 36% 23% 11% Source: Authors’ calculations using data from the RGTE (27).21 Substantial revenue increases occurred in all the scenarios that were considered in the simulation. These results reaffirm much of what is already known, i.e. that higher tax increases generate higher revenue gains, and that these gains increase with the increasing inelasticity of demand. Even when demand is relatively price elastic (-1.2), the simulation predicts a gain in revenue. The tax share in price also affects revenue potential. The lower the tax share in price, the larger the revenue potential. This suggests that revenue reductions as a result of an excise tax increase will occur only if the scenario is extreme (i.e. a very elastic demand coupled with a very high current tax share). It is important to note that the vast empirical literature 20 These projections use 2018 data from 185 countries. The countries were classified according to World Bank income group, with the average total tax share, excise tax share and VAT/sales tax share for each country weighted according to the number of current adult cigarette smokers. To calculate the projected revenue for each stated elasticity, it was assumed that there would be full pass-through of the excise tax increase, along with constant percentages of non-excise taxes (VAT/sales tax) as a share of the retail price. The consequent changes in price were multiplied against the respective elasticities to derive the expected change in consumption. The projected revenues could be easily computed by multiplying the new consumption figures against the increased excise tax rates. 21 These calculations do not take into account brand substitution (cross-price elasticities), income ef- fects or illicit trade. The excise tax was assumed to be a specific tax, while the non-excise taxes (VAT and others) were bundled and treated as an ad valorem tax with retail price as the tax base. The difference between retail price minus all taxes was also assumed to be constant, with full pass-through of the tax increase to consumers. CHAP T ER 4. PO LI T I C AL ECO N OMY 219 shows tobacco to be universally inelastic; thus the extreme scenario should not be given credence by policy-makers. Furthermore, as demonstrated by the data in Table 4.7, tax shares in most countries are relatively low and reinforce the revenue potential of tobacco tax increases. The revenue potential of tobacco taxes is indeed quite significant. It is estimated that in 2018, excise taxes on cigarettes generated a total of US$ 361 billion in revenues worldwide, including US$ 162 billion in LMICs. If all countries were to raise excise rates by the equivalent of US$ 1 per pack of cigarettes, the amount of excise revenue would increase by between US$ 178 billion and US$ 219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from these tax increases, with excise revenues in these countries increasing by US$ 133 billion to US$ 167 billion, or by 82–103%.22 Revenue reduction in the countries examined was due to other causes, not the tax increase per se. For example, Tonga significantly increased its excise tax on cigarettes in 2016 and saw a very sharp decrease in its consumption (40% decrease), followed by a revenue decrease. This occurred because 20% of smokers switched to an untaxed, cheap local loose tobacco product called Tapaka Tonga (102). The lesson learned was that Tonga needed to tax all its tobacco products at the same level to avoid substitution to lower-price/untaxed tobacco products. Another example of revenue decrease that was not related to tax increases but rather to tax administration mismanagement is the case of South Africa (see explanation in the case study later in this section). Finally, declines in revenue due to long-term declining trends in tobacco use should not be confused with being beyond the revenue-maximizing point of the Laffer curve. For example, in the United Kingdom, where long-term declines in tobacco use are being experienced, a nominal decline in revenues occurred between 2017 and 2018 even though excise taxes remained unchanged. Conversely, even countries with very high tobacco excise rates experience increases in revenues as a result of tobacco tax increases (see the case study of Australia below) (98, 100). This suggests that few countries, if any, are beyond the revenue-maximizing point on the Laffer curve. Tobacco consumption is expected to be tax inelastic, even if demand becomes effectively price elastic as a result of successful tobacco control interventions. Taxation serves as an instrument for both fiscal and public health objectives. If after successful tobacco control interventions, prices reach levels where demand is elastic, the tax base is still most likely to be inelastic due to tax undershifting, since overshifting is not a good pricing policy when demand is elastic (for a more detailed discussion on the shifting of tax, see section 2.2.2). In other words, a tax rate increase in combination 22 Goodchild M, Perucic AM, Paul J. Tobacco taxation as a strategy to achieve global targets for smoking prevalence. Unpublished manuscript. October 2020. 220 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N with non-price tobacco control measures, which make consumers more sensitive to price (tax) increases, leads to declining but still positive marginal revenues. In the long run, tobacco control policies, including price and tax measures, may be so successful in reducing consumption that revenues will plateau or fall. This is ultimately the long-term policy goal. Reducing the impact of the policies and ending the global tobacco epidemic is the aim of tobacco control and not something to be avoided. However, ending the global tobacco epidemic is unfortunately not foreseen in the short to medium term; therefore, governments can currently rely upon tobacco taxes as a reliable source of revenue (103). 4.4.2 THE REVENUE IMPACT OF EXCISE TAX INCREASES: CASE STUDIES The following case studies illustrate four key points: (1) large and regular tax increases result in large and consistent revenue increases; (2) countries with high taxes and falling prevalence of tobacco use can still increase revenue with tax increases; (3) countries that reduce taxes experience revenue declines; and (4) countries that increase taxes in the face of illicit trade still increase revenue. Large and regular tax increases usually mean large and consistent revenue increases South Africa’s experience shows how successive tax increases, well above inflation and year after year, generate additional revenues even after taxes have been increased substantially. After two decades of declining real revenue in the 1970s and 1980s as real excise per pack declined, South Africa implemented successive excise tax increases from 1994 until 2011 (Fig. 4.4.1) (98, 104). After adjusting for inflation, this resulted in a real excise tax revenue increase of 245% (98). Revenues began to plateau from 2012 as tax increases stalled. They began to decline after 2015 – not due to tax increases, however, but due to a dramatic decline in administrative capacity and enforcement measures exacerbated by large-scale corruption in the government, including the tax administration authority (105). The rapid and catastrophic decline in tax administration and enforcement has been the subject of much attention (106). CHAP T ER 4. PO LI T I C AL ECO N OMY 221 Fig. 4.4.1 Real excise tax per pack of cigarettes and real excise tax revenue in South Africa, 1961–2020 Source: Data shared by University of Cape Town, 2020. Similarly, the Philippines provides a compelling example of how large and regular tax increases alongside reforms to tax structure can lead to large and consistent revenue increases – in this case, also through an accompanying reform to the tax structure (Fig. 4.4.2) (98). The 2012 Sin Tax Law consolidated the country’s four tax tiers into two by 2013 and established a uniform structure by 2017. The same law provided for large, progressive increases across the board, but in particular for the lowest tax categories (98). Not only were the revenue gains substantial, they exceeded all the projections for 2013–2017 made prior to the law’s passage (98). Excise tax per pack Excise tax revenue Ra nd s pe r p ac k (c on st an t 2 02 0 ra nd s) Excise revenue (constant 2020 rands) 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11 20 13 20 15 20 17 20 19 0 0 5 6 10 10 15 14 20 18 222 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.2 Real excise tax per pack of cigarettes (multitier, two-tier, unitary) and total tobacco real excise tax revenue in the Philippines, 2012–2018 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2012 as the base year. Sources: (107, 108 and data shared by the Philippines Department of Finance, September 2020). Ukraine is another example of a country that has regularly increased taxes over the past 10 years and has experienced increased revenues along with decreases in consumption and the number of smokers. Figure 4.4.3 the shows the trends in excise tax, revenues, cigarette sales and number of smokers in 2008–2017. Increases in excise rates were consistently accompanied by increases in revenues. In 2014–2015, excise tax was not increased above inflation (and inflation, especially in 2015, was very high, at 48.7%), so real values of excise and revenues went down. But it is evident from the data that revenues closely follow the path of excise levels even when sales go down. 12 Re al e xc is e ta x pe r p ac k, P hi lip pi ne p es os (2 01 2 ba se ) Billions (Philippine pesos) in tobacco real excise tax revenue (2012 base) 00 20 60 100 140 2012 HIGH PREMIUM SIN TAX LAW RA 10354 TRAIN LAW RA 10963 UNITARY RATE 2013 2014 2015 2016 2017 2018 5 10 15 20 25 30 32 .1 9 68 .6 6 71 .0 9 93 .5 2 87 .2 0 96 .0 1 11 4. 39 24.4 25.4 26.2 26.8 26.9 27.5 11.7 16 23.1 19.6 LOW Real tobacco excise revenues Dierent levels of real tobacco excise tax per pack MEDIUM 7.6 2.7 28.3 CHAP T ER 4. PO LI T I C AL ECO N OMY 223 Fig. 4.4.3 Average real cigarette excise tax rates, real cigarettes excise tax revenues (base year 2008) and cigarette sales and number of cigarette smokers in Ukraine, 2008–2017 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2008 as the base year. Source: Data provided by Konstantin Krasovsky, July 2020. Countries with high tax and falling prevalence of tobacco use can still increase revenue with tax increases Countries with already high tobacco taxes and rapidly diminishing tobacco use can still increase revenue by increasing taxes (98, 109). Australia has implemented comprehensive tobacco control policies and enacted consistent tobacco tax increases on top of what were already some of the highest tax rates in the world (see Fig. 4.4.4). Between 2001 and 2010, revenue increased with increasing tax rates, but in real terms (inflation-adjusted) it remained static (109). Then, in 2010, a 25% excise tax increase was introduced, with large annual increases scheduled from 2013 onward (98, 109). The result of this tax policy has been consistent and large increases in revenue year after year for nearly a decade, even when the increases were being made on already high tax rates.23 23 The apparent reduction in revenues in 2012 and 2013 was due to a change in the source of the data for 2001–2011 and 2012–2016. Data for 2012 and 2013 do not include customs duty, while all other years do. 73 95 125 112 88 82 75 76 67 11.8 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Number of daily smokers, in millions Real tobacco excise revenue, in billions of Ukrainian hryvnia Real average excise per cigarette packs of 20, Ukrainian hryvnia Number of taxed cigarettes (sales), in billions of sticks 0.6 3.6 10.1 9.2 8.7 8.6 8.4 8.1 7.3 6.2 6.5 6.3 7.8 10.3 11.2 12.1 13 1.4 2.2 2.5 2.9 3.5 3.2 2.7 3.3 4 12.7 13.4 74 9.7 224 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.4 Real excise tax and customs duty per stick and real total revenue (all tobacco products) in Australia, 2001–2019 Notes: Rates published by Australian Taxation Office and Australia Department of Immigration and Border Protection, adjusted using Australian Bureau of Statistics Consumer Price Index rates. The 2011–2012 and 2012–2013 figures do not include customs duty, which explains the apparent decline in revenue. Using official disclosures, Scollo and Bayly estimate that duties in these years were $7397.2 and $7687.2 respectively (110). Sources: (109, 110). Countries that reduced taxes and saw revenues decline Prior to 1982, Canada lowered taxes on cigarettes and experienced declining revenues as well as increased smoking – particularly among youth. Subsequent fivefold in- creases in cigarette taxes between 1982 and 1992 resulted in more revenue, increases in retail price and substantial reductions in consumption, with teenage smoking declining by nearly two thirds (5). In the early 1990s, a growing illicit trade in ciga- rettes emerged in which Canadian cigarettes exported to the United States were then smuggled back into Canada (5). The tobacco industry – which was later found to be complicit in and profiting from this illicit trade – sought to frame Canada’s high tax rates as the cause of smuggling (111–112) and succeeded in convincing the federal government, as well as six provincial governments, to make massive reductions in the tobacco tax (111–112). As a result, federal tax revenues fell significantly – more than twice as much as the government had predicted – and smoking rates among both adults and youth began to increase (5, 112). The Canadian government later changed its strategy, and the federal excise tax was restored, resulting in increased Specic excise/ duty per stick (cigarettes/cigars less than 0.8 g) Total revenue (all tobacco products) To ta l c us to m s/ du ty re ve nu e (in a tio n, a dj us te d, 20 19 A us tr al ia n do lla rs , m ill io ns ) Total excise/duty per stick (ination, adjusted, 2019 A ustralian dollars) 0 3 000 6 000 9 000 12 000 $0.8 $0.6 $0.4 $0.2 $0 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 CHAP T ER 4. PO LI T I C AL ECO N OMY 225 revenues and decreased smoking (5). Canada’s focus then shifted to using customs enforcement, rather than tax rates, as the best means of countering illicit trade (5). Countries that increased taxes in the face of illicit trade and still increased revenue As discussed in section 4.1, the tobacco industry exploits illicit trade as a strategy to undermine tobacco tax policy, with the goal of deterring governments from increasing tobacco taxes. The narrative that has been created is that higher tobacco tax rates result in increased illicit trade and undermine the policy goals by resulting in lower (or no) declines in tobacco use or lower (or no) increases or even decreases in revenue. However, as shown in section 4.1, the empirical evidence does not sup- port the industry arguments. Furthermore, the evidence shows that the industry and its allies have consistently overstated and exaggerated the scale and extent of illicit trade (see section 4.1). As was the case in Canada, Brazil’s tobacco tax policy suffered from a fear that the illicit market would expand unless it was undercut by price competition in the legal market, which it was thought could be best encouraged through tax cuts (20). Real excise tax rates declined from 1999 until the mid 2000s, as nominal increases were below the rate of inflation. This resulted in declines in real tax revenues (20). In these years, the tobacco industry used the tax cuts to increase profit margins rather than decrease prices and outcompete the illicit market, while also exaggerating the size and scope of the illicit trade problem (20). This caused the industry’s argument on illicit trade and revenue to lose credibility and resulted in increases in tax rates from 2007 onwards, with a major reform passed in 2011 (20). Tobacco excise rates and minimum prices were scheduled by the law to increase at levels above expected inflation from 2011 until 2015 (20). This resulted in substantial increases in the tobacco excise per pack, as well as overall revenue, which by 2015 had more than doubled from its low point in 2013 – equating to more than 50% in real terms (see Fig. 4.4.5). The success of this reform shows that revenues can be increased by higher rates despite the presence of a sizeable illicit market (113). More recent data show that revenues in Brazil declined in 2015 and 2016, coinciding with an increase in illicit trade, but also with an exceptionally bad economic recession that saw GDP decline by more than 3% in those years. 226 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.5 Average real excise tax per pack, real excise tax revenues and domestic cigarette sales in Brazil, 1999–2014 Notes: Data based on domestic sales and tobacco excise revenues, Federal Revenue Secretariat, indexed to 2013 Brazilian reals, using Consumer Price Index. Revenue collection indexed to 2013 reals, using Consumer Price Index. Source: (20). Improvements in tax administration and enforcement can also generate increases in revenues. In Kenya, several measures, including fiscal markings and, later, an advanced tracking and tracing system, improved collection, resulting in increases in both legal sales and tax revenues and a reduction in illicit sales (114). Moreover, these examples of poor governance indicate that attention should be focused on countries where a significant loss in administrative and enforcement capacity un- dermined revenue collection. 4.4.3 CONCLUSIONS The tobacco industry uses revenue concerns as a SCARE tactic to avoid, dilute and/or delay tobacco tax increases. The argument that higher taxes will decrease revenue is theoretically plausible, but real-world examples have demonstrated that this has not occurred. Furthermore, simulations show that even large tax increases in current average tax shares yield substantial revenue gains. The use of the Laffer curve by the tobacco industry should be challenged and refuted. The relatively price inelastic nature of cigarette demand combined with the Real excise tax amount per pack Domestic sales Real excise tax revenue D om es tic s al es (b ill io n pa ck s) Re al e xc is e ta x re ve nu es (b ill io n 20 13 re ai s) Real excise tax am ount per pack (2013 reais) 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 0 1 2 3 4 5 6 0 1 2 3 CHAP T ER 4. PO LI T I C AL ECO N OMY 227 low tax share and no overshifting of the tax means that most – if not all – countries are still far from the revenue-maximizing point, indicating that increases in taxes will lead to increases in revenues. The case studies in this section refute each of the tobacco industry’s arguments regarding alleged potential revenue loss due to tax increases. The experiences of South Africa, the Philippines and Ukraine demonstrate that large and regular tax increases result in large and consistent revenue increases. Well-designed tax structures have also proven to play an important role in generating revenues. The experience of Australia shows that even countries with already high tax rates and declining prevalence of tobacco use can increase revenues with regular, large tax increases. The experience of Canada warns against following the advice of the tobacco industry to decrease taxes as a way to fight illicit trade. It demonstrates clearly that decreasing tobacco taxes will decrease revenue and encourage consumption, rather than counteract illicit trade. The experience of Brazil shows that countries with substantial illicit trade issues can still increase revenue by increasing taxes. Finally, in the few cases where revenue decreases were seen, the reasons for the decreases were not strictly linked to tax increases. This was the case in Tonga, where the increase in tax was applied only to cigarettes and not to their close substitute, loose tobacco – leading smokers to switch products. In South Africa, a decrease in revenue was the result of the weakening of government institutions. And in Ukraine, real revenues decreased only during the two years when taxes were not increased above inflation. 228 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.5 SCARE TACTIC E: EMPLOYMENT 4.5.1 INTRODUCTION In opposing tax increases, the tobacco industry often seeks to frame tobacco taxes as an economic rather than a public health issue (5, 48, 53, 115). Particular emphasis is placed on the alleged threat that tax increases pose to employment in tobacco farm- ing and manufacturing, as well as other related industries (5). This so-called choice between health and jobs is, however, based on several false premises, including (5): 1. tobacco is a significant source of jobs within the context of broader labour markets, and domestic tobacco tax increases will have a drastic effect on domestic employment (48, 53, 116–117); 2. tobacco consumption is an indispensable engine for job creation (5, 48, 54); and 3. tobacco provides highly prosperous, sustainable and irreplaceable livelihoods (5, 53, 118). In reality, the relationship between tobacco taxation and employment is consider- ably more complex than the industry makes it out to be. In fact, there is ample evidence to show that tobacco taxes are a win-win for public health and the fiscal space, without measurable risks to employment. 4.5.2 THE LINK BETWEEN TOBACCO EMPLOYMENT AND TOBACCO TAX RATES Tobacco farming, production and manufacturing (including hand-rolling in some countries, most of them in South-East Asia) constitute a small proportion of the labour force, even in countries where the industry is most heavily concentrated (5, 48, 103, 116). Employment in tobacco farming and manufacturing has been declining globally due to advances in technology, trade liberalization, market consolidation and the privatization of formerly state-owned tobacco companies (5, 103, 119). These same trends have led to the heavy concentration of tobacco growing and manufacturing in only a handful of countries – and within these countries, often in only a small number of regions (103, 119–121). Even in those countries that lead in tobacco growing and manufacturing, tobacco’s overall share of total agricultural and manufacturing employment is relatively small and is often decreasing as efficiencies in production reduce labour intensity (5, 117–118, 122). Similarly, the industry’s claim that tobacco taxes reduce employment is exag- gerated and typically overlooks wider trends driving tobacco industry employment. Indeed, tobacco industry developments and innovations have played a greater role in the reduction of employment in the tobacco industry than have tobacco control policies (103). Despite industry claims that tobacco taxes can affect employment, CHAP T ER 4. PO LI T I C AL ECO N OMY 229 characteristics of the location of production – such as market size, labour costs, growing conditions and leaf preferences – have much more to do with tobacco industry interests than with the tobacco tax rate (103, 123). Moreover, jobs in countries that produce tobacco primarily for export are not greatly affected by reductions in local consumption resulting from tax increases (5, 103, 116, 123). Finally, it has been demonstrated that tobacco tax increases do not have a significant effect on employment in the retail sector, as most retail businesses sell other goods (103). Estimates of the gross employment impact of tobacco tax hikes demonstrate that job losses that do occur can be more than compensated for by increases in revenue. A 2018 World Bank study estimated that in Indonesia, for example, an ambitious tax reform that would simplify tiers and increase prices by close to 50% would reduce gross employment in the tobacco manufacturing sector by less than 0.5% (a loss of 2 914 jobs). The government could provide income support to the displaced workers (for example, through training, temporary transport/mobility or income support) with less than 2% of the revenue gained from the tax increase (117). Similarly, a 2019 study by Bangladesh’s National Board of Revenue estimated that a substantial increase in tobacco taxation would cause 7 012 lost jobs, but that the total income associated with these job losses in the bidi industry would amount to only 3.5% of the revenue gained (120). Accordingly, increased revenue can more than compensate for the expenditure of supporting those who lose jobs and need to acquire new skills before transitioning to new employment (120). Box 4.5.1 Employment fears deployed to frustrate tobacco tax reform in Indonesia In 2017, the Indonesian Ministry of Finance decided to implement tobacco tax in- creases by 2019 and tier simplification by 2021 (124). This resolution was, however, abandoned within a year, after a concerted campaign by tobacco industry actors and their allies to reframe the increase as an economic issue with a focus on, among other things, the effect the tobacco tax increase would have on employment (124). This defeat for the tobacco tax initiative came despite estimations of how the loss of income associated with lost jobs would be dwarfed by the additional revenue gained by the tax (117). Earlier analysis had estimated an overall large net positive impact on employment from tobacco tax increases (125), which illustrates how evidence that challenges assumptions around the negative socioeconomic impacts of tobacco control tends to be discounted (126). In this case, tobacco industry arguments seem to have resonated strongly with politicians from the electoral districts of West Java, East Java, Central Java and West Nusa Tenggara, where employment in tobacco farming and manufacturing is concentrated (124). Although tobacco manufacturing 230 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N represented only 5.13% of total manufacturing employment, the concentration of the job and economic activity meant that arguments regarding employment were particularly salient (117,119). Accordingly, tobacco industry arguments that tobacco taxation would negatively impact employment and farmer livelihoods prevailed, despite strong opposing evidence (119). Concentration of tobacco industry activity within countries poses particular obstacles to overcoming industry arguments on employment and needs to be given careful attention. 4.5.3 THE EVIDENCE ON THE NET EFFECT OF TOBACCO TAX INCREASES ON EMPLOYMENT A proper analysis of the effect of tobacco tax increases on employment must examine their impact on net or economywide employment. Decreases in expenditures on tobacco associated with tobacco control do not mean that expenditures simply disappear; rather, they are redistributed towards consumption of other goods and services, thereby generating employment elsewhere in the economy (5, 103, 123). Similarly, though the effect of higher tobacco taxes on net consumption is arguably more ambiguous, revenues from this intervention do generate spending, invest- ment and employment in public services such as health and education (5). Tobacco control polices usually have a marginal neutral or positive effect on net employment, particularly in countries that are net importers of raw or manufactured tobacco products, as expenditures on these imported items tend to flow out of the country (5). Export-oriented tobacco producers are less sensitive to local demand and are not significantly affected by domestic tobacco tax measures, which likely have a near-neutral net impact (5). In some cases, the net employment impact is a very small negative number, typically less than 1% (127–128). A recent study estimated that in the United Republic of Tanzania – a large tobacco-producing and exporting country – a 30% reduction in smoking prevalence would result in a net employment decline of just 0.5% across the economy as a whole (129). A similar study of Pakistan found that, with some variance depending on where spending was redistributed from tobacco consumption, the overall net effect on employment from a significant reduction in expenditure on cigarette employment – 1 billion rupees – would be a gain of between 6 651 and 5 803 jobs (122). This increase would occur because expenditure on cigarettes produces much less employment in the broader economy than expenditure on food and education (122). In the United Republic of Tanzania, as elsewhere, increased revenue could be used to assist those who lose employment with transitioning to new livelihoods. CHAP T ER 4. PO LI T I C AL ECO N OMY 231 4.5.4 THE VIABILITY OF BETTER LIVELIHOODS In arguing against tobacco tax increases, the tobacco industry advances the myth that people employed in tobacco production – particularly tobacco farming, but also manufacturing – lack any other prospect for a comparably attractive livelihood. However, studies based on extensive survey data in Indonesia, Kenya, Malawi, the Philippines and Zambia have shown that despite needing to commit significant amounts of labour to their crop, tobacco farmers often suffer losses rather than gain profits (119, 121, 123, 130–132). Furthermore, the Indonesian studies demonstrate that tobacco farming has a negative impact on household income and opportunity compared with the experience of other farming households that have given it up (119). Declines in consumption as a result of tobacco tax increases are gradual and susceptible to the same progressive adaptation that has occurred for decades (5, 103). While there will be a need in some countries for the government to help farmers transition to other crops or industries in the longer term, this process will not be a major short-term shock to employment or the wider economy (123). Because tobacco growing and manufacturing can be concentrated in just a few locations within a country, job losses within the tobacco industry might have a disproportionate effect in one location, while employment gains from reduced consumption may be spread across the whole country (120, 123). A study of the employment effects of tobacco tax increases in Bangladesh estimated that up to 60% of all job losses would occur in only two districts – among the poorest in the country – due to the high level of industry concentration (120). Studying the need for support, as well as the means of delivery and funding of support, is particularly necessary in these circumstances. Beyond the need to ensure equity and support employment, a failure to provide for targeted relief can exacerbate fear of job losses and may prove fatal to a tobacco tax proposal (120). Box 4.5.2: Supporting alternative livelihoods in the Philippines The Philippines earmarked 15% of the revenue from a 2012 increase in tobacco taxes to supporting economically viable alternative livelihoods for tobacco farmers and workers (5). Tobacco farming in the Philippines is regionally concentrated, and the tobacco industry had previously been successful in deploying concern for smallholder tobacco farmers to undermine tobacco control measures (136). The provision of economic support was a politically effective countermeasure to tobacco industry SCARE tactics and eased the tax increase’s passage. Given the Philippines’ integration with global tobacco markets and demand, tobacco farmers have not been seriously affected by the tobacco tax increase and 232 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N reduced domestic demand (108, 132). Nevertheless, transfers to tobacco-growing regions from the earmarked tax have been substantial (108). These funds are support- ing gradual transitions to alternative livelihoods, with farmers being encouraged to take up alternative crops, as well as establishing infrastructure, such as market-to-farm roads, that will make these alternative crops more economically viable (108, 133–134). Supporting alternative livelihoods for farmers and other tobacco workers is important because it can offset the political effect of industry arguments, even though domestic tax increases usually have only a modest and gradual effect on employment. There are various models for supporting alternative livelihoods when employment in the tobacco sector gradually diminishes due to decreases in either global or national demand. The Philippines is exemplary, but many other countries have either implemented or experimented with supporting crop transitions. Turkey’s alternative crop pro- gramme, implemented in anticipation of the privatization of the country’s cigarette monopoly, has proven effective in supporting many tobacco farmers’ move to other crops (135). Smaller-scale crop substitution projects in Kenya and Yunnan Province in China have shown how financial, regulatory and infrastructure support from government can contribute to crop transitions (5, 53). Argentina, Bangladesh, Mexico and the state of Maryland in the United States provide additional case studies of how governments can support these transitions (5, 136). 4.5.5 CONCLUSIONS The tobacco industry exaggerates the importance of tobacco employment and over- states the impact that domestic demand reduction due to local taxes will have on tobacco farmers serving a global market. The industry also simplifies employment’s relationship with taxation by focusing only on gross employment in tobacco, which ignores the reality that expenditures on tobacco do not disappear but rather are redistributed for other consumption that can produce a similar or higher number of jobs. Many detailed studies have found that tobacco growing is much less profitable and sustainable than the tobacco industry claims. Tobacco farmers throughout the world have successfully transitioned to other crops, although the transition often requires temporary or additional support from the government or other stakeholders. The extent of such support is moderated by the reality that transition from tobacco to other crops is a long-term consideration. CHAP T ER 4. PO LI T I C AL ECO N OMY 233 4.6 EARMARKING TOBACCO TAX REVENUES TO FUND HEALTH 4.6.1 INTRODUCTION Earmarking tax revenues involves the separation of all or a portion of revenue from a tax or group of taxes to be put aside for a specific purpose (137). Globally, more than 80 countries earmark for health (138), and 37 earmark tobacco tax revenues for health (27). There are two main types of earmarks: hard – also called substantive – and soft, or symbolic (139). Hard earmarks link the expenditure with a revenue source in legislation. This can limit funding if the earmarked revenues are the main source of funding, or it can cause surpluses to accrue wastefully when more revenues are raised than may be expended for the earmarked purpose. Soft earmarks include dedicated funds or commitments to use funds for a particular purpose. They are not necessarily legally binding. For example, in France, the ma- jority of tobacco tax revenue is used to fund social security (which includes health insurance and health care), but there is no hard, formal earmark (140). Earmarks can also be some combination of hard and soft. In the Philippines, tobacco tax earmarks are legally binding, but earmarked revenues go to the general fund, and the Department of Health must submit an annual budget for covered programs as part of its budget request. Earmarking is a broad and contentious topic that goes beyond the specifics of tobacco tax earmarking. Discussions on the topic fall within the ambit of public financial management, and earmarking generally is not encouraged. From a tobacco control perspective, however, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. It is a tool to improve the political economy of tobacco taxation; it is a secondary issue only, after the primary goal of reducing demand for tobacco. One way to use earmarking to improve the political economy of tobacco taxation is to link the payment of tax by tobacco users to benefits they will receive through the funding of complementary tobacco control programmes, such as cessation support, or through increased funding for health programmes on which they will rely disproportionately. This is known as the benefit principle. Earmarking for tobacco control makes sense, as its financial cost is relatively small and tobacco tax reduces demand more effectively when implemented within a package of complementary tobacco control measures. Another way earmarking improves the political economy of tobacco taxation is by safeguarding against any perceived or potential negative ramifications of the tax itself. This is important for neutralizing erroneous but often convincing tobacco industry arguments against effective tobacco tax policies. For example, the Philippines earmarks the bulk of the additional revenues from sin taxes for the health insurance premiums of the poor. In addition, a portion 234 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the country’s tobacco tax revenues is earmarked to provide for the economic well-being of tobacco growers and tobacco growing regions, with the general aim of promoting economically viable alternatives to tobacco farming and manufacturing as a safeguard against the potential for reduced domestic tobacco demand (141). Tobacco tax earmarks are complex, however, and care is required when determin- ing whether a particular earmark is needed on the grounds of political economy and justified on the grounds of equity and economic efficiency. In assessing tobacco tax earmarks, many of the same criteria that have been used to assess the appropriateness of generic earmarks also apply. In the rest of this section, these criteria are set out and matched with reasons for the ability of well-designed tobacco tax earmarks to fulfil them. The types and structures of tobacco tax earmarking are explored alongside descriptions of country experiences to provide guidance on how tobacco tax earmarks are used, when they are justified and the best ways to design them. 4.6.2 CIRCUMSTANCES IN WHICH EARMARKS MAY BE SUITABLE Scepticism about earmarking is both long-standing and justified, but much of the debate concerns earmarking generally and is not specifically concerned with the merits of tobacco tax earmarking (138). The main concerns raised about earmark- ing are listed in Table 4.8, accompanied by suggestions for how earmarks may be structured to address these concerns. Table 4.8 Concerns about earmarking and suggested safeguards to avoid the concerns MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Democratic accountability and oversight: earmarks undermine democratic processes by impeding legislative and executive oversight over expenditure. Establishing proper oversight and accountability procedures is important to ensure funds are not mismanaged (138). Additionally, if a soft earmark structure, which transfers revenue to the general fund from which it is then allocated, is adopted, this will not be a concern. Budget rigidity: earmarking may create budget rigidity that can lead to inefficient allocation of resources (138). An earmark’s particular design determines how much rigidity is introduced (138). Flexible soft earmarks are less prone to introducing rigidity than hard earmarks. Concerns about rigidity can be reduced by the inclusion of a sunset clause that ensures that the earmark is automatically discontinued or reviewed after a set period of time has elapsed (138). A further safeguard is to establish the earmark as a waterfall account, with any excess revenue over a set amount being allocated to the general fund. CHAP T ER 4. PO LI T I C AL ECO N OMY 235 MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Fragmentation: earmarking can result in fragmented and uncoordinated expenditures. This means policies complementary to the earmarked purpose but outside of its purview may be unfunded (138, 142). This is a legitimate concern. The negatives of fragmentation cannot be entirely eliminated, but they may be outweighed by the other merits of tobacco tax earmarking. That said, proposals for tobacco tax earmarks should be scrutinized to ensure that the funded purpose is at least cost-effective. Decreased equity: equity will decrease if individual access to benefits is narrowly defined according to payments made. This issue is not likely to arise with tobacco tax earmarks but is conceivable and something that should be guarded against in an earmark’s design. Capture by special interests: because earmarks are often the result of political expediency, an earmarked purpose may be determined by powerful special interests promoting a tax’s passage rather than careful prioritization of resources (138). Well-designed earmarks will guarantee funding for underresourced programmes and high- priority programmes. While the above concerns may be valid and design does matter, tobacco tax and other health-promoting taxes are not subject to the same concerns when it comes to the justifiability of earmarking their revenue (138, 143). Some of the factors that distinguish tobacco tax earmarks from more general critiques of earmarking are listed in Table 4.9 (138). Table 4.9 Concerns about earmarking and distinguishing factor for tobacco tax earmarks GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Procyclicality: earmarked revenues are often procyclical and susceptible to booms and busts (138–139, 142). Tobacco tax revenues are generally not cyclical (they are recession-proof ), and revenue is predictable relative to most other indirect and direct taxes (103). Budget rigidity Tobacco tax earmarks necessarily involve only a relatively small proportion of the budget; therefore, the effect of any rigidity will be relatively insignificant. Partly because of the relatively small amounts involved, there is only limited real-world evidence of tobacco tax earmarks having introduced harmful rigidity (143).24 24 See also the subsection on the amount of money associated with tobacco tax earmarks in section 4.6.3. 236 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Capture by special interests All earmarks should be scrutinized to ensure that their funded purpose is cost-effective. In the case of tobacco tax earmarks, however, political economy considerations may mean that it is sufficient for a low-priority purpose to be funded if the funding will unlock the political will needed for effective tobacco tax increases. In these cases, special interests are being purposefully catered to in order to ensure that tobacco tax increases occur. Of course, arguments against tobacco taxes and tobacco tax earmarking are led by special interests such as the tobacco industry (144–147). Insufficient revenue: the earmarked revenue source may become insufficient for funding its purpose (139, 142). Even though revenue may decrease in the long term when more tobacco users quit, such decrease is expected to be gradual. See Section 4.4 for details on how, with effective design, revenue will generally increase even with declining consumption. In addition to these reasons why general concerns about earmarking do not fully apply to well-designed tobacco tax earmarks, there are a number of compelling reasons for tobacco tax earmarking to finance tobacco control or public health that argue in favour of its implementation: • Significant increases in excise taxes are the most effective, as well as the most cost-effective mechanism for reducing consumption, but they are best implemented as a part of a package of complementary tobacco control measures, such as the WHO MPOWER package. Earmarking tobacco tax revenue for interventions that may not be funded otherwise can strengthen overall tobacco demand reduction (148). • The political economy of tobacco tax increases also makes earmarks attractive: – People have been shown to be more supportive of tobacco tax increases when they know the revenues will be used for targeted social programmes (143, 149–150). Earmarking tax revenue for health or tobacco control frames tobacco tax as a public health intervention in the minds of the public, which may otherwise view it as merely a revenue source (138). Research has shown that using earmarking to link a tobacco tax to health can also help raise awareness about the dangers of tobacco use (143). – When tobacco tax revenue is earmarked for programmes that benefit vulnerable groups, the tax becomes more equity-enhancing. Although lower socioeconomic groups and young adults receive disproportionate health and economic benefits from tobacco tax increases over the medium term, these groups will expend a greater share of their income in the CHAP T ER 4. PO LI T I C AL ECO N OMY 237 short term because of tobacco taxes. Earmarking tobacco tax revenue for programmes such as UHC or cessation services that provide immediate benefits to these groups neutralizes some critiques of tobacco taxation (e.g. the 2009 United States federal excise tax increase and the 2012 Philippines Sin Tax Reform illustrate how equity-enhancing earmarking facilitated passage of substantial tax rises) (93, 138, 143, 150). 4.6.3 EARMARKING PRACTICES AND COUNTRY EXAMPLES Earmarking tax revenues for health is a common practice in 80 countries. In 2018, 37 countries from all regions of the world earmarked tobacco tax revenues for health purposes.25 Case studies in the political economy of tobacco tax earmarking In 2012, the Philippines comprehensively reformed tobacco and alcohol excise taxes. Tobacco taxes were increased significantly, and numerous tax tiers were reduced to only one tier by 2017. Although increasing revenue was a foremost motive for some officials, the reform was explicitly framed around boosting UHC funding and advancing public health by reducing alcohol and tobacco consumption. Earmark- ing of tax revenue for UHC was essential to the political compromise that made this trailblazing tax increase a reality. It ensured that the increase, which may have otherwise been perceived as regressive, was framed as a progressive public health measure in the public imagination, while also appeasing tobacco growers and their political representatives. Earmarking was also important because the earmark en- sured high-level support for the tax by achieving a key political priority (151). Its soft-earmark structure meant it was not a blank cheque to the Ministry of Health, and this addressed concerns within the Ministry of Finance. Similarly, in Australia, earmarking of revenue helped overcome community objections to tobacco taxes and tobacco control more generally that resulted from the tobacco industry’s sponsorship of sports and the arts in the 1980s. Attempts to completely ban tobacco advertising and sponsorship had been unsuccessful due to strong pressure from sports, arts and racing lobbies that claimed that a ban would harm these activities. States, starting with Victoria, responded by earmarking funding for Health Promotion Foundations that took over the tobacco industry’s sponsorship activities and also paid for antismoking campaigns. In 1997, these earmarks ended after a High Court ruling that the Constitution did not allow states to collect excise taxes. However, in recognition of the successful work of the Foundations, the federal 25 Details about earmarked taxes by country are provided at https://www.who.int/tobacco/global_report/ Table-9-4-Use-of-earmarked-tobacco-taxes.xls?ua=1. 238 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government began funding them directly from the federal budget (152). Although this example may be difficult to replicate precisely, it shows how earmarks with built-in sunset clauses for piloting cost-effective interventions can potentially graduate to funding from the general budget once they have proven their effectiveness. It also more generally shows how earmarks can disarm community objections, reframe tobacco tax increases and unlock the political will needed to advance effective tobacco control measures. Structures for managing earmarked tobacco taxes A 2016 review of nine countries’ tobacco tax revenue earmarking experiences identified three arrangements for governance and allocating revenue (151). Table 4.10 presents some examples of these allocation arrangements. In some countries, earmarked tobacco tax revenues are combined with alcohol tax revenues. Table 4.10 Illustrative arrangements for allocating earmarked tobacco tax revenues POSSIBILITIES FOR ALLOCATING TOBACCO TAX EARMARK REVENUE Forms of budget allocation Revenue goes to the general fund and is later assigned to the official actor(s) specified in the earmark. Revenues do not go through the general budget but are instead paid into a separate account belonging to the official actor(s) specified in the earmark. Earmarked tax revenue is paid directly to the account of the entity managing an autonomous or semi- autonomous fund. Examples In the Philippines, revenue goes to the general fund before being allocated to the Ministry of Health following submission of a budget for its use (140). In Romania, revenue goes directly into a Ministry of Health account that is distinct from the general fund. In Panama, revenue is paid into subaccounts of the three recipient agencies (the Ministry of Health, the National Cancer Institute and the Customs Authority). In Thailand, ThaiHealth directly receives the earmarked revenues in its own account. In Viet Nam, the Viet Nam Tobacco Control Fund receives the revenues directly into a subaccount it manages but that belongs to the Ministry of Health. Source: (151). Where is the money being spent? Earmarked tobacco tax revenues are used for a variety of health purposes, including tobacco control, health promotion and UHC. A wide variety of other programmes have also been funded with earmarks from tobacco taxes, including disaster relief (e.g. hospital medical supplies and equipment to treat COVID-19 in India), youth pro- grams, sports and craft jobs in Yemen, social cohesion in Morocco, health and social programs in areas dependent on tobacco growing in Argentina, health promotion CHAP T ER 4. PO LI T I C AL ECO N OMY 239 and tobacco control in Thailand and alternative livelihood programs for tobacco farmers as well as economic projects in tobacco-growing provinces in the Philippines. Table 4.11 shows the three main categories of health programmes to which ear- marked tobacco tax revenue is allocated, as well as a fourth miscellaneous category, with country-specific examples for each.26 Table 4.11 Programmes to which earmarked tobacco tax revenue is allocated TOBACCO CONTROL NCD PREVENTION AND CONTROL PROGRAMMES (otherwise indicated between brackets) HEALTH COVERAGE EXPANSION (e.g. through health insurance coverage) OTHER, MORE GENERAL OR UNSPECIFIED HEALTH PROGRAMMES Costa Rica, Côte d’Ivoire, the Islamic Republic of Iran, Madagascar, Panama (tobacco cessation and fighting illicit trade), Switzerland, Viet Nam Cook Islands, Costa Rica, Mauritania (anti-cancer research), Palau (NCD prevention only), Panama (National Institute of Oncology), Paraguay Colombia, Congo, Egypt, Palau, Philippines Algeria, Argentina, Bangladesh, Botswana, Cabo Verde, Chad (programmes delivering antiretroviral drugs), Colombia (sports), Comoros (sports, hospital emergencies), Congo (sports), Côte d’Ivoire (AIDS programme), El Salvador, Estonia (sports), Guatemala, Indonesia, the Islamic Republic of Iran (sports), Ireland, Jamaica, Lithuania (sports), Madagascar (sports), Morocco, Nepal, Paraguay (sports), Republic of Korea (health promotion), Romania, Thailand (health promotion), United States, Yemen (sports) Note: Countries appear in more than one column when their earmarked tax revenues are used in more than one specific health programme. Source: (27). The amount of money associated with tobacco tax earmarks Case studies of the experiences of nine countries in tobacco tax earmarking show that earmarked funds are relatively small in comparison with government spend- ing on health (see Table 4.12) and, consequently, even smaller in terms of GDP. Therefore, the argument that tobacco tax earmarks would introduce rigidity into public financial management may not apply. 26 Details about how the tobacco tax revenues are earmarked are given in Annex 4.2. 240 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.12 Proportion of earmarked tobacco tax funds in government expenditures COUNTRY ESTIMATED ANNUAL TOTAL FUNDS FROM EARMARKED TAX ANNUAL FUNDS FROM TOBACCO TAX EARMARKS as a % of general government expenditure on health in 2013 Botswana 2014–2015: 4 million pula (US$ 0.48 million) NA Egypt 2013–2014: 392 million Egyptian pounds (US$ 52.06 million); earmarked taxes only 1.8% of total taxes on cigarettes 1.086% Iceland 2014: 108.3 million kronor (US$ 0.89 million) 0.083% Panama 2014: US$ 27.8 million 1.322% Philippines 2014: 50.18 billion Philippine pesos (US$ 1.18 billion) NA Poland 2013: 1 million złoty (US$ 0.316 million) from general budget  0.001% Romania 2014: 1.1 million lei (US$ 0.33 million); 14.4% of total health budget 0.004% Thailand 2014: 4064.74 million baht (US$ 125.15 million); 1.78% of Ministry of Health budget and 1.84% of National Health Security Fund 0.932% Viet Nam 2014: 299.171 billion dong (US$ 13.91 million); 0.5% of national health budget 0.335% Source: (151). 4.6.4 CONCLUSIONS Despite the initial principled resistance to earmarking by some ministries of finance, experience has shown that the use of revenue from tobacco taxes and other taxes on the consumption of products that have negative externalities can ensure political as well as public support. Successful earmarking needs a well-developed structure for the use of funds for health purposes. Even intergovernmental organizations that are opposed to earmarking (e.g. the IMF) have acknowledged the justifiability of well-designed tobacco tax earmarks when revenue is directed to specific cost- effective programmes (153–154). The amounts of tobacco tax revenue effectively earmarked for health have been relatively small and could hardly introduce the feared rigidity in government budgets. Moreover, in some countries, those funds have helped to implement much-needed health programmes (e.g. Australia, the Philippines, Thailand). More governments are considering this option as a stable medium-term source of secure funding for programmes such as tobacco control. The payoffs will be seen in the future as fewer people fall ill and less medical care for tobacco-related illnesses is needed. In Australia, CHAP T ER 4. PO LI T I C AL ECO N OMY 241 an earmarked tax was used to fund a needed and underresourced programme that proved to be successful, effective and impactful; the programme is now sustainably funded, embedded in the federal budget. Earmarking is desirable in a particular political economy when it enables the implementation of effective tobacco taxation that will increase price and reduce consumption. It will, however, also be rational as a matter of public financial man- agement, economic efficiency and democratic governance when concerns such as the following are considered. Although not every question needs an affirmative answer, policy-makers who can answer yes to many of the following questions will likely be considering an effective and rational tobacco tax earmark:27 • Does the tobacco tax earmark’s purpose rationally connect with the recipient programme’s purpose? Earmarks that fund tobacco control or other health programmes are more economically rational under the benefit principle than those that fund unrelated programmes such as childhood education, even when the popularity of the unrelated programmes may make a tax increase politically palatable. • Does the tobacco tax earmark’s amount rationally connect with the needs of the recipient programme? Earmarked funds that cannot be absorbed by the recipient programme are, in effect, money taken away from other needs. • When a tobacco tax earmark funds health programmes, is this clearly com- municated to the public to ensure that the framing of the tobacco tax increase as a health measure reinforces the demand-reduction effect? • Is the programme being funded by the tobacco tax earmark a politically neglected but highly cost-effective or crucially needed programme that, once established as a proof of concept, has a chance of being funded out of the general budget? • Does the tobacco tax earmark’s purpose rationally connect with the effects of the tax itself? Earmarks that fund programmes that disproportionately benefit lower socioeconomic groups or that fund alternative livelihoods for former tobacco workers and farmers will have equity-enhancing effects that will reinforce the already progressive nature of tobacco taxes. • Is the scope of the earmark’s purpose narrow enough that it can be funded mostly from the tobacco tax earmark, to ensure that the revenue is additive and does not merely substitute for spending that would otherwise come from the general fund? • Does the design of the tobacco tax earmark provide for flexibilities that ensure that windfall revenue collection is not squandered on a purpose already saturated with overfunding? 27 Adapted from and informed by References 138–139, 143, 150. 242 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Does the design of the tobacco tax earmark include a sunset clause that triggers its automatic end or review? The presence of such a clause will guard against inefficiency in allocation arising and being left uncorrected due to political inertia. CHAP T ER 4. PO LI T I C AL ECO N OMY 243 REFERENCES 1. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tob Control. 2013; 22(2): 144–153 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 31 January 2021). 2. LaFaive MD, Drenkard S, Nesbit T. Strong link between cigarette tax and illegal smuggling rates [website]. Midland: Mackinak Center for Public Policy; 2018 (https://www.mackinac.org/strong- link-between-cigarette-tax-and-illegal-smuggling-rates#:~:text=An%20increase%20of%2011%20 percentage,packs%20to%2030.6%20million%20packs.&text=Tax%20evasion%20and%20avoidance%20 is,high%20excise%20taxes%20on%20cigarettes., accessed 31 January 2021). 3. Ross H, Blecher E. Illicit trade in tobacco products need not hinder tobacco tax policy reforms and increases. Chicago: University of Illinois at Chicago; 2019 (Tobacconomics White Paper; https:// tobacconomics.org/research/illicit-trade-in-tobacco-products-need-not-hinder-tobacco-tax- policy-reforms-and-increases-white-paper/#:~:text=additional%20related%20research-,Illicit%20 Trade%20in%20Tobacco%20Products%20Need%20Not%20Hinder%20Tobacco,Reforms%20 and%20Increases%20( White%20Paper)&text=The%20paper%20cites%20research%20 that,consumption%2C%20and%20improves%20overall%20health, accessed 31 January 2021). 4. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 12 October 2020). 5. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 6. Joossens L, Chaloupka FJ., Merriman D, Yürekli A. Issues in smuggling of tobacco products. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac00Box 370064B00PUBLIC0.pdf, accessed 18 February 2021). 7. Merriman D, Yürekli A., Chaloupka, FJ. How big is the worldwide cigarette smuggling problem. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac 00Box370064B00PUBLIC0.pdf, accessed 18 February 2021). 8. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 9. Merriman D. Understand, measure, and combat tobacco smuggling. Washington (DC): World Bank Group; 2003 (Economics of tobacco toolkit 7; https://openknowledge.worldbank.org/bitstream/ handle/10986/16267/805800NWP070Sm0Box0379807B00PUBLIC0.pdf?sequence=1&isAllowed=y, accessed 31 January 2021). 10. Progress report on the implementation of the Commission Communication. Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products – a comprehensive EU strategy. Brussels: European Commission; 2017 (COM(2017) 235 final; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:52017DC0235&from=en, accessed 31 January 2021). 11. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 12. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Reuter P, Majmundar M. (editors). Washington (DC): National Academies Press; 2015. 13. von Lampe K. The illegal cigarette trade. In: Natarajan M, editor. International crime and justice. New York: Cambridge University Press. 2010:148–54. 14. Joossens L, Raw M, From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–34 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 31 January 2021). 15. Gomis B, Lee K, Carrillo Botero N, Shepherd P, Iglesias RM. “We think globally”: the rise of Paraguay’s 24 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tabacalera del Este as a threat to global tobacco control. Global Health. 2018;14(1):110 (https:// globalizationandhealth.biomedcentral.com/articles/10.1186/s12992-018-0412-3, accessed 31 January 2021). 16. Communication from the commission to the Council and the European Parliament: Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products - A comprehensive EU Strategy. Brussels: European Commission; 2013 (https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52013DC0324&from=EN, accessed 31 January 2021). 17. Ross H, Vellios N, Smith KC, Ferguson J, Cohen JE. A closer look at ‘cheap white’ cigarettes. Tob Control. 2016;25(5):527–31 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/527.full.pdf, accessed 31 January 2021). 18. Nagelhout GE, van den Putte B, Allwright S, Mons U, McNeill A, Guignard R, et al. Socioeconomic and country variations in cross-border cigarette purchasing as tobacco tax avoidance strategy. Findings from the ITC Europe Surveys. Tob Control. 2014;23:i30–i38 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/23/suppl_1/i30.full.pdf, accessed 1 February 2021). 19. Gilmore AB, Rowell A, Gallus S, Lugo A, Joossens L, Sims M. Towards a greater understanding of the illicit tobacco trade in Europe: a review of the PMI funded ‘Project Star’ report. Tob Control. 2014;23:e51–e61 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/e1/e51.full.pdf, accessed 1 February 2021). 20. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Revista Panam Salud Publica. 2016;40(4):243–9 (https://www.scielosp.org/pdf/ rpsp/2016.v40n4/243-249, accessed 1 February 2021). 21. Operação Huno combate mercado clandestino de cigarros em sete Estados [Operação Huno fights clandestine cigar market in seven states]. Brasilia: Federal Police of Brazil; 2015 (in Portuguese) (http:// www.pf.gov.br/agencia/noticias/2015/11/operacao-huno-combate-mercado-clandestino-de-cigarros- em-seis-estados, accessed 12 October 2020). 22. Gilmore AB, Gallagher AWA, Rowell A. Tobacco industry’s elaborate attempts to control a global track and trace system and fundamentally undermine the Illicit Trade Protocol. Tob Control. 2019;28:127–40 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/28/2/127.full.pdf, accessed 1 February 2021). 23. Technical assessment of the experience made with the Anti-Contraband and Anti-Counterfeit Agreement and general release of 9 July 2014 among Philip Morris International and affiliates, the Union and its Member States. Brussels: European Commission; 2016 (SWD (2016) 44 final; https://ec.europa.eu/anti- fraud/sites/antifraud/files/technical_assessment_pmi_24022016_en.pdf, accessed 1 February 2021). 24. Gallagher AWA, Evans-Reeves KA, Hatchard JL, Gilmore AB. Tobacco industry data on illicit tobacco trade: a systematic review of existing assessments. Tob Control. 2019;28(3):334–45 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/28/3/334.full.pdf, accessed 1 February 2021). 25. Joossens L, Merriman D, Ross H, Raw M. How eliminating the global illicit cigarette trade would increase tax revenue and save lives. Paris: International Union Against Tuberculosis and Lung Disease; 2009 (https://theunion.org/sites/default/files/2020-08/How%20Eliminating%20the%20Global%20Illicit%20 Cigarette%20Trade%20Would%20Increase%20Tax%20Revenue%20and%20Save%20Lives%20-%20 EN.pdf, accessed 1 February 2021). 26. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 27. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 28. Prieger JE, Kulic J. Cigarette taxes and illicit trade in Europe. Econ Inquiry. 2018;56(3):1706–23 (https:// papers.ssrn.com/sol3/papers.cfm?abstract_id=2718519, accessed 18 February 2021). 29. Stoklosa M. Prices and cross-border cigarette purchases in the EU: evidence from demand modelling. Tob Control. 2020;29:55–60. 30. Passport: illicit trade in tobacco products. In: Euromonitor International [website]. London: Euromonitor International; 2018 (restricted access). 31. The Global Illicit Trade Environment Index. London: The Economist Intelligence Unit; 2018 (http:// illicittradeindex.eiu.com/documents/EIU%20Global%20Illicit%20Trade%20Environment%20Index%20 CHAP T ER 4. PO LI T I C AL ECO N OMY 245 2018%20-%20Overall%20Results%20White%20Paper%20June%206%20FINAL.pdf, accessed 1 February 2021). 32. Chaloupka FJ, Yürekli A, Fong GT. Tobacco taxes as a tobacco control strategy. Tob Control. 2012;21:172– 80 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/172.full.pdf, accessed 1 February 2021). 33. Chernick H, Merriman D. Using littered pack data to estimate cigarette tax avoidance in NYC. Natl Tax J. 2013;66(2):635–68 (https://www.ntanet.org/NTJ/66/3/ntj-v66n03p635-668-littered-pack-data- cigarette-tax.pdf, accessed 1 February 2021). 34. Liber AC, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia Survey. Tob Control. 2015;24:iii83–iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed 1 February 2021). 35. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 36. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 37. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 38. Paraje G, Araya D, Drope J. Illicit cigarette trade in metropolitan Santiago de Chile. Tob Control 2020;29:68–73. 39. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 40. Saenz de Miera Juarez BS, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020(0):1-7 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 1 February 2021). 41. Lee K, Collin J. ‘Key to the future’: British American Tobacco and cigarette smuggling in China. PLoS Medicine. 2006;3(7):e228 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1502159/pdf/pmed.0030228. pdf, accessed 2 February 2021). 42. Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38– i43 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i38.full.pdf, accessed 2 February 2021). 43. Aziani A, Kulick J, Norman N, Prieger JE. Empty discarded pack data and the prevalence of illicit trade in cigarettes; 2017. (https://ssrn.com/abstract=2906015, accessed 12 October 2020). 44. Philip Morris International. The Perspective of PM International on Smoking and Health Issues. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1985. Bates no. 2023268328-2023268349 (https://www.industrydocuments.ucsf.edu/tobacco/docs/#id=pxpb0040, accessed 5 February 2021). 45. Mandal S, Gilmore A, Collin J, Weishaar H, Smith K, McKee M. Block, amend, delay: tobacco industry efforts to influence the European Union’s Tobacco Products Directive (2001/37/EC). Smokefree Partnership/Cancer Research UK; 2009 (http://www.smokefreepartnership.eu/images/Reports-and- Position-Papers/EU_TI_TPD_report_May_2012.pdf, accessed 15 October 2020). 46. Bishop DJ. Key area paper: indirect taxation – British American Tobacco. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1994. Bates no. 503911771-503911780 (http:// bat.library.ucsf.edu/data/h/f/k/hfk11a99/hfk11a99.pdf, accessed 15 October 2020). 47. Zhou S, Liberman JD, Ricafort E. The impact of the WHO Framework Convention on Tobacco Control in defending legal challenges to tobacco control measures. Tob Control. 2009:28:s113-s118 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/28/Suppl_2/s113.full.pdf, accessed 2 February 2021). 48. Gilmore AB, Fooks G, Drope J, Bialous SA, Jackson RR. Exposing and addressing tobacco industry 246 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N conduct in low-income and middle-income countries. Lancet. 2015;385:1029–43. 49. Ulucanlar S, Fooks G, Gilmore A. The policy dystopia model: an interpretive analysis of tobacco industry political activity. PLoS Med. 2016:13(9):e1002125 (https://journals.plos.org/plosmedicine/ article?id=10.1371/journal.pmed.1002125, accessed 2 February 2021). 50. Apollonio D, Glantz S. Tobacco manufacturer lobbying to undercut minimum price laws: an analysis of internal industry documents. Tob Control. 2020;29:e10-e17 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC7374022/pdf/nihms-1590806.pdf, accessed 2 February 2021). 51. Crosbie E, Glantz S. Tobacco industry argues domestic trademark laws and international treaties preclude cigarette health warning labels, despite consistent legal advice that the argument is invalid. Tob Control. 2012;23:e7 (https://www.researchgate.net/publication/233769722_Tobacco_industry_ argues_domestic_trademark_laws_and_international_treaties_preclude_cigarette_health_warning_ labels_despite_consistent_legal_advice_that_the_argument_is_invalid, accessed 2 February 2021). 52. Puig S. Tobacco litigation in international courts. Harv Int Law J. 2016;57(2):1–68. 53. Crooked nine: nine ways the tobacco industry undermines health policy. New York: Vital Strategies; 2019 (https://exposetobacco.org/wp-content/uploads/2019/09/Crooked-9-STOP.pdf, accessed 2 February 2021). 54. Balwicki L, Stokłosa M, Balwicka-Szczyrba M, Tomczak W. Tobacco industry interference with tobacco control policies in Poland: legal aspects and industry practices. Tob Control. 2016;25(5):521–26 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/521.full.pdf, accessed 2 February 2021). 55. Sud SR, Brenner JE, Shaffer ER. Trading away health: the influence of trade policy on youth tobacco control. J Pediatrics. 2015;166(5):1303–07 (https://www.jpeds.com/action/showPdf?pii =S0022-3476%2815%2900017-7, accessed 2 February 2021). 56. Lee S, Ling PM, Glantz SA. The vector of the tobacco epidemic: tobacco industry practices in low and middle-income countries. Cancer Cause Control. 2012;23:117–29 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC3332051/pdf/nihms-367996.pdf, accessed 2 February 2021). 57. Zhou S, Liberman J. The global tobacco epidemic and the WHO Framework Convention on Tobacco Control—the contributions of the WHO’s first convention to global health law and governance. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018 (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3339718, accessed 18 February 2021). 58. Legislative Consultation with Constitutional Division of the Supreme Court, No. 2012-003918. Constitutional Division of the Supreme Court of Costa Rica; 20 March 2012 (https://www. tobaccocontrollaws.org/files/live/litigation/830/CR_Legislative%20Consultation%20with%20.pdf, accessed 2 February 2021). 59. British American Tobacco Kenya Ltd v Ministry of Health, Petition No. 143 of 2015. High Court of Kenya; 2016 (https://www.tobaccocontrollaws.org/files/live/litigation/2643/KE_British%20American%20 Tobacco%20Kenya.pdf, accessed 2 February 2021). 60. British American Tobacco v Ministry of Health, et al, Civil Appeal No. 112 of 2016. Kenyan Court of Appeal; 2017 (https://www.tobaccocontrollaws.org/files/live/litigation/2521/KE_British%20 American%20Tobacco%20Ltd%20v.pdf, accessed 2 February 2021). 61. Thailand – customs and fiscal measures on cigarettes from the Philippines: Report of the Panel. Geneva: World Trade Organization; 2010 (WT/DS371/R; https://docs.wto.org/dol2fe/Pages/SS/directdoc. aspx?filename=Q:/WT/DS/371R.pdf&Open=True, accessed 2 February 2021). 62. Tribunal Constitucional de Chile [Constitutional Court of Chile] (1995 Rol No. 219-95; https://www. tribunalconstitucional.cl/descargar_sentencia2.php?id=3074, accessed 4 April 2020.) 63. McGrady B. International investment law. In: Voon T, Mitchell AD, Liberman J, editors. Regulating tobacco, alcohol and unhealthy foods: the legal issues. New York: Routledge; 2014. 64. Marvin Roy Feldman Karpa v United Mexican States, Case No. ARB(AF)/99/1 International Centre for Settlement of Investment Disputes; 2002 (https://www.italaw.com/sites/default/files/case-documents/ ita0319.pdf, accessed 2 February 2021). 65. Feldman Karpa v Mexico (2002 ICSID Arbitral Tribunal) (https://www.italaw.com/sites/default/files/ case-documents/ita0319.pdf, accessed 18 February 2021). 66. Undang-Undang Nomor 11 Tahun 1995 tentangCukai (Lembaran Negara Republik Indonesia Tahun1995 CHAP T ER 4. PO LI T I C AL ECO N OMY 247 Nomor 76) (Excise Law,in Indonesian). Article 5. 67. Excise Regulation (2011) Reference No. PMK Number 167/PMK.011/2011 (in Indonesian) (https:// peraturan.bpk.go.id/Home/Details/145897/pmk-no-167pmk0112011, accessed 3 February 2021). 68. Mahkamah Agung. Indonesia Supreme Court, Case No. 73/PDT/2018/PT SBY (https://putusan3. mahkamahagung.go.id/direktori/putusan/74efff9cfe18d60990fcde19583ecd5f.html, accessed 6 February 2021). 69. Maya D. Analisis kebijakan kenaikam tarif cukai dan penyesuaian batasan harga jual eceran hasil tembakau [Policy analysis on excise tax increases and price limit adjustments retail tobacco products]; 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-09/S45937-Deacy%20Maya accessed 3 March 2020). 70. Indriawati EN, Susanti L, Qurnia MA. Kebijakan tarif cukai hasil tembakau terhadap perekonomian [Tobacco excise tariff policy on the economy]; 2014 (in Indonesian). (https://www.academia. edu/6611977/MAKALAH_CHT_kel_6, accessed 15 October 2020). 71. Arroyani DW. Evaluasi Kebijakan Penetapan Tarif Cukai Hasil Tembakau Berdasarkan PMK [Evaluation of Tobacco Excise Tariff Determination Policy Based on PMK].167/PMK.011/2011. 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-08/S44664-Dwira%20Wanti%20Arroyan, accessed 15 October 2020). 72. Ngo Ngo Ha and Another v. State of New South Wales and Others (Matter No S 45 of 1996); Walter Hammond & Associates Pty Ltd v. State of New South Wales and Others (Matter No S 165 of 1996). Constitutional law; (1997) 146 ALR 355: 5 August 1997 (https://staging.hcourt.gov.au/assets/ publications/judgments/1997/029--NGO_NGO_HA_AND_ANOR_v_STATE_OF_NEW_SOUTH_WALES_&_ ORS_(Matter_No_S_45_of_1996);_WALTER_HAMMOND--(1997)_146_ALR_355.html, accessed 2 February 2021). 73. McGrady B. Trade and public health: the WTO, tobacco, alcohol, and diet. New York: Cambridge University Press; 2011. 74. Giorgetti C. Health and international investment law. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018. 75. British American Tobacco v Attorney General and Center for Human Rights and Development. Constitutional Court of Uganda Petition No.46 of 2016; 2017 (https://media.ulii.org/files/judgments/ ugcc/2019/4/2019-ugcc-4_0.pdf, accessed 18 February 2021). 76. Lim CL, Ho J, Paparinskis M. International investment law and arbitration: commentary, awards and other materials. Cambridge: Cambridge University Press; 2018. 77. Dominican Republic – measures affecting the importation and internal sale of cigarettes : Report of the Panel. Geneva: World Trade Organization; 2004 (WT/DS302/R; https://docs.wto.org/dol2fe/ Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302R.pdf&Open=True, accessed 21 February 2021) 78. Dominican Republic — measures affecting the importation and internal sale of cigarettes: Report of the Appellate Body. Geneva: World Trade Organization; 2005 (WT/DS302/AB/R; https://docs.wto. org/dol2fe/Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302ABR.pdf&Open=True, accessed 21 February 2021). 79. Gilmore A. Tobacco and transition: understanding the impact of transition on tobacco use and control in the former Soviet Union [Thesis]. London: London School of Hygiene and Tropical Medicine, June 2005 (https://escholarship.org/uc/item/3rw2c04c, accessed 2 February 2021). 80. MacKenzie R, Collin J, Sopharo C, Sopheap Y. ‘Almost a role model of what we would like to do everywhere’: British American Tobacco in Cambodia. Tob Control. 2004;13(suppl II): ii112–ii117 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/13/suppl_2/ii112.full.pdf, accessed 2 February 2021). 81. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tobacco Control 2013;22(2):144-53 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 21 February 2021). 82. Price and tax. In: TobaccoTactics [website]. Bath: University of Bath; 2020 (https://tobaccotactics.org/ topics/price-and-tax/, accessed 15 December 2020). 83. Vulovic V, Chaloupka FJ. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation—commentary. Tob Control. 24 June 2020. 248 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 84. Verguet S, Kearns PKA, Rees VW. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation. Tob Control. 23 June 2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/08/11/tobaccocontrol-2019-055315. full.pdf, accessed 2 February 2021).. 85. Palipudi KM, Gupta PC, Sinha DN, Andes LJ, Asma S, McAfee T. Social determinants of health and tobacco use in thirteen low and middle income countries: evidence from Global Adult Tobacco Survey. PLoS ONE. 2012;7(3):e33466 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3306395/pdf/ pone.0033466.pdf, accessed 2 February 2021). 86. Systematic review of the link between tobacco and poverty. Geneva: World Health Organization; 2014 (http://www.who.int/tobacco/publications/economics/9789241507820/en, accessed 21 October 2020). 87. Goodchild M, Sinha P, Gill Munish V, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/bitstream/handle/10665/331836/whoseajph2020v9i1p73. pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 88. Faruque GM, Ahmed M, Huq I, Parven R, Wadood SN, Chowdhury SR, et al. The economic cost of tobacco use in Bangladesh: a health cost approach. Dhaka: Bangladesh Cancer Society; 2020 (https://www. researchgate.net/publication/342445467_The_Economic_Cost_of_Tobacco_Use_in_Bangladesh_A_ Health_Cost_Approach, accessed 2 February 2021). 89. John RM, Sung H, Max WB, Ross H. Counting 15 million more poor in India, thanks to tobacco. Tob Control. 2011;20:349–52 (https://www.researchgate.net/publication/49809290_Counting_15_million_ more_poor_in_India_thanks_to_tobacco, accessed 2 February 2021). 90. Liu Y, Rao K, Hu TW, Sun Q, Mao Z. Cigarette smoking and poverty in China. Soc Sci Med. 2006;Dec;63(11):2784–90. 91. Fuchs A, González Icaza F, Paz D. Distributional effects of tobacco taxation – a comparative analysis. Washington (DC): World Bank Group; 2019 (Policy research working paper 8805; https://openknowledge. worldbank.org/bitstream/handle/10986/31534/WPS8805.pdf?sequence=4&isAllowed=y, accessed 2 February 2021). 92. Smith CE, Hill SE, Amos A. Impact of population tobacco control interventions on socioeconomic inequalities in smoking: a systematic review and appraisal of future research directions. Tob Control. 29 September 2020 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/09/28/ tobaccocontrol-2020-055874.full.pdf, accessed 2 February 2021). 93. Furman J. Six lessons from the U.S. experience with tobacco taxes – World Bank conference “Winning the tax wars: global solutions for developing countries”. Washington (DC): Obama White House Archives; 24 May 2016 (https://obamawhitehouse.archives.gov/sites/default/files/page/files/20160524_cea_ tobacco_tax_speech.pdf, accessed 21 October 2020). 94. Önder Z, Yürekli A. Who pays the most cigarette tax in Turkey. Tob Control. 2016;25(1):39–45. 95. Verguet S, Gauvreau CL, Mishra S, MacLennan M, Murphy SM, Brouwer ED, et al. The consequences of tobacco tax on household health and finances in rich and poor smokers in China: an extended cost-effectiveness analysis. Lancet Glob Health. 2015;3:e206–16 (https://www.thelancet.com/action/ showPdf?pii=S2214-109X%2815%2970095-1, accessed 2 February 2021). 96. Fuchs Tarlovsky A, González Icaza F. The welfare and distributional effects of increasing taxes on tobacco in Vietnam. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/ curated/en/758391561477648302/pdf/The-Welfare-and-Distributional-Effects-of-Increasing-Taxes- on-Tobacco-in-Vietnam.pdf, accessed 2 February 2021). 97. Expanding the global tax base: taxing to promote public goods: tobacco taxes. Washington (DC): World Bank Group; 2016 (http://pubdocs.worldbank.org/en/799981480947591462/Taxing-to-Promote- Public-Goods-tobacco-taxation-chapter-conference-25-nov-2016-final-version-pmarquez-002.pdf, accessed 2 February 2021). 98. Blecher E. Increasing tobacco taxes significantly will increase revenues. Chicago: University of Illinois at Chicago; 2018. (Tobacconomics policy brief; https://tobacconomics.org/files/research/470/Laffer- Curve-Policy-Brief_v2.2.pdf, accessed 2 February 2021). 99. Savedoff W, Lopert R. Sugar, rum, and tobacco: domestic resource mobilization for low-income countries through excise taxes. Washington (DC): Centre for Global Development; 2020 (CGD Policy Paper CHAP T ER 4. PO LI T I C AL ECO N OMY 249 164, https://www.cgdev.org/sites/default/files/PP164-Savedoff-Health-Taxes-DRM.pdf, accessed 2 February 2021). 100. Lavado R, Postolovska I, Sotomayor RE. Beyond the perceived regressivity of higher tobacco taxes, turning short-term losers into long-term winners? In: Tobacco reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 101. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 102. Tonga NCD related taxation policy assessment preliminary findings. Presentation made at the innovations and actions against NCDs conference ADB headquarters, Manila, 17 July 2018. 103. Moreno-Dodson B. Raising tobacco taxes: why, what, and how: where public health and economics converge. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e- library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 104. Fuchs A, Del Carmen G, Mukong AK. Long-run impacts of increasing tobacco taxes: evidence from South Africa. Policy Research Working Paper 8369. Washington [DC]: World Bank; 2018 (https://openknowledge. worldbank.org/bitstream/handle/10986/29497/WPS8369.pdf?sequence=4&isAllowed=y, accessed 18 February 2021) . 105. van Walbeek C. Case studies in illicit tobacco trade: South Africa. Tobacconomics fact sheet. Chicago: University of Chicago, Tobacconomics, Health Policy Center; 2020 (https://tobacconomics.org/files/ research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 106. van Loggerenberg, J. Tobacco Wars. Cape Town: NB Publishers; 2019. 107. Marquez P. Global tobacco control: a development priority. In: Tobacco tax reform at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads. pdf, accessed 2 February 2021). 108. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington [DC]: World Bank; 2017 (https:// tobacconomics.org/files/research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 109. Scollo M, Bayly M. Revenue from tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia. org.au/chapter-13-taxation/13-6-revenue-from-tobacco-taxes-in-australia, accessed 21 October 2020). 110. Scollo M, Bayly M. Tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia.org.au/ chapter-13-taxation/13-2-tobacco-taxes-in-australia, accessed 21 October 2020). 111. Breton E, Richard L, Gagnon F, Jacques M, Bergeron P. Fighting a tobacco-tax rollback: a political analysis of the 1994 cigarette contraband crisis in Canada. Journal of Pub Health Policy. 2006;27:77–99. 112. Fanta E, Garcia M. Preventing and controlling illicit tobacco flows. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB- Report-CrossRoads.pdf, accessed 2 February 2021). 113. Goodchild M, Perucic AM, Zheng R, Blecher E, Paul J. The health impact of raising tobacco taxes in developing countries. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/ taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 114. Ross H. Tracking and tracing tobacco products in Kenya. Prev Med. 2017;105:S15–S18 (https://reader. elsevier.com/reader/sd/pii/S0091743517301457?token=424D9EC1D42A9C69BDB55B90090C3F7853AA 20A4970D0E379C91EE64C73A273A0FA949557F4F120A0B636701581EC055, accessed 2 February 2021). 115. Mamudu H, Hammond R, Glantz S. Tobacco industry attempts to counter the world bank report curbing the epidemic and obstruct the WHO framework convention on tobacco control. Soc Sci & 250 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Med. 2008;67(11):1690–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2662513/pdf/nihms81164. pdf, accessed 2 February 2021). 116. Gilmore A. Understanding the vector in order to plan effective tobacco control policies: an analysis of contemporary tobacco industry materials. Tob Control. 2012;21:119–26 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/21/2/119.full.pdf, accessed 2 February 2021). 117. The economics of tobacco taxation and employment in Indonesia: policy implications technical brief. Washington (DC): World Bank Group; 2018 (https://openknowledge.worldbank.org/bitstream/ handle/10986/29814/126158-WP-P154568-PUBLIC-add-series-WBGPNIndoEmploymentUPDATEFin alWeb.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 118. Sahadewo GA, Drope J, Li Q, Nargis N, Witoelar F. Tobacco or not tobacco: predicting farming households’ income in Indonesia. Tob Control. 2020. 119. The economics of tobacco farming in Indonesia: health, population, and nutrition global practice. Washington (DC): World Bank Group, 2017 (http://documents1.worldbank.org/curated/ en/161981507529328872/pdf/120307-REVISED-PUBLIC-WBGIndoEconomicsTobaccoFarming.pdf, accessed 2 February 2021). 120. The revenue and employment outcome of biri taxation in Bangladesh. Dhaka: National Board of Revenue, Government of the People’s Republic of Bangladesh; 2019 (https://www.who.int/docs/ default-source/searo/bangladesh/pdf-reports/cat-2/biri-study-report-03-12-2019.pdf?sfvrsn=b8fea69_2, accessed 2 February 2021). 121. Magati P, Lencucha R, Li Q, Drope J, Labonte R, Appau A, et al. Costs, contracts and the narrative of prosperity: an economic analysis of smallholder tobacco farming livelihoods in Kenya. Tob Control. 2019;28(3):268–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6512316/pdf/nihms-1012064. pdf, accessed 2 February 2021). 122. Ghaus K, Iqbal MA, Sabir M, Saleem W, Ali A, Aamir N. Macroeconomic impacts of tobacco use in Pakistan. Karachi: The Social Policy and Development Centre; 2018 (https://tobacconomics.org/files/ research/483/Macroeconomic-impact-of-Tobacco-in-Pakistan.pdf, accessed 2 February 2021). 123. Ross H. The effects of tobacco taxes on employment and labor productivity. In: Tobacco tax reform: at the crossroads of health and development. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021) 124. Bigawanto M. Tobacco industry interference undermined tobacco tax policy in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance (SEATCA); 2019 (https://seatca.org/dmdocuments/ Indonesia%20TII%20in%20Tax.pdf, accessed 2 February 2021). 125. Ahsan A, Wyono N. Analysis of the impact of higher cigarettes prices on employment in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance; 2007 (https://seatca.org/dmdocuments/An%20 Analysis%20of%20the%20Impact%20of%20Higher%20Cigarette%20Prices%20on%20Employment%20 in%20Indonesia.pdf, accessed 2 February 2021). 126. Higashi H, Khuong TA, Ngo AD, Hill PS. Evidence and decision making: tobacco control policy and legislation in Vietnam. Int J of Health Plann. 2013;2(1):e72–e94. 127. Buck D, Raw M, Godfrey CA, Sutton M. Tobacco and jobs: the impact of reducing consumption on employment in the UK. York: University of York; 1995 (https://www.researchgate.net/ publication/4814523_Tobacco_and_jobs_the_impact_of_reducing_consumption_on_employment_ in_the_UK, accessed 2 February 2021). 128. Petkova R, Ileva R and Rusenov A. Tobacco control policies and tobacco industry employment in Bulgaria. In: Economic, social and health issues in tobacco control. Report of a WHO international meeting, Kobe, Japan, 3-4 December 2001. Geneva: World Health Organization; 2003. 129. Jha V, Narayanan BG, Wadhwa D, Tesche J. Economic and environmental effects of reduction in smoking prevalence in Tanzania. Tob Control; 2020;29:24–8 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/29/1/24.full.pdf, accessed 2 February 2021). 130. Lencucha R, Makoka D, Drope J, Appau A, Labonté R, Li Q, et al. Costs, revenues and profits: an economic analysis of smallholder tobacco farmer livelihoods in Malawi. Tob Control. 2017;26(6)634–40 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC5957074/pdf/nihms965842.pdf, accessed 2 February 2021). 131. Goma FM, Labonté R, Drope J, Li Q, Zulu R, Kangwa E. The economics of tobacco farming in Zambia: CHAP T ER 4. PO LI T I C AL ECO N OMY 251 tobacco farmers survey report. Lusaka: University of Zambia and American Cancer Society; 2019 (https://www.cancer.org/content/dam/cancer-org/research/economic-and-healthy-policy/zambia- report-7-16-19.pdf, accessed 28 October 2020). 132. Chavez JJ, Drope J, Li Q, Aloria MJ. The economics of tobacco farming in the Philippines. Quezon City: Action for Economic Reforms and Atlanta: American Cancer Society, 2016 (https://www.cancer. org/content/dam/cancer-org/research/economic-and-healthy-policy/the-economics-of-tobacco- farming-in-the-philippines.pdf, accessed 22 October 2020). 133. Augstin CP, Cardenas RP, Cortez JB, Gimongala GC, Ramos AR. The effects of the sin tax reform law of 2012 to tobacco farmers of Amulung, Cagayan. IJARMSS. 2017;6(12):57–75. 134. Mercado A. Tobacco tax: good for the tobacco farmers, too. Business World; 2019 [website] (https:// www.bworldonline.com/tobacco-tax-good-for-the-tobacco-farmers-too/, accessed 25 October 2020). 135. Yürekli A, Önder Z, Elibol M, Erk N, Cabuk A, Fisunoglu M, et al. The economics of tobacco and tobacco taxation in Turkey. Paris: International Union Against Tuberculosis and Lung Disease; 2010 (https:// www.who.int/tobacco/en_tfi_turkey_report_feb2011.pdf, accessed 2 February 2021). 136. Latif E, Warner V. Legislation and smoking prevention. In: Loddenkemper R, Kreuter M, editors. The Tobacco Epidemic: progress in Respiratory Research. 2015;42:136–148. 137. Tandon A, Cashin C. Assessing public expenditure on health from a fiscal space perspective. Washington (DC): World Bank Group; 2010 (http://documents1.worldbank.org/curated/en/333671468330890417/ pdf/560530WP0Box341penditureFiscalSpace.pdf, accessed 2 February 2021). 138. Cashin C, Sparkes S, Bloom D. Earmarking for health: from theory to practice. Geneva: World Health Organization; 2018 (https://apps.who.int/iris/bitstream/handle/10665/255004/9789241512206-eng. pdf?sequence=1, accessed 2 February 2021). 139. Wilkinson M. Paying for public spending: is there a role for earmarked taxes? Fiscal Studies. 1994;15(4):119–35 (https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1475-5890.1994.tb00213.x, accessed 2 February 2021). 140. ITAF: impôts et taxes affectées à la protection sociale. Les impôts et taxes affectés (ITAF) [taxes and duties allocated to social protection]. Vie publique: au coeur du débat public [website]. Paris: Directorate of Legal and Administrative Information; 2018 (in French) (https://www.vie-publique.fr/fiches/22010- impots-et-taxes-affectees-itaf-protection-sociale, accessed 31 October 2020). 141. Seventeenth congress of the Philippines, Metro Manila, RA 11346, 23 July 2018 (https://www. officialgazette.gov.ph/downloads/2019/07jul/20190725-RA-11346-RRD.pdf, accessed 31 October 2020). 142. McCleary W. The earmarking of government revenue: a review of some World Bank experience. World Bank Res Obs. 1991;6(1):81–104 (http://documents1.worldbank.org/curated/en/109871468764672767/ pdf/multi-page.pdf, accessed 2 February 2021). 143. Strategic investment of tobacco tax revenue. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/global/pdfs/en/strategic_investment_tobacco_tax_revenue. pdf, accessed 31 October 2020). 144. Bialous SA, Glantz SA. Arizona’s tobacco control initiative illustrates the need for continuing oversight by tobacco control advocates. Tob Control. 1999;8:141–51 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/8/2/141.full.pdf, accessed 2 February 2021). 145. Goldman LK, Glantz SA. The passage and initial implementation of Oregon’s measure 44. Tob Control. 1999;8:311–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1763939/pdf/v008p00311.pdf, accessed 2 February 2021). 146. Heiser PF, Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health. 1997;87(6):968–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380932/pdf/amjph00505-0082. pdf, accessed 2 February 2021). 147. Landman A, Bialick P. Tobacco industry involvement in Colorado. Chicago: American Lung Cancer Association; 2004 (https://escholarship.org/uc/item/9d880448#main, accessed 18 February 2021). 148. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010. (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 149. Vardavas CI, Filippidis FT, Agaku IT, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the 252 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N importance of earmarking the revenue to health care and tobacco control. Tob Induc Dis. 2012;10:21 (http://www.tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the- revenue-to-health-care-and-tobacco,66110,0,2.html, accessed 31 October 2020). 150. Bird RM. Tobacco and alcohol excise taxes for improving public health and revenue outcomes: marrying sin and virtue?. Washington (DC): World Bank Group; 2015 (Policy Research Working Paper 7500; https://openknowledge.worldbank.org/bitstream/handle/10986/23464/ Tobacco0and0al0ing0sin0and0virtue00.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 151. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 152. The establishment and use of dedicated taxes for health. Manila: World Health Organization Western Pacific Regional Office; 2004 (https://www.who.int/health_financing/documents/dedicated_taxes. pdf?ua=1, accessed 2 February 2021). 153. Teja RS. The case for earmarked taxes: theory and example. Washington (DC): International Monetary Fund; 1988 (IMF Working Paper; https://www.elibrary.imf.org/doc/IMF001/14514- 9781451922677/14514-9781451922677/Other_formats/Source_PDF/14514-9781455212002.pdf, accessed 2 February 2021). 154. Keen M. Taxation and development – again. Washington (DC): International Monetary Fund; 2012(IMF Working Paper WP/12/220O; (https://www.imf.org/external/pubs/ft/wp/2012/wp12220.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 253 ANNEX 4.1 METHODS TO ASSESS THE NATURE AND SIZE OF THE ILLICIT TOBACCO TRADE A4.1 DIRECT MEASUREMENT A4.1.1 SMOKER INTERCEPT AND PACK OBSERVATION SURVEYS Illicit trade can be measured directly by examining the cigarette packs of smokers. The smokers themselves can provide information on purchasing patterns, brand preferences and prices paid. Researchers can select individuals or retailers to survey based on a convenience sample (i.e. a sample that may not be representative) or a probability-based sample (i.e. a sample selected to be statistically representative of an underlying population). Data collected from a pack could reveal whether the pack is compliant or non- compliant with the local tax laws. Information can be obtained from objective markings such as brands, public health warning labels, tax stamps, foreign language labels or duty-free labels. During these stops, researchers can record demographic information (e.g. age and gender of the smoker), smoking-related history (e.g. number of cigarettes smoked per day) and price information. This is helpful in understand- ing the profile of smokers who are able and willing to avoid cigarette taxes. Pack observations can be used in conjunction with population-based household surveys to obtain population-based estimates of the illicit tobacco trade (1,2). For example, as part of a regular national health survey, Kaplan et al. conducted a cross-sectional study of smokers in Turkey, using a face-to-face interviewer-administered survey and pack observation (3). They were able to collect sociodemographic, lifestyle and medical details along with pack observations as part of the study protocol. Advantages and disadvantages of smoker intercept and pack observation surveys A primary advantage of conducting pack observation is that it is direct and objective, and smokers are not subject to any value judgements (2). Paired with survey data, pack observation can appropriately account for respondents who are not residents of the area in which they are surveyed (4). Disadvantages include the difficulty of identifying areas that are representative of the tobacco use population and the difficulty of sampling important subpopulations such as elderly and immobile smok- ers. Also, surveys conducted in the daytime may discount the number of youthful smokers who are in school. Another disadvantage is that a sizeable number of smokers may refuse to show their last-purchased pack (2). Kaplan et al. found that 24% of smokers sampled in Turkey did not show their cigarette pack to the study interviewer (3). This issue may be mitigated by asking users to provide information on the brand purchased, whether any public health warnings were posted and the 254 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N price paid (5). Although the responses are based on recall, they may still yield useful information. For example, Joossens et al. allowed smokers who did not show their packs to provide self-reported information and found no statistical differences in illicit packs between these respondents and those who did show their packs (2). Another obstacle to accurate measurement based on smoker intercepts is the inability to determine the tax payment of smokers who purchase single cigarettes, since these smokers are generally not given packs. However, information can still be captured in self-report surveys by asking smokers to report the brand purchased and price paid. Key study for readers to refer to for additional guidance: Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–447. A4.1.2 PACK RETURN AND SWAP SURVEYS Pack return and pack swap surveys fall within the broader category of pack ob- servation studies that use survey sampling techniques to examine smokers’ pack characteristics and to determine whether they are tax compliant. For these surveys, the unit of analysis is the individual. The main differences between pack swap and pack return surveys is that swap surveys offer the smoker a replacement pack, whereas pack returns are built into mail surveys and allow respondents to mail in their unopened packs. Pack swap and pack return surveys use probability and nonprobability sampling procedures. Probability sampling allows researchers to generalize to the broader population. Governments can use this method to rapidly assess the availability of illicit products in a given geographic area or to measure the share of the illicit market- place. Rapid assessment may be performed in instances where there is an emerging tobacco product (e.g. a new cheap white brand) or suspected counterfeiting of tax stamp features. Rapid assessment using a convenience sampling strategy could place researchers near busy intersections where they could ask smokers for permission to look at their cigarette packs or to take photographs that could be analysed later. A population-based study requires a sample that closely mirrors the tobacco use population. Advantages and disadvantages of pack return and pack swap surveys Pack swap and pack return surveys may help to overcome the stigma associated with traditional smoking surveys. For example, when researchers ask smokers to see their cigarette packs (or when they take photographs), no value judgements are made. These surveys are good rapid-assessment tools that can be used to examine CHAP T ER 4. PO LI T I C AL ECO N OMY 255 the effectiveness of physical features of a pack designed to deter illicit trade (e.g. packs that have tracking and tracing technology or high-tech stamps). In addi- tion, they can be supplemented with population-based tobacco use surveys. When coupled with such survey data, these methods allow researchers to obtain relevant information about the context of illicit purchases, including, for example, sources (e.g. street, peer networks, retail stores) and prices. Mail-in surveys are filled out in the comfort of the respondent’s home without the presence of family members or passers-by, which may assure them that responses will be kept confidential. A potential disadvantage is that smokers who purchase both illicit cigarettes and tax-paid cigarettes may disproportionately mail back compliant packs. In addition, in LMICs, this mode of survey distribution may be unreliable because of issues associated with mail delivery systems. Key study for readers to refer to for additional guidance: Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61-66. A4.1.3 LITTERED-PACK SURVEYS Littered-pack surveys, also known as empty discarded pack surveys, are used pre- dominantly in high- and middle-income countries (e.g. the United States, France, Canada, New Zealand, Mexico and Poland). This unobtrusive method relies on the premise that smokers publicly discard packs (e.g. on streets, sidewalks and in public trash cans). The packs bear characteristics that illustrate whether they are tax compliant (e.g. tax stamps, health warnings). For example, an Albanian health warning label on a cigarette pack discarded in Greece provides evidence that the pack was destined for the Albanian market. The pack may have been smuggled into Greece by criminal entrepreneurs or it may have been brought by a visitor. Collecting discarded packs from a representative geographic sample and examining these characteristics can provide estimates of tax compliance. Operationally, this data collection method uses an ecological approach whereby geographies are the units of analysis. Geographical units are meant to represent the smokers in the city/ country and can be administratively defined (e.g. by the country’s census bureau or transit zones) or may reflect researcher-defined neighbourhoods (e.g. half-mile buffer zones near bus stops or activity spaces). Researchers in Canada have expanded the littered-pack method to include col- lection and analysis of cigarette butts on 25 postsecondary campuses. The cigarette butts provide information on the brands sold (or lack thereof) and allow researchers to distinguish between legal and illegal products (6). A recent innovative expansion 256 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the littered-pack methodology is the collection of packs from cigarette retailers. John and Ross collected empty packs of tobacco products from a sample of registered and unregistered retailers in India (7). Collecting packs from retailers was relevant given that single cigarettes dominate the illicit market in India. Smokers who buy single cigarettes would be unable to provide a pack in a pack swap or street intercept survey, so collecting littered packs from the ground would undercount sales of single cigarettes. The feasibility of this method is dependent on the relationship between researchers and retailers (enhanced trust) and the efforts taken to ensure confidentiality. In some countries, retailer compliance with this research method might be strained because of concerns regarding confidentiality, since retailers may face criminal and civil penalties, depending on the research findings. Advantages and disadvantages of littered-pack surveys Littered-pack surveys are generally advantageous for governments because they facilitate comparison with industry estimates. This is one of the most-preferred methods because it yields estimates that are less likely to be biased due to issues of social desirability, recall error and confidentiality that plague survey research, and they are much less expensive than face-to-face interviews used in smoker intercept or household surveys. However, there are some issues regarding these surveys, espe- cially in high-income countries, including the inability to differentiate between tax avoidance and tax evasion (8). For example, a pack in Berlin that bears a Vietnamese tax stamp may have been smuggled in mass quantity or brought in by a temporary visitor. Researchers have circumvented this issue and broadened the umbrella to measure cigarette tax noncompliance considering the potential biases introduced by tourism. Another disadvantage of littered-pack surveys is that larger budgets are needed to employ field researchers to collect, code and analyse the data. Not all countries employ tax stamps on their cigarette packaging, which may make it difficult to measure tax compliance. These surveys also can underestimate the markets in low-income countries such as India, where the main item of illegal trade is single cigarettes (7). The surveys may also overestimate illicit trade if littering behaviour is correlated with willingness to engage in illicit trade. Finally, littered-pack surveys and butt collections provide information on the proportion of butts and packs that are illegal, not the proportion of smokers that purchase illegal cigarettes (6). Key studies for readers to refer to for additional guidance: Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38–i43. Merriman D. The micro-geography of tax avoidance: evidence from littered cigarette CHAP T ER 4. PO LI T I C AL ECO N OMY 257 packs in Chicago. Am Econ J Econ Policy. 2010;2(2):61–84. Stoklosa M., Paraje G., Blecher E., A Toolkit on Measuring Illicit Trade in Tobacco Products. A Tobacconomics and American Cancer Society Toolkit. Chicago, IL:Tobacconomics, Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/621/uic-illicit-trade-tool-kit-eng-v2.0-2. pdf, accessed 18 February 2021). A4.1.4 SELF-REPORT POPULATION SURVEYS Self-report surveys, when distributed to a representative sample of the population, can provide meaningful data on the prevalence of tax noncompliance. The surveys can be distributed to individuals or households in various ways, including face-to-face, telephone, mail and internet. Questions that specifically address illicit purchases can be added as supplementary questions to existing health or tobacco surveys. Some countries include such questions in their adult and youth tobacco surveys to estimate tax evasion/avoidance. For example, Canada’s annual Youth Smoking Survey asks smokers about the frequency of their purchases of First Nations/Native brand cigarettes (9-10). Davis et al. used data from the New York Adult Tobacco Survey to measure the source of purchase of the last cigarette pack purchased (i.e. Native American Reservations, lower-tax neighbouring states or countries, toll-free telephone numbers, the internet, duty-free shops) and the price paid (11). Twenty- eight nations currently use surveys to measure tax noncompliance as part of the ITC Project (12). Similar analyses can be conducted using questions from the Global Adult Tobacco Use Surveys. For example, Iglesias et al. used the Brazil Global Adult Tobacco Use Surveys to compare self-reported prices with a defined threshold retail price to estimate the proportion of illicit cigarette use among smokers in Brazil (13). Countries are encouraged to use existing global health surveys or to incorporate similar types of questions pertaining to illicit trade in their annual health surveys. Asking respondents about price paid per pack (including taxes), brand name and location where cigarettes were purchased (e.g. duty free shop, unlicensed vendor, internet) can contribute to a better understanding of the illicit tobacco trade. Advantages and disadvantages of self-report surveys Self-report surveys can be repeated over time to measure purchasing trends and progress associated with increases in cigarette taxation. Well-designed surveys can also provide generalizable estimates at the national level. Depending on the size of the sample, a self-report survey can provide comparable data across geographies that can help governments target resources. For example, findings that illicit cigarettes are more common in urban areas could lead to additional education campaigns and targeted enforcement. 258 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Limitations of self-report surveys include the possibility of bias due to the social stigma associated with participating in the illicit trade, which could lead survey respondents to underreport participation. Additionally, surveys may be unable to gauge whether individuals are associated with tax avoidance versus tax evasion. Finally, there is evidence that self-report household surveys may underrepresent smokers. Key studies for readers to refer to for additional guidance: Callaghan RC, Veldhuizen S, Ip D. Contraband cigarette consumption among adolescent daily smokers in Ontario, Canada. Tob Control. 2011;20(2):173–174. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany (NY): New York State Department of Health, Tobacco Control Program; 2006. A4.1.5 COVERT-PURCHASES SURVEYS A number of studies in high-, middle- and low-income countries use covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes in public and semi-private spaces (14-17). This method is also used by the tobacco industry in the United States to identify retailers who sell counterfeit cigarettes (18). Covert-purchases surveys do not provide estimates of the size of the illicit trade (i.e. market volume). Instead, they serve as a surveillance tool to identify where illicit cigarettes are sold and the extent to which they have infiltrated legal businesses. For example, a covert-purchases survey can examine whether illicit cigarettes are sold through legal retailers. It can also be used to measure compliance with emerging tobacco control policies that focus on, for example, product standardization or new regulations on flavours (e.g. plain packaging or bans on flavoured tobacco products). Covert-purchases surveys use trained researchers to visit a selected sample of retailers and directly purchase or inquire about the availability of illicit tobacco products. Retailers are not informed about the goals of the studies. Methods for determining the availability of illicit product vary. For example, in some studies, covert buyers do not directly inquire about illicit products. Instead, they purchase packs of tobacco products, paying full price, to determine whether retailers are selling illicit products under the guise that they are licit (14). The research team then examines the packs to determine whether they are legal. In the United States, researchers have observed that some consumers are paying full price for illicit untaxed packs smuggled from lower-tax states (14). Other research protocols directly ask retailers for illicit products (14,16-17). In Guatemala, Arevalo et al. specifically asked retailers for “imported cigarettes” (17). The ways covert buyers ask for illicit products may also vary geographically. For example, in some countries covert buyers may ask for “cheaper” packs or for illicit whites such as Jin Ling. CHAP T ER 4. PO LI T I C AL ECO N OMY 259 Advantages and disadvantages of covert-purchases surveys Covert purchasing allows researchers to directly identify sources of illicit cigarettes. It also allows them to measure and test the dynamics between buyer and seller. For example, researchers can experiment to see if repeated attempts to purchase products increase the likelihood of purchase (known as the familiarity protocol) (16). One methodological challenge associated with covert purchases is that it is difficult to create a sampling frame for illicit sources because some may be unknown (e.g. pubs or homes). The traditional approach is to make purchases in legal outlets, which may bias estimates. Another issue with this method is that it is difficult for buyers (also called raters) to purchase products if they are unfamiliar with the seller or do not fit the typical demographics of purchasers. Therefore, researchers using covert- purchases surveys must have detailed knowledge of the marketplace, including the ways individuals specifically ask for illicit tobacco products, and they must know whether they mirror the demographics of the neighbourhood. For example, in a study of South Bronx smokers, von Lampe et al. found that smokers looked for certain clues to assess whether they were being sold illicit cigarettes (19). Overall, this method can be quite costly because it requires training researchers, travelling to retailers and purchasing product. Covert-purchases surveys do not enable researchers to estimate the level of illicit trade, but they can provide information on availability of supply. Key studies for readers to refer to for additional guidance: Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Global Health. 2016;12(1):78. A4.1.6 SEIZURES OF GOODS Seizures are the result of enforcement activity carried out by local, national and international organizations that confiscate tobacco products that are illegally manu- factured, transported and sold. Seizures are meant to reduce the profits associated with illicit trade by confiscating proceeds (e.g. cash, cars or houses) and the tools of the trade (e.g. print and tobacco machinery). Seizures can occur at various points in the supply chain. Seizure data are often tallied by the responsible agencies and used to measure program effectiveness or as justification for requesting additional resources (e.g. personnel). Some of the data may be supplied to international customs organiza- tions, including the WCO (20). The quality of recordkeeping varies. For example, some agencies may maintain criminal files in databases that detail dates of seizure, brand names and laboratory testing. 260 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Seizures provide preliminary data on the scope of criminal activity and can help identify key trends to guide law enforcement agencies’ efforts. For example, seizures can identify trends on the modus operandi of smugglers, including transporta- tion methods (e.g. sea cargo versus trucks), point of entry and brand preference. Seizure statistics can also be used as a preliminary test to measure the efficacy of interventions. For example, Stoklosa and Ross used seizure data from the Canadian province of Nova Scotia to test the impact of a 2015 menthol ban. He found no statistically significant change in the number of menthol cigarettes seized before and after the ban (1). Advantages and disadvantages of seizures of goods Generally, seizure statistics can be readily obtained from law enforcement agencies through formal requests to agency gatekeepers (e.g. public information officers). Seizure data, however, generally do not provide a representative picture of illicit activity. For example, certain geographies may yield higher seizures because that is where the bulk of operations are being conducted. Police agencies may focus on certain geographies (e.g. locations near borders) rather than randomly inspecting, and their findings may be limited to those specific regions. Seizure data may also be skewed by the type of investigation procedures utilized. Large seizures may be the result of long-term investigations (i.e. wiretaps or culling confidential infor- mants), while smaller seizures may come from anti-smuggling cases that involve cross-border purchases of low quantities of cigarettes (less than 1 000) (2). Seizures can also be skewed by industry cooperation with law enforcement agencies. For example, the tobacco industry may be more likely to support law enforcement on counterfeit seizures rather than smuggling cases because counterfeiting impacts their brand integrity. A4.2 RESIDUAL METHODS Because the illicit tobacco trade is often decentralized, it can be difficult to observe directly. However, researchers are sometimes able to make inferences about its size without direct observation by comparing observed tobacco tax revenues with the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. The difference between observed and hypothetical revenues is called the residual and can be used as an indicator of the magnitude of illicit trade. Even when the residual is only an approximate measure, changes in its size may be a reliable indicator of changes in the size of the trade. When actual tobacco tax revenues are reliably observed, the main challenge for residual methods is that of producing accurate estimates of the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. CHAP T ER 4. PO LI T I C AL ECO N OMY 261 A4.2.1 GAP ANALYSIS Gap analysis is the preferred residual methodology because it is intuitive, straight- forward and relatively easy to explain to policy-makers and the general public, and it has been widely employed in government studies (4). Researchers using gap analysis compare survey-based self-reported consumption data with observed (usually administrative) data on tax-paid sales. The basic premise is that if both self-reports and observed data are accurate, any difference between reported consumption and tax-paid sales can be explained by legal imports of non-taxed cigarettes (such as duty-free sales), exports of taxed cigarettes, tax evasion or tax avoidance. The greatest research challenge in implementing gap analysis – as with most residual methods – is obtaining reliable and accurate estimates of tobacco consump- tion. In its simplest implementation, gap analysis calculates the residual as the difference (which should be a minimum of zero) between the amount of tobacco consumption reported in surveys and tax-paid sales, which are generally available from administrative sources, minus exports. This simple calculation, however, is generally flawed, since surveys of reported tobacco consumption underestimate true consumption. Underreporting of tobacco consumption may result from survey respondents’ reticence about disclosing behaviour that is viewed as unhealthy and potentially socially undesirable. Depending on the legal and cultural context, cer- tain groups (e.g. women or youth) may be more likely than others to underreport consumption. Other groups (e.g. rebellious young men and women) may accurately report or even overestimate consumption. Reuter and Majmundar measured actual consumption by total national taxed sales in the United States, where both legal untaxed imports and exports of taxed tobacco are widely believed to be very small, and found that the ratio of self-reported consumption to actual consumption was only 65% (4). After incorporating this survey underreporting into their analyses and considering the evidence from their gap analysis and the literature, Reuter and Majmundar found that the illicit market in the United States, which largely consists of avoidance or evasion of subnational state taxes, is between 8.5% and 21% of consumption (4). The higher range of the estimate is consistent with prior estimates using population-based pack observation studies (21). Researchers using gap analysis for countries or regions where legal imports of untaxed tobacco (such as duty-free products) or (legal or illegal) exports of taxed cigarettes are more significant should attempt to incorporate data about, or estimates of, these factors into their calculations. Legal untaxed imports of tobacco should be subtracted from reported consumption (after adjustment for underreporting), and exports of taxed tobacco should be subtracted from taxed sales. Obtaining data about legal untaxed imports and exports of taxed tobacco may be challenging, because these imports and exports may be the result of decentralized decisions of individual 262 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N travellers as they cross tax borders. Data on these activities will not necessarily be collected through normal administrative activities. Despite these challenges, gap analyses may still prove useful. For example, if researchers have reason to believe that misreporting of tobacco consumption and the amounts of legal untaxed imports and taxed exports are relatively stable or follow known trends (e.g. are falling) over time, gap analyses can be used to provide estimates or lower (or upper) bounds on illicit trade when several years of data on taxed sales and reported consumption are available. Data sources may be country tax administrators who have access to sales data and health departments that have access to population-level studies of reported tobacco con- sumption. In this context, multiple years of data on reported consumption and tax-paid sales can allow researchers to estimate changes in the size of the illicit trade even when it is difficult to measure the absolute level. Paraje used the 2008 Global Adult Tobacco Survey and the 2013 National Health Survey to measure reported tobacco consumption in Brazil (22-23). Advantages and disadvantages of gap analysis A major advantage of gap analysis is that when quality data are available, it is simple, easily reproduced and explainable to policy-makers and the general public. How- ever, high-quality data on reported consumption may not be available, especially in low-income countries. In many cases, gap analysis does not provide reliable information on the size of the illicit market but only on changes in the size over time (22). Additionally, some low-income countries may not have reliable estimates of tax-paid cigarette sales, and secondary data repositories of cigarette sales may not be transparent about their methodology (24). Another disadvantage of gap analysis is that it generally cannot be used to obtain separate estimates of tax avoidance and tax evasion. Biased estimates may also result if surveys of tobacco consumption are not representative of the population (25). Moreover, it is generally not possible to quantify the precision of the estimates or uncertainty associated with the estimates, because of both statistical uncertainty resulting from the use of samples to imperfectly represent populations (e.g. the share of the population that smokes) and uncertainty about key facts such as the degree to which survey respondents understate their tobacco consumption. Key studies for readers to refer to for additional guidance: Szklo A, Iglesias RM, Carvalho de Souza M, Szklo M, Maria de Almeida L. Trends in illicit cigarette use in Brazil estimated from legal sales, 2012–2016. Am J Public Health. 2018;108(2):265–269. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine and Tob Res. 2019;21(8):1079–86 CHAP T ER 4. PO LI T I C AL ECO N OMY 263 A4.2.2 ECONOMETRIC MODELLING There is a long tradition of using data to estimate parameters of demand functions that relate the quantity of goods consumed to the prices faced by consumers, their incomes and other variables. Because of the addictive nature of tobacco – and because of important public health and public policy concerns relating to tobacco use – economists have paid particular attention to the estimation of cigarette demand functions (26). As the literature on this topic developed, it became apparent that taxed tobacco sales would be a biased indicator of tobacco consumption if some consumers obtained their tobacco in illicit markets. Similarly, the price of cigarettes in the legal market might overestimate the price paid by consumers if some sales were not tax-paid. While economists generally cannot observe sales in the illicit market, they have been able to develop models that predict conditions under which consumers avoid tobacco taxes. They reason that the relative size of illicit tobacco markets depends primarily on two variables: the relative price of taxed and untaxed consumption and the ease of obtaining lower-cost (untaxed) tobacco. Other variables, includ- ing the social stigma from evading tax laws and the perceived relative quality of illicit tobacco, could also influence the demand for it. While illicit trade cannot be directly observed, it can be estimated from the difference between tax-paid sales and predicted consumption. Tax-paid sales can be less than predicted consump- tion when retailers or consumers evade taxes. They can be greater if some tax-paid cigarettes are bought within the jurisdiction and then consumed in areas where after-tax prices are higher. Econometric modelling estimates of illicit trade must be tailored to the situation in the country that is being studied, and therefore the data requirements may differ substantially from case to case. Researchers using this method should be familiar with the literature and should also understand the conditions in the areas they are researching. They must always include some measure of tobacco consumption or sales and some measure of the price of tobacco in the home country, as well as other variables (e.g. income) that are known to affect the demand for tobacco. It is also generally necessary to include variables that measure the availability and relative price of illicit tobacco, which can often be measured by comparing tobacco taxes in the home country with those in areas that are the source of illicit tobacco. Advantages and disadvantages of econometric modelling The major advantage of econometric modelling is that it is consistent with a long tradition of economic theory and practice, and the quality of the modelling techniques and empirical estimates can therefore be evaluated against widely accepted criteria. Empirical analyses provide estimates of price elasticities, income elasticities and price elasticities of tax avoidance. A substantial literature base makes it possible to 264 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N rigorously quantify uncertainty about the estimates and to test their robustness to various assumptions made in the modelling process. Estimates can be compared to other estimates available in the literature, and the results of these analyses can be used to simulate the impacts of policy changes (including tax and/or enforcement increases) on both consumption and tax avoidance. Because this methodology provides direct estimates of the uncertainty of the results, researchers can specify their level of confidence in the findings. A shortcoming of econometric modelling is that it requires high-quality data on a variety of important variables over a period of time, as well as advanced econo- metric modelling expertise. Also, because results from the econometric models are based on statistical inference and economic theory rather than direct observation (e.g. the proportion of packs without tax stamps), it can be difficult to explain to policy-makers and the general public. Key studies for readers to refer to for additional guidance: Becker GS, Grossman M, Murphy KM, (1994). An empirical analysis of cigarette addiction. Amer Econ Review. 1994;84(3):396–418. Schafferer C, Yeh CY, Chen SH, Lee JM, Hsieh CJ. A simulation impact evaluation of a cigarette excise tax increase on licit and illicit cigarette consumption and tax revenue in 36 European countries. Public Health. 2018;162:48–57. A4.2.3 EXPERT OPINION (KEY-INFORMANT SURVEYS AND INTERVIEWS) Insight on illicit trade dynamics can come from experts in the field, including researchers (e.g. in economics, criminal justice and public health), taxation depart- ments, enforcement agencies, product manufacturers, wholesalers and retailers. Other key informants include journalists and academics who have secured confi- dential informants. Experts can provide novel information about emerging trends (e.g. new smuggling routes). In some cases, researchers can obtain interviews with incarcerated or active offenders (27-29). For example, researchers studying cigarette smuggling in eastern Africa conducted interviews with more than 150 Ugandan tobacco smugglers (29). Experts can be queried through surveys or semi-structured interviews. When sampling frames are available (e.g. directories of tax department employees), surveys are more expedient than interviews. However, when experts are hard to find, non- random sampling strategies coupled with interviews are recommended. Identifying experts may require recruiting a gatekeeper who is tasked with helping researchers find additional experts; or purposive sampling, where individuals are identified based on set criteria (e.g. they are taxation experts employed by local governments) (30). CHAP T ER 4. PO LI T I C AL ECO N OMY 265 Advantages and disadvantages of key-informant surveys and interviews Informant interviews can be a useful starting point for identifying trends in the marketplace (e.g. venues where illicit cigarettes are sold or modes of entry). One of the disadvantages of relying on informants is that the information solicited from them may not be generalizable. Expert knowledge may be outdated or limited by the informants’ experience. Furthermore, the opinions of experts are subjective and may be biased by the experts’ employment status and the sampling methods used. For example, persons working in law enforcement may overestimate the extent of bootlegging in order to secure additional funding for future operations. Similarly, manufacturers looking to defeat taxes may overestimate the illicit trade to illustrate the links between taxation and illicit behaviour. Alternatively, tobacco control advocates may underestimate illegal market measures in order to support the argument that taxes do not increase illicit trade. Key studies for readers to refer to for additional guidance: Joossens L, Raw M. Cigarette smuggling in Europe: who really benefits? Tob Control. 1998;7:66–71. doi:10.1136/tc.7.1.66 PMID: 9706757. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20(3):226–232. A4.3 MIXED AND MULTIMETHOD STUDIES Given the shortcomings of the aforementioned methods for assessing the nature and size of the illicit tobacco trade, governments may want to validate their findings by using mixed or multiple methodologies. Mixed methods use two methodological paradigms, qualitative and quantitative, as tools for exploration and explanation. For example, mixed method studies can use littered-pack surveys to measure the size of the market along with self-report surveys of smokers to understand patterns of purchasing, including sources, frequency and social norms. For example, Stoklosa and Ross estimated the share of the illicit market in Poland using a population- based self-report survey and a littered-pack survey (1). Using both types of survey simultaneously enables governments to assess their validity in estimating the size of the illicit market. Alternatively, governments can employ multimethod research, i.e. the use of multiple methods that are similar in tradition (e.g. focus groups and semi-structured interviews) (31). Saenz de Miera et al. used face-to-face interviews (households), litter collection and observation of single-stick sellers, which enabled them not only to cross-validate the two major methodologies, but also to see if the brand of the single stick was a good measure of licit versus illicit trade (33). 266 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Advantages and disadvantages of mixed and multimethod studies Mixed and multimethods studies enable researchers to check the validity of their findings. Multiple methods are preferred in contexts where illicit trade estimates are politicized. For example, low estimates may be challenged by the tobacco indus- try, while high estimates may be challenged by tobacco control researchers and/or proponents. Mixed and multiple methods (e.g. littered-pack surveys and informant interviews) can enable governments to understand the situational context in which the illicit trade operates, including the actors involved and venues of sale. One disadvantages of using mixed and multiple methods is cost. Governments that are constrained by tight budgets may choose to use a single method that provides the most accurate information. However, given the issues faced by each method, this may not be feasible – each method has limitations. Instead, governments can pair a high-cost method with a lower-cost method (e.g. pairing interviews with empty pack surveys, or law enforcement seizure data with face-to-face consumer surveys). Key study for readers to refer to for additional guidance: Zaloshnja E, Ross H, Levy DT. The impact of tobacco control policies in Albania. Tob Control. 2010;19:463–468. CHAP T ER 4. PO LI T I C AL ECO N OMY 267 REFERENCES 1. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 2. Joossens L, Lugo A, La Vecchia C, Gilmore AB, Clancy L, Gallus S. Illicit cigarettes and hand-rolled tobacco in 18 European countries: a cross-sectional survey. Tob Control. 2014;23:e17–e23 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC3812425/pdf/nihms491463.pdf, accessed 2 February 2021). 3. Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–7. 4. Reuter P, Majmundar M. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Washington (DC): National Academies Press; 2015. 5. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 6. Barkans M, Lawrance KA. Contraband tobacco on post-secondary campuses in Ontario, Canada: analysis of discarded cigarette butts. BMC Pub Health. 2013;13(1):335 (https://bmcpublichealth. biomedcentral.com/articles/10.1186/1471-2458-13-335, accessed 2 February 2021). 7. John RM, Ross H. Illicit cigarette sales in Indian cities: findings from a retail survey. Tob Control. 2018;27(6), 684–688. 8. Davis KC, Grimshaw V, Merriman D, Farrelly MC, Chernick H, Coady MH, et al. Cigarette trafficking in five northeastern US cities. Tob Control. 2014;23(e1):e62–e68. 9. Callaghan RC, Veldhuizen S, Leatherdale S, Murnaghan D, Manske S. Use of contraband cigarettes among adolescent daily smokers in Canada. CMAJ. 2009;181(6-7):384–6 (https://www.cmaj.ca/content/ cmaj/181/6-7/384.full.pdf, accessed 2 February 2021). 10. Guindon GE, Burkhalter R, Brown KS. Levels and trends in cigarette contraband in Canada. Tob Control. 2017;26(5):518–25 (https://www.researchgate.net/publication/307890937_Levels_and_trends_in_ cigarette_contraband_in_Canada, accessed 2 February 2021). 11. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany: New York State Department of Health; 2006 (https://www. health.ny.gov/prevention/tobacco_control/docs/cigarette_purchasing_patterns.pdf, accessed 2 February 2021). 12. International Tobacco Control Evaluation Project (2018). Surveys. Waterloo: University of Waterloo; 2018 (http://www.itcproject.org/surveys, accessed 12 October 2020). 13. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 14. Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. 15. Brown J, Welding K, Cohen JE, Cherukupalli R, Washington C, Ferguson J, et al. An analysis of purchase price of legal and illicit cigarettes in urban retail environments in 14 low-and middle-income countries. Addiction. 2017;112:1854–60 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5600117/pdf/ADD- 112-1854.pdf, accessed 2 February 2021). 16. Scollo M, Bayly M, Wakefield M. Availability of illicit tobacco in small retail outlets before and after the implementation of Australian plain packaging legislation. Tob Control. 2015;24(e1):e45–e51. 17. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Globalization and Health. 2016;12(1):78 (https://globalizationandhealth. biomedcentral.com/articles/10.1186/s12992-016-0219-z, accessed 2 February 2021). 18. Phillip Morris USA Inc. v. Shalabi. United States, District Court, C.D. California; 2004 (https://www. casemine.com/judgement/us/5914b6cbadd7b0493477b3da, accessed 2 February 2021). 19. von Lampe K, Kurti M, Johnson J, Rengifo AF. ‘I wouldn’t take my chances on the street’ navigating illegal cigarette purchases in the South Bronx. J Res Crime Delinq. 2016;53(5):654–80 (https://www. 268 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N researchgate.net/publication/298515449_I_Wouldn’t_Take_My_Chances_on_the_Street_Navigating_ Illegal_Cigarette_Purchases_in_the_South_Bronx, accessed 2 February 2021). 20. Customs and Tobacco Report 2009. Brussels: World Customs Organization; 2009. (http://www.wcoomd. org/en/media/newsroom/2010/june/~/media/83967DFEB9F74D388924A4C61F279DC4.ashx, accessed 12 October 2020). 21. Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ, et al. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61–i66 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3984758/pdf/nihms567943.pdf, accessed 2 February 2021). 22. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine Tob Res. 2019;21(8):1079–86. 23. National Survey of Health. Instituto Brasileiro de Geografia e Estatística; 2013 (in Portuguese) (https:// www.ibge.gov.br/en/statistics/social/health/16840-national-survey-of-health.html?=&t=downloads, accessed 27 November 2020). 24. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 25. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 26. Chaloupka FJ, Warner KE. The economics of smoking. In: Arrow KJ and Intriligator MD, editors. Handbook of Health Economics. Amsterdam: Elsiver; 2000. pp1539–1627. 27. Antonopoulos GA. Cigarette smugglers: a note on four ‘unusual suspects’. Glob. Crime. 2007;8(4):393–8 (https://www.researchgate.net/publication/248955367_Cigarette_Smugglers_A_Note_on_ Four_’Unusual_Suspects’, accessed 2 February 2021). 28. Antonopoulos GA. The Greek connection(s): the social organization of the cigarette-smuggling business in Greece. Eur J Criminol. 2008;5(3):263–88 (https://www.researchgate.net/publication/249752218_ The_Greek_ConnectionsThe_Social_Organization_of_the_Cigarette-Smuggling_Business_in_Greece, accessed 2 February 2021). 29. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20:226–32 (https://www.researchgate.net/publication/49809289_ Blood_cigarettes_Cigarette_smuggling_and_war_economies_in_central_and_eastern_Africa, accessed 2 February 2021). 30. Babbie, ER. The basics of social research. Boston: Cengage Learning; 2013. 31. Tashakkori A, Teddlie C., editors. Sage handbook of mixed methods in social & behavioral research. New York: Sage; 2010. 32. Saenz de Miera Juarez B, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 269 ANNEX 4.2 HOW ARE THE TOBACCO TAX REVENUES EARMARKED? The introduction of earmarking of tobacco tax revenue is almost always combined with an increase in excise taxes (or a new surcharge) rather than reallocation of existing revenues (1). Table A4.1 provides examples of the different approaches used by several countries to earmark tobacco tax revenues. Table A4.1 Approaches used to earmark tobacco tax revenues FUNDING SOURCES/ TYPE OF TAX EXAMPLES OF TAX BASE AND RATES As part of the excise system (tobacco, alcohol) Specific Republic of Korea: 841 won (US$ 0.75) per pack or 29% of the specific excise rate Costa Rica: 467.8 cólones (US$ 0.83) per pack or 100% of the specific excise rate Congo: 20 CFA francs (US$ 0.036) per pack or 50% of the specific excise rate Ad valorem Colombia: 10% of retail price (equivalent to 100% of the ad valorem rate) New levy (surcharge on the existing excise or completely new levy) Specific Egypt: additional 0.75 Egyptian pounds (US$ 0.042) per pack Ad valorem Thailand: surcharge of 2% over the excise tax base Botswana: new tobacco levy of 30% of the cost of production or CIF Percentage of excise revenue Cook Islands: 50% of revenues from the excise tax on tobacco Guatemala: 100% of revenues from the excise tax on tobacco Note: Conversions of amounts from the local currency were made using the official exchange rates from the IMF as of 31 July 2018 (date of the data collection). Source: (2). REFERENCES 1. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 2. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 270 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 271 CHAPTER 5. Best practices in tobacco tax policy and administration TAX POLICY Use excise tax increases to achieve the public health goal of reducing the death and diseases caused by tobacco use Extensive research has clearly demonstrated the effectiveness of higher tobacco product taxes and prices in reducing tobacco use and its harmful consequences, particularly among the poor and the young. In fact, tobacco excise tax increases are the single most effective and cost-effective policy for reducing tobacco use. Excise taxes are the most significant taxes applied on tobacco products because of their ability to raise both absolute and relative prices. Tobacco excise tax increases also generate sizeable new revenues that will be sustained in the short to medium term. In the long term, continued increases in tobacco taxes – coupled with implementa- tion of other evidence-based tobacco control policies and programmes – will lead to even larger reductions in tobacco use and its consequences. Include significant tobacco excise tax increases as part of a comprehensive strategy to reduce tobacco use Governments should adopt a comprehensive tobacco control strategy that includes objectives for reducing adult tobacco use and preventing youth tobacco use. Rais- ing excise taxes significantly is the most effective, as well as the most cost-effective, measure for reducing consumption. When combined with other demand reduction interventions, the impact of tax increases on tobacco use is even stronger. Such interventions include comprehensive smoke-free policies in all public spaces, total bans on tobacco advertising, promotion and sponsorship by tobacco companies, large graphic health warnings about the consequences of tobacco use, plain packaging, broad efforts to help current users quit and mass media public education campaigns. Implementation of a comprehensive strategy to reduce tobacco use leads to greater reductions in the harmful consequences of tobacco use, builds public and political support for higher taxes and maximizes the effectiveness of tax increases in achieving public health objectives. 272 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Involve the competent authority from the start when considering the revision of a tax policy Competent authorities such as tax administrations and customs authorities are key partners in the effective implementation of a tax policy. Policy-makers need to ensure that those competent authorities are consulted and involved in the tax policy revision process so that their concerns about the impacts of policy change on enforcement can be taken into account from the beginning. This can also help identify and address possible loopholes early on in the enforcement process. Coordination among relevant bodies, including close cooperation and sharing of information, will optimize enforcement of tax policy and tax collection. To streamline the process of cooperation and exchanges of information, a basis in law needs to be established. Additionally, the involvement of tax administration authorities in the entirety of the tax revision policy process is important to ensure effective implementation of the policy. Promote greater policy coherence across sectors such as agriculture, industry, trade, finance and labour Greater multisectoral integration and policy coherence is needed at the country level to achieve effective health improvements. In particular, it is important to ensure that public policies and interventions in non-health sectors (e.g. agriculture, industry, trade, finance and labour) do not act against the intended public health impact of tobacco control and taxation (such interventions include providing subsidies to tobacco growing or manufacturing). TAX DESIGN Tax structure matters and simpler is better Complex tax structures are difficult to administer, create opportunities for tax avoid- ance and evasion and are less effective than simpler structures in achieving public health and revenue goals. Simplifying the structure of tobacco excise taxes will facilitate tax administration, reduce tax avoidance and evasion, enhance revenues and have a greater impact on tobacco use by reducing incentives to substitute among tobacco products or brands in response to tax increases. Countries with multiple tiers of tobacco tax rates based on product characteristics (e.g. price level, length, weight, type of tobacco) should reduce and eventually eliminate these differential tax rates. An appropriate transition strategy is to reduce the variations in tax rates over time with the aim of implementing a uniform tax (i.e. a single rate applies whether excise is ad valorem or specific) on a given tobacco product. Applying a uniform tax to all brands of a given tobacco product also sends a clear message that they are equally harmful. CHAP T ER 5. B E S T PR AC T I CE S 273 Rely more on specific tobacco excises to drive price increases Greater reliance on specific excise taxes maximizes the impact of tobacco taxes on public health by reducing the gap in prices between premium and low-priced alternatives and limiting opportunities for users to switch down in response to tax increases. For countries that currently rely on an ad valorem tax, an appropriate first step would be to shift to a mixed system by adding a sizeable specific component or introducing a high minimum specific excise tax (an excise tax floor). For countries that rely on a mix of ad valorem and specific taxes, the specific tax component should be increased regularly so that it accounts for a greater share of the total excise tax. Increase tobacco taxes significantly to reduce the affordability of tobacco products To maximize the public health impact of higher tobacco taxes while at the same time generating higher revenues, governments should significantly raise taxes to increase prices and reduce the affordability of tobacco products. In many LMICs, tobacco use increases with incomes, and since incomes rise faster than tobacco product prices, these products are becoming more affordable. To reduce afford- ability, tax increases need to result in real price increases that are higher than the increases in real incomes. Where revenue increases are a goal, rely on regular excise tax increases If governments want to increase tobacco revenues, they must increase excise taxes regularly. From the tax revenue perspective, the important determinant is the tax base elasticity, which has three key components: the price elasticity of demand of tobacco, the share of the tax in the retail price and the degree of pass-through of the excise tax rate increase on to retail price. Tax increases will increase revenues at least in the short to medium term, because demand is price inelastic, tax levels are generally low as a proportion of retail prices and the pass-through of tax increases on to retail prices is unlikely to be higher than the tax increase itself (i.e. there is no overshifting). In addition, increasing tax rates is the only policy measure that can reverse reduced revenues in a declining market that has strong tobacco control policies. Automatically adjust specific tobacco taxes for inflation and income growth Unless specific tobacco taxes are regularly adjusted, their real value will fall over time as general price levels increase. When this happens, their effectiveness in reducing tobacco use will be diminished. Governments should establish a mechanism for automatically adjusting specific taxes to keep pace with inflation. Recently, some governments have begun to extend this indexation to include income growth as well, further ensuring that tobacco does not become more affordable over time. 274 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Pricing regulations cannot be considered an alternative to excise tax. However, in some specific contexts, pricing regulations could be used in conjunction with excise taxes to help ensure the effective implementation of tax increases In certain contexts where increasing taxes is challenging or the tax structure is weak, non-tax policies such as pricing regulation (specifically, minimum mark-ups and price floors/minimum prices) may be seen as a second-best alternative to ensure a high price level and dissuade consumption of tobacco products. These policies, however, do not necessarily lead to the desired price level, nor do they protect consumers and government from industry manipulation. However, in the context of powerful multinationals that sell brands across all market segments and could easily undershift a tax increase to cheaper brands – or where price promotions cannot be banned – minimum price policies may help increase the effectiveness of tax increases, especially if the minimum prices are increased regularly. Implement nontax policies affecting price levels, such as banning promotional discounts for tobacco products and the sale of single sticks of cigarettes The banning of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Do not allow concerns about the inflationary impact of higher tobacco taxes to deter tax increases Given that wages or some government spending may be tied to a price index, govern- ments can reduce concerns about the inflationary impact of a tobacco tax increase by using a price index that excludes tobacco products. TAX PARITY Tax all tobacco products in a comparable way Increasing excise taxes on some tobacco products but not on others results in changes in the relative prices of these other products. This induces substitution towards relatively less-expensive products – for example, from expensive manufactured cigarettes to other, cheaper tobacco products such as RYO tobacco, bidis, cheroots or chewing tobacco. As a result, the overall reduction in tobacco use is smaller than it would have been had all taxes increased by comparable amounts. Comparable increases in the taxes on all tobacco products maximize the public health impact of tobacco tax increases by minimizing opportunities for substitution. Moreover, increases in taxes on all tobacco products will generate larger increases in revenues. CHAP T ER 5. B E S T PR AC T I CE S 275 Strictly regulate new and emerging tobacco and nicotine products where they are not banned and impose an excise tax In recent years, the world has been experiencing the rise of new and emerging tobacco and nicotine products including ENDS, ENNDS and HTPs. The tobacco industry claims these new products are safer than traditional tobacco products, but the evidence so far suggests that they could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of newer products – as well as initiation, smoking cessation and switching behaviour among different socioeconomic groups – are not yet clear. Best practices for taxing new and emerging tobacco and nicotine products, based on current knowledge, are that: 1. HTPs should be taxed at the same level as cigarettes and, in terms of structure, through a specific excise per unit regardless of tobacco content. HTPs contain tobacco and should be treated as a tobacco product. 2. ENDS/ENNDS products should be taxed in a manner that discourages up- take by youth and non-users. Nicotine- and non-nicotine-delivery systems containing e-liquids should be taxed equally. 3. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While these newer products create additional challenges for tobacco control, it is important to remember that cigarettes remain by far the predominant tobacco product and that raising taxes and prices on cigarettes – and thereby reducing their use – should remain the top priority. MONITORING AND EVALUATION Know your market Know your market well. The type of tax structure you choose and the impacts it will have on consumption and tax revenue are shaped by the particular dynamics of your market. Understanding the nature and degree of competition in your market is vital to selecting the appropriate type of tax structure and policies to achieve your public health and revenue objectives. This knowledge will also facilitate more accurate estimates of the impacts of a tax increase, as well as better anticipation of industry responses. Assess the impact of your policies to design and implement the most effective tobacco excise tax policies Monitoring and evaluation are essential for effective tobacco taxation, and they 276 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N should be built into the initial design – or redesign – of tobacco tax policies. A number of tools exist to help policy-makers pre-emptively assess the effects of a proposed tobacco control policy on consumption, smoking prevalence and lives saved. The WHO TaXSiM uses target simulations to assist governments in predict- ing how specific tax changes will impact consumer prices, consumption and tax revenue in their market. Adopt indicators that help you measure improvements in tax policy and its impact Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the improvement of their policies and determine if those policies have an impact on tobacco use over time. The tax share of the retail price for a particular tobacco product is a key indicator that should be used in conjunction with an af- fordability indicator. A recommended target for countries to aspire to is to have an excise tax that represents at least 70% of the retail price of tobacco products. Another useful indicator to assess the performance of the tax policy overall is the use of a tax scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). TAX ADMINISTRATION Implement best practice approaches in general tax administration to make tobacco tax administration more effective and efficient Best practice approaches include (1) defining clearly the roles and responsibilities of competent authorities, (2) ensuring effective coordination among relevant bodies at the national and international levels and (3) undertaking evaluation of performance and accountability against pre-agreed indicators to identify points for improvement. Ensure compliance and accuracy of information on the tax compliance cycle To achieve this, implement the following actions: • Require licences for manufacturing, importing, exporting, retailing, growing, transporting, wholesaling, brokering, warehousing and distributing tobacco products. This will help secure the supply chain while obtaining valuable infor- mation, e.g. through access to companies’ accounting and inventory systems. • Make sure all persons and entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment keep complete and accu- rate records of all relevant transactions and details of materials used in the production of tobacco products. CHAP T ER 5. B E S T PR AC T I CE S 277 • Ensure that tax declarations collect as much information as possible on the taxpayer. • Collect taxes close to the point of production and import to limit the number of taxpayers a competent authority needs to manage. • Maintain a system of authorization for warehousing to carry out controls in production and storage facilities to ensure that taxes are paid. • Use electronic methods, through the best available IT, for declarations and collection of taxes. This allows for cross-check of information provided in dec- larations with information from other government agencies and third parties. Ensure control and enforcement on the supply chain To achieve this, implement the following actions: • Include control and enforcement as a fundamental pillar in the strategic plan of the tax administration overall. • Use a risk-based approach by choosing defined targets for enforcement and control, such as those who have a higher probability of noncompliance. • In the licensing process, ensure that purchases from unlicensed suppliers or sales to unlicensed purchasers are not allowed. Ensure also that the validity of licences is limited in time and require renewals or reapplication to maintain a high level of control. • Use tax stamps with strong security features to reduce the risk of stamp counterfeiting. These markings facilitate the collection of excise taxes, audits and enforcement actions. • Implement a tracking and tracing system for tobacco products. A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, and monitoring and control- ling the movement of tobacco products and their legal status. • Implement anti-forestalling measures so that forestalling does not delay a tax increase and its intended effect on revenues and consumer behaviour. • Control import and export of tobacco products and manufacturing equipment by allowing only duly licensed natural persons or legal entities to conduct such activities. • Strengthen border control, e.g. by utilizing non-invasive tools such as X-ray scanners and dogs to detect tobacco products. • Limit or tightly control and, ideally, ban activities related to production and trade of tobacco products in tax-free zones to avoid opportunities for tax evasion. • Prohibit intermingling of tobacco products with non-tobacco products in a single container or any other similar transportation unit when removed from tax-free zones. 278 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Prohibit the sale to or import by international travellers of tax-free or duty- free tobacco products. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and adversely affect government revenues by creating a loophole in the tax structure. Clearly define procedures to follow after detecting illicit trade of tobacco • Take immediate action to seize and destroy smuggled and/or illicit tobacco and collect due taxes. • Ensure certain, swift and severe sanctions for those caught engaging in illicit trade in tobacco products, such as penalties, fines and withdrawal of licences. It can also be effective to consider illicit trade in tobacco products by law as a source of money-laundering. Become a Party to and/or implement the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products The WHO FCTC Protocol provides a blueprint of best practices and policies for dealing with illicit trade and should be part of any strategy to fight it. Implement, to the extent possible, the same rules and regulations for tax administration and enforcement for all tobacco products, as well as new and emerging nicotine and tobacco products Implement broad policies for ensuring a good tax system that will trickle down to good tax administration of tobacco products by: • ensuring proper resourcing of competent authorities; • having strict rules and regulations to detect corruption and to punish both personnel and taxpayers who are engaged in corrupt practices; and • ensuring a strong judicial system that is independent in fact and in perception, where disputes are solved quickly. The appeal process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially for illicit trade. POLITICAL ECONOMY Beyond the technical soundness of best practices in tax policy and administration, a critical factor in advancing tobacco taxes is the ability to get the political buy-in of the highest instances in the government. One key strategy is to address concerns around the political economy of tobacco taxation, which are often exploited by the tobacco industry to block major reforms. CHAP T ER 5. B E S T PR AC T I CE S 279 SCARE tactics The tobacco industry uses SCARE tactics to dissuade governments from implement- ing tobacco tax increases. These include smuggling and illicit trade (S), court and legal challenges (C), anti-poor rhetoric (A), revenue reduction (R) and employment impact (E). Best practices for countering these tactics are described below. S: Smuggling and illicit trade Do not allow concerns over the impact of increasing excise taxes on illicit trade in tobacco affect your decision to increase them. Rely on your own estimates of the level and nature of illicit trade and not on the industry’s estimates. Illicit trade in tobacco products continues to be a major concern for tax administrators because of the difficulties associated with accurate and independent measurement of it, as well as with its elimination. Industry figures provide a distorted understanding of the extent of the problem, along with a monocausal explanation of the link between illicit trade and tobacco taxation. It is therefore recommended that governments (1) assess independently and with the best statistical practices the size of the illicit trade to assess the scope of the problem; (2) address directly the country-specific institutional and/or governance challenges, including multilateral coordination, and improve tax and customs administrations practices; and (3) implement best practices to fight illicit trade, contained in the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products. Ideally, accede to the Protocol if not yet a Party. C: Court and legal challenges Do not let tobacco industry threats of court and legal challenges to tax increases or reforms prevent you from improving your tax policy. Closely follow legal requirements for design, procedure and consultation to strengthen your legal position and minimize the possibility that any challenge will be raised. Health-protective and non-discriminatory tobacco excise taxes are legally defensible, and industry threats will usually be baseless. Your legal position can be strengthened, however, by exercising care with a tax measure’s procedure, design and consultation: (1) determine the standard of consultation required under domestic law and any applicable international obligations; (2) distance the tobacco industry from the policy-making process to the extent that this is permissible; (3) avoid unnecessary and unjustified discrimination towards foreign tobacco products or investors in the design, implementation or enforcement of a tax measure; and (4) do not offer investment incentives in the form of inducements or contractual undertakings, as these may be binding in and of themselves or grounds for a challenge under an international investment agreement. 280 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A: Anti-poor rhetoric Do not allow concerns about the regressivity of higher tobacco taxes prevent tobacco tax increases. In recent years, there has been an overwhelming increase in the evidence concern- ing the positive distributional impact of tobacco taxes and tax increases. Indeed, tobacco taxation and tax increases are actually a progressive or pro-poor policy once these wider considerations are properly accounted for. In its effort to lobby against tax increases, the tobacco industry often claims that tobacco taxation will hurt the poor. This argument is based on the concept of regressivity in relation to taxation. Conceptually, a tax is regressive if it means lower-income people must pay a relatively greater proportion of their household income to meet the tax liability than wealthy people. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use that are disproportionately experienced by lower socioeconomic groups. Second, higher tobacco taxes and prices can induce behavioural change in the population, as reflected in the price elasticity of demand, which means that lower-income smokers will curtail their smoking the most and thus will benefit disproportionately in terms of health gains from reduced tobacco consumption and use. In fact, these broader considerations make tobacco taxation a progressive, rather than regressive, public health intervention. R: Revenue reduction Do not let fears of potential revenue reductions prevent you from increasing excise taxes on tobacco products. Tax increases, even in countries with already high taxes, bring in additional revenue. Arguments by tobacco control opponents that tax increases will not result in increases in revenue are unfounded. The relatively price inelastic nature of cigarette demand, combined with the low tax share and no overshifting of the tax, means that for most, if not all, countries, increases in revenues will accompany increases in taxes. If tax increases are carefully designed and tax administration is functional, it is extremely unlikely that tax increases will lead to revenue decreases. E: Employment impact Do not allow concerns about employment impact to prevent tobacco tax increases. The tobacco industry often seeks to frame tobacco taxes as an economic issue rather than a public health issue. Particular emphasis is placed on the alleged threat tax increases pose to employment in tobacco farming and manufacturing, as well as related industries. This so-called choice between health and jobs, however, is largely based on exaggeration. The tobacco industry exaggerates the importance of tobacco CHAP T ER 5. B E S T PR AC T I CE S 281 employment relative to total national employment and overstates the impact that domestic demand reduction from local taxes will have on tobacco farmers serving a global market. The argument used by the industry also ignores the fact that expendi- tures on tobacco do not disappear but rather are redistributed to other consumption that can produce a similar or higher number of jobs. Case studies demonstrate the possibility and methods for governments to support farmers in transitioning to other crops that provide similar and often better returns with greater sustainability. Earmarking Consider earmarking tobacco tax revenues for health-focused programmes, especially if it helps advance tobacco control efforts and, more specifically, efforts to implement large tobacco tax increases and tax reforms. This could have the additional benefit of funding health programmes where they are poorly funded or not prioritized. From a tobacco control perspective, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. Earmarking is a tool to improve the political economy of tobacco taxation; it is only a secondary issue, after the primary goal of reducing demand for tobacco. Evidence shows that public support for higher tobacco taxes is greater when at least some of the increased revenues are explicitly used to support health-focused programmes. Current evidence shows that the amounts effectively earmarked for health have been relatively small and therefore unlikely to introduce rigidity in government budgets. At the same time, in some countries, those funds have helped to imple- ment much needed underresourced health programmes. The payoffs will be seen in the future as fewer people fall ill and need less medical care for tobacco-related illnesses. Earmarking tobacco tax revenues to fund high-burden/low-priority health programmes could pave the way for raising awareness about the importance of such programmes and their effectiveness, thereby convincing governments to redefine their priorities and commit to including the programmes in their regular budget. 282 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) Focus on tobacco taxation’s purposes Tobacco tax policy should aim not only to increase revenues but also to decrease consumption and improve health. To both raise revenue and reduce consumption, you need to (1) simplify tobacco tax structures, (2) significantly increase rates to impact price levels, and (3) regularly adjust rates to at least account for inflation and income growth. Analyse your tax structure and identify its weaknesses You need to analyse and identify the problems of your current tax structure to know which steps to consider next. Which tax structure do you have: specific, ad valorem, mixed, or no excise? Identify the improvements to be made to the existing tax policy/structure Your present tax structure and tax situation will provide you with the steps you would ideally take next to achieve the aims in Step 1. Specific: 1. Ensure that the tax automatically adjusts for inflation and income growth effects. 2. Ensure that all price promotions are banned. Ad Valorem: 1. Ensure that the tax base of the ad valorem is retail price. 2. Introduce a high specific excise component (and a minimum specific excise). 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. Mixed: 1. Ensure that the tax base of the ad valorem component is retail price. 2. Ensure that you are using a high specific excise component and a minimum specific excise. 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. No Excise: 1. Introduce a high specific excise. 2. Ensure that the rate automatically adjusts for inflation and income growth effects. 3. Ensure that all price promotions are banned. Assess tobacco taxation’s political economy Reform must begin with an assessment of tobacco taxation’s political economy: (1) learn from past successes and failures – what went wrong, what went right, what you can do differently this time; (2) assess the reform’s strengths and weaknesses, likely opportunities and risks; (3) determine who the main supporters and opponents of reform inside and outside of government have been and may be, based on past reforms and current situation; and (4) anticipate argu- ments that will be used against the reform (refer to SCARE tactics). STEP 1 STEP 2 STEP 3 STEP 4 Prepare a plan for realizing the reform Focusing on the overall aims identified in Step 1, the steps for achiev- ing them identified in Steps 2 and 3 and the political economy around this reform as identified in Step 4, prepare your plan: 1. Be clear on the non-negotiable objectives for the reform and the trade-offs you are prepared to make to realize them. 2. Develop a plan to approach potential allies and win them over to the reform efforts. 3. Develop the counterarguments that will be needed in response to the SCARE arguments identified earlier. 4. Prepare the evidence you will need ahead of time. To do this, get support from academics and relevant intergovernmental agencies. Mobilize a coalition for reform 1. Formulate a strategic communications plan: aim for political support both at the highest levels and among the public (framing tobacco taxation as a health issue has helped win political support in many countries). 2. Identify champions in government: ensure that finance and health officials are on the same page; involve implementing departments, such as enforcement agencies, from the start. 3. Mobilize allies from academia, civil society and the private sector to counter the anticipated pushback from the tobacco industry, its proxies and its allies. Monitor and evaluate To make the most well-informed policy decisions, a reform effort should be monitored to assess its overall impact and its effect on key indicators; this will help identify issues to be fixed while also creating a strong evidence base for further reform efforts. Get and analyse the relevant data to better understand the market situation and its dynamics: 1. Monitor the market and its evolution (e.g. retail prices, duty-paid sales, market shares). 2. Get regular estimates of price elasticity (including cross-price elasticity), income elasticity and tax base elasticity to evaluate any changes in tobacco demand. Use relevant tools to assess the impact of the tax policy on consumption and revenue: 1. Use specific tools on the impact of excise on price, consumption and revenue (e.g. the WHO TaXSiM). 2. Use global tools to assess the tax increase’s impact on prevalence (e.g. the WHO ISPT). Monitor key indicators closely to assess improvements over time: 1. Tax as a percentage of retail price. 2. Change in affordability of tobacco products over time. 3. Change in the tobacco tax scorecard, which combines a mix of best practices in tax policy. 4. Change in sales, prevalence and illicit trade in tobacco products. 5. Improvements in MPOWER package achievement. STEP 5 STEP 6 STEP 7

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION WHO technical manual on tobacco tax policy and administration ISBN 978-92-4-001918-8 (electronic version) ISBN 978-92-4-001919-5 (print version) © World Health Organization 2021 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/ rules/). Suggested citation. WHO technical manual on tobacco tax policy and administration. Geneva: World Health Organization; 2021. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Document design by Ana Sabino. iii Contents Foreword v Acknowledgements vii Acronyms viii Executive summary xi CHAPTER 1. Why this manual? 1 CHAPTER 2. Tobacco excise tax policy 11 2.1 Global overview of tobacco tax practices 11 2.2 Designing excise tax policy 18 2.3 Domestic and regional policy integration 55 2.4 New and emerging nicotine and tobacco products 59 2.5 Conclusions 75 ANNEX 2.1 Countries that apply different types of excise tax structures 89 ANNEX 2.2 Analytics of the tax base elasticity 90 ANNEX 2.3 Elements of the devices that make up ENDS/ENNDS products 91 CHAPTER 3. Tobacco tax administration 93 3.1 Introduction 93 3.2 Institutional arrangements 94 3.3 The tax compliance cycle 102 3.4 Control and enforcement 115 3.5 Tax administration of other tobacco products 152 3.6 The broader elements of a good tax system 155 3.7 Conclusions 156 ANNEX 3.1 Composition of tobacco products 166 ANNEX 3.2 Example of forestalling and countermeasures 172 CHAPTER 4. Political economy 175 4.1 SCARE tactic S: Smuggling and illicit trade 176 4.2 SCARE tactic C: Court and legal challenges 198 4.3 SCARE tactic A: Anti-poor rhetoric or regressivity 212 4.4 SCARE tactic R: Revenue reduction 217 4.5 SCARE tactic E: Employment impact 228 4.6 Earmarking tobacco tax revenues to fund health 233 ANNEX 4.1 Methods to assess the nature and size of the illicit tobacco trade 253 ANNEX 4.2 How are the tobacco tax revenues earmarked? 269 CHAPTER 5. Best practices 271 TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) 283 iv v Foreword In 1999, the World Bank’s Curbing the Epidemic was the first report by an interna- tional organization to recognize that increasing tobacco excise taxes was the most effective and cost-effective measure to reduce tobacco use and save lives. Over the two decades since, the evidence base supporting this claim, especially in low- and middle-income countries, has been steadily growing. Meanwhile, the credibility of the tobacco industry’s arguments against tobacco taxation has been slowly waning. In short, health-promoting tobacco taxation has come of age, and the evidence has consistently shown that it is a win for public health, a win for revenue and a win for the economy overall. But we must be cautioned against complacency. Although the evidence on tobacco taxation is irrefutable and there are now signs that the tide is turning on the global tobacco epidemic, tobacco taxation was, in 2018, the WHO MPOWER1 measure that was least implemented at the highest level of achievement. Even more concerningly, cigarettes have become more, rather than less, affordable in many low- and middle-income countries over the past decade. Many countries set rates at insufficient levels and increase them too infrequently, while others still use complex and inefficient taxation structures. This failure to advance tobacco taxation able to effect significant price increases constitutes a loss for governments in revenues, a loss for public health and a win for the tobacco industry. To overcome this inertia, this manual charts the way forward for policy-makers, finance officials and others involved in tobacco tax policy development. It equips them with the information and evidence needed for the realization of their coun- tries’ tobacco tax policy objectives. It also analyses the tobacco industry’s tactics for influencing the political economy of tobacco taxation and shows the limitations and exaggerations of the arguments used against tax increases. The manual serves as an update of the 2010 WHO technical manual on tobacco tax administration by adding new evidence on the successes of tobacco taxation in all parts of the world and broadening its scope to capture more material relevant to developing and implementing more effective tobacco tax policy. 1 The WHO MPOWER package of technical measures and resources that comprises (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco. vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Its contributions are particularly timely in a COVID-19-stricken world. As the pandemic has been worsened by the global burden of noncommunicable disease, and revenue is now desperately needed, taxing tobacco should be more palatable than ever. This manual shows policy-makers how to seize this unique opportunity to use tobacco taxation to build back better, save lives and strengthen health systems while increasing revenue. Dr Naoko Yamamoto Assistant Director-General UHC/Healthier Populations Division vii Acknowledgements This manual was developed under the direction of Jeremias Paul Jr and Anne-Marie Perucic. The following contributed to the content of the manual (in alphabetical order): • WHO: Evan Blecher, Annerie Bouw, Mark Goodchild, Roberto Iglesias, Juliette McHardy, Jeremias Paul Jr, Anne-Marie Perucic and Robert Totanes. • External authors: Mauricio Cardenas (Columbia University), Sophia Delipalla (University of Macedonia), Luk Joossens (tobacco control expert, Belgium), Marin Kurti (Eastern Connecticut State University), Enrique Fanta (former World Bank senior specialist), David Merriman (University of Illinois at Chicago) and Jean Tesche (University of Cape Town). WHO would like to thank the following reviewers for their invaluable comments (in alphabetical order): • External: Jo Birckmayer (Bloomberg Philanthropies), Adriana Blanco Mar- quizo (WHO Framework Convention on Tobacco Control), Frank Chaloupka (University of Illinois at Chicago), Yoni Dekker (WHO Framework Conven- tion on Tobacco Control), Jeffrey Drope (University of Illinois at Chicago), Ceren Ozer (World Bank), Corne van Walbeek (University of Cape Town), Chonlathan Visaruthvong (Ministry of Finance, Thailand) and Rodrigo Santos Feijo (WHO Framework Convention on Tobacco Control). • WHO: – Headquarters: Douglas Bettcher, Itziar Belausteguigoitia, Ranti Fayokun, Joseph Kutzin, Benn McGrady, Vinayak Prasad and Susan Sparkes. – Regional offices: Nina Dela Cruz (WPRO), Fatimah El-Awa (EMRO), Charles Frasier (EMRO), Lee Lily Joung-Eun (WPRO), Jagdish Kaur (SEARO), Elizaveta Lebedeva (EURO), Maxime Roche (AMRO) and Rosa Sandoval (AMRO). WHO would also like to thank Amal Amoune-Naal for the administrative support, Alison Goldstein for the technical editing and Janet DeLand for the copy-editing. Production of this document has been supported by a grant from Bloomberg Phi- lanthropies. The contents of this document are the sole responsibility of WHO and should not be regarded as reflecting the position of Bloomberg Philanthropies. viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ACRONYMS AEO authorized economic operator AFRO WHO Regional Office for Africa AMRO WHO Regional Office for the Americas ATO Australian Taxation Office BAT British American Tobacco CCTV closed-circuit television CIF cost, insurance and freight COP Conference of the Parties CPI consumer price index CVA Customs Valuation Agreement (Thailand) DGCE Directorate General of Customs and Excise (Indonesia) DIY do-it-yourself ECBA World Bank Extended Cost-Benefit Analysis EIU Economist Intelligence Unit EMRO WHO Regional Office for the Eastern Mediterranean ENDS electronic nicotine delivery systems ENNDS electronic non-nicotine delivery systems EU European Union EURO WHO Regional Office for Europe FDA Food and Drug Administration (United States) FET fair and equitable treatment GCC Cooperation Council for the Arab States of the Gulf GDP gross domestic product HTP heated tobacco product HMRC Her Majesty’s Revenue and Customs (United Kingdom) IARC International Agency for Research on Cancer IIA international investment agreement IMF International Monetary Fund IRS Internal Revenue Service (United States) ISO International Organization for Standardization IT information technology ITC International Tobacco Control JTI Japan Tobacco International KRA Kenya Revenue Authority LMICs low- and middle-income countries ACRONYMS ix MFN most favoured nation MOP Meeting of the Parties (to the Protocol) MPOWER (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco NCDs noncommunicable diseases NCI National Cancer Institute NT national treatment OECD Organisation for Economic Co-operation and Development OST other smoking tobacco PMI Philip Morris International PPP purchasing power parity QR quick response RGTE WHO Report on the global tobacco epidemic RYO roll-your-own SACU Southern African Customs Union SCARE (S) smuggling and illicit trade; (C) court and legal challenges; (A) anti-poor rhetoric; (R) revenue reduction; and (E) employment impact SDGs Sustainable Development Goals SEARO WHO South-East Asia Regional Office SII Internal Revenue Service of Chile TADAT Tax Administration Diagnostic Assessment Tool TTC transnational tobacco company UAE United Arab Emirates UHC universal health coverage VAT value added tax WAEMU West African Economic and Monetary Union WCO World Customs Organization WHO World Health Organization WHO FCTC WHO Framework Convention on Tobacco Control WHO ISPT WHO interactive smoking projection and target-setting tool WHO TaXSiM WHO tobacco tax simulation model WPRO WHO Regional Office for the Western Pacific WTO World Trade Organization x W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N xi Executive summary This WHO technical manual on tobacco tax policy and administration builds upon the 2010 WHO technical manual on tobacco tax administration by further detailing the strategies for effective tobacco tax policy development, design, implementation and administration. This 2021 edition also serves as an update to the 2010 manual, incorporating the latest developments in science, technology and policy, as well as providing illustrative recent examples from a variety of countries. The best practices laid out in this manual are designed to inform governments on the development of their tobacco taxation policy, facilitating the achievement of their health and revenue objectives while also supporting their overall development strategy. Tobacco taxes have long been seen as a source of revenue for governments, but as evidence of the harms caused by tobacco has accumulated over the years, public perception has evolved. Increasingly, governments, as well as the general public, are recognizing that taxation of tobacco is not only a revenue source but also an effective public health intervention to reduce tobacco consumption and its associated harms. The profile of tobacco taxation as a health policy tool has increased greatly since the publication of the 2010 WHO technical manual on tobacco tax administration. Multiple global commitments have been adopted over the past decade to address tobacco use specifically – as well as noncommunicable diseases (NCDs) and the Sustainable Development Goals (SDGs) more broadly – through tax and price measures to reduce demand for tobacco products, save lives and fund develop- ment. Global development institutions, including the World Bank, the International Monetary Fund (IMF) and major philanthropic foundations, also agree with WHO on the importance of emphasizing and strengthening tobacco taxation as a key health policy tool. The COVID-19 pandemic has further fuelled this shift in the narrative on tobacco taxation by revealing how the global economy is inextricably linked with population and planetary health. Investing in health is fundamental to any economic recovery, and fiscal policy will be a key driver in addressing the socioeconomic consequences of COVID-19. Interventions such as tobacco taxation – which leads to reduced tobacco consumption, improved population health and increased revenues for governments – should be part of a comprehensive strategy for a build back better recovery. The evidence is clear: significant increases in excise taxes that lead to price increases have consistently proven to be the most effective, as well as the most xii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cost-effective, mechanism for reducing tobacco consumption. This manual will guide readers through the necessary steps to create and implement the strongest tobacco taxation policies for their specific countries. There are many factors to consider when developing tobacco taxation policy. Understanding the market is a fundamental step before deciding which form of taxation to use. Specifically, the choice between ad valorem and specific taxation is influenced by the market structure in a given country. At the same time, tax structure also shapes the market structure. Over the past decade, there has been a trend towards countries adopting specific excise taxes or mixed excise systems that rely more heavily on the specific component, which the latest global data associate with the highest average prices. Tobacco tax structures can be simple, with one flat rate across the board, or com- plex, with multiple tiers for products with different characteristics. In 2018, 31 coun- tries used complex, multitiered structures. But evidence demonstrates that simpler excise tax structures – utilized in all high-income countries – leave the least room for industry manipulation or tax avoidance and brand/product switching by consumers. Not only is it important to set taxes at a high level to discourage consumption, specific excise tax policies must include regular adjustments to increase the tax rate so that it keeps up with inflation and income growth in a country over time. Excise tax increases should aim to reduce the affordability of tobacco products. The base on which the tax is applied is also important. For specific taxation, the tax base should be the quantity in clearly defined units. For ad valorem (or mixed) taxation, the best practice is to use the retail price as the tax base and introduce a minimum excise tax. With regard to non-tax regulations that affect the price of tobacco products, pric- ing regulation may be considered to prevent the tobacco industry from exercising differential tax shifting, which it uses to ensure that large price gaps exist between premium and cheap cigarettes. However, pricing policies cannot be used alone. If con- sidered, they should be used only as complements to significant excise tax increases. Other non-tax regulations include banning promotional discounts for tobacco products and banning the sale of single cigarettes. To assuage concerns that tax increases will increase inflation – as well as to reflect the declining trend in consump- tion of tobacco products – it is good practice to exclude tobacco products from the basket of items that are used to develop consumer price indexes. Finally, in order to make excise tax on tobacco products more effective in reducing overall tobacco use, all tobacco products must be taxed in a comparable way. Regular assessment, evaluation and monitoring of the impact of tobacco tax policies over time are essential components of effective tax policy development and analysis. Governments need to have accurate estimates of price, income and tax base elasticities in order to anticipate the impact of a tax increase on consumption and EXECUTIVE SUMMARY xiii tax revenue. Ideally, other factors such as non-price policies should also be taken into account when estimating price and income elasticities for a specific country. A variety of tools and indicators exist to measure impact and monitor progress, and these are described in Chapter 2 of this manual. When developing tobacco tax policy, it is also important to take the broader policy context into consideration at both the domestic and the regional level. Domestically, cooperation is needed across sectors to ensure that policies and interventions in the areas of agriculture, trade, finance and labour do not work against the public health objectives of tobacco control and taxation. For countries that are part of a regional bloc, regional harmonization of tobacco taxation is a useful tool to prevent tax revenue erosion, tax avoidance and tax evasion, as well as to protect population health. Tax harmonization must be designed carefully, however, to be effective. The experience of the European Union (EU) demonstrates that both a declining consumption trend and stable revenues can be achieved with harmonized minimum excise tax rates. Discussions of policy development and implementation for new and emerging nicotine and tobacco products such as heated tobacco products (HTPs) and electronic nicotine and non-nicotine delivery systems (ENDS/ENNDS), are complicated, by their constantly changing technology and market dynamics. Policies and regulations need to be developed carefully and adjusted accordingly. Where HTPs are not banned, the current recommendation is to tax them at the same level as cigarettes on a per-unit basis, regardless of tobacco content. Early evidence from the United States shows that demand for e-cigarettes, a subcategory of ENDS/ENNDS products, is possibly even more price-responsive than the demand for conventional cigarettes, meaning that taxes can be used as an effective deterrent to ENDS/ENNDS products use. While there is preliminary evidence of substitutability between conventional cigarette use and e-cigarette use, further research is needed to understand substitutability effects among users of both conventional cigarettes and ENDS/ENNDS products. It is essential to imple- ment regulation of ENDS/ENNDS products along with any tax policy to safeguard public health. In countries where they are not banned, ENDS/ENNDS products must be regu- lated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS taxation. Nicotine-containing and non-nicotine-containing e-liquids should be taxed equally. Ultimately, while the policy implications of these newer products require careful consideration, the fact remains that conventional tobacco products constitute the overwhelming share of consumption (more than 97% in 2018). Tobacco tax administration must be both efficient and effective to ensure that health objectives are met and the desired level of tax revenue is raised. Since the xiv W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation of tobacco taxation often involves numerous agencies within a country, clearly defined roles and responsibilities are essential to maximize efficiency. Coordination among the different agencies involved, as well as with neighbouring countries, is required for tobacco tax administration to be effective. Performance evaluation and accountability for competent authorities is also necessary, and many tools and indicators exist to facilitate these processes (described in Chapter 3). There are a number of steps authorities should take to ensure efficiency and effectiveness at each stage throughout the tax compliance cycle (see Chapter 3, section 3.3). Control and enforcement are the main functions of tax administration, and these can best be achieved through the use of a strategic plan and a risk-based approach. Controls can be exercised through licensing and due diligence, fiscal markings (e.g. tax stamps), tracking and tracing, implementation of anti-forestalling measures, national audits and specific controls for imports and exports, as well as for free zones and transhipment points. Once smuggling or illicit trade is detected, actions such as seizing and destroying smuggled and/or illicit tobacco and col- lecting due taxes must be taken immediately. To deter further illegal activities, a comprehensive audit must also be carried out, including all those involved in the illicit acts. Penalties and sanctions must be sufficient to deter illegal activities. The Protocol to Eliminate Illicit Trade in Tobacco Products provides invaluable guidance for tobacco tax administration, control and enforcement that is applicable even for countries that are not Parties to it. The broader elements of a good tax system include proper resourcing of competent authorities, strict rules and regulations to detect and punish corruption and a strong judiciary system capable of resolving disputes as soon as possible. In its efforts to oppose tobacco tax increases, the tobacco industry utilizes many SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric, (R) revenue reduction and (E) employment impact – to influence the political economy of tobacco. Chapter 4 provides detailed analyses of these issues with supporting evidence that belies the SCARE tactics, as well as guidance for tax and other relevant authorities on how to anticipate and respond to industry arguments. This manual also provides tools and methodologies to help tax authorities define and evaluate the problem of illicit trade of tobacco products in their countries, inde- pendent of the tobacco industry’s generally inflated estimates. Price (and tax) levels are not a key determinant of illicit trade; rather, the problem is exacerbated by the lack of governance and tax administration capacity. Refraining from increasing taxes is not the solution. Countries should instead respond with a comprehensive strategy to fight illicit trade, including undertaking independent estimates of illicit trade levels and implementing good tax administration practices such as those discussed in Chapter 3 and contained in the WHO Framework Convention on Tobacco Con- trol (WHO FCTC) Protocol to Eliminate the Illicit Trade in Tobacco Products. When it comes to court challenges, the tobacco industry is less likely to challenge excise taxes than other tobacco control measures, because taxation is a comparatively well-established regulatory measure. The industry will, however, exploit the slightest vulnerability in the design, adoption or implementation of tax measures. For this reason, measures to strengthen regulators’ legal position are described that will enable authorities to protect themselves from potential legal challenges. The industry argument of regressivity, or the notion that tobacco tax increases hurt the poor because they have to pay a larger share of their income in taxes than the rich, has two fundamental limitations. First, the notion of regressivity does not take into consideration the broader health and economic harms caused by tobacco use that exacerbate the impoverishment of lower-income smokers. These harms are actually reduced when tobacco consumption decreases following a tax increase. Second, the tobacco industry argument ignores the fact that higher tobacco taxes and prices can induce behaviour change as is reflected in the price elasticity of demand. Evidence consistently shows that lower-income smokers are more sensitive to price and therefore more likely to reduce smoking in response to a tax and price increase. Including these factors shows tobacco taxation to be, in fact, a progressive public health intervention that disproportionately benefits the poor. While essentially admitting that a tobacco tax increase may have the desired effect of reducing consumption, the industry also tries to argue that a tax increase will also reduce revenues. In fact, the price inelastic demand for tobacco makes tobacco tax increases a win-win for both public health and finance. This manual presents several country examples that demonstrate how well-designed and well-implemented tobacco tax increases lead to increases rather than decreases in revenue in the short to medium term. In addition, the reduced consumption resulting from a tax increase results in reductions of other tobacco-related government expenditures as well. The final tactic used by the tobacco industry to challenge proposed tax increases is to frame tobacco taxes as an economic rather than a public health issue. This false choice between health and jobs is based on faulty assumptions that 1) tobacco is a significant source of domestic employment; 2) job creation relies on tobacco consumption and 3) tobacco-related livelihoods are prosperous, sustainable and irreplaceable. Earmarking can be a useful tool for improving the political economy of tobacco tax increases. While the primary goal of tobacco tax increases is to reduce demand for tobacco, setting aside portions of tax revenue to fund other tobacco control efforts or relevant health programmes can help convince the public, politicians and officials of the value of significant tobacco tax increases. Earmarking can also be EXECUTIVE SUMMARY xv x vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N used to counter tobacco industry arguments about potential negative effects of tax increases – for example, by supporting tobacco farmers in transitioning to other crops. Tobacco taxes work. This is why the industry invests so much money and effort in blocking large tax increases and other effective tax policy reforms. Policy-makers must not be swayed by industry pressure but need only to follow the facts. This manual provides all the information policy-makers need to make the right deci- sions at each step of the process – from designing, evaluating, implementing and administering tax policy to refuting specious industry attacks and communicating the value of tobacco taxation to legislators and the broader population. An effectively designed and efficiently administered tobacco tax policy will not only produce the direct results of reducing tobacco consumption among smokers and raising revenue for governments, its effects will be felt much more broadly. Indeed, raising tobacco taxes is a SMART policy: it Saves lives; Mobilizes resources; Addresses health inequities; Reduces burdens on health systems; and Targets tobacco use, a major risk factor for NCDs. • Saves lives: Tobacco use is the leading cause of preventable deaths globally – it claims 8 million lives each year. Tobacco taxation is the most effective mechanism for reducing tobacco consumption and its associated health burden worldwide. • Mobilizes resources: Despite being the single most effective tobacco control measure, tobacco taxation is largely underutilized as a policy mechanism. Based on available data on the price and taxation of cigarettes, it is estimated that excise taxes on cigarettes generated a worldwide total of US$ 361 billion in revenues in 2018, including US$ 162 billion in revenues for low- and middle- income countries (LMICs). If all countries were to raise cigarette excise tax rates by the equivalent of US$ 1 per pack, the amount of excise revenue from cigarettes would increase by US$ 178–219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from such tax increases, with excise revenues increasing by 82–103%, providing governments in these countries with an extra US$ 133–167 billion. This shows the substantial revenue potential of tobacco taxes. • Addresses health inequities: Tobacco taxation and tax increases are effectively progressive or pro-poor policies because of their positive distributional impact. Lower-income smokers benefit disproportionately from reduced tobacco consumption and use in terms of health gains and income retention. • Reduces burdens on health systems: The worldwide economic cost of tobacco use was US$ 1.4 trillion in 2012. Tobacco taxes reduce tobacco-related burdens on governments and health systems through population-based preventive measures. • Targets tobacco use: Tobacco taxation directly targets and reduces tobacco use, which is a major risk factor for several deadly NCDs. In summary, significant tobacco tax increases, designed and implemented according to the latest guidance and best practices presented in this technical manual – and as a strong component of a comprehensive tobacco control strategy – will bring about substantial reductions in tobacco use and the health and economic harms it causes. EXECUTIVE SUMMARY xvii x viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 1 CHAPTER 1. Why this manual? BACKGROUND Tobacco taxes are not new. Governments around the world have been applying taxes on tobacco and tobacco products practically since the idea of excise was conceived. And rightly so: tobacco is not a necessity, it is easy to tax and the demand for it is relatively inelastic. These characteristics, along with the substantial revenues tobacco taxes generate, have made tobacco a highly appropriate object of taxation. As evidence of the harms of tobacco has accumulated over the years, the public perception of tobacco taxes has evolved. Now tobacco taxes are not only seen as a revenue source, but, more importantly, they are recognized as an effective public health intervention to reduce tobacco consumption. This trend reflects the reasons excise taxes exist in the first place – to discourage harmful behaviour and to mitigate the associated negative externalities (1–2). Many governments view tobacco taxes as a significant and stable source of rev- enue, which may explain why there is often a degree of hesitation whenever tobacco tax reform is proposed. Historically, many governments have relied on revenues from tobacco taxes and have even adjusted the level of taxation according to their revenue needs (3). However, some countries are beginning to recognize the value of applying high tobacco taxes primarily as a public health tool, viewing revenues as a secondary consideration (4). Arguments against tobacco tax hikes or improvements to the tax structure are often economic in nature: such tax changes will allegedly decrease revenues, wipe out jobs, increase illicit trade and harm local industries, among other claims. But the evidence has consistently shown that such claims are simply not true in an overwhelming majority of situations. The tobacco industry, in particular, frequently portrays this conflict as a false dichotomy between public health and the economy – as if prioritizing health comes at the expense of the economy. In fact, studies and real-world experiences have shown that increasing tobacco taxes not only improves public health but also has a net positive impact on the economy and development of a country – a true win-win scenario (5–6). As an update to the first WHO technical manual on tobacco tax administration published in 2010, this manual aims to help readers better navigate the various 2 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N issues surrounding tobacco taxes and their implementation. The primary intended audience includes policy-makers, finance officials, tax authorities, customs officials and other relevant persons/bodies involved in the formulation and implementation of tobacco tax policy. The manual provides a detailed guide to the design of tobacco tax policy and describes how to effectively administer these taxes to maximize impact. Detailed discussions of the political economy considerations and the hurdles that need to be overcome before and during implementation are included as well. The overarching goal is to equip those working in the tax policy and implementation spheres with sufficient information to help realize the health and revenue objectives of a government’s tobacco tax policy in line with its overall development strategy. THE CONSEQUENCES OF TOBACCO USE AND THE NEED FOR INTERVENTION Most people are aware that smoking and tobacco use are harmful to health, but few truly comprehend the scale of this harm. The tobacco epidemic claimed more than 100 million lives in the last century (7), with updated estimates now reaching 8 million deaths annually from tobacco use and exposure to second-hand smoke (8). As much as 80% of these deaths occur in low- and middle-income countries (LMICs) (6), revealing how the developing world carries much of the global burden. Tobacco use is a major risk factor for many chronic conditions, including heart disease, cancer, diabetes and chronic lung disease – collectively known as noncom- municable diseases (NCDs). NCDs account for about 15 million premature deaths (between ages 30 and 69) worldwide, killing people in their most productive years. As the leading cause of preventable deaths, tobacco use remains one of the foremost public health challenges of our time. The consequences of tobacco use also present enormous economic, development and social costs that wreak havoc on families, communities and societies. The annual economic cost of smoking was estimated at US$ 1.4 trillion in 2012, equivalent to 1.8% of the global gross domestic product (GDP) (9). With these figures likely to have increased since then, the massive health and economic burdens of tobacco use provide justification for governments to intervene and strictly regulate the market for tobacco products. The purview of tobacco control extends beyond the strong imperative to protect people’s health and well-being; it should also strive to contain the market failures and negative externalities of tobacco use, particularly since these effects can significantly impact a country’s development trajectory. The mounting evidence of the enduring destruction caused by tobacco in the 20th century provided compelling reasons for a strong global response, which led countries to negotiate the World Health Organization Framework Convention on Tobacco Control (WHO FCTC). The WHO FCTC came into force in 2005 as the CHAP T ER 1. WHY T HIS M ANUAL? 3 first public health treaty under the auspices of WHO. To facilitate its implementation at the country level, WHO packaged a set of demand-reduction measures directly taken from the treaty (7). These interventions, collectively known as MPOWER, are as follows: (M) monitoring tobacco use and prevention policies; (P) protecting people from tobacco smoke (smoke-free laws); (O) offering help to quit tobacco use (cessation services); (W) warning about the dangers of tobacco (including graphic pack warnings and plain packaging); (E) enforcing bans on tobacco advertising, promotion and sponsorship; and (R) raising taxes on tobacco products. Specifically, under Article 6 of the WHO FCTC, Parties recognized that price and tax measures are an effective and important means of reducing tobacco consumption for various segments of the population – in particular, among young persons (10). The severity of the tobacco epidemic and its ongoing damage to health and economies are clear justifications for governments to actively intervene and correct market failures. The scale of the burden and the rate at which lives are being destroyed necessitates urgent and aggressive action on tobacco control, using measures that most countries have committed to implementing and that are proven to be effective in reducing tobacco use. WHY TOBACCO (EXCISE) TAXES ARE CRUCIAL Among the different tobacco control interventions, raising excise taxes has been identified as the most effective as well as the most cost-effective measure to reduce consumption (6). While other interventions are certainly important components of a comprehensive tobacco control strategy, the direct impact of significant tax increases on consumption is by far the strongest. On average, a tax increase that causes prices to go up by 10% reduces consumption by 4% in high-income countries and 5% in LMICs (6). When implemented at scale, this demonstrates the enormous power of tobacco taxation and its potential to save lives. Tobacco taxes differ from other interventions in that their impact can increase and build over time – even if taxes are already relatively high, their rates need to be continuously increased to retain and amplify their effectiveness. However, this should not be taken as a suggestion that governments considering tobacco control interventions should focus solely on taxes. Taxes are even more effective when implemented as part of a comprehensive package of measures such as MPOWER, which covers distinct but complementary intervention points. Among the different taxes applied on tobacco products, excise taxes are the most significant because they raise both absolute and relative prices (6). This is important when considering health objectives, since it is the magnitude of the price increase of tobacco products that determines the reduction in consumption. An excise tax is typically applied on a limited set of products, designed to discourage their use by 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N raising the price significantly over that of other products available in the market. This is in contrast to value added taxes (VAT) or sales taxes, which apply to most goods and services. Seeking to raise the prices of tobacco products through VAT or sales taxes would fail to increase relative prices, making this an ineffective and inefficient method. Customs or import duties on tobacco products are also utilized, but their impact is waning with the global trend towards bilateral and regional agreements aimed at trade facilitation. The application of these duties varies across countries, but overall, they are not applicable to locally produced tobacco products. As a tool to increase prices, import duties cannot substitute for excise taxes, since they are not specifically designed to reduce consumption. When viewed as a public health policy tool, tobacco taxation is highly cost-effective, since it delivers significant impact yet is relatively inexpensive to implement (11). The costs of implementing tobacco taxation are much lower than those of clinical NCD interventions such as cancer treatments or maintenance medications, since the commodity and human capital requirements are less substantial (12). Moreover, increasing tobacco taxes actually generates additional revenue for a government. Tobacco taxes are also very effective in pre-empting or reducing consumption among groups of people who are especially price-sensitive – youth in particular, who are prevented from initiating a lifelong addiction if taxes and prices are sufficiently high (13). This is also true for the poor, who are more prone to catastrophic health expenditures than the wealthy are. Preventing initiation or encouraging cessation by imposing high taxes provides an escape route from the vicious cycle of tobacco use and poverty (6). A DECADE OF PROGRESS AND COMMITMENT TO ACTION Since 2010, when the first WHO technical manual on tobacco tax administration was published, numerous developments have raised the profile of tobacco taxes as an essential public health intervention. The Conference of the Parties (COP) to the WHO FCTC adopted guidelines for implementation of Article 6 of the treaty, which focuses on price and tax measures to reduce the demand for tobacco. Also within this period, three high-level meetings on the prevention and control of NCDs by the United Nations General Assembly, as well as the endorsement of the Global NCD Action Plan in 2013 by the World Health Assembly, have resulted in strong global commitments to implement measures, such as increased tobacco taxes to protect people’s health. The 2030 Agenda for Sustainable Development, which contains 17 goals known as the Sustainable Development Goals (SDGs), describes the global development strategy for the next decade. Within the SDGs, two specified targets are highly relevant for tobacco control: strengthening the implementation of the WHO FCTC (target 3.a) CHAP T ER 1. WHY T HIS M ANUAL? 5 and reducing premature mortality from NCDs by 30% (target 3.4). Furthermore, the Addis Ababa Action Agenda1, which aims to provide a global framework for financing the SDGs, also highlights tax and price measures on tobacco as key mechanisms to reduce demand and save lives while increasing domestic resources for develop- ment. Another important milestone was the 2018 entry into force of the Protocol to Eliminate Illicit Trade in Tobacco Products. These key events, along with several outcome documents and policy declarations in the area of tobacco control and the wider development sphere, have introduced tobacco taxation into the consciousness of a much larger share of policy-makers. As detailed in subsequent chapters, numerous countries have imposed sufficiently high tobacco tax rates while applying best practices in tax policy design and imple- mentation over the past decade (8, 10). For example, sustained and substantial tax increases have reduced tobacco use in LMICs such as Brazil (14), Turkey (15) and the Philippines (16). High-income countries also continued their leadership in this area, comprising 23 of the 38 countries judged to have sufficiently high tobacco taxes in 2018 (8). However, much remains to be done. The 2019 WHO report on the global tobacco epidemic (RGTE) shows that tobacco taxes are still the most underutilized tobacco control policy among the MPOWER measures (8), with only 14% of the world’s population being covered by sufficiently high tobacco taxes. Substantial progress has also been made in building the tools and evidence base for tobacco taxation. Volume 14 of the International Agency for Research on Cancer (IARC) handbooks of cancer prevention, Effectiveness of tax and price policies for tobacco control, published in 2011, is a key review of the literature published as of May 2010 on the effectiveness of tax and price policies in reducing tobacco use. The National Cancer Institute (NCI)-WHO Monograph on the economics of tobacco and tobacco control, published in 2016, details the evidence accumulated over the years from various countries, focusing not only on tax and price policies, but on all aspects of the economics of tobacco and tobacco control. In addition, numerous published studies from LMICs provide a comprehensive picture of the impact of tobacco taxation in different contexts. The updated Appendix 3 of the Global NCD Action Plan explains the cost-effectiveness of tobacco taxation (11), while the Global NCD Business Plan, Saving lives, spending less, built on this work by estimating a dollar figure for the return on investment expected from implementing the best-buy interventions for tobacco control, including taxation (12). The past decade has seen major steps forward for tobacco taxation in terms of global commitments, the number of countries implementing best practices and the 1 The Addis Ababa Action Agenda of the Third Conference on Financing for Development. Third Inter- national Conference, 13-16 July 2015, Addis Ababa, Ethiopia (https://sustainabledevelopment.un.org/ content/documents/2051AAAA_Outcome.pdf, accessed 17 February 2021). 6 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N expansion of the evidence base on tobacco taxes, particularly in LMICs. Within the development sphere, institutions such as the World Bank, the International Monetary Fund (IMF) and many other multilateral agencies are aligned with WHO on the importance of tobacco taxation and the need to improve its implementation (17–18). Considerable challenges remain; although it appears that the world is headed in the right direction, progress needs to accelerate at a much quicker pace in order to achieve the SDG targets by 2030. SHAPING A “NEW NORMAL” FOR TOBACCO TAXATION The global upheaval caused by the COVID-19 pandemic has cast an unprecedented spotlight on how well governments around the world prepared for and responded to the crisis. It has exposed glaring health systems vulnerabilities and highlighted the struggles of many countries to control the spread of the virus. But perhaps more than anything, the pandemic has demonstrated how the economy, trade, science, politics and many other aspects of our societies are very much interdependent and interconnected with the health of the population. It is clear that an individual’s state of health can significantly determine their susceptibility to disease and their ability to overcome it. People with NCDs are more vulnerable to becoming severely ill with a number of conditions, which also appears to be the case with COVID-19 (19). Tobacco use is a major risk factor for NCDs, and available research suggests that smokers are at higher risk of developing severe illness and dying from COVID-19 (20). Just as the different aspects of society are interconnected, so too are people’s health, the existence of health-promoting environments and the government policies and agencies that shape these environ- ments. This critical moment presents a unique opportunity and renewed motivation to discourage the use of harmful products such as tobacco and to further improve tobacco control measures, especially tobacco tax policy. Moving forward, a business-as-usual approach to tobacco taxation will not be sufficient. Responding to this new reality and preparing for the next pandemic entails implementing measures that promote healthier populations. Like the COVID-19 pandemic, any future pandemic will likely exacerbate health inequities, bring about more economic uncertainty and put pressure on governments’ fiscal capacities. Interventions such as higher tobacco taxes, which protect people’s health while generating more revenues and economic benefits, become even more important in such crisis situations. Given this context and the stakes involved, ministries of finance and tax authorities are in a unique and powerful position – one of saving not only livelihoods, but also lives. The importance of increasing tobacco taxes – one of the most effective public health tools available – cannot be overstated. The traditional approach of treating CHAP T ER 1. WHY T HIS M ANUAL? 7 tobacco tax exclusively as a revenue source has no place in the new normal. One cannot deny the scale of the tobacco epidemic, the necessity to correct market failures and the overwhelming evidence of tobacco taxation’s benefits to health and to the economy. The positive trend in the changing narrative around tobacco taxation needs to continue. Tobacco taxation should not be viewed in isolation from the rest of government policies, but rather as an important part of the whole, an essential piece in working towards our common goal of better health for all. OVERVIEW OF SUBSTANTIVE CHAPTERS This manual is primarily designed for policy-makers, finance officials, tax authorities and customs officials. It may also be useful for officials within health ministries or other government agencies, as well as nongovernmental organizations working in this area, including tobacco control advocates. Significant effort is made to present real-world examples and recent experiences from a wide range of countries to demonstrate success stories and lessons learned in raising tobacco taxes. A sub- stantial amount of evidence has been generated in LMICs over the past few years that supports and augments the existing evidence base, providing a much broader body of knowledge than was available when the first WHO technical manual on tobacco tax administration was released. Chapter 2 delves into the theory, practice and empirical evidence on tobacco excise tax policy, including current global trends. The chapter offers a detailed analysis of the various elements that constitute tax structure, aiming to provide policy-makers with a comprehensive understanding of the factors affecting prices, consumption and the market. It describes the key components to keep in mind when designing tobacco tax policy to maximize the impact of tax increases and improve the tax structure. The chapter also includes updated global price and tax data, specific examples from various countries and a discussion of tax base elasticity, automatic excise tax adjustments and pricing regulations, as well as descriptions of new and emerging nicotine and tobacco products, including electronic nicotine- and non- nicotine delivery systems (ENDS/ENNDS) and heated tobacco products (HTPs). Chapter 3 focuses on tobacco tax administration. It provides an in-depth discus- sion of the fundamental components that make tobacco tax collection effective and efficient, ensuring achievement of the health and revenue objectives of tax policy. It highlights the importance of cooperation among the various agencies involved in the implementation of tobacco taxes within countries and across borders. Building on country and regional experiences from previous decades, the chapter outlines specific measures and recommendations to maintain oversight of the whole tax compliance cycle. Also included are actions to facilitate control and enforcement, such as licensing, fiscal markings (e.g. tax stamps), tracking and tracing systems 8 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N and import and export controls. Finally, the discussion pivots towards the broader elements of a good tax system, such as proper resourcing of competent authorities, a strong judiciary and strict rules regarding corruption. Chapter 4 deals with the important challenges in the area of political economy when countries attempt to increase tobacco taxes or simplify the tax structure. The tobacco industry often relies on identified patterns of argumentation and tactics to obstruct such reforms, i.e. SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric (regressivity), (R) revenue reduc- tion and (E) employment impact – each of which is discussed thoroughly in this chapter. Also included is a detailed discussion of the measurement of illicit trade and a discussion of earmarking tobacco tax revenue for health purposes. Finally, Chapter 5 presents a comprehensive list of the best practices in tobacco tax policy and administration discussed throughout this manual. The list is intended to serve as a practical guide and quick reference to the salient points presented. CHAP T ER 1. WHY T HIS M ANUAL? 9 REFERENCES 1. Ranson K, Jha P, Chaloupka FJ, Nguyen SN. The effectiveness and cost-effectiveness of price and other tobacco control policies. In: Jha P, Chaloupka FJ, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000:427–447 (https://www.paho.org/hq/dmdocuments/2010/ Cost-effectiveness%20of%20price%20increases.pdf, accessed 4 February 2021). 2. Effectiveness of tax and price policies for tobacco control. Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications.iarc.fr/Book-And-Report-Series/Iarc- Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And-Price-Policies-For-Tobacco-Control-2011, accessed 4 February 2021. 3. Tobacco taxation in the United States. In: Lynch BS, Bonnie RJ, editors. Institute of Medicine (US) Committee on Preventing Nicotine Addiction in Children and Youths. Growing up tobacco free: preventing nicotine addiction in children and youths. Washington (DC): National Academies Press (US); 1994 (https://www.ncbi.nlm.nih.gov/books/NBK236771/, accessed 10 November 2020). 4. Sin tax reform. Manila: Department of Finance (Philippines); 2012 (https://www.dof.gov.ph/advocacies/ sin-tax-reform/, accessed 10 November 2020). 5. Goodchild M, Perucic AM, Nargis N. Modelling the impact of raising tobacco taxes on public health and finance. Bull World Health Organ. 2016; 94:250–7 (https://www.who.int/bulletin/ volumes/94/4/15-164707.pdf, accessed 4 February 2021). 6. The economics of tobacco and tobacco control. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute, NIH Publication No. 16-CA-8029A; 2016 (National Cancer Institute tobacco control monograph 21; https://cancercontrol.cancer.gov/ brp/tcrb/monographs/monograph-21, accessed 17 December 2020). 7. WHO report on the global tobacco epidemic, 2008: the MPOWER package. Geneva: World Health Organization; 2008 (https://www.who.int/tobacco/mpower/mpower_report_full_2008.pdf, accessed 10 November 2020). 8. WHO report on the global tobacco epidemic, 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/tobacco/global_report/en/, accessed 10 November 2020). 9. Goodchild M, Nargis N, Tursan d’Espaignet E. Global economic cost of smoking-attributable diseases. Tob Control. 2018;27:58–64 (https://tobaccocontrol.bmj.com/content/27/1/58, accessed 4 February 2021). 10. Guidelines for implementation of Article 6 of the WHO FCTC. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 4 February 2021). 11. Tackling NCDs: ‘best buys’ and other recommended interventions for the prevention and control of noncommunicable diseases. Geneva: World Health Organization; 2017 (https://apps.who.int/iris/ handle/10665/259232, accessed 10 November 2020). 12. Saving lives, spending less: a strategic response to noncommunicable diseases. Geneva: World Health Organization; 2018 (https://www.who.int/publications/i/item/WHO-NMH-NVI-18.8, accessed 4 February 2021). 13. Chaloupka FJ, Warner KE. The economics of smoking. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics. Elsevier; 2000;1(1):1539–1627. 14. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Rev Panam Salud Publica. 2016;40(4):243–9 (https://iris.paho.org/bitstream/ handle/10665.2/31306/v40n4a09_243-9.pdf?sequence=1&isAllowed=y, accessed 17 February 2021). 15. Cetinkaya V, Marquez PV. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 16. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington (DC): World Bank Group; 2016 (http://documents.worldbank.org/curated/en/638391468480878595/pdf/106777-PUB-PUBLIC- PUBDATE-7-26-2016.pdf, accessed 10 November 2020). 17. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 10 November 2020). 10 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 18. Irwin A, Marquez P, Jha P, Peto R, Moreno-Dodson B, Goodchild M, et al. Tobacco tax reform: at the crossroads of health and development – a multisectoral perspective. Washington (DC): World Bank Group; (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report- CrossRoads.pdf, accessed 2 February 2021). 19. Information note: COVID-19 and NCDs. Geneva: World Health Organization; 23 March 2020 (https:// www.who.int/publications/m/item/covid-19-and-ncds, accessed 10 November 2020). 20. WHO statement: tobacco use and COVID-19. Geneva: World Health Organization; 11 May 2020 (https:// www.who.int/news-room/detail/11-05-2020-who-statement-tobacco-use-and-covid-19, accessed 10 November 2020). 11 CHAPTER 2. Tobacco excise tax policy 2.1. GLOBAL OVERVIEW OF TOBACCO TAX PRACTICES A well-designed tax policy is key to having an effective tax policy. Any government that is planning to reform its tax policy must first understand the fundamental components of a good tax policy, as well as consider the strengths and weaknesses of different approaches to taxation, how they impact price and the requirements for tax administration. Understanding how the tobacco market operates in a country is equally important for policy-makers because of the inevitable interaction between the market and tax structures. Beyond the political considerations that strongly influence tobacco tax policy development, this chapter focuses on the technical aspects of tobacco taxation – excise tax in particular. Section 2.1 provides an overview of tobacco tax practices at the global level, focusing on the different ways countries structure excise tax. Section 2.2 emphasizes the importance of carefully designing excise tax policy, highlighting not only the significance of tax increases but also excise tax structure and its impact on prices, taking into account how market structure influences trends. This section also discusses the importance of measuring impact as another aspect of tax policy development, and it presents the crucial elements for performing measurement, as well as the relevant indicators available to monitor progress. Section 2.3 describes external policy considerations in the design phase to ensure that the goals of tobacco control and taxation are achieved. Intersectoral policy integration and coherence at the domestic level is discussed as a strategy to ensure that policies of other sectors do not inhibit or obstruct public health policy objectives. This section also reviews the current state of regional tax harmonization based on the experience of existing regional blocs and draws conclusions on the best policy approaches to preserve the public health interests of individual countries. Section 2.4 discusses new and emerging nicotine and tobacco products, in par- ticular HTPs and ENDS/ENNDS. It reviews the latest evidence on the health impacts of these products and current approaches to regulation. Key policy considerations are identified, and recommendations are provided for adopting an appropriate excise tax policy for these products. Section 2.5 summarizes the issues covered in the chapter and the key takeaways. 12 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.1.1 TAXES APPLIED ON TOBACCO PRODUCTS Taxes are classified as either direct or indirect. Direct taxes are imposed on the profit, income, property or wealth of persons or companies, whereas indirect taxes are imposed on the price of goods and services. Indirect taxes are most relevant to tobacco products taxation, because they directly influence price. A variety of types of indirect taxes can be applied to tobacco products. These include: • excise taxes – taxes that apply to a few selected commodities (they can also be applied to alcohol, fuel, sugar-sweetened beverages, etc.). • VAT or sales taxes – VAT is a multistage tax on all consumer goods and services that is applied proportionally to the price the consumer pays for a product. It is a tax on the amount by which the value of an article has been increased at each stage of its production or distribution. Some countries impose sales taxes instead of VAT. Unlike VAT, which is collected at every stage of the supply chain, sales taxes are generally levied at the point of retail on the total value of goods and services purchased. Ultimately, the consumer ends up paying the tax, whether it is a VAT or a sales tax. • import duties – taxes on selected goods imported into a country to be consumed in that country (i.e. goods that are not in transit to another country). In general, import duties are collected from the importer at the point of entry into the country. • other taxes – other indirect taxes, such as environmental taxes, that do not fall into any of the categories listed above. One of the most well-established and widely understood points in tax policy is that tobacco products should be subject to excise taxation. The focus of this chapter – and of this manual overall – is on excise taxes. They are the most important type of indirect taxes for tobacco control because they are applied directly to tobacco products and contribute the most to increasing the price of tobacco products relative to other goods and, subsequently, to reducing consumption. There are two basic types of excise taxes: • specific – levied as a monetary value per quantity of the product being taxed (e.g. 1 000 cigarettes, pack of 20 sticks, kilogram of tobacco); and • ad valorem – levied as a percentage of the value (e.g. retail price, or the producer/ex-factory price or the cost, insurance and freight [CIF] value1) of the product being taxed. These types of excise tax can be applied at a uniform or a differential (tiered) rate and on their own or in combination (i.e. a mixed system). 1 CIF is the value of an imported product as declared to customs upon entry into a territory. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 13 Given the widespread use of cigarettes – almost the only tobacco product used in some parts of the world – and the scarce availability of data for other tobacco products, this chapter focuses mainly on cigarettes. But there are a few examples and recommendations for other tobacco products, including those that are more prevalent in specific parts of the world (e.g. bidis or smokeless tobacco in South-East Asia and waterpipe tobacco in the Eastern Mediterranean region). 2.1.2 CIGARETTE TAXES AND RECENT TRENDS WORLDWIDE Tax and national income levels: the higher the income level, the higher the taxes and prices At the global level, cigarette price and tax levels correlate positively with a country’s income level: prices and taxes are higher in higher-income countries and lower as income level decreases. This trend has not changed over the years since 2008. Figure 2.1 presents the levels of price and tax by income groups for 2018, using the World Bank classification of income groups. Fig. 2.1 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by income group, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in purchasing power parity (PPP) adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Excise tax per pack Other taxes Retail price minus taxes Pr ic es a nd ta xa tio n pe r p ac k (P PP $ ) High-income Low-income 4.25 0.68 Middle-income 2.06 Total taxes: 5.30 (67.9% of pack price) Total taxes: 2.91 (58.3% of pack price) Total taxes: 1.18 (38.1% of pack price) PPP $ 7.80 PPP $ 4.99 PPP $ 3.09 Global 2.48 PPP $ 5.53 Total taxes: 3.36 (60.8% of pack price) 0 1 2 3 4 5 6 7 8 14 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tax and price levels: the higher the tax share, the higher the price Globally, cigarette prices correlate positively with tax percentage levels: as the total tax share (of which excise represents the largest part) as a percentage of retail price increases, the price of cigarettes generally also increases (see Fig. 2.2 below).2 This indicates that taxes do influence prices. Fig. 2.2 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by total tax levels, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Tax level and WHO regional classification: taxes and prices are highest in the European region, followed by South-East Asia, the Americas and the Western Pacific, with the lowest levels in the Eastern Mediterranean and African regions At the regional level (WHO regional classification), average levels of prices and taxes vary greatly. The highest level can be seen in the European region, which includes the European Union (EU) countries. The EU’s unified tax structure includes high levels of minimum taxes – which lead to high prices – and encourage member 2 This is a general trend and does not apply for every country; there are countries that have a large tax share but low prices for cigarettes. Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax ≥ 75% Total tax ≤ 25% 0 7 4 1 2 3 6 5 50% ≤ Total tax < 75% 25% ≤ Total tax < 50% Excise tax Other taxes Retail price minus taxes 4.44 2.24 1.57 PPP $ 5.07 PPP $ 5.33 PPP $ 7.07 PPP $ 2.60 0.17 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 15 countries to regularly increase their taxes to meet their obligations. However, with the current minimum level now being reached by all EU member countries, the motivation to increase excise taxes may wane. Indeed, the minimum cigarette excise amount in the EU has not been adjusted since it went into effect on 1 January 2014, and it is suffering from inflation erosion. Member States of the EU acknowledged this in June 2020 by stating that action at the EU level is required to ensure that minimum excise duty rates regain traction to effectively reduce the consumption of tobacco products and that the minimum rates of excise duties on a number of tobacco products would be increased (2). Excise taxes are lowest in the African and Eastern Mediterranean regions. And China – reported separately due to its size – has lower tax rates than the Western Pacific region (see Fig. 2.3). Fig. 2.3 Weighted average retail prices and taxation (excise and total) of most-sold brand of cigarettes, by region, 2018 Notes: China is represented separately from the Western Pacific Regional Office (WPRO) average because of its exceptionally large number of smokers compared with the number in other countries in the region. AFRO is the African Region, AMRO is the Region of the Americas, EMRO is the Eastern Mediterranean Region, EURO is the European Region, SEARO is the South-East Asia Region. Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Retail price Excise tax Total tax share % AFRO AMRO EMRO EURO SEARO WPRO w/o China China All 3. 80 4. 02 4. 24 4. 89 4. 02 1. 45 2. 89 2. 65 2. 48 7. 53 7. 27 5. 61 2. 51 1. 95 1. 05 42.2% 55.8% 61.2% 72.9% 63.2% 55.7% 60.8% 56.9% Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax share % 5. 53 0 1 2 3 4 5 6 7 8 9 10 40 30 50 60 70 80 20 16 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Global tax structures trend: more countries are adopting specific excise taxes or mixed excise systems that rely more on the specific component Since 2008, the number of countries that rely solely on ad valorem taxes or apply no excise taxes at all has decreased as more countries have adopted specific or mixed systems. More of the countries that have implemented a mixed system have increased the specific component of the tax structure relative to the ad valorem component (see Figs. 2.4 and 2.5).3 Fig. 2.4 Changes in excise tax structure, 2008–2018 Fig. 2.5 Changes in reliance on specific versus ad valorem component in mixed systems, 2008–2018 Source: (1). 3 For information about countries that applied each type of excise tax structure in 2018, see Annex 2.1. N um be r o f c ou nt ri es Specic excise Ad valorem Mixed excise No excise 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 54 56 57 63 50 45 24 23 24 21 19 15 49 47 44 42 41 55 57 59 56 60 63 62 Mixed excise Relying more on specic Relying more on ad valorem N um be r o f c ou nt ri es 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 45 22 23 23 24 22 26 27 27 27 32 35 37 50 54 56 57 63 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 17 The imposition of a minimum specific excise tax: half of the countries that have a mixed or ad valorem structure impose an excise tax floor A minimum specific excise tax ensures that at least a certain minimum amount of tax is paid, irrespective of price level. Almost half of the 101 countries that impose either ad valorem or a mixed excise for which data on minimum excise are available (47 countries) set a minimum specific excise tax.4 Nearly two thirds of those that set a minimum specific excise tax (29 countries) are high-income countries; most of them are in the EU, which requires its members to impose a minimum specific excise tax. The choice of tax base worldwide: almost half of the countries that apply a mixed or ad valorem excise system use retail price as the base Setting the base for applying a specific excise is relatively easy: most countries use a defined quantity of sticks for cigarettes, the weight in kilograms for tobacco and the weight in grams for other tobacco products (1). Different bases for ad valorem excises are applied in different countries. Nearly half of the 105 countries that implement either ad valorem or a mixed excise for which data are available (47 countries) use the retail price5 as the tax base for the ad valorem part, and most of those (28 countries) are high-income countries. Using the retail price as the base for the excise ad valorem tax is more effective than using the producer price or the CIF value. Unlike retail prices, which are easy for tax administrators to ascertain by monitoring the market, the producer price or CIF value is prone to undervaluation by producers or importers, who may pass on their margins to related parties further down the supply chain and successfully reduce their tax burden. This tactic is also known as transfer pricing. Additionally, global-level data show that the excise ad valorem on the retail price seems to lead to higher retail prices on average compared with an excise ad valorem applied on other bases, such as the producer price or CIF value (see Fig. 2.8 below). On complex tiered structures: 31 countries still apply complex, multitiered excise taxes on tobacco products As of 2018, 31 countries imposed excise taxes that varied according to defined char- acteristics of cigarettes, including price level, type of production, type of package and length of cigarette (Table 2.1). Some countries use more than one criterion to differen- tiate the tax rates. Indonesia, for example, imposes differential rates based on volume 4 This means that countries with a mixed system impose an overall minimum specific excise tax (where the yield of the specific plus the ad valorem excise cannot be below the set minimum specific excise tax), in addition to the excise on a specific component. 5 Countries that impose ad valorem on retail price exclusive of VAT are also included, since retail prices are easy to determine and VAT rates are known variables. 18 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N produced, type of cigarette and price level. In other countries, such as Member States of the EU, differential rates for cigarettes are prohibited by law, and the rate of the ad va- lorem tax and the amount of specific excise duty must be the same for all cigarettes (3). Table 2.1 Criteria used by countries for tiered excise taxes, 2018 BASE OF TIERS COUNTRY Retail price Bangladesh, Belarus, Indonesia, Jordan, Mozambique, Myanmar, Pakistan, Thailand Cigarette grade (e.g. premium, mid-grade, economy) Egypt, Japana, Mali Producer price China, Lao People’s Democratic Republic Production volume Indonesia Type filter/non-filter Belarus, Georgiaa, India, Kenya, Republic of Moldova, Nepal, Papua New Guinea hand/machine made India, Indonesia kretek/white cigarette Indonesia tobacco content (dark/ blonde or dark/light) Algeria, Bolivia (Plurinational State of ) Packaging soft/hard Mozambique, Uganda Cigarette length India, Nepal, Sri Lanka Trade (domestic/imported) Iran (Islamic Republic of ), Lebanon, Myanmar, Solomon Islands, Tonga, Uzbekistan Leaf content (domestic/imported) Fiji, United Republic of Tanzania a Japan and Georgia were using a tiered excise tax structure when these data were collected in 2018, but as of 2020, that is no longer the case. Source: (1). 2.2 DESIGNING EXCISE TAX POLICY Significantly increasing the taxes on and prices of tobacco products is the most effective and most cost-effective policy to control tobacco use (4). Increased taxes that are passed on to tobacco users as higher prices reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face challenges ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues such as addressing the SCARE6 tactics of the tobacco industry. This section provides guidance for policy-makers regarding the best tax structure to use from a health perspective, taking into consideration all the appropriate tax designs. It also proposes recommended indicators to consider when formulating policy change. 6 SCARE tactics are the tactics most commonly used by the tobacco industry when countries plan to increase tobacco taxes. They are described, and refuted, in detail in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 19 2.2.1 THE IMPORTANCE OF THE TYPE OF EXCISE TAX STRUCTURE The existing theoretical and empirical evidence on approaches to the choice of (uniform) specific and ad valorem excises is reviewed below, along with their effects on price, consumption, perceived quality and variety of tobacco products, govern- ment revenue and tax administration. The use of the word “quality” in this chapter does not refer in any way to the health impact of a tobacco product. It refers rather to the consumers’ perceptions of quality and their decision to buy a product, which they may evaluate based on the packaging, the blend used for the cigarette or anything that makes the product more appealing to them. Just to be clear, from a public health perspective, all cigarettes are equally harmful even if perceived by consumers as having higher or lower quality. The choice between ad valorem and specific taxation is influenced by the market structure, i.e. the nature and degree of competition in the market for goods and services. Although each country has its own specific characteristics, the tobacco market structure is typically a monopoly or an oligopoly where firms have the power to control prices – and hence exploit the tax structure – to their benefit. For example, China, the largest producer and consumer of tobacco products in the world (5), has a state monopoly. In Viet Nam, foreign brands are produced under licence by the state monopoly. In Thailand and Egypt, despite the presence of foreign companies, the market is dominated by the state-owned company. In Uruguay, the oligopoly is led by a domestically owned company. In Bangladesh, the oligopoly consists of domestically owned companies competing with foreign companies (6). In most of Africa, the market consists of transnational tobacco companies (7). The impact of tax structure on final price: uniform specific versus uniform ad valorem The choice between specific and ad valorem taxes is a long-standing issue in tax policy, as the level and structure of excises have different implications for the interests and goals of various groups. Given the market structure of the tobacco industry – typically a monopoly or oligopoly for most products in most countries – different excises may have different effects on government revenue, manufacturer profit, consumer price, perceived product quality and variety and tax administration (8–16). Consequently, the two types of excise taxes – specific and ad valorem – may have different implications for public health to the extent that they affect individual consumption via their impact on perceived product quality, variety and prices. Moreover, governments have the potential to influence tobacco excises to manage demand, raise revenue and promote public health. 20 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax on prices and in turn on consumption is also influenced by price and income elasticities, as well as consumer perceptions of quality (“perceived quality”) and the variety of available products, which, in turn, are closely linked to the type of tax structure adopted. Tax structure is affected by both the price elasticity of demand and the price elasticity of supply. The price elasticity of demand measures the responsiveness of consumer demand to changes in prices. The price elasticity of supply measures how sensitive producers are to changes in prices. Tobacco tax structure is also influenced by market structure. In a monopoly, the profit-maximizing firm sets the price, considering the price elasticity of demand: the lower the price elasticity (in absolute value) – i.e. the less sensitive the consumer is to price changes – the higher the price the monopolist can set. Profits are typically abnormal in a monopolist market structure, meaning total sales revenue is higher than total cost (where total cost includes a normal profit). A monopolist producer therefore receives more than the minimum reward required to invest its (physical and human) capital and undertake business risks. Economic theory predicts that in a private monopoly, prices are higher than in an oligopolistic market. This is not, however, necessarily true when the monopoly is owned by the state and the government’s objective is not straightforward profit maximization: the government might have other considerations, such as preserving jobs (e.g. in China) or keeping prices low for low-income consumers (e.g. in Egypt). Under a monopoly, an ad valorem taxation structure enables the monopolist producer to set prices lower than would be possible under a specific tax structure. This is feasible because under ad valorem taxation, when supply increases and price falls, the price reduction is not fully borne by the producer. Rather, the price reduc- tion is partly shared by the government since, as supply increases, the tax per unit of product sold falls. In other words, ad valorem taxation leads to lower prices and higher consumption relative to revenue-equivalent specific taxation. Technically, this means that the supply function is less elastic under ad valorem taxation. In contrast, under a specific taxation structure, any increase in the monopolist producer price will go to the producers as revenue, which incentivizes them to increase prices. The same logic also applies to an oligopolistic market structure, where profits again are, in general, abnormal. KEY TAKEAWAY 1 In a monopoly or an oligopoly, specific taxation incentivizes industry to set prices higher than it would with ad valorem taxation. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 21 Understanding oligopolies, however, is more complicated, since they are characterized by strategic interdependence among a few firms. This strategic interdependence extends to the relationship between industry and regulators. Industry anticipates the government’s regulatory policy – whether through tax or other interventions – and acts accordingly. For example, competitors may coordinate and lobby against a certain tax structure reform or tax rate increase. Under an oligopoly market structure, ad valorem taxation is a relatively more efficient tool for transferring part of the profits to the government as tax revenue, since it acts like both an excise and a profit tax. In contrast, a specific tax has a smaller (negative) effect on profits. This explains why we observe multinationals that are leaders in high-priced brands (e.g. Philip Morris International [PMI]) lobbying in favour of specific taxation (17). As an example, in the countries of the Cooperation Council for the Arab States of the Gulf (GCC), the tobacco industry has been trying for a long time to lobby governments to introduce a specific excise (18–21). After years of consideration and discussions on the possible introduction of excise taxes, the GCC adopted the Common Excise Tax Agreement of the States of the Gulf Cooperation Council in November 2016 (21), which introduced an ad valorem excise on tobacco products. Tobacco companies’ support for excise tax structures ultimately depends on the market segments they control in a particular country. A company selling mainly premium brands will favour specific excises, whereas a company that sells mid-priced or economy brands would favour ad valorem excise (17). When oligopolistic firms produce identical products, a specific tax has a stronger positive effect on price and is more likely to be overshifted to consumer prices than an ad valorem tax (13). Overshifting means that the price increases by more than the tax increase itself. Empirical evidence supports this (22–26). KEY TAKEAWAY 2 In an oligopoly, prices are likely to increase by more than the amount of the specific tax increase when demand is relatively inelastic. In general, demand for a product depends not only on prices but also on consumer perceptions of quality and preferences for variety. For example, the most popular brand in GCC countries is Marlboro, a premium brand (1). Consumers differ in their willingness to pay, depending on their respective perceptions of quality, which influence whether they ultimately purchase high- or low-priced brands. A tax-induced price increase can cause the following plausible responses from consumers or users of tobacco products: (1) a group of consum- ers will quit; (2) a group of consumers will reduce their overall consumption; 22 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N (3) another group, most likely high-income users, will switch to an upgraded version of the tobacco product if its relative price (compared to the cheaper brand) has been reduced, which is the case under a specific tax increase; there might also be a group of consumers in the lower income range who switch to lower-priced variants of the tobacco product if the price gap increases, as is the case under an ad valorem tax increase; and (4) another group might switch to the illegal market or buy products in a neighbouring country with a lower tax where possible. Consumers’ decisions to purchase are also affected by their preferences for variety – meaning preferences among products that consumers perceive as equal in quality but are given different characteristics by the producers to account for consumer taste preferences. Thus, it is possible that a tax increase that leads to an increase in average prices will lead to an increase in the total quantity demanded in the market because of an increase in the variety of product choices available to consumers. Variety enables new consumers to be captured, especially in an environment lacking certain regulations (e.g. without plain packaging and flavour bans). The tobacco industry was able to capture a new group of consumers when it introduced menthol cigarettes into the tobacco market. There is more than sufficient evidence that menthol cigarettes increased youth smoking initiation, increased nicotine dependence and reduced adult smoking cessation (27). To prevent this from happening in their countries, Member States of the EU have prohibited characterizing flavours other than tobacco in tobacco products (28). It is therefore important to consider the broader effects that the structure and level of an excise tax can have on average price, perceived quality and the variety of cigarette brands and other emerging substitutes. When consumers make choices based on dimensions other than quantity, the two types of tax structures are not equivalent, even in a perfectly competitive market where firms have no market power (29–30). To illustrate this point, consider a US$ 1 cost to improve consumer perceptions of quality for a tobacco product. This will lead to an equivalent price increase under specific taxation but not under ad valorem taxation. At an ad valorem rate of 20%, the price must increase by more than US$ 1, or by 1/(1 – 0.2) to cover the US$ 1 cost of improvement, due to the multiplier effect. A specific tax induces consumers to reduce the quantity demanded, but they might still choose to pay a higher price in exchange for a product that they perceive to be of better quality. An ad valorem tax, on the other hand, leads to a reduction in both quantity and perceived quality, not a substitution between them. An ad valorem tax has only an income effect and – unlike specific taxation – does not lead to substitution between perceived quality and quantity. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 23 KEY TAKEAWAY 3 Under specific taxation, the industry has incentives to create upgraded variants of tobacco products that attract new consumers and encourage consumption. When firms produce differentiated products, as the tobacco industry does, economic theory provides ambiguous results regarding specific versus ad valorem taxation. The relative effects of the two types of tax are not as straightforward as in the case of oligopolistic firms producing a homogeneous product. With differentiated products, the relative effects of the tax types depend on various assumptions: whether or not firms face symmetric costs, whether the number of firms is fixed or new firms can enter the market and the level of the tax revenue requirement. When firms face different costs, ad valorem taxes exacerbate the absolute differences in marginal costs between them. The high-cost brand is not considered a perfect substitute for the low-cost brand. A sufficiently high ad valorem tax rate may lead to a relative underproduction of the high-cost products (31). The effect of specific and ad va- lorem taxes on consumer perceptions of quality depends on market structure and the price and income elasticities of demand across various qualities. The relative price of the cheapest product does not necessarily remain unchanged or increase; it might fall (32). Empirical evidence showing that increases in the specific tax lead to a lower market share for the cheaper generic brands and an upward shift to premium brands (33–34) usually considers gradual tax increases and ignores income effects. Chaloupka et al. (35) found that in 21 EU countries that impose a mixed tax system, the price gap between premium and low-priced brands – while not reflecting the full distribution of cigarette prices – is smaller when the specific component of the mixed structure dominates.7 Although the price gaps are narrower under specific taxation, there is evidence that firms sometimes respond by introducing new, very cheap (subvalue) brands, or they exercise differential tax shifting. This practice has been evident in India for quite some time, with the Indian Tobacco Company launching a number of cheaper variants of its flagship cigarette brand, Gold Flake, to take advantage of a lower excise tax rate in the so-called microcigarette (< 60 mm length) market (36). Consequently, the cheapest end of India’s cigarette market has expanded significantly in recent years due in part to the marketing of new brand variants like Gold Flake Century. 7 The EU countries impose a mixed tax structure with a minimum tax floor. Some countries rely on the specific component more than others, but they remain within a given range (the specific component must be between 5% and 76.5% of total tax share of the weighted average price). 24 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As another example, transnational tobacco companies, which have sold ultra- low-priced brands in the United Kingdom since 2006, have managed to double their market share in a few years: their real price did not increase, since they absorbed part of the tax increases (37). The share of ultra-low-priced brands increased between 2001 and 2009 from 5% to 10%, while the market share of economy brands increased from 40% to 50% and the market share of premium brands and mid-priced brands decreased during the same period (from 35% to less than 25% for the former and from 15% to 5% for the latter). In order to keep the price of discount brands low and certain consumers in the market, firms may overshift the tax for premium products and undershift it for the lower-priced products (37–40). KEY TAKEAWAY 4 Evidence suggests that the price gap between brands is narrower under a specific tax structure. As the tobacco industry simultaneously consolidates producers and widens its portfolio of products, evidence is emerging that it is introducing cheaper brands while increasing the price of its expensive brands, therefore paradoxically widening the price gap within its products. The extent of the impact is still unclear, however, and this evidence does not negate the overall conclusion that a specific tax structure reduces price gaps. The impact of tax structure on final price: uniform specific, ad valorem and mixed systems Evidence from the 2019 RGTE (1) data suggests that the average price of the most- sold brand of cigarettes – weighted by the number of smokers – is the highest in countries implementing a mixed system that relies more on specific excise, followed by countries applying specific excise taxes only, followed by countries applying a mixed system that relies more on ad valorem and then by countries that apply ad valorem excise only (Fig. 2.6). The price is lowest in countries that have no excise at all. In past WHO reports on the global tobacco epidemic, countries that applied specific excise only had the highest price, on average. The trend may have changed partly because more countries are adopting mixed excise systems. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 25 Fig. 2.6 Weighted average price and excise for a pack of the most-sold brand of cigarettes, by excise tax structure, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Shang et al. (41) describe and compare price distributions, using data from 16 countries of the International Tobacco Control (ITC) Project that impose different cigarette tax structures. Specific uniform taxation tends to result in less variability in prices than all other structures (ad valorem tax, mixed tax, tiered tax). In general, structures other than uniform specific tax give rise to more opportunities for brand switching and tax avoidance. Reliance on complicated systems is likely to be as- sociated with wider price distribution, leading to greater tax avoidance, as there are more opportunities for substitution with cheaper brands when taxes rise. KEY TAKEAWAY 5 Evidence suggests that the tax structures most likely to lead to higher prices are uniform specific excise tax structures or mixed systems that rely more on specific excises. Mixed system relying more on specic excise Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Mixed system relying more on ad valorem excise Ad valorem excise No excise 2.21 Specic excise 2.62 3.26 0 7 6 5 4 3 2 1 1.66 Excise tax Other taxes Retail price minus taxes PPP $ 7.56 PPP $ 5.66 PPP $ 4.84 PPP $ 3.73 PPP $ 2.52 26 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax structure on final price: uniform excise versus tiered tax systems Another aspect of tax structure that impacts final price is the use of tiered taxation, i.e. tax rates that vary according to product characteristics. The characteristics can vary, from price level to the type of tobacco leaf contained in the cigarette, the size of production volume, the packaging, etc. Table 2.1 (earlier) lists the criteria used by 31 countries as the basis for different tax rates. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax (see Fig. 2.7). Fig. 2.7 Weighted average price and excise for a pack of the most-sold brand of cigarettes for countries with and without tiered taxation, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Uniform specific tax structures are likely to lead to relatively higher prices with less variability in price distribution. Compared with tiered tax structures that have differential rates based on brand characteristics, uniform taxation may reduce consumers’ incentive to switch to cheaper brands (leading to higher quit rates and lower prevalence), as well as decreasing manufacturers’ incentive to reduce their tax liabilities by changing their pricing strategies, production process or size (42–45). In Indonesia, for example, where small producers were taxed more fa- vourably, manufacturers had an incentive to reduce their scale of production but increase the number of affiliated small companies. The issue was resolved when tax authorities considered the aggregate production of all affiliated companies in the application of differential tax rates. By 2017, there were 786 active factories, while Excise uniform Price and taxation per pack of 20 sticks (PPP $) Excise tiers 3.28 1.94 Excise tax Other taxes Retail price minus taxes PPP $ 4.99 PPP $ 6.30 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 27 there had been 4 198 factories in 2006.8 Abolishing the differential tax rates would have been even more effective in removing the incentive for tax avoidance, as well as helping improve both public health and government finances. KEY TAKEAWAY 6 Evidence suggests that applying a uniform excise tax on cigarettes is not only easier to administer than tiered systems but also more likely to lead to higher cigarette prices. The impact of tax structure on final price: the significance of the choice of the tax base It is important for the excise tax to be applied to the base that leads to the greatest possible effect on price and revenue. For specific taxation, the tax base is the quantity of tobacco products. The quantity of cigarettes, cigars and bidis is measured in number of sticks; for other tobacco products, such as smokeless tobacco or roll-your-own (RYO), it is measured in the weight of the tobacco. When the tax is ad valorem, the choice of the tax base is important not only for health considerations – due to its effect on consumption – but also for tax revenue generation and industry profits. An ad valorem tax that is based on the ex-factory price (or CIF value) provides tobacco manufacturers with opportunities to reduce their tax liability, especially when they control the distribution system. Tobacco producers may sell cigarettes to distributors who are related parties at a reduced price, which then serves as the basis for calculating their ad valorem tax liability. Distributors, however, can then set high prices and share the extra profit with the producers (46). Because of the potential for such trade mispricing, the best practice is to use the retail price as the tax base and introduce a minimum excise tax per pack. Data in the 2019 WHO RGTE (1) show that, on average, the price level of a pack of cigarettes and the excise level are both much higher in countries that use retail price as the base for their ad valorem excise (Fig. 2.8). The maximum retail sales price, which includes all taxes, is used as the ad valorem tax base in the EU. That price also forms the tax base for ad valorem taxes in a growing number of LMICs, including Brazil, Egypt, Thailand, Turkey and Rwanda. 8 Indonesian Ministry of Finance, personal communication, 2017. 28 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.8 Weighted average price of the most-sold brand of cigarettes in countries that use retail price as the base for their ad valorem excise, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 7 The base on which the excise is applied is important. For specific excise, the base needs to be clearly defined (for cigarettes, cigars and bidis, it is the number of sticks; for other tobacco products, such as smokeless tobacco or RYO, it is the weight of tobacco). For ad valorem excise – where the base is typically either retail price, CIF value or producer price – evidence suggests that countries that apply the excise tax on the retail price of cigarettes tend to have higher prices than those that apply the tax on other bases. CIF and producer prices are difficult for government authorities to ascertain and are prone to undervaluation. The tax impact on final price: the significance of the minimum excise tax The use of a minimum excise tax in countries with ad valorem or mixed systems is another important factor in determining final price. On average, the price of a pack of cigarettes – as well as the excise level – is much higher in countries that impose a minimum specific excise than in those that do not (see Fig. 2.9). While more than half of the 47 countries that apply a minimum excise are members of the EU, removing EU countries from the average calculations produces the same conclusions. Ad valorem/mixed with retail price as base 3.72 PPP $ 5.01 PPP $ 6.41 Ad valorem/mixed with other base 1.79 Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 29 Fig. 2.9 Weighted average price of the most-sold brand of cigarettes in countries with and without a minimum specific excise tax, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 8 Among countries that apply an ad valorem or mixed excise tax on cigarettes, evidence suggests that those that impose a minimum specific excise tax tend to have higher prices than those that do not. The minimum excise tax also helps guarantee minimum excise revenues. Summarizing the advantages, disadvantages and impacts of the choice of excise tax structure for tobacco products Table 2.2 summarizes the characteristics of different types of tobacco excise taxes and the advantages and disadvantages of each type in relation to its impact on quantity demanded, perceived quality of brands offered, price, certainty and stability of revenue, administration and enforcement and opportunities for tax avoidance and tax evasion as they are predicted by the economic theory of imperfect competition and observed in real life. Ad valorem/mixed with minimum speci c 1.80 4.23 Ad valorem/mixed without minimum speci c Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes PPP $ 4.82 PPP $ 7.41 30 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 2.2 Characteristics of different types of tobacco excise taxes Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) TA X B A SE The unit of product (e.g. 1 000 cigarettes) The value of the product (e.g. retail, wholesale or manufacturer price) Excise is calculated on an ad valorem basis; however, if the calculated tax falls below a specified minimum amount, a specific tax rate applies Unit and value of product Both unit and value, unless the calculated tax falls below a specified minimum, in which case the tax base is the unit A D M IN IS TR AT IV E R EQ U IR EM EN TS The tax should be collected at the point of manufacturing or at the time of importation Low, as only the volume of the products needs to be ascertained Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as with a pure ad valorem regime Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises, as well as minimum specific excise tax compliance U N D ER VA LU AT IO N Not an issue Susceptible to undervaluation Provides an easy tool to prevent undervaluation of low-priced brands subject to the minimum specific excise The ad valorem part of the excise collection may be susceptible to undervalua- tion, depending on the choice of tax base The minimum specific excise prevents possible ad valorem tax base undervaluation of low-priced brands CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 31 Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IM PA C T O N P ER C EI V ED PR O D U C T Q U A LI TY Upgrading effect tends to reduce the relative tax on higher-priced brands Multiplier effect provides a disincentive to costly so-called quality improvement No incentive to upgrade higher- priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands IM PA C T O N P R IC E Tends to lead to relatively higher prices, particularly for low-priced cigarettes Tends to lead to relatively lower prices; price reductions will be subsidized if the multiplier effect is strong Tends to lead to relatively higher price increases for low-priced cigarettes An increase in the specific tax will to lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem payment if the base of the ad valorem includes excise An increase in the specific tax will lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem tax amount if the base of the ad valorem includes excise. Increases in the ad valorem and /or specific tax will raise the minimum tax paid if the minimum is a percentage of the total tax on, for example, weighted average price; they will reduce price gaps, given impact on perceived quality 32 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IN FL AT IO N The real value of the excise tax will be eroded unless the tax is adjusted in line with inflation The real value of the excise tax will be preserved as prices increase, at least to the extent that tobacco product prices follow inflation The real value of the minimum specific excise will be eroded over time unless the excise is adjusted in line with inflation The real value of the specific excise will be eroded unless the excise is adjusted in line with inflation The real value of the specific excise and the minimum specific excise will be eroded unless the excises are adjusted in line with inflation H EA LT H B EN EF IT S Will discourage consumption of tobacco products irrespective of the price band May encourage more trading down in favour of cheaper cigarettes, reducing the health benefit The minimum specific excise reduces incentives for trading down May reduce trading down Reduces trading down Source: (47). 2.2.2 OTHER TAX DESIGN CONSIDERATIONS The significance of automatic adjustments and indexation of specific tax to inflation Specific taxation does not depend on price and therefore, unlike the ad valorem tax, is not automatically adjusted for inflation. The real value of a specific tax is eroded over time as the price of the taxed product increases. Therefore, especially in countries with rapid growth in inflation, the nominal value of the specific tax must be increased regularly in order for the tax to maintain its real value. This is of great importance for both public health and public revenues, especially in countries where manufacturers do not increase prices regularly and/or low-priced tobacco products are the dominant products in the market. Table 2.3 lists countries that include automatic adjustments to their excise in order to avoid the erosion of the specific excise over time, using different units of adjustment and based on different frequencies. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 33 Table 2.3 Countries that include automatic adjustments to the specific excise COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Argentina Inflation (consumer price index [ CPI]), on a quarterly basis Armenia Minimum specific excise set to increase in the Tax Code by 15% on average between 2019 and 2021 Australia Wages – excise rates on tobacco and tobacco products increase in March and September each year, based on average weekly ordinary time earnings Bosnia and Herzegovina Specific excise rate is increased annually by at least 7.50 convertible marks per 1 000 cigarettes; minimum excise tax is increased annually to be at least 60% of the weighted average price Canada Inflation – federal tobacco tax rates are to be increased every five years, indexed to Canada’s CPI starting in 2019 Chile Inflation Colombia Specific tax set to 1 400 pesos, increased to 2100 pesos in 2018; starting in 2019, it will increase yearly by the CPI plus 4 points Costa Rica Inflation Dominican Republic Inflation, on a quarterly basis France Increase from 2017 to reach an average price for cigarettes of €10 per pack by 2020 Honduras Inflation, annually to December of the previous year Italy Minimum tax burden calculated every year in March on the basis of the weighted average price of cigarettes sold in the previous year New Zealand Inflation annually plus 10% annually from 2017 to 2020 Nicaragua Updated annually as of 1 January 2017, taking the highest among the annual devaluation of the official exchange rate of Cordoba with respect to the US dollar, published by the Central Bank of Nicaragua, and the annual inflation rate of the CPI published by the National Development Information Institute, observed in the last 12 months available North Macedonia Specific and minimum specific rate increase by 0.2 denars per cigarette on 1 July each year until 2023 Philippines Agreed tax increases and rates for specific excise tax between 2020 and 2023, with a 5% indexation thereafter Romania Inflation, annual (1 January) adjustment of the total excise according to inflation calculated on 1 October of the previous year Serbia Inflation, every six months Southern African Customs Union (SACU) – Botswana, Eswatini, Lesotho, Namibia and South Africa Inflation, on an annual basis9 Sweden Inflation 9 While the adjustment is not strictly automatic in the SACU, it is greatly informed by the inflation rate. The Treasury has some discretion. In recent years, the increases have typically been slightly above inflation. 34 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Turkey Producer Price Index Ukraine Agreement to increase the specific component 20% annually between 2019 and 2025 United Kingdom Increase by 2% above the retail price index (measure of inflation) for the 2015–2020 Parliament Source: (1). KEY TAKEAWAY 9 To avoid erosion of their specific excise tax, countries need to regularly – and, ideally, automatically – adjust the excise to inflation. The significance of automatic adjustments and indexation of specific tax to income growth In addition to the risk of erosion due to inflation, the effect of a (specific) tax can be significantly reduced if the tax is not adjusted for increases in consumer income. Income growth makes products more affordable – thereby encouraging consump- tion – especially in countries with rapid income growth. Australia is one of the rare countries that explicitly adjusts its specific excise rates according to wage growth (see Table 2.3). However, a number of countries have adopted automatic adjustments that are higher than inflation and sometimes largely cover income growth as well (see also Table 2.3). Adjusting tax for income growth contributes to increases in prices that make tobacco products less affordable (see section 2.2.3). KEY TAKEAWAY 10 The specific excise tax needs to be adjusted to reflect income growth so that tobacco products do not become more affordable over time. Measures for specific contexts: the role of pricing and other non-tax regulation Emerging evidence indicates that the tobacco industry finds ways to mitigate the impact of higher taxes on prices. For example, despite the heavy reliance on specific taxation in the United Kingdom, a price differential between premium and cheap cigarettes still exists. There is evidence that the tobacco industry does not always pass tax increases on to cheaper products (37, 48). Differential shifting among price categories is also observed in the EU (49–50), New Zealand (38) and the United States (51). Therefore, the public health community has suggested that pricing regulation could be considered as a method of eliminating inexpensive tobacco products that are often used by the young and the poor (52). Three types of pricing regulation are described below: minimum mark-up, price floor and price ceiling. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 35 Pricing regulation • Minimum mark-up – It has been argued that a minimum mark-up of wholesale tobacco prices could be a better strategy to raise prices of tobacco products than excise tax increases. Minimum mark-up laws aim to discourage the sale of products below an assessed cost by imposing a mark-up to the cost declared at different levels of the supply chain. Some studies in the United States have shown that minimum mark-ups do not increase average cigarette prices (53–54). However, a recent study of the impact of minimum mark-up/ price laws has shown that these laws are linked with higher prices, especially for the cheapest brands, and could be used as an effective tool to mitigate the impact of the industry’s price-reducing promotions (55). Another concern related to minimum mark-ups is that they can be manipulated by manufac- turers and are likely to lead to higher profits for the industry, as well as extra administrative costs for the government (56). • Price floor – A few studies suggest that setting a price floor, or a minimum price, is an alternative strategy for increasing tobacco taxes, particularly with respect to reducing health inequities (57–60). A price floor, imposed by the government or as a vertical restraint imposed by the supplier upon retailers, is a price that firms cannot legally undercut. Governments impose price floors to restrain unfair competition or, in the case of services, to increase quality. It is difficult, however, to find the right floor or to anticipate unintended conse- quences or an industry’s adjustments. A study in Malaysia, where a minimum price for cigarettes was imposed in 2010, found that the policy did not seem to have a meaningful impact on prices: licit brand prices remained well above the minimum price, while illicit brands remained well below it. This outcome may be a result of the floor being set too low or the proportion of illicit trade being high, either of which would reduce the effectiveness of the policy (52). In the EU, imposing minimum retail sale prices for cigarettes could be a breach of harmonized legislation concerning the internal market, as minimum prices would distort competition. Therefore, increasing minimum excise duties is recommended instead, to discourage consumption (61). Increasing the minimum excise duties would also result in the additional revenue going to the governments instead of contributing to industry profits. A price floor would probably lead to increased industry profits – giving the industry greater funds for its marketing strategies – and lower tax revenue for governments, reducing their ability to cover costs associated with tobacco use. By reducing price competition, the price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifica- tions). Competition among firms may prevent them from raising their prices, 36 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N but a government that imposes a price floor does this for them. Minimum pricing is likely to create windfall profits for manufacturers and retailers. It can even help manufacturers sustain a cartel. If the industry uses the money to increase promotions, advertising or grant proposals for related research, this could undo some of the potential benefits of the policy.10 Some recent evidence shows that, at least in the case of the United Kingdom, increased concentration of power among a handful of multinational corpora- tions is enabling them to undermine tax increases through increased price segmentation, and that requiring minimum prices might be a good way to address the problem. A longitudinal analysis of price data from the United Kingdom (48) has shown that despite regular excise tax increases over time, average real prices for cheaper segments of the tobacco market (in this case, cigarettes and RYO) did not increase – indicating an undershifting of the tax increases in those segments that resulted in increased sales volume. At the same time, average prices for more-expensive market segments increased, indicating overshifting of the tax increases that resulted in decreased sales volume. This industry strategy ensures that the most price-sensitive consumers remain addicted, while encouraging initiation and discouraging cessation. Furthermore, segmenting the market further by overshifting the tax increase on premium brands while undershifting it for cheaper brands mitigates the impact of declining consumption resulting from higher taxes while increasing overall industry margins and profitability. Another situation where setting minimum prices can be a useful policy is specific to the United States. Banning marketing and promotions11 is not possible under the freedom of expression protections of the Constitution of the United States (Amendment I), and it was estimated in 2008 that more than 82% of all advertising and promotional spending by the tobacco industry was focused on reducing the price of their products at the point of sale (62). This limitation on how government can set policy has paved the way for the implementation of minimum price policies in many states and cities to counter the detrimental impact of price promotions on consumption and on the tax policy itself. Huang et al. (55) found that the presence of minimum price laws was associated with higher cigarette prices. They also noted that cigarette prices were even higher than prices resulting from minimum price laws in states that also prohibit industry from engaging in other price-reduction strategies, 10 See, for example, the PMI strategy of setting up the Foundation for a Smoke-Free World and grant proposals for related research. 11 See section below on banning promotional discounts for tobacco products for further discussion about marketing and promotions. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 37 such as below-cost combination sales, using trade discounts to reduce the base cost of cigarettes and distributing below-cost coupons to consumers. In the contexts of both price segmentation and price promotions, the imposi- tion of minimum prices as a complementary policy to excise tax increases – not as a policy alternative – may help guarantee that taxes do indeed lead to the intended reduction in consumption. Nonetheless, more evidence is needed to support the effectiveness of this policy. • Price ceiling – Concerns about differential tax shifting have led to suggestions that a price cap may benefit public health by limiting the tobacco industry’s ability to reduce average prices by differentially shifting tax increases among various price segments (63–64). Because tobacco manufacturers operate across international markets, however, they could maintain low prices in one coun- try but maintain overall profitability by selling more premium products in another country. Additionally, limiting price increases does not fit the public health purpose of reducing consumption. It is worth noting that maximum retail prices are sometimes used as a base for calculating the ad valorem tax payments in countries with ad valorem or mixed tax systems. In a systematic review of the literature on non-tax policy approaches to raising prices, Golden et al. (63) hypothesized how such policies would influence price dispersion and average prices. Their study found that minimum price policies combined with promotion bans have the potential to increase average prices. This is, of course, relevant in a context where price promotions are present. From either a theoretical or a practical standpoint, however, it is clear that price policies cannot be used alone and should always be considered as complements to excise tax increases. Significantly increasing taxes is the most effective way to dissuade consumption, correcting whatever bias may exist. Significant tax increases also provide the added benefit of raising money for the government rather than profits for the tobacco industry. Nonetheless, a minimum price might help narrow the gap between cheap and pre- mium cigarettes when applied to all tobacco products to avoid product substitution. Other non-tax regulation • Banning promotional discounts for tobacco products – The sale of tobacco products at a discount rate – such as through reduced-price coupons or buy-one-get-one-free offers – encourages consumption and undermines tax increases. Such practices should be completely banned. They often exist outside the realm of the finance sector because they are considered a type of marketing – promotional discounts are usually addressed in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. 38 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N According to the 2019 RGTE (1), 118 countries out of 195 with all levels of income had such a provision implemented as of 31 December 2018. • Banning the sale of single sticks of cigarettes – Article 16 of the WHO FCTC, “Sales to and by minors,” paragraph 3, requires Parties to “prohibit the sale of cigarettes individually or in small packets which increase the affordability of such products to minors”. Some smokers opt for buying single sticks partly because of the lower im- mediate costs of buying cigarettes individually (65). Internal (unpublished) analysis of single-stick prices collected by WHO for the 2012, 2014 and 2016 editions of the WHO RGTE shows that, in fact, the aggregate price of 20 single sticks of cigarettes sold separately is generally higher than the price of a 20-cigarette pack sold in the market of a specific country. Despite this fact, single-stick sales – and sales of small-sized packs – make cigarettes accessible to consumers with limited disposable income. De Ojeda (66) found in a study conducted in Guatemala that single-cigarette sales are associated with increased cigarette accessibility for less-educated, lower-income populations and minors. Single-stick sales are also a feature of many markets in South-East Asia, including most notably Bangladesh and India, but also in other parts of the world, e.g. South Africa. Single-stick sales also reduce the impact of a tax increase, since the in- crease per stick is much smaller than the increase per pack (67). In a study investigating how smokers in New York City responded to a tax increase of US$ 1.25 per pack in 2008, Coady et al. (68) found that 15% of smokers bought more single cigarettes than they had previously.12 By allowing single stick sales, governments risk losing part of the ad valorem taxes if the tax base is the retail selling price; the retail price of single sticks is much more difficult to monitor than the retail price of packs of cigarettes, on which, for example, tax stamps with prices can be applied. An internal WHO analysis of the most recent tobacco control laws in 2018 in 195 countries found that 86 countries impose by law a ban on the sale of single sticks of cigarettes (36% of the countries are high-income, and 64% are LMICs). In addition to banning the sale of single sticks of cigarettes, 67 of the 86 countries specify a minimum size for packs of cigarettes. Most (52 countries) use the 20 cigarettes per pack standard, but minimum sizes 12 Before 2018, the use of single sticks was possible, but it has since been banned. See New York City Administrative Code. chapter 7: regulation of tobacco products, subchapter 1: Tobacco Product Regula- tion Act, §17-704.a-1. New York: New York Legal Publishing Corporation; 2020 (http://library.amlegal. com/nxt/gateway.dll/New%20York/admin/title17health/chapter7regulationoftobaccoproducts?f=tem plates$fn=default.htm$3.0$vid=amlegal:newyork_ny$anc=JD_T17C007, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 39 range from five sticks in Ghana up to 25 sticks in Papua New Guinea. Another 15 countries do not specifically ban the sale of single sticks but do specify the minimum size for packs of cigarettes. According to the WHO FCTC, in order to reduce affordability, single stick sales should be banned and a minimum number of cigarettes contained per pack should also be defined. KEY TAKEAWAY 11 A number of non-tax measures are closely connected to tax policies, including price regulations, bans on promotional discounts for tobacco products and bans on the sale of single sticks of cigarettes. The price policies discussed are (1) minimum mark-up, (2) price floors and (3) price ceilings. Current evidence does not yet demonstrate that minimum mark-ups and price floors lead to increases in average price. Nonetheless, they may be relevant in some specific contexts as complementary policies to excise tax increases. Price ceilings limit price increases, which can mitigate their impact on consumption. Price marketing strategies such as promotional discounts and the sale of single sticks undermine the effect of tax policies and should be banned. A minimum pack size should also be required by regulators. Tax increases and their possible impact on inflation At times, the inflationary impact of tax increases on cigarettes and other tobacco products is raised as an argument for not increasing these taxes. This may be a concern in countries where wages and/or a significant share of government spend- ing is indexed to inflation (e.g. for public pension payments) or where government policy is to keep inflation low. The extent to which tobacco product tax increases lead to increases in inflation depends on several factors, most notably the share of these taxes in prices and the weight tobacco prices are given in computing a price index. For example, if taxes account for 25% of tobacco product prices, a doubling of the tax (100% increase) will increase prices by 25%. If the weight given to tobacco products in the price index is 3%, the index will rise by 0.75% in response to the tax increase. As tobacco taxes account for a larger share of tobacco product prices, the inflationary impact of a tax increase will be greater. Similarly, as tobacco products are given more weight in computing a price index, a given tax increase will have a greater inflationary effect. In general, for most countries, the inflationary impact of tobacco product tax increases would be relatively small (47). 40 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Consumer price indexes have multiple purposes. They are an important economic indicator for most countries and are often a key determinant of monetary policy. Inflation rates have a direct impact on interest rates and exchange rates. In many countries, changes in wages, social security benefits and other payments are tied to inflation as measured by a price index. Price indexes are used to provide more accurate comparisons of changes in expenditures, incomes and prices for specific goods over time, as well as to allow comparisons across countries. Given the many uses of consumer price indexes and the potential inflationary impact of tobacco tax increases, some governments have developed alternatives that exclude tobacco (and sometimes other goods) for some uses. For example, since 1992, France has excluded tobacco products from the price index used for adjust- ing minimum wages (47). However, many countries continue to include tobacco product prices in their consumer price indexes. Excluding tobacco products from the basket of goods used in developing key price indexes would greatly reduce concerns about their impact on inflation. In addition, with declining consumption of tobacco products, the inclusion of their prices in key price indexes results in a distorted measure of price for many consumers. KEY TAKEAWAY 12 If governments are concerned about the potential inflationary impact of a tobacco tax increase because wages or some government spending may be tied to a price index, they can use a price index that excludes tobacco products. The importance of taxing cigarettes and other tobacco products in a comparable way While cigarettes are the most commonly used tobacco product globally, other tobacco products are as prevalent and sometimes more prevalent than cigarettes in some parts of the world. Bidis and smokeless tobacco are the main products consumed in some countries in South-East Asia – Bangladesh and India in particular – and waterpipes are widely used for smoking tobacco in the Eastern Mediterranean region (4). These products, as well as RYO, have historically been taxed much less than cigarettes (see, for example, Fig. 2.10 for Bangladesh and India, where the excise tax and prices of bidis and smokeless tobacco are much lower than those for cigarettes). This differential taxation undermines the health impact of excise taxes on tobacco products because (1) it encourages users to switch from cigarettes to the lower-taxed product (see the case of Thailand below); (2) it is not effective in reducing tobacco use in general, especially if the most widely used product in the country is not cigarettes; (3) it can encourage tax avoidance by companies that may CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 41 redefine products that are similar to cigarettes so that they fall within the lower- taxed product category (see the EU example below) and (4) it reduces the size of government revenues since those products could have been taxed at higher levels. Fig. 2.10 Price and tax of the most-sold brand of cigarettes, bidis and smokeless tobacco in Bangladesh and India, 2018 Source: (1). In Thailand, for example, the price of cigarettes has been raised quite successfully through taxation over a number of years, while taxes and prices of loose or RYO tobacco have until very recently remained unchanged. Indigenous tobacco used for RYO cigarettes has historically been exempt from excise, while foreign tobacco was taxed at a low level relative to that of manufactured cigarettes. Consequently, Thailand experienced growth in the RYO market even though cigarette consumption had been falling.13 The Thai government eventually took strong action to address this issue. First, the exemption for indigenous tobacco was removed in 2018. The Cabinet then approved an increase in the excise rate on small producers (of indigenous tobacco) from 0.005 baht per gram to 0.025 baht per gram in 2020, with another increase to 0.1 baht per gram scheduled for 2021 (69). In the EU, the minimum excise duty levels for cigars and cigarillos is significantly lower than that for cigarettes. The Member States of the EU are required to levy an 13 WHO Country Office for Thailand, personal communication, 2019. 12.8 0 190 54.9 64 24 Price Excise tax amount 80 12.8 3.8 44 Lo ca l C ur re nc y 0 50 100 150 200 Cigarettes (20 sticks) Bangladesh, taka India, rupees Bidis (20 sticks) Bidis (20 sticks) Cigarettes (20 sticks) Smokeless (20 g) 42 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N excise duty of at least €90 per 1 000 cigarettes, which should be 60% of the weighted average retail selling price of cigarettes released for consumption. For cigars and cigarillos, only €12 per 1 000 items, or an excise duty of 5% of the retail selling price, is required. As a result, the excise tax share on cigarettes is much higher in many EU countries than the share for cigars and cigarillos. In response, some companies started to market so-called borderline cigarillos. These products have characteristics similar to cigarettes but can be sold at a lower price because for excise purposes, they are considered as cigarillos. Although this issue seems to be largely solved by amendments to the definitions of these products at the EU level and a change in tax structures in some countries, it is important to be aware of the unintended incentives that can be created by large gaps in excise tax levels between product categories (70–71). For more details on industry tactics to undermine tax increases, see Box 2.1. KEY TAKEAWAY 13 To make excise tax on tobacco products more effective in reducing overall tobacco use and to avoid substitution between products, all tobacco products need to be taxed in a comparable way. The Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce the demand for tobacco) (73) recommend that all tobacco products should be taxed in a comparable way. Box 2.1 Industry tactics used to undermine tax increases Tax increases reduce the demand for tobacco products and present a threat to the tobacco industry’s high profits. The industry responds by using various strategies (17, 46, 48), including the following (46): Stockpiling (forestalling/front-loading) – Before the implementation of an an- nounced tax increase, manufacturers overproduce tobacco products, paying the pre-tax-increase rate. As a consequence, sales and tax revenue decline immediately but temporarily after the tax increase (while sales and revenues had increased sub- stantially just before the tax increase) and the industry attributes this drop in revenue to the emergence or increase of illicit trade. This practice results in tax avoidance if there is no law prohibiting it (see also the discussion on anti-forestalling in Chapter 3). Changing certain product characteristics (for example, weight or length) and/ or adjusting the production process – When tobacco products are taxed at different rates or are subject to different tax increases, the industry can, for example, re-label one type of tobacco product as another product that has a lower tax burden (as in the example of the EU above). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 43 Choosing the time of a price increase announcement strategically – The industry may raise prices in anticipation of a tax rate increase, generating extra profits in the period until the tax is actually implemented. When the tax increase is implemented, consumption and tax revenue will fall, but prices will not change, so the industry can claim that the tax policy was ineffective in reducing demand. Adopting price-discriminating strategies or price-related promotions – The industry may offer discounts, retailer rebates or added value (gifts) to tobacco purchases to minimize the loss of price-sensitive consumers. This, however, is not possible in countries where strict bans on tobacco advertising, promotions and sponsorship are implemented. Using brand proliferation (for example, launching a low-priced brand) and price segmentation – Manufacturers can choose to reduce prices of certain brands or introduce new, even cheaper ones to keep price-sensitive consumers in the market. There is evidence that firms introduce new cheaper products and use price-marking – printing the price directly on packs of tobacco products – to lock in their price (48). Such practices compromise both public health and revenue objectives. Differential shifting of tax increases across different price segments, depending on the market circumstances – The industry may increase the price of a product by more than the amount of the tax increase (tax overshifting) and blame the govern- ment for the total increase. Tax overshifting is profitable when demand is inelastic, that is, when the price increase more than offsets the reduction in sales. The industry may overshift the tax increase for higher-priced brands, which are expected to be more price inelastic than lower-priced brands. Additionally, to keep price-sensitive consumers in the market, the industry may temporarily absorb part (or all) of the tax increase on lower-priced brands. The differential tax shifting will lead to different responses in the demand for the different brands (37, 48). Lobbying government to distort interventions – Government policy might be influenced by tobacco industry lobbying, directly or indirectly. Policy-makers are not simply welfare or revenue maximizers; they also value political support. Industry lobbying might lead to adopting a favourable type of taxation, postponing tobacco tax increases or distorting the tax rate downwards (17). Article 5.3 of the WHO FCTC, “On the protection of public health policies with respect to tobacco control from commercial and other vested interests of the tobacco industry”, and its guidelines provide useful guidance on how to address tobacco industry interference. In fact, all 181 countries that are Parties to the WHO FCTC have a legal obligation to implement the requirements of Article 5.3. Having correct expectations about industry responses is important for estimating the impact of a tax increase on consumption and tax revenue. 4 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.2.3 MEASURING IMPACT AND RECOMMENDED INDICATORS Governments need to consider a number of indicators when formulating policy changes. Inappropriate assumptions about consumer behaviour, market structure and industry behaviour can lead to faulty policy analysis. Measuring impact on price and demand Consumption habits, local traditions and industry characteristics – such as the number of different brands offered, the possibilities of cross-border shopping and the presence and level of illicit trade – all affect the shape of the demand and supply of tobacco products, thereby determining the value of the price elasticities. Price elasticity, together with the industry’s pricing strategies – for example, the degree of tax shifting – and the tax share in the retail price, determine the elasticity of the tax base, regardless of whether the base is determined by quantity (for specific taxation) or transaction value (for ad valorem taxation). The importance of elasticity estimates Different types of elasticity should be considered: • price elasticity – own-price elasticity – measures the response of consumers’ demand for a product following a change in the price of the product. – cross-price elasticity – measures the response of consumers’ demand for a product when the price of another product changes. Cross-price elasticity can also occur between different brands or price segments for the same product. • income elasticity – the response of consumers’ demand for a product when their income level changes. Correct estimates of price and income elasticities are important for policy-makers who need to anticipate the impact of a tax increase on consumption and tax rev- enue. Estimates will vary depending on a number of factors, including whether responses are considered in the short run versus the long run, the functional form of the demand function used, whether factors such as addiction or tax evasion are accounted for and the way data are constructed. For example, details such as the degree of aggregation of data, whether gender- or age-specific data are used, the time span covered and which estimation procedures are used (e.g. ordinary least squares, two-stage least squares or generalized method of moments) will all affect the results of the estimate (72). Price elasticities may change over time, as well because of changes in any of the other factors affecting demand, such as income or tobacco control measures, and also because of changes in estimation techniques and the types or sources of data used. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 45 Moreover, what is of most interest is the price elasticity of total demand. A tax increase may reduce tax-paid retail sales but not necessarily total consumption. For example, smuggling can significantly bias price elasticities when the elasticities are estimated using legal sales data; not accounting for possible illicit trade might lead to overestimation. Similarly, when cross-border shopping is included, the price elasticity of demand is lower (in absolute value) (74). Estimating the total price elasticity of demand for legal and illegal consumption can be done by using cross- sectional data from nationally representative household surveys. However, this approach also has its weaknesses. For example, respondents tend to underreport their consumption of tobacco, which leads to bias in the size of demand. Price endogeneity14 is another technical problem that can be challenging to address. To comprehensively estimate the total effect of a tax increase on demand for all tobacco products as well as on tax revenue, the degree of substitutability between them needs to be estimated (55). Cross-price elasticity measures how the quantity demanded of a particular tobacco product changes when the price of another tobacco product increases. When this elasticity is positive, the products are substitutes; the higher the value of the elasticity, the closer substitutes the products are to one another. For example, positive cross-price elasticity between RYO and manufactured cigarettes implies that the demand for RYO increases as the price for cigarettes increases. Substitutability may also arise between different cigarette brands – when the relative price of economy brands increases, demand for premium brands may increase. This effect can be exacerbated when differential (tiered) taxation is ap- plied on different types of cigarettes, further widening the gap in prices between brands and segments and encouraging substitution. The substitutability between traditional and new and emerging tobacco and nicotine products is currently of great interest (see section 2.4 below). In some countries, different tobacco products can also be complementary rather than substitute goods. This means that when the price of a tobacco product increases, the demand for its complement drops because users are unlikely to use the complementary tobacco product alone. For example, a number of studies have found manufactured and indigenous bidi cigarettes to be complementary goods in India (75–76). The sign and magnitude of income elasticity vary across time, countries and demographic groups. For example, in the United States, a high-income country, income elasticity over time has changed from positive to negative, and cigarettes have switched from being a normal good to an inferior good (77–78). On the other hand, among LMICs, where prevalence of smoking tends to be relatively higher, 14 Price is endogenous because it is not an independent variable: it is estimated by dividing expenditure on tobacco by consumption of tobacco, with consumption being a dependent variable in the estimation of price elasticity. 46 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarettes might still be a normal good, with consumption increasing as income levels increase (positive income elasticity) (79–80). There are unobservable characteristics that differentiate higher-income smok- ers from lower-income smokers, such as differences in time and risk preferences, differences in associating a social stigma with smoking and differences in taste for smoking as a pleasurable activity. When these characteristics are ignored, estimates of the correlation between income and smoking-related outcomes are biased. Kenkel et al. (81), using techniques that estimate the causal effect of income on smoking among low-income adults, found that tobacco is a normal (even a luxury) good: higher income is associated with a higher probability of smoking participation and a lower probability of smoking cessation. These results are consistent with those regarding the impact of the business cycle – periods of expansion or recession in economic activity – on health be- haviour and outcomes. Ruhm (82–83), for example, found that smoking declines during temporary economic downturns and increases during economic expansions. Tarantilis et al. (84) found that estimates of income elasticities of demand in Greece were higher after the economic crisis of 2010 than before it. The financial crisis and the austerity measures shifted the demand for cigarettes downwards and turned cigarettes into a more income-elastic good. Interestingly, evidence from Germany suggests that the propensity to become a smoker significantly increases during an economic downturn. However, among those who are already smokers, cigarette consumption actually decreases (85). Ideally, when estimating price and income elasticities, the effect of non-price policies should also be accounted for. A recent study from South Africa shows that failing to take non-price policies into account will overstate the price effect (86). The NCI/WHO Monograph (4) suggests that price elasticity of demand for tobacco is on average -0.4 in high-income countries (ranging from -0.2 to -0.6). Estimates for LMICs are more variable, clustering around -0.5 (ranging from -0.2 to -0.8). A price elasticity of -0.5 means that a 10% increase in price would lead to a 5% reduction in consumption. KEY TAKEAWAY 14 Policy-makers need to know the elasticity of demand – including price elasticity (own-price and cross-price) and income elasticity – for tobacco products in their country in order to correctly assess the impacts of potential policy changes on consumption and subsequent revenues. These estimates need to be made on a regular basis to capture changes in demand over time. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 47 The importance of the tax base elasticity The tax base elasticity measures the sensitivity to a change in the tax rate of the base on which the tax is imposed – the base being tobacco consumption in the case of specific taxation and tobacco expenditure in the case of ad valorem taxation. The magnitude of the elasticity of the tax base depends on price elasticity of demand, the tax structure, the level of the tax rate and its share in price, along with the industry response through its decision to absorb, pass through or overshift the tax on to the retail price. Consumers’ preferences and income, the availability of substitutes and other non-price tobacco control measures also influence the tax base elasticity, essentially through the price elasticity of demand. The magnitude of the elasticity of the tax base also depends on social motivations, including price and tax expectations, which are ultimately impacted by successful tobacco control measures that affect consumers’ willingness to pay taxes or prices. In addition, the tax-base elasticity depends on smokers’ perceptions of the prob- ability of detection and tax enforcement when using illegal products, as well as the availability and accessibility of opportunities for tax evasion and avoidance. Finally, consumers’ willingness to pay taxes depends on their perceptions regarding the use of the tax revenue (87). Therefore, the tax base elasticity is largely influenced by government policy choices. Increasing the tax share in prices is recommended by WHO as a tool to achieve the public health objective of reduced tobacco use: a higher tax share in prices increases the tax base elasticity, all else remaining constant, and therefore increases the reduction in the tax base through the resulting reduction in smoking. However, manufacturers can be expected to attempt to manipulate the tax base elasticity through their pricing policies, such as tax shifting. As discussed earlier in this chapter, industry behaviour is itself affected by government tax policy and regulations. A number of factors need to be taken into account when considering tax pass- through. As discussed earlier, tax is more likely to be overshifted within a specific tax structure than within an ad valorem structure. There is also evidence of industry overshifting the tax for premium or expensive cigarette brands while undershifting the tax for cheaper brands. This indicates that within a given market, the industry’s decision on the extent of tax pass-through will vary based not only on the tax structure but also on the structure of the market. It will also vary by brand. But this does not give an indication about the impact of the tax increase on the average price of a tobacco product. In the context of the tax base elasticity and the impact of tax increases on revenues, it is important to assess how tax increases affect average prices. The example of South Africa is very useful here. Over the past two decades, South Africa has been consistently increasing its specific excise tax on cigarettes, which has led to large price increases. An analysis of the effect of excise tax increases on 48 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarette prices between 2001 and 2015 (26) shows that while there is evidence of tax overshifting, at least within a specified period of time, overall, the tax has been undershifted in real terms. This undershifting is due partly to increased competitive- ness in the market and partly to the introduction of low-priced brands. Of course, this encouraged some consumers to downshift their consumption to cheaper products, but it also pushed manufacturers of more expensive brands to absorb part of the tax increase to reduce the impact on price. A change in the level of the tax rate – with all other factors that influence con- sumption held constant – will result in a change in the tax revenue.15 Estimates of tax base elasticity help governments predict changes in tobacco tax revenues following a tax increase (see details in Annex 2.2). Under specific taxation: • tobacco consumption – the tax base – is expected to be price inelastic (17, 47); • prices increase by less than the tax increase, on average (there is no tax overshifting overall); and • consumption – the tax base – is also expected to be tax inelastic: the quantity of consumption falls less than proportionately to the tax increase, and the tax revenue increases. Under ad valorem taxation: • the tax base is the total consumer expenditure (or, equivalently, the industry sales revenue) on (legal) tobacco consumption – that is, the tax base under ad valorem taxation is determined by both price and quantity, which is itself a function of price; • the sign of the tax base elasticity – which can be either negative or positive – depends on the magnitude of the price elasticity of demand; • since evidence suggests that tobacco demand is price inelastic, the tax base elasticity is positive; • when the ad valorem tax rate increases, both price and quantity adjust, but quantity falls less than proportionately to the price increase, and tax revenue increases; and • a tax rate increase leads to both a higher level of revenue and a lower level of consumption; the value of the elasticity – and hence the tax revenue – increases with the degree of tax shifting. 15 This concept has been used by Laffer to argue that tax increases that are too high will reduce excise tax revenues (the so-called Laffer curve). For a detailed discussion on the Laffer curve, see section 4.4 in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 49 Taxation serves as an instrument for achieving both fiscal and public health objectives. If, after successful tobacco control interventions, prices reach levels where demand becomes elastic, the tax base is still most likely to be inelastic due to tax undershift- ing, since overshifting is not a good pricing policy when demand is elastic. In other words, a tax rate increase – in combination with non-price tobacco control measures that make consumers more sensitive to price (tax) increases – leads to decelerating but still positive marginal revenues. For an example of a tobacco taxation success story, see Box 2.2. For further details of countries’ experiences with tax increases and their impact on revenues, see section 4.4 in Chapter 4. KEY TAKEAWAY 15 Policy-makers’ key policy tool to control demand is tax. Therefore, it is essential they assess not only the impact of price on demand but, more appropriately, the impact of tax on demand: this is the tax base elasticity. The tax base elasticity is essentially determined by (1) the price elasticity of demand, (2) the degree to which the industry will pass the tax on to the retail price and (3) the tax as a share of the retail price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. Box 2.2 A tobacco taxation success story: Turkey Turkey is an example of a country that has been increasing taxes regularly and sig- nificantly over a relatively short period of time and has reaped the benefits of this policy. As shown in Fig. 2.11, the excise tax per pack of cigarettes more than doubled in real terms over 10 years, with the real price almost doubling as well. In parallel, tobacco excise revenues increased by 67% and cigarette sales decreased by 20%. Since the beginning of the country’s Health Transformation Program in 2003, Turkey has successfully increased public health spending and collected more tobacco tax revenue. According to the latest available figures, in 2015, tobacco tax revenue was equivalent to 42% of the country’s public health expenditure and 1.5% of GDP (88). 50 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.11 Tobacco excise revenue and consumption in Turkey (base year 2008), 2008–2018 Sources: Reference 1 for the price of the most-sold brand, Ministry of Finance for the sales and revenue data and IMF world economic outlook, April 2020. See https://www.imf.org/en/Publications/WEO/weo- database/2020/April for the adjustment for inflation. Impact on affordability While price increases clearly have an impact on consumption, when the effects of increasing per capita income of a population are not considered, the price impact may not be as strong as expected. Increases in a population’s income also increase its purchasing power. And, as indicated earlier, tobacco products generally behave like a normal good. Consequently, as income increases, it is expected that tobacco consumption will increase as well. To mitigate this effect, price increases (following tax increases) need to be greater than increases in income. This is where the concept of affordability comes in. Affordability examines the effects of both increasing prices and increasing incomes on consumer behaviour. A common and easy way to calculate affordability, made popular by Blecher and van Walbeek (89), is to use the percentage of GDP per capita required to buy 2000 cigarettes (or 100 packs of 20 cigarettes) in a given year. An increase in this proportion over time will indicate that cigarettes are becoming less affordable and should lead to reductions in consumption. Changes in trends in affordability of cigarettes over time help policy-makers understand how prices are evolving 0 1 2 3 4 5 6 Tu rk is h lir as p er p ac k N um ber of sticks, 10 000 000 Turkish liras 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real excise tax amount Real price Real tobacco excise revenues Cigarettes, per capita sales 1 100 500 1 700 2 300 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 51 relative to a population’s ability to purchase cigarettes and enable them to revise their policies accordingly. Recent studies in India, for example, highlight the wide price differential between manufactured cigarettes and indigenous tobacco products such as bidis and chewing tobacco, as well as the propensity for these indigenous products to become more affordable over time due to favourable or more lenient tax policies towards them (90–91). Figure 2.12 shows the change in affordability of a pack of the most-sold brand of cigarettes by country income group between 2008 and 2018. During this time period, affordability declined in almost 70% of high-income countries, while it declined in slightly more than 35% of middle-income countries and only 26% of low-income countries. Fig. 2.12 Number of countries that have experienced a change in affordability of cigarettes between 2008–2018, by income level Note: Change in affordability was computed as the least squares rate of change in the per capita GDP required to purchase 2000 cigarettes of the most-sold brand in local currency in a given year. The trend rate of growth was computed for countries with four or more years of data, including 2018. Affordability was assessed as not having changed if the least squares trend in the per capita GDP required to purchase 2000 cigarettes over the period 2008–2018 was not statistically significant at the 5% level. Source: (1). KEY TAKEAWAY 16 From a health perspective, in addition to examining the impact of a tax increase on the levels of price, demand and revenues, policy-makers should consider a tax hike that will lead to prices rising more than increases in their population’s income; a tax increase should make tobacco products less affordable to consumers so that demand will be effectively reduced. Cigarettes became less aordable Cigarettes became more aordable Aordability did not change Could not be assessed due to insucient data High income Middle income Low income 39 36 9 13 5 7 37 23 6 13 2 5 52 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Projecting impact on consumption, smoking prevalence and lives saved The WHO interactive smoking projection and target-setting tool (WHO ISPT) The WHO ISPT enables national policy-makers and tobacco control experts to explore the potential impact of proposed tobacco control policies. It uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. The WHO ISPT provides projections of (1) tobacco smoking rates and (2) tobacco-smoking-related deaths in a country under different policy settings and for different time periods. It was designed to promote multisec- toral collaboration within countries by enabling experts from various ministries (for example, health, education, finance, national statistics), civil society, academia and media to explore options for medium- and long-term tobacco control planning together with WHO experts. Use of the WHO ISPT enables strong partnerships for policy change advocacy, program development and evaluation.16 In particular, it can help policy-makers in the Ministry of Finance assess the specific contribution of tax policies – within overall tobacco control policies – towards achieving specific targets in tobacco prevalence reduction. Projecting impact on excise revenue The WHO tobacco tax simulation model (WHO TaXSiM) The WHO TaXSiM is a simple but data-intensive Excel-based tool that helps policy- makers analyse their tobacco tax policy and assess the impact of any excise tax increase or change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Using detailed data about the market – including the majority of brands found in the market, their market share and price levels and the applicable tax – and assumptions about price elasticity of demand, the WHO TaXSiM predicts the impact of tax changes on consumer prices, consumption volume and tax revenues generated by each brand and market segment for the following year. The exercise can be done for multiple years.17 By exploring market data in detail, in addition to assessing the potential rev- enue impact of changes in excise tax, the WHO TaXSiM is a useful instrument for highlighting weaknesses and opportunities in an existing tax system and market. It can also encourage policy-makers to create administrative databases that can be periodically updated to monitor the dynamics of the cigarette market. 16 The WHO ISPT is not available publicly, but WHO will work directly with interested countries upon request to use it to produce data-to-action-type plans. 17 For more information about the methodology, see https://www.who.int/tobacco/economics/tax- sim_background.pdf, accessed 29 September 2020. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 53 KEY TAKEAWAY 17 Policy-makers can use available tools to simulate the impact of tax increases on prices, consumption and revenues, as well as smoking prevalence and lives saved. Recommended indicators to monitor tobacco taxation progress MPOWER WHO publishes a biannual RGTE, which monitors global progress in tobacco con- trol. In particular, the report focuses on the implementation of the policy package MPOWER, a set of proven demand-reduction measures in line with the key provi- sions of the WHO FCTC (1). While raising taxes on tobacco (component R) is proven to be the most effective and cost-effective policy to reduce tobacco use (4), implementing the entire MPOWER package at the best practice level will reinforce the impact of R. For example, as mentioned earlier in this chapter, banning promotional discounts as part of the E measure (enforce bans on tobacco advertising, promotion and sponsorship) will favour price increases following a tax increase. If all the MPOWER tobacco control measures except R were implemented at the best practice level, all else remaining constant, revenues would be expected to decline. Thus, in order to maintain revenue levels, it is important to raise excise taxes on tobacco products regularly to compen- sate for the decline in tobacco use from the other four tobacco control measures. Tax share The main indicator in the R policy in the RGTE (1) is the total share of indirect taxes in the retail price of the most-sold brand of cigarettes.18 Countries whose most-sold brand of cigarettes has a total tax that is equal to or greater than 75% of the retail price are considered to be at the highest level of achievement. While total taxes include excise taxes, VAT (or sales taxes), import duties (when applicable) and other indirect taxes (where applicable), it is preferable to focus on excise taxes, since they are the component that most influences the relative price of tobacco. The share of excise tax in the retail price can be extracted from the RGTE database.19 The 2010 WHO technical manual on tobacco tax administration recommended making excise taxes account for at least a 70% share of excise taxes in the retail price of tobacco products (47). 18 For more details about how this indicator was compiled, see Technical Note III of the RGTE 2019 (https://www.who.int/tobacco/global_report/Technical-Note-III.pdf?ua=1). 19 See taxes and retail price for a pack of 20 cigarettes, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.1-Taxes-and-retail-price-for-a-pack- of-20-cigarette-most-sold-brand.xls?ua=1, accessed 29 September 2020). 54 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Affordability As discussed previously, the share of tax in the retail price is not enough to ensure that a policy will be successful in reducing demand. Any tax increase should lead to an increase in price that will effectively discourage consumption. While global trends indicate that a high tax share is positively correlated with a high price level (see Fig. 2.2 in section 2.1.2), this may not necessarily apply to a particular country; a tax share can be high, while at the same time tobacco products remain afford- able. For this reason, it is important to monitor not only tax increases but also whether those increases led to a price increase that is greater than income increases. As described in section 2.2.3 of this chapter, a common indicator is the percentage of GDP per capita required to buy 100 packs of 20 cigarettes in a given year.20 Other indicators As discussed in detail in sections 2.2.1 and 2.2.2, a good tax structure can make a tax policy more effective in increasing prices and decreasing affordability of tobacco products. Indicators can include whether a uniform excise is applied, whether it is a specific excise and whether it is adjusted regularly for inflation. A number of such indicators are also monitored through the RGTE and can be downloaded online.21 A tobacco tax indicator compiled in 2020 combines the various elements that form a good tobacco tax policy. The Tobacconomics Cigarette Tax Scorecard (92) rates a country’s tobacco tax policy performance based on best practices. The four components that determine the level of performance are (1) cigarette price (in PPP), (2) changes in the affordability of cigarettes over time, (3) the share of taxes (total and excise) in retail cigarette prices and (4) the structure of cigarette taxes (i.e. whether excise is applied; whether it is uniform or tiered; whether excise is specific, ad valorem or mixed; and, for the ad valorem component, if the tax is applied on the retail price and if there is a minimum specific excise and, for the specific component, if tax is automatically adjusted upwards). Each of the four components is given a score, using a five-point index, with the total score reflecting an average of the four component scores. The closer the total score is to 5, the bet- ter the tobacco tax policy performance is in a given country. While this published scorecard is currently applied only on cigarettes, it can be easily applied on other tobacco products, provided the needed data are available. 20 This indicator has also been compiled in the RGTE; see (https://www.who.int/tobacco/global_report/ Table-9.6-Affordability.xls?ua=1, accessed 29 September 2020). 21 See supplementary information on taxation, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.5-Supplementary-information-on-taxation. xls?ua=1, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 55 KEY TAKEAWAY 18 Tobacco taxation works best if implemented as part of a comprehensive MPOWER package. MPOWER is an overall indicator that incorporates all the key demand-side tobacco control measures. KEY TAKEAWAY 19 In addition to assessing the potential impact of a tax increase, policy-makers need to monitor progress over time. The share of the tax in the retail price is an indicator of progress. However, it is important to remember that an effective tax increase must translate into higher prices in order to make tobacco products less affordable. Combining all the components of a good tax policy into one scorecard can also be useful for assessing tobacco tax policy as a whole. 2.3 DOMESTIC AND REGIONAL POLICY INTEGRATION While it is essential to design tobacco tax policies with the utmost consideration of all the aforementioned factors, it is also important to consider how external factors can impact or even impede public health policy objectives. As Chapter 3 explains, cooperation among the various agencies that are directly involved in tax administration, collection and enforcement is important for effective and efficient tax policy implementation. But in the design phase, it is also essential to engage with agencies and other policy-makers that are not directly involved with taxation. Domestically, coordination is required to ensure that policies in non-health sectors do not negatively impact or even counteract tobacco control initiatives. For countries that are part of a regional bloc, harmonization of tobacco taxation is essential to protect the single market – as well as the health of the population – and to prevent tax revenue erosion, tax avoidance and tax evasion. 2.3.1 INTERSECTORAL COOPERATION ON DOMESTIC POLICY Domestic policies in agriculture, industry, trade, finance and labour have the po- tential to create or support incentives at different stages of tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. For example, subsidies provided to farmers or manufacturers involved in growing or processing tobacco can reduce prices and incentivize continued participation or even increase development in these areas, which is counterproductive to the goals of making tobacco products less affordable and reducing tobacco consumption. Multisectoral integration and policy coherence are needed at the country level to ensure that public policies and interventions in non-health sectors do not act against the intended public health impact of tobacco control and taxation. 56 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 20 Greater policy coherence in agriculture, industry, trade, finance and labour should be promoted to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. 2.3.2 REGIONAL TOBACCO TAX HARMONIZATION Policy integration is driven by the recognition that cooperation on domestic policies can substantially increase the gains from forming a regional bloc. Harmonization is desirable and may be necessary in certain areas with spillover effects, such as tax policy, the possibility of a so-called “race to the bottom” or threats to public health. Harmonization could be as simple as setting minimum standards and requirements based on global norms and best practices. Harmonization of tobacco taxation is required to ensure the establishment and proper functioning of a single market, prevent tax revenue erosion, prevent tax avoidance and tax evasion and protect people. When barriers to trade between countries are removed, harmonized tax rates support the single market because they improve the ability of consumers, producers and investors to make decisions that are not distorted by taxation but reflect real opportunity costs. Tax competition – where countries simply undercut each other’s tax rate – could prevent governments from raising sufficient funds to pursue social policy. To avoid such a race to the bottom, countries can establish minimum tax rates within the customs union (93). Even if tax competition is not present, when substantial tax differences exist in neighbouring countries, there is a clear incentive to trade across borders in order to reduce tax payments legally or illegally. The experiences of established regional economic communities offer important policy lessons, not only in terms of the general integration process but also for the process and extent of tax policy coordination. The EU implemented a successful regional tax harmonization scheme. Over the years, the focus in harmonization of tobacco taxes has broadened from the elimination of tax obstacles to the fight against harmful tax competition, tax avoid- ance and tax evasion and, more recently, to public health protection. Naturally, addressing these issues requires increasing convergence in fiscal policy and tax administration. Although price differentials still exist, setting a minimum on the share of taxes in the final price of tobacco products as well as a minimum excise tax has helped countries reach some level of harmonization. The EU experience confirms that both a declining tobacco consumption trend and stable revenues can be achieved with harmonized minimum excise rates (94). Moreover, the harmonization process has CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 57 offered opportunities for the enforcement authorities (tax or customs) to obtain information that can be used in the fight against fraud and tax evasion. On the other hand, the experience of harmonization efforts in the West African Economic and Monetary Union (WAEMU) shows how the absence of a supranational body (like the EU) or a hegemonic member state (see the SACU example below) can slow down policy integration that would benefit all member countries (95). The eight countries of the WAEMU are bound by a Tax Directive22 that requires them to impose an ad valorem excise on the CIF value or producer price of tobacco products, which is subject to under-declaration and is difficult to ascertain. Additionally, a maximum excise rate is imposed, and some members apply additional taxes to deal with this constraint. The Directive was revised in 2017 (96), but unfortunately the tax structure remains the same, and the maximum rate was not removed but rather has been increased. The Southern African Customs Union (SACU), which has five member countries, is the oldest existing customs union, established in 1910. Thanks to the hegemonic lead of South Africa, a country with a sophisticated administration system and an aggressive tobacco tax policy, SACU adopted a well-integrated tax policy that has benefited all its members (95). The GCC, established in 1981, is a regional intergovernmental political and economic union consisting of six states of the Persian Gulf. Home to one fifth of the global oil supply (97), the GCC has never relied on taxation as a source of revenue; no direct or indirect taxes were applicable in the region. Although there was no excise on tobacco products, as a customs union, the GCC countries have a common external tariff. This common tariff includes harmonized rates but also a harmonized structure. The import duty is 100% of the CIF value of tobacco products imported in the region, with a minimum tax per quantity imported. However, in recent years, to reduce their dependence on income from oil, GCC countries have considered diversifying their sources of income, including by de- veloping reliance on indirect taxes such as excise and VAT. In 2015, a decision was adopted at the 36th GCC summit meeting to implement selective taxes on all imported tobacco products and cultivated raw tobacco grown domestically (GCC Decision number 963/1). A follow-up decision in December 2016 formally agreed to the introduction of an excise tax on tobacco and other products such as sugary and energy drinks, as well as special goods (alcohol and pork meat), in all GCC countries. The decisions at the national level to implement this subregional decision came into force gradually in all GCC countries, starting with Saudi Arabia, which began 22 Directive No. 03/98/CM/WAEMU on the harmonization of Member States’ legislation of excise duties was adopted 22 December 1998. It was amended by Directive No. 03/2009/CM/WAEMU of 27 March 2009 with the objective of harmonizing excise duties within WAEMU. 58 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation in June 2017, followed by Bahrain and the United Arab Emirates (UAE) later that year, Qatar in 2018 and Oman in 2019 (1). Only Kuwait has yet to adapt its national laws accordingly. The excise introduced by the GCC countries has a structure somewhat similar to the import duty on tobacco products: the rate is 100%, but the base was changed from the CIF value to the retail price excluding taxes. The introduction of the excise led to large increases in the price of the most- sold brand of cigarettes in member countries between 2016 and 2018 – by 33% in Bahrain, more than 80% in the UAE and more than 100% in Saudi Arabia (1). In federations such as Canada and the United States – where the central govern- ment has real taxing power and some financial and regulatory control over the states or provinces – tobacco taxes are not harmonized (98–99). Even though there are significant interjurisdictional differences in taxes and prices, and tax harmonization holds great potential to reduce the scope of illicit transactions in the tobacco market, there is little evidence that Canadian provinces or individual states in the United States are interested in tobacco tax harmonization. Tax harmonization is most relevant in the context of further economic integration within a group of countries that are already part of a customs union, but it needs to be planned well to be effective. Discrepancies in law interpretation and a lack of standardization of tobacco product definitions and tax base lead to suboptimal situations. Tax rate alignment, or setting minimum rates, should come after tax structure alignment. It is important that governments support the move towards harmonization and are committed to dedicating enough financial resources and skilled personnel to oversee the entire process. KEY TAKEAWAY 21 In the context of regional economic integration and ongoing discussions regarding the possibility of harmonizing tobacco excise taxation among member countries, the experiences of existing groups can be instructive. So far, only the EU, SACU, WAEMU and, more recently, the GCC have effectively implemented a harmonized approach to excise taxation of tobacco products. Lessons learned indicate that harmonization should be planned well and should not come at the expense of tobacco control. Setting a common minimum specific excise tax, adjusted over time, is the best approach. This ensures that taxes and prices are above a minimum level, encouraging equalization of price levels and at the same time reducing affordability across countries. On the other hand, agreeing on maximum tax rates is a bad policy. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 59 Countries that wish to raise their taxes further for revenue purposes, health concerns or both should be given the space to do so. Concerns about illicit trade provoked by higher tax rates are best dealt with by strong cooperation in administration and enforcement, information sharing and adoption of new technology with common or interoperable information systems. 2.4 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In recent years, awareness of tobacco risks and harms, implementation of tobacco control provisions – especially under the WHO FCTC – and tightening of regulations have resulted in declining sales of cigarettes, primarily in high-income economies. This has changed the dynamics of the tobacco market. In response to these effective tobacco control measures, the tobacco industry has diversified its business by promot- ing a new portfolio of products, which they claim to be technological innovations that supposedly reduce the harms and risks associated with conventional tobacco products, particularly cigarettes. So-called novel tobacco products have been promoted by the tobacco industry as “cleaner alternatives,” “safer alternatives” and “reduced harm/risk products” with no smoke and no ash. On the basis of these claims, they negotiate for less-restrictive regulatory environments within countries. Some of the new products are also mar- keted or promoted for smoking cessation, despite the evidence of this outcome being inconclusive. Where these products are not banned, one of the debates in the global health community concerns the issue of their regulation and taxation. 2.4.1 HEATED TOBACCO PRODUCTS (HTPs) HTPs are tobacco products that produce aerosols containing nicotine and toxic chemicals upon heating of the tobacco or activation of a device containing the tobacco. These aerosols are inhaled by users sucking on or smoking the device. They contain the highly addictive substance nicotine (found in tobacco) as well as non-tobacco additives and are often flavoured. The tobacco in HTPs may be in the form of specially designed cigarettes (e.g. so-called heat sticks or Neo sticks) or pods or plugs. These products include IQOS from PMI, Ploom TECH from Japan Tobacco International (JTI), glo from British American Tobacco (BAT) and PAX from PAX Labs. HTPs differ not only from con- ventional cigarettes but also from ENDS – some of which are called e-cigarettes – as ENDS do not contain tobacco but rather a nicotine solution (see next subsection). However, the boundaries between the different products are becoming increasingly 60 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N difficult to define, given the emergence of so-called hybrid tobacco products that contain both nicotine solution and tobacco. HTPs are currently available in more than 40 countries and are banned in fewer than 10 countries. Even in countries where they are regulated, there is significant variation in the approaches taken to regulation. A variety of factors affect a country’s ability to control and regulate the use of HTPs, including national regulatory pow- ers, enforcement capacity regulatory frameworks, country capacity and tobacco industry interference (1). Most countries tax HTPs at a lower rate than cigarettes and on the kilogram of tobacco as a base when applying a specific or mixed excise (see Table 2.4). The use of such a base may be quite challenging for tax collection, especially because of the difficulty of checking the tobacco content in each stick. In the past, some countries taxed cigarettes per kilogram of tobacco, but today it is common practice to tax them per stick regardless of tobacco content. Table 2.4 Excise taxation of HTPs, first collected for July 2018 – updated for July 2020 OVERALL COMPARISON WITH CIGARETTES Type of excise Base unit is kg, overall rate lower than cigarettes Base unit is sticks, rate is the same as cigarettes Base unit is sticks, rate is lower than cigarettes Other Specific excise Albania, Austria, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Greece, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Montenegro, Netherlands, New Zealand, North Macedonia, Romania, Russian Federation, Slovakia, Slovenia, Sweden, United Kingdom Azerbaijan a, Japan, Ukraine b Armenia, Hungary, Jordan, Italy c, Philippines, Republic of Korea d Montenegro e, Republic of Moldova f , Serbia g Ad valorem excise (base is retail price unless specified other- wise between brackets) Spain, Switzerland Saudi Arabia and United Arab Emirates (base is retail price exclusive of excise and VAT) Indonesia h CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 61 Mixed system (ad valorem compo- nent based on retail price unless specified other- wise between brackets) France, Germany, Poland, Portugal Colombia, Georgia, Israel and West Bank and Gaza Strip (ad valorem excise base is wholesale price) a The specific excise rate applied is the same as for imported cigarettes, higher than the rate applied to domestically produced cigarettes. b The rate is the same as the minimum excise on cigarettes per 1 000 pieces. Rate and structure were effective as of 1 January 2021. c The specific excise rate is defined as 25% of the excise tax on cigarettes based on an equivalency used between cigarettes and HTPs. There are planned increases of this proportion to 40% by 2023. d In 2020 the specific excise rate was only 11% lower than cigarettes. e The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. f Specific excise rate is higher than for cigarettes but, unlike HTPs, cigarettes also face an ad valorem excise. Overall effect of excise is a slightly lower for HTPs. g The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. There is a planned phased increase of this proportion aiming equalization with cigarettes by 2025. h While cigarettes face a specific excise tax rate, HTPs face an ad valorem rate, the highest rate as defined by law, on the basis of a pre-defined minimum price. Sources: (1, 100, 101, WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021 and the Campaign for Tobacco-Free Kids website on Taxation and Price for Heated Tobacco Products https://www.tobaccofreekids.org/what-we-do/global/taxation-price/staging-tax-gap). A study by Liber (102) compared prices of HTPs and cigarettes in 34 countries and showed that while taxes have been systematically lower for HTPs than for cigarettes, prices were higher in half of the countries surveyed. KEY TAKEAWAY 22 HTPs, when taxed, are usually taxed lower than cigarettes, although they generally seem to be priced higher than cigarettes. It is important to remember that HTPs are tobacco products, and the same provisions that apply to tobacco products should apply to them as well. This is articulated in WHO’s information sheet on HTPs (103), which provides guidance on how these products should be regulated, as well as Decision FCTC/COP8(22) for novel and emerging tobacco products. Moreover, MPOWER measures, which help WHO Member States to implement the demand-reduction articles of the WHO FCTC, are applicable to HTPs, in particular, Article 6 for taxation. Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. Furthermore, HTPs contain chemicals not found in cigarette smoke, the health effects of which are not yet known. Independent assessment of industry data demonstrates that more than 20 harmful and potentially harmful chemicals 62 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N are significantly higher in HTP emissions than in cigarette smoke (104). Therefore, there is a need to learn more about these products and the health impacts of their emissions, as well as the impacts of exposure to these emissions. KEY TAKEAWAY 23 Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. From both public health and tax administration perspectives, HTPs should be taxed at the same level and in the same way as tobacco cigarettes. Some countries have already adopted this approach and are taxing HTPs at the same rate per stick as cigarettes (Azerbaijan, Colombia, Georgia, Israel, Japan, Ukraine and West Bank and Gaza Strip). Saudi Arabia and the UAE, which have recently introduced an excise tax on tobacco products as part of the GCC, are now applying the same import duty rate and excise tax structure for cigarettes and HTPs. Continuing developments in technology and changes in products have led to a recommendation to tax HTPs per unit. The definition of unit may vary by product within the HTP category. For example, one unit of IQOS is one heat stick, for Glo it is one Neo Stick and for Ploom TECH it is one tobacco pod. Governments will need to determine the exact definition of a unit for each product allowed on the market. The potential complexity of the market strongly supports limiting the types of HTPs allowed in a country and setting strict regulations to standardize the products as much as possible. Countries can also consider taxing the devices used to consume HTPs, i.e. the holder and the charger (see product description in Annex 3.1). KEY TAKEAWAY 24 HTPs are tobacco products, and they need to be treated as such. Where they are not banned, HTPs need to be strictly regulated and taxed. The recommendation is to tax them at the same level as cigarettes on a per-unit basis. Countries can also consider taxing the devices used for HTP consumption. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 63 2.4.2 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS)23 Products like ENDS and ENNDS have evolved rapidly over the past decade. ENDS heat a solution (e-liquid) containing nicotine, but not tobacco, and other chemicals that may be toxic to people’s health to create an aerosol, which is inhaled by the user. Examples of ENDS include Juul from Juul Labs, Vype from BAT and blu from Imperial Brands (1). Electronic non-nicotine delivery systems (ENNDS) are essentially the same as ENDS, but the e-liquid used generally does not contain nicotine. Upon testing, however, many so-called zero-nicotine solutions are found to contain nicotine (105–107). While generally considered as a single product class, ENDS products constitute a diverse group with potentially significant differences in the production of toxicants and delivery of nicotine. There are several coexisting types of devices for ENDS/ ENNDS on the market, including first-generation or so-called cigalikes, second- generation tank systems and even-larger third-generation or personal vaporizers. Collectively, they are also often referred to as e-cigarettes, vapes or vape pens. Other categories of ENDS include e-hookahs, e-pipes and e-cigars – hence, ENDS is an all-encompassing term for multiple product categories. Some of the products resemble their conventional tobacco counterparts – cigarettes, cigars, cigarillos, pipes or hookahs – while others are shaped more generically like pens, USB memory sticks or basic cylinders. Different forms of nicotine are also used in these ENDS, the most recent one being nicotine salts, which deliver high levels of nicotine (1). There are two types of ENDS/ENNDs products: open systems and closed systems. Open systems are devices that allow the user to buy e-liquids and fill their device with the mixtures they want (with no nicotine, different nicotine concentrations and/or flavours). Closed systems are products that come with a prefilled container (called a cartridge, pod or tank). For the past decade, divisive debates have been waged over the effectiveness of ENDS as smoking cessation aids – especially for tobacco users who are unable to give up the habit – as well as the possibility of ENDS playing a role in public health. However, the evidence remains inconclusive. Despite the tobacco industry and other related industries promoting these products as tools for quitting smoking, current evidence does not support their use as part of a population-based cessation strategy (108). Accordingly, the United States Surgeon General, in January 2020, concluded that 23 It is worth noting that ENDS are not tobacco products and not exactly new products – the technology has been around since the late 1980s (e.g. Premier, Eclipse and Accord). However, the recent generation of these products is new and has more or less piggybacked on the success of e-cigarettes. 64 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N E-cigarettes, a continually changing and heterogeneous group of products, are used in a variety of ways. Consequently, it is difficult to make generalizations about efficacy for cessation based on clinical trials involving a particular e- cigarette, and there is presently inadequate evidence to conclude that e-cigarettes, in general, increase smoking cessation (109). The evidence on the adverse health effects associated with use of ENDS is mounting, and when ENDS are used in combination with smoking – which is the practice of the majority of ENDS users (110) – the adverse health effects of two or more products are combined. However, there are insufficient data to understand the full breadth of these effects, as ENDS have not been on the market long enough for their long-term effects to be established. Nevertheless, the evidence is clear that the aerosols of the majority of ENDS and ENNDS, some of which are cancer causing chemicals. ENDS also contain nicotine, which is highly addictive. In addition, ENDS are associated with increased risk of cardiovascular diseases and lung disorders, as well as adverse effects on the developing fetus during pregnancy (108, 110). For adolescents, the use of nicotine can lead to dependence and may harm brain development. Use of ENDS could also lead to a new generation of nicotine and tobacco users, as seen in some countries, especially since these products are designed to appeal to young people. Although the specific level of risk associated with ENDS has not yet been determined conclusively, these products are undoubtedly harmful. Therefore they should be strictly regulated if allowed to be sold in domestic markets, and must be kept away from children. Taxation will be a key component of regulation, since it is an effective tool for influencing consumer behaviour. Some countries have taken the bold decision to completely ban these products. Approaches that have been taken range from partial to comprehensive bans, and ENDS/ENNDS products were banned in more than 30 countries in 2018.24 In other countries, they are regulated as, for example, consumer products, pharmaceutical products or tobacco products, or they are completely unregulated. WHO recom- mends that where ENDS/ENNDS are not banned, they should be regulated to achieve the following objectives: 1. prevent the initiation of ENDS/ENNDS by non-smokers and youth, with special attention to vulnerable groups; 2. minimize as much as possible potential health risks for ENDS/ENNDS users and protect non-users from exposure to their emissions; 3. prevent unproven health claims being made about ENDS/ENNDS; and 24 Data collected for the WHO RGTE 2019. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 65 4. protect tobacco control activities from all commercial and other vested interests related to ENDS/ENNDS, including the interests of the tobacco industry. ENDS/ENNDS present a risk to youth, who have taken up their use in high numbers in some countries, including Canada and the United States (1, 111). The Juul brand, for example, has quickly gained a significant e-cigarette market share in the United States (112–113). Its marketing and popularity have led the United States Food and Drug Administration (FDA) to raise serious concerns and to seek solutions to effectively prevent youth from taking up the use of ENDS/ENNDS (114). The city of San Francisco banned the sale of e-cigarettes in June 2019 (115). In addition to posing a risk for initiation by youth, ENDS can attract non- tobacco users or prevent current smokers from quitting. Taxation could play a role in preventing the uptake of these products, specifically among non-smokers, vulnerable groups, children and adolescents. KEY TAKEAWAY 25 The long-term health effects of ENDS/ENNDS products are still unknown, but they are clearly harmful to health. Furthermore, evidence on the effectiveness of ENDS products as a smoking- cessation aid remains inconclusive. Taxing these products could play a role in preventing their uptake, specifically among non-smokers, vulnerable groups, children and adolescents. Price elasticity of demand for ENDS products In the context of taxation, it is important to ask whether demand for ENDS is price-responsive. Preliminary evidence, although almost exclusively focused on e-cigarette data from the United States, indicates that this is the case: demand for e-cigarettes may be even more price-responsive than the demand for conventional cigarettes, so taxes can be used to deter initiation by never-users (116–123). Most of the studies of price elasticity of demand for ENDS products also demonstrate that e-cigarettes and conventional cigarettes are partial substitutes – that is, they show positive cross-price elasticity. The magnitude of the elasticity indicates the degree of substitutability between products: the higher its value, the closer the products are to being substitutes, with higher cigarette prices being associated with increased e-cigarette sales. Some of the studies also show a substitutability effect in the other direction, with increased prices for e-cigarettes leading to an increase in conven- tional cigarette use (117, 120). All of the studies show evidence of substitutability except for one (124), which differentiates between exclusive and dual users and shows no evidence of substitution between e-cigarettes and conventional cigarettes. 66 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The presence of concurrent (or dual) use – consumers using both conventional cigarettes and ENDS products – complicates results and highlights the need for more research in this area to better disentangle the different effects. Liber et al. (125) analysed sales prices in a sample of 45 countries and concluded that comparable units of conventional cigarettes cost less than disposable e-cigarettes. The units considered for pricing e-cigarettes included both the e-liquid and the rechargeable device. Taken alone, the price of e-liquids is on average much lower than that of cigarettes in high-income countries and the same in LMICs. The time needed to buy back a rechargeable device is estimated to be less than two weeks in most countries. One can argue that increasing price differentials by further increasing taxes on regular cigarettes could be effective in driving current smokers of regular cigarettes to e-cigarettes (126) as a potentially lower-risk alternative (127). However, the ef- fectiveness of ENDS as smoking cessation devices is still being debated; a study by Sweet et al. (128) shows that dual use of e-cigarettes as a potential tool for cessation was effective only in the short term. Moreover, significantly more smokers said they would quit if cigarette prices doubled and e-cigarettes were not available (122) or that they would never have become addicted to nicotine if e-cigarettes had not been so readily available (129). Once an e-cigarette user is addicted to nicotine, there is a risk of initiating traditional tobacco products use (130). In general, cessation can be better facilitated by governments via stronger implementation of the other tobacco control policies that have been proven effective at reducing use. KEY TAKEAWAY 26 Few studies are available on the price elasticity of ENDS products, and the available data come almost exclusively from the United States. These early studies indicate that demand for e-cigarettes will go down as the price of e-cigarettes increases. Generally, the results also show that cigarettes and e-cigarettes are partial substitutes, where an increase in cigarette price would increase the demand for e-cigarettes while reducing demand for cigarettes. But these results do not differentiate between people who are exclusive cigarettes or e-cigarette users and those who are users of both products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 67 Tax structure Different countries impose different tax structures on ENDS/ENNDS products25 (see details in Table 2.5). The Republic of Korea, for example, imposes a specific tax per millilitre of ENDS/ENNDS e-liquid (131), while Indonesia imposes an ad valorem tax on the retail price of the e-liquid; the maximum rate allowed by law for tobacco products (132). In the United States, there is no common way to tax e-cigarettes among the states that do tax them (133–134). The situation is similar in the EU, where new and emerging nicotine and tobacco products are not currently covered by the tobacco tax directive, and Member States may apply a national tax as they see fit under their own rules. All the EU countries that tax ENDS products apply a specific excise per millilitre of e-liquid. These different tax treatments have the potential to distort the functioning of the internal market. In February 2020, the European Commission concluded that the current provisions of the harmonized directive are no longer relevant for the taxation of ENDS and HTPs, and this is a source of concern from the internal market perspective (135). In June 2020, the Member States of the EU reiterated that it is urgent and necessary to upgrade the EU regulatory framework by harmonizing defini- tions and the tax treatment of novel products such as ENDS/ENNDS and HTPs (2). Table 2.5 Types of excises applied on ENDS/ENNDS products e-liquids globally and in individual states in the United States, as of July 2019 (updated as of July 2020 for all countries except the United States) TYPE OF EXCISE COUNTRIES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Albania, Kazakhstan, Kyrgyzstan, Pakistan, Portugal, Republic of Korea, Romania, Russian Federation, Slovenia, Sweden Azerbaijan, Cyprus, Estonia, Finland, Georgia, Greece, Hungary, Italyb, Latvia, Lithuania, Montenegro, Morocco, North Macedonia, Philippines, Serbia Ad valorem (% of retail price or import value) Bahraina Indonesia, Jordan, United Arab Emirates, Yemen TYPE OF EXCISE INDIVIDUAL STATES IN THE UNITED STATES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Delaware, Illinois (Chicago), Cook County, Louisiana, Ohio, Puerto Ricoc, Connecticutc Kansas, North Carolina, Washington, West Virginia, Wisconsin 25 The focus is on the e-liquid used for ENDS/ENNDS products. 68 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ad valorem (% of wholesale/ distributor price) Alaska (Juneau, Matanuska- Susitna Borough), California, Illinois, Maine, Maryland (Montgomery County), Minnesota, Nevada, Pennsylvania, Vermont, Washington DC, Virgin Islandsc New York Mixed New Jersey, New Mexico a Tax applied to e-shisha (or e-hookah) because e-cigarettes are banned in Bahrain. b Italy imposes differential rates for nicotine and non-nicotine containing liquids. c States in which it is unclear if only ENDS or both ENDS and ENNDS products are taxed with an excise. Sources: (1, 135, complementary data from Frank Chaloupka and WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021). Table 2.6 provides reference material on the pros and cons of different considerations for determining the tax structure and base of ENDS/ENNDS products e-liquids. Table 2.6 Excise tax options for ENDS/ENNDS products e-liquids TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Specific Volume of nicotine- containing e-liquid (regardless of concentration) 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only the volume needs to be determined 1. Difficult to compare if tax equivalencya with cigarettes is sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarettes and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed 5. May encourage more do-it- yourself (DIY) products where e-liquids are mixed by the users themselves, which increases the risk of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 69 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of e-liquid regardless of nicotine presence 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only volume needs to be determined 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Does not require laboratory capacity to detect the presence of nicotine in liquids 1. Difficult to compare if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed Volume of all e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL are affected the most; also reduces the price gap between different products 2. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 3. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 4. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 5. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to user 1. Difficult to compare with cigarettes if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed 70 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of nicotine- containing e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL affected the most; also reduces the price gap between different products 2. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as the nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 3. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 4. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to the user 1. Difficult to compare if tax equivalency with cigarettes is being sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids as there is no simple way currently available to determine whether the e-liquid contains nicotine; self-declarations by industry are not sufficient 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed Ad valorem Producer price/ CIF value of nicotine- containing e-liquid (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 71 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Producer price/ CIF value of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Retail price of nicotine- containing e-liquids (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 2. Requires capacity to monitor the market to assess market prices 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled in an e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risk of accidents, illness and death 72 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Retail price of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires capacity to monitor the market to assess retail prices 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Ad valorem with minimum specific or mixed Ad valorem + min specific 1. Volume of nicotine- containing e-liquids will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of nicotine- containing e-liquids will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up. b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products 1. Requires capacity to monitor the market to assess retail prices 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotin; (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Difficult to set a minimum specific excise amount/specific excise amount, especially if tax equivalency with cigarettes is sought 5. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 73 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Ad valorem + min specific 1. Volume of all e-liquids regardless of nicotine presence will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of all e-liquids regardless of nicotine presence will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 1. Requires capacity to monitor the market to assess retail prices 2. Difficult to set a minimum specific excise amount/specific excise amount, especially if equivalency with cigarettes is sought 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels a Tax equivalency can be measured in different ways: (1) in terms of tax burden (as % of the retail price) or (2) as the exact amount of tax for equivalent quantities (assuming an equivalency between a certain volume of e-liquid and a pack of cigarettes). b There is also the option to use the producer price/CIF value as a base for the ad valorem component, but it is a weaker option because the base is difficult to ascertain and therefore prone to undervaluation. Note: Table compiled following a WHO Expert Meeting on Taxation of Electronic Nicotine and Non-Nicotine Delivery Systems (ENDS/ENNDS), Geneva, Switzerland, 2–4 September 2019. There is currently a lack of evidence on the practical challenges being faced by countries favouring one approach over the other. Furthermore, such data are difficult to obtain because the nature of the market is constantly changing. However, a clear recommendation can be made with regard to which e-liquids to tax. As indicated in Table 2.5, some countries tax all e-liquids – whether or not they contain nicotine (ENDS and ENNDS products) – while some tax only nicotine- containing e-liquids (ENDS products). As shown in Table 2.6, there is evidence that in a number of instances, ENNDS products do contain some nicotine. Additionally, ENNDS products are not harmless (136–137). It is therefore recommended that all e-liquids be taxed for both ENDS and ENNDs products. The question of whether to employ differential taxation based on nicotine content seems reasonable from a health perspective, since nicotine is a toxic substance. However, this would likely create an additional burden for tax administrators as 74 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N they would need to determine the nicotine concentration of e-liquids on the market. Additionally, this may no longer be relevant, as advancements in technology indicate that other features of the product can influence nicotine delivery beyond the actual concentration of the e-liquid. It is now possible to increase nicotine delivery at low nicotine concentrations by increasing battery power (by reducing resistance or increasing voltage) (138). In terms of implementation, while most countries seem to have adopted a specific excise tax on ENDS/ENNDS e-liquids per millilitre, one benefit of implementing ad valorem taxation is that it seems relatively easier to regulate in the context of a large heterogeneity of products. However, it is essential that the tax be applied on the retail price value of the products, as this base is easier to ascertain than any other value that could be declared by the manufacturer. It is also important to add that regulation of the characteristics of ENDS/ENNDS products is essential, and it should be implemented along with any tax policy adopted. Regulations should include: 1. setting a maximum nicotine concentration per millilitre to safeguard public health, including reducing the risk of dependence, especially among youth; 2. setting a maximum volume for cartridges to reduce toxicants exposure and possibly limit use; 3. setting a maximum capacity for refill containers to reduce toxicants exposure and possibly limit use; 4. setting a maximum battery power to reduce the possibility of influencing nicotine and toxicant delivery; and 5. taxing nicotine regardless of its source (e.g. tobacco, eggplant, synthetic). Countries may choose to impose an excise tax on ENDS and ENNDS devices26 as well. The easiest type of tax would be an ad valorem tax based on the declared retail price. If countries choose not to impose an excise tax on these products, they should at least impose the regular VAT or sales tax rate. Imposing an excise tax on devices can be challenging from an administrative perspective, as all components need to be clearly defined and classified as devices for ENDS/ENNDS consumption. For example, if the device is assembled after importation and some parts may be used for other purposes than ENDS/ENNDS consumption, authorities may face a challenge in detecting and differentiating which component parts would be subject to excise tax and which would not. 26 See Annex 2.3 for an overview of elements of devices used in ENDS/ENNDS products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 75 KEY TAKEAWAY 27 There is currently not enough evidence to recommend one tax structure over another for ENDS/ENNDS products. However, it is clear that taxing the e-liquids used for consumption is key. The excise tax should be applied on all e-liquids, whether or not they contain nicotine. If the preferred type of excise tax is ad valorem, it should be applied to the retail price. Countries can consider taxing devices as well, but they need to adequately assess their administrative capacity to do so. Policy-makers need to be mindful of the diversity and rapid evolution of ENDS/ ENNDS products and adjust accordingly. Regulation must reflect this reality so that loopholes will not be exploited by the industry. For example, ENDS/ENNDs products include not only e-cigarettes, vapes and vape pens but also other categories such as e-hookahs, e-pipes and e-cigars. Lawmakers need to be clear about how ENDS/ENNDS products are defined so that subcategories do not fall under the radar when regulation comes into effect. Definitions will also be relevant when it comes to taxation. An unclear definition can lead to a seemingly contradictory situation, such as in Bahrain, where e-cigarettes are banned but e-hookahs are not.27 Finally, while policy-makers need to be mindful of the emergence of new products and must take appropriate actions to protect the health of their citizens, it is important to remember that the overwhelming share of nicotine consumption remains that of tobacco products, especially cigarettes. The total market value of ENDS/ENNDS and HTPs sales in 2018 was less than 2.2% of the total market value, while cigarette sales alone accounted for 91% of the same total market value (139–140). 2.5 CONCLUSIONS An overview of excise tax application globally reveals a broad variety of price and tax levels, as well as structures used for taxing tobacco products, in particular, cigarettes. Some trends, however, indicate that tax and price levels are higher among higher- income countries. The rate of taxes also matters: higher tax rates are correlated with higher prices, and higher prices change behaviour, which leads to a reduction in consumption. More countries are moving away from ad valorem taxes and towards either mixed or specific excise systems, and there are few countries that do not impose any excise tax on cigarettes. 27 In Bahrain, the Ministry of Production and Trade Decision 38 of 2013 banned e-cigarettes, while the official list of excisable products from the Ministry of Finance includes e-shishas (or e-hookahs), making them apt to be taxed and therefore considered legal. 76 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Significantly increasing the taxes and prices of tobacco products is the most effective and cost-effective policy to control tobacco use. Increased taxes – which are passed on to smokers as higher prices – reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face several challenges, ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues, as well as the perceived contribution of the tobacco sector to economic development. In designing tax policy, the tax structure adopted not only affects consumption overall, it also shapes the market structure. Ad valorem taxation incentivizes industry to set prices lower than specific taxation does. Evidence suggests that under a specific tax, the price gap between premium and lower-priced products is narrower, therefore reducing incentives for substitution to lower-priced products following a tax increase. However, as industry consolidates producers and widens its portfolio of products, new evidence indicates that the industry is introducing cheaper brands while increasing the price of its expensive brands, therefore, paradoxically, widening the price gap between its products. Evidence also suggests that prices are higher under a specific excise tax structure. Additionally, from a tax administration perspective, a specific tax is easier to imple- ment, since only the quantity produced needs to be ascertained rather than the value of the product. Another aspect of tax structure is the use of tiered taxation – that is, tax rates that vary on the basis of different product characteristics. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax. Tiered taxation encourages substitution from premium to cheaper brands, maintaining smoking prevalence and reducing the health im- pact of tax rate increases. In addition to leading to lower prices, tiered taxation is difficult to administer and creates opportunities for the tobacco industry to avoid and evade taxes. The design of a tax structure must also consider the base on which tax is applied. The choice of base should lead to the highest possible effect on price and revenue. For specific taxation, the tax base is the quantity. When the tax is ad valorem, the CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 77 choice of the tax base is important not only for health considerations, through its effect on consumption, but also for tax revenue generation, as well as industry profits. An ad valorem tax based on the producer price, or CIF value, gives tobacco manufacturers opportunities to reduce their tax liability, especially when they control the distribution system through related parties. The best practice in an ad valorem (or mixed) excise structure is to use the retail price as the tax base and introduce a minimum excise tax per pack. Other tax design considerations include the importance of using automatic adjust- ments and indexation to inflation and income growth for the specific excise tax in order to avoid erosion of the tax over time. Emerging evidence indicates that tobacco taxation does not always achieve the intended results, because the tobacco industry finds ways to circumvent it. Non- tax policies such as pricing regulation (in particular, minimum mark-ups or price floors/minimum prices) may be seen as a complementary approach to ensuring a high price level and discouraging consumption of tobacco products. So far, these policies have not proven to increase average prices. A price floor is likely to lead to increased industry profits, giving the industry greater funds for its marketing strategies (such as the introduction of new products), and lower tax revenues for governments. By reducing price competition, a price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifications). However, where powerful multinationals are operating in certain markets with presence in all market segments and with the capability to overshift a tax on some brands while undershifting the tax on others, or where price promotions cannot be banned, minimum price policies may help increase the effectiveness of tax increases. Other non-tax policies affecting price levels are those relating to promotional dis- counts for tobacco products and the sale of single sticks of cigarettes. Both should be completely banned. The ban of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Higher taxes are the most effective way to dissuade consumption, with the added benefit of raising money for the government – money that can be earmarked for health and education programs, rather than going as profits to the tobacco industry. Additionally, in order to make excise tax on tobacco products more effective in reduc- ing overall tobacco use and in line with the recommendation of the Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce 78 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N the demand for tobacco), all tobacco products need to be taxed in a comparable way; the focus should not be on cigarettes only. Tax choices and reforms have various and sometimes conflicting consequences for the market. For example, there might be a trade-off between quantity and variety or perceived quality implications. It is important for the government to recognize that firms respond strategically to changes in tax policy. Close monitoring of the market is necessary to form correct expectations about industry responses and enable estimates of the impact of a tax increase on consumption and tax revenue. To estimate the total effect of a tax increase on demand for tobacco products and tax revenue, it is important to use correct estimates of the own-price elasticity of demand, the cross-price elasticity and the income elasticity of demand. It is also important to use updated estimates of demand elasticities, as the environment within which consumers make decisions continues to change. For example, financial crises or successful tobacco control interventions can be expected to shift demand and change elasticity. Another key measure of the impact of tax policy is the tax base elasticity. Policy- makers need to be mindful of the three key components of tax base elasticity: (1) the price elasticity of demand of tobacco, (2) the share of the tax in the consumer price and (3) the degree of pass-through of the excise tax rate increase to consumer price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. It is important to acknowledge that if tax increases lead to increases in prices be- low concurrent increases in income levels, they will not be effective in reducing consumption, as tobacco remains a normal good in most countries. Policy-makers need to account for affordability when considering tax increases. They should ensure that tax increases are high enough to increase prices above income growth so that consumption goes down effectively. When designing tax policy and deciding on the right level to impose, policy-makers need to assess and project the impact of their policy decisions. Monitoring and evaluation are important. Tools for measuring impact can be very helpful, and several such tools exist. The WHO ISPT, for example, looks not only at the impact of tax policy but also at a set of tobacco control policies, and this enables national policy-makers and other tobacco control experts to explore the potential impact of CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 79 future tobacco control policies. The tool uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. More specific to tobacco tax policy, the WHO TaXSiM assesses the impact of any excise tax increase and change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the effectiveness of their policies and whether those policies have an impact on tobacco use over time. The implementation of the MPOWER package is one useful indicator for assessing tobacco control overall. Tobacco taxation works best if it is implemented as part of a comprehensive MPOWER package. The tax share in the retail price of a selected tobacco product is one indicator of the effectiveness of tax policy, but a more important one is affordability, that is, whether tax increases do lead to price increases that are above income and general price increases. A useful indicator to assess the performance of the tax policy overall is the Tobacconomics Cigarette Tax Scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). Domestic policies in agriculture, industry, trade, finance and labour all have the potential to create or support incentives at different stages in tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. Greater domestic policy coherence should be pro- moted across different sectors of the government to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. Differential tax structures and rates have the potential to distort the functioning of the internal market. Harmonization of tobacco taxation ensures the establishment and proper functioning of a single market; prevents tax revenue erosion, tax avoidance and tax evasion; and protects people’s health. In this context, tax competition, where countries simply undercut each other’s tax rate, might prevent governments from achieving their tobacco control objectives and raising sufficient funds to pursue public health policies. To avoid such a race to the bottom, countries can establish minimum tax rates on all tobacco products. A common high minimum specific excise tax is the best approach to ensure that taxes and prices are above a minimal level. 80 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N In recent years, the world has experienced the rise of new and emerging tobacco and nicotine products, including ENDS/ENNDS and HTPs, which the industry claims are safer than traditional tobacco products. The evidence so far suggests that these products could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of these products as well as initiation, cessation and switching of tobacco use behaviours among different socioeconomic groups, are not yet clear. Until more evidence for the claimed benefits of these tobacco products is available, caution should be taken in developing tax policy. Therefore, the current recommendation is for HTPs to be taxed at the same level as cigarettes on a per-unit basis regardless of tobacco content. In countries where they are not banned, ENDS/ENNDS products must be regulated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS products taxation. Nicotine- and non-nicotine-containing e-liquids should be taxed equally. If ad valorem excise is chosen as the structure, the base should be applied on the retail price. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While the evolution of new and emerging tobacco and nicotine delivery systems merits the attention of administrators and regulators, it is worth remembering that the bulk of the world’s tobacco consumption is still in conventional cigarettes. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 81 REFERENCES 1. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 2. Outcome of proceedings: Council conclusions concerning the structure and rates of excise duty applied to manufactured tobacco. Brussels: Council of the European Union; 2 June 2020 (8483/20; https://www.consilium.europa.eu/media/44235/st08483-en20.pdf, accessed 28 September 2020). 3. Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco: Article 7, paragraph 2. Brussels: OJEU. 2011;176:24–36 (https://eur-lex.europa. eu/LexUriServ/LexUriServ.do?uri=OJ:L:2011:176:0024:0036:EN:PDF, accessed 28 September 2020). 4. U.S. National Cancer Institute and World Health Organization.The economics of tobacco and tobacco control. NIH Publication No. 16-CA-8029A. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute; 2016 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/brp/tcrb/monographs/monograph-21, accessed 28 September 2020). 5. Drope J, Schluger N, Cahn Z, Drope J, Hamill S, Islami F, et al. The tobacco atlas. Atlanta: American Cancer Society and Vital Strategies; 2018 (https://tobaccoatlas.org/wp-content/uploads/2018/03/ TobaccoAtlas_6thEdition_LoRes_Rev0318.pdf, accessed 24 January 2021). 6. Sector reports, cigarettes. In: GlobalData [website]. London: John Carpenter House; 2019 (restricted access). 7. Vellios N, Ross H, Perucic AM. 2018 trends in cigarette demand and supply in Africa. PLOS ONE. 2018;13(8):e0202467 (https://doi.org/10.1371/journal.pone.0202467, accessed 28 September 2020). 8. Keen MJ. The balance between specific and ad valorem taxation. Fiscal Studies. 1998;19(1):1–37 (https://www.ifs.org.uk/fs/articles/fsmick1.pdf, accessed 24 January 2021). 9. Kay JA, Keen MJ. The structure of tobacco taxes in the European community. IFS report series. Institute for Fiscal Studies; 1982. 10. Kay JA, Keen MJ. How should commodities be taxed?: Market structure, product heterogeneity and the optimal structure of commodity taxes. Euro Econ Rev. 1983;23(3):339–58. 11. Kay JA, Keen MJ. Commodity taxation for maximum revenue. Public Finan Q. 1987;15(4):371–85. 12. Kay JA, Keen MJ. Product quality under specific and ad valorem taxation. Public Finan Q. 1991;19(2):238–47. 13. Delipalla S, Keen MJ. The comparison between ad valorem and specific taxation under imperfect competition. J Public Econ. 1992;49(3):351–67. 14. Suits DB, Musgrave RA. Ad valorem and unit taxes compared. Q J Econ. 1953;67(4):598–604. 15. Skeath SE, Trandel GA. A pareto comparison of ad valorem and unit taxes in noncompetitive environments. J Public Econ. 1994;53(1):53–71. 16. Myles G. Imperfect competition and the optimal combination of ad valorem and specific taxation. Int Tax Public Finance. 1994;3:29–44. 17. Effectiveness of tax and price policies for tobacco control. Geneva: WHO Press, World Health Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications. iarc.fr/Book-And-Report-Series/Iarc-Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And- Price-Policies-For-Tobacco-Control-2011, accessed 28 January 2021) 18. Bevan J. GCC taxation. Aylesbury (UK): Rothmans Exports Limited; 1992 (https://www.industrydocuments. ucsf.edu/docs/yxpy0198, accessed 29 September 2020). 19. Morton IT. Note from IT Morton to I Birks regarding agreement on specific taxation; 8 February 1992 (https://www.industrydocuments.ucsf.edu/docs/#id=nxpy0198, accessed 29 September 2020). 20. Galgut R. Oman. BAT (UK and Export) Limited; 16 April 1992 (https://www.industrydocuments.ucsf. edu/docs/xxpy0198, accessed 29 September 2020). 21. Common excise tax agreement of the states of the Gulf Cooperation Council [website]. Abu Dhabi: UAE Ministry of Finance; 2016 (https://www.mof.gov.ae/En/lawsAndPolitics/govLaws/Documents/ Common%20Excise%20Tax%20Agreement.pdf, accessed 29 September 2020). 82 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 22. Brock B, Choi K, Boyle RG, Moilanen M, Schillo B. Tobacco product prices before and after a statewide tobacco tax increase. Tob Control. 2016;25(2):166–73. 23. Hanson A, Sullivan R. The incidence of tobacco taxation: evidence from geographic micro-level data. Natl Tax J. 2009;62(4):677–98. 24. Delipalla S, O’Donnell O. Estimating tax incidence, market power and market conduct: the European cigarette industry. Int J Ind Organ. 2001;19(6):885–908. 25. Keeler TE, Hu TW, Barnett PG, Manning WG, Hai-Yen S. Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxation. J Health Econ. 1996;15:499–512. 26. Linegar DJ, van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2017;0:1–7 (https://tobaccocontrol.bmj.com/content/27/1/65, accessed17 February 2021). 27. Villanti AC, Collins LK, Niaura RS, Gagosian SY, Abrams DB. Menthol cigarettes and the public health standard: a systematic review. Public Health. 2017;17:983 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5747135/, accessed 24 January 2021). 28. Directive 2014/40/EU of the European Parliament and of the Council of 3 April 2014 on the approximation of the laws, regulations and administrative provisions of the member states concerning the manufacture, presentation and sale of tobacco and related products and repealing Directive 2001/37/EC: Article 7. Brussels: OJEU. 2014;57:1–38 (https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=OJ:L:2014:127:FULL&from=EN, accessed 24 January 2021). 29. Delipalla S, Keen MJ. Product quality and the optimal structure of commodity taxes. J Public Econ Theory. 2006;8:547–54. 30. Obara T, Tsugawa S. The welfare comparison of ad-valorem tax and specific tax with both quality and quantity choice of a consumer. Theor Econ Lett. 2017;7:1797–1813 (https://www.scirp.org/journal/ paperinformation.aspx?paperid=79862, accessed 24 January 2021). 31. Anderson S, de Palma A, Kreider B. Tax incidence in differentiated product oligopoly. Journal of Public Econ. 2001;81(2):173–92. 32. Razzolini L, Shughart WF, Tollison RD. On the third law of demand. Econ Inq. 2003;41:2. 33. Sobel RS, Garrett TA. Taxation and product quality: new evidence from generic cigarettes. J Polit Econ. 1997;105:880–7 (http://faculty.citadel.edu/sobel/All%20Pubs%20PDF/Generic%20Cigarettes. pdf, accessed 24 January 2021). 34. Nargis N, Fong GT, Chaloupka FJ, Li Q. The choice of discount brand cigarettes: a comparative analysis of International Tobacco Control (ITC) surveys in Canada and the United States (2002-2005). Tob Control. 2014;23:i86–i96 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/suppl_1/i86. full.pdf, accessed 24 January 2021). 35. Chaloupka FJ, Peck R, Tauras JA, Xu X, Yurekli A. Cigarette excise taxation: the impact of tax structure on prices, revenues and cigarette smoking. Cambridge: National Bureau of Economic Research; 2010 (NBER Working Paper 16287; https://www.nber.org/system/files/working_papers/w16287/w16287. pdf, accessed 24 January 2021). 36. Cigarettes in India, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 37. Gilmore AB, Tavakoly B, Taylor G, Reed H. Understanding tobacco industry pricing strategy and whether it undermines tobacco tax policy: the example of the UK cigarette market. Addiction. 2013;108(7):1317–26 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3746125/, accessed 24 January 2021). 38. Marsh L, Cameron C, Quigg R, Hoek J, Doscher C, McGree R, et al. The impact of an increase in excise tax on the retail price of tobacco in New Zealand. Tob Control. 2016;25(4):458–634. 39. Brock B, Schillo BA, Moilanen M. Tobacco industry marketing: an analysis of direct mail coupons and giveaways. Tob Control. 2015;24:505–8. 40. Chaloupka FJ, Warner KE. The economics of smoking. Handbook Health Econ. 2000;1:1539–1627. 41. Shang C, Chaloupka FJ, Fong G, Thompson M, O’Connor RJ. The association between tax structure and cigarette price variability: findings from the ITC Project. Tob Control. 2015;24:iii88–iii93 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/24/Suppl_3/iii88.full.pdf, accessed 24 January 2021) 42. Li Q, Hyland A, Fong GT, Jiang Y, Elton-Marshall T. Use of less expensive cigarettes in six cities in China: findings from the International Tobacco Control (ITC) China Survey. Tob Control. 2010;19(suppl CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 83 2):i63–i68 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/19/Suppl_2/i63.full.pdf, accessed 24 January 2021). 43. Hyland A, Laux FL, Higbee C, Hastings G, Ross H, Chaloupka FJ, et al. Cigarette purchase patterns in four countries and the relationship with cessation: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15(suppl III):iii59–iii64 (https://tobaccocontrol.bmj.com/ content/tobaccocontrol/15/suppl_3/iii59.full.pdf, accessed 24 January 2021). 44. Tsai YW, Yang CL, Chen CS, Liu, TC, Chen PF. The effect of Taiwan’s tax induced increases in cigarette prices and brand switching and the consumption of cigarettes. Health Econ. 2005;14:627–41. 45. Cummings KM, Hyland A, Lewit E, Shopland D. Use of discount cigarettes by smokers in 20 communities in the United States, 1988-1993. Tob Control. 1997;6(suppl 2):S25–S30 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/6/suppl_2/S25.full.pdf, accessed 24 January 2021). 46. Ross H, Tesche J. Undermining government tax policies: common strategies employed by the tobacco industry in response to increases in tobacco taxes. Economics of Tobacco Control Project, School of Economics, University of Cape Town and Tobacconomics, Health Policy Center, Institute for Health Research and Policy. Chicago: University of Illinois at Chicago; 2016 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC5747349/, accessed 26 January 2021). 47. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 48. Hiscock R, Branston JR, McNeill A, Hitchman SC, Partos TR, Gilmore AB. Tobacco industry strategies undermine government tax policy: evidence from commercial data. Tob Control. 2017;0:1–10 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 49. Filippidis F, Laverty A, Hone T, Been JV, Millett C. Association of cigarette price differentials with infant mortality in 23 European Union countries. JAMA Pediatr. 2017;171(11):1100–1106 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5710372/#:~:text=Cigarette%20price%20increases%20across%20 23,priced%20cigarettes%20during%20this%20period, accessed 26 January 2021). 50. López-Nicolás A, Belén Cobacho M, Fernández E. The Spanish tobacco tax loopholes and their consequences. Tob Control. 2013;22(E1):e21. 51. Ballester L, Auchincloss A, Robinson L, Mayne S. Exploring impacts of taxes and hospitality bans on cigarette prices and smoking prevalence using a large dataset of cigarette prices at stores 2001–2011, USA. Int J Environ Res Public Health. 2017;14(3):e318 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5369154/, accessed 24 January 2021). 52. Liber A, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia survey. Tob Control. 2015;24:iii83-iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed on 26 January 2021). 53. Feighery E, Ribisl K, Schleicher N, Zellers L, Wellington N. How do minimum cigarette price laws affect cigarette prices at the retail level? Tob Control. 2005;14:80–85 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/14/2/80.full.pdf, accessed 26 January 2021). 54. Tynan MA, Ribisi KM, Loomis BR. Impact of cigarette minimum price laws on the retail price of cigarettes in the USA. Tob Control. 2012;22:e1. 55. Huang J, Chriqui JF, DeLong H, Mirza M, Diaz MC, Chaloupka FJ. Do state minimum markup/price laws work? Evidence from retail scanner data and TUS-CPS. Tob Control. 2016;25:i52–i59 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i52.full.pdf, accessed 26 Jan 2021). 56. Whitehead R, Brown L, Riches E, Rennick L, Armour G, McAteer J, et al. Strengths and limitations of tobacco taxation and pricing strategies. Edinburgh: NHS Health Scotland; 2018 (http://www. healthscotland.scot/media/1829/rapid-evidence-review-strengths-and-limitations-of-tobacco- taxation-and-pricing-strategies.pdf, accessed 27 January 2021). 57. Golden S, Farrelly M, Luke D, Ribisl K. Comparing projected impacts of cigarette floor price and excise tax policies on socioeconomic disparities in smoking. Tob Control. 2016;25:i60-i66 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i60.full.pdf, accessed 27 January 2021). 58. Doogan NJ, Wewers ME, Berman M. The impact of a federal cigarette minimum pack price policy on cigarette use in the USA. Tob Control. 2018;27(2):203–08. 84 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 59. Marynak K, Xu X, Wang X, Holmes CB, Tynan MA, Pechacek T. Estimating the impact of raising prices and eliminating discounts on cigarette smoking prevalence in the United States. Public Health Rep. 2016;131:536–43 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4937114/#:~:text=We%20found%20 that%20each%20price,smokers%20aged%20%E2%89%A526%20years., accessed 27 January 2021). 60. McLaughlin I, Pearson A, Laird-Metke E, Ribisl K. Reducing tobacco use and access through strengthened minimum price laws. Am J Public Health. 2014;104(10):1844–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4937114/#:~:text=We%20found%20that%20each%20price,smokers%20aged%20 %E2%89%A526%20years., accessed 27 January 2021). 61. European Commission v. French Republic, C-197/08, European Court of Justice 4 March 2019 (https:// eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A62008CC0197, accessed on 27 January 2021). 62. Tobacco price promotion: policy responses to industry price manipulation. Boston: Public Health and Tobacco Policy Center; October 2011 (http://tobaccopolicycenter.org/documents/Price%20 Promotion%20Overview%20FINAL.pdf, accessed 28 September 2020). 63. Golden SD, Holt Smith M, Feighery EC, Roeseler A, Rogers T, Ribisl KM. Beyond excise taxes: a systematic review of literature on non-tax policy approaches to raising tobacco product prices. Tob Control. 2016;25:377–85. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/4/377.full.pdf, accessed 27 January 2021). 64. Branston J, Gilmore S. The case for Ofsmoke: the potential for price cap regulation of tobacco to raise ₤500 million per year in the UK. Tob Control. 2014;23(1):45–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3812939/, accesed 27 January 2021). 65. von Lampe K, Kurti M, Johnson J. “I’m gonna get me a loosie” understanding single cigarette purchases by adult smokers in a disadvantaged section of New York City. Prev Med Rep. 2018;12:182–5. (https:// www.sciencedirect.com/science/article/pii/S2211335518302195, accessed 27 January 2021). 66. de Ojeda A, Barnoya J, Thrasher JF. Availability and costs of single cigarettes in Guatemala. Nicotine Tob Res. 2013;15(1):83–7 . 67. Lal P, Kumar R, Ray S, Sharma N, Bhattarcharya B, Mishra D, et al. The single cigarette economy in India - a back of the envelope survey to estimate its magnitude. Asian Pac J Cancer Prev. 2015;16(13):5579-82 (http://journal.waocp.org/article_31295_9ebd3981652934bbf5d78ec066b6a39d.pdf, accessed 27 January 2021). 68. Coady MH, Chan CA, Sacks R, Mbamalu IG, Kansagra SM. The impact of tax on purchasing behaviors among New York City smokers. Am J Public Health. 2013;103(6):e54-e60 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC3698745/, accessed 27 January 2021). 69. Cigarettes in Thailand, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 70. Cigars in Germany, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 71. Report from the Commission to the Council on Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2018 (COM(2018) 17 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/report_excise_duty_manufactured_ tobacco_12012018_en.pdf, accessed 28 September 2020). 72. Gallet CA, List JA. Cigarette demand: a meta-analysis of elasticities. Health Econ. 2003;12(10):821–35. 73. Guidelines for implementation of article 6 of the WHO FCTC. Geneva: World Health Organization; 2016 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 29 September 2020). 74. DeCicca P, Kenkel D, Liu F. Excise tax avoidance: the case of state cigarette taxes. J Health Econ. 2013;32(6):1130–41 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3855906/, accessed 27 January 2021). 75. Jawad M, Lee JT, Glantz S, Millett C. Price elasticity of demand of non-cigarette tobacco products: a systematic review and meta-analysis. Tob Control; 2018;27(6):689–95 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6056338/, accessed 27 January 2021). 76. John RM. Price elasticity estimates for tobacco products in India. Health Policy Plan. 2008;23(3):200–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2800993/, accessed on 27 January 2021). 77. Wasserman J, Manning WG, Newhouse JP, Winkler JD. The effects of excise taxes and regulations on CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 85 cigarette smoking. Journal of Health Econ, Elsevier. 1991;10(1):43–64. 78. Cheng K, Kenkel D. U.S. cigarette demand: 1944-2004. B.E. Journal of Econ Analysis & Policy. 2010;10(1):10.2202/1935-1682.2438 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3820165/, accessed 28 January 2021). 79. Bobak M, Jha P, Nguyen S, Jarvis M. Poverty and smoking. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000. 80. Peck, RM. Equity issues, tobacco, and the poor. Yurekli A, de Beyer J, editors. Washington (DC): World Bank Group; 2013. (Economics of Tobacco Toolkit, Tool 6; https://openknowledge.worldbank.org/ handle/10986/16265, accessed 28 January 2021). 81. Kenkel DS, Schmeiser MD, Urban C. Is smoking inferior?: evidence from variation in the earned income tax credit. J Hum Res. 2014;49:1094–1120. 82. Ruhm, CJ. Healthy living in hard times. Journal of Health Econ. 2005;24(2):341–63. 83. Ruhm, CJ. Are recessions good for health? The Quarterly Journal of Econ. 2000;115(2):617-50. 84. Tarantilis F, Athanasakis K, Zavras D, Vozikis A, Kyriopoulos I. Estimates of price and income elasticity in Greece. Greek debt crisis transforming cigarettes into a luxury good: an econometric approach. BMJ Open. 2015;5(1):e004748. (https://bmjopen.bmj.com/content/bmjopen/5/1/e004748.full.pdf, accessed 28 January 2021). 85. Kaiser M, Reutter M, Sousa-Poza A, Strohmayer K. Smoking and the business cycle: evidence from Germany. Discussion Paper Series. Bonn, Germany: IZA Institute of Labour Economics; 2017 (http:// ftp.iza.org/dp10953.pdf, accessed 28 September 2020). 86. Tingum EN, Mukong AK, Mdege N. The effects of price and non-price policies on cigarette consumption in South Africa. Tob Induc Dis. 2020;18(July):62 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC7398599/, accessed 28 January 2021). 87. Vardavas CI, Filippidis F, Agak, I, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the importance of earmarking revenue to health care and tobacco control. Tob Induc Dis. 2012;10(1):21 (http://www. tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the-revenue-to- health-care-and-tobacco,66110,0,2.html, accessed 28 January 2021). 88. Marquez P, Catinkaya V. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 89. Blecher EH, van Walbeek CP. An international analysis of cigarette affordability. Tob Control. 2004;13:339– 46 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/13/4/339.full.pdf, accessed 28 January 2021). 90. Goodchild M, Sinha P, Munish VG, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/handle/10665/331836, accessed 28 January 2021). 91. John RM, Dauchy E. Trends in affordability of tobacco products before and after the transition to GST in India. Tob Control Online First. 20 March 2020. 92. Chaloupka FJ, Drope J, Siu E, Vulovic V, Stoklosa M, Mirza M, et al. Tobacconomics cigarette tax scorecard. Chicago: Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/636/uic-tobacco-scorecard-report-final.pdf, accessed 28 January 2021). 93. Kanbur R, Keen MJ. Jeux sans frontieres: tax competition and tax coordination when countries differ in size. Am Econ Rev. 1993;83(4):877–92 (https://www.researchgate.net/publication/4901220_Jeux_ Sans_Frontieres_Tax_Competition_and_Tax_Coordination_When_Countries_Differ_in_Size, accessed 28 January 2021). 94. Bouw A. Tobacco taxation in the European Union: an overview (English). Washington (DC): World Bank Group; 2017 (http://documents.worldbank.org/curated/en/493581492415549898/Tobacco- taxation-in-the-European-Union-an-overview, accessed on 28 September 2020). 95. Blecher E, Drope J. The rewards, risks and challenges of regional tobacco tax harmonization. Tob Control. 2014;0:1–5. 96. Directive No.n° 01/2017/CM/UEMOA of 22 December 2017. Paris: Droit-Afrique; 2017 (https://www. 86 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N droit-afrique.com/uploads/UEMOA-Directive-2017-01-droits-accise-tabacs.pdf, accessed 29 September 2020). 97. Mirzoev T, Zhu L, YangY, Pescatori A, Matsumoto A, Callen T. The future of oil and fiscal sustainability in the GCC region. Middle East and Central Asia Research Department. Washington (DC): International Monetary Fund; 2020 (https://www.imf.org/-/media/Files/Publications/DP/2020/English/FOFSGCCEA. ashx, accessed 28 January 2021). 98. State cigarette excise tax rates & rankings. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/factsheets/0097.pdf, accessed 28 September 2020). 99. A map and table comparing cigarette prices in Canada (March 2018). Toronto: Non-Smokers’ Rights Association/Smoking and Health Action Foundation; 2018 (https://nsra-adnf.ca/wp-content/ uploads/2018/03/180302_map_and_table.pdf, accessed 4 February 2021). 100. European Commission excise duty rates on manufactured tobacco. Brussels: European Commission; 31 December 2017 (https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco- energy/excise-duties-tobacco_en, accessed 1 July 2018). 101. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2019 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study-on-the-tobacco-taxation-directive-2019_en.pdf, accessed 28 September 2020). 102. Liber AC. Heated tobacco products and combusted cigarettes: comparing global prices and taxes. Tob Control; 2018;28(6). 103. Heated tobacco products: information sheet - 2nd edition. Geneva: World Health Organization; 2020 (https://www.who.int/publications/i/item/WHO-HEP-HPR-2020.2, accessed 28 January 2021). 104. Kass Lempert L, Glantz S. Analysis of FDA’s IQOS marketing authorisation and its policy impacts. Tob Control 2020;0:1–9. 105. Chivers E, Janka M, Franklin P, Mullins M, Larcombe A. Nicotine and other potentially harmful compounds in “nicotine-free” e-cigarette liquids in Australia. Med J Aust. 2019;210(3):118–9 (https://onlinelibrary. wiley.com/doi/epdf/10.5694/mja2.12059, accessed 28 September 2020). 106. Wenzl T, Zelinkova Z, Regueiro J, Giri A. Scientific-technical support activities to DG TAXUD-C-2 on the option to include e-cigarettes within the scope of excisable goods for the Impact Assessment on a possible revision of Directive 2011/64/EU. Administrative Arrangement TAXUD/2015/DE/329. Luxembourg: Publications Office of the European Union; 2017 (https://publications.jrc.ec.europa.eu/ repository/bitstream/JRC105828/kjna29062enn.pdf, accessed 28 September 2020). 107. Lisko JG, Tran H, Stanfill SB, Blount BC, Watson CH. Chemical composition and evaluation of nicotine, tobacco alkaloids, pH, and selected flavors in e-cigarette cartridges and refill solutions. Nicotine Tob Res. 2015;17(10):1270–78. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573955/, accessed 28 January 2021). 108. Ghebreyesus TA. Progress in beating the tobacco epidemic. Lancet. 17 August 2019;394:548–49 (https:// www.thelancet.com/pb-assets/Lancet/pdfs/S0140673619317301.pdf?hss_channel=tw-27013292, accessed 29 September 2020). 109. Smoking cessation, a report of the Surgeon General. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2020. (https://www.hhs.gov/sites/default/ files/2020-cessation-sgr-executive-summary.pdf, accessed 8 October 2020). 110. E-cigarettes are harmful to health. Geneva: World Health Organization. 5 February 2020 (https://www. who.int/news-room/detail/05-02-2020-e-cigarettes-are-harmful-to-health, accessed 29 September 2020). 111. Shan L, Manzione LC, Azagba S. Psychological well-being and dual-use of cigarettes and e-cigarettes among high school students in Canada. J Affect Disord. 2020;265:357–63 112. King BA, Gammon DG, Marynak KL, Rogers T. Electronic cigarettes sales in the United States, 2013- 2017. JAMA. 2018;320(13):1379-80 (https://jamanetwork.com/journals/jama/fullarticle/2705175, accessed 28 January 2021). 113. LaVito A. Popular e-cigarette Juul’s sales have surged almost 800 percent over the past year [website]. CNBC. 2 July 2018 (https://www.cnbc.com/2018/07/02/juul-e-cigarette-sales-have-surged-over-the- past-year.html, accessed 28 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 87 114. Statement from FDA Commissioner Scott Gottlieb, M.D., on new enforcement actions and a youth tobacco prevention plan to stop youth use of, and access to, JUUL and other e-cigarettes. Washington (DC): US Food and Drug Administration; 23 April 2018 (www.fda.gov/NewsEvents/Newsroom/ PressAnnouncements/ucm605432.htm, accessed 29 September 2020). 115. Paul K. San Francisco becomes first US city to ban sale of e-cigarettes. The Guardian. 25 June 2019 (https://www.theguardian.com/us-news/2019/jun/25/san-francisco-e-cigarettes-vaping-juul-ban, accessed 29 September 2020). 116. Pesko MF, Courtemanche CJ, Maclean JC. The effects of traditional cigarette and e-cigarette tax rates on adult tobacco product use. J Risk Uncertain. 2020;60:229–258. 117. Cotti CD, Courtemanche CJ, Maclean JC, Nesson ET, Pesko MJ, Tefft N. The effects of e-cigarettes taxes on e-cigarettes prices and tobacco product sales: evidence from retail panel data. Cambridge: National Bureau of Economic Research; 2020 (NBER Working Paper 26724; https://www.nber.org/ system/files/working_papers/w26724/w26724.pdf, accessed 28 January 2021). 118. Saffer H, Dench DL, Grossman M, Dave DM. E-cigarettes and adult smoking: evidence from Minnesota. Cambridge: National Bureau of Economic Research; 2019 (NBER Working Paper 26589; https://www. nber.org/system/files/working_papers/w26589/w26589.pdf, accessed 28 January 2021). 119. Huang J, Gwarnicki C, Xu X, Caraballo RS, Wada R, Chaloupka FJ. A comprehensive examination of own- and cross-price elasticities of tobacco and nicotine replacement products in the U.S. Prev Med. 2018;117:107–14 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6195827/, accessed 28 September 2020). 120. Zheng Y, Zhen C, Dench D, Nonnemaker JM. U.S. demand for tobacco products in a system framework. Health Econ. 2017;26:1067–86. 121. Stoklosa M, Drope JM, Chaloupka FJ. Prices and e-cigarette demand: evidence from the European Union. Nicotine Tob Res. 2016;18(10):1973–80. 122. Grace RC, Kivell BM, Laugesen M. Estimating cross-price elasticity of e-cigarettes using a simulated demand procedure. Nicotine Tob Res. 2015;17(5):592–8. (https://www.researchgate.net/ publication/270291421_Estimating_Cross-Price_Elasticity_of_E-Cigarettes_Using_a_Simulated_ Demand_Procedure, accessed 28 January 2021). 123. Huang J, Tauras J, Chaloupka FJ. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob Control. 2014;23:iii41–iii47 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/23/suppl_3/iii41.full.pdf, accessed 17 February 2021). 124. Cheng KW, Chaloupka FJ, Shang C, Ngo A, Fong GF, Borland R, et al. Prices, use restrictions and electronic cigarette use—evidence from wave 1 (2016) US data of the ITC Four Country Smoking and Vaping Survey. Addiction. 2019;114(suppl 1):115–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC6661221/, accessed 28 January 2021). 125. Liber AC, Drope JM, Stoklosa M. Combustible cigarettes cost less to use than e-cigarettes: global evidence and tax policy implications. Tob Control. 2016;0:1–6 (https://www.researchgate.net/ publication/299473181_Combustible_cigarettes_cost_less_to_use_than_e-cigarettes_Global_ evidence_and_tax_policy_implications, accessed 28 January 2021). 126. Chaloupka FJ, Sweanor D, Warner KE. Differential taxes for differential risks — toward reduced harm from nicotine yielding products. N Engl Journal Med. 2016;373(7):594-7. 127. Goniewicz ML, Knysak J, Gawron M, Kosmider L, Sobczak A, Kurek J, et al. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob Control. 2014;23:133–9. 128. Sweet L, Brasky TM, Cooper S, Doogan N, Hinton A, Klein EG, et al. Quitting behaviors among dual cigarette and e-cigarette users and cigarette smokers enrolled in the tobacco user adult cohort. Nicotine Tob Res. 2019;21(3):278–84 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6379027/, accessed 28 September 2020). 129. Wills TA, Knight R, Williams RJ, Pagano I, Sargent JD. Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics. 2014;135(1):e43–e51 (https://www. ncbi.nlm.nih.gov/pmc/articles/PMC4279062/, accessed 28 January 2021). 130. Electronic Nicotine Delivery Systems: report by WHO. Geneva: World Health Organization; 2014. (FCTC/ COP/6/10 Rev.1; https://apps.who.int/gb/fctc/PDF/cop6/FCTC_COP6_10Rev1-en.pdf, accessed 28 January 2021) 88 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 131. WHO report on the global tobacco epidemic 2017: monitoring tobacco use and prevention policies. Geneva: World Health Organization; 2017 (https://www.who.int/tobacco/global_report/2017/en/, accessed 28 January 2021). 132. Regulation of the Minister of Finance of the Republic of Indonesia, number 146, PMK.010/2017 on excise tax rates of tobacco products. Jakarta: Ministry of Finance Indonesia; 2017. 133. Gourdet CK, Chriqui JF, Chaloupka FJ. A baseline understanding of state laws governing e-cigarettes. Tob Control. 2014;23:iii37-iii40 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4078672/, accessed 28 January 2021). 134. An act to ensure that nicotine products are taxed equally. 127th Maine Legislature: first regular session-2015. H.P. 670, Legislative Document No. 973. 2015 (https://www.mainelegislature.org/legis/ bills/getPDF.asp?paper=HP0670&item=1&snum=127&PID=, accessed 28 January 2021). 135. Evaluation of the Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco. Commission Staff Working Document. Brussels: European Commission; 2020 (SWD(2020) 32 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/10- 02-2020-tobacco-taxation-report.pdf, accessed 28 September 2020). 136. Electronic nicotine delivery systems and electronic non-nicotine delivery systems (ENDS/ENNDS): report by WHO. Geneva: World Health Organization; 2016. (FCTC/COP/7/11; https://www.who.int/ fctc/cop/cop7/FCTC_COP_7_11_EN.pdf, accessed 28 September 2020). 137. E-cigarette use among youth and young adults: a report of the Surgeon General. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2016 (https://e-cigarettes. surgeongeneral.gov/documents/2016_SGR_Full_Report_non-508.pdf, accessed 28 September 2020). 138. Wagener TL, Floyd EL, Stepanov I, Driskill LM, Frank SG, Meier E, et al. Have combustible cigarettes met their match? The nicotine delivery profiles and harmful constituent exposures of second-generation and third-generation electronic cigarette user. Tob Control 2017;26:e23–e28 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/26/e1/e23.full.pdf, accessed 28 September 2020). 139. Vapour products data. In: Euromonitor International [website]. London: Euromonitor International; 2019 (restricted access). 140. Global market data: segment insights. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 89 ANNEX 2.1 Table A2.1 Countries that apply different types of cigarette excise tax structures, 2018 SPECIFIC EXCISE (65 COUNTRIES) AD VALOREM EXCISE (42 COUNTRIES) MIXED EXCISE (63 COUNTRIES) NO EXCISE (15 COUNTRIES) Albania, Andorra, Australia, Azerbaijan, Barbados, Belarus, Belize, Bolivia (Plurinational State of ), Burundi, Canada, Cook Islands, Dominica, Ecuador, Eswatini, Fiji, Gambia, Honduras, Iceland, India, Indonesia, Jamaica, Japan, Jordan, Kazakhstan, Kenya, Kiribati, Kyrgyzstan, Lesotho, Malaysia, Mauritius, Mongolia, Mozambique, Myanmar, Namibia, Nepal, New Zealand, Nicaragua, Norway, Pakistan, Palau, Papua New Guinea, Peru, Philippines, Republic of Korea, Saint Lucia, Saint Vincent and the Grenadines, Samoa, Seychelles, Singapore, Solomon Islands, South Africa, Sri Lanka, Suriname, Tajikistan, Timor-Leste, Tonga, Trinidad and Tobago, Uganda, United Republic of Tanzania, USA, Uruguay, Uzbekistan, Vanuatu, Yemen, Zimbabwe Argentina, Armenia, Bahrain, Bangladesh, Benin, Burkina Faso, Cabo Verde, Cambodia, Cameroon, Chad, Comoros, Côte d’Ivoire, Cuba, Democratic Republic of the Congo, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Ghana, Grenada, Guatemala, Guinea-Bissau, Liberia, Madagascar, Mali, Mauritania, Niger, Panama, Paraguay, Saint Kitts and Nevis, Saudi Arabia, Senegal, Sierra Leone, Sudan, Syrian Arab Republic, Togo, Turkmenistan, Tuvalu, United Arab Emirates, Venezuela (Bolivarian Republic of ), Viet Nam, Zambia Algeria, Austria, Belgium, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Central African Republic, Chile, China, Colombia, Congo, Costa Rica, Croatia, Cyprus, Czechia, Denmark, Dominican Republic, Egypt, El Salvador, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iran (Islamic Republic of ), Ireland, Israel, Italy, Lao People’s Democratic Republic, Latvia, Lebanon, Lithuania, Luxembourg, Malta, Mexico, Montenegro, Morocco, Netherlands, Nigeria, North Macedonia, Poland, Portugal, Republic of Moldova, Romania, Russian Federation, Rwanda, Sao Tome and Principe, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Thailand, Tunisia, Turkey, Ukraine, United Kingdom, West Bank and Gaza Strip Afghanistan, Angola, Antigua and Barbuda, Democratic People’s Republic of Korea, Iraq, Kuwait, Libya, Maldives, Marshall Islands, Micronesia (Federated States of ), Nauru, Niue, Oman,a Qatar,a Somalia a This table shows the status of cigarette excise tax structures as of July 2018 and does not account for changes occurring after that date, in particular for the cases of Qatar and Oman, which introduced excise on tobacco in January 2019 and June 2019, respectively. Source: WHO RGTE. 90 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 2.2 THE ANALYTICS OF THE TAX BASE ELASTICITY Assume tax revenue R = tsQ or R = tvPQ , where Q is the quantity consumed, ts is the specific tax, tv is the ad valorem tax and P is the consumer price. The following equations can help to illustrate the different components of the tax base elasticity. Under a specific excise regime, change in revenue depends essentially on the change in consumption: where R is the tobacco tax revenue, is the specific excise tax and is the tobacco tax base elasticity. The tax base elasticity is made of: where ε, the price elasticity = , is the degree of pass-through of the specific excise tax rate increase on consumer price and is the tax-price ratio. Under an ad valorem excise regime, change in revenue depends essentially on the change in tobacco expenditure: where R is the tobacco tax revenue, tav is ad valorem excise tax and ηav is the tobacco tax base elasticity. The tax base elasticity here is made of: where is the degree of pass-through of the ad valorem excise tax rate increase on consumer price, is the tax-price ratio and ε the price elasticity = . ∂R = Q (1+ηs)∂ ts ηs = ε ts P ∂P ∂ts tav P ηav = (1 + ε) tav P ∂P ∂tav∂P ∂tav ∂P ∂ts ΔQ ΔP P Q ΔQ ΔP P Q ts P ∂R = P Q (1+ηav)∂tav CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 91 ANNEX 2.3 ELEMENTS OF THE DEVICES THAT MAKE UP ENDS/ENNDS PRODUCTS The main components of any ENDS/ENNDS kit include essentially: • USB charger (not a car charger) • Inbuilt battery Additionally, For open systems • Tanks (refillable containers) with removable atomizer (often sold bundled with atomizers) • Clearomizers/refillable pods (no removable atomizer) • E-liquid For closed systems • Disposable e-cigarettes: not rechargeable, thrown away after e-liquid is finished • Nondisposable e-cigarettes: – Pre-filled cartomizers (cartridges designed to go with the cigalike kit) – Pre-filled tank refills/pods (pods or cartridges designed to go with the prefilled tank/pod kits) Some definitions: • Atomizer: uses a heating element to vaporize the e-liquid • Cartomizer: combines the cartridge/tank and the atomizer • Clearomizer: same as cartomizer, uses different technology • Cartridge/tank/pod: container that includes the e-liquid In summary, ENDS/ENNDS product devices include the following: • USB charger (not a car charger) • Inbuilt battery • Disposable e-cigarettes • Atomizer • Cartomizer/clearomizer • Cartridge/tank/pod with or without atomizer • Pre-filled cartridge/tank/pod (for closed systems, includes e-liquid) • E-liquid (added in the cartridge/tank/pod in open systems) Source: ECigIntelligence, 2020. Information also obtained from vaping websites, including https://www. misthub.com/blogs/vape-tutorials/76788357-tutorial-atomizer-vs-cartomizer-vs-clearomizer, http:// www.bestclearomizer.com/clearomizer-vs-cartomizer-vs-atomizer/, https://wayofleaf.com/accessories/ vapes/atomizer-vs-clearomizer-vs-cartomizer, https://wayofleaf.com/accessories/vapes/atomizer-vs- clearomizer-vs-cartomizer, accessed 15 July 2020. 92 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 93 CHAPTER 3. Tobacco tax administration 3.1 INTRODUCTION Imposing excise taxes on tobacco products usually serves more than one purpose. Governments often find themselves balancing interests between financial and public health objectives. Both objectives can best be achieved by an efficient and effective competent authority with strong technical capacity to enforce and collect taxes. A competent authority is the agency, organization or department that is legally as- signed to complete a particular activity; in the case of administering tobacco taxes, the competent authority is often a tax administration, revenue authority, customs department or ministry of finance. Article 6 of the WHO FCTC (1), along with its guidelines (2), provides a solid foundation for sound tax administration. As stated under section 1.5 in the guidelines, tobacco tax systems should be efficient and effective. They should be structured to minimize the costs of compliance and administration, while ensuring that the desired level of tax revenue is raised and health objec- tives are achieved. Efficient and effective administration of tobacco tax systems enhances tax compliance and collection of tax revenue while reducing tax evasion and the risk of illicit trade. Efficiency in tax administration refers to minimizing the costs per unit of tax revenue collected. It is measured by comparing the resources used with the revenues gener- ated. Effectiveness in tax administration refers to a high level of compliance – also described as taxpayers meeting their obligations. Thus, an efficient and effective competent authority collects the tax at a minimum cost while ensuring conformity to the rules. Tobacco taxation is the single most effective tobacco control measure for re- ducing tobacco use and is best implemented as part of a comprehensive tobacco control plan (3). Illicit trade – including smuggling and illicit manufacturing – and tax avoidance undermine the effectiveness of tax policies and their objectives (4). The impact on illicit trade is often cited by opponents of tax increases, who argue that increasing taxes increases illicit trade. They contend that illicit trade can lead 94 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N to lower revenues for governments and lower prices. The challenge faced by the competent authorities is to ensure that due taxes are declared and collected on all tobacco products that are manufactured in and/or imported into its jurisdiction, while at the same time detecting tobacco products that are illegally manufactured in and/or imported into its jurisdiction, stopping such activity and prosecuting the responsible parties. This chapter describes the shared characteristics of good tax administrations, including best practices based on country experiences. It regularly refers to the WHO FCTC, and – given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade – draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (5). Any practice or conduct prohibited by law and related to production, shipment, receipt, possession, distribution, sale or purchase of tobacco products – including any practice or conduct intended to facilitate such activity – is considered as illicit trade (Article 1). The objective of the Protocol is to eliminate and prevent all forms of illicit trade in tobacco products. At the same time, the Protocol includes measures for tobacco tax administration based on international best practices, which makes it relevant for all countries, even those that are not Parties to it. The Protocol was adopted at the fifth session of the COP to the WHO FCTC in 2012 and entered into force on 25 September 2018. As indicated in the Preamble, it was developed in response to the increasing international illicit trade in tobacco products (5). The Protocol covers three main areas: (1) measures to control the supply chain (Part III); (2) measures dealing with offences, including sanctions (Part IV) and (3) international cooperation (Part V). Different provisions of the Protocol are discussed in detail throughout this chapter, and section 3.4 is devoted specifically to control and enforcement. 3.2 INSTITUTIONAL ARRANGEMENTS Competent authorities that collect taxes effectively in an efficient way share a number of attributes. The organizational structures of these authorities contain clearly defined roles, responsibilities and rules for coordination among relevant bodies. Moreover, competent authorities collect data regularly and manage information needed for assessing risks. The key to successful risk management is to share this information among relevant authorities both within a country and between countries. Effective and efficient competent authorities also regularly evaluate their performance and accountability according to key performance indicators to identify areas for improve- ment. These characteristics are discussed in greater detail in the following sections. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 95 3.2.1. CLEARLY DEFINED ROLES AND RESPONSIBILITIES OF COMPETENT AUTHORITIES The designation of competent authorities for the implementation and enforcement of tax laws – including clear definitions of the boundaries of authority among numerous agencies within a country – is essential for efficient collection of taxes. Areas where different agencies need to cooperate and share data must also be defined. Overlap of activities by different authorities leads to inefficient use of resources, whereas gaps create opportunities for fraud, leading to ineffective tax laws. The importance of clearly defined roles and responsibilities applies not only to tax authorities and customs but also to law enforcement agencies, including police and border control forces. The implementation and enforcement of taxation is organized differently in various countries. The most common structure separates customs and tax administration. The trend since the 1990s, however, has been to combine these functions into one agency, such as Her Majesty’s Revenue and Customs (HMRC) in the United King- dom, SUNAT in Peru1 and AFIP in Argentina.2 Several countries have increased coordination between tax and customs by creating a revenue secretariat and also implementing systems to share tax records as a single taxpayer account. Coordina- tion between tax policy and tax administration authorities has also increased. One can think of combining both into one department within the ministry of finance or ensure that tax administration authorities are consulted during the tax policy process. Some tasks, such as licensing, may be handled by other ministries such as the ministries of health, agriculture or trade. For example, the Ministry of Health of Brunei and the Health Science Authority of Singapore are responsible for the licensing of importers of tobacco products (6). In some federal countries, including Colombia and the United States, excise taxes – including tobacco taxes – are collected and enforced by local or state tax administrations. Other countries have organized the administration of national taxes by establishing a single unified revenue body. Particularly in larger economies, that body is often responsible for both direct and indirect taxes, including excise taxes, and reports to the ministry of finance. All the functions needed for effective and efficient tax administration are established within these bodies (7). No matter what the institutional arrangements may be, it is vital that the agencies cooperate and exchange information and that their competencies find their basis in law. More information on this topic is provided in section 3.2.2. 1 Law Decreto Supremo 061-2002-PCM - Disponen fusión por absorción de la Superintendencia Na- cional de Administración Tributaria – SUNAT con la Superintendencia Nacional de Aduanas - Aduanas [Supreme decree year 2002 about the merger between Tax and Customs Administration]. Lima: El Peruano, 12, July 2002 (in Spanish) (http://www.sunat.gob.pe/legislacion/sunat/ds061-2002-PCM.pdf, accessed 13 November 2020). 2 Administracion Federal de Ingresos Publicos, Decreto 618/1997 [Federal Administration of Public Revenue, Decree 618] (in Spanish) (http://servicios.infoleg.gob.ar/infolegInternet/an- exos/40000-44999/44432/norma.htm, accessed 13 November 2020). 96 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Many countries, however, have separate bodies for the collection of taxes and customs duties. A 2015 survey of 135 tax administrations worldwide found that only 36% of them were responsible for both tax administration and customs ad- ministration (8). In most countries, customs authorities are more likely to collect excise duties on imports, and in many countries, VAT or sales tax is collected jointly with tobacco tax, particularly for imported products. This simplifies controls and creates synergy by unifying common processes and procedures, resulting in cost savings for tax administrations and taxpayers. The involvement of multiple bodies in tax collection requires especially good collaboration and information-sharing to ensure efficient and effective collection of taxes and duties. KEY TAKEAWAY 1 Institutional arrangements with clearly defined roles and responsibilities – designed to prevent overlaps and voids – contribute to effective and efficient tax administration. 3.2.2. EFFECTIVE COORDINATION AMONG RELEVANT BODIES Coordination at the national level Coordination among relevant bodies is key to effective tobacco tax administration. This means not only clearly defined roles and responsibilities, as described in the previous section, but also coordination among the competent authority, customs and those responsible for formulating, analysing and implementing tax policy. Regardless of the institutional arrangements – whether the responsible parties are all within the ministry of finance or in separate government agencies – all parties need to cooperate and exchange information to optimize tax collection and enforcement of tax policy. In practice, this means that information should be shared among, for example, customs, local government units that issue licences and health authori- ties – particularly those that regulate the sale of tobacco products. For tax authorities, the most relevant information concerning excise taxes in- cludes the identity of taxpayers and those involved in the trade of tobacco (import and export data, licences, criminal records, tax returns, bank statements, etc.); the category, quantity, value and location of manufactured goods; and the movement of those goods until all taxes are paid. Legal impediments to obtaining this informa- tion – such as bank secrecy or privacy regulations – should be kept in mind, and where needed, exceptions for fiscal procedures should be incorporated into law. Seizure data are also a valuable source of information; more details on this are provided in section 3.4. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 97 Tax authorities should regularly coordinate with law enforcement agencies – such as the police and border control forces, depending on a country’s laws – to properly monitor tobacco-related activities and enforce the tax laws. Often, the competent tax authority and customs authorities work in close cooperation with anti-fraud teams.3 Coordination and sharing of information can be required in legislation or regulations to ensure a streamlined process and avoid confusion. This can be done on an ad hoc basis as needed or with formal planned exchanges of information and regular meetings. It is recommended that at least a legal basis for exchange or access to information among government bodies be established to prevent claims during legal procedures that evidence was obtained unlawfully. Some countries go beyond exchanging information and cooperation. In the Neth- erlands, for example, customs authorities not only carry out work for the Ministry of Finance, they also carry out non-fiscal tasks for seven other departments, including the Ministry of Agriculture, Nature and Food Quality; the Ministry of Justice and Security; and the Ministry of Foreign Affairs (9). These activities are often based on bilateral agreements between the Ministry of Finance and the other departments. In other countries, such as the United States and Canada, Customs and Border Protection are not part of the Ministry of Finance; they are part of the Department of Homeland Security in the United States and the Ministry of Public Safety and Emer- gency Preparedness in Canada. These agencies also carry out many non-fiscal tasks. Along with the implementation of new tobacco control and tax laws, several countries have also created high-level committees to ensure good coordination and implementation of the laws. Led by health and finance ministries, committees ensure coordination and fine-tuning to achieve desired results. Botswana, Chile, Colombia, Indonesia and Senegal, among other countries, have successfully started with coordina- tion, planning and monitoring of tobacco laws’ implementation through periodic com- mittee meetings. The committees usually include representatives from the ministries of health, finance, tax and customs, police, transport and, in some cases, education. Coordination across borders Effective approaches to control smuggling in tobacco products require interventions at the borders of jurisdictions and therefore must involve the border agencies. However, with the globalization of trade, there is a need for close coordination not only between tax and border control authorities but also between different jurisdictions. Recent cases have demonstrated that an absence of formal cooperation frameworks may expose a market to financial crime, including money-laundering and financing of terrorism (10). 3 See, for example, Focus on tax fraud. Customs administration of the Netherlands, tax and customs administration. 2017;2 (https://customsnl-insight.nl/article/309563676, accessed 3 October 2020). 98 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Accession to international cooperation agreements such as the Protocol, the Organisation for Economic Co-operation and Development (OECD) Multilateral Convention on Mutual Administrative Assistance in Tax Matters and other regional arrangements will contribute greatly to the effective exchange of information and cooperation among enforcement agencies. An effective exchange of market data and information from participating jurisdictions can prevent potential cross-border crimes and loss of domestic revenue. International cooperation reinforces domestic measures to stop illicit trade and raise much-needed revenues. Parties to the Protocol have a commitment to cooperate with one another and to share information to meet their obligations under the Protocol (Article 20). The Protocol itself is the legal instrument that allows Parties to cooperate and share information across borders. Authorities of governments that are not Parties to the Protocol or another coopera- tion agreement that represents a legal instrument to exchange information could conclude a mutual assistance agreement or exchange of information agreement to guide the procedures under which information exchange can take place effectively. The Revised Kyoto Convention of 2010 promulgated by the World Customs Organization (WCO) recommends that jurisdictions that enter into bilateral agree- ments require the other jurisdiction to provide pre-arrival information on goods bound for their customs territory. A survey of 87 WCO members in 2013 found that the vast majority of customs administrations had the legal authority to share information related to the supply chain of tobacco products with other administrations (11). Some economic blocs have also established harmonized legislation applying to all of their Member States to provide administrative cooperation to efficiently cooperate on tax matters (12). Coordination can include the establishment of a special agency to ensure the safety and proper functioning of external borders, such as the European Border and Coast Guard Agency, also known as Frontex (from the French frontières extérieures, “external borders”). In some of the Frontex-led operations, EU and non-EU countries cooperate together with international organizations to target cross-border crime, including the smuggling of cigarettes and raw tobacco (13). Criminals who engage in illicit trade of tobacco products are usually also en- gaged in related criminal activities such as bribery, money laundering, corruption, obstruction of justice and even financing of terrorist organizations (14). A number of international treaties provide the legal framework for addressing such conduct through mechanisms that tackle illicit trade from a criminal justice perspective, such as the United Nations Convention against Transnational Organized Crime, the United Nations Convention against Corruption and the International Convention for the Suppression of the Financing of Terrorism. Table 3.1 summarizes the types of structures available for such coordination. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 99 Table 3.1 Structures for coordinating mechanisms TYPE OF COORDINATION BASIS INVOLVED ACTORS National coordination Agreements with a basis in law between national agencies Customs authorities, ministries of finance and those responsible for formulating, analysing and implementing tax policy; law enforcement agencies, such as police and border control forces; and anti-fraud teams Agreements between ministries or a basis in law or regulation on the establishment of high- level committees Ministry of health, finance, revenue, justice, transport and sometimes education and enforcement entities such as customs and police Bilateral coordination Bilateral cooperation agreements National governments Regional coordination Regional arrangements such as • Harmonized legislation applying to all Member States of an economic bloc to provide administrative cooperation in taxation to efficiently cooperate on tax matters • Regulation to jointly establish a special agency to ensure the safety and functioning of external borders EU Member States, the European Border and Coast Guard Agency (Frontex), customs, law and border enforcement agencies International coordination International treaties or conventions such as • The Protocol • OECD multilateral Convention on Mutual Administrative Assistance in Tax Matters • United Nations Convention against Transnational Organized Crime • United Nations Convention against Corruption • International Convention for the Suppression of the Financing of Terrorism Parties to international treaties and conventions, law and border enforcement agencies KEY TAKEAWAY 2 Regardless of differing institutional arrangements, coordination and cooperation within a country and across jurisdictions are essential to optimize tax collection and enforcement of tax policy. 100 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 3.2.3. EVALUATION OF PERFORMANCE AND ACCOUNTABILITY Key strategic indicators are useful for assessing the performance of a competent authority. Performance indicators can include measures such as net revenue col- lected, total expenditures compared with budgeted amounts, the ratio of costs to collection, measures of filing and payment compliance and taxpayer satisfaction (15). Several international organizations, including the IMF, the World Bank, the Inter-American Development Bank and OECD have developed tools to evaluate tax and customs with key performance indicators. This section provides information on some of the indicators that are particularly useful for measuring performance related to tobacco taxes, including the cost of collection ratio, tax gap analysis and tax revenue targets. Cost of collection ratio Collection costs vary among countries. The cost of collection ratio is the total ex- penditure as a percentage of the total net taxes collected. This ratio is often used as a measure of efficiency and effectiveness of competent authorities. In Table 3.2, the cost of collection ratio is calculated for country groups by income level, based on an annual IMF survey. The numbers in the table give an indication of resources used and revenues collected for taxes in general. The same definition of cost of collection was used for all countries. The tax revenue excludes VAT and excise taxes on imported products, so it reflects internal taxes only: personal and corporate income taxes, VAT and excise on domestic production. Customs duties are also not included. The results show the differences among countries at various income levels. Other contributing factors include differences between tax systems, economic situations and compliance levels. Table 3.2 Cost of collection ratio in 2015 per 100 units (ratio of average recurrent budget to revenue collecteda) GROUP (SAMPLE SIZE) 2015 Low-income countries (6) 1.3 Lower-middle-income countries (15) 1.6 Upper-middle income countries (18) 0.9 High-income countries (36) 0.9 All (76) 1.1 a Does not include VAT or excise on imports Source: (Reference 8, Appendix Table 12). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 101 As one would expect, given lower levels of automation and resources, the ratio is higher for low- and lower-middle-income countries, greater than 1.0 (more than 1.0 currency unit needed to collect 100 currency units). The ratios for upper-middle income and high-income countries are below 1.0, indicating more efficient and/or effective collection systems. The cost of collection might be less relevant for taxes that are introduced with other than solely financial objectives, such as influencing a change in behaviour. In particular, in the case of excise taxes applied on tobacco products, the cost of tax collection does not reveal the full picture. If excise tax rates are increased substan- tially – or at least increased above inflation and income growth – consumption will be reduced. As a result, health care costs will be reduced due to reduced tobacco- related mortality and morbidity and increased productivity. These savings are not factored into the ratio of cost of collection to revenue, but governments do benefit from these lower expenses overall. Nevertheless, the cost of collection can be used as an indicator of the efficiency of a competent authority. Tax gap analysis Tax gap analysis is another method of determining how effectively taxes on tobacco products are collected. The tax gap is the difference between the tax due and the tax that is collected. For example, the theoretical tax due under an ad valorem tax on the retail price of cigarettes would be the average price of a pack of cigarettes multiplied by the number of packs sold (estimated from household expenditure surveys, for example) multiplied by the tax rate. This outcome can then be compared to the actual revenues collected (16). The effectiveness of tax collection can also be determined by using the macro- economic input-output matrix, measuring the added value of the economic sector – tobacco in this case – and the theoretical VAT due and then comparing the result with the real VAT collection. This methodology is valid for measuring domestic tax evasion (more information on the use of this method to estimate illicit trade is provided in Chapter 4, section 4.1). Tax revenue target The performance of a competent authority can also be evaluated by determining whether the tax revenue target has been met, if mandated, for a given tax period. Although revenue forecasts are often used as targets, caution is advised. Forecast revenues could include assumptions such as economic growth, inflation and amount collected. Forecasting is a good practice, however, and competent authorities should provide input to the government for the forecasting of revenues to improve the quality of the estimates. Competent authorities should monitor the actual collections 102 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N in comparison with the forecasted revenues, but the theoretical base may not be attainable for a variety of reasons. In addition, a revenue target could provide an incentive for some customs and competent authorities to simply aim to reach the target amount, rather than making efforts to collect the maximum amount possible with the available resources. 3.3 THE TAX COMPLIANCE CYCLE For any tax, there are associated compliance, control and enforcement processes. The compliance cycle usually includes registration and licensing, tax declarations, recordkeeping, storage in warehouses, duty suspension, collection of tax and tax refunds. Figure 3.1 illustrates the typical stages of the tax compliance cycle. Fig. 3.1 Tax compliance cycle 3.3.1 REGISTRATION AND LICENSING Along with regulating and ensuring the integrity of those who deal with controlled substances or goods, the main objective of licensing is to regulate the supply chain. Licensing is a powerful tool for obtaining more information and securing the supply chain of tobacco products. Parties to the Protocol are committed to licensing the manufacturing, import and export of tobacco products and manufacturing equipment (Article 6). In addition, Parties are committed to endeavouring to license – as considered appropriate – the persons involved in the growing of tobacco and the retailing, transporting, wholesal- ing, brokering, warehousing and distribution of tobacco products or manufacturing Registration and licensing (renewal) Tax declaration Authorities: audit and control Taxpayers: recordkeeping Audit and control of information provided in tax declaration Payment and collection of tax Refund Licensed activities, for example: import, production, transport, storage, export, etc. 1 2 3 456 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 103 equipment (Article 6). To ensure an effective licensing system, Parties shall monitor and collect, where applicable, any licence fees that may be levied and consider using them in effective administration and enforcement of the licensing system, for public health or for any other related activity in accordance with national law. If feasible, each Party shall require that retailers and tobacco growers – except for traditional growers working on a noncommercial basis – maintain complete and accurate records of all relevant transactions in which they engage, in accordance with its national law (Article 9.4). Article 6.3(b) of the Protocol provides a list of information to be requested from the applicant of the licence, including: • relevant identity information on the applicant • business location of the manufacturing unit or warehouse and production capacity • detailed list of tobacco products and equipment used • description of where the manufacturing equipment will be installed and used • documentation or declaration of any criminal records • information on bank accounts to be used for transactions and payments • description of intended use and intended market of sale of the tobacco products. To make it easier for authorities to collect all the information they need, rules of confidentiality could be exempted in the licensing process. Licences can be general – covering all activities requiring a licence – or issued for each activity separately, such as different licences for manufacturing, importing and retail. A general licence is less burdensome for the licensing authority, whereas licences for each type of activity offer greater control but at the cost of more adminis- tration (17). The cost of implementing the licensing system should be proportionate to the potential impact of the system. Not only should the type of licences be taken into consideration, the process and information needed to obtain a licence should be carefully considered to ensure proportionality. The more stringent the process is – in terms of the information required and the obligations the system imposes on licensees – the more burdensome the regime will be on both businesses and the authorities who must administer and enforce it. The more information is collected, the higher the compliance and administrative burden will be. It is recommended that the added value of the information be balanced with the additional compliance, administrative and/or enforcement burden. The level of stringency should be decided with consideration of factors such as the level of risk of the activity and the availability of enforcement capacity. A more stringent regime might be justifiable for activities that pose a higher risk for the 104 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government in terms of potential loss of tax revenues – such as the import, production and handling of excisable products on which the excise taxes have not yet been paid. Authorities could consider setting licence fees at a high enough level to cover the costs of administering and enforcing the system. For an example of a system that relies mainly on licensing and permissions, see the case study of Australia in Box 3.1. Wholesalers, distributors and retailers of tobacco products could also be required to obtain a licence before they can engage in the trade of those products. This would enable the competent authority to require reports on, for example, transactions relating to the purchase and sale of tobacco products. Moreover, it would allow the authorities to complete the audit trail of the entire supply chain and to obtain data that will help tax and health policy-makers properly and effectively monitor tobacco products. Governments could also require a licence for entities dealing with raw materials or growing tobacco, including farmers. If licensing of tobacco farmers is deemed appropriate and subsequently required in a country, the farmers have to identify and register their farm areas and location to obtain a licence. The benefit of requiring licences for farmers is that the control of the legitimate supply chain is extended to the identification of the source of the raw material for tobacco products. It also makes it more difficult to divert raw tobacco from the licit to the illicit supply chain. Licences are issued by different agencies across the world. In Brazil, for example, the Health Surveillance Agency is responsible for providing licences. Operators need to obtain approval of the layout of manufacturing and warehousing facilities before they can operate. In addition, they must demonstrate how they will comply with other laws and regulations – for example, by showing the design of product packaging, including the pack, carton and master case. The factory location must be identified before manufacturers can obtain a licence. Finally, a licence is required for the importation of machinery to produce tobacco products (18). Licences can be a source of useful information if authorities establish the informa- tion that applicants must supply in order to obtain the licence. Such information could include the quantity, price and how the tobacco harvests are disposed, as well as the identity of the buyers. It is recommended that an effective licensing regime collect information to establish both the identity and characteristics of applicants by requiring criminal records on relevant offences, such as previous noncompliance with tobacco licences or fraud. To avoid loopholes for monitoring raw tobacco, importers of tobacco leaf could also be licensed or at least required to register and report information on quanti- ties, sources and sales. In some countries, this information is already collected by a government agency other than tax authorities, for example, by the ministry of agriculture. Duplication of requirements and reporting should be avoided through CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 105 legislation and coordination among agencies. If licensing of (small-scale) farmers is difficult to implement, subsequent purchasers (first processors) in the supply chain could be licensed and regulated instead. Licensing first processors is often less burdensome to enforce for competent authorities because, in general, there are far fewer first processors than there are growers. For example, in the EU, between 50 and 100 first processors have been identified, compared with 55 000 farmers (19). Countries could also consider requiring registration of persons or entities engaged in the manufacture and import or sale of materials used for the manufacturing of tobacco products, such as cigarette papers, tobacco leaves, additives, adhesives, acetate or any other type of filters used for cigarettes, tipping paper and cellophane or plastic wraps, as well as materials for packing the cigarettes into packs, reams and master cases. In addition, tobacco manufacturers could be required to obtain a licence before they can purchase these materials. The Parties to the Protocol should decide on appropriate measures, depending on research as to whether key inputs that are essential for manufacturing of tobacco products exist and can be identified and subject to effective controls. KEY TAKEAWAY 3 The objective of licensing is to regulate and secure the supply chain. It is a powerful tool for obtaining information for verification, further investigation and audits. Ideally, all persons involved in the growing of tobacco and retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. Licensing requisites Based on case studies and best practices – including experiences from managing bonded warehouses where the value of merchandise or suspended duties or taxes is high – the following kinds of information could be required to obtain a licence, in particular, for producers, warehouses and distributors of tobacco products: • certification of safety of installations, perimeter security for production and storage (may include CCTV [closed-circuit television]4 access for tax administration) • certification of financial solvency • detailed online, real-time inventory of tobacco products and main raw materi- als, accessible by tax administration • electronic accounting systems 4 The term “closed-circuit television” is used generically to describe surveillance camera systems. 106 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • detailed lists of owners and managers • banking and other financial records • periodic electronic reports of transactions for tobacco products • anytime tax administration right of entry for inventories • mandatory electronic tax returns and payments • mandatory prior-to-arrival customs declarations for tobacco products • declarations of compliance with the tax stamp system (if applicable) • for those involved in import or export, authorized economic operator (AEO) certification • proof of compliance with the bond or guarantee regime • agreement to finance reasonable cost of inspections and tracking and tracing. Box 3.1 Case study of licensing in Australia Australia has taken an approach to controlling tobacco taxes that differs from that in many other countries. It has not used fiscal marks or tracking and tracing.5 Instead, it administers tobacco taxes through licensing and permission-based systems aimed at facilitating operations by lower-risk entities while preventing or tightly controlling commerce involving higher-risk entities. The domestic tax agency, the Australian Taxation Office (ATO), is responsible for most of the controls. These controls cover tobacco that is grown or manufactured and imported as finished goods or as leaf for manufacturing in Australia. In fact, the legal tobacco market in Australia consists only of imported finished tobacco products. In 2006, all tobacco-growing licences were cancelled by the ATO because manufacturers switched to cheaper leaf from external suppliers. The last domestic cigarette manufacturers closed in 2015 and 2016, and there has been no legal domestic tobacco growing or manufacture since then. The ATO administers all other functions relating to the import of tobacco and tobacco products, including licensing of bonded warehouses used to store imported products and issuance of permissions to undertake movement of bonded tobacco products between licensed bonded warehouses or to places of export. Importers must apply for a licence for a bonded warehouse to store imported tobac- co. The applicant must meet general criteria such as fitness, recordkeeping and security.6 These criteria are designed to ensure that only low-risk entities are able to enter the ex- cise tax system. Risk levels are also kept at an acceptable level through provisions allow- ing the suspension or cancellation of licences, subject to appeal. Licences are valid for a 5 See sections 3.4.4 and 3.4.5 for detailed discussions on fiscal marks and tracking and tracing. 6 The entity must not have been charged with an offence under the Excise Act or any Commonwealth, State or Territory Act that carries a penalty in excess of US$ 105 000 in the previous 12 months (or convicted in the previous 10 years), has shown a history of compliance with tax law in the previous four years, has had no previous cancellation of a licence, has adequate financial resources and is not in receivership. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 107 three-year period, with automatic renewal for licensees with demonstrated compliance. The permission system relies on post-transaction audits of commercial records. Criteria used to assess risk include the size of the duty liability, the compliance record of both parties and the possibility of diversion into the market. When there is a perceived risk of revenue loss, the application can be denied or a financial security deposit can be required. Exports of tobacco products are also subject to an export declaration process with the Australian Border Force. An approved export declaration is required for the products to be able to leave the country. Following recommendations from a government task force in 2017, the status of tax-suspended, bonded tobacco was eliminated as of 1 July 2019. In addition, an import licensing regime was introduced, and commercial tobacco imports without a licence are banned. Importers are required to identify their duty liabilities at import and make immediate payment; there are no credit terms available. Full payment of duties and taxes to the Australian Border Force are required prior to a release of tobacco products into the country. Sources: (20–21). As mentioned above, certification as an AEO could be requested as part of the licensing process. Most customs authorities are familiar with the concept of AEOs. Created by the WCO, AEO principles were initially focused on security concerns (22). Having a special licensing regime for operators of the tobacco supply chain is recom- mended due to the special nature of the product. For countries that have no system in place, AEO certification could be a starting point for setting up such a regime. An AEO is defined by the WCO SAFE Framework of Standards (22) as a party involved in the international movement of goods – in whatever function – that has been approved by, or on behalf of, a national customs administration as complying with WCO or equivalent supply chain security standards. AEOs include, inter alia, manufacturers, importers, exporters, brokers, carriers, consolidators, intermediaries, ports, airports, terminal operators, integrated operators, warehousers and distributors. For many years – in some cases, even since the 1970s – customs administrations have been increasingly involved in the security of the international trade supply chain. More recently, customs administrations have developed security programmes in a global context. The AEO is part of these programmes, and in 2005, the WCO adopted the SAFE Framework of Standards. Since then, a number of traders have been required to make substantial investments in order to obtain AEO status and must continue to invest to maintain that status. The AEO program is also recognized by the Trade Facilitation Agreement, a multilateral agreement signed by 174 countries (23). 108 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Some regional blocs have further specified the standards for AEOs and provide clear and well-structured information on their websites to guide and encourage operators to apply for AEO status. A good example of this practice is the website of the Revenue Commissioners of the Republic of Ireland, which contains the in- formation shown in Box 3.2. Box 3.2 AEO: Republic of Ireland Tax and Customs What are AEOs? AEO status is a certified standard authorization issued by customs administrations in the European Union (EU). It certifies that an economic operator has met certain standards in relation to: • safety and security • systems to manage commercial records • compliance with customs rules • financial solvency • practical standards of competence or professional qualifications. This is primarily a trade facilitation measure that recognizes reliable operators and encourages best practices in the international supply chain. As an AEO, an operator could benefit from: • recognition worldwide as a safe, secure and compliant business partner in international trade; • lower risk scores in risk analysis systems when profiling; • priority treatment if physical controls are conducted; • mutual recognition of AEO programmes under Joint Customs Cooperation Agreements, which could result in faster movement of goods through third- country borders; • reduced data sets for entry and exit summary declarations (this applies only to AEO safety and security); • easier access to simplified procedures; • reduction or waiver of comprehensive guarantees. The conditions for AEO status apply to all businesses regardless of size. Manufacturers, exporters, freight forwarders, warehouse keepers, clearance agents, carriers and importers may all apply for AEO status. Source: (24). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 109 3.3.2 DATA COLLECTION, DECLARATIONS AND ACCOUNTING The effectiveness of risk analysis depends on the quality and reliability of the available data. This is also the case for risk analysis in relation to tobacco taxes. Obtaining reliable data can be a challenge in many countries, but the use of electronic sys- tems to collect and manage data is increasing in most competent authorities. The introduction of VAT in many countries around the world has greatly improved the availability of data that can be used for tobacco tax analysis, since reporting is done along the supply chain on, for example, the value, quantity of goods and transaction date. Most countries applying excise duties also have a VAT system in place. In addition, more countries are becoming Parties to the Protocol. With the implementation of the Protocol, more data will become available because countries will be obliged to implement, among other measures, licensing systems with report- ing requirements and tracking and tracing systems. More information on tracking and tracing systems is provided in section 3.4. The obligations of the Protocol will also assist in monitoring the stock of tobacco products. Ideally, all entities involved in the tobacco product supply and distribution chains should be licensed and required to record every transaction that occurs. As this might be burdensome for both tax authorities and taxpayers, the use of automated and electronic systems is recommended in order to decrease the costs of compliance. An accurate inventory system for all raw materials, machinery, goods in process and finished products can be required. It is even more important to have good recordkeeping of the required data. As the volume of reported data increases, a good information technology (IT) system will be needed. The use of IT for periodic tax declarations, accounting, inventory and financial data is critical for obtaining accurate information and decreasing costs for the entire reporting system. Most countries now have some level of automation that can facilitate data analysis. An emerging trend is the use electronic invoices, issued by traders, as part of online real-time information for tax administration. Countries generally start by using electronic invoices at public utility companies and then later expand the use to large companies. Electronic invoices minimize the use of paper, contribute to automated recordkeeping and give accurate and timely information about transactions for tax administration. Several countries began using electronic invoices for companies on a voluntary basis and later made their use mandatory, especially for large companies with a high number of transactions – including the tobacco industry. Electronic invoices have been implemented successfully in EU countries and almost all Latin American countries, as well as several Asian countries.7 7 Electronic Invoicing in Latin America: English Summary of the Spanish Document; Inter-American Development Bank, Inter-American Center of Tax Administrations, 2018 (https://publications.iadb.org/ publications/english/document/Electronic-Invoicing-in-Latin-America.pdf ). 110 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N To verify that information is accurate, competent authorities could systematically cross-check declared information against third-party information (e.g. from banks, financial institutions, employers) or match the data with the information in registers of other government agencies. Processes of cross-checking and data matching could also be automated to minimize the administrative burden (25). KEY TAKEAWAY 4 Reliable data are essential for effective risk analysis. While obtaining these data can be challenging, electronic systems can help reduce the burden by automating procedures of data collection and cross-checking of information with different sources. 3.3.3 RECORDKEEPING Parties to the Protocol are committed to requiring, as appropriate, that all persons or entities engaged in the supply chain of tobacco, tobacco products and manu- facturing equipment keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of tobacco products (Article 9). Relevant information includes market volumes, trends, forecasts of tobacco products and quantities of tobacco products and manufacturing equipment kept in stock in tax and customs warehouses in transit, transhipment and under duty suspension. This information should be required from the persons and entities engaged in the supply chain and submitted to the competent authority on a regular basis, as provided for in the law. The competent authority can use the submitted information to monitor compliance with tobacco regulations and payment of taxes. A registry with this level of detail can realistically be kept only in electronic form. Records must provide full accountability for materials used in the production of tobacco products. The intention is that tax authorities and manufacturers should be able to reconcile the production quantities with the inputs used in production – thereby providing confidence that no unrecorded or illicit production has occurred. Obligations should also be imposed on suppliers of key inputs to show that supply is commensurate with demand (17). KEY TAKEAWAY 5 To monitor compliance and payment of taxes, all persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions, as well as details of materials used in the production of tobacco products. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 111 3.3.4 WAREHOUSING, STORAGE AND DISTRIBUTION According to Article 6.2 of the Protocol, all Parties shall endeavour to license persons involved in any wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Maintaining a system of authorization allows the authorities to carry out controls in production and storage facilities to ensure that taxes are paid (2). The approval process to obtain an authorization could include an evaluation of the layout of the plant or warehouse, the machinery that will be used and the flow of production, warehousing and shipping, including the points of entry and exit of raw materials and finished products. The basic method of monitoring production and ensuring that only tax-paid products are released to the market from the premises is to identify the production facilities and to control the entry and exit points. From time to time, the competent authority should conduct a physical inventory of the goods contained therein to check whether all documentation was duly prepared and approved and to determine the accuracy and completeness of the records kept. If the jurisdiction requires tax stamps to be placed on the tobacco products, only products with the proper stamps affixed can be withdrawn. Generally, tobacco products for which the required taxes have not been paid and, if required, fiscal marks have not been affixed should not be allowed into warehouses. For practical reasons, many countries allow suspension of excise duties, meaning that prior authorized persons can produce, send, receive and store tobacco products on which the excise duty has not yet been paid. The relevant authorities could also require that products on which the taxes have been paid should not be stored in the same areas as the products under duty suspension. Obviously, products under suspension of payment of excise duties are at high risk, which could justify stricter requirements for production, trade, storage and handling. Australia, which has a strict system of licensing and requirements for permission to move tobacco products, has migrated to a new system that eliminates bonded warehouses from the supply chain as of 1 July 2019. Importers are required to have an import licence and to pay excise taxes on cigarettes immediately upon import (see Box 3.1 in section 3.3.1). 3.3.5 DUTY SUSPENSION Many countries require authorization of natural or legal persons (as authorized warehouse keepers) to produce, process, hold, receive and dispatch products sub- ject to excise duty during their business. Producing, processing, holding, receiving and dispatching excise goods often take place under suspension of the excise duty. Guarantees can be requested from authorized persons to secure the payment of taxes. Features of such a system may include strict criteria for granting authorization, 112 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N warehouse pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. The use of a computerized system for monitoring movements of excise goods under suspension of excise duty can be a control as well. Different licences for products under duty suspension could also be considered. This would make enforcement easier and less burdensome for both authorities and operators. In general, it is recommended to allow the handling of excise goods under suspension of duties only if strict criteria are met. Such criteria could include pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. In principle, the movements of tobacco products should also be covered by the tracking and tracing system. Considering the high risk related to these products, additional monitoring could be considered appropriate, such as a computerized system monitoring the movements of excise goods under suspension of excise duty. In the design of such a system, it is recommended that close attention be paid to customs procedures for import and export to ensure alignment and avoid a vacuum in monitoring. An example of a computerized system is the EU’s Excise Movement and Control System, which follows the movement of all excise products – including manufactured tobacco products – for which excise taxes have not been paid. The system records the movement in real time and is thereby an important tool for combatting fraud. In addition, this system is indispensable for the exchange of information and co- operation between the relevant authorities of Member States of the EU (26). Finally, authorization is required before tobacco products can be produced, imported or stored under suspension of excise duties (27). KEY TAKEAWAY 6 Products under duty suspension of excise taxes are at a higher risk of tax evasion, which can justify stringent measures such as requesting guarantees to ensure the payment of taxes, additional licensing requirements, compliance with computerized systems to monitor the movement of excise goods under suspension and on-site authorization and audits. 3.3.6 COLLECTION OF TAXES To reduce the complexity of tax collection systems, it is recommended that excise taxes be imposed at the point of manufacture, import or release from storage or production warehouses for consumption. This is common practice in the majority of countries that impose excise taxes. Collecting taxes at this level of the supply chain CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 113 greatly limits the number of taxpayers and thus the resources needed to control them. Encouraging taxpayers to use electronic payment methods can also increase the chances of collecting all taxes. The same applies to requiring guarantees for certain high-risk activities, such as the handling of goods under duty suspension. Many countries decide on a case- by-case basis the level of the guarantee, depending on the situation of the requestor and the level of risk (quantity or value and potentially due excise taxes) that the regular business activities represent in a given time frame. Some countries allow a reduction of guarantees for operators with a track record of good compliance. It should be noted that a guarantee is not a limitation of the liability; taxpayers can still be requested and liable to pay an amount far above the level of the guarantee. Tax payments should be required by law to be remitted at fixed intervals after sales or on a fixed date each month (2). Many countries have a specialized collection enforcement unit that works full-time on the collection of taxes. It is important to have a stop-filer or payment control that can act immediately when noncompliance occurs, by sending a message and phone call of late declaration or late payment to the taxpayer. This increases the likelihood of keeping taxpayers compliant. If nondeclaration or nonpayment persists, the bond or guarantee could be executed. Another reason for collecting excise taxes around the time of production or import is that quantities can be monitored more effectively at these points. There are different options for monitoring the supply chain of tobacco products. The decision about what kind of monitoring system to use depends on the country’s financial, technical and human resources. The weakest form of monitoring is in- dustry self-declaration. Activities to verify compliance and ensure the collection of the full amount of taxes due can include, for example, physical checks, audits, cross-checking of declared information with third-party data and inspection of administration and recordkeeping. In general, in countries with poor administration systems, enforced compliance is carried out by imposing physical control over the production or manufacturing process. The cost of physical control increases when there is a potential for fraud by excise officers. However, fraud can be diminished significantly when excise officers are rotated frequently among different locations and supervisors make surprise visits. Historically, some countries (e.g. India) have posted tax administration staff at production facilities to monitor production and removals. In India, a staff member of the competent authority is placed in cigarette and large bidi manufacturing facili- ties around the clock. Each officer records the daily production and the quantity of cigarettes/bidis that leaves the factory and reports to the next officer.8 8 Ministry of Finance India, personal communication, 2009. 114 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A better option is to monitor production remotely. The competent authority can require the installation of CCTV cameras in strategic places throughout the manufacturing and warehousing facilities. With these, the authority can establish a central command post from which the facilities and activities can be continu- ously monitored and documented. In addition, the competent authority can carry out physical inventory controls from time to time and – if electronic invoices are implemented – cross-checking between invoices and declared inventory. This is also an effective way to prevent collusion between staff of a competent authority and manufacturers or importers. For example, in 2015, the Bureau of Internal Revenue of the Philippines required all tobacco companies to install CCTV cameras in their production lines and warehouses. This decision was taken in response to large seizures of untaxed cigarettes, with the objective of monitoring production to ensure the payment of all taxes. The collection process must also be supported by IT systems. These systems must provide for transparency and accuracy to ensure a safe process for the flow of payments from taxpayers to the tax treasury. Most countries have implemented automated electronic systems for tax payments linked to each declaration, for both domestic and import. It is key for tax administrations to have a comprehensive agreement with the banking system in order to obtain lower transaction costs, if applicable. Some countries have implemented a state payment web portal that allows citizens to pay their taxes and other fees such as county fees, fees for car permits and licences and agricultural, health and environmental fees online. KEY TAKEAWAY 7 Excise taxes should be imposed at the point of manufacture, import or release from storage or production warehouses for consumption, to ensure that quantities can be monitored effectively. This also reduces the complexity of tax collection systems by limiting the number of taxpayers and thus the resources needed to control them. 3.3.7 TAX REFUNDS Refunds for VAT, excise taxes and customs duties are a common process in most countries, under the principle that consumption taxes are not exported. Frequency and methods of refund vary by country. It is common to have monthly refunds (if there are exports during the period), and the reimbursements may be sent directly to the exporter or reserved as a credit to pay other taxes. An alternative used by some countries that have a high volume of exports is a so-called zero rate, or suspension, meaning that indirect taxes (VAT, excise taxes and customs duties) are suspended for the whole chain – from import of raw materials to production and packing until CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 115 export. This regime requires a special licensing process. Since the tobacco industry has an export component, the refund process for this sector requires special atten- tion for tax administration. 3.4 CONTROL AND ENFORCEMENT Control and enforcement are the main functions of tax administration. In fact, most tax laws include the objective “to control and enforce tax compliance” and, for customs, “to control and enforce tax and duty payments at the border” or similar phrases. The Protocol provides guidance for control and enforcement of tobacco taxes. Efficient and effective competent authorities often have a strategic plan to ensure compliance, a risk-based approach to identify the problematic points in the chain and the ability to direct resources accordingly to high-risk or high-value areas. Tasks that can play a role in control and enforcement include controlling the registration and licensing process, due diligence, verifying declarations and collec- tion of taxes. Production and distribution controls including tracking and tracing, fiscal markings, audits and import and export controls all play a role in control and enforcement. This section describes the main activities for improving control and enforcement, focusing on the tobacco supply chain. The procedures and penalties that can be enacted once illicit trade in tobacco has been detected are also discussed. 3.4.1 CONTROL AND ENFORCEMENT PLANNING Strategic plan In modern tax administrations, it is common to have a strategic plan, with control and enforcement as pillars. Appropriate control of the compliance cycle is key to keeping taxpayers in compliance and preventing illicit trade and tax avoidance. For this reason, most tax administrations focus a majority of their resources on preventive policy. Some examples of this can be found in the strategic plans of the United Kingdom and the United States’ Internal Revenue Service (IRS). In the United Kingdom, HMRC has had a well-developed strategic plan for years. A key pillar of the plan focuses on keeping taxpayers compliant. This is the concept of prevention: controlling initial minor noncompliant behaviour for the majority of taxpayers, while using strong enforcement for the minority on the noncompliant side (28). The strategic plan of the IRS has a similar approach, with a focus on control. If noncompliance is detected, data analysis and behavioural insights are used to identify the best way to address noncompliance. Early intervention or self-correction are examples of ways to address detected noncompliance. The IRS also highlights the importance of resolving noncompliance to ensure taxpayer confidence in the tax system and protecting the integrity of the system (29). 116 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Risk-based approach Following the establishment of a strategic plan, an enforcement and control plan must be drafted. This plan should include definitions of the activities that will be enforced, the taxpayers upon whom they will be enforced and the circumstances under which they will be enforced, as well as allocating resources for staffing, audit- ing, infrastructure and IT. Targets must also be defined, including the number of interventions and the amount of additional collected revenue or reduction of tax evasion. Several tax administrations elaborate annual plans with periodic perfor- mance reviews aimed at improving results, and they correct allocations and targets as needed. Clear targeting of interventions is needed for better results, more efficient use of resources, lower costs for taxpayers and more effective collection. In other words, the point is to focus interventions on those who have a higher probability of noncompliance. Using a risk-based approach can be particularly beneficial. Tax risk management is a key element of control strategy in modern tax admin- istration. A risk is a possible threat to reaching objectives such as collecting taxes in an effective and efficient way for competent authorities. Risk assessment is the process of analysing risks and deciding on the best way to manage an identified risk. The responses can vary from acceptance to mitigation to avoidance. Proper risk assessment allows competent authorities to use their available resources most efficiently and to become more effective in dealing with risks. It can be used to improve compliance by identifying taxpayers or types of activities with a high risk of noncom- pliance. Groups of taxpayers with the same characteristics often have similar risks. Groups with a high risk of noncompliance could then be subject to greater review. Areas of potentially greater risk of noncompliance in the tobacco supply chain include import, export and transfers to and from warehouses, particularly when they take place under duty suspension. Gathering risk-related information from internal and external sources is a best practice in compliance risk management. Such sources could include third-party information (e.g. from banks, credit card companies, transport companies), studies on taxpayer behaviour and research on compliance issues, tax gap analysis, tax audits and declarations (30). Risk management uses these different sources of data along with algorithms to find patterns of high noncompliance. Risk analysis can indicate reduced risk as well. Lower-risk areas are likely to need less governance to ensure compliance, which allows for resources to be directed elsewhere. Risk assessment can therefore help with strategic allocation of limited resources to the areas of greatest risk while at the same time reducing the burden on lower-risk taxpayers. Risk management has always been done by competent authorities, but data availability and statistical methods to identify patterns have changed the way risks CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 117 can be assessed. Although many risk assessment systems are still done manually or include manual elements, the use of intensive data techniques allows systematic, deeper and more targeted analysis (31). Modern risk assessment makes use of electronic data on taxpayers, tax payments, declarations from other taxes, such as VAT, and third-party information. With these data, tax authorities can identify indicators that suggest where further activities might be required to ensure compliance. For example, VAT invoices can be used to match reported purchases of inputs of tobacco leaf to sales invoices of tobacco leaf wholesalers. For taxpayers (i.e. those who are licensed and provide required reports), competent authorities can create a business analytics program to determine whether the data reported are consistent on each side of the transactions. Moreover, in countries with a VAT system, competent authorities can compare the data reported by taxpayers under the VAT system with data reported under the tobacco excise tax system to detect any inconsistency. VAT invoices can also be used to verify inputs and sales data. If VAT is collected at all levels of the supply chain, it is easier for govern- ments to monitor the supply chain for the enforcement of excise duty obligations. Regular surveys on tobacco consumption that use the same methodology can also provide indications about the level of compliance with excise tax policy. A sudden drop in revenue that is not reflected in consumption data could be an indication of illicit manufacturing, illegal imports, cross-border shopping or forestalling. In addition, seizure data can provide valuable information on areas and activities at high risk of noncompliance. The structure of tobacco tax policies should also be taken into account when con- ducting risk analyses. If excise tax rates are increased, there might be a greater risk of forestalling or front-loading (see the discussion on anti-forestalling later in this section). Differentiated excise tax rates based on product or packaging characteristics – such as distinctions between soft and hard packs or filter and nonfilter cigarettes – are also prone to manipulation by operators, which could affect tax revenues. One of the options to mitigate these risks is to amend the excise tax policy and apply a uniform tax rate. For customs transactions, the use of risk management is a key element in target- ing merchandise and support declarations to be inspected. Before the 1990s, most customs agencies used random criteria for selecting targets for inspection. Since that time, many countries have implemented risk-based approaches for selecting inspections. Historical data on importers and trade communities, complemented by artificial intelligence technology, show that risk management tools dramatically increased the effectiveness of physical inspections. Most modern customs agencies have implemented such techniques, allowing for more effective control processes while facilitating smoother processes for those transactions that are in compliance. Box 3.3 details some of the recent changes in risk management processes. 118 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.3 Changes in risk management The OECD developed Fig. 3.2 to show the framework and key steps for understanding compliance risks in 2004 (32). The same approach is still used to identify, assess and prioritize risk. However, many competent authorities now use new technologies and advanced data analytics, along with more information sources, including external data from banks, employers and sales invoices for VAT, for example. Fig. 3.2 Compliance risk management process Source: (32). The methods of identifying risks and the analysis of compliance behaviour have also changed. Traditionally, competent authorities used audits to identify high-risk cases. With more diverse and better data, competent authorities can now use more evidence-based approaches to examine risk patterns. Success of compliance activi- ties is now more often measured in terms of their impact on the overall compliance environment, rather than only on increased revenues.9 9 For more information on effective risk management with several indicators and a checklist of questions, see the Tax Administration Diagnostic Assessment Tool (68). OPERATING CONTEXT Assess and prioritize risks Evaluate compliance outcomes: - Registration - Filing - Reporting - Payment Monitor performance against plan Analyse compliance behaviour (causes, options for treatment) Determine treatment strategies Identify risks Plan and implement strategies CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 119 In Indonesia, use of the compliance risk management process reduced the share of illicit trade in total consumption of cigarettes from 12% to 3%. More information on this can be found in the case study of Indonesia in Box 3.13 later in this chapter. Understanding the products – as well as the supply and distribution chains – al- lows competent authorities to identify which areas along a chain pose the greatest risk and therefore require more resources. Detailed information on the composition of selected tobacco products is given in Annex 3.1. KEY TAKEAWAY 8 Risk analysis helps identify the points of intervention that have higher probabilities of noncompliance. A risk-based approach with targeted interventions allows for better results and more efficient use of resources to ensure effectiveness of tax collection. 3.4.2 CONTROLS OVER THE TOBACCO SUPPLY CHAIN As defined in Article 1 of the Protocol, the supply chain covers the manufacture of tobacco products and manufacturing equipment – as well as their import or export – and may be extended, where relevant, to one or more of the following activities when so decided by a Party: 1. retailing of tobacco products 2. growing of tobacco, with the exception of traditional small-scale growers, farmers and producers 3. transporting of commercial quantities of tobacco products or manufacturing equipment 4. wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Article 4.1 of the Protocol requires parties to “adopt and implement effective mea- sures to control or regulate the supply chain of tobacco products to prevent, deter, detect, investigate and prosecute illicit trade in such goods and to cooperate with one another to this end”. Concrete measures to regulate the supply chain, as well as best practices in this regard, are discussed further below. Figure 3.3 shows the main places for reporting and monitoring along the supply chain: import, ex-factory and removals from warehouses. Manufacturers could be required to report imported inputs at the border, as importers of finished products do. If components are subject to licensing, information can be required as part of the licensing process. The arrows in Fig. 3.3 represent transporting, which is also part of the supply chain. 120 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 3.3 Cigarette supply chain from manufacture or import to retail sale 3.4.3 LICENSING10 AND DUE DILIGENCE A licensing system is effective only if it is properly controlled. Most tax administra- tions have experience with licensing processes for excise taxes on products such as alcoholic beverages and energy products. It is strongly recommended that lessons learned with the licensing process of such products be applied when implementing and enforcing tobacco-related licensing. Licensing provides timely and accurate data that can serve as the basis for audits because it identifies and controls legitimate operators. For new operators, the process to obtain a licence could include visits and verification of production factories, storage facilities and distribution premises. Countries that have no licensing system in place and would like to start applying licences could allow a transitional period for existing operators to comply with the new licensing requirements. The process of licensing control must be carried out and updated periodically, in particular by controlling the validity of bonds or guarantees, the proper functioning of the required systems (CCTV, for example) and recordkeeping. 10 Licensing is discussed here in the context of due diligence and enforcement. Details about how licensing can be set up and what information could be requested are presented in section 3.3.1. Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blend d. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: wher he tobac o will be roll d into st cks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 121 Where licences are required, the law should include a provision specifying that purchases from unlicensed suppliers – or sales to unlicensed purchasers – are not allowed. This means that both suppliers and purchasers would need to verify those with whom they are doing business. This requirement provides enforcement au- thorities with an entry point to enforce the licensing system at both ends. Also, a licensing requirement for manufacturing equipment assists authorities in identifying and prosecuting illegal manufacturing of tobacco products, reducing the burden of proof substantially. In many countries, the presence of manufacturing equipment is not sufficient proof that illegal manufacturing is taking place; the machinery has to be in operation and producing illegal tobacco products when authorities inspect the location. With a licensing requirement, however, the presence of machinery without a licence is sufficient for authorities to act. The validity of licences should be time-limited, requiring renewals or reapplica- tion, to maintain a high level of control. Adherence to the conditions required for a licence should be controlled by the authorities, and penalties for noncompliance – for example, suspension or withdrawal of a licence – should be severe enough to act as a deterrent (33). Regulations for licensing should provide for inspection of the licensee’s products and premises, with penalties for noncompliance, which could include criminal and civil prosecution for serious or repeated offences. As stated in Article 6.3(a) of the Protocol, Parties need to establish or designate a single authority or multiple authorities to issue, renew, suspend, revoke and/or cancel licences. In accordance with Article 7 of the Protocol, persons engaged in the supply chain are required by law to conduct due diligence before and during business relationships. They also must report to the competent authorities any evidence that a customer is engaged in activities in contravention of its obligations arising from the Protocol. This requirement includes customer identification, monitoring of sales to ensure that the quantities are commensurate with demand for such products within the intended market and taking measures to ensure compliance. Knowledge of the demand of a market is indispensable for determining if there is a case of oversupplying. If the supply of tobacco products to a lower-taxing foreign market exceeds the demand, it creates a higher risk that these products will be smuggled back into a higher-taxing country, undermining the objectives and effectiveness of the higher-taxing jurisdiction. In the past, some governments decided to impose a fine on tobacco companies if the quantities supplied were significantly higher than the demand and the risk of being smuggled back into their jurisdiction was judged to be high (34,35).11 11 Excise duty rates applied in all the EU countries can be found on the European Commission’s webpage: https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco-energy/ excise-duties-tobacco_en. 122 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 9 Licensing helps to identify and control legitimate operators. The data obtained from licensing can serve as a basis for audits. Licences should be controlled on a regular basis and updated periodically to ensure their validity. 3.4.4 FISCAL MARKINGS (E.G. TAX STAMPS) Fiscal markings are another important tool for controlling and monitoring pro- duction and import of tobacco products. Their use is generally considered to be appropriate for increasing compliance with tax laws. Fiscal markings can also be of help in distinguishing between genuine and illicit tobacco products. Tax stamps or other fiscal markings affixed to packs of cigarettes or tobacco products facilitate the collection of excise taxes, as well as audits and enforcement actions. The pres- ence of fiscal markings enables both the competent authority and the public to monitor whether the taxes on tobacco products were properly paid. It thus assists the competent authority in investigating illicit trade and prosecuting violations. Fiscal markings include tax stamps, enhanced tax stamps (banderols) and digital tax stamps. Examples of fiscal marks are tobacco stamps, tax stamps, excise stamps, tax stickers and banderols. Box 3.4 presents details on the different types and features of tax stamps. Tobacco products for export are often required to be marked that they are for export. Box 3.5 provides useful information regarding the International Organization for Standardization (ISO) standard for excise tax stamps. The terminology “fiscal mark” holds no indication of the characteristics of the mark. A fiscal marking is affixed to each pack of tobacco product. Requiring a standard package size can facilitate the application of the markings (2). Fraudsters can be deterred from attempting to re-use fiscal markings (in particular stamps) by having the marking affixed to each pack of cigarettes (or other tobacco product) before the pack is wrapped with cellophane (36). In most cases, tax stamps are purchased by the producer or importer and applied to each product sold as proof of excise tax payment (33). Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Box 3.4 Types and features of tax stamps Over time, tax stamps and markings have become more sophisticated. In the past, tax stamps were often paper-based and easy to counterfeit. New tax stamps use additional security features to make them more difficult to counterfeit. Authentica- tion solutions against counterfeiting can utilize various security features, including: • overt features – features that can be verified by the naked eye; CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 123 • covert features – features that can be authenticated only by using dedicated and specialized electronic readers; • semi-covert features – features requiring a simple tool that does not involve extensive training; and • forensic features – features that can be identified through laboratory analysis. Tax markings can be either physical or digital: • physical markings – the information is contained in the document or device attached to the package. • digital markings – information is obtained through a link with a database and by decrypting with the tools and keys used for creation of the data. The term “digital tax stamp” sometimes leads to confusion, as some paper-based stamps with digital components are also described as digital tax stamps. Tax markings that are fully digital do not contain information in the document or device attached to the package. It is probably too simplistic to say that digital tax stamps are more secure than paper- based tax stamps. Both types have advantages and weaknesses. For instance, both physical and digital tax stamps can be weak or strong on security features. Neverthe- less, the management, production, sales, transport and monitoring of physical tax stamps require increased attention. Tax stamps have the same value as banknotes and are a possible attraction for theft, loss and fraud. In Belgium, for example, the Court of Audit severely criticized the lack of control of the production and stock management of tax stamps in 2015. The Court of Au- dit concluded that tax stamps issuance should operate under recognized security practices and procedures relative to the security risk associated with the various production, distribution and issuance processes. Moreover, it was noted that new printing technology of digital tax stamps on packs may facilitate stock management and lead to less fraud. In March 2016, Belgium changed its stamps. The printing became an in-house process by the financial federal government department. The new stamps are still printed with a watermark, but they also have a digital component. The change resulted in a cost reduction by standardizing the sizes and optimizing the production process. Some of the more advanced fiscal marking technologies include embedded threads and watermarks; special inks and coatings, such as so-called invisible inks, holograms and foils; and calculated or changeable content. Because of their enhanced security features, these stamps can be more expensive than traditional stamps. In the state of California in the United States, the traditional stamps cost US$ 0.42 per 1 000 stamps. The cost of the first generation of high-tech stamps was 10 times higher, at US$ 4.77 per 1 000 stamps. This price nearly doubled for the second-generation encrypted stamp, to US$ 8.20 per 1 000 stamps. Nevertheless, California collected 124 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N about US$ 450 million of additional tax revenue in the first decade following the implementation of encrypted tax stamps. This additional revenue was far greater than the costs of implementation and enforcement. Other jurisdictions have also revised their tax stamps to incorporate new technolo- gies. The state of Michigan, for instance, replaced heat-applied cigarette tax stamps with digital pressure-applied stamps in 2015. Michigan deployed a tax stamp with several overt and covert security features and a unique quick response (QR) code and serial number. QR codes (machine-readable codes consisting of an array of black and white squares, typically used for storing URLs or other information) can have purposes beyond tracking and tracing. The QR code can be read by consumers with a smartphone or tablet application to access information on smoking-cessation programs, report violations of the state’s youth access policies, connect to a tip line to report noncompliant packs and learn about the harms from illicit tobacco sales and purchases. Enforcement authorities can validate stamps using the smartphone- based eTRACS (Electronic Tax Reporting and Audit Compliance System). As part of the system’s implementation, the Michigan State Police department created teams of enforcement officers in each of the state’s seven districts and the state Department of Treasury created its own enforcement team. Sources: (37–42). Box 3.5 The ISO standard on excise tax stamps The ISO published its excise tax stamp standard (ISO/TC 292/SC) in October 2018. The purpose of the ISO standard is to assist tax and finance authorities in enhancing compliance with excise tax regulations. A tax stamp is defined as a visible tax stamp, label or mark placed on certain types of consumer goods to show that the applicable excise tax has been paid. The ISO standard applies to tax stamps that are physical in nature – not to digital markings, which are directly printed on to packs without a physical component. “Authentication” in this standard refers to the authentication of the tax stamp, not the product on which the tax stamp is affixed. In other words, authentication of a tax stamp on a cigarette pack means that the tax stamp is authentic but does not guarantee that the pack is authentic. In addition, control measures are needed at the time of the application of the stamp to verify the conformity of the tax stamp with the corresponding product. The standard provides guidance on the content, security, issuance and examination of physical tax stamps used to indicate that the required taxes have been paid and that the tax stamp is authentic. The use of stamps to facilitate tracking and tracing within the supply chain is not described. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 125 Specifically, the ISO standard deals with the following issues: • defining the functions of a tax stamp • identifying and consulting with stakeholders • planning the procurement process and selection of suppliers • the design and construction of tax stamps • the overt and covert security features that provide protection of the tax stamp • the finishing and application processes for the tax stamp • security of the tax stamp supply chain • serialization and unique identifier codes for tax stamps • examination of tax stamps • monitoring and assessing tax stamp performance. A stamp may fulfil many functions, but the core business of tax stamps is to ensure and facilitate the collection of revenue. The tax stamp must use a combination of security features. The tax authority should ensure that the tax stamp can be authen- ticated and that counterfeit, altered, tampered or otherwise fraudulent tax stamps can be detected. The standard provides detailed information on the different components of the tax stamp such as the substrate, inks, adhesives, laminate, authentication or security features and the unique identifier that should enable checks on the payments of the required tax. The process of procurement is discussed in detail in the standard. The tax author- ity should ensure that the procurement process is open, transparent and meets the sustainability objectives. The tax authority should set out the goals and requirements to give tendering organizations more leeway in proposing optimum solutions that might be different from those the authority would specify. The standard is not prescriptive; rather, it provides a catalogue of options. It does not, for instance, recommend specific security features, but it does describe the different types of features that are necessary for a tax stamp to be secure. Tax officials still need to make decisions and choose the option that suits them best, but the standard remains recommended reading for those who would like to introduce tax stamp programmes in their jurisdiction. ISO standards are not freely available but can be purchased at the ISO Store (www.ISO.org) or from an ISO national member body. Source: (43). According to Article 8 of the Protocol, each Party shall require that unique, secure and nonremovable identification markings – such as codes or stamps – are affixed to or form part of all unit packets, packages and any outside packaging of cigarettes for the implementation of the tracking and tracing system within a period of five 126 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N years, and of other tobacco products within a period of 10 years, of entry into force of the Protocol for that Party. The Protocol specifies that at least the following information shall form part of the unique marking: • date and location of manufacture • manufacturing facility • product description • where available, the intended market of retail sale. In several countries, QR codes are used as fiscal markings for tobacco and alcohol tax control. Each stamp has a unique identifier code and a QR code. The data stored in the QR code provide the following product information: • manufacturer • production location • stamp order date • tax status and class • brand • intended market • unique identifier (serial number). KEY TAKEAWAY 10 The use of fiscal markings is generally considered to be an appropriate tool for increasing compliance with tax laws. Fiscal markings can also be helpful for distinguishing between genuine and illicit tobacco products. 3.4.5 TRACKING AND TRACING A tracking and tracing system assists authorities in determining the origin of tobacco products – and the point of diversion, if applicable – as well as monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to enable authorities to have information on all transactions through the entire tobacco product supply chain until duties are paid or other obligations are discharged. Traceability is not used only for tobacco products. It is also used to improve the supply chain function, as in the case of parcel services, as well as for product safety reasons, to manage potential product recalls and for regulatory reasons. Tracking is the process that monitors where a product is at all times while also creating a time and location record for all movements. Tracing is the ability to identify the past locations of a product, so that the product’s route can be followed CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 127 back to its origin (44). In other words, traceability is “the ability to trace the history, application or location of an object” (45). A tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it is possible to unambiguously register that product’s movements. Other necessary characteristics include the ability to share the registered move- ment information and to authenticate products. This enables a product’s status to be captured through the supply chain and its history to be identified and verified retrospectively. According to Article 8.4.1 of the Protocol, Parties should require the following information to be available: • the date and location of manufacture • the manufacturing facility • the machine used • the production shift or time of manufacture • the name, invoice, order number and payment records of the first customer not affiliated with the manufacturer • the product description and intended market of retail sale • any warehousing and shipping • the identity of any known subsequent purchaser • the intended shipment route, date, destination, point of departure and consignee. A good tracking and tracing system enables the government to properly monitor the supply chain, improves its ability to ensure collection of the proper duties and taxes, provides it with the ability to authenticate whether the identification marking is genuine and matches the product and improves its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. The following elements are required for an effective tracking and tracing system (46): • A serialized unique identification marking for each package of product. These identifiers are a distinctive combination of numbers, letters or both. They cannot be predictable or used more than once. The representation of the identifier on the package can be human-readable (letters or numbers) or machine-readable (barcodes). Generation of codes and encryption that are part of a tobacco industry patent should be excluded. • A data carrier with the serialized unique identifier and other information such as date and location of manufacture, manufacturing facility, product description and, where available, the intended retail market. This informa- tion should be readable by authorized agencies of any Party to the Protocol. The data carrier should comply with quality standards and be suitable for 128 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N high-speed production lines. Two-dimensional barcodes, for example, meet these standards and are readable with inexpensive equipment. • A link and parent-child relationship (called aggregation) between different packaging units that offers the option to trace a pallet without the need to scan all the packs and master cases of that pallet. • Recordkeeping of all shipping and receiving events along the supply chain. This includes, for example, the departure location and the arrival location, as well as the involved operators. International standards from the ISO are recommended for the capture and exchange of data and events. • The use of international standards for key information that is encoded in the data carrier (5). An example of a unique and internationally recognized identifier for products is a Global Trade Item Number. The following details on information storage and sharing are drawn from various sections of the Protocol. Data and events along the supply chain must be stored in an independent database that is controlled by competent government authorities. At the global level, national and/or regional databases can be interconnected to facilitate international inquiries by competent authorities. Parties to the Protocol agree to establish a global information-sharing focal point located at the Conven- tion Secretariat of the WHO FCTC, accessible to all Parties, enabling them to make enquiries and receive relevant information. Each Party shall ensure that the information recorded under paragraph 5 of Article 8 of the Protocol is accessible to the global information-sharing focal point on request, subject to paragraph 9, through a standard electronic secure interface with its national and/or regional central point. The global information-sharing focal point shall compile a list of the competent authorities of Parties and make the list available to all Parties. The cost of tracking and tracing systems is a concern for many countries, but as indicated in paragraph 14 of Article 8 of the Protocol, jurisdictions may require the tobacco industry to bear any costs associated with putting in place the tracking and tracing system in a country (46). In Brazil, the cost for cigarette manufacturers was US$ 0.0185 per pack (42). In Kenya, the cost for manufacturers was US$ 0.024 per pack (42). Along with considering the characteristics of a tracking and tracing system in selecting a particular one, it is important to avoid conflicts of interest, ensure fair and transparent dealing with suppliers, implement a zero-tolerance policy for corruption or anti-competitive behaviour and ensure compliance. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, which deals with industry interference, and Article 8 of the Protocol. Article 8.13, which states that “each Party shall ensure that its competent authorities, in participating CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 129 in the tracking and tracing regime, interact with the tobacco industry and those representing the interests of the tobacco industry only to the extent strictly necessary in the implementation of this Article”. Box 3.8 provides a cautionary example of a tracking and tracing system that is not compliant with Article 5.3 of the WHO FCTC. While the Protocol contains a great deal of information on the requirements that a tracking and tracing system should meet, questions come up in relation to the implementation of such systems. To achieve the objectives of the Protocol, the Meeting of the Parties (MOP) to the Protocol, as the governing body of the treaty, has the prerogative to establish subsidiary bodies, such as expert groups and working groups. In decision FCTC/MOP1(6), the MOP established a working group for the development and implementation of tracking and tracing systems in accordance with Article 8 of the Protocol, including the global information-sharing focal point (Article 8.1) and unique identification markings for cigarette packets and pack- ages (Article 8.3), to further elaborate on the next steps. The working group will produce a comprehensive report compiling good practices and experiences on the implementation of tracking and tracing systems, as well as unique identification markings for cigarette packets and packages at national or regional levels. The working group was also given a mandate to prepare a conceptual analysis of how a global information-sharing focal point could be set up. Implementing a complete tracking and tracing system with fiscal markings takes time. In most of the countries that have already implemented tracking and tracing, it took several years from starting with the legal framework to final implementation. Several hurdles need to be overcome: • Legal framework approval is usually delayed by the tobacco industry. • Knowledge of tracking and tracing and associated technologies is scarce at tax administrations. • Tender and bidding processes are complex. • Coordination between domestic tax authorities and customs is weak. Even though the process might be lengthy, the investment in a tracking and tracing system will be repaid with the amount of tobacco taxes that are not lost due to evasion. When implementing a new tracking and tracing system, tax administration should ask for collaboration and technical assistance from intergovernmental organizations and countries that have successfully implemented such systems, in order to speed up and ensure success of the process. Examples of tracking and tracing systems implementation in Chile, Kenya and the EU are detailed in Boxes 3.8, 3.9 and 3.10. For countries that already have fiscal markings in place, the potential interaction between the markings and the implementation of a tracking and tracing system should be taken into account. Further information on this interaction is presented in Box 3.6. 130 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.6 Tracking and tracing and fiscal markings It is becoming increasingly common for stamps to contain some tracking and trac- ing features, such as unique identification markings and basic information on the product that describes the company, tax status or the product itself. The intention is to mark each pack with a unique identification marking so it can be monitored from the point of production to the retailer, including each step in between, thereby creating a complete time and location history. Although a tax stamp could meet the requirements of Article 8 of the Protocol and have tracking and tracing features, in general, the focus of tax stamp systems differs from that of tracking and tracing systems. Tracking and tracing is more than the unique, secure and nonremovable identification markings on the packages of tobacco products. It implies reading or scanning the codes; linking the codes between packs, cartons, master cases and pallets; uploading the information to a database; recording of any shipping and receiving events along the supply chain; and interconnecting the different databases. While new tax stamp programs contain tracking and tracing features, they are primarily intended to facilitate tax collection on the domestic market and not to track duty-suspended cross-border trade or the export of products. The focus of tax stamp systems is on authenticity and the proof that taxes are paid. The focus of tracking and tracing systems is on unique identification and on control of the movements in the supply chain by monitoring and investigating the past and future location of products. Tax stamp programs focus on stock management, verification (that the stamps correspond to the product) and authentication (that the stamps are genuine), while the focus of tracking and tracing systems is on the origin, intended route, first customer and final destination. The focus of tax stamps is primarily on individual packs intended for the duty-paid domestic market, while the focus of tracking and tracing systems is on all packaging (packs, cartons, master cases, pallets) and certainly – but not exclusively – for the duty-suspended export market. Nevertheless, sometimes there can be synergies. For example, the EU countries that require a tax stamp or national identification mark for fiscal purposes have the option to use it as the security feature for tracking and tracing purposes, provided that the requirements are met. In summary, tax stamps can be converted to or be part of a tracking and tracing system when the converted system provides aggregation between packs, cartons and master cases and records all movement along the sup- ply chain. For the export market, a unique identification marking should be added. Sources: (39–42). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 131 Box 3.7 What not to do: use the industry solution to tracking and tracing Codentify is a serialization system used to produce cigarette pack markers for the purpose of verifying whether cigarette packs are legal. It was patented by PMI but subsequently licensed at no cost to other major cigarette manufacturers. In 2016, Codentify was transferred to Inexto, which is an affiliate of the French group Impala. What is the problem with Codentify/Inexto? The main issue is that Codentify/Inexto’s links to the tobacco industry make it incom- patible with the Protocol, which came into force in September 2018. The Protocol specifies that obligations assigned to a Party shall not be performed by or delegated to the tobacco industry. Additionally, many elements indicate it is an ineffective means of authentica- tion. For example, the 12-character digital codes generated by Codentify can be easily duplicated or cloned and used as originals on either a counterfeit or genuine pack, which can then pass the system’s basic verification test. The codes are also produced by relatively unsecured, commercially available equipment and do not include high-security features capable of protecting the authenticity of identifier numbers. Systems that use multilayered, advanced security solutions that enable distributors, retailers, customers and authorities to identify noncompliant products are more secure. Another problem is that Codentify/Inexto cannot track products as efficiently as other available systems. It requires a much larger enforcement capacity to achieve the same detection rates as other systems that are not linked to the tobacco indus- try. Authorities would have to inspect significantly more packs marked under the Codentify system than is necessary under some other systems to achieve the same certainty of not missing a fraudulent pack. In addition, not all stakeholders will be able to verify that a pack marked under the Codentify system is genuine, while other available systems do offer this possibility. Source: (47). 132 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.8 A successful tobacco traceability system: SITRAF, Chile The Servicio de Impuestos Internos (SII, Internal Revenue Service of Chile) has success- fully coordinated and implemented a tracking and tracing system for tobacco products. Application of a compliance management model In the framework of a compliance management model, a traceability system for tobacco products utilizes a structural measure to reduce tax evasion. It is estimated that evasion of taxes (VAT and excise taxes) in the cigarette market in Chile amounts to 16.6% of the country’s total market annually – approximately US$ 300 million. The traceability system implemented in Chile (SITRAF (TAB2)) allows authorities to know in a certain and timely manner the quantities of cigarettes produced or imported into the country. Moreover, it helps authorities to distinguish between counterfeit products and original products that did not comply with payment of the tax. In 2018, the implementation of the traceability system was awarded to a company through public bidding. The company is in charge of the implementation and operation of the system for five years, according to a contract signed with the SII, and must maintain a team of 20 people available for the project. Direct markings are applied to items produced in Chile for national consump- tion, and stamps are used for imported products. For both types of product, the marking is based on a data matrix code, which is printed using security ink that is distinguishable from any other type of ink with specific devices that are provided by the awarded company. Although products for export are not subject to marking, they are controlled and accounted for by the traceability system. For national production, devices are installed on each production line that rec- ognize the type of pack being produced, print a unique code on each pack and then read it (activation) to save all the information on servers located in the production plant. This information is transmitted to the central servers of the system and then to the SII. It is also available for on-site inspection. In the case of imported products, the stamps must be acquired in Chile by each importer and then sent to its producer abroad, which is responsible for adhering them to each pack of cigarettes prior to wrapping the packs with cellophane, us- ing applicators on the production lines. Once the cigarettes enter Chile, the tax determination process has been completed in the service (Provisional Free Transit Guide) and the corresponding taxes have been paid to customs (Import Declara- tion), the importer must enter the data on the stamps used by the importer on the platform of the traceability system. After validation, the stamps can be activated in the system – that is, they are recognized as valid for commercialization. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 133 In addition to the devices provided by the company, SII has developed a smart- phone application for verification by citizens. Although the application cannot verify the authenticity of the ink used, it is able to verify whether a code is correctly gener- ated and display the information contained in the traceability system for the code (brand, variety, quantity of cigarettes, products/importer) so that the taxpayer can verify its consistency. Progress of the compliance management model Implementation of the traceability system has required a coordinated effort both within the service – for the generation of instructions, procedures and computer developments – and with other institutions, such as the National Customs Service and the Ministry of Health. It is an unprecedented project in terms of coordinating the implementation of the system in the production lines of the different tobacco companies in the country and the provider company. Some of the main milestones of the project are: September 2014: Law 20,780 on Tax Reform establishes an obligation to implement the system within a term of six months, after the publication of the resolution determining the obligated tax payers. May 2015: Resolution No. 47 determines obligated taxpayers. June 2015: Circular No. 47 describes obligation to incorporate stamps or distinctive marks as a traceability mechanism. February 2016: Law 20,899 on Tax Reform simplifies the definition of the system, allowing the system to be outsourced or provided by the SII, in addition to making the type of traceability more flexible. August 2016: Traceability system regulation D.S. 1,027 is issued (published on 28 December 2016). March 2017: Exempt Resolution No. 49 of the Ministry of Finance authorizes the SII to outsource all or part of the traceability system. June 2017: Bidding bases in public market are published. February 2018: Tender is awarded to selected company. June 2018: Decision of contract is made by General Comptroller of the Republic. August 2018: Resolution No. 61 determines taxpayers obliged to apply the trace- ability system. August 2018: Holding of first workshops for detailed definitions of the project, with the participation of Customs, Ministry of Health, provider and SII. September – October 2018: Visits of plants to coordinate with producers and define adaptations to production lines for system implementation. January 2019: Resolutions No. 6, 7 and 8 passed, with response to a request for an 134 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N extension of producers; start of system implementation in all production lines in the country. February 2019: Resolution No. 16 establishes a term to commercialize the remaining stock without marking. 2019: Resolution No. 24, with request for extension to importer, includes training of customs staff. March 2019: The traceability system is started up. June 2019: Stamp process begins for all imported cigarettes. Currently, the system is installed and operating in all production lines in the coun- try, placing traceability markings on virtually 100% of the cigarettes produced and imported. During 2019, the traceability system enabled controlling approximately 1 175 million packs: 744 million produced for national consumption, 409 million produced for export and 22 million imported yearly. Source: (48). Box 3.9 Case study of Kenya’s implementation of a tracking and tracing system Kenya’s current tracking and tracing system was preceded by a series of reforms in both tax structure and administration of excise taxes. The reforms included electronic cargo monitoring of exports, which allowed for automatic monitoring and reporting. The system appears to be highly effective because it requires less capacity and is less prone to manipulation than earlier systems. The experience of Kenya shows that a lower-middle-income country can successfully implement a sophisticated system capable of decreasing illicit trade. It also shows the importance of other measures such as strengthening enforcement, increasing cooperation and communication among different agencies and increasing penalties for noncompliance. Illicit trade, as measured by the Kenya Revenue Authority (KRA), was estimated to be around 15% of total consumption in the market during the initial reform period. After the introduction of the new system in 2015, it dropped to 5%. Timeline of the major reforms: 2003: In this period, the paper tax stamps used had a unique identifier and were colour-coded to indicate the type of product. Regular compliance checks were in- troduced. In 2007, the cost of a stamp was 2.124 Kenyan shillings or US$ 0.023 per pack. However, the stamps were found to be easily counterfeited and could not be linked to specific brands. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 135 2008: The KRA proposed a tracking and tracing system and increasing tax rates. The new system was introduced gradually. 2010: Enhanced security features, including ultraviolet markings, were added to the paper stamps. The stamps were to be clearly visible when packs were displayed for sale and placed so that opening a pack would destroy the stamp. The stamps were verified at four different points along the supply chain. The costs were just slightly higher than those of the previous stamps at US$ 0.024 per pack. Licensing was introduced for domestic manufacturers, subject to annual renewal. Importers were required to register with the KRA. Licences required submission of details on the company directors, inventories and equipment, accounting systems, input-to-production ratios and brands produced. Penalties for noncompliance were increased and included up to three years in prison. An electronic cargo tracking system was launched. Electronic seals were affixed on containers or trucks, and GPS technology was used for tracking. A bond was payable on exports to cover excise and VAT taxes. The bond was released only when the goods reached the final destination and taxes were paid. Verification involving both countries of the business deal takes place at the bor- ders. The electronic system provides information about the departure and arrival of the goods and the disarming of the seals. Authorities in the importing country are notified before the shipment leaves the domestic production facility. The system reduces the number of checkpoints and staff needed and generates arrival reports that can be verified with VAT refund requests. As a result of these changes, three factories and seven of the 10 importers were shut down due to noncompliance. Exports to Côte d’Ivoire, Eritrea, Mali and Sudan stopped because companies could not provide evidence that the goods reached the final destination and taxes were paid. More than US$ 11 million in excise tax losses was recovered in 2011. The KRA estimated that illicit trade dropped to 8%. 2013: A contract was signed to introduce a tracking and tracing system for tobacco and alcohol, the Excisable Goods Management System, in April. The system added production counting, tracking and tracing, stock control, processing and other data collection to the existing system. Infrastructure requirements included high-speed broadband internet at production facilities, warehouses, the KRA and ports, along with reliable power or backup generators at those points. Implementation was planned in three stages: • Stage 1 – A new electronic digital stamp with a unique identifier was introduced. It included a data matrix code plus overt markings (holograms, fluorescent fibres, a security link for KRA authentication and visible two-dimensional codes for verification and activation), semi-covert markings (UV features, fluorescent prints detectable by specialized devices, mini text printing for retailers and 136 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N distributors) and forensic taggants for use in prosecutions. The stamp also included human-readable codes for verification by short message service using the KRA web portal. • Stage 2 – Control and monitoring systems were automated in February 2014. Manufacturers had to install photosensitive readers on production lines, with data automatically sent to the KRA in real time. Each stamp was activated and associated with a brand and package size on the line. The KRA database is automatically updated every 15 minutes. • Stage 3 – Market surveillance began, with 83 officers given powers to seize illicit cigarettes and make arrests. The officers were equipped with hand-held devices that transmitted data to the KRA for authentication. Distributors and retailers became liable for selling products without an excise stamp and were subject to fines plus prison sentences of up to three years for noncompliance. In 2016, a smartphone application became available with which the public could authenticate cigarette packs. Importers must now buy digital stamps and send them to export facilities in other countries to be affixed. Tax liability is due at removal from a factory or at import. The electronic cargo monitoring system is still in effect. 2016: The Excise Duty and Tax Procedures Acts clarified new obligations and penalties. 2017: A new integrated customs management system was launched. The KRA estimates that illicit trade levels are now around 5%. The current, more comprehensive digital system is cheaper than the previous paper tax stamp system. Manufacturers pay for the production monitoring system, but it counts as a busi- ness expense on corporate tax returns. In 2018, two manufacturers and 10 licensed importers were operating in Kenya. In 2018, aggregation between the markings of packs, cartons and master cases had not yet been implemented but was expected to be forthcoming. Source: (36) Box 3.10 Case study of the new EU tracking and tracing system, May 2019 Cigarette smuggling and other forms of illicit trade in the EU is estimated to cause a loss of €10 billion in revenue annually. In 2018, 4.2 million packs (20 sticks per pack) of illegal cigarettes were seized by customs in the EU. Illicit tobacco production was also increasing: an illegal factory in Ireland, dismantled in 2018, was capable of producing 250 000 cigarettes per hour. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 137 The EU tobacco control policy is described in the Tobacco Products Directive and is influenced by the Protocol. Article 15 of the Directive calls for the traceability of cigarettes and RYO tobacco products by May 2019 and of other tobacco products by May 2024. The EU tracking and tracing system is sufficiently flexible to be imple- mented at both the regional and the single-country level. Countries can choose among providers as long as the basic requirements are met. The policy provides a high level of protection against any attempts at manipulating the data. The report- ing obligations cover all the economic operators involved in the manufacture and distribution of tobacco products. The EU system requires all unit packets of tobacco products to be marked with a unique identifier Information on the movements of those products is to be stored by third-party data storage providers. The suppliers of the unique identifiers and data stor- age are to be financially and legally independent from the tobacco industry. The data are to be fully accessible to authorities of EU Member States for enforcement purposes. The generation of unique identifiers, as well as all other codes required for pre- registration of economic operators, facilities and machines, will be done at the Member State level by designated identifier issuers. Manufacturers and importers are required to supply information relating to the product and production lines when requesting unique identifiers from the issuers. The issuers will then generate and deliver batches of unique identifiers. On the production line, manufacturers of tobacco products will complete each unique identifier with a marking indicating the date and time. The unique identifier will be a machine-readable, optical, one- or two-dimensional barcode. An anti-tampering device, capable of creating an unalterable independent record of the verification process, must have been installed previously. This additional record will be accessible to public authorities for potential investigation and inspection. Unit packets, as well as aggregated packages such as cartons, master cases or pallets, can be tracked and traced throughout the supply chain. Tracking is also allowed at an aggregated packaging level as long as unit packets remain traceable. During transport, each dispatch and arrival up to the final dispatch to the first retail outlet must be recorded and reported. All recorded information must be submitted to the independent third-party data storage facility, generally within three hours, and 24 hours before dispatch and transloading. Costs, including operational costs, are shifted to the tobacco industry, in line with Article 8 of the Protocol. The EU system of tobacco traceability and security features became operational on 20 May 2019. Sources: (36, 49–50). 138 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 11 A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as monitoring and controlling the movement of tobacco products and their legal status. 3.4.6 ANTI-FORESTALLING “Forestalling” is a term that describes increases in the production or stock of products in anticipation of a tax increase (2). Other terms referring to this practice include “stockpiling” and “front-loading”. Forestalling occurs when manufacturers or importers increase their tax-paid stock or oversupply the market by increasing production or imports in order to pay the previous lower rate. It reduces and delays the effective- ness of tax measures. The effective starting date of the new tax rate will be delayed, revenues will be lower and the possible effect on prices and thus consumer behaviour will also be postponed. To illustrate how anti-forestalling measures function, an example is presented in Annex 3.2. A legal basis must exist for anti-forestalling measures; otherwise, the govern- ment cannot prevent the industry from forestalling. Legal measures to deal with forestalling include (51–52): 1. Limiting the amount of tobacco products that can be released subject to the old tax rate and levying the new tax on the products exceeding that limit. 2. Levying the new tax rate on all goods that are still in stock and not yet sup- plied to the final consumer. 3. Limiting the number of tax stamps issued at the rate that was in effect before the increase or limiting the time that products with a tax stamp with the old rate can be sold. 4. Requiring producers and importers to buy new tax stamps annually or after a tax increase. Under the first three measures, the competent authority determines the limit for taxation at the previous (lower) rate. The quantity allowed may be based on the shelf life of tobacco products – around six months for cigarettes – or normal inventory levels, such as an average over the previous three years. The first measure, limiting the amount of tobacco products that can be released, requires resources from the competent authority for enforcement. Authorities may decide to post inspectors in each production facility, but even without posting inspectors, procedures are necessary for determining when the allowed quantity has been exceeded and what subsequent actions to take. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 139 The second measure, levying the new tax on goods in stock and not yet supplied to the final consumer, might be difficult to implement. The competent authority is required to monitor the manufacturing process and, at the very least, to conduct a stock-taking of all players in the supply chain – including the various manufacturers, importers, wholesalers and retailers – in a very short window of time. If monitoring covers only the stock of manufacturers and importers, this measure could easily be circumvented by ensuring the stock is sold to others in the supply chain or by setting up separate distribution companies to purchase the stock. Controlling stock at the retail level is burdensome and might not be administratively feasible given the large number of cigarette retailers. It becomes even more burdensome if there is no licensing requirement for retailers because they will first have to be identified. The third measure accomplishes the same thing as the first if a country uses tax stamps. The fourth also requires tax stamps and is simpler for the competent authority but somewhat more burdensome for the tobacco companies, since stamps must be purchased every year. Box 3.11 provides examples of anti-forestalling measures in EU countries. Box 3.11 Examples of anti-forestalling measures in EU countries Several EU Member States have taken measures to limit forestalling. A cautious ap- proach seems to be required for designing such measures to ensure that they comply with EU legislation and the general principles of EU law – in particular, the principle of proportionality. No disputes have occurred concerning the right of initiative of EU Member States to implement anti-forestalling measures. Nevertheless, several EU Member States had to defend their measures in front of the Court of Justice of the European Union, the institution that ensures all national legislation is in line with EU law and a consistent application of that law (53). The Court acknowledged that anti-forestalling measures are appropriate to combat tax evasion and tax avoidance. Moreover, the Court emphasized that fiscal legislation is an important and effective instrument for discouraging consumption of tobacco products and therefore for protecting public health (54–55). However, the measures taken should be proportion- ate to the objectives. The principle of proportionality means that only the action needed to achieve the objective should be taken, and it should not exceed what is necessary. This principle regulates the measures taken within the EU and is included in the Treaty on the European Union. The Court demanded that Portugal amend its legislation to ensure compliance with the principle of proportionality. Belgium, Estonia and Hungary were also urged to change their anti-forestalling measures to bring them in line with EU legislation (56–57). 140 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The majority of the EU Member States have anti-forestalling measures in place, but there is no harmonization of these measures. The following are some examples:12 • Portugal limits the quantity of cigarettes that can be released in the last four months of a year to the average of the previous 12 months plus 10%. In addition, manufacturers and importers must sell cigarette packages with a tax marking of the preceding calendar year within three months. For other tobacco products, longer limits apply. • Denmark limits the number of tax stamps issued before a tax increase at the old rate to 20% more than are usually purchased in the two months before the end of the year. • In Poland, tax stamps are valid only for the current calendar year, and cigarettes with the old stamp can be sold only through February of the following year. • In Romania, companies must apply for approval to release for consumption from the customs office. Source: (58). KEY TAKEAWAY 12 Forestalling reduces and delays the effectiveness of tax measures. Implementing anti-forestalling measures can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. 3.4.7 ADDITIONAL NATIONAL AUDITS AND CONTROLS In addition to the previously described measures, several periodic audits and controls could be implemented to increase compliance with tax laws. The most common audits and controls are the following: • Cost audit – The cost audit method provides expected VAT and tobacco tax collection by simulating the intermediate and final cost of cigarettes. It starts with the inventories of raw materials and estimates added values and final cost, then matches the results with real collection from the tobacco supply chain. Annex 3.1 provides more information about the components that make up some selected (tobacco) products. • Transfer pricing audit – To ensure companies pay their fair share of tax, prices of transactions between related companies should be assessed, and when prices are not in line with the market conditions, they should be corrected. Companies that operate at the international level (transnational companies), including many tobacco companies, can manipulate import or export prices 12 Considering the frequency of court cases concerning anti-forestalling incidents, these measures might have been replaced or amended. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 141 of merchandise or raw material to related companies or branches in other countries, with the objective of lowering profits in countries with higher tax rates, and can transfer those profits to countries with lower taxes. • Price and market monitoring – Retail price surveys can provide information about variance from the market price in certain locations, highlighting areas of potential tax avoidance or illicit trade. Physical control of such locations requires rapid response teams, as implemented in the Philippines. To monitor the tax compliance of its taxpayers, the competent authority in the Philippines needs to understand the tobacco market; it must have information on brands, market segments and prices of products. This information enables authori- ties to estimate the impact of tax and price changes on consumer behaviour and revenues. Market data can be analysed as part of risk management and anti-fraud analysis to determine whom to investigate for noncompliance and when to do so. Sales data can be triangulated to validate other data sources, such as household surveys on prevalence. Market data and trends are also useful indicators for determining whether there is a case of oversupplying. See also section 3.4.3. • Consumer control – Involving the public via awareness campaigns has also been shown to be effective. Consumers have the right to be assured that the products available in the market are authentic and come from legitimate sources. Thus, it is in the consumers’ interest to understand and be able to verify that they are buying genuine products. The features of the fiscal marks on tobacco products should help consumers distinguish between genuine and illicit products. Some countries – Kenya, for example (see the case study in Box 3.9) – use a smartphone application to allow anyone to check both covert and overt features and to report any cigarettes with incorrect markings. Other countries, such as the Netherlands, have developed a smartphone application that allows anyone to report a suspected case of excise tax fraud. • Cross-check controls – Competent authorities should consider using multiple sources to obtain market data and determine if these data are consistent with tax declarations. VAT declarations can be used to verify that suppliers and purchasers of raw materials and final products are reporting the same amounts. Bank information can be used to verify both sides of transactions along the supply chain. Any discrepancy can alert the competent authority to conduct further investigation for possible illicit trade of tobacco or tax evasion. 142 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 13 Several different types of periodic audits and controls that can be carried out to increase compliance, including cost audits, transfer pricing audits, price and market monitoring, consumer controls and cross-check controls. 3.4.8 IMPORT AND EXPORT CONTROLS Parties to the Protocol should allow import and export of tobacco products and manufacturing equipment only by duly licensed natural persons or legal entities (Article 6.1). A well-known strategy used by fraudsters is to declare products for export so that no duties are due to the country of export. These products are subse- quently transported through other countries, using the in-transit regime that allows temporary suspension of duties until the goods arrive at their final destination. Before arriving at their final destination – where the excise duties would be due – the goods disappear or are lost while being diverted to the illegal supply chain. The goods may never leave the country, or they may be smuggled back into the country from which they were exported without declaring or paying duties. This risk of loss of revenue can be mitigated by requiring a guarantee or bond, which will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless the required fiscal marking (such as a tax stamp or export label) is affixed on the pack, according to the national law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Illicit tobacco trade could be decreased significantly if the various competent authorities that have jurisdiction over manufacturers and exporters of tobacco prod- ucts and manufacturing equipment would provide the competent authority at the destination with prior information when a shipment has been authorized and is about to take place. The information could include the name of the consignee, a description of the item shipped and the quantity. Also, the competent authority at the destination should inform the competent authority having jurisdiction over the shipper that a shipment was received along with the pertinent information relating to the shipment. A good IT system is also required for import and export. Electronic processing of prior-to-arrival manifest and import declarations is recommended. Most countries have implemented an online customs system to process import and export declara- tions, including all required data such as country of origin or export, description of merchandise, value, weight, cargo insurance, carrier, importer or exporter and broker identification, detailed tax duties to pay and final destination. The World Trade Organization (WTO) Trade Facilitation Agreement provides sev- eral tools for better controls, including collaboration between customs administrations, CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 143 exchange of information, use of non-invasive devices and prior-to-arrival import declaration (23). Non-invasive detection equipment at customs posts is highly capable of detect- ing contraband merchandise. The most common tools are X-ray scanners that are used for small parcels, containers, trucks and trains. Most modern ports have also implemented the use of X-ray scanners, and such technology is improving the speed of controls to as little as two minutes per container. Although the cost of scan- ners is declining, it remains inaccessibly high for countries with limited resources. Fortunately, scanners are often available for lease, making them accessible for tax administrations in those countries. Less sophisticated and less costly detection equipment includes endoscopes, mirrors, night vision equipment, cameras and automatic licence plate readers (33). A still less expensive alternative is the use of dogs, which can be trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied in order to reduce contra- band. These include the separation of processing operations from the sealed storage of taxed and untaxed products. Physical and direct control by officials of the excise authority during a part or the whole of an operation can be applied (for example, physical escort of the transit consignment from border to border by individual trucks or in a convoy, or application of radio or satellite tracking systems such as GPS-enabled devices to goods, conveyances, vehicles or containers). Control at borders is essential and should include integrated technology and cooperation with agencies at the border station. Frontline officers should be sup- ported by appropriate intelligence, guidance and supervision from management, as well as technical aids to enforcement. Within a country, mobile excise control units are helpful for verifying excisable goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. These operations require close coordination between police, border guards and other public services. Exports also require special attention, in particular if VAT and tax refunds are granted to the export of tobacco products. Validating the real exit in the declared amount is essential to avoid illegal re-entry to a territory and the improper refund of taxes. For any tax refund, an audit including tax credit information must be carried out. The audit may include the invoices for the whole chain involved in the export, including tobacco farmers, first processors, manufacturers, wholesalers, storage and transport. 14 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 14 To ensure control of import and export, it is recommended that only duly licensed persons and entities be allowed to import and export tobacco products and manufacturing equipment. 3.4.9 FREE ZONES AND TRANSHIPMENT POINTS The term “free zone” is very broad and can refer to a number of different types of areas. The Financial Action Task Force listed the following types in its 2010 report (59): free trade, export processing, enterprise, free ports, foreign trade, special economic zones and bonded warehouses. A number of these areas can include tobacco manufacturing and trade. By definition, controls such as regulation and oversight within free zones are less strict than in other areas. This can make them appealing to persons involved in illegal cigarette manufacturing or trade (17). In fact, illicit activities related to free zones (not limited to tobacco) are regularly documented by organizations that recognize the linkage. These activities include money laundering, tax evasion and trade in counterfeit goods or other illicit goods (60). A report from the European Parliament (61) referring to free ports in particular, mentioned that the motivation for using them included a “high degree of secrecy and deferral of import duty and indirect taxes”. The report even proposed the “urgent phasing out of free ports”. In the European Parliament report, free ports are free zones that function as (semi-) permanent storage areas for high-value goods. The Protocol includes a time-bound provision of effective controls on all manu- facturing and transactions of tobacco products in free zones (Article 12). Free zones are defined as a part of the territory of a Party where goods are considered to be outside the customs territory for import duties and taxes (Article 1.5). This is the same definition used in the International Convention on the Simplification and Harmonization of Customs Procedures (Revised Kyoto Convention) (62). Parties to the Protocol must implement effective controls in free zones within three years of entry into force of the Protocol. For countries not yet Parties to the Protocol, stringent controls of manufacturing and transactions involving tobacco products in free zones are an important component of an effective and efficient tax administration. One of the measures for dealing with free zones within the Protocol includes implementing “effective controls on all manufacturing of, and transactions in, tobacco and tobacco products, in free zones, by use of all relevant measures as provided in this Protocol”. As indicated in an Interpol report (17), a significant vulnerability of free zones is the fact that different economic operations (e.g. manufacturing, assembly, re-packaging and warehousing) take place outside the control of authorities. It is therefore essential for customs administrations to exercise their authority in free zones CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 145 to effectively identify and fight illicit trade in tobacco products. Relevant measures listed in the Protocol should be applied. These include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a track- ing and tracing regime. Removing exemptions on excise taxes is an additional way to increase control and remove incentives for using free zones as a means for tax evasion. Indonesia, for example, imposes excise taxes on cigarette manufacturing in its free trade zone.13 Parties to the Protocol shall also prohibit the intermingling of tobacco products with non-tobacco products in a single container or any other such similar transporta- tion unit when removed from free zones. Finally, each Party is to “adopt and apply control and verification measures to the international transit or transhipment of tobacco products and manufacturing equipment in conformity with the provisions of the Protocol”. Article 13 of the Protocol, which covers all duty-free sales of tobacco products, requires Parties to the WHO FCTC to consider prohibiting or restricting the sale to or import by international travellers of tax-free or duty-free tobacco products, as mentioned in Article 6 of the WHO FCTC. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect government revenues by creating a loophole in the tax structure (2). KEY TAKEAWAY 15 Customs administrations should exercise their authority in free zones to prevent different economic operations from taking place outside the control of authorities. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a tracking and tracing regime. 3.4.10 PROCEDURES AFTER DETECTING ILLICIT TRADE OF TOBACCO The procedures described in previous sections are intended to increase compliance and to prevent illicit trade. When smuggling or illicit trade is detected – through, for example, audits, tracking and tracing systems, verification of declarations or border control – actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. To deter further illegal behaviour, a comprehensive audit of everyone and everything involved in the illicit acts must also be carried out. Assets and vessels involved in the illicit activity can be seized, and financial accounts can be frozen. Some countries, including the United Kingdom, Canada and Chile, have also adopted a strategy known as “follow the 13 Indonesian Ministry of Finance, personal communication, January 2020. 146 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N money” to obtain more information on those who finance illicit trade. This strategy is intended to have a further-reaching effect by targeting those who finance the transport, production and storage of illicit products. The United Kingdom, Canada and Chile all created special teams tasked with identifying and targeting the offenders. Furthermore, Article 18 of the Protocol provides for the confiscation and destruction of tobacco, tobacco products and manufacturing equipment. One of the difficulties faced by competent authorities in exercising the authority to seize and forfeit products and/or equipment used in the manufacture or distribution of tobacco products is the cost of keeping or storing the goods and/or machinery before destruction. Thus, the law should also provide for a mechanism and timetable for the disposal and/or destruction of seized and forfeited goods or machinery, while prescribing a mechanism by which these properties can still be presented as admissible evidence in a judicial proceeding. Boxes 3.12 and 3.13 provide examples of successful efforts to combat illicit trade in the United Kingdom and Indonesia. KEY TAKEAWAY 16 As soon as smuggling or illicit trade in tobacco products is detected, actions such as collecting taxes and seizing and destroying smuggled and/or illicit tobacco must be taken. Box 3.12 The United Kingdom’s experience in fighting illicit trade in tobacco products In 2000, illicit cigarettes accounted for 22% of the cigarette market in the Unit- ed Kingdom. To deal with the problem, Her Majesty’s Customs and Excise14 implemented a major anti-smuggling effort. The strategy was refreshed with ad- ditional resources and measures in 2011 and reviewed in 2015. The result was a steady decline in the illicit cigarette market to 10% by 2013/2014. The measures taken were comprehensive and included hiring 1 000 new customs officers and investigators. In addition, tobacco supply chain legislation was introduced, aimed at discouraging tobacco manufacturers from facilitating smuggling. Tougher sanctions included increased fines of up to £5 million levied on a manu- facturer, criminal prosecution with sentences up to seven years, confiscation of assets as part of the proceeds of the crime, payment of duty on the confiscated goods plus penalties up to 100% of the duty, prohibition of the sale of tobacco products for 14 By the time of the renewed strategy, the respective bodies were the HMRC and the United Kingdom Border Agency. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 147 up to six months, unlimited fines for selling tobacco without the United Kingdom duty-paid fiscal mark after 13 March 2015, referral for withdrawal of the transporter’s licence, use of immigration sanctions to refuse entry to the United Kingdom for tobacco smugglers and civil action, including bankruptcy. The cost of these measures was £209 million over the first three years of the program and around £100 million annually by 2008/2009. This figure covers only HMRC and excludes any costing of the United Kingdom Border Agency. In 2013/2014, tobacco tax revenues were £9.5 billion. After a review of the strategy in 2015, the controls on raw tobacco were strength- ened by the introduction of an approval system in 2017. Anyone who manufactures, purchases, acquires, owns or is in the possession of a tobacco products manufacturing machine must be licensed with customs as of 1 August 2018 (63). The United Kingdom ratified the Protocol on 27 June 2018. It was the 40th country to ratify, which was the trigger point for the Protocol to enter into force. Fig. 3.4 Estimate of the illicit cigarette market and United Kingdom tax-paid consumption Sources: (16, 64). Bi lli on c ig ar et te s UK tax paid consumption Illicit market 20 00 -01 20 01 -02 20 02 -03 20 03 -04 20 04 -05 20 05 -06 20 06 -07 20 07 -08 20 08 -09 20 09 -10 20 10 -11 20 11 -12 20 12 -13 20 13 -14 20 14 -15 20 15 -16 20 16 -17 20 17 -18 0 10 20 30 40 50 60 148 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.13 A success story: Indonesia reduced illegal cigarettes from 12% to 3% of the market In some countries, most of the illegal cigarettes are imported, but in Indonesia most of them are produced within the country by unregistered manufacturers that are usually home-based and relatively small. In several territories, specifically on the Island of Java, illegal cigarettes have been produced for generations. This practice is supported by the availability of raw tobacco materials and cloves, as well as cheap labour costs, especially for female workers. Indonesia also produces cigarette products that are not available in most other countries. These products – known as handmade clove cigarettes (sigaret kretek tangan) – contain cloves, and the production process covers blending, rolling and packing. Parts of the process are done by hand, and 99% of the labourers are women. The strategy for combatting the illegal cigarette trade in Indonesia is divided into two main parts: preventive actions and responsive actions. Indonesia’s success in effectively tackling illicit trade is attributed to the following key factors. Monitoring and surveillance Preventive actions consist of administrative measures – such as issuance of permits and the excise stamp purchasing mechanism – that use risk management by optimizing the Excise Service Information System (ExSis). With this IT system, the Directorate General of Customs and Excise (DGCE) can oversee both daily transactions and daily production from factories. When information of suspicious activities is obtained, DGCE can suspend the purchase of excise stamps. The efforts to fight the illegal cigarette trade also invite stakeholders to be involved by supplying information regarding high-risk areas and regional governments. Strategic communications and community involvement DGCE continuously disseminates information and conducts public education to fight illegal cigarettes. These efforts are conducted every year, using a special campaign slogan. In 2019, the slogan was “Gempur Rokok Ilegal” (“Fight Illegal Cigarettes”). Key performance indicators for DGCE units and offices Parallel with the above-mentioned preventive actions, DGCE also continuously conducts responsive actions: enforcement, investigation and audit activities in cigarette factories. Enforcement activities are planned and measured by consider- ing the limited human resources and the large scale of the monitored territories. To demonstrate the effectiveness of administrative and enforcement measures in curbing the trade of illegal cigarettes, both types of activities are translated into key CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 149 performance indicators for all DGCE working units and offices, including regional offices and personnel. Use of technology and intelligence Coordination between different DGCE offices responsible for monitoring the produc- tion and marketing of cigarettes is maintained by sophisticated IT applications that enable efficient distribution of information and investigation activities. The applica- tions are the Customs Intelligence and Tactical Centre for data and analysis and the Centre for Command and Control (Pusat Komando dan Pengendalian/Puskodal) for ensuring that sea patrols work effectively and efficiently. Independent evaluation To evaluate the efforts and activities to reduce the circulation of illegal cigarettes in Indonesia (e.g. cigarettes without stamps, with fake stamps or with used stamps), in 2016, the government commissioned the University of Gadjah Mada in Yogyakarta to conduct a survey using a stratified random sampling method. To maintain objectiv- ity and independence, an independent body from this well-known university was appointed to conduct the survey. The survey results showed that the level of illegal cigarette circulation in Indonesia was 12.1% of total consumption. In 2018, the DGCE commissioned the University of Gadjah Mada to conduct another survey. Results showed that circulation of illegal cigarettes had been reduced to 7.0%. In 2019, using the same method the university used, DGCE conducted a survey that showed a reduction to 3.0%. Fig. 3.5 shows the results of the surveys. Fig. 3.5 Share of illicit trade in total cigarette consumption in Indonesia, 2016–2019 Source: Customs and excise department, Ministry of Finance, Indonesia, personal communication, 2020. 2016 2017 2018 2019 12.1% 10. 9% 7% 3% 150 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The results of the actions taken can be used as feedback for DGCE in developing excise policies for both service and enforcement, including policies for excise tariffs. DGCE will provide recommendations for an optimum policy format for controlling consumption, maintaining labour protections, optimizing revenue and, most impor- tantly, constantly reducing the consumption of illegal cigarettes on a national level. 3.4.11 PENALTIES Penalties and sanctions must be sufficient to deter illegal activities. Otherwise, finan- cial penalties may simply be paid as a cost of doing business while the illegal activity continues. The Protocol specifies commitments for Parties and provides information on best practices for non-Parties. Article 14.1 in Part IV of the Protocol requires each Party to establish unlawful activities, including manufacturing, wholesaling, broker- ing, selling, transporting, distributing, storing, shipping and importing or exporting tobacco products or manufacturing equipment without the payment of applicable duties or taxes or without using fiscal stamps or other required markings or labels. Articles 14.2 and 15 of the Protocol mandate Parties to determine which of the types of unlawful conduct set out in Article 14.1 shall be criminal offences. Parties must adopt legislative and other measures to give effect to such determinations, as well as to define whether the liability for committing illicit trade in tobacco is a criminal, civil or administrative offence. Article 17 further provides that the Parties shall consider adopting measures as needed to authorize competent authorities to levy penalties in an amount proportionate to lost taxes and duties resulting from the commission of illicit trade. Box 3.14 provides a case study of how Colombia used penalties to fight illicit trade. Box 3.14 The use of penalties to combat illicit trade in Colombia In 2017, the specific tax on cigarettes in Colombia was doubled, increasing from COL$ 700 per pack in 2016 to COL$ 1 400 in 2017. The tax rate was tripled from 2016 levels in 2018, reaching COL$ 2 100 per pack. A provision was added to increase taxes annually after 2018 at the rate of inflation plus 4%. In 2015, before the tax increase, Law 1762 introduced a number of measures to fight illicit trade more effectively. The length of imprisonment for dealing in contraband cigarettes was increased from 3–5 years to 4–12 years. Moreover, government officials who facilitate illicit trade – or anyone involved in transporting or retail sales of illicit tobacco – face similar prison terms. The law allows vehicles used for smuggling to be confiscated, and penalties were increased for illicit trade that is conducted through areas such as special economic zones. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 151 Under the law, illicit trade is considered to be a source of money laundering, which means that the Financial Intelligence Unit can use the same methods it uses to investigate other illegal financial activities. This practice is not common. The law specifically created new sanctions related to alcohol and tobacco excise tax evasion, including the seizure of goods, fines, closure of retail outlets and the suspension or cancellation of licences, authorizations or registries. Arrests and seizures have increased under the new law. Indeed, since its enactment, law authorities reported that between 2016 and 2018, five criminal organizations were dismantled, 53 individuals were apprehended and 72 assets were confiscated. In addition, 2 236 individuals were apprehended and 503 vehicles transporting smuggled goods were confiscated, as transport of such goods is now also considered a crime under the law. More importantly, thanks to the large tax increases, consumption decreased while revenues increased substantially in 2017 and 2018 (see Fig. 3.6). It is estimated that illicit trade in cigarettes in five Colombian cities in 2016 constituted 3.5% of total consumption, a much lower estimate than the industry data suggest. In 2017, after nine months of the tax increase implementation, a similar study found that illicit cigarettes remained low, at 6.4% of total consumption. Fig. 3.6 Packs sold and tobacco tax revenue before and after the tax increase in Colombia, 2016–2018 Sources: (65–67 and Ministry of Finance, Colombia (Direccion de Apoyo Fiscal), personal communication, 2020). 2016 2017 2018 Cigarette sales, million packs Cigarette excise revenue, million US$ current 0 100 200 300 400 500 600 700 800 673.7 521 446.4 195.1 301.2 386.2 33.7% reduction 97.9% increase 152 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N For consumers in possession of illicit tobacco, the minimum penalty should be confiscation and destruction of the illicit tobacco products found in their possession, with payment required for the unpaid tax and duties on those products. In the state of California in the United States, it is illegal to possess a tobacco product on which taxes are due and not yet paid. The burden of proving that taxes have been paid is on those who have the products in their possession. The provi- sion is enforced by the California Department of Tax and Fee Administration and local law enforcement agencies. A violation is a misdemeanour, with a maximum fine of US$ 5 000 and/or up to one year in prison. Illegal packages are subject to seizure and forfeiture. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are using this type of legislation to address illicit trade of tobacco. Asset confiscation and increased penalties for involvement in illicit trade are becoming more common as well. Withholding or even confiscation of trucks involved in smuggling is also common in several countries. KEY TAKEAWAY 17 Penalties and sanctions imposed should be sufficient to deter illegal tobacco trade activities. Penalties should be levied in amounts proportionate to lost taxes and duties resulting from illicit trade. 3.5 TAX ADMINISTRATION OF OTHER TOBACCO PRODUCTS In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. However, there is a lack of standard- ization of other products and sometimes large informal markets. For example, it is estimated that two thirds of waterpipe tobacco in the EU is non-duty-paid (19). Other tobacco products – such as bidis in South-East Asia, waterpipe tobacco in the Eastern Mediterranean region and snus in Sweden – are considered part of a country’s traditions. This sometimes leads to situations where governments are hesitant to strongly regulate and tax these products. Some products, such as kreteks (clove cigarettes) in Indonesia and bidis in India and Bangladesh, are mainly sold in one market. Other tobacco products are more likely to be produced by hand on a small scale, making it difficult to detect and collect taxes on them. The same applies to RYO tobacco, which can be produced on a small scale by hand or with the use of small machinery. The trade in raw tobacco and small-scale home production of tobacco often take place outside of monitoring and control systems (19). As mentioned in section 3.3.1, countries have found various solutions to address this problem, including prior approval for purchase or sale of raw materials and reg- istering, authorizing or licensing of all operators and growers that handle raw tobacco. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 153 3.5.1 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In principle, adding a product to an existing tax framework is not likely to impose significant costs. It is reasonable to expect that challenges similar to those faced in dealing with conventional tobacco products will be faced in the collection of taxes on new products, as market players will attempt to use loopholes in tax regulation to avoid or evade taxes whenever possible. However, new challenges are expected to arise when those new products involve rapidly changing technology and where their market dynamics are widely unknown. Furthermore, taxation of new tobacco products may require additional capacity, as a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. 3.5.2 HEATED TOBACCO PRODUCTS (HTPs) Many countries apply a specific excise on tobacco products according to tobacco weight (see Table 2.4 in Chapter 2). With HTPs, this is likely to impose a challenge, since assessing the content of tobacco in a heated tobacco stick will be an additional burden. From a tax administration perspective, it will be easier for authorities to apply taxes per stick or per unit, as is done for cigarettes. 3.5.3 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS) PRODUCTS Some countries tax only nicotine-containing e-liquids while others tax both nicotine- and non-nicotine-containing e-liquids. Taxing only nicotine-containing e-liquids re- quires laboratory capacity to detect the presence of nicotine (see Table 2.5 of Chapter 2). Self-declarations by industry are not sufficient, since some e-liquids labelled as nicotine-free have been found to contain nicotine (see section 2.4.2). Therefore, it is simpler to tax both nicotine- and non-nicotine-containing e-liquids. One likely challenge of taxing all e-liquids will be the capacity to detect and differentiate whether the e-liquids used in ENDS/ENNDS are falsely declared as being for other purposes at the import and manufacturing levels. More information on advantages and disadvantages of different excise tax policies is given in Table 2.6 of Chapter 2. The challenge in taxing the other components of ENDS/ENNDS products is their diversity (see section 2.4.2 of Chapter 2) and the possibility that some parts may be used for other purposes (e.g. in batteries). As indicated earlier, rapidly changing technology and the lack of control and knowledge of the market make taxation of ENDS/ENNDS devices challenging. It may be for this reason that the majority of countries that tax those products address only the e-liquids. When applying a tax on these newer products, countries should be aware that many customers buy their products online. It is therefore recommended that countries 154 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N deciding to tax these products draw up a proper implementation plan, including how taxes will be collected on imported products and online sales. Online cross- border sales of tobacco products are not permitted in all countries. Several EU countries have banned such sales, which include online sales from retail outlets to consumers in another country. Of course, such bans makes sense only if there is also capacity to enforce them. As in the case of taxing tobacco products, the following actions will be important to more effectively impose taxes on these products: 1. implementing strong enforcement mechanisms such as licensing, recordkeep- ing and control of the supply chain, which can include but is not limited to: a) imposing strict licensing of retailers, importers and manufacturers; ideally, licensing of all those involved in the supply chain and developing a tracking and tracing regime for ENDS/ENNDS products and HTPs (to share cost, this can be done in tandem with the system developed for cigarettes); b) exercising the right to set the frequency and type of audits or controls; c) exercising the right to confiscate goods; and d) imposing sanctions such as penalties, fines and/or withdrawal of licences (if applicable) if legislation is not respected. Specific to ENDS/ENNDS products: 2. implementing highly consequential sanctions for producers who declare nicotine-containing e-liquids as “non-nicotine-containing”; and 3. requiring a fee (contribution to the costs) for laboratory tests when a new product is brought on the market or when there are significant modifications to an existing one. More information on the policy options to apply excise taxes on ENDS/ENNDS is given in Chapter 2, section 2.4.2, Table 2.6. KEY TAKEAWAY 18 In principle, the administration of taxes on new and emerging nicotine products and tobacco products should be similar to that for cigarettes. Due to the lack of standardization of these products, however, a rapid and constantly evolving understanding of them and their supply chain will be required to achieve effective and efficient administration of taxes. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 155 3.6 THE BROADER ELEMENTS OF A GOOD TAX SYSTEM 3.6.1 PROPER RESOURCING OF COMPETENT AUTHORITIES In addition to legal tools and a legal basis on which to act and enforce, the authori- ties in charge of implementing excise tax laws should be provided with sufficient resources to hire the necessary staff to properly implement and enforce them. The necessary staffing could encompass multiple agencies and will often require coop- eration between agencies, since some aspects – such as regulation, licensing and border control – may be performed by agencies other than the competent authority. Staff of competent authorities need the necessary tools, equipment, training and supplies to carry out their functions. This requirement includes the means to build or purchase and maintain a software system that will allow taxpayers to submit required information electronically. Electronic filing has benefits for both taxpayers and authorities. It minimizes the compliance cost for taxpayers and can therefore support voluntary compliance (68). To identify risks of noncompliance, the software system should offer competent authorities the ability to analyse the data submitted by taxpayers and cross-check it with data from other taxes – such as VAT – and third-party sources, such as banks and household surveys. Another option for authorities is to make more efficient use of existing resources. For example, an authority could optimize the risk management system by switching to a risk-based approach: resources could be saved by auditing taxpayers who are more likely to be noncompliant based on risk analysis rather than auditing all of them. Other problems that challenge the effective functioning of a competent authority are lack of a coherent strategy and problems with professionalism related to lack of training or corruption (69). Having a strategy avoids directing resources towards less- important areas. The strategy should always be aligned with the objectives, so that com- petent authorities can identify which steps they should take and in which order they should take them to reach these objectives. A strategy is indispensable to prioritizing and organizing resources so that identified issues or risks can be addressed efficiently. 3.6.2 CORRUPTION Competent authorities should implement tax laws with integrity and have strict rules and regulations for detecting corruption. Strict rules and regulations should also be in place for the punishment of both agency personnel and taxpayers who engage in corrupt practices. Corruption within a competent authority results in the improper monitoring of tax compliance and is one of the causes of the proliferation of illicit trade in tobacco products. It also erodes confidence in competent authorities and ultimately in governments overall. In addition to effective laws and regulation, strong internal audits covering prevention, investigation and sanctions should be 156 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implemented. To improve prevention, a risk map should be created that highlights areas of misconduct and possible leakages. An action plan to update controls should be established to improve areas of weakness detected in procedures and systems. The audits should also be scheduled regularly. Prepared internal auditors with pow- ers to conduct investigations are necessary. Sanctions for corruption, including administrative sanctions and criminal prosecution, must be strong. 3.6.3 A STRONG JUDICIARY The judicial system should be honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. The appeals process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially in the context of illicit trade. KEY TAKEAWAY 19 Broader elements of a good tax system include (1) proper resourcing of competent authorities to hire staff and obtain necessary equipment and systems, (2) strict rules and regulations to detect and punish corruption among both agency personnel and taxpayers and (3) ensuring that the judicial system is honest and independent, with disputes being solved as quickly as possible. 3.7 CONCLUSIONS Policies are more effective if they are properly implemented and enforced. Com- petent authorities have a key role in the achievement of financial and public health objectives of excise taxes. Given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade, this chapter draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (i.e. the Protocol). The Protocol provides a blueprint of measures to address the problem of illicit trade and can be used as a model even by countries that are not Parties to it. Qualities of an effective and efficient tax administration include institutional ar- rangements where roles and responsibilities of competent authorities are clearly defined to avoid overlap and voids. Additionally, effective collaboration among relevant bodies must be facilitated. At the national level, within any organizational arrangement, it is vital that agencies cooperate and exchange information and that their competences find their basis in law. A legal basis for exchange or access to information between government bodies should be ensured. At the international level, especially for border control, the role of customs is key, and access to international CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 157 cooperation agreements such as the Protocol is very useful. An organizational tax administration structure must include a system of performance evaluation and accountability through pre-defined key indicators. To ensure compliance, the accuracy of information for the tax compliance cycle is key, including clear and straightforward taxpayer registration and licensing, declara- tion, recordkeeping, warehousing, distribution, collection and tax refund processes. • Licensing is a powerful tool for obtaining information and securing the supply chain of tobacco products. Ideally, all persons involved in the growing of tobacco and the retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. • Collecting as much information as possible on the business of tobacco and recording all transactions are key to reducing tax evasion, but this may be burdensome for authorities. The use of IT for periodic tax declarations, ac- counting, inventory and financial information is critical for obtaining accurate information and can help decrease the cost of the whole reporting system. • Recordkeeping should be ensured. All persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of those products. • Maintaining a system of authorization for warehousing allows the authori- ties to carry out controls in production and storage facilities to ensure that taxes are paid. Ideally, bonded warehouses should be eliminated from the supply chain. • Duty suspension – which is often applied during the producing, processing, holding, receiving and dispatching of excise goods – should be granted only if strict criteria are met (e.g. for granting authorization, warehouse pre-authoriza- tion visits, adequate stock control measures, checking the origin of excise prod- ucts and the entire production process and coding and marking of products). • To limit the number of taxpayers a competent authority has to manage, tax collection should take place close to the point of production and import. • Refunds for VAT, excise taxes and customs duties are common in most coun- tries, under the principle that taxes are not exported. The refund process must be closely monitored to avoid opportunities for tax evasion. Control and enforcement – key components of tax administration – include a number of measures to secure the supply chain: licensing and due diligence, fiscal markings, 158 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N track and trace, anti-forestalling measures, audits and controls, import and export control and attention to free zones and transhipment points. Control and enforce- ment need to be included as pillars in the strategic plan of the tax administration. Enforcement and control plans must be designed to define the activities and taxpayers that are subject to enforcement and to allocate staffing, auditing, infrastructure and IT resources. Targets must be defined, including the number of interventions and any additional collection or reduction of tax evasion. This includes choosing interventions for those who have a higher probability of noncompliance (the risk- based approach). In the tobacco supply chain, import, export and transfers to and from warehouses may be areas at greater risk of noncompliance. • Licensing provides timely and accurate data that can serve as the basis for audits, since it identifies and controls legitimate operators. The process of licensing control must be carried out and updated periodically – in particular, by controlling the validity of bonds or guarantees and the proper functioning of the required systems and recordkeeping. Where licences are required, the law should include a provision that disallows purchases from unlicensed suppliers or sales to unlicensed purchasers. This means that both suppliers and purchasers will need to verify those with whom they are doing business. This will substantially help to reduce the burden of proof for authorities. In addition, to maintain a high level of control, the validity of licences should be limited in time, making renewals or reapplication required. • Another important measure for controlling and monitoring production and import of tobacco products is the use of fiscal markings (e.g. tax stamps). In addition to increasing compliance with tax laws, fiscal markings can help distinguish between genuine and illicit tobacco products. The use of fis- cal marks enables both the competent authority and the public to monitor whether the taxes on tobacco products have been properly paid. In addition to locally produced and imported products, tobacco products for export should also be required to be marked, but with an indication that they are for export. Requiring a standard package size can facilitate the application of fiscal markings. To lower the chance that fraudsters attempt to re-use fis- cal markings (in particular, stamps) the marking should be applied to each pack of cigarettes (and other tobacco products) before the pack is wrapped with cellophane. Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Fiscal markings should include several security features to make them more difficult to counterfeit. These can include overt, covert, semi-covert and/or forensic features. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 159 • Tracking and tracing systems assist authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as in monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to provide authorities with information on all transactions throughout the entire tobacco product supply chain until duties are paid or other obligations are discharged. Any tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it becomes possible to unambiguously register that product’s movements. A good tracking and tracing system enables the government to properly monitor the supply chain, improve its ability to ensure collection of the proper duties and taxes, authenticate whether the identification marking is genuine and matches the product, improve its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. To reduce the financial burden of implementing such a system, jurisdictions could require the tobacco industry to bear the cost. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, and governments should ensure that the system is independent from the tobacco industry. While the objectives of fiscal markings and tracking and tracing systems are different, stamps increasingly contain tracking and tracing features. • Implementing legal measures to prevent forestalling can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. Forestalling, stockpiling or front-loading occur when manufacturers or im- porters increase their tax-paid stock or oversupply the market by increasing production or imports before a tax increase in order to pay the previous lower rate. • Periodic audits and controls can be implemented to increase compliance. These include cost audits, transfer price audits, price and market monitoring, consumer controls and cross-check controls. • Import and export of tobacco products and manufacturing equipment should be allowed only for duly licensed natural persons or legal entities. The risk of loss of revenue can be mitigated by requiring a guarantee or bond that will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless required fiscal markings (such as tax stamps or export labels) are affixed on the pack, according to the law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Exchange of information between jurisdictions on the movement of goods can also reduce the risk of evasion. 160 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Non-invasive detection equipment (such as X-ray scanners) can be used at customs posts to detect contraband merchandise. A cheaper alternative is the use of dogs that are trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied to reduce contraband. Such measures include the separation of processing operations from the sealed storage of taxed and untaxed products. Within a country, mobile excise control units are helpful in verifying excis- able goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. Physical control operations require close coordination between police, border guards and other public services. • Controls such as regulation and oversight are usually less strict in free zones and transhipment points. This can make free zones appealing to persons involved in illegal cigarette manufacturing or trade. Customs administrations should exercise their authority in free zones to effectively identify and fight illicit trade in tobacco products. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing tracking and tracing regimes and removing exemptions from excise taxes. Other actions include the prohibition of intermingling of tobacco products with non-tobacco products in a single container or other similar transportation unit when the products are removed from free zones. Sale of tax-free or duty-free tobacco products to international travellers should be prohibited, as these sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect govern- ment revenues by creating a loophole in the tax structure. Procedures after detection of illicit trade of tobacco products should be clearly defined. If smuggling or illicit trade is detected through audits, tracking and tracing systems, verification of declarations or border control, actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. It is also important that penalties and sanctions be sufficient to deter illegal activities. Low financial penalties may simply be paid as a cost of doing business while the illegal activity continues. The minimum penalty for consumers in possession of illicit tobacco products should be confiscation and destruction of the products found in their possession and required payment for the unpaid tax and duties on those products. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are taking advantage of this type of legislation and using it to address illicit trade of tobacco as well. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 161 In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. The challenge for taxation of other products includes the lack of standardization of those products and sometimes large informal markets. Knowledge of the product and the supply chain greatly help to facilitate effective tax administration. The trade in raw tobacco and small-scale home production of RYO and other products such as bidis often takes place outside of monitoring and control systems. The best way to address this challenge is to enforce prior approval for purchase or sale of raw materials and a requirement to register, obtain an authorization or license all operators and growers handling raw tobacco. In principle, adding new and emerging nicotine and tobacco products to an exist- ing tax framework is not expected to impose significant costs. It is reasonable to expect that similar challenges will be faced in the collection of taxes on these newer products, as market players will attempt to use the current loopholes in tax regulation to avoid or evade taxes on these products whenever possible. However, challenges are expected to arise, as newer products involve rapidly changing technology, and their market dynamics are widely unknown. Furthermore, additional capacity may be required, since a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. Because the newer nicotine and tobacco products are widely purchased online, countries deciding to tax these products should draw up a proper implementation plan that includes rules on how taxes will be collected on imported products and online sales. Online cross-border sales are not permitted in some countries. The elements of a good tax system include (1) proper resourcing of competent authorities sufficient for hiring the necessary staff to properly implement and en- force excise tax laws; (2) implementation of tax laws with integrity and with strict rules and regulations to detect corruption and for the punishment of both agency personnel and taxpayers who are engaged in corrupt practices; and (3) ensuring that the judicial system is honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. 162 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N REFERENCES 1. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 2. WHO FCTC guidelines for implementation of Article 6 Price and tax measures to reduce the demand for tobacco. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/treaty_instruments/ Guidelines_Article_6_English.pdf?ua=1, accessed 5 October 2020). 3. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 4. Joossens L, Raw M. From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–234 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 5 October 2020). 5. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 6. Reyes IP. Measures to control the tobacco supply chain in the ASEAN: a review of excise tax stamps, track and trace mechanisms, and licensing systems in the ASEAN region. Bangkok: Southeast Asia Tobacco Control Alliance; 2014 (https://seatca.org/dmdocuments/SEATCA%20MEASURE%20BOOK%20. pdf, accessed 5 October 2020). 7. Italy’s tax administration: a review of institutional and governance aspects. Paris: Organisation for Economic Co-operation and Development; 2016 (https://www.oecd.org/tax/administration/italy- tax-administration-a-review-of-institutional-and-governance-aspects.pdf, accessed 5 October 2020). 8. Crandall W, Gavin E, Masters A. ISORA 2016: understanding revenue administration. Washington (DC): International Monetary Fund; 2019 (Departmental Paper Series No. 19/05; (https://www.imf.org/en/ Publications/Departmental-Papers-Policy-Papers/Issues/2019/03/07/ISORA-2016-Understanding- Revenue-Administration-46337, accessed 5 October 2020). 9. Dutch customs in 2018. Heerlen: Customs Administration of the Netherlands; 2019 (https://over-ons. belastingdienst.nl/wp-content/uploads/2019/04/Dutch-Customs-in-2018.pdf, accessed 8 October 2020). 10. ESAs transform the way competent authorities cooperate with each other on AML/CFT matters [press statement]. European Banking Authority; 16 December 2019 (https://eba.europa.eu/esas-transform- way-competent-authorities-cooperate-each-other-amlcft-matters, accessed 5 October 2020). 11. Illicit trade report 2013. Brussels: World Customs Organization; 2014 (http://www.wcoomd.org/-/ media/wco/public/global/pdf/topics/enforcement-and-compliance/activities-and-programmes/ illicit-trade-report/itr_2013_en.pdf?la=en, accessed 5 October 2020). 12. Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation and repealing Directive 77/799/EEC. Brussels: OJEU; 2011. (L 64/1; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:32011L0016&from=en, accessed 29 January 2021). 13. 439 stolen cars seized in 12 days as part of Frontex-led operation. Frontex News Release; 14 October 2019 (https://frontex.europa.eu/media-centre/news-release/439-stolen-cars-seized-in-12-days-as- part-of-frontex-led-operation-vvvirI, accessed 5 October 2020). 14. Focus on: the illicit trafficking of counterfeit goods and transnational organized crime. Vienna: United Nations Office on Drugs and Crime; 2014 (https://www.unodc.org/documents/counterfeit/FocusSheet/ Counterfeit_focussheet_EN_HIRES.pdf, accessed 5 October 2020). 15. Crandall W. Revenue administration: performance measurement in tax administration. Washington (DC): International Monetary Fund; 2010 (Technical Notes and Manuals 10/11; https://www.imf.org/ external/pubs/ft/tnm/2010/tnm1011.pdf, accessed 7 October 2020). 16. Measuring tax gaps 2019 edition – tax gap estimates for 2017-18. London: HM Revenue and Customs; 2019 (https://webarchive.nationalarchives.gov.uk/20200701215139/https://www.gov.uk/government/ statistics/measuring-tax-gaps, accessed 29 January 2021). 17. Countering illicit trade in tobacco products: a guide for policy-makers. Lyon: INTERPOL; 2014 (Legal handbook series; https://www.aph.gov.au/DocumentStore.ashx?id=704d9168-2aa7-4df4-bb81- CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 163 0fe1f357dfa6&subId=408832, accessed 29 January 2021). 18. Resolução- RDC No. 226 de 30 de Abril de 2018 [Resolution of April 30, 2018]. Ministério da Saúde/ Agência Nacional de Vigilância Sanitária/Diretoria Colegiada; 2018 (in Portuguese) (https://www.in.gov. br/materia/-/asset_publisher/Kujrw0TZC2Mb/content/id/12527411/do1-2018-05-02-resolucao-rdc- n-226-de-30-de-abril-de-2018-12527407, accessed 17 February 2021). 19. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 20. Preece R. Addressing the illicit flow of tobacco products in Australia. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 21. Excise on tobacco [website]. Canberra: Australian taxation office; 2019 (https://www.ato.gov.au/ business/excise-on-tobacco/, accessed 5 February 2021). 22. Safe Framework of Standards. Brussels: World Customs Organization; 2018 (http://www.wcoomd. org/-/media/wco/public/global/pdf/topics/facilitation/instruments-and-tools/tools/safe-package/ safe-framework-of-standards.pdf?la=en, accessed 29 January 2021). 23. The Trade Facilitation Agreement. Geneva: World Trade Organization; 2017 (https://www.wto.org/ english/docs_e/legal_e/tfa-nov14_e.htm, accessed 29 January 2021). 24. Authorised economic operator (AEO) [website]. Kildare: Irish Tax and Customs; 2020 (https://www. revenue.ie/en/customs-traders-and-agents/authorised-economic-operators/index.aspx, accessed 7 October 2020). 25. Performance outcome area 1: integrity of the registered taxpayer base. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www. tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 26. Commission implementing regulation (EU) 2016/323 of 24 February 2016 laying down detailed rules on cooperation and exchange of information between member states regarding goods under excise duty suspension pursuant to Council Regulation (EU) No 389/2012. Brussels: OJEU; 2018 (http://data. europa.eu/eli/reg_impl/2016/323/2018-02-15, accessed 7 October 2020). 27. Council Directive 2008/118/EC concerning the general arrangement for excise duty repealing Directive 92/12/EEC. Brussels: OJEU; 2009 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:3 2008L0118&from=en, accessed 29 January 2021). 28. Our strategy. London: HM Revenue & Customs, 2017 (https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/629941/HMRC-Strategy.pdf, accessed 31 January 2021). 29. Strategic plan FY2018-2019. Washington (DC): Internal Revenue Service, 2017 (https://www.irs.gov/ pub/irs-pdf/p3744.pdf, accessed 31 January 2021). 30. Performance outcome area 2: effective risk management. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/ assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 31. Tax administration 2017: comparative information on OECD and other advanced and emerging economies. Paris: Organisation for Economic Co-Operation and Development; 2017. 32. Compliance risk management: managing and improving tax compliance. Paris: Organisation for Economic Co-Operation and Development; 2017 (https://www.oecd.org/tax/administration/33818656. pdf, accessed 7 October 2020). 33. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 34. Evans P. U.K. Fines British American Tobacco for Oversupplying Cigarettes in Belgium. The Wall Street Journal. 13 November 2014 (https://www.wsj.com/articles/u-k-fines-british-american-tobacco-for- oversupplying-cigarettes-in-belgium-1415873470, accessed 31 January 2021). 35. Doward J. BAT fined for oversupplying tobacco in low-tax European jurisdictions. The Guardian. 164 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 16 November 2014 (https://www.theguardian.com/business/2014/nov/16/bat-fined-for-oversupplying- tobacco-in-low-tax-european-jurisdictions, accessed 31 January 2021). 36. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021) 37. Cour des comptes. Accises sur les produits du tabac, Bruxelles [Court of Auditors, Excise duties on tobacco products, Brussels]. Bruxelles: Cour des comptes; 2015 (in French) (https://www.ccrek.be/ docs/2015_27_AccisesProduitsDuTabac.pdf, accessed 7 October 2020). 38. Chaloupka FJ, Edwards SM, Ross H, Diaz M, Kurti M, Xu X, et al. Preventing and reducing illicit tobacco trade in the United States. Atlanta: Centers for Disease Control and Prevention; 2015 (https://www. cdc.gov/tobacco/stateandcommunity/pdfs/illicit-trade-report-121815-508tagged.pdf, accessed 29 January 2021). 39. The Tax Stamp Forum 2015. Hexham: Reconnaissance International (https://10times.com/tax-stamp- forum-miami, accessed 10 September 2020). 40. Tax stamps: a technical study and market report. London: Reconnaissance international; 2012. 41. ЕАНС/2013/Health/ll concerning the provision of an analysis and feasibility assessment regarding EU systems for tracking and tracing of tobacco products and for security features. Brussels: European Commission, 2015 (https://ec.europa.eu/health/sites/health/files/tobacco/docs/2015_tpd_tracking_ tracing_frep_en.pdf, accessed 29 January 2021). 42. Ross H. Controlling illicit tobacco trade: international experience. Economic research informing tobacco control policy. Cape Town: University of Cape Town; 2015 (https://tobacconomics.org/ uploads/misc/2015/05/Ross_International_experience_05.28.15.pdf, accessed 18 December 2020). 43. Security and resilience — authenticity, integrity and trust for products and documents — guidelines for the content, security, issuance and examination of excise tax stamps. Geneva: International Organization for Standardization; 2018. 44. Borkowski F, Twomey C. European Union: an update on EU policies. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 45. Quality management systems — Fundamentals and vocabulary. ISO 9001:2015(en). Geneva: International Organization for Standardization; 2015 (https://www.iso.org/obp/ui/#iso:std:iso:9001:ed- 5:v1:en, accessed 16 February 2021). 46. Protocol to Eliminate Illicit Trade in Tobacco Products: questions & answers. Geneva: World Health Organization; 2019 (https://www.who.int/fctc/protocol/faq/en/, accessed 7 October 2020). 47. Ross H, Eads M, Yates M. Why governments cannot afford Codentify to support their tracking and tracing solutions. Tob Control. 2018;27(6):706–08. 48. Ser vicio de Impuestos Internos, Chi le, Noticias, 28 May 2019. (http://www.si i .c l/ noticias/2019/280519noti01er.htm, accessed 20 October 2020). 49. Directive 2014/40/EU of the European Parliament and of the Council. OJEU. 2014;127:1–38 (https:// ec.europa.eu/health/sites/health/files/tobacco/docs/dir_201440_en.pdf, accessed 7 October 2020). 50. Report on the EU customs enforcement of intellectual property rights: results at the EU border. Luxembourg: Publications Office of the European Union, 2019 (https://ec.europa.eu/taxation_customs/ sites/taxation/files/2019-ipr-report.pdf, accessed 7 October 2020). 51. Matthews P. Forestalling ahead of property tax changes. London: Office for Budget Responsibility; 2016 (OBR Working Paper No. 10; https://obr.uk/docs/dlm_uploads/Working-paper-No.10-1.pdf, accessed 29 January 2021). 52. Ross H, Tesche J, Vellios N. Undermining government tax policies: common strategies employed by the tobacco industry in response to tobacco tax increases. Prev Med. 2017;105S:S19–S22 (https://www. sciencedirect.com/science/article/pii/S0091743517302165?via%3Dihub, accessed 17 February 2021). 53. European Commission v Portugal supported by Belgium, Estonia and Poland, Case C-126/15. Judgement of the Court (Fifth Chamber); 29 June 2017 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?u ri=CELEX:62015CJ0126&from=EN, accessed 31 January 2021). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 165 54. Amalia Valesko v Zollamt Klagenfurt, Case C-140/05. Judgement of the Court (Second Chamber); 5 October 2006 (http://curia.europa.eu/juris/showPdf.jsf;jsessionid=02DEE17CC11845BAE0F42FC7 CFD24C85?text=&docid=65599&pageIndex=0&doclang=en&mode=lst&dir=&occ=first&part=1&c id=2591459, accessed 29 January 2021). 55. European Commission v French Republic, Case C-197/08. Judgement of the Court (Third Chamber); 4 March 2010 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:62008CJ0197&qid =1612223322830&from=EN, accessed 31 January 2021). 56. Commission requests that Belgium amend its legislation on tobacco: letter of formal notice. Brussels: European Commission; 8 November 2018. (Memo 18/6247; https://ec.europa.eu/commission/ presscorner/detail/EN/MEMO_18_6247, accessed 29 January 2021). 57. Commission requests that Belgium amend its legislation on tobacco, MEMO/18/6247 of 8 November 2018. 58. Pederson H, Floristean A, Iseppi L, Dawkins R, Smith C, Mørup C, et al. Study on the measuring and reducing of administrative cost for economic operators and tax authorities and obtaining in parallel a higher level of compliance and security in imposing excise duties on tobacco products. Brussels: European Commission; 2014 (https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/ ramboll-tobacco-study.pdf, accessed 3 October 2020). 59. Money laundering vulnerabilities of free trade zones. Paris: Financial Action Task Force; 2010. https:// www.fatf-gafi.org/media/fatf/documents/reports/ML%20vulnerabilities%20of%20Free%20Trade%20 Zones.pdf, accessed 7 October 2020). 60. Omi K. ‘Extraterritoriality’ of free zones: the necessity for enhanced customs involvement. World Customs Organization; 2019 (WCO research paper no. 47; http://www.wcoomd.org/-/media/wco/public/global/ pdf/topics/facilitation/ressources/permanent-technical-committee/225-226/itemixb_fzs_wco_e. pdf?la=en, accessed 29 January 2021). 61. European Parliament resolution of 26 March 2019 on financial crimes, tax evasion and tax avoidance. European Parliament; 2019 (2018/2121(INI)); http://www.europarl.europa.eu/doceo/document/TA- 8-2019-0240_EN.html, accessed 7 October 2020). 62. International Convention on the Simplification and Harmonization of Customs Procedures. Brussels: World Customs Organization; 2008 (http://www.wcoomd.org/Topics/Facilitation/Instrument%20 and%20Tools/Conventions/pf_revised_kyoto_conv/Kyoto_New, accessed 7 October 2020). 63. Evidence Notice 2004: tobacco duty – tobacco products manufacturing machine licensing scheme. London: HM Revenue & Customs, 2018 (https://www.gov.uk/government/publications/excise-notice- 2004-tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme/excise-notice-2004- tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme, accessed 31 January 2021). 64. Tackling illicit tobacco: from leaf to light: the HM Revenue & Customs and Border Force strategy to tackle tobacco smuggling. London: HM Revenue and Customs; 2015 (https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/418732/Tackling_illicit_ tobacco_-_From_leaf_to_light__2015_.pdf, accessed 7 October 2020). 65. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 66. Cardenas M. Case studies in illicit tobacco trade: Colombia. Chicago: University of Illinois at Chicago; 2020 (Tobacconomics Fact Sheet; https://tobacconomics.org/files/research/606/UIC_Colombia-Illicit- Trade-Fact-Sheet_v1.4.pdf, accessed 29 January 2021). 67. Maldonado N, Llorente BA, Escobar D, Iglesias RM. Smoke signals: monitoring illicit cigarettes and smoking behaviour in Colombia to support tobacco taxes. Tob Control. 2019;29:s243-s248 ( https:// tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/s243.full.pdf, accessed 29 January 2021). 68. Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 69. Pellechio A, Tanzi V. The reform of tax administration. Washington (DC): International Monetary Fund; 1995 (Working Paper no. 95/22; https://www.elibrary.imf.org/view/IMF001/07242- 9781451843941/07242-9781451843941/07242-9781451843941_A001.xml?language=en&redirect=true, accessed 29 January 2021). 166 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.1 COMPOSITION OF TOBACCO PRODUCTS To implement and enforce tobacco taxes in the most efficient way, competent authori- ties should be familiar with all of the components of tobacco products, including each of the raw materials used in their manufacture, production inputs and tobacco manufacturing machinery. Knowledge of the components of excisable products and machinery provides valuable information to identify activities at high risk for non- compliance, implement measures to ensure all taxes are paid and prevent illicit trade. Nicotine, non-nicotine and tobacco products and their component parts In most countries, the ministry of finance determines tax policy, including which tobacco products are taxed, while the ministry of health is responsible for product and use regulation. This could lead to different definitions of the same product, depending on which ministry is responsible for a given law or regulation. Wherever possible, a clear and common definition should be developed to simplify procedures and avoid confusion. Tobacco products take various forms, and not all may be regulated or subject to excise tax in a specific jurisdiction. In addition to cigarettes, other traditional tobacco products include smokeless tobacco – such as chewing tobacco, snuff and snus – as well as bidis and kreteks (clove cigarettes), which can be hand-rolled or manufactured, pipes, hookah or waterpipe and cigars. Cigarettes It is important to understand the materials and component parts of the tobacco products most commonly used in a particular country. Cigarettes are the most common and significant tobacco products in terms of volume and tax revenues in most jurisdictions. A cigarette stick is composed of: • the tobacco blend of various types of tobacco plant (leaves and stem and other plant parts) and additives (including flavours); • the cigarette paper used to wrap the tobacco blend to make up the tobacco rod; • the acetate filter that forms the white portion at the tip of a filtered cigarette, which is in direct contact with the smoker’s mouth; • the tipping paper or wraps around the filter; and • the adhesive that secures the cigarette paper around the tobacco blend and the tipping (1). Each manufacturer follows a specific process to produce cigarettes. Aside from the tobacco blend, manufacturers also vary the size of cigarette paper and tipping paper and the length of acetate filter used per stick (1). In some countries, these elements are standardized. In an ideal regulatory framework, a manufacturer would CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 167 be required to submit information on the specific process for each brand and variant of the tobacco product that it manufactures to the competent authorities as part of the licensing requirement (see section 3.4.3). Authorities could, for example, require manufacturers to submit this information in order to obtain a licence. The minimum requirements of the administration and the information that should be included could be laid down in law or lower regulation to ensure that authorities have the information in their possession for all licensed manufacturers. This information contributes to verifying whether a company is reporting the actual quantity of cigarettes manufactured for sale and sold cigarettes by comparing the amount of materials used for production and the quantities used per cigarette with the total number of manufactured cigarettes. Eventually, this information also contributes to validating whether the taxes are properly paid. Figure A3.1 shows the component parts of a typical machine-made traditional cigarette. Fig. A3.1 Component parts of a machine-made cigarette Source: Author’s compilation. Photo by Walter Klerx. Not all parts of tobacco products are subject to the same level of control. According to Article 6.5 of the Protocol, five years following the entry into force of this Protocol, the MOP shall ensure at its next session that evidence-based research is conducted to ascertain whether any key inputs exist that are essential to the manufacture of tobacco products, are identifiable and can be subject to an effective control mechanism. On the basis of such research, the MOP shall consider appropriate action. In addition to the component parts of tobacco products, materials needed for packaging a specific number of sticks into a pack of cigarettes, usually 20 per pack, can be monitored. These materials include the foil paper, the package paper (which could bear the brand name, design and health warnings), the fiscal marking (if required) and the plastic or cellophane wrap. A fixed number of packs of cigarettes, Filter Tipping paper Cigarette paper Tobacco Tobacco rodFiltration zone 168 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N normally 10 packs, are packed into cartons, also called reams. These cartons are usu- ally made of soft paperboard or cardboard, possibly with branding, and are wrapped in plastic wrap or cellophane. Fifty cartons are packed in master cases, which are made of sturdier and thicker paperboard and stacked on pallets (usually 50 master cases to a pallet). An effective regulatory framework would require manufacturers and importers to provide information to the competent authorities on packaging and design, as well as the number of sticks per pack, carton and master case. Prior approval for purchase or sale of materials used in the cigarette production process can also be required. In the Philippines, suppliers of such raw materials, including those providing tobacco papers and filter components, are required to have a licence (2). In some of the Member States of the EU, raw tobacco is also subject to fiscal and legal requirements. For example, in Slovakia and Poland, raw tobacco can be handled only by authorized operators. While authorized operators do not have to pay excise duties on raw tobacco, if raw tobacco is detected by an unauthorized operator, excise duties will be due. Hungary, Italy and the United Kingdom require registration or authorization for all operators and growers handling raw tobacco (3). In addition to knowing the quantities of inputs required to produce a specific amount of a regulated product (e.g. cigarettes), the competent authority also needs to understand the supply, manufacturing and distribution chains to be able to properly monitor, regulate and determine whether taxes have been paid (see also Fig. 3.3). Novel and emerging nicotine, non-nicotine and tobacco products In recent times, new products have been introduced to several markets, namely, ENDS, ENNDS and HTPs. ENDS usually comprise a nicotine-containing e-liquid but do not contain to- bacco. ENNDS are essentially the same but do not (ostensibly) contain nicotine. The WHO COP requested the Convention Secretariat to invite Parties to monitor and report on scientific, regulatory and market developments such as initiation, cessation, advertising and promotion of ENDS and ENNDS. Furthermore, the COP requested WHO to report on the development of methods by regional and international standards-development organizations for the testing and measuring of contents and emissions of these products (4). There are different types of e-cigarettes – the most common type of ENDS and ENNDS – and currently there are four generations of products. However, they can be divided into two broad categories: open systems and closed systems. Both types of e-cigarette use a wick and a heat source to generate an aerosol. The wick is saturated with e-liquid, and a microprocessor is used to control operations (not all include this). Some e-cigarettes also have an LED light to imitate the burning end of a conventional cigarette (5). Fig. A3.2 presents examples of open and closed systems. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 169 E-liquid pod cover Rechargeable battery Heating element (heats solution, aerosolizing nicotine) Mouthpiece E-liquid pod Fig. A3.2 Examples of open and closed systems of ENDS/ENNDs products Open ENDS/ENNDS system (e-cigarette) Closed ENDS system (e-cigarette) Source: (6). Unlike ENDS/ENNDS, HTPs do contain tobacco. HTPs produce aerosols containing nicotine and toxic chemicals when tobacco is heated or when a device containing tobacco is activated (7). HTPs are composed of two elements: the sticks or pods that contain the tobacco and the device used to heat the tobacco. Both are necessary for the product to be used. Fig. A3.3 shows an example of a heated tobacco product. HTPs are tobacco products and are therefore subject to the regulatory measures contained in the WHO FCTC. Rechargeable battery Power button (to start vaping) Mouthpiece Atomizer / Heating element (heats solution, aerolizing nicotine) E-liquid tank (refillable e-liquid nicotine tank) 170 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. A3.3 Components of an HTP Holder: Tobacco stick: Note: PLA: polyactic acid, MPF: mouthpiece filter. Sources: (8-9). More information on tax administration of other tobacco products is presented in section 3.5. Tobacco stick Holder Charger Casing Control Electronics Battery Heating Blade Heatstick total lenght: 45mm MPF (7 mm) PLA (18 mm) Tobacco plug (12 mm) Hollow acetate tube (8 mm) Outer paper Diameter max 7.42 mm Tipping paper CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 171 REFERENCES 1. Memorandum by Gallaher Group Plc: the tobacco industry and the health risks of smoking. Select Committee on Health Minutes of Evidence. London: UK Parliament, 2000 (https://publications. parliament.uk/pa/cm199900/cmselect/cmhealth/27/0011323.htm, accessed 7 October 2020). 2. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 3. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco.. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 4. Electronic nicotine delivery systems and electronic non-nicotine delivery systems. Geneva: World Health Organization; 2016 (Decision FCTC/COP7(9) of the Conference of the Parties to the WHO Framework Convention on Tobacco Control; https://www.who.int/fctc/cop/cop7/FCTC_COP7_9_EN.pdf?ua=1, accessed 7 October 2020). 5. Brown CJ, Cheng JM. Electronic cigarettes: product characterization and design considerations. Tob Control. 2014;23:ii4-ii10 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3995271/pdf/ tobaccocontrol-2013-051476.pdf, accessed 31 January 2021). 6. E-cigarettes and vapor products [webpage]. King County, Washington (USA); 2019 (https://www. kingcounty.gov/depts/health/tobacco/data/e-cigarettes.aspx, accessed 9 October 2020). 7. Heated tobacco products. Geneva: World Health Organization; 2020 (WHO/HEP/HPR/2020.2 Information sheet; https://apps.who.int/iris/bitstream/handle/10665/331297/WHO-HEP-HPR-2020.2- eng.pdf?sequence=1&isAllowed=y, accessed 7 October 2020). 8. Tobacco heating system (IQOS) briefing document. Silver Spring: US Federal Drug Administration; 2018 (https://www.fda.gov/media/110377/download, accessed 31 January 2021). 9. Premarket tobacco product application: technical project lead review. Silver Spring: US Federal Drug Administration; 2017 (https://www.fda.gov/media/124247/download, accessed 7 October 2020). 172 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.2 EXAMPLE OF FORESTALLING AND COUNTERMEASURES This example is hypothetical but inspired from the situation in the Philippines. The amounts and prices have been altered, however, and it is assumed that the normal inventory kept by a tobacco manufacturer is two months.  The excise tax imposed on a pack of cigarettes in the current year is US$ 3.00; it will be increased to US$ 3.30 at the beginning of the new fiscal year (January in this example). The monthly production of Brand Y cigarettes of company X, which it declares for tax purposes, is as follows: MONTH OF THE CURRENT YEAR PACKS OF CIGARETTES January 10 000 000 February 10 500 000 March 9 900 000 April 11 000 000 May 10 200 000 June 10 600 000 July 9 700 000 August 10 100 000 September 9 900 000 October 10 100 000 November 20 000 000 December 25 000 000 Since the normal inventory is two months, the quantity in the two months prior to the implementation of the new excise tax rate is disregarded. The shelf life of tobacco products is approximately six months. The average of the six months prior to November is computed to obtain the quantity presumed to be produced or imported if there was no tax increase. The quantity from May to October (inclusive) divided by 6 is 10 100 000 packs. Thus, any quantity produced beyond 10 100 000 packs for the months of November and December (the months prior to the implementation of the new tax rate) is assessed using the new tax rate. In this example, 10 100 000 of the packs produced in November will be taxed at the old rate of $3.00, and 9 900 000 packs will be taxed at the new rate of $3.30. For December, 10 100 000 packs will be taxed at $3.00, while 14 900 000 packs will be taxed at $3.30. Without imposing these measures, the government would have been deprived of the excise tax increase on 24 800 000 packs. In addition, the effect of the increase on prices and consumers would have been delayed by approximately two months. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 173 In countries using tax stamps, the withholding of the issuance of stamps is a well-known approach to counter forestalling. Another practical solution is to allow the competent authorities to request advances from the industry to cover revenue shortfalls, provided there is a legal basis for such requests. 174 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 175 CHAPTER 4. Political economy As with any proposed government action, policy-makers need to navigate the political environment of tobacco taxation at every stage of policy development, implementa- tion and administration. While every country’s distinct history, culture, systems and structural forces shape its unique political landscape, there are some universal themes when it comes to tobacco control and particularly tobacco taxation. These themes boil down to the distribution of money, power and resources. The tobacco industry, as both a political and economic player, understands these themes well. The industry has been effective in using principles of the political economy of tobacco taxation in its efforts to block important advancements in tobacco control. Nevertheless, the savvy policy-maker can see through the industry arguments by considering who benefits from industry-favoured policy measures and interven- tions. The industry’s challenges to tobacco tax policies can be organized into the five categories of SCARE tactics. This chapter provides a road map to help policy-makers navigate the political economy of tobacco taxation through each of these themes. The first five sections dissect the tobacco industry framing of each issue, pinpoint- ing the flaws in each argument, identifying the extent to which each concern has merit and suggesting how a responsible government can address each one. These discussions are supported by unbiased evidence from independent, peer-reviewed research, as well as specific examples from country experiences. Sections 4.1 through 4.5 on SCARE tactics will equip policy-makers with the tools they need to proceed with confidence that their tobacco tax policy – developed and implemented fol- lowing the guidelines spelled out in this technical manual – will bring about the greatest health and economic benefits for their constituents, regardless of industry attempts to thwart them. Section 4.6 further buoys policy-makers’ efforts to ensure the beneficial impacts of their policies, as it describes how earmarking can improve the political economy of tobacco taxation by funding programmes and initiatives that promote and support the health and well-being of the population. 176 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.1 SCARE TACTIC S: SMUGGLING AND ILLICIT TRADE 4.1.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases will inevitably result in an increase in the illicit trade in tobacco products (1–2). They claim that higher tax rates and higher prices strengthen the financial incentives for criminal enterprises to supply cigarettes from lower-tax jurisdictions, boost domestic tax evasion and encourage smokers to seek cheaper illegal cigarettes. The industry also challenges the argument that tobacco tax hikes increase government revenue by claiming that the presence of an illicit tobacco market will actually reduce revenue collection following a tax increase. More recent versions of this argument – adapted to address public health concerns about tobacco use – claim that illicit market growth also offsets reductions in smoking prevalence that would otherwise be brought about by tobacco tax increases. In summary, the tobacco industry and its allies claim that raising tobacco taxes are ineffective – and even counterproductive – because they are circumvented by illicit markets, which prevents the government from achieving its public health objective and reduces rather than increases tax revenues. When a country considers a proposal to increase tobacco excises, the tobacco industry and its allies frequently make exaggerated claims about the size and scope of illicit tobacco trade in that country. Opponents of tobacco tax increases argue that price differentials are the exclusive – or at least the dominant – cause of illicit trade. Influenced by this fear-inducing faulty diagnosis, tax authorities frequently find it difficult to make decisions about tobacco taxes. However, the industry diagnosis always contains the same erroneous elements. First, the illicit trade in a country is frequently less than the industry portrays it to be, and the country’s tax enforcement policy towards tobacco products is rarely unique or in any way different from the norm in the country (3). Second, the scale of illicit trade in tobacco is not exclusively or even primarily determined by tax or price differentials. Typically, it results from a set of governance problems characterized by government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, lack of dissuasive sanctions and/or weak norms regarding participation in illegal and informal markets (4–7). This section provides guidance for tax and other relevant authorities on how to respond to the tobacco industry SCARE tactic that increasing tobacco taxes will lead to smuggling and illicit trade in their countries. Tax authorities need to know the nature, causes and extent of illicit trade so that they can define the problem properly and formulate an appropriate response. This chapter addresses the available tools for better defining and understanding specific illicit trade problems, particularly tools that facilitate independent assessment of the magnitude of that trade. Improvements CHAP T ER 4. PO LI T I C AL ECO N OMY 177 to governance within the realm of tax authorities – such as best practices in tobacco tax administration and policies to improve the effectiveness of fiscal regulations or norms regarding participation in informal and illegal markets – are discussed in Chapter 3. This section first describes the nature of the illicit tobacco trade to highlight some of its complexities and identify complementary policies for tackling the problem. Next, evidence that calls into question the link between illicit trade and high prices or tax rate changes is discussed. Finally, to help tax authorities assess their own situation, several different methodologies are presented to estimate the scope of the illicit tobacco trade and to evaluate estimates of that trade for a particular country or tax jurisdiction. 4.1.2 THE NATURE AND EXTENT OF THE ILLICIT TOBACCO TRADE The WHO FCTC defines illicit trade as: any practice or conduct prohibited by law and which relates to production, ship- ment, receipt, possession, distribution, sale or purchase, including any practice or conduct intended to facilitate such activity (8). Non-duty-paid tobacco products found in a jurisdiction (i.e. through littered-pack surveys) could be the result of either of two related but distinct activities: tax evasion and tax avoidance. Tax evasion is a set of unlawful actions seeking the non-payment of tobacco taxes and duties, whereas tax avoidance comprises legal actions with the purpose of avoiding payment of some or all taxes, such as bringing an amount of cigarettes up to the legal allowance from a lower- into a higher-tax jurisdiction. Tax avoidance is not illegal and is therefore not considered part of illicit trade in tobacco products.1 The focus of this section is on tax evasion activities, which can occur in the movement across borders or in domestic production and distribution. When tax evasion happens across borders, it is known as smuggling (9) and can be done on a large scale or a small scale. Tax evasion in the domestic market can be partial, when licensed and authorized producers or distributors comply with only part of their tax obligations, or total, when the whole production and distribution system is illegal and out of sight of tax administrators (5, 10–14).2 Large-scale tax evasion schemes can be run by different types of producers and their associated distributors, such as the transnational tobacco companies (TTCs) and their national subsidiaries, 1 Tax avoidance practices –common among states in the United States and countries in the EU – are not analysed in this section. 2 Tax evasion is normally considered as illicit manufacturing in the literature (14). 178 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N other local tobacco companies producing their own brands and illegal factories that normally counterfeit other brands or produce illicit brands. Large-scale smuggling involves, for example, taking advantage of tax-free zones and mislabelling shipping products prior to or during transit (11) or using so- phisticated clandestine networks. This form of tax evasion is systemic and can be carried out by TTCs (12), by local companies producing in countries with low tax enforcement, e.g. Paraguay (15) that feed neighbouring countries and regional illicit hubs through a network of clandestine distributors (15) or by companies located in tax-free zones like Jebel Ali and Dubai in the UAE (16) or such zones in Russia or Cyprus (17). The origins, routes and quantities of large-scale smuggling frequently change as affected countries or markets react by strengthening enforcement and seeking bilateral cooperation with the jurisdictions of origin. For example, the 2013 comprehensive strategy of the EU (10) enhanced bilateral cooperation with major source and transit countries of illicit cigarettes coming to Europe, including Russia, Belarus and Ukraine. These agreements improved day-to-day cross-border coopera- tion, reduced illegal flows and introduced gradual tobacco excise rate adjustments in those three countries to bring them to European levels. Small-scale smuggling (also known as ant smuggling or bootlegging) is the cross-border trafficking of cigarettes in quantities that are larger than the allowable limits (e.g. two cartons) but smaller than large shipments (e.g. truckloads, cargo containers), normally for the purpose of selling at a profit (11). This type of illicit trade may exist in places where there are opportunities within neighbouring tax jurisdictions. For example, small-scale smuggling is commonly done by individuals living in French and German provinces near lower-taxed countries (e.g. Belgium, Luxembourg, Switzerland, Spain, Poland and Czechia) (18). Counterfeiting is a form of illicit manufacturing that involves the production of tobacco products (including packaging and tobacco filler) without the approval of the trademark holder (13). Another product of illicit manufacturing is so-called cheap or illicit whites. Cheap whites are branded (e.g. Jin Ling) or unbranded cigarettes that are legally or illegally produced3 and knowingly sold in the illicit market (17). Cheap whites are not usually produced by TTCs (17, 19).4 They are produced by small tobacco produc- tion companies in one country and often sold in illegal markets of neighbouring 3 Ross et al. (17) analysed this issue and found that the sale to the first purchaser is usually legal. Their analysis covers the production in free zones (i.e. in the UAE, Russia and Cyprus) and production exported from Viet Nam, Indonesia and China. In those cases, there is no need to make the first sale illegally. However, cheap white production in Paraguay is sold to domestic distributors, and most of those sales are completely illegal. 4 Ross et al. (17) and Gilmore et al. (19) identify some cheap white brands sold by TTCs, such as President (PMI), produced in Ukraine, and Esse (Korea Tobacco & Ginseng Company, KT&G), produced in Indonesia. CHAP T ER 4. PO LI T I C AL ECO N OMY 179 countries. For example, in Paraguay, cheap whites are produced on a large scale by a few companies under the guise that they are marketed domestically, but a large share is smuggled into Uruguay and Brazil (4). Iglesias et al. (20) showed how TTCs’ cheap brands were illicitly shipped through Paraguay to be sold in the Brazilian and Argentine markets in the 1990s. This contributed to increased production of cheap whites in Paraguayan firms, which continued the illicit business even after Brazilian legislation obstructed the illegal activity of the TTCs. Domestic tax evasion is a pervasive phenomenon, particularly in LMICs. Partial tax evasion in tobacco products can be found at any level of tax rates or prices and is generally the result of defective legislation or weak tax enforcement.5 Complete or total tax evasion occurs when producers and distributors are clandestine or when there are serious institutional challenges to tax enforcement between two tax jurisdictions, such as between the United States and Native American Reservations. Evidence of illicit manufacturing has increased in recent years in several places in the world, including the EU (10) and Brazil (21). TTCs were predominant in illicit trade activity until the end of the 20th century, and even with the entrance of new actors into the illicit business, TTCs have not entirely exited. Gilmore et al. (22) analysed industry-funded data and seizure data and concluded that TTCs are still involved in illicit trade in Europe, despite the Anti-Contraband and Anti-Counterfeit Agreements (the “Agreements”) signed between the four TTCs and the EU (23).6 Using industry-funded data, Gilmore et al. show that 58% of illicit EU cigarettes can be attributed to the four main TTCs. When seizure data are used, 69% to 73% of illicit EU cigarettes can be attributed to these firms (22). It is always difficult to assess the extent of the global illicit tobacco trade because of its illegality, its global and changing nature and problems with data collection (24). Before the 21st century, when TTCs were almost unique actors in the large-scale smuggling of well-known cigarette brands, the difference between global exports and imports of cigarettes could provide a good approximation of the size of this problem globally (7). However, with the growth of illicit manufacturing in general, the manufacturing of cheap whites and the illegal movements of those products 5 This occurs when licensed and authorized producers underreport actual quantities and sell the non- duty-paid produced quantities through illegal channels. It can also include instances when producers do not report quantities at all, as in many ad valorem systems of LMICs. 6 From 2000 on, the European Commission and 10 Member States launched court cases regarding smuggling and money laundering against several TTCs. To end the court cases, the Anti-Contraband and Anti-Counterfeit Agreements were signed, which required the TTCs to exercise stringent control over their supply chain (through tracking and tracing, due diligence and anti-money-laundering and reporting obligations), share operational intelligence with Member States and the EU and pay penalties for seizures, as well as annual payments over a period of 12 years. The agreement with PMI has ended, the one with JTI will end in 2022 and the others with Imperial Tobacco and BAT will run until 2030 (23). 180 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N over the past two decades, trade statistics are no longer as useful as they were in the past. Joossens et al. (25) tried to estimate the size of the global illicit cigarette market by adding different types of national estimations prepared around 2007. They found that the estimated size was 657 billion cigarettes per year, or 11.6% of the global cigarette market. According to Joossens et al., illicit trade costs govern- ments US$ 40.5 billion in tax revenue worldwide, and eliminating illicit tobacco trade would recover US$ 13 billion in immediate revenue in high-income countries and US$ 18.3 billion in LMICs.7 Descriptions of the types of illicit trade are useful for developing the first com- ponent of a strategy to fight it: assess the nature and size of the problem. Table 4.1 presents all the main components of a strategy to fight illicit trade. To make progress in this first component – knowing the problem – authorities could use and adapt existing instruments of health surveillance or seek partnerships with academia and independent specialists to investigate the issues involved, using different methodolo- gies (see subsection 4.1.4 and Annex 4.1 on methodologies to assess the nature and size of the problem). Knowing the nature of the problem requires the cooperation of different government actors – for example, to investigate both the financial and criminal operations of organized crime behind the illicit trade. The gathering of qualitative information on the nature of the illicit trade should start simultaneously with the statistical work of measuring the magnitude of the problem. Table 4.1 Components of a strategy to fight the illicit tobacco trade 1. Assess the nature and size of the problem Use and adapt existing health surveillance and other existing national surveys to assess the problem Seek partnerships with academia and independent specialists to find ways to rigorously study illicit trade Use financial and police investigations to identify and fight organized crime operating in illicit trade 2. Start identifying and implementing appropriate country-specific policies and strategies to address illicit trade Improve tax and customs administration to close the legal and administrative loopholes facilitating illicit trade Implement other appropriate policies to deal with country-specific problems 3. Become a Party and/or implement the Protocol to Eliminate Illicit Trade in Tobacco Products Adapt the Protocol supply-chain control obligations Adjust national penalties for illicit trade offences Seek and build international cooperation 7 The WCO publishes an Illicit Trade Report annually, with the main characteristics and trends of illicit flows in key products, including tobacco, using data based on customs seizures. CHAP T ER 4. PO LI T I C AL ECO N OMY 181 Methodologies available to estimate the nature and size of illicit trade are discussed in subsection 4.1.4. This is the first step for dealing with SCARE tactic S. Chapter 3 discusses at length the relevant tax administration measures and best practices to minimize opportunities for illicit trade in tobacco products. Table 4.2 presents examples of appropriate policies and strategies targeted to address specific types of illicit trade in addition to the best practices described in Chapter 3. After completing the first step of this strategy, tax, health and justice authorities should discuss how to face country-specific problems, considering not only tax and customs administration measures but also social, law enforcement and international cooperation policies and strategies. Table 4.2 Suggested policies and strategies to address country-specific illicit trade problems MAIN TYPE OF ILLICIT TRADE IN THE JURISDICTION PROBLEMS POLICIES/STRATEGIES TO USE Bootlegging Neighbouring low-tax jurisdiction Bilateral negotiations to harmonize tobacco tax systems Difficulty of controlling people’s movements in countries with extensive land borders Identify and establish suitable social protection or employment policies for targeted populations in border regions Extensive land border with multiple accesses Bilateral cooperation with law enforcement and border control forces, monitoring of access routes to main consumption markets Large-scale smuggling from neighbouring jurisdiction Neighbouring low-tax jurisdiction and difficulties in controlling borders Bilateral negotiations to harmonize tobacco tax systems and bilateral law enforcement cooperation Producers and distributors in the lower-tax jurisdiction aiming to supply the high-tax jurisdiction Bilateral cooperation to harmonize tax systems and control producers and distributors in the origin country, create conditions for legal exports and taxed imports Large-scale smuggling from a third country or tax-free zones Producers and distributors aiming to supply non-duty-paid tobacco products wherever possible Customs and other forms of international cooperation to control and monitor exports from identified areas Domestic tax evasion Existence of many small informal or semi-formal producers Encourage business concentration through producer associations and cooperatives, create incentives for formalization and establish licensing rules and basic electronic information systems for raw material and production 182 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Underreporting from formal producers Improve tax administration with policies such as basic electronic information systems for inputs and production, establish neutral procedures to verify production, improve audit systems, increase third-party information on inputs and production of tobacco products Clandestine factories Law enforcement investigation of commercial associations with raw- material and machine producers and distributors Governments should try to identify the incentives and governance problems that encourage and allow illicit trade movement inside their country. As seen in Table 4.2, the design and implementation of policies to deal with those problems do not depend exclusively on tax and customs authorities; they also depend on the efforts of the police and law enforcement, the Justice Department and the judicial apparatus. In other words, a great deal of coordination and consultation among different types of government bodies and expertise is needed to produce an adequate response. It is also clear from Table 4.2 that domestic tax evasion by formal producers can be tackled by tax authorities and is mainly related to the supply-chain-control provisions of the Protocol. The Protocol (26) builds upon and complements Article 15 of the WHO FCTC, which addresses means of countering illicit trade in tobacco products as a key aspect of a comprehensive tobacco control strategy. The Protocol is a blueprint of measures to deal with this problem, and its provisions should be part of any strategy for fighting the illicit market. It is a legally binding treaty in its own right that entered into force on 25 September 2018. As described in detail in Chapter 3, the Protocol has three main lines of action: supply-chain controls, recommendations on how to treat unlawful conduct related to the illicit tobacco trade and suggested mechanisms to seek and build international cooperation to fight that trade. Countries can start implementing Article 15 of the WHO FCTC and the appropriate polices or strate- gies recommended by the Protocol even before acceding to it, selecting those most suitable to the nature and extent of their particular problem. Such transitional work will facilitate the eventual implementation of the Protocol, because any plan to correct loopholes in tax and customs practices will bring government authorities closer to the best practices recommended in the Protocol. 4.1.3 DETERMINANTS OF TAX EVASION: THE ROLE OF PRICE LEVELS The argument that price and tax rates are the main determinants of the illicit tobacco trade has persuaded some governments (e.g. Uruguay and Georgia in the past) to avoid policies that may lead to cigarette price increases (e.g. excise tax rate increases) (4). CHAP T ER 4. PO LI T I C AL ECO N OMY 183 Some governments (e.g. Canada in 1994, Brazil in 1999 and Pakistan in 2017) (20, 27) have even reduced tax rates in attempt to reduce the illicit trade. The wider scholarly literature demonstrates that illicit trade is not a monocausal phenomenon (7) but is the result of many factors, most of them related to gover- nance issues. Government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, weak norms regarding participation in illegal and informal markets and conflicts between neighbouring countries (5) all contribute to the existence and growth of the illicit tobacco trade. It is difficult to isolate the role of price from each of the other factors because (1) obtaining prices and quantity measures of illicit trade is inherently challenging; (2) in most countries, there are many cigarette brands, and prices vary between and even among brands; and (3) there is a lack of good measures to deal with nonprice factors affecting illicit trade, such as government corruption and ineffective criminal justice. These constraints make it challenging to develop rigorous empirical evidence about how price and other factors affect illicit trade. Despite these fundamental challenges, the economic literature has produced credible evidence that price is always only one factor – and often not the most important factor – determining the extent of illicit trade. Many econometric studies about the influence of price and other factors have focused on cross-border shopping (or small-scale bootlegging from low- to high-tax jurisdictions), given the availability in the United States and Europe8 of sales data for low- and high-tax jurisdictions, classified in a convenient way by geographical zones – i.e. close to or far from the borders. Those studies attempted to explain the illicit trade flows or the relatively higher sales in low-tax jurisdictions as a function of price and tax differentials between the lower-tax and surrounding higher-tax jurisdictions, after controlling for other important factors affecting cross-border sales such as proximity to borders and levels of corruption (6, 7, 11, 28).9 The main conclusion of the studies is that illicit trade flows are not linked solely to price (29). Some show a significant effect of price differentials together with other factors, but others do not find significant price differential effects. The important policy implication of these analyses is that decreasing tobacco tax rates and real prices in higher-tax jurisdictions could have minimal or no effect on illicit market shares.10 8 This was a traditional strand of the literature in the United States on trade among states, and to a lesser extent in European countries, most of which used conventional but inaccurate illicit trade measurements. 9 Recently, PMI-Altria financed some studies of factors affecting cross-border sales. One of those studies, Prieger and Kulic (28), criticized Merriman et al. (2000) (9) and arrived at the conclusion that in cross-border shopping, price differentials are important for determining the magnitude of illicit trade. 10 Brazil decreased tax rates and real prices at the beginning of the 21st century to fight illicit trade coming from lower-tax jurisdictions. After this action, however, the government lost revenues, and the size and scope of illicit trade remained unaltered, according to industry sources (20). 184 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Observational and case studies provide information that may improve public policy even when they are unable to produce compelling evidence of causal relation- ships. Some observational studies have correlated price levels with illicit market shares, using large samples of countries. Joossens et al. (25) found that countries with high taxes and prices normally have lower shares of illicit trade than countries with lower tax shares and prices. In their sample, high-income countries generally have relatively high cigarette prices and tax shares, but their favourable results (i.e. lower levels of illicit trade) are related to effective tax administration and lower corruption levels. In contrast, LMICs generally have lower prices and tax shares, along with significant illegal market shares. Joossens et al. attribute difficulties in fighting illicit trade to weak tax and customs administrations and, in most cases, institutional and legal challenges (25). Figure 4.1.1 illustrates the relationship between price and illicit trade, using the price (in US$) per pack of the most-sold brand of cigarettes and the estimated level of illicit trade for 94 countries in 2018.11 There is no apparent unique association between the two variables. Running a linear regression with retail price as the explana- tory variable and share of illicit trade as the dependent variable shows an inverse, but not statistically significant, relationship between price and illicit market share.12 Figure 4.1.1 illustrates some particular cases: • Many countries with low prices (i.e. lower than US$ 2 per pack) have the highest levels of illicit trade in the sample, e.g. Brazil (BRA) ($1.33 and 46.3% illicit share), Pakistan (PAK) ($0.39 and 40%), Ethiopia (ETH) ($0.55 and 32.9%), Ghana (GHA) ($1.06 and 29%) and Cameroon (CMR) ($0.89 and 25%). • In contrast, many of the countries with prices between US$ 4 and $8 – which could be considered high enough for financial incentives to operate – have illicit trade shares of less than 10% of total consumption. These countries include the Republic of Korea (KOR) ($4.02 and 0.8%), Czechia (CZE) ($4.31 and 2.9%) and Sri Lanka (LKA) ($6.89 and 1.6%). • All countries that have very high prices – higher than US$ 8 – except for Ireland, register illicit trade shares below 20%. These countries include France (FRA) ($9.39 and 17.8%), Switzerland (CHE) ($8.71 and 5.5%), Singapore (SGP) ($10.35 and 3.7%) and Norway (NOR) ($14.51 and 9.6%). 11 National estimates of the magnitude of illicit trade are controversial. The tobacco industry’s numbers overestimate the problem and are based on questionable methodologies. Estimates with a rigorous and transparent methodology are not available for a large sample of countries for the same year. In order to compare price levels with illicit market shares, Euromonitor’s estimations of illicit market share were selected, for two reasons: they are comparable estimates for a large sample of countries in a given year, and no one could argue that they are biased towards tobacco control’s points of view. The use of Euromonitor data does not imply that WHO fully agrees with all the details and methodologies used to obtain them. 12 Other factors must be taken into account to transform this observational analysis into a rigorous analysis of cause and effect. CHAP T ER 4. PO LI T I C AL ECO N OMY 185 Fig. 4.1.1 Share of illicit trade versus retail price of the most-sold brand of cigarettes in US$, by country, 2018 Note: The extent of illicit trade in cigarettes is measured by Euromonitor as the estimated quantity of illegal cigarettes consumed in a country divided by the estimated total consumption of cigarettes in that country. Sources: (27, 30). 51 2 3 4 6 7 8 9 10 11 12 13 14 1615 0 10 20 25 15 5 30 35 40 45 50 MYS BRA ECU ECUPAK ETH PAN CRI URY GRCIND GTM SLV ARE LVAHND NGA ZAF CAN FRA LBNAGO IRL DOM GHA CMR Retail price, USD per pack Ili ci t t ra de % IRQ VNM COL MMR AUT SWE EST PER ESP ROM POL CHL ISR GBR KEN LAO KHM EGY MKD TZA THA TUN BGD UZB UKR GEO KAZ CHN KWT HRVJOR CZE SVK KOR JPN BLR SRB TUR OMN FIN NLD CHE SGP SAU USA LKA DNK DEU ITA HUN SVN PRT BOL DZAIDN CIV BEL MAR ARG AUS NOR NZL LTU BIH MEX BGRPHLAZE RUS 186 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As indicated in numerous studies and analyses of illicit trade in tobacco products (4–5), the most effective way to tackle the problem is not to forgo tax increases but rather to strengthen the capacity to fight the trade. Therefore, it is important to consider the relationship between good governance and illicit trade. The more capacity a country has to counter illicit trade in general, the lower the level of that trade will be. An index compiled by the Economist Intelligence Unit (EIU), the Global Illicit Trade Environment Index, measures countries’ structural capacity to fight illicit trade overall. The EIU indicator is a combination of four indicators or categories designed to assess countries’ performance in those areas; the closer the overall indicator is to 100, the better the country’s capacity to fight illicit trade. The four categories are:13 1. government policy, which measures the government’s commitment to pro- actively monitoring and preventing illicit trade; 2. supply and demand, which measures the extent to which the domestic en- vironment discourages or encourages supply and demand for illicit goods; 3. transparency and trade, which measures transparency and the degree of governance applicable to free-trade zones and transhipments; and 4. customs environment, which measures how effectively customs services facilitate legitimate trade while at the same time preventing illicit trade. Figure 4.1.2 illustrates the relationship between the EIU indicator and the estimated level of illicit trade in cigarettes in a set of countries. There is an inverse and statisti- cally significant relationship between the indicator and the estimated level of illicit trade in cigarettes. This suggests that as the capacity to fight illicit trade in general increases, the illicit trade in cigarettes falls.14 13 For more details about this indicator, visit http://illicittradeindex.eiu.com/. 14 The association was significantly different from zero at a 90% confidence level, using a linear regression between the two variables. CHAP T ER 4. PO LI T I C AL ECO N OMY 187 Fig. 4.1.2 Share of illicit trade versus the EIU indicator in 70 countries, by country, 2018 Sources: (30–31). % Il ic it tr ad e ci ga re tt es MYS 0 10 15 5 20 25 30 35 40 45 50 55 60 EIU GIT indicator 20 3010 40 50 60 70 80 90 JPN KOR HRV SAU UKR BLR KHM LAO MMR IRQ GTM DOM VNM MAR SRB PER BGR MEX ROM COL ZAF BIH PHL IDN ITA PRT HUN BEL DEU ESP ARG AUS SWECHL TW POL ISR AUTLTU FRA CAN IRL LVA ARE GRC IND URY CRI PAN PAK ECU BRA EST GBR FIN SVN NLD SVK TUN RUS DZA TUR CHNKAZ THA CZE USA SGP NZL DNK 188 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Rigorous independent research has established that despite the challenges of illicit trade, taxation of tobacco products is an effective public health intervention that substantially reduces tobacco use and generates government revenue (5). Further, when cigarette taxes increase, governments generate higher revenue and consumption is reduced (32–33). However, ineffective tax administration can allow illicit trade to grow and can undermine some of the benefits of tobacco taxation by making cheaper cigarettes available. For example, the average street price of smuggled cigarettes in Malaysia is 55% lower than its legal tax-paid equivalent (34). Illicit tobacco trade also reduces government tax revenue and may increase health costs associated with smoking and costs associated with policing. 4.1.4 MEASURING ILLICIT TRADE IN TOBACCO PRODUCTS The magnitude of illicit trade is a powerful argument in tax policy discussions, and for this reason the tobacco industry funds estimation of illicit trade in countries or regions of particular interest to itself (i.e. Project Sun and Project Star in the EU and Oxford Economics in East Asia). However, a recent systematic review of industry data on illicit trade finds substantial methodological weaknesses in industry-commissioned reports (24). Furthermore, Blecher et al. (35) argue that industry-funded studies tend to systematically overestimate the size of illicit trade to persuade authorities to abandon tobacco tax reforms. Independent researchers have also uncovered inconsistencies in tobacco-industry-funded estimates (36). Some examples of inflated industry-linked illicit trade estimates are given in Table 4.3, which compares peer-reviewed and independent studies with estimates funded by the tobacco industry. Because some countries have several industry estimates from different sources or years, Table 4.3 presents the estimate included in the article that published the independent study, because it was considered as representative and adequate to illustrate the overestimation. In all cases, the industry estimates exceed those of the independent studies. Measuring the scale of illicit trade can be a daunting task for governments because different methods are employed by independent researchers, governments and the tobacco industry. Nonetheless, it is worth investing in these studies because they drive policy discussions and can be used to evaluate the impact of policies (e.g. tax increases, plain packaging and health warnings). CHAP T ER 4. PO LI T I C AL ECO N OMY 189 Table 4.3 Illicit market share estimated in independent studies compared with estimates in tobacco-industry-funded studies COUNTRY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDEPENDENT STUDIES SOURCE OF THE INDEPENDENT STUDY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDUSTRY- FUNDED STUDIES INSTITUTION RESPONSIBLE FOR THE INDUSTRY- FUNDED STUDIES Colombia 3.5% of the total market in five cities, 2016 Maldonado et al., 2018 (37) 13% of the total market, 2014 FND and INVAMER, 2015 Chile 16.3% of the total market in the Metropolitan Region of Santiago, 2017 Paraje et al., 2020 (38) 24.3% of the total market, 2017 Observatorio del Comercio Ilícito BATC, 2017 Brazil 28.8% of the total market, 2014 Iglesias et al, 2017 (39) 34%, of the total market, 2014 BAT public statement, 2015 Mexico 8.8% of the total market in eight major cities, 2017 Saenz de Miera Juarez et al., 2020 (40) 16.6% of the total market, 2012 Confederación de Cámaras Industriales, 2012 As shown in Table 4.4, methodologies to measure illicit trade can be grouped into three types: (1) direct measurement; (2) residual methods and (3) expert opinion (12). Direct measurements rely on evidence directly linked to actual illicit behaviour and pack observation; residual methods infer evasion based on theory and evidence about consumption and legal sales; and expert opinion distills information garnered from talking to individuals with the most direct knowledge of the tobacco market. Each method has advantages and disadvantages. No single method is unambigu- ously superior to others, but direct measurement and residual methods are more conducive to determining the size of the illicit market, whereas expert opinion could provide insight into the details of the market’s operations.15 Table 4.4 presents the relative amount of resources and the degree of expertise required to implement each main measurement method, as well as the primary purpose, data collection characteristics, sampling features and unit of analysis. A brief description of each of the methods is presented in Annex 4.1. Merriman (11) and Ross (9) provide more expansive details. There is no simple selection rule for deciding what measurement method to use. The major factors to consider when selecting a method or methods include (1) the nature and characteristics of the illicit trade problem (i.e. where and how the 15 In interviews with experts from the tobacco industry, provisions of Article 5.3 of the FCTC and its Guidelines need to be followed. 190 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N problem manifests and whether domestic tax evasion or illegal inflows of foreign brands or a combination of both predominates), (2) previously collected data, (3) available budget and (4) expertise of available analysts. Available budget and staff skills are often the main restrictions that governments face. Therefore, Table 4.4 orders the measurement methods according to resources needed and available expertise. For example, residual methods and expert opinion can provide crude but useful estimates at low cost and require the lowest levels of technical sophistication. Another low-cost option for countries that employ popula- tion health surveillance surveys is to add questions to measure illicit trade, such as brand name, value and quantities of the last purchase. In contrast, the direct measurement approach often requires sophisticated research designs and expensive (and time-consuming) field research. Table 4.4 Overview of resources and expertise needed and main purpose of measurement methods METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Seizures (D) $ Low Identify trends in types of products, transporta- tion methods, points of entry and brand names Secondary data use Non- probability Shipments Use of existing health sur- veillance surveys – self- reported consump- tion (D) $ Low, only additional questions Size of illicit trade, adding or improving questions on brands, value and quantities of the last purchase Additional primary data collection Probability Individuals Gap analysis (R) $ Medium Provides a measure of changes in illicit trade Secondary data use Universe Nations Econo- metric modelling (R) $ High Estimation of price elasticity of substitution from tax- paid to illicit products Secondary data use Universe Geography Expert interviews (E) $ Low Characteristics of the illicit trade Primary data collection Non- probability Individuals Smoker intercepts and pack observa- tion surveys (D) $$$ Medium Size and characteristics of illicit trade, probability-based sample to be representative of population Primary data collection Probability Individuals CHAP T ER 4. PO LI T I C AL ECO N OMY 191 METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Pack return and swap surveys (D) $$$ Medium Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Littered- pack surveys (D) $$$ Medium Size and characteristics of illicit trade, compa- rability with industry estimation using empty-pack surveys Primary data collection Probability Individuals Covert purchases (D) $$$ Medium Type of products and trade channels of illicit trade Primary data collection Probability Geography Self-report consumer surveys (D) $$$ High Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Notes: Universe includes total population; D = direct measurement, R = residual method, E = expert opinion. Scale for resource costs assuming a moderately sized study (e.g. a representative study of a region of several million): $ (cheapest) – weeks of skilled labour hours; $$ (moderately expensive) – 1 to 2 months of skilled labour hours; and $$$ (most expensive) – 6 to 12 months of skilled and unskilled labour hours. A more detailed description of the different measurement methods is given in Annex 4.1. To further assist responsible authorities in deciding which method to select, Table 4.5 presents the key characteristics of each of the measurement methods, along with the main advantages and disadvantages of each. Countries may begin with methods that require fewer resources and less skills to obtain an overview of the problem. Seizures – which are a by-product of law enforcement efforts – pro- vide a first step, and countries can analyse the information obtained (origin of the products, brands, location, etc.) and report the results to increase public awareness of the problem.16 Alternatively, countries can add questions related to illicit trade to existing and funded health surveillance surveys conducted regularly by health surveillance authorities and statistical authorities. In that way, cooperation in using existing measurement methods between health authorities – the tobacco control office and health surveillance unit – tax and customs authorities and the national 16 Seizures are useful for obtaining qualitative information about the illegal activity, but they have to be treated very cautiously in projecting the size of the problem. Countries may think they have a very large problem because they have competent authorities doing an extraordinary job at finding illicit goods. On the other hand, countries can have less-efficient authorities making few seizures, and in these environments, seizures tell nothing about the size and nature of the problem. 192 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N statistical office could be a starting point for identifying the nature and size of the illicit trade problem in the country. Direct observation of packs has been increasingly implemented in many LMICs, through different types of surveys such as intercepts of smokers or retailers, pack return, littered-pack inspections and covert purchases of cigarettes. These activities have expanded the skills of independent researchers and academia and increased knowledge of these methods. Also, increasingly cheaper digital technologies allow interviewers to take pictures and record pack characteristics in direct observation surveys or in larger national self-report consumer surveys. Table 4.5 Key characteristics, advantages and disadvantages of illicit trade measurement methods METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Seizures (D) Statistics of tobacco products confiscated by local and national authorities Readily available from law enforcement agencies May not provide a representative picture of the size and/or nature of illicit trade Using existing health surveillance surveys to obtain self-reported consumption (D) Adding or improving questions about brand names, quantities, prices, locale of purchase and other factors Produces good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Gap analysis (R) Compare self-reported consumption data with observed (usually administrative) data about tax-paid sales When quality data are available, is simple and easily reproduced (providing for measurements over time) and explainable Data on tax-paid sales and/or consumption are frequently inaccurate and in many cases do not provide information on the size of the illicit market, but only on changes over time Econometric modelling (R) Estimated according to the difference between tax-paid sales and predicted consumption given by the model Because it is consistent with a long tradition of economic theory, empirical estimates can be evaluated Requires high-quality data on a variety of important variables over a period of time and advanced econometric modelling expertise CHAP T ER 4. PO LI T I C AL ECO N OMY 193 METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Expert interviews (E) Experts include researchers (e.g. in economics, criminal justice and public health), journalists, tax and enforcement specialists, product manufacturers and wholesalers Useful for identifying the nature of and trends in the marketplace (e.g. venues where illicit cigarettes are sold, modes of entry), and the interviews can be useful for defining the method to assess the size of the illicit trade Information obtained may not be generalizable, and expert knowledge may be outdated or limited by the experts’ experience; also, experts often have strong biases Smoker/retailer intercepts and pack observation surveys (D) Examining the packs of smokers and cigarette retailers, convenience or probability-based sample Is direct and objective, and smokers do not suffer from any value judgements The difficulty of identifying areas representative of the tobacco use population and sampling important subpopulations such as elderly and immobile smokers, but household surveys could overcome sampling issues Pack return and swap surveys (D) Also a pack observation study using survey sampling techniques to examine smokers’ pack characteristics May decrease the stigma associated with traditional smoking surveys In LMICs, survey distribution may be unreliable because of the mail delivery system Littered pack surveys (D) Also known as empty-discarded-pack surveys; publicly discarded packs bear characteristics (e.g. tax stamps, public health warnings) that indicate whether they are tax compliant Yields estimates that are less likely to be biased from issues of social desirability, recall error and confidentiality Significant budgets could be needed to employ field researchers to collect, code and analyse the data; surveys do not provide information about the smoker and the price paid Covert purchases (D) Uses covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes Directly identifies sources of illicit cigarettes It is difficult to create a sampling frame of retailers for illicit sources or to know what smokers are actually buying and how much Self-report consumer surveys (D) Surveys can be distributed to individuals or households, using various modes of distribution Good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Notes: D = direct measurement, R = residual method, E = expert opinion. 194 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ultimately, when capacity allows, more solid estimates will need to be made using more than one methodology at a given point in time. Ideally, estimates will be made on a regular basis in order to assess the evolution of illicit trade over time and its possible connection to policy changes. 4.1.5 CRITIQUING STUDIES THAT MEASURE THE SIZE OF THE ILLICIT MARKET Measuring illicit trade is a challenge for researchers, industry and governments, because the trade is, by definition, hidden from plain sight. Buyers and dealers sometimes go to great lengths to ensure that their participation in illicit activity is concealed. Over the years, researchers and government agencies have been increas- ingly interested in estimating the size of illicit markets and identifying effective interventions. As consumers of research, governments should critically examine available studies and evaluate them on their scientific rigor and methodological transparency. Measurement issues are particularly acute with respect to the illicit tobacco trade because it is a politicized topic. High estimates may raise questions about the tobacco industry’s ability to control the supply chain, its involvement in illicit diversion, the impact of taxation policies and the effectiveness of enforcement strategies. While the industry has portrayed itself as taking an active stance in measuring and fighting illicit trade (e.g. Project Star, conducted by KPMG LLC but paid for by PMI, later followed by Project Sun), in the past it has used smuggling as a strategy to enter closed markets – for example, in China and Russia (19, 41). Govern- ments should carefully scrutinize evidence about the illicit tobacco trade produced by industry or quasi-industry sources and are advised to seek alternative evidence. Quasi-industry reports are studies commissioned by the industry but published by private research companies (e.g. Ernst and Young, Oxford Economics) (11). Characteristics of good analyses One of the main characteristics of a good analysis is scientific rigor, which involves the use of relevant theoretical frameworks, sound statistical methods and examination of the robustness of findings (e.g. sensitivity analyses). High-quality research reports provide transparent explanations about their methodology and statistical analysis steps undertaken, as well as supplementary analyses that established the robustness of the findings. For example, Joossens et al. (25) clearly describe the data sources used (limitations and advantages and where they can be found) and calculations performed on the number of lives that would be saved if the global market share of illicit cigarettes was eliminated. Explanations should be detailed enough to allow future researchers to scrutinize the analysis and replicate the findings. Replicability is another hallmark of good science. For example, littered-pack studies should detail where and when data collection took place, how many packs CHAP T ER 4. PO LI T I C AL ECO N OMY 195 were collected per geographical unit, the protocol of identifying the illicit packs (e.g. characteristics of the warning labels, brands, tax stamps, etc.) and details of statistical analyses. There should also be explanations of the representativeness of the selected geographical areas. Failure to provide this depth of information may call into question the generalizability of a study and whether there are faults with the chosen method. In the context of policy decisions regarding illicit trade, the most useful data provide information about a representative sample of individuals and geographies. Studies that are limited to, for example, one group of individuals based on specific characteristics or a given geography may yield biased information. Research reports also should be clear about the study’s limitations. For example, studies that measure illicit trade often do not measure product counterfeiting and do not include non- cigarette tobacco products in their estimates (42). When statistical estimates are included, they should provide confidence intervals as well as point estimates to account for uncertainty resulting from simple random chance (11). Characteristics of flawed analyses Flawed analyses can convolute and distort scientific knowledge about illicit trade. Flaws usually manifest in the data, methodology, statistical analysis and/or interpreta- tion of the results (11). Studies may be purposefully designed with methodological flaws to yield high or low estimates of the trade. For example, research showing that illicit trade constitutes a large share of the total market may be used to support arguments that taxes cause sharp increases in illicit trade, whereas lower estimates may be used to support arguments that certain governmental interventions (e.g. increased retail inspections) are effective. Pressures to skew data may also be tied to funding. For example, high estimates can sway governments to provide more resources for law enforcement activities. Analyses can be purposefully skewed by using data sources or data collection methods that will provide biased estimates. Flawed studies sometimes provide incomplete or inaccurate descriptions of their methodology. They may lack detail regarding the quality of the data used or information about how the data were collected and analysed. For example, a common weakness in industry-funded research on discarded packs is that the methods of collection and forensic analysis are not reported, ostensibly because they are “proprietary” information (43). However, these methodological details are key to assessing whether a study’s findings are biased by sampling error, model misspecification, measurement error, non-response or other flaws. It may be impossible to assess measurement error if researchers fail to disclose questions included in a survey instrument. Survey items used to measure the illicit tobacco trade may be imprecise. For example, asking respondents the frequency with 196 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N which they purchase “cheap” cigarettes may yield biased estimates, particularly if consumers can purchase cigarettes at discounted prices by using coupons. To more ac- curately measure tax evasion, surveys must include questions about the location of last purchase, purchase price, presence of public health warning labels and brand names. There are other ways that flawed studies can inadvertently or purposefully distort estimates of illicit trade. For example, data collectors can intentionally oversample areas known to be hot spots of illicit sales or sites that residents from lower-tax jurisdictions visit. Researchers can collect discarded cigarette packs close to the borders of countries with lower taxes to (inadvertently or purposefully) demonstrate the undesired side effects of tax policies. Studies published in non-peer-reviewed or lightly peer-reviewed outlets such as edited book volumes or policy briefs should be viewed with more scepticism than those published in highly regarded peer- reviewed outlets. 4.1.6 CONCLUSIONS Globally, the illicit tobacco trade continues to be a major concern for tax admin- istrators because of the challenges it generates to collecting higher revenues as well as the challenges to accurate and independent measurement. Industry figures provide distorted conclusions regarding the extent of the problem – frequently with a monocausal explanation of the link between illicit trade and tobacco taxation. Illicit trade comprises multisystemic issues and requires multiple strategies. Worldwide, countries at different levels of economic development have implemented a variety of effective measures to combat the illicit trade in tobacco products. The Philippines and the United Kingdom, for example, have addressed illicit trade as part of their overall tobacco tax reform (4). Price (and tax) levels are not a key determinant of illicit trade, the presence of which is exacerbated by the lack of tax administration capacity. Refraining from increasing taxes is not the solution; countries should instead respond with a com- prehensive strategy that includes at least these three main components: 1. It should identify – independently from the industry – the nature and dimen- sions of the problem. It is necessary to assess scientifically and with the best statistical practices the size of the illicit trade to understand the characteristics and scope of the problem. 2. It should identify and implement appropriate policies and strategies targeted at addressing the specific type of illicit trade the country is experiencing. It should address directly the country-specific institutional and/or governance challenges – as well as the lack of multilateral coordination that can exacer- bate illicit trade – and improve tax and customs administration practices as described in Chapter 3. CHAP T ER 4. PO LI T I C AL ECO N OMY 197 3. It should implement best practices contained in the WHO FCTC Protocol to Eliminate the Illicit Trade in Tobacco Products and accede to the Protocol if the country is not yet a Party. There are proper methods and policies with which to address the illicit tobacco trade. If countries start implementing the appropriate policies, they can raise tobacco taxes and reap health and revenue benefits even in the presence of illicit trade. 198 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.2 SCARE TACTIC C: COURT AND LEGAL CHALLENGES 4.2.1 INTRODUCTION The tobacco industry views well-designed and significant tax increases as a threat to the profit, growth and long-term sustainability of its business. As noted by PMI in 1985: Of all the concerns there is one – taxation – that alarms us the most. While [other restrictions] … do depress volume, in our experience taxation depresses it much more severely (44). The industry is, however, less likely to launch direct legal challenges to excise taxes than to other tobacco control measures (see Box 4.2.1 for details), because taxation – and excise tax in particular – is a comparatively well-established regulatory measure; in many jurisdictions, taxes have been levied on tobacco products for more than a century. There is also less unanimity in opposition to taxes among tobacco industry actors, because differences in the market position of different tobacco companies affect their interests in tax policy. This, in turn, decreases the likelihood that they will act collectively on the issue (45). BAT’s stated strategy in the early 1990s was to influence governments with regard to the level and structure of tobacco taxation in order to promote market growth and to secure competitive advantage (46). Nevertheless, tobacco industry actors will still legally challenge, or at least legally threaten, significant tax measures when vulnerabilities in their design, adoption or implementation are apparent. Box 4.2.1 Court and legal challenges to tobacco tax measures Evidence suggests that the tobacco industry and its allies instigate fewer legal actions against tax measures than against other tobacco control measures: 1. The Campaign for Tobacco-Free Kids’ tobacco control laws database contains only a handful of cases concerning tobacco tax measures, but hundreds on other tobacco control topics. This pattern can also be seen in a 2018 review of tobacco control legal challenges that examined this and two other databases to select 96 cases relevant to the question of the WHO FCTC’s usefulness in litigation (47). Only 6 of these 96 cases were challenges related to tax measures. CHAP T ER 4. PO LI T I C AL ECO N OMY 199 2. A 2013 systematic review of empirical studies on tobacco industry interference with tobacco tax policy found that only 9 of 36 relevant articles reported the specific use of litigation as a tobacco industry tactic (1). All 9 concerned constitutional challenges to earmarking provisions for tobacco tax initiatives in the United States (1). 3. A 2015 study on industry interference in LMICs cited legal challenges to tobacco control measures in 15 countries as examples of industry interference, but none of the challenges concerned a tobacco tax measure (48). 4. A 2016 analysis of papers published in systematic reviews of industry inter- ference with tax and marketing measures found that only 5 of 65 papers concerning tobacco tax related to the use of litigation or threats of litigation to interfere with tobacco tax measures (49). The tobacco industry makes extensive use of legal experts (1, 50–52) who study all relevant laws and regulations closely to determine their likely and arguable boundar- ies for the purpose of manipulating regulations and regulators (1, 50–52). Based on this expert advice, tobacco companies know when regulations remain within the bounds of both international and domestic obligations but can still argue that legally permissible tobacco control measures would be defeated in litigation if passed (48, 51–52). As the threat of a legal challenge alone can be used to the industry’s advantage, recourse to litigation is seldom needed or desirable (1, 45, 48, 51, 53–56). Even when litigation is launched, the objective may be to delay or weaken a measure rather than to win on the merits of the case (1, 45, 48, 53). To counter actual and threatened legal challenges, policy-makers need to be aware of relevant legal obliga- tions when preparing and implementing tobacco control measures. Fortunately, the tobacco industry playbook is relatively predictable. Tax and other tobacco control measures can thus be designed to strengthen the regulators’ legal position against genuine threats and enable them to dismiss baseless industry threats. 4.2.2 COUNTRY EXPERIENCES WITH LEGAL CHALLENGES TO TOBACCO TAXATION Legal obligations that are relevant to tobacco taxation include those under do- mestic law and international instruments such as international trade agreements and international investment agreements (IIAs).17 Some of the legal issues that a tax measure may encounter are outlined in Table 4.6. Case studies from various countries illustrate how these legal issues have and have not been avoided in the 17 Relevant international trade agreements include the WTO Agreement and custom unions such as the EU, the East African Customs Union and Mercosur. Relevant IIAs include bilateral investment treaties and the investment chapters in free trade agreements and within custom unions. 200 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N passage, design and implementation of tobacco taxes. These issues are not the norm, however, and should not give rise to undue apprehension. The case studies are rated as positive, mixed or negative based on the extent to which the legal decision upheld the taxation measure in question. Table 4.6 Potential legal issues for tobacco tax measures VULNERABILITIES LEGAL OBLIGATIONS CASE STUDIES Inadequate consultation and other procedural vulnerabilities Domestic procedural law 1, 2 Due process protections for investors under IIAs None Procedural requirements under WTO Agreements and Custom Unions 3 Discrimination against imports or investors Nondiscrimination obligations under WTO Agreements and Customs Unions 8, 9, 10 Nondiscrimination obligations under IIAs 11 Investment incentives or inducements Arbitration mechanisms under investor-state contracts 12 Fair and equitable treatment clauses of IIAs None Other substantive breaches Constitutional rights and restrictions on taxation 4 Statutory restrictions on the imposition of taxation 6 Expropriation clauses of IIAs 5 Ultra vires (the scope of legal authority) 7 Avoiding procedural vulnerabilities in tax laws Procedural defects can be avoided by taking great care in progressing and imple- menting regulatory or legislative provisions. Procedural concerns pose a dilemma for tobacco control regulators. Article 5.3 of the WHO FCTC and the COP guide- lines for its implementation state that policy-makers and regulators should interact with the tobacco industry only when and to the extent strictly necessary (57). For taxation measures, interaction might be necessary because consultative and de- liberative processes could be prescribed under domestic constitutional provisions and procedures for good governance, due process requirements of IIAs and some international trade agreements. The tobacco industry may use these requirements as leverage to delay, distort or hijack the rule-making process in contravention of Article 5.3. Accordingly, interactions with the tobacco industry should be limited to strictly necessary consultation conducted in a transparent or public manner but with care that this does not come at the expense of a measure’s defensibility. The proper balance will depend on the jurisdiction in question, since constitutional, statutory and applicable international legal obligations vary. CHAP T ER 4. PO LI T I C AL ECO N OMY 201 CASE STUDY 1 (MIXED): Industry manipulation of legislative procedures In 2012, a bill stipulating, among other things, the creation of a new specific excise tax on cigarettes passed its final reading in Costa Rica’s Legislative Assembly. Passage of the bill had, however, proceeded under “urgency” and notwithstanding a pending constitutional enquiry (a constitutional query is meant to prevent passage of a bill).18 ISSUE MAJORITY DECISION MINORITY DECISION LESSON Whether the court could consider the enquiry despite passage of the bill and the effect the bill’s passage could have despite the enquiry. The enquiry was taken up by the Supreme Court’s Constitutional Division’s majority (58). The signing and publication of the bill by the executive was suspended by the Constitutional Division pending their decision on the merits of the case – which, in the end, found any question of the bill’s constitutionality baseless (58). The enquiry was inadmissible by reason of having been filed too late and notice of its filing having not been received by the legislature prior to the reading of the bill (58). In disagreement with the majority, the minority held that the court could not consider the enquiry or suspend the bill’s signing by the executive – the final step in becoming law. This challenge demonstrates how the tobacco industry’s defenders may attempt to frustrate and impede a tax measure’s passage. In this case, the challenge seemed to have been a delaying tactic, as it was posted on the same day as the final reading of the bill. Its authors may have either wanted its pending nature to cause the legislature to delay or, as occurred, to create conditions for a procedural and constitutional challenge in the absence of delay. All the grounds of the challenge itself were found to be without merit. Although such frivolous challenges cannot be prevented, they can and should be anticipated to ensure that they do not lead to a tax measure’s defeat. 18 “Urgency” is a procedure under which a bill is progressed through a legislature in an expedited fashion. 202 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N CASE STUDY 2 (POSITIVE): Adhering to domestic procedural requirements Kenya’s tobacco control regulations required the tobacco industry to pay a levy to compensate the state for health care and other negative externalities of smoking. In a 2016 challenge brought against these regulations, the plaintiff, BAT, was unsuccessful on every count (59–60). Even though the levy was not considered a tax measure by the court, the case study is instructive on how regulators may safeguard tax measures against procedural challenges. ISSUES LAWS AND ARGUMENTS DECISIONS LESSONS Whether the government’s consultations on the measure were adequate. Asserting that the Constitution and the Statutory Instruments Act together meant that “appropriate consultations with persons who are likely to be affected” were required because of the measure’s likely substantial effect on business. BAT claimed that this standard was not met. Kenya’s government claimed that it was under no obligation to undertake special or extensive consultation with the tobacco industry. The judge found in favour of Kenya’s government, noting that (1) the requirement to consult does not imply that any particular view needs to prevail; (2) dissatisfaction with the level of consultation is not decisive; (3) on the facts, industry was allowed, and often invited, to send representatives to all relevant public consultative meetings and parliamentary committee hearings; and (4) consultation on the regulations was adequate (59). The tobacco industry carefully scrutinizes legislative and regulatory processes for defects. In this case, Kenyan government officials appropriately distanced themselves from the tobacco industry by not permitting its representatives special consideration but did permit their attendance at public meetings and the ability to submit their views under usual procedures. In this way, both the principles behind WHO FCTC Article 5.3 and the requirement for consultation under Kenyan law were observed. CHAP T ER 4. PO LI T I C AL ECO N OMY 203 Avoiding procedural issues in tax administration CASE STUDY 3 (NEGATIVE): Contravening procedural requirements in international obligations In 2010, a WTO panel held that Thailand violated the Customs Valuation Agreement (CVA) by the process it used to value cigarettes that Phillip Morris (PM) Thailand imported into the country from a related party, PM Philippines. Customs values are important as they are the tax base for tariffs and can feed into the base for other taxes levied against the value of the good, such as ad valorem excise taxes and VAT. Transaction values declared by PM Thailand were rejected by Thai tax authorities as influenced by the relationship between the parties and a customs value determined by deduction was substituted (61). ISSUE LEGAL OBLIGATION DECISION LESSONS Whether Thailand adequately consulted with PM Philippines before rejecting its declared transaction value (61). The CVA requires good faith exchange of reasons and information, with opportunities for response (61). Thailand had failed to properly explain its reasons for rejecting the transaction value, as well as its belief that price was influenced by the relationship between the two parties (61). This was a violation of the CVA. Thailand did not appeal these findings. Thailand’s authorities needed to take greater care in their dealings with the tobacco industry to ensure they met the pertinent procedural obligations. In this instance, a specific and high standard of consultation – the provision of detailed reasons and an opportunity for response – was prescribed by the CVA and Thailand failed to meet it. 204 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ensuring compliance with substantive requirements Rules found in domestic and international law also establish substantive obliga- tions. This subsection focuses on substantive obligations found in domestic law. International obligations concerning discrimination and investment incentives are considered in the next subsections. CASE STUDY 4 (POSITIVE): Tax measure found to be consistent with the Constitution The Chilean government introduced a substantial increase in tobacco and fuel excise, and in 1995, a coalition of taxpayers brought a Constitutional challenge to the measure (62). ISSUE DECISION LESSON Whether the tax was, per article 19 of the Chilean Constitution, “obviously disproportionate or unjust” (62). The excise tax increase did not violate the Constitution, as it was neither confiscatory nor manifestly irrational. Generally applicable excise taxes are not vulnerable to challenges for being excessive, unfair or disproportionate. CASE STUDY 5 (POSITIVE): Failure to grant tax rebates not an expropriation under an IIA This case study is an example of a claim for breach of an expropriation clause in an investment treaty. Such clauses protect foreign investors against measures that can be construed as directly or indirectly seizing an investment or depriving it of its value (63). In the case, an investor was, for more than a decade, denied tax rebates by the Mexican government. This affected the profitability of the business of purchasing and reselling Mexican cigarettes abroad, and the investor brought the claim to an investment agreement arbitral tribunal in 2002 (64). ISSUE DECISION LESSONS Whether Mexico’s failure to grant rebates to the investor exceeded the bounds of valid regulation to constitute indirect expropriation of the investor’s investment (65). There was no expropriation. The arbitral tribunal noted that not all business problems are violations: the investor had no right to participate in the “grey market” export of cigarettes and there were sound reasons to restrict that market (65). Further, the investor was able to participate in other business ventures and actually continued to have business success (65). Claims of indirect expropriation made under IIAs are unlikely to be successful, as generally applicable tax measures are a legitimate form of regulation. A mere loss of profit will not suffice. Claims of expropriation will not succeed unless a substantial or significant deprivation of the investment results. CHAP T ER 4. PO LI T I C AL ECO N OMY 205 CASE STUDY 6 (NEGATIVE): A regulation contrary to superior domestic legislation In 2011, an Indonesian tobacco industry association group, FORMASI, challenged a new excise regulation. Since 2009, the government had been implementing a tiered specific excise tax system based on a set of characteristics (size of production, type of cigarettes and price levels). In 2011, excise rates were increased in nearly all of the 19 tiers, but the reference prices were not accordingly adjusted. This gave rise to a legal issue. ISSUE DECISION LESSONS Whether new excise regulations breached a 57% ceiling for the rate of excise on the retail sale price of tobacco products under the superior Excise Law (66–70). The challenge specified that excise exceeded this ceiling for hand-rolled domestic clove cigarettes (kreteks) (68, 71). The Court found in favour of the tobacco industry association, and the government was required to immediately revoke the 2011 regulation. (69–70). It is advisable to stay within the rules and be aware of legal hierarchies – including superior domestic legislation. The tobacco industry scrutinizes all increases in tobacco taxes. In this case, a breach of a legislative requirement for a single category of tobacco product resulted in Indonesia suffering lost revenue and a setback in its efforts to reduce tobacco consumption. Ensuring a tax measure is within an authority’s legal power A tax measure is ultra vires when it goes beyond the legal power of the enacting body. As with case study 6, this is a legal issue that involves legal hierarchies. In ultra vires cases, however, instead of centring on conflict between inferior and superior law, the issue is whether an authority that enacts a tax measure is authorized to do so. This issue may arise when a tax measure is enacted by a subnational jurisdiction or by an executive acting under a statutory delegation. CASE STUDY 7 (NEGATIVE): Tobacco taxation contrary to the Australian Constitution ISSUE DECISION LESSONS Whether New South Wales’ licensing and penalty fees regime constituted an excise tax by other means contrary to the Australian Constitution’s exclusive grant of that power to the federal government (72). The court found that state licensing fees were excise taxes and that this was contrary to the Australian Constitution (72). Authorities enacting tobacco tax measures must act within the scope of their legal power. 206 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Avoiding discrimination against imports and foreign investments Although inherently discriminatory, customs duties may be used subject to the agreed upper limits in a country’s trade agreements. Excise and other taxes designed with the aim of raising tobacco prices to reduce demand and advance human health should be origin-neutral: they should not seek to tax local products less than foreign products or aim to treat foreign products differently from one another. Tobacco tax measures are pursuing objectives other than health when they aim to raise the price of imports more than that of local products or seek to burden favoured market participants less than others. Solely health-protective tobacco taxes will not ordinarily violate Articles III:2 and I:1 of the GATT (the WTO’s General Agreement on Tariffs and Trade), which prohibit discriminatory taxation (in light of general exceptions). Nor will solely health-protective tobacco taxes directly violate anti-discrimination protections for investors found in the national-treatment (NT), most-favoured-nation (MFN), expropriation and fair-and-equitable-treatment (FET) clauses of IIAs (63, 73–74). It is possible to make claims for breach of international obligations on grounds other than discrimination, but such claims are generally highly unlikely to succeed. CASE STUDY 8 (NEGATIVE): BAT v Uganda (2017 East African Court of Justice) DISCRIMINATION LESSON Uganda established a higher level of excise taxes on imported cigarettes – including those from Partner states of the East African Customs Union (75) – than on local cigarettes. Its implementation was discrimination contrary to Article 15 of the Customs Union Protocol (75). Differential taxation explicitly based on origin can be construed as protectionist discrimination in violation of international obligations. The tobacco industry can also turn to international trade agreements outside of the WTO – in particular, customs union mechanisms. Difficulties arise when ostensibly origin-neutral and health-protective tobacco taxes result in dissimilar taxation of tobacco products (73). Discrimination does not exist simply because there is dissimilar taxation – the taxation must adversely impact imported goods more than local products, the imports of one nation more than another or a particular investor’s products more than comparable products. Where dissimilar taxation between product categories results in discrimination, the tax will ordinarily still be lawful if the dissimilar taxation is based solely on a legitimate regulatory distinction between the product categories in question.19 19 The precise applicable rules vary depending on the nature of the legal obligations in question. Under the GATT, dissimilar taxation of like or directly competitive products can be justified based on scientifically grounded distinctions between products under Article III:2 and, in the alternative, discrimination that is necessary under the explicit carve-out for health-protective measures, Article XX(b) (73). For the MFN and NT clauses of IIAs, differential taxation can be argued as nondiscriminatory on the basis that difference in harm means the products are not “alike” or, in the alternative, discrimination is justified based on scientific evidence of differences in harm and rational reasons for the health-protective role of differential taxation (63). CHAP T ER 4. PO LI T I C AL ECO N OMY 207 Where discrimination is inadvertent, lack of an intention to discriminate is not sufficient as a defence for breach of obligations under IIAs or the GATT (73, 76). Policy-makers should carefully scrutinize measures to determine: 1. whether an aspect of a tax measure’s design or implementation may be more to the detriment of imports or foreign investors than of local products or domestic investors; 2. whether the potentially discriminatory aspect of the tax measure serves any useful purpose in supporting the tax measure (i.e. it is needed to achieve the health goal); 3. whether there is any reasonable alternative that could achieve the same effect without the potential for discrimination; (i.e. it is indispensable) and 4. when it is needed and indispensable there is a good chance that it will be defensible. The case studies below provide examples of discrimination arising in connection with a tobacco tax measure. CASE STUDY 9 (NEGATIVE): Thailand – Customs and Fiscal Measures on Cigarettes from the Philippines (2010 WTO panel) The facts of this case are presented in case study 3. This case study examines claims of discrimination rather than the procedural issues. DISCRIMINATION EXPLANATION LESSONS Thailand implemented its policy for determining the tax base for VAT on cigarettes inconsistently (61). Thailand applied a methodology in fixing the tax base, in particular a marketing cost component, of imported cigarettes that differed from that for local products (61). This resulted in the marketing cost component for the imported cigarettes being higher than it would have been under the general methodology. This difference in treatment was insufficiently justified and therefore considered discriminatory. As there is potential for inadvertent discrimination when the base for an ad valorem tax is fixed, tax base determinations must be consistent and well- reasoned (61). This case study demonstrates how policy- makers need to take care in designing and implementing ad valorem taxes to ensure they are nondiscriminatory and legally defensible. 208 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Thailand’s VAT rebate policy imposed a potentially higher tax burden and also created more burdensome administrative requirements for imported cigarettes (61). Resellers of cigarettes produced by a government entity were granted an exemption from VAT (61). Although resellers of imported cigarettes would be eligible for a tax credit on their VAT, this was not an automatic process (61). The distinct treatment of resellers of imported cigarettes and those of local cigarettes resulted in the risk that there would be a higher VAT burden for the former (61). The distinct treatment also imposed an additional administrative burden on resellers of imported cigarettes and altered conditions of competition (61). Rules for the collection and enforcement of tax obligations should be the same, or as similar as practicable, in both form and effect for domestic and imported tobacco products. CASE STUDY 10 (NEGATIVE): Dominican Republic – Measures Affecting the Importation and Internal Sale of Cigarettes (2004 WTO panel; 2005 WTO Appellate Body) Under article XX(d) of the GATT, discrimination that is necessary to secure com- pliance with a legitimate tax measure will be justified provided there is no less- discriminatory alternative. In this case, this justification was used unsuccessfully. DISCRIMINATION EXPLANATION LESSONS The Dominican Republic’s tax stamp regulations were discriminatory towards imported goods (77). Under the regulations, all cigarette packs had to be affixed with tax stamps, but imported cigarettes were to be affixed with tax stamps under the supervision of local tax authorities following importation, while locally manufactured cigarettes could be affixed with a tax stamp in the course of production. This de facto distinction between local and imported products modified the conditions of competition to the detriment of imported cigarettes by (1) increasing costs for importers and (2) impairing the aesthetics of imported products (77). The panel did not consider this discrimination justified: it was not necessary for the enforcement of tax measures, because less restrictive alternatives were available such as permitting importers to affix tax stamps during the course of production (Dominican Republic – measures affecting) (77). The panel’s findings were upheld on appeal (78). Policies crafted to ensure compliance with tax measures need to also be nondiscriminatory. Discrimination claims can arise when compliance costs are higher for imports than for local products and this de facto distinction is avoidable. It is important to consider whether less burdensome alternatives may achieve the same objective. CHAP T ER 4. PO LI T I C AL ECO N OMY 209 CASE STUDY 11 (NEGATIVE): Feldman Karpa v Mexico (2002 ICSID [International Centre for Settlement of Investment Disputes] Arbitral Tribunal) Arbitral tribunals have accepted differences in treatment accorded to investors protected by IIAs when there is a legitimate connection between the distinctions drawn and public welfare objectives (76). The facts of this case are presented in case study 5. This case study examines aspects of the case involving the investor’s claim of discrimination, rather than the substantive issue of expropriation. Claims of discrimination are made on different grounds than claims for expropriation, which is why the case was decided differently on this claim. DISCRIMINATION LESSONS Denial of foreign investors’ claims for tax rebates. Tax rebate claims were granted to similar local investors (65), which was a violation of an IIA’s national treatment clause (65). Foreign and local investors must be treated similarly, and consistent and well-documented policies must be used to guide administrative decisions. The denial of the rebates may have been justified, but the government was unable to establish this due to a lack of documentation. Avoiding the investment incentives trap Investor-state contracts between the tobacco industry and governments should be avoided. They are not merely “contractual” in the domestic law sense, as even in the absence of an applicable IIA, they can be internationalized to provide inves- tors the right to (1) remove dispute settlement from the state’s court in favour of independent arbitration and (2) remove the dispute from the state’s legal framework in favour of general principles of law (63, 76). Commitments under these clauses cannot, therefore, be legislatively moderated or extinguished, nor can liability be limited within domestic courts that may be more likely to favour the state’s right to regulate in favour of public health (76). Investor-state contracts and other noncontractual inducements can be further internationalized by umbrella clauses within IIAs. Such clauses make reneging on undertakings assumed towards investors a breach of the IIA (76). Moreover, even in the absence of an umbrella clause, contracts and inducement can underpin a claim for legitimate expectation and breach of fair and equitable treatment and can also strengthen an investor’s claim for indirect expropriation (63). Arbitral awards make clear that although taxes can be expected to vary and tobacco will be regulated, investors can have the legitimate expectation that states will abide by formal inducements and written contractual undertakings. A common clause within investor-state contracts, the stabilization clause, is ruinous to evidence-based tobacco control’s most effective measure: excise tax in- creases. Stabilization clauses purport to freeze specific domestic law from the time 210 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of investment (63). Seemingly less onerous, economic equilibrium clauses require contracting states to compensate for regulatory changes that negatively affect an investment’s value (63). There is little difference in effect between these two types of clauses: liability for the cost of breaching an equilibrium economic clause can be onerous enough to make it fiscally challenging and politically unpalatable. CASE STUDY 12: An investor-state contract A state entered into an investment agreement with a TTC in 2001 on the privatiza- tion of its state-owned tobacco enterprise and creation of a joint venture. This investment was to provide economic benefits under the agreement: the joint venture would increase exports and profit using the TTC’s cash and expertise while also ensuring prioritization of local employment, manufacturing and resources. The final investor-state contract included a form of economic equilibrium clause under which any increase in the excise tax rates applied to the company’s tobacco products before a set date would be compensable. While the agreement was not removed from the state’s law, it provided for independent arbitration in case of a dispute over its compensation. In addition, there is a bilateral investment treaty between the host state and another state in which the TTC’s subsidiary has residence that includes a FET clause – this could buttress, if needed, the protection provided by the stand-alone arrangements of the investor-state contract. There were similar less-formal inducements offered to a separate TTC. The extent to which incentives have been granted to the tobacco industry is unknown, but contracts and inducements are likely to be offered in the context of the privatization of state-owned tobacco interests and in dealings between investors and state-owned tobacco enterprises (63). Although countries have been entrapped by their incentives to industry, the investor-state contract provides the clearest example of how undertakings and inducements with the tobacco industry under- mine tobacco control (56, 79–80). States should avoid offering industry incentives and, in particular, entering into contractual undertakings with the industry. More systematically, government should consider avoiding IIAs that elevate incentives and inducements above sensible and reasonable regulation. 4.2.3 CONCLUSIONS Health-protective and origin-neutral tobacco excise taxes are legally defensible, and industry threats are usually baseless. There are, however, certain rules governing procedure, design and consultation that governments may need to consider: 1. Governments should be aware of the standard of consultation required under do- mestic law and any applicable international obligations (case studies 1, 2 and 3). CHAP T ER 4. PO LI T I C AL ECO N OMY 211 It is important to distance the tobacco industry from the policy-making process to the extent that this is permissible. Do not grant the industry special consideration, but do ensure that it is consulted with as required – for example, by providing public meetings, timely information and the ability to submit industry views – while being aware of potential procedural manipulation (case studies 1, 2 and 3). 2. Excise tax is generally safe from challenges that claim it is confiscation or expropriation under domestic or international law (case studies 4 and 5). But express limits on taxation can be found in other laws or a country’s constitution or in the limits of the power to tax granted to an authority (case studies 6 and 7). 3. Explicit and de facto discrimination against foreign tobacco products or investors must be avoided in the design, implementation or enforcement of tax measures (case studies 8, 9, 10 and 11). Legal issues may arise not from the tax measure itself, but rather from ancillary measures that support its implementation (case studies 9 and 10). 4. Explicit differentiation between products based on their effect on health may be challenged as discrimination if it falls heaviest on imported products and has to be justified on the basis of evidence of impact on health and a lack of alternatives. 5. Investment incentives in the form of inducements or contractual undertak- ings should not be offered, as these may be binding (case study 12) or may ground a challenge under an IIA; they are also contrary to the WHO FCTC Article 5.3 Guidelines. 212 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.3 SCARE TACTIC A: ANTI-POOR RHETORIC (REGRESSIVITY) 4.3.1 INTRODUCTION In their efforts to lobby against tax increases, the tobacco industry and its affiliates often claim that increases in tobacco taxation will hurt the poor (81–82). This argu- ment is based on the concept of regressivity in relation to taxation. Conceptually, a tax can be regressive if it means that lower-income people pay a greater proportion of their household income to meet the tax burden than do wealthy people. In other words, the tax burden tends to be relatively higher for lower-income households than for middle- and high-income households. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use. Second, higher tobacco taxes and prices can induce behaviour change among the population, as reflected in the price elasticity of demand (83–84). In combination, these broader considerations effectively make tobacco taxation a progressive – rather than regressive – public health intervention. 4.3.2 REGRESSIVITY AND THE BROADER PERSPECTIVE In a narrow sense, tobacco taxation can be seen as regressive because lower-income people must allocate a relatively greater proportion of their household income than wealthy people to pay for tobacco products when those products become more expensive following a tax increase. In many countries, people from lower-income groups use tobacco more than other people (85). A systematic literature review by WHO found a robust association between lower income and a higher prevalence of current smoking among adults, both men and women (86). This finding was consistent across three decades of studies, across most geographic regions and across countries of different income classifications. For example, in India, high rates of tobacco use – i.e. use by more than 30% of the adult population – are found only in lower-income states such as Assam and Odisha, where net state domestic product is still below 100 000 rupees per capita (see Fig. 4.3.1) (87). CHAP T ER 4. PO LI T I C AL ECO N OMY 213 Fig. 4.3.1 Relationship between adult tobacco use and net state domestic product per capita in states and union territories of India, 2016–2017 Source: (87). However, this finding does not account for broader health and economic factors that determine the full impact on households. Tobacco taxation can in fact be viewed as a progressive – or pro-poor – policy when these wider considerations are properly ac- counted for and explained. In terms of health concerns, the relatively high use of tobac- co among low-income populations translates into a much greater burden of tobacco- attributable diseases for these populations, including higher morbidity and mortality. Low-income groups are also less able to afford medical care to treat tobacco- attributable diseases, and large out-of-pocket medical expenditures can further impoverish many families. Consequently, many poor individuals do not get or even seek the medical care they need. One study found that in Bangladesh, 55% of patients diagnosed with a tobacco-attributable illness did not seek further medical care. This lack of health care utilization was attributed in part to prohibitively high out-of-pocket treatment costs (88). The combination of high rates of tobacco use and lack of access to affordable medical care means that tobacco use measurably contributes to the poverty rate in a number of high-tobacco-burden countries, including China and India (89–90). A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India 214 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The poor are also known to be more price-sensitive than the wealthy; lower-income smokers exhibit higher price elasticities than their higher-income counterparts. This is demonstrated in recent studies by the World Bank, findings of which are shown in Fig. 4.3.2 (91). The poor respond more strongly to higher tobacco taxes and prices by reducing their use of tobacco products more than others, and thus they benefit disproportionately in terms of avoiding tobacco-related deaths, diseases and associated medical costs. A similar conclusion was drawn in a systematic review of the population impact of tobacco control policies on socioeconomic inequities in high-income countries at the late stage of the tobacco epidemic (92). The review found 16 relevant studies relating to taxation, only one of which found a regressive association between tax and the social economic gradient (seven found a progressive impact, while the others produced mixed results). Fig. 4.3.2 Price elasticity of tobacco consumption, medium estimate, by decile Source: (91). This wider economic perspective is explained in the World Bank’s Extended Cost- Benefit Analysis (ECBA) framework, which assesses the distributional impact of tobacco tax increases on health, among other factors (82, 83). That is, the ECBA framework looks beyond the simple or partial definition of regressivity (i.e. impact El as tic it y Deciles -1.2 1 2 3 4 5 6 7 8 9 10 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina CHAP T ER 4. PO LI T I C AL ECO N OMY 215 on household expenditure by income levels) to capture the full distribution of benefits, including improved health and income. The ECBA framework has been applied in studies of various countries, including Bangladesh, Bosnia and Herzegovina, Chile, Indonesia, Republic of Moldova, South Africa, the Russian Federation, Ukraine and Viet Nam. The evidence from these studies supports the view that effective tobacco tax policies can generate pro-poor and welfare-improving outcomes. When reductions in medical expenditures and additional years of working life that result from lower smoking-related mortality are taken into account, the overall policy of tobacco tax increases becomes progressive rather than regressive (see Fig. 4.3.3) (84). A similar conclusion has been reached in studies of high-income countries, such as the United States, where a tobacco tax increase was enacted in 2009 (93). Fig. 4.3.3 Impact of a 100% price increase, with medium elasticities, by deciles Source: (91). Tobacco tax increases will also often lead wealthier smokers to contribute relatively more than poorer smokers to the overall amount of tax revenue collected. This is because poorer smokers reduce their consumption the most, since they are more price- sensitive and wealthier smokers also tend to purchase premium (higher-priced and In co m e ga in s (% ) Deciles -1 2 3 54 6 7 8 9 10 0 1 2 3 4 5 1 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina 216 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N taxed) tobacco products (94). Hence, higher tobacco taxes can be seen as progressive in terms of additional revenue collection and health gains resulting from cessation, as well as from preventing the young from taking up smoking in the first place. One study from China suggests that a 50% tax increase would raise US$ 703 billion over 50 years, with just 14% of this increase being borne by smokers in the lowest income quintile (95). In addition, the tax increase would yield a savings of US$ 24 billion in expenditures on tobacco-related diseases, with about 28% of these savings being enjoyed by smokers in the lowest income quintile. The ECBA framework actually presents a rather conservative profile of the net benefits of raising tobacco taxes, since it does not include other sources of gain, such as reduced harm from exposure to second-hand smoke, increased productivity and the potential for poor households to benefit from social programmes funded through increased tax revenues (96). Assessments of the distributive impact of the 2009 tobacco tax increase in the United States found that the overall progressivity of the increase was enhanced by the tax rev- enue being used to expand health insurance coverage for children of low- and middle- income families (97). Accounting for this expanded coverage added to the progressiv- ity of the overall legislative package, the bottom line being that the impacts are positive for lower-income quintiles and greatest, on average, for low-income households (93). Similarly, a large proportion of the tobacco tax revenues from the Philippines’ so-called Sin Tax Reform was used to subsidize universal health coverage (UHC) for poor and near-poor families. Globally, 37 countries are known to earmark some tobacco tax revenues for health programs, with many of these programs indirectly benefiting the poor and less-advantaged disproportionately more than other groups (27) (for details on earmarking, see section 4.6). 4.3.3 CONCLUSIONS Contrary to the perception of tobacco taxation being regressive, it is a strong pro- poor policy when the broader economic impacts are taken into consideration. The tax burden is not a complete indicator of regressivity, since it does not include the negative health and economic impacts of tobacco-attributable diseases or the positive impacts of behaviour change in response to tax and price increases. The health and economic burdens of tobacco-attributable diseases fall dispropor- tionately on the poor, who tend to have higher tobacco use and are also the least able to afford the necessary medical care. Because the poor tend to be more price-sensitive, they curtail their use and consumption more significantly than wealthier smokers in response to tax increases, which in turn reduces their downstream health and economic costs. Tobacco taxation can be made even more progressive by earmarking or allocating tobacco tax revenues for social goods and services that benefit the poor (see section 4.6). CHAP T ER 4. PO LI T I C AL ECO N OMY 217 4.4 SCARE TACTIC R: REVENUE REDUCTION 4.4.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases result in reduced tax revenues for the government. According to them, the reduction in revenues is caused either by substitution to cheaper, lower-taxed or smuggled tobacco products or by reductions in consumption overall (98–99). The tobacco industry often refers to the Laffer curve to make this argument. According to this curve, revenues increase along with tax rates up to a certain point, after which further increasing tax rates leads to declining revenues. When considering tobacco taxes, the tobacco industry assumes that countries are already approaching or are even beyond the critical tax rate level (98). However, the argument rests on a narrow theoretical and empirically unsubstanti- ated foundation (98–100). The price inelastic demand for tobacco and the relatively low tax share in prices in many countries explain the win-win for public health and finance, i.e. that declines in consumption and increases in revenues can occur simultaneously (98, 101). Furthermore, many country examples (see case studies below) demonstrate that well-designed and well-implemented tobacco tax increases lead to increases in revenue, at least in the short to medium term (98, 100). Although consumption will diminish with a tobacco tax increase, the percentage increase in excise tax per unit is greater than the percentage decrease in tobacco consumption, cancelling out at least some of the effect of reduced consumption on revenue (98–99). A change in the tax rate, with all other factors influencing consumption kept constant, corresponds to a change in the tax revenue and is represented by a move- ment along the Laffer curve. As the tax rate changes, so does the elasticity of the tax base; each point on the Laffer curve corresponds to a different tax base elasticity. When one or more of the other factors changes, this affects the position of the curve, and the tax base elasticity changes at a given tax rate. For example, a successful smoke-free policy or advertising ban that reduces the demand for tobacco shifts the curve down, reducing the tax revenue potential for each tax rate. To demonstrate that few, if any, countries are beyond the revenue-maximizing point on the Laffer curve, Table 4.7 shows the revenue impact of increasing excise taxes under different scenarios, using different price elasticities of demand, different levels of tax increases and different starting tax shares, depending on country income levels. This is the tax base elasticity approach from which the Laffer curve is derived (for more details, see section 2.2.3 and Annex 2.2). The total and excise tax shares shown are weighted averages for each country income group, calculated from the RGTE dataset. The revenue gains were simulated using progressive levels of excise tax increases (25%, 50%, 75% and 100%) and varying price elasticities of demand (-0.4 to -1.2). 218 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.7 Percent increase in excise revenues under different scenarios of tax levels, tax increases and price elasticities20 Income group Total tax as % of retail price Excise tax as % of retail price Increase in excise tax Increase in excise revenue when price elasticity of demand is: -0.4 -0.6 -0.8 -1.0 -1.2 LOW INCOME 38% 22% 25% 22% 20% 19% 17% 16% 50% 43% 39% 36% 33% 29% 75% 63% 57% 52% 46% 41% 100% 82% 74% 66% 59% 51% MIDDLE INCOME 58% 41% 25% 19% 17% 14% 11% 9% 50% 37% 31% 26% 20% 15% 75% 54% 45% 36% 27% 19% 100% 71% 57% 45% 34% 23% HIGH INCOME 68% 55% 25% 18% 15% 11% 8% 5% 50% 35% 27% 21% 14% 8% 75% 50% 39% 29% 19% 10% 100% 65% 50% 36% 23% 11% Source: Authors’ calculations using data from the RGTE (27).21 Substantial revenue increases occurred in all the scenarios that were considered in the simulation. These results reaffirm much of what is already known, i.e. that higher tax increases generate higher revenue gains, and that these gains increase with the increasing inelasticity of demand. Even when demand is relatively price elastic (-1.2), the simulation predicts a gain in revenue. The tax share in price also affects revenue potential. The lower the tax share in price, the larger the revenue potential. This suggests that revenue reductions as a result of an excise tax increase will occur only if the scenario is extreme (i.e. a very elastic demand coupled with a very high current tax share). It is important to note that the vast empirical literature 20 These projections use 2018 data from 185 countries. The countries were classified according to World Bank income group, with the average total tax share, excise tax share and VAT/sales tax share for each country weighted according to the number of current adult cigarette smokers. To calculate the projected revenue for each stated elasticity, it was assumed that there would be full pass-through of the excise tax increase, along with constant percentages of non-excise taxes (VAT/sales tax) as a share of the retail price. The consequent changes in price were multiplied against the respective elasticities to derive the expected change in consumption. The projected revenues could be easily computed by multiplying the new consumption figures against the increased excise tax rates. 21 These calculations do not take into account brand substitution (cross-price elasticities), income ef- fects or illicit trade. The excise tax was assumed to be a specific tax, while the non-excise taxes (VAT and others) were bundled and treated as an ad valorem tax with retail price as the tax base. The difference between retail price minus all taxes was also assumed to be constant, with full pass-through of the tax increase to consumers. CHAP T ER 4. PO LI T I C AL ECO N OMY 219 shows tobacco to be universally inelastic; thus the extreme scenario should not be given credence by policy-makers. Furthermore, as demonstrated by the data in Table 4.7, tax shares in most countries are relatively low and reinforce the revenue potential of tobacco tax increases. The revenue potential of tobacco taxes is indeed quite significant. It is estimated that in 2018, excise taxes on cigarettes generated a total of US$ 361 billion in revenues worldwide, including US$ 162 billion in LMICs. If all countries were to raise excise rates by the equivalent of US$ 1 per pack of cigarettes, the amount of excise revenue would increase by between US$ 178 billion and US$ 219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from these tax increases, with excise revenues in these countries increasing by US$ 133 billion to US$ 167 billion, or by 82–103%.22 Revenue reduction in the countries examined was due to other causes, not the tax increase per se. For example, Tonga significantly increased its excise tax on cigarettes in 2016 and saw a very sharp decrease in its consumption (40% decrease), followed by a revenue decrease. This occurred because 20% of smokers switched to an untaxed, cheap local loose tobacco product called Tapaka Tonga (102). The lesson learned was that Tonga needed to tax all its tobacco products at the same level to avoid substitution to lower-price/untaxed tobacco products. Another example of revenue decrease that was not related to tax increases but rather to tax administration mismanagement is the case of South Africa (see explanation in the case study later in this section). Finally, declines in revenue due to long-term declining trends in tobacco use should not be confused with being beyond the revenue-maximizing point of the Laffer curve. For example, in the United Kingdom, where long-term declines in tobacco use are being experienced, a nominal decline in revenues occurred between 2017 and 2018 even though excise taxes remained unchanged. Conversely, even countries with very high tobacco excise rates experience increases in revenues as a result of tobacco tax increases (see the case study of Australia below) (98, 100). This suggests that few countries, if any, are beyond the revenue-maximizing point on the Laffer curve. Tobacco consumption is expected to be tax inelastic, even if demand becomes effectively price elastic as a result of successful tobacco control interventions. Taxation serves as an instrument for both fiscal and public health objectives. If after successful tobacco control interventions, prices reach levels where demand is elastic, the tax base is still most likely to be inelastic due to tax undershifting, since overshifting is not a good pricing policy when demand is elastic (for a more detailed discussion on the shifting of tax, see section 2.2.2). In other words, a tax rate increase in combination 22 Goodchild M, Perucic AM, Paul J. Tobacco taxation as a strategy to achieve global targets for smoking prevalence. Unpublished manuscript. October 2020. 220 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N with non-price tobacco control measures, which make consumers more sensitive to price (tax) increases, leads to declining but still positive marginal revenues. In the long run, tobacco control policies, including price and tax measures, may be so successful in reducing consumption that revenues will plateau or fall. This is ultimately the long-term policy goal. Reducing the impact of the policies and ending the global tobacco epidemic is the aim of tobacco control and not something to be avoided. However, ending the global tobacco epidemic is unfortunately not foreseen in the short to medium term; therefore, governments can currently rely upon tobacco taxes as a reliable source of revenue (103). 4.4.2 THE REVENUE IMPACT OF EXCISE TAX INCREASES: CASE STUDIES The following case studies illustrate four key points: (1) large and regular tax increases result in large and consistent revenue increases; (2) countries with high taxes and falling prevalence of tobacco use can still increase revenue with tax increases; (3) countries that reduce taxes experience revenue declines; and (4) countries that increase taxes in the face of illicit trade still increase revenue. Large and regular tax increases usually mean large and consistent revenue increases South Africa’s experience shows how successive tax increases, well above inflation and year after year, generate additional revenues even after taxes have been increased substantially. After two decades of declining real revenue in the 1970s and 1980s as real excise per pack declined, South Africa implemented successive excise tax increases from 1994 until 2011 (Fig. 4.4.1) (98, 104). After adjusting for inflation, this resulted in a real excise tax revenue increase of 245% (98). Revenues began to plateau from 2012 as tax increases stalled. They began to decline after 2015 – not due to tax increases, however, but due to a dramatic decline in administrative capacity and enforcement measures exacerbated by large-scale corruption in the government, including the tax administration authority (105). The rapid and catastrophic decline in tax administration and enforcement has been the subject of much attention (106). CHAP T ER 4. PO LI T I C AL ECO N OMY 221 Fig. 4.4.1 Real excise tax per pack of cigarettes and real excise tax revenue in South Africa, 1961–2020 Source: Data shared by University of Cape Town, 2020. Similarly, the Philippines provides a compelling example of how large and regular tax increases alongside reforms to tax structure can lead to large and consistent revenue increases – in this case, also through an accompanying reform to the tax structure (Fig. 4.4.2) (98). The 2012 Sin Tax Law consolidated the country’s four tax tiers into two by 2013 and established a uniform structure by 2017. The same law provided for large, progressive increases across the board, but in particular for the lowest tax categories (98). Not only were the revenue gains substantial, they exceeded all the projections for 2013–2017 made prior to the law’s passage (98). Excise tax per pack Excise tax revenue Ra nd s pe r p ac k (c on st an t 2 02 0 ra nd s) Excise revenue (constant 2020 rands) 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11 20 13 20 15 20 17 20 19 0 0 5 6 10 10 15 14 20 18 222 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.2 Real excise tax per pack of cigarettes (multitier, two-tier, unitary) and total tobacco real excise tax revenue in the Philippines, 2012–2018 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2012 as the base year. Sources: (107, 108 and data shared by the Philippines Department of Finance, September 2020). Ukraine is another example of a country that has regularly increased taxes over the past 10 years and has experienced increased revenues along with decreases in consumption and the number of smokers. Figure 4.4.3 the shows the trends in excise tax, revenues, cigarette sales and number of smokers in 2008–2017. Increases in excise rates were consistently accompanied by increases in revenues. In 2014–2015, excise tax was not increased above inflation (and inflation, especially in 2015, was very high, at 48.7%), so real values of excise and revenues went down. But it is evident from the data that revenues closely follow the path of excise levels even when sales go down. 12 Re al e xc is e ta x pe r p ac k, P hi lip pi ne p es os (2 01 2 ba se ) Billions (Philippine pesos) in tobacco real excise tax revenue (2012 base) 00 20 60 100 140 2012 HIGH PREMIUM SIN TAX LAW RA 10354 TRAIN LAW RA 10963 UNITARY RATE 2013 2014 2015 2016 2017 2018 5 10 15 20 25 30 32 .1 9 68 .6 6 71 .0 9 93 .5 2 87 .2 0 96 .0 1 11 4. 39 24.4 25.4 26.2 26.8 26.9 27.5 11.7 16 23.1 19.6 LOW Real tobacco excise revenues Dierent levels of real tobacco excise tax per pack MEDIUM 7.6 2.7 28.3 CHAP T ER 4. PO LI T I C AL ECO N OMY 223 Fig. 4.4.3 Average real cigarette excise tax rates, real cigarettes excise tax revenues (base year 2008) and cigarette sales and number of cigarette smokers in Ukraine, 2008–2017 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2008 as the base year. Source: Data provided by Konstantin Krasovsky, July 2020. Countries with high tax and falling prevalence of tobacco use can still increase revenue with tax increases Countries with already high tobacco taxes and rapidly diminishing tobacco use can still increase revenue by increasing taxes (98, 109). Australia has implemented comprehensive tobacco control policies and enacted consistent tobacco tax increases on top of what were already some of the highest tax rates in the world (see Fig. 4.4.4). Between 2001 and 2010, revenue increased with increasing tax rates, but in real terms (inflation-adjusted) it remained static (109). Then, in 2010, a 25% excise tax increase was introduced, with large annual increases scheduled from 2013 onward (98, 109). The result of this tax policy has been consistent and large increases in revenue year after year for nearly a decade, even when the increases were being made on already high tax rates.23 23 The apparent reduction in revenues in 2012 and 2013 was due to a change in the source of the data for 2001–2011 and 2012–2016. Data for 2012 and 2013 do not include customs duty, while all other years do. 73 95 125 112 88 82 75 76 67 11.8 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Number of daily smokers, in millions Real tobacco excise revenue, in billions of Ukrainian hryvnia Real average excise per cigarette packs of 20, Ukrainian hryvnia Number of taxed cigarettes (sales), in billions of sticks 0.6 3.6 10.1 9.2 8.7 8.6 8.4 8.1 7.3 6.2 6.5 6.3 7.8 10.3 11.2 12.1 13 1.4 2.2 2.5 2.9 3.5 3.2 2.7 3.3 4 12.7 13.4 74 9.7 224 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.4 Real excise tax and customs duty per stick and real total revenue (all tobacco products) in Australia, 2001–2019 Notes: Rates published by Australian Taxation Office and Australia Department of Immigration and Border Protection, adjusted using Australian Bureau of Statistics Consumer Price Index rates. The 2011–2012 and 2012–2013 figures do not include customs duty, which explains the apparent decline in revenue. Using official disclosures, Scollo and Bayly estimate that duties in these years were $7397.2 and $7687.2 respectively (110). Sources: (109, 110). Countries that reduced taxes and saw revenues decline Prior to 1982, Canada lowered taxes on cigarettes and experienced declining revenues as well as increased smoking – particularly among youth. Subsequent fivefold in- creases in cigarette taxes between 1982 and 1992 resulted in more revenue, increases in retail price and substantial reductions in consumption, with teenage smoking declining by nearly two thirds (5). In the early 1990s, a growing illicit trade in ciga- rettes emerged in which Canadian cigarettes exported to the United States were then smuggled back into Canada (5). The tobacco industry – which was later found to be complicit in and profiting from this illicit trade – sought to frame Canada’s high tax rates as the cause of smuggling (111–112) and succeeded in convincing the federal government, as well as six provincial governments, to make massive reductions in the tobacco tax (111–112). As a result, federal tax revenues fell significantly – more than twice as much as the government had predicted – and smoking rates among both adults and youth began to increase (5, 112). The Canadian government later changed its strategy, and the federal excise tax was restored, resulting in increased Specic excise/ duty per stick (cigarettes/cigars less than 0.8 g) Total revenue (all tobacco products) To ta l c us to m s/ du ty re ve nu e (in a tio n, a dj us te d, 20 19 A us tr al ia n do lla rs , m ill io ns ) Total excise/duty per stick (ination, adjusted, 2019 A ustralian dollars) 0 3 000 6 000 9 000 12 000 $0.8 $0.6 $0.4 $0.2 $0 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 CHAP T ER 4. PO LI T I C AL ECO N OMY 225 revenues and decreased smoking (5). Canada’s focus then shifted to using customs enforcement, rather than tax rates, as the best means of countering illicit trade (5). Countries that increased taxes in the face of illicit trade and still increased revenue As discussed in section 4.1, the tobacco industry exploits illicit trade as a strategy to undermine tobacco tax policy, with the goal of deterring governments from increasing tobacco taxes. The narrative that has been created is that higher tobacco tax rates result in increased illicit trade and undermine the policy goals by resulting in lower (or no) declines in tobacco use or lower (or no) increases or even decreases in revenue. However, as shown in section 4.1, the empirical evidence does not sup- port the industry arguments. Furthermore, the evidence shows that the industry and its allies have consistently overstated and exaggerated the scale and extent of illicit trade (see section 4.1). As was the case in Canada, Brazil’s tobacco tax policy suffered from a fear that the illicit market would expand unless it was undercut by price competition in the legal market, which it was thought could be best encouraged through tax cuts (20). Real excise tax rates declined from 1999 until the mid 2000s, as nominal increases were below the rate of inflation. This resulted in declines in real tax revenues (20). In these years, the tobacco industry used the tax cuts to increase profit margins rather than decrease prices and outcompete the illicit market, while also exaggerating the size and scope of the illicit trade problem (20). This caused the industry’s argument on illicit trade and revenue to lose credibility and resulted in increases in tax rates from 2007 onwards, with a major reform passed in 2011 (20). Tobacco excise rates and minimum prices were scheduled by the law to increase at levels above expected inflation from 2011 until 2015 (20). This resulted in substantial increases in the tobacco excise per pack, as well as overall revenue, which by 2015 had more than doubled from its low point in 2013 – equating to more than 50% in real terms (see Fig. 4.4.5). The success of this reform shows that revenues can be increased by higher rates despite the presence of a sizeable illicit market (113). More recent data show that revenues in Brazil declined in 2015 and 2016, coinciding with an increase in illicit trade, but also with an exceptionally bad economic recession that saw GDP decline by more than 3% in those years. 226 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.5 Average real excise tax per pack, real excise tax revenues and domestic cigarette sales in Brazil, 1999–2014 Notes: Data based on domestic sales and tobacco excise revenues, Federal Revenue Secretariat, indexed to 2013 Brazilian reals, using Consumer Price Index. Revenue collection indexed to 2013 reals, using Consumer Price Index. Source: (20). Improvements in tax administration and enforcement can also generate increases in revenues. In Kenya, several measures, including fiscal markings and, later, an advanced tracking and tracing system, improved collection, resulting in increases in both legal sales and tax revenues and a reduction in illicit sales (114). Moreover, these examples of poor governance indicate that attention should be focused on countries where a significant loss in administrative and enforcement capacity un- dermined revenue collection. 4.4.3 CONCLUSIONS The tobacco industry uses revenue concerns as a SCARE tactic to avoid, dilute and/or delay tobacco tax increases. The argument that higher taxes will decrease revenue is theoretically plausible, but real-world examples have demonstrated that this has not occurred. Furthermore, simulations show that even large tax increases in current average tax shares yield substantial revenue gains. The use of the Laffer curve by the tobacco industry should be challenged and refuted. The relatively price inelastic nature of cigarette demand combined with the Real excise tax amount per pack Domestic sales Real excise tax revenue D om es tic s al es (b ill io n pa ck s) Re al e xc is e ta x re ve nu es (b ill io n 20 13 re ai s) Real excise tax am ount per pack (2013 reais) 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 0 1 2 3 4 5 6 0 1 2 3 CHAP T ER 4. PO LI T I C AL ECO N OMY 227 low tax share and no overshifting of the tax means that most – if not all – countries are still far from the revenue-maximizing point, indicating that increases in taxes will lead to increases in revenues. The case studies in this section refute each of the tobacco industry’s arguments regarding alleged potential revenue loss due to tax increases. The experiences of South Africa, the Philippines and Ukraine demonstrate that large and regular tax increases result in large and consistent revenue increases. Well-designed tax structures have also proven to play an important role in generating revenues. The experience of Australia shows that even countries with already high tax rates and declining prevalence of tobacco use can increase revenues with regular, large tax increases. The experience of Canada warns against following the advice of the tobacco industry to decrease taxes as a way to fight illicit trade. It demonstrates clearly that decreasing tobacco taxes will decrease revenue and encourage consumption, rather than counteract illicit trade. The experience of Brazil shows that countries with substantial illicit trade issues can still increase revenue by increasing taxes. Finally, in the few cases where revenue decreases were seen, the reasons for the decreases were not strictly linked to tax increases. This was the case in Tonga, where the increase in tax was applied only to cigarettes and not to their close substitute, loose tobacco – leading smokers to switch products. In South Africa, a decrease in revenue was the result of the weakening of government institutions. And in Ukraine, real revenues decreased only during the two years when taxes were not increased above inflation. 228 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.5 SCARE TACTIC E: EMPLOYMENT 4.5.1 INTRODUCTION In opposing tax increases, the tobacco industry often seeks to frame tobacco taxes as an economic rather than a public health issue (5, 48, 53, 115). Particular emphasis is placed on the alleged threat that tax increases pose to employment in tobacco farm- ing and manufacturing, as well as other related industries (5). This so-called choice between health and jobs is, however, based on several false premises, including (5): 1. tobacco is a significant source of jobs within the context of broader labour markets, and domestic tobacco tax increases will have a drastic effect on domestic employment (48, 53, 116–117); 2. tobacco consumption is an indispensable engine for job creation (5, 48, 54); and 3. tobacco provides highly prosperous, sustainable and irreplaceable livelihoods (5, 53, 118). In reality, the relationship between tobacco taxation and employment is consider- ably more complex than the industry makes it out to be. In fact, there is ample evidence to show that tobacco taxes are a win-win for public health and the fiscal space, without measurable risks to employment. 4.5.2 THE LINK BETWEEN TOBACCO EMPLOYMENT AND TOBACCO TAX RATES Tobacco farming, production and manufacturing (including hand-rolling in some countries, most of them in South-East Asia) constitute a small proportion of the labour force, even in countries where the industry is most heavily concentrated (5, 48, 103, 116). Employment in tobacco farming and manufacturing has been declining globally due to advances in technology, trade liberalization, market consolidation and the privatization of formerly state-owned tobacco companies (5, 103, 119). These same trends have led to the heavy concentration of tobacco growing and manufacturing in only a handful of countries – and within these countries, often in only a small number of regions (103, 119–121). Even in those countries that lead in tobacco growing and manufacturing, tobacco’s overall share of total agricultural and manufacturing employment is relatively small and is often decreasing as efficiencies in production reduce labour intensity (5, 117–118, 122). Similarly, the industry’s claim that tobacco taxes reduce employment is exag- gerated and typically overlooks wider trends driving tobacco industry employment. Indeed, tobacco industry developments and innovations have played a greater role in the reduction of employment in the tobacco industry than have tobacco control policies (103). Despite industry claims that tobacco taxes can affect employment, CHAP T ER 4. PO LI T I C AL ECO N OMY 229 characteristics of the location of production – such as market size, labour costs, growing conditions and leaf preferences – have much more to do with tobacco industry interests than with the tobacco tax rate (103, 123). Moreover, jobs in countries that produce tobacco primarily for export are not greatly affected by reductions in local consumption resulting from tax increases (5, 103, 116, 123). Finally, it has been demonstrated that tobacco tax increases do not have a significant effect on employment in the retail sector, as most retail businesses sell other goods (103). Estimates of the gross employment impact of tobacco tax hikes demonstrate that job losses that do occur can be more than compensated for by increases in revenue. A 2018 World Bank study estimated that in Indonesia, for example, an ambitious tax reform that would simplify tiers and increase prices by close to 50% would reduce gross employment in the tobacco manufacturing sector by less than 0.5% (a loss of 2 914 jobs). The government could provide income support to the displaced workers (for example, through training, temporary transport/mobility or income support) with less than 2% of the revenue gained from the tax increase (117). Similarly, a 2019 study by Bangladesh’s National Board of Revenue estimated that a substantial increase in tobacco taxation would cause 7 012 lost jobs, but that the total income associated with these job losses in the bidi industry would amount to only 3.5% of the revenue gained (120). Accordingly, increased revenue can more than compensate for the expenditure of supporting those who lose jobs and need to acquire new skills before transitioning to new employment (120). Box 4.5.1 Employment fears deployed to frustrate tobacco tax reform in Indonesia In 2017, the Indonesian Ministry of Finance decided to implement tobacco tax in- creases by 2019 and tier simplification by 2021 (124). This resolution was, however, abandoned within a year, after a concerted campaign by tobacco industry actors and their allies to reframe the increase as an economic issue with a focus on, among other things, the effect the tobacco tax increase would have on employment (124). This defeat for the tobacco tax initiative came despite estimations of how the loss of income associated with lost jobs would be dwarfed by the additional revenue gained by the tax (117). Earlier analysis had estimated an overall large net positive impact on employment from tobacco tax increases (125), which illustrates how evidence that challenges assumptions around the negative socioeconomic impacts of tobacco control tends to be discounted (126). In this case, tobacco industry arguments seem to have resonated strongly with politicians from the electoral districts of West Java, East Java, Central Java and West Nusa Tenggara, where employment in tobacco farming and manufacturing is concentrated (124). Although tobacco manufacturing 230 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N represented only 5.13% of total manufacturing employment, the concentration of the job and economic activity meant that arguments regarding employment were particularly salient (117,119). Accordingly, tobacco industry arguments that tobacco taxation would negatively impact employment and farmer livelihoods prevailed, despite strong opposing evidence (119). Concentration of tobacco industry activity within countries poses particular obstacles to overcoming industry arguments on employment and needs to be given careful attention. 4.5.3 THE EVIDENCE ON THE NET EFFECT OF TOBACCO TAX INCREASES ON EMPLOYMENT A proper analysis of the effect of tobacco tax increases on employment must examine their impact on net or economywide employment. Decreases in expenditures on tobacco associated with tobacco control do not mean that expenditures simply disappear; rather, they are redistributed towards consumption of other goods and services, thereby generating employment elsewhere in the economy (5, 103, 123). Similarly, though the effect of higher tobacco taxes on net consumption is arguably more ambiguous, revenues from this intervention do generate spending, invest- ment and employment in public services such as health and education (5). Tobacco control polices usually have a marginal neutral or positive effect on net employment, particularly in countries that are net importers of raw or manufactured tobacco products, as expenditures on these imported items tend to flow out of the country (5). Export-oriented tobacco producers are less sensitive to local demand and are not significantly affected by domestic tobacco tax measures, which likely have a near-neutral net impact (5). In some cases, the net employment impact is a very small negative number, typically less than 1% (127–128). A recent study estimated that in the United Republic of Tanzania – a large tobacco-producing and exporting country – a 30% reduction in smoking prevalence would result in a net employment decline of just 0.5% across the economy as a whole (129). A similar study of Pakistan found that, with some variance depending on where spending was redistributed from tobacco consumption, the overall net effect on employment from a significant reduction in expenditure on cigarette employment – 1 billion rupees – would be a gain of between 6 651 and 5 803 jobs (122). This increase would occur because expenditure on cigarettes produces much less employment in the broader economy than expenditure on food and education (122). In the United Republic of Tanzania, as elsewhere, increased revenue could be used to assist those who lose employment with transitioning to new livelihoods. CHAP T ER 4. PO LI T I C AL ECO N OMY 231 4.5.4 THE VIABILITY OF BETTER LIVELIHOODS In arguing against tobacco tax increases, the tobacco industry advances the myth that people employed in tobacco production – particularly tobacco farming, but also manufacturing – lack any other prospect for a comparably attractive livelihood. However, studies based on extensive survey data in Indonesia, Kenya, Malawi, the Philippines and Zambia have shown that despite needing to commit significant amounts of labour to their crop, tobacco farmers often suffer losses rather than gain profits (119, 121, 123, 130–132). Furthermore, the Indonesian studies demonstrate that tobacco farming has a negative impact on household income and opportunity compared with the experience of other farming households that have given it up (119). Declines in consumption as a result of tobacco tax increases are gradual and susceptible to the same progressive adaptation that has occurred for decades (5, 103). While there will be a need in some countries for the government to help farmers transition to other crops or industries in the longer term, this process will not be a major short-term shock to employment or the wider economy (123). Because tobacco growing and manufacturing can be concentrated in just a few locations within a country, job losses within the tobacco industry might have a disproportionate effect in one location, while employment gains from reduced consumption may be spread across the whole country (120, 123). A study of the employment effects of tobacco tax increases in Bangladesh estimated that up to 60% of all job losses would occur in only two districts – among the poorest in the country – due to the high level of industry concentration (120). Studying the need for support, as well as the means of delivery and funding of support, is particularly necessary in these circumstances. Beyond the need to ensure equity and support employment, a failure to provide for targeted relief can exacerbate fear of job losses and may prove fatal to a tobacco tax proposal (120). Box 4.5.2: Supporting alternative livelihoods in the Philippines The Philippines earmarked 15% of the revenue from a 2012 increase in tobacco taxes to supporting economically viable alternative livelihoods for tobacco farmers and workers (5). Tobacco farming in the Philippines is regionally concentrated, and the tobacco industry had previously been successful in deploying concern for smallholder tobacco farmers to undermine tobacco control measures (136). The provision of economic support was a politically effective countermeasure to tobacco industry SCARE tactics and eased the tax increase’s passage. Given the Philippines’ integration with global tobacco markets and demand, tobacco farmers have not been seriously affected by the tobacco tax increase and 232 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N reduced domestic demand (108, 132). Nevertheless, transfers to tobacco-growing regions from the earmarked tax have been substantial (108). These funds are support- ing gradual transitions to alternative livelihoods, with farmers being encouraged to take up alternative crops, as well as establishing infrastructure, such as market-to-farm roads, that will make these alternative crops more economically viable (108, 133–134). Supporting alternative livelihoods for farmers and other tobacco workers is important because it can offset the political effect of industry arguments, even though domestic tax increases usually have only a modest and gradual effect on employment. There are various models for supporting alternative livelihoods when employment in the tobacco sector gradually diminishes due to decreases in either global or national demand. The Philippines is exemplary, but many other countries have either implemented or experimented with supporting crop transitions. Turkey’s alternative crop pro- gramme, implemented in anticipation of the privatization of the country’s cigarette monopoly, has proven effective in supporting many tobacco farmers’ move to other crops (135). Smaller-scale crop substitution projects in Kenya and Yunnan Province in China have shown how financial, regulatory and infrastructure support from government can contribute to crop transitions (5, 53). Argentina, Bangladesh, Mexico and the state of Maryland in the United States provide additional case studies of how governments can support these transitions (5, 136). 4.5.5 CONCLUSIONS The tobacco industry exaggerates the importance of tobacco employment and over- states the impact that domestic demand reduction due to local taxes will have on tobacco farmers serving a global market. The industry also simplifies employment’s relationship with taxation by focusing only on gross employment in tobacco, which ignores the reality that expenditures on tobacco do not disappear but rather are redistributed for other consumption that can produce a similar or higher number of jobs. Many detailed studies have found that tobacco growing is much less profitable and sustainable than the tobacco industry claims. Tobacco farmers throughout the world have successfully transitioned to other crops, although the transition often requires temporary or additional support from the government or other stakeholders. The extent of such support is moderated by the reality that transition from tobacco to other crops is a long-term consideration. CHAP T ER 4. PO LI T I C AL ECO N OMY 233 4.6 EARMARKING TOBACCO TAX REVENUES TO FUND HEALTH 4.6.1 INTRODUCTION Earmarking tax revenues involves the separation of all or a portion of revenue from a tax or group of taxes to be put aside for a specific purpose (137). Globally, more than 80 countries earmark for health (138), and 37 earmark tobacco tax revenues for health (27). There are two main types of earmarks: hard – also called substantive – and soft, or symbolic (139). Hard earmarks link the expenditure with a revenue source in legislation. This can limit funding if the earmarked revenues are the main source of funding, or it can cause surpluses to accrue wastefully when more revenues are raised than may be expended for the earmarked purpose. Soft earmarks include dedicated funds or commitments to use funds for a particular purpose. They are not necessarily legally binding. For example, in France, the ma- jority of tobacco tax revenue is used to fund social security (which includes health insurance and health care), but there is no hard, formal earmark (140). Earmarks can also be some combination of hard and soft. In the Philippines, tobacco tax earmarks are legally binding, but earmarked revenues go to the general fund, and the Department of Health must submit an annual budget for covered programs as part of its budget request. Earmarking is a broad and contentious topic that goes beyond the specifics of tobacco tax earmarking. Discussions on the topic fall within the ambit of public financial management, and earmarking generally is not encouraged. From a tobacco control perspective, however, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. It is a tool to improve the political economy of tobacco taxation; it is a secondary issue only, after the primary goal of reducing demand for tobacco. One way to use earmarking to improve the political economy of tobacco taxation is to link the payment of tax by tobacco users to benefits they will receive through the funding of complementary tobacco control programmes, such as cessation support, or through increased funding for health programmes on which they will rely disproportionately. This is known as the benefit principle. Earmarking for tobacco control makes sense, as its financial cost is relatively small and tobacco tax reduces demand more effectively when implemented within a package of complementary tobacco control measures. Another way earmarking improves the political economy of tobacco taxation is by safeguarding against any perceived or potential negative ramifications of the tax itself. This is important for neutralizing erroneous but often convincing tobacco industry arguments against effective tobacco tax policies. For example, the Philippines earmarks the bulk of the additional revenues from sin taxes for the health insurance premiums of the poor. In addition, a portion 234 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the country’s tobacco tax revenues is earmarked to provide for the economic well-being of tobacco growers and tobacco growing regions, with the general aim of promoting economically viable alternatives to tobacco farming and manufacturing as a safeguard against the potential for reduced domestic tobacco demand (141). Tobacco tax earmarks are complex, however, and care is required when determin- ing whether a particular earmark is needed on the grounds of political economy and justified on the grounds of equity and economic efficiency. In assessing tobacco tax earmarks, many of the same criteria that have been used to assess the appropriateness of generic earmarks also apply. In the rest of this section, these criteria are set out and matched with reasons for the ability of well-designed tobacco tax earmarks to fulfil them. The types and structures of tobacco tax earmarking are explored alongside descriptions of country experiences to provide guidance on how tobacco tax earmarks are used, when they are justified and the best ways to design them. 4.6.2 CIRCUMSTANCES IN WHICH EARMARKS MAY BE SUITABLE Scepticism about earmarking is both long-standing and justified, but much of the debate concerns earmarking generally and is not specifically concerned with the merits of tobacco tax earmarking (138). The main concerns raised about earmark- ing are listed in Table 4.8, accompanied by suggestions for how earmarks may be structured to address these concerns. Table 4.8 Concerns about earmarking and suggested safeguards to avoid the concerns MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Democratic accountability and oversight: earmarks undermine democratic processes by impeding legislative and executive oversight over expenditure. Establishing proper oversight and accountability procedures is important to ensure funds are not mismanaged (138). Additionally, if a soft earmark structure, which transfers revenue to the general fund from which it is then allocated, is adopted, this will not be a concern. Budget rigidity: earmarking may create budget rigidity that can lead to inefficient allocation of resources (138). An earmark’s particular design determines how much rigidity is introduced (138). Flexible soft earmarks are less prone to introducing rigidity than hard earmarks. Concerns about rigidity can be reduced by the inclusion of a sunset clause that ensures that the earmark is automatically discontinued or reviewed after a set period of time has elapsed (138). A further safeguard is to establish the earmark as a waterfall account, with any excess revenue over a set amount being allocated to the general fund. CHAP T ER 4. PO LI T I C AL ECO N OMY 235 MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Fragmentation: earmarking can result in fragmented and uncoordinated expenditures. This means policies complementary to the earmarked purpose but outside of its purview may be unfunded (138, 142). This is a legitimate concern. The negatives of fragmentation cannot be entirely eliminated, but they may be outweighed by the other merits of tobacco tax earmarking. That said, proposals for tobacco tax earmarks should be scrutinized to ensure that the funded purpose is at least cost-effective. Decreased equity: equity will decrease if individual access to benefits is narrowly defined according to payments made. This issue is not likely to arise with tobacco tax earmarks but is conceivable and something that should be guarded against in an earmark’s design. Capture by special interests: because earmarks are often the result of political expediency, an earmarked purpose may be determined by powerful special interests promoting a tax’s passage rather than careful prioritization of resources (138). Well-designed earmarks will guarantee funding for underresourced programmes and high- priority programmes. While the above concerns may be valid and design does matter, tobacco tax and other health-promoting taxes are not subject to the same concerns when it comes to the justifiability of earmarking their revenue (138, 143). Some of the factors that distinguish tobacco tax earmarks from more general critiques of earmarking are listed in Table 4.9 (138). Table 4.9 Concerns about earmarking and distinguishing factor for tobacco tax earmarks GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Procyclicality: earmarked revenues are often procyclical and susceptible to booms and busts (138–139, 142). Tobacco tax revenues are generally not cyclical (they are recession-proof ), and revenue is predictable relative to most other indirect and direct taxes (103). Budget rigidity Tobacco tax earmarks necessarily involve only a relatively small proportion of the budget; therefore, the effect of any rigidity will be relatively insignificant. Partly because of the relatively small amounts involved, there is only limited real-world evidence of tobacco tax earmarks having introduced harmful rigidity (143).24 24 See also the subsection on the amount of money associated with tobacco tax earmarks in section 4.6.3. 236 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Capture by special interests All earmarks should be scrutinized to ensure that their funded purpose is cost-effective. In the case of tobacco tax earmarks, however, political economy considerations may mean that it is sufficient for a low-priority purpose to be funded if the funding will unlock the political will needed for effective tobacco tax increases. In these cases, special interests are being purposefully catered to in order to ensure that tobacco tax increases occur. Of course, arguments against tobacco taxes and tobacco tax earmarking are led by special interests such as the tobacco industry (144–147). Insufficient revenue: the earmarked revenue source may become insufficient for funding its purpose (139, 142). Even though revenue may decrease in the long term when more tobacco users quit, such decrease is expected to be gradual. See Section 4.4 for details on how, with effective design, revenue will generally increase even with declining consumption. In addition to these reasons why general concerns about earmarking do not fully apply to well-designed tobacco tax earmarks, there are a number of compelling reasons for tobacco tax earmarking to finance tobacco control or public health that argue in favour of its implementation: • Significant increases in excise taxes are the most effective, as well as the most cost-effective mechanism for reducing consumption, but they are best implemented as a part of a package of complementary tobacco control measures, such as the WHO MPOWER package. Earmarking tobacco tax revenue for interventions that may not be funded otherwise can strengthen overall tobacco demand reduction (148). • The political economy of tobacco tax increases also makes earmarks attractive: – People have been shown to be more supportive of tobacco tax increases when they know the revenues will be used for targeted social programmes (143, 149–150). Earmarking tax revenue for health or tobacco control frames tobacco tax as a public health intervention in the minds of the public, which may otherwise view it as merely a revenue source (138). Research has shown that using earmarking to link a tobacco tax to health can also help raise awareness about the dangers of tobacco use (143). – When tobacco tax revenue is earmarked for programmes that benefit vulnerable groups, the tax becomes more equity-enhancing. Although lower socioeconomic groups and young adults receive disproportionate health and economic benefits from tobacco tax increases over the medium term, these groups will expend a greater share of their income in the CHAP T ER 4. PO LI T I C AL ECO N OMY 237 short term because of tobacco taxes. Earmarking tobacco tax revenue for programmes such as UHC or cessation services that provide immediate benefits to these groups neutralizes some critiques of tobacco taxation (e.g. the 2009 United States federal excise tax increase and the 2012 Philippines Sin Tax Reform illustrate how equity-enhancing earmarking facilitated passage of substantial tax rises) (93, 138, 143, 150). 4.6.3 EARMARKING PRACTICES AND COUNTRY EXAMPLES Earmarking tax revenues for health is a common practice in 80 countries. In 2018, 37 countries from all regions of the world earmarked tobacco tax revenues for health purposes.25 Case studies in the political economy of tobacco tax earmarking In 2012, the Philippines comprehensively reformed tobacco and alcohol excise taxes. Tobacco taxes were increased significantly, and numerous tax tiers were reduced to only one tier by 2017. Although increasing revenue was a foremost motive for some officials, the reform was explicitly framed around boosting UHC funding and advancing public health by reducing alcohol and tobacco consumption. Earmark- ing of tax revenue for UHC was essential to the political compromise that made this trailblazing tax increase a reality. It ensured that the increase, which may have otherwise been perceived as regressive, was framed as a progressive public health measure in the public imagination, while also appeasing tobacco growers and their political representatives. Earmarking was also important because the earmark en- sured high-level support for the tax by achieving a key political priority (151). Its soft-earmark structure meant it was not a blank cheque to the Ministry of Health, and this addressed concerns within the Ministry of Finance. Similarly, in Australia, earmarking of revenue helped overcome community objections to tobacco taxes and tobacco control more generally that resulted from the tobacco industry’s sponsorship of sports and the arts in the 1980s. Attempts to completely ban tobacco advertising and sponsorship had been unsuccessful due to strong pressure from sports, arts and racing lobbies that claimed that a ban would harm these activities. States, starting with Victoria, responded by earmarking funding for Health Promotion Foundations that took over the tobacco industry’s sponsorship activities and also paid for antismoking campaigns. In 1997, these earmarks ended after a High Court ruling that the Constitution did not allow states to collect excise taxes. However, in recognition of the successful work of the Foundations, the federal 25 Details about earmarked taxes by country are provided at https://www.who.int/tobacco/global_report/ Table-9-4-Use-of-earmarked-tobacco-taxes.xls?ua=1. 238 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government began funding them directly from the federal budget (152). Although this example may be difficult to replicate precisely, it shows how earmarks with built-in sunset clauses for piloting cost-effective interventions can potentially graduate to funding from the general budget once they have proven their effectiveness. It also more generally shows how earmarks can disarm community objections, reframe tobacco tax increases and unlock the political will needed to advance effective tobacco control measures. Structures for managing earmarked tobacco taxes A 2016 review of nine countries’ tobacco tax revenue earmarking experiences identified three arrangements for governance and allocating revenue (151). Table 4.10 presents some examples of these allocation arrangements. In some countries, earmarked tobacco tax revenues are combined with alcohol tax revenues. Table 4.10 Illustrative arrangements for allocating earmarked tobacco tax revenues POSSIBILITIES FOR ALLOCATING TOBACCO TAX EARMARK REVENUE Forms of budget allocation Revenue goes to the general fund and is later assigned to the official actor(s) specified in the earmark. Revenues do not go through the general budget but are instead paid into a separate account belonging to the official actor(s) specified in the earmark. Earmarked tax revenue is paid directly to the account of the entity managing an autonomous or semi- autonomous fund. Examples In the Philippines, revenue goes to the general fund before being allocated to the Ministry of Health following submission of a budget for its use (140). In Romania, revenue goes directly into a Ministry of Health account that is distinct from the general fund. In Panama, revenue is paid into subaccounts of the three recipient agencies (the Ministry of Health, the National Cancer Institute and the Customs Authority). In Thailand, ThaiHealth directly receives the earmarked revenues in its own account. In Viet Nam, the Viet Nam Tobacco Control Fund receives the revenues directly into a subaccount it manages but that belongs to the Ministry of Health. Source: (151). Where is the money being spent? Earmarked tobacco tax revenues are used for a variety of health purposes, including tobacco control, health promotion and UHC. A wide variety of other programmes have also been funded with earmarks from tobacco taxes, including disaster relief (e.g. hospital medical supplies and equipment to treat COVID-19 in India), youth pro- grams, sports and craft jobs in Yemen, social cohesion in Morocco, health and social programs in areas dependent on tobacco growing in Argentina, health promotion CHAP T ER 4. PO LI T I C AL ECO N OMY 239 and tobacco control in Thailand and alternative livelihood programs for tobacco farmers as well as economic projects in tobacco-growing provinces in the Philippines. Table 4.11 shows the three main categories of health programmes to which ear- marked tobacco tax revenue is allocated, as well as a fourth miscellaneous category, with country-specific examples for each.26 Table 4.11 Programmes to which earmarked tobacco tax revenue is allocated TOBACCO CONTROL NCD PREVENTION AND CONTROL PROGRAMMES (otherwise indicated between brackets) HEALTH COVERAGE EXPANSION (e.g. through health insurance coverage) OTHER, MORE GENERAL OR UNSPECIFIED HEALTH PROGRAMMES Costa Rica, Côte d’Ivoire, the Islamic Republic of Iran, Madagascar, Panama (tobacco cessation and fighting illicit trade), Switzerland, Viet Nam Cook Islands, Costa Rica, Mauritania (anti-cancer research), Palau (NCD prevention only), Panama (National Institute of Oncology), Paraguay Colombia, Congo, Egypt, Palau, Philippines Algeria, Argentina, Bangladesh, Botswana, Cabo Verde, Chad (programmes delivering antiretroviral drugs), Colombia (sports), Comoros (sports, hospital emergencies), Congo (sports), Côte d’Ivoire (AIDS programme), El Salvador, Estonia (sports), Guatemala, Indonesia, the Islamic Republic of Iran (sports), Ireland, Jamaica, Lithuania (sports), Madagascar (sports), Morocco, Nepal, Paraguay (sports), Republic of Korea (health promotion), Romania, Thailand (health promotion), United States, Yemen (sports) Note: Countries appear in more than one column when their earmarked tax revenues are used in more than one specific health programme. Source: (27). The amount of money associated with tobacco tax earmarks Case studies of the experiences of nine countries in tobacco tax earmarking show that earmarked funds are relatively small in comparison with government spend- ing on health (see Table 4.12) and, consequently, even smaller in terms of GDP. Therefore, the argument that tobacco tax earmarks would introduce rigidity into public financial management may not apply. 26 Details about how the tobacco tax revenues are earmarked are given in Annex 4.2. 240 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.12 Proportion of earmarked tobacco tax funds in government expenditures COUNTRY ESTIMATED ANNUAL TOTAL FUNDS FROM EARMARKED TAX ANNUAL FUNDS FROM TOBACCO TAX EARMARKS as a % of general government expenditure on health in 2013 Botswana 2014–2015: 4 million pula (US$ 0.48 million) NA Egypt 2013–2014: 392 million Egyptian pounds (US$ 52.06 million); earmarked taxes only 1.8% of total taxes on cigarettes 1.086% Iceland 2014: 108.3 million kronor (US$ 0.89 million) 0.083% Panama 2014: US$ 27.8 million 1.322% Philippines 2014: 50.18 billion Philippine pesos (US$ 1.18 billion) NA Poland 2013: 1 million złoty (US$ 0.316 million) from general budget  0.001% Romania 2014: 1.1 million lei (US$ 0.33 million); 14.4% of total health budget 0.004% Thailand 2014: 4064.74 million baht (US$ 125.15 million); 1.78% of Ministry of Health budget and 1.84% of National Health Security Fund 0.932% Viet Nam 2014: 299.171 billion dong (US$ 13.91 million); 0.5% of national health budget 0.335% Source: (151). 4.6.4 CONCLUSIONS Despite the initial principled resistance to earmarking by some ministries of finance, experience has shown that the use of revenue from tobacco taxes and other taxes on the consumption of products that have negative externalities can ensure political as well as public support. Successful earmarking needs a well-developed structure for the use of funds for health purposes. Even intergovernmental organizations that are opposed to earmarking (e.g. the IMF) have acknowledged the justifiability of well-designed tobacco tax earmarks when revenue is directed to specific cost- effective programmes (153–154). The amounts of tobacco tax revenue effectively earmarked for health have been relatively small and could hardly introduce the feared rigidity in government budgets. Moreover, in some countries, those funds have helped to implement much-needed health programmes (e.g. Australia, the Philippines, Thailand). More governments are considering this option as a stable medium-term source of secure funding for programmes such as tobacco control. The payoffs will be seen in the future as fewer people fall ill and less medical care for tobacco-related illnesses is needed. In Australia, CHAP T ER 4. PO LI T I C AL ECO N OMY 241 an earmarked tax was used to fund a needed and underresourced programme that proved to be successful, effective and impactful; the programme is now sustainably funded, embedded in the federal budget. Earmarking is desirable in a particular political economy when it enables the implementation of effective tobacco taxation that will increase price and reduce consumption. It will, however, also be rational as a matter of public financial man- agement, economic efficiency and democratic governance when concerns such as the following are considered. Although not every question needs an affirmative answer, policy-makers who can answer yes to many of the following questions will likely be considering an effective and rational tobacco tax earmark:27 • Does the tobacco tax earmark’s purpose rationally connect with the recipient programme’s purpose? Earmarks that fund tobacco control or other health programmes are more economically rational under the benefit principle than those that fund unrelated programmes such as childhood education, even when the popularity of the unrelated programmes may make a tax increase politically palatable. • Does the tobacco tax earmark’s amount rationally connect with the needs of the recipient programme? Earmarked funds that cannot be absorbed by the recipient programme are, in effect, money taken away from other needs. • When a tobacco tax earmark funds health programmes, is this clearly com- municated to the public to ensure that the framing of the tobacco tax increase as a health measure reinforces the demand-reduction effect? • Is the programme being funded by the tobacco tax earmark a politically neglected but highly cost-effective or crucially needed programme that, once established as a proof of concept, has a chance of being funded out of the general budget? • Does the tobacco tax earmark’s purpose rationally connect with the effects of the tax itself? Earmarks that fund programmes that disproportionately benefit lower socioeconomic groups or that fund alternative livelihoods for former tobacco workers and farmers will have equity-enhancing effects that will reinforce the already progressive nature of tobacco taxes. • Is the scope of the earmark’s purpose narrow enough that it can be funded mostly from the tobacco tax earmark, to ensure that the revenue is additive and does not merely substitute for spending that would otherwise come from the general fund? • Does the design of the tobacco tax earmark provide for flexibilities that ensure that windfall revenue collection is not squandered on a purpose already saturated with overfunding? 27 Adapted from and informed by References 138–139, 143, 150. 242 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Does the design of the tobacco tax earmark include a sunset clause that triggers its automatic end or review? The presence of such a clause will guard against inefficiency in allocation arising and being left uncorrected due to political inertia. CHAP T ER 4. PO LI T I C AL ECO N OMY 243 REFERENCES 1. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tob Control. 2013; 22(2): 144–153 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 31 January 2021). 2. LaFaive MD, Drenkard S, Nesbit T. Strong link between cigarette tax and illegal smuggling rates [website]. Midland: Mackinak Center for Public Policy; 2018 (https://www.mackinac.org/strong- link-between-cigarette-tax-and-illegal-smuggling-rates#:~:text=An%20increase%20of%2011%20 percentage,packs%20to%2030.6%20million%20packs.&text=Tax%20evasion%20and%20avoidance%20 is,high%20excise%20taxes%20on%20cigarettes., accessed 31 January 2021). 3. Ross H, Blecher E. Illicit trade in tobacco products need not hinder tobacco tax policy reforms and increases. Chicago: University of Illinois at Chicago; 2019 (Tobacconomics White Paper; https:// tobacconomics.org/research/illicit-trade-in-tobacco-products-need-not-hinder-tobacco-tax- policy-reforms-and-increases-white-paper/#:~:text=additional%20related%20research-,Illicit%20 Trade%20in%20Tobacco%20Products%20Need%20Not%20Hinder%20Tobacco,Reforms%20 and%20Increases%20( White%20Paper)&text=The%20paper%20cites%20research%20 that,consumption%2C%20and%20improves%20overall%20health, accessed 31 January 2021). 4. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 12 October 2020). 5. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 6. Joossens L, Chaloupka FJ., Merriman D, Yürekli A. Issues in smuggling of tobacco products. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac00Box 370064B00PUBLIC0.pdf, accessed 18 February 2021). 7. Merriman D, Yürekli A., Chaloupka, FJ. How big is the worldwide cigarette smuggling problem. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac 00Box370064B00PUBLIC0.pdf, accessed 18 February 2021). 8. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 9. Merriman D. Understand, measure, and combat tobacco smuggling. Washington (DC): World Bank Group; 2003 (Economics of tobacco toolkit 7; https://openknowledge.worldbank.org/bitstream/ handle/10986/16267/805800NWP070Sm0Box0379807B00PUBLIC0.pdf?sequence=1&isAllowed=y, accessed 31 January 2021). 10. Progress report on the implementation of the Commission Communication. Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products – a comprehensive EU strategy. Brussels: European Commission; 2017 (COM(2017) 235 final; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:52017DC0235&from=en, accessed 31 January 2021). 11. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 12. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Reuter P, Majmundar M. (editors). Washington (DC): National Academies Press; 2015. 13. von Lampe K. The illegal cigarette trade. In: Natarajan M, editor. International crime and justice. New York: Cambridge University Press. 2010:148–54. 14. Joossens L, Raw M, From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–34 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 31 January 2021). 15. Gomis B, Lee K, Carrillo Botero N, Shepherd P, Iglesias RM. “We think globally”: the rise of Paraguay’s 24 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tabacalera del Este as a threat to global tobacco control. Global Health. 2018;14(1):110 (https:// globalizationandhealth.biomedcentral.com/articles/10.1186/s12992-018-0412-3, accessed 31 January 2021). 16. Communication from the commission to the Council and the European Parliament: Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products - A comprehensive EU Strategy. Brussels: European Commission; 2013 (https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52013DC0324&from=EN, accessed 31 January 2021). 17. Ross H, Vellios N, Smith KC, Ferguson J, Cohen JE. A closer look at ‘cheap white’ cigarettes. Tob Control. 2016;25(5):527–31 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/527.full.pdf, accessed 31 January 2021). 18. Nagelhout GE, van den Putte B, Allwright S, Mons U, McNeill A, Guignard R, et al. Socioeconomic and country variations in cross-border cigarette purchasing as tobacco tax avoidance strategy. Findings from the ITC Europe Surveys. Tob Control. 2014;23:i30–i38 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/23/suppl_1/i30.full.pdf, accessed 1 February 2021). 19. Gilmore AB, Rowell A, Gallus S, Lugo A, Joossens L, Sims M. Towards a greater understanding of the illicit tobacco trade in Europe: a review of the PMI funded ‘Project Star’ report. Tob Control. 2014;23:e51–e61 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/e1/e51.full.pdf, accessed 1 February 2021). 20. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Revista Panam Salud Publica. 2016;40(4):243–9 (https://www.scielosp.org/pdf/ rpsp/2016.v40n4/243-249, accessed 1 February 2021). 21. Operação Huno combate mercado clandestino de cigarros em sete Estados [Operação Huno fights clandestine cigar market in seven states]. Brasilia: Federal Police of Brazil; 2015 (in Portuguese) (http:// www.pf.gov.br/agencia/noticias/2015/11/operacao-huno-combate-mercado-clandestino-de-cigarros- em-seis-estados, accessed 12 October 2020). 22. Gilmore AB, Gallagher AWA, Rowell A. Tobacco industry’s elaborate attempts to control a global track and trace system and fundamentally undermine the Illicit Trade Protocol. Tob Control. 2019;28:127–40 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/28/2/127.full.pdf, accessed 1 February 2021). 23. Technical assessment of the experience made with the Anti-Contraband and Anti-Counterfeit Agreement and general release of 9 July 2014 among Philip Morris International and affiliates, the Union and its Member States. Brussels: European Commission; 2016 (SWD (2016) 44 final; https://ec.europa.eu/anti- fraud/sites/antifraud/files/technical_assessment_pmi_24022016_en.pdf, accessed 1 February 2021). 24. Gallagher AWA, Evans-Reeves KA, Hatchard JL, Gilmore AB. Tobacco industry data on illicit tobacco trade: a systematic review of existing assessments. Tob Control. 2019;28(3):334–45 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/28/3/334.full.pdf, accessed 1 February 2021). 25. Joossens L, Merriman D, Ross H, Raw M. How eliminating the global illicit cigarette trade would increase tax revenue and save lives. Paris: International Union Against Tuberculosis and Lung Disease; 2009 (https://theunion.org/sites/default/files/2020-08/How%20Eliminating%20the%20Global%20Illicit%20 Cigarette%20Trade%20Would%20Increase%20Tax%20Revenue%20and%20Save%20Lives%20-%20 EN.pdf, accessed 1 February 2021). 26. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 27. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 28. Prieger JE, Kulic J. Cigarette taxes and illicit trade in Europe. Econ Inquiry. 2018;56(3):1706–23 (https:// papers.ssrn.com/sol3/papers.cfm?abstract_id=2718519, accessed 18 February 2021). 29. Stoklosa M. Prices and cross-border cigarette purchases in the EU: evidence from demand modelling. Tob Control. 2020;29:55–60. 30. Passport: illicit trade in tobacco products. In: Euromonitor International [website]. London: Euromonitor International; 2018 (restricted access). 31. The Global Illicit Trade Environment Index. London: The Economist Intelligence Unit; 2018 (http:// illicittradeindex.eiu.com/documents/EIU%20Global%20Illicit%20Trade%20Environment%20Index%20 CHAP T ER 4. PO LI T I C AL ECO N OMY 245 2018%20-%20Overall%20Results%20White%20Paper%20June%206%20FINAL.pdf, accessed 1 February 2021). 32. Chaloupka FJ, Yürekli A, Fong GT. Tobacco taxes as a tobacco control strategy. Tob Control. 2012;21:172– 80 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/172.full.pdf, accessed 1 February 2021). 33. Chernick H, Merriman D. Using littered pack data to estimate cigarette tax avoidance in NYC. Natl Tax J. 2013;66(2):635–68 (https://www.ntanet.org/NTJ/66/3/ntj-v66n03p635-668-littered-pack-data- cigarette-tax.pdf, accessed 1 February 2021). 34. Liber AC, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia Survey. Tob Control. 2015;24:iii83–iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed 1 February 2021). 35. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 36. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 37. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 38. Paraje G, Araya D, Drope J. Illicit cigarette trade in metropolitan Santiago de Chile. Tob Control 2020;29:68–73. 39. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 40. Saenz de Miera Juarez BS, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020(0):1-7 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 1 February 2021). 41. Lee K, Collin J. ‘Key to the future’: British American Tobacco and cigarette smuggling in China. PLoS Medicine. 2006;3(7):e228 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1502159/pdf/pmed.0030228. pdf, accessed 2 February 2021). 42. Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38– i43 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i38.full.pdf, accessed 2 February 2021). 43. Aziani A, Kulick J, Norman N, Prieger JE. Empty discarded pack data and the prevalence of illicit trade in cigarettes; 2017. (https://ssrn.com/abstract=2906015, accessed 12 October 2020). 44. Philip Morris International. The Perspective of PM International on Smoking and Health Issues. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1985. Bates no. 2023268328-2023268349 (https://www.industrydocuments.ucsf.edu/tobacco/docs/#id=pxpb0040, accessed 5 February 2021). 45. Mandal S, Gilmore A, Collin J, Weishaar H, Smith K, McKee M. Block, amend, delay: tobacco industry efforts to influence the European Union’s Tobacco Products Directive (2001/37/EC). Smokefree Partnership/Cancer Research UK; 2009 (http://www.smokefreepartnership.eu/images/Reports-and- Position-Papers/EU_TI_TPD_report_May_2012.pdf, accessed 15 October 2020). 46. Bishop DJ. Key area paper: indirect taxation – British American Tobacco. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1994. Bates no. 503911771-503911780 (http:// bat.library.ucsf.edu/data/h/f/k/hfk11a99/hfk11a99.pdf, accessed 15 October 2020). 47. Zhou S, Liberman JD, Ricafort E. The impact of the WHO Framework Convention on Tobacco Control in defending legal challenges to tobacco control measures. Tob Control. 2009:28:s113-s118 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/28/Suppl_2/s113.full.pdf, accessed 2 February 2021). 48. Gilmore AB, Fooks G, Drope J, Bialous SA, Jackson RR. Exposing and addressing tobacco industry 246 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N conduct in low-income and middle-income countries. Lancet. 2015;385:1029–43. 49. Ulucanlar S, Fooks G, Gilmore A. The policy dystopia model: an interpretive analysis of tobacco industry political activity. PLoS Med. 2016:13(9):e1002125 (https://journals.plos.org/plosmedicine/ article?id=10.1371/journal.pmed.1002125, accessed 2 February 2021). 50. Apollonio D, Glantz S. Tobacco manufacturer lobbying to undercut minimum price laws: an analysis of internal industry documents. Tob Control. 2020;29:e10-e17 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC7374022/pdf/nihms-1590806.pdf, accessed 2 February 2021). 51. Crosbie E, Glantz S. Tobacco industry argues domestic trademark laws and international treaties preclude cigarette health warning labels, despite consistent legal advice that the argument is invalid. Tob Control. 2012;23:e7 (https://www.researchgate.net/publication/233769722_Tobacco_industry_ argues_domestic_trademark_laws_and_international_treaties_preclude_cigarette_health_warning_ labels_despite_consistent_legal_advice_that_the_argument_is_invalid, accessed 2 February 2021). 52. Puig S. Tobacco litigation in international courts. Harv Int Law J. 2016;57(2):1–68. 53. Crooked nine: nine ways the tobacco industry undermines health policy. New York: Vital Strategies; 2019 (https://exposetobacco.org/wp-content/uploads/2019/09/Crooked-9-STOP.pdf, accessed 2 February 2021). 54. Balwicki L, Stokłosa M, Balwicka-Szczyrba M, Tomczak W. Tobacco industry interference with tobacco control policies in Poland: legal aspects and industry practices. Tob Control. 2016;25(5):521–26 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/521.full.pdf, accessed 2 February 2021). 55. Sud SR, Brenner JE, Shaffer ER. Trading away health: the influence of trade policy on youth tobacco control. J Pediatrics. 2015;166(5):1303–07 (https://www.jpeds.com/action/showPdf?pii =S0022-3476%2815%2900017-7, accessed 2 February 2021). 56. Lee S, Ling PM, Glantz SA. The vector of the tobacco epidemic: tobacco industry practices in low and middle-income countries. Cancer Cause Control. 2012;23:117–29 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC3332051/pdf/nihms-367996.pdf, accessed 2 February 2021). 57. Zhou S, Liberman J. The global tobacco epidemic and the WHO Framework Convention on Tobacco Control—the contributions of the WHO’s first convention to global health law and governance. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018 (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3339718, accessed 18 February 2021). 58. Legislative Consultation with Constitutional Division of the Supreme Court, No. 2012-003918. Constitutional Division of the Supreme Court of Costa Rica; 20 March 2012 (https://www. tobaccocontrollaws.org/files/live/litigation/830/CR_Legislative%20Consultation%20with%20.pdf, accessed 2 February 2021). 59. British American Tobacco Kenya Ltd v Ministry of Health, Petition No. 143 of 2015. High Court of Kenya; 2016 (https://www.tobaccocontrollaws.org/files/live/litigation/2643/KE_British%20American%20 Tobacco%20Kenya.pdf, accessed 2 February 2021). 60. British American Tobacco v Ministry of Health, et al, Civil Appeal No. 112 of 2016. Kenyan Court of Appeal; 2017 (https://www.tobaccocontrollaws.org/files/live/litigation/2521/KE_British%20 American%20Tobacco%20Ltd%20v.pdf, accessed 2 February 2021). 61. Thailand – customs and fiscal measures on cigarettes from the Philippines: Report of the Panel. Geneva: World Trade Organization; 2010 (WT/DS371/R; https://docs.wto.org/dol2fe/Pages/SS/directdoc. aspx?filename=Q:/WT/DS/371R.pdf&Open=True, accessed 2 February 2021). 62. Tribunal Constitucional de Chile [Constitutional Court of Chile] (1995 Rol No. 219-95; https://www. tribunalconstitucional.cl/descargar_sentencia2.php?id=3074, accessed 4 April 2020.) 63. McGrady B. International investment law. In: Voon T, Mitchell AD, Liberman J, editors. Regulating tobacco, alcohol and unhealthy foods: the legal issues. New York: Routledge; 2014. 64. Marvin Roy Feldman Karpa v United Mexican States, Case No. ARB(AF)/99/1 International Centre for Settlement of Investment Disputes; 2002 (https://www.italaw.com/sites/default/files/case-documents/ ita0319.pdf, accessed 2 February 2021). 65. Feldman Karpa v Mexico (2002 ICSID Arbitral Tribunal) (https://www.italaw.com/sites/default/files/ case-documents/ita0319.pdf, accessed 18 February 2021). 66. Undang-Undang Nomor 11 Tahun 1995 tentangCukai (Lembaran Negara Republik Indonesia Tahun1995 CHAP T ER 4. PO LI T I C AL ECO N OMY 247 Nomor 76) (Excise Law,in Indonesian). Article 5. 67. Excise Regulation (2011) Reference No. PMK Number 167/PMK.011/2011 (in Indonesian) (https:// peraturan.bpk.go.id/Home/Details/145897/pmk-no-167pmk0112011, accessed 3 February 2021). 68. Mahkamah Agung. Indonesia Supreme Court, Case No. 73/PDT/2018/PT SBY (https://putusan3. mahkamahagung.go.id/direktori/putusan/74efff9cfe18d60990fcde19583ecd5f.html, accessed 6 February 2021). 69. Maya D. Analisis kebijakan kenaikam tarif cukai dan penyesuaian batasan harga jual eceran hasil tembakau [Policy analysis on excise tax increases and price limit adjustments retail tobacco products]; 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-09/S45937-Deacy%20Maya accessed 3 March 2020). 70. Indriawati EN, Susanti L, Qurnia MA. Kebijakan tarif cukai hasil tembakau terhadap perekonomian [Tobacco excise tariff policy on the economy]; 2014 (in Indonesian). (https://www.academia. edu/6611977/MAKALAH_CHT_kel_6, accessed 15 October 2020). 71. Arroyani DW. Evaluasi Kebijakan Penetapan Tarif Cukai Hasil Tembakau Berdasarkan PMK [Evaluation of Tobacco Excise Tariff Determination Policy Based on PMK].167/PMK.011/2011. 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-08/S44664-Dwira%20Wanti%20Arroyan, accessed 15 October 2020). 72. Ngo Ngo Ha and Another v. State of New South Wales and Others (Matter No S 45 of 1996); Walter Hammond & Associates Pty Ltd v. State of New South Wales and Others (Matter No S 165 of 1996). Constitutional law; (1997) 146 ALR 355: 5 August 1997 (https://staging.hcourt.gov.au/assets/ publications/judgments/1997/029--NGO_NGO_HA_AND_ANOR_v_STATE_OF_NEW_SOUTH_WALES_&_ ORS_(Matter_No_S_45_of_1996);_WALTER_HAMMOND--(1997)_146_ALR_355.html, accessed 2 February 2021). 73. McGrady B. Trade and public health: the WTO, tobacco, alcohol, and diet. New York: Cambridge University Press; 2011. 74. Giorgetti C. Health and international investment law. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018. 75. British American Tobacco v Attorney General and Center for Human Rights and Development. Constitutional Court of Uganda Petition No.46 of 2016; 2017 (https://media.ulii.org/files/judgments/ ugcc/2019/4/2019-ugcc-4_0.pdf, accessed 18 February 2021). 76. Lim CL, Ho J, Paparinskis M. International investment law and arbitration: commentary, awards and other materials. Cambridge: Cambridge University Press; 2018. 77. Dominican Republic – measures affecting the importation and internal sale of cigarettes : Report of the Panel. Geneva: World Trade Organization; 2004 (WT/DS302/R; https://docs.wto.org/dol2fe/ Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302R.pdf&Open=True, accessed 21 February 2021) 78. Dominican Republic — measures affecting the importation and internal sale of cigarettes: Report of the Appellate Body. Geneva: World Trade Organization; 2005 (WT/DS302/AB/R; https://docs.wto. org/dol2fe/Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302ABR.pdf&Open=True, accessed 21 February 2021). 79. Gilmore A. Tobacco and transition: understanding the impact of transition on tobacco use and control in the former Soviet Union [Thesis]. London: London School of Hygiene and Tropical Medicine, June 2005 (https://escholarship.org/uc/item/3rw2c04c, accessed 2 February 2021). 80. MacKenzie R, Collin J, Sopharo C, Sopheap Y. ‘Almost a role model of what we would like to do everywhere’: British American Tobacco in Cambodia. Tob Control. 2004;13(suppl II): ii112–ii117 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/13/suppl_2/ii112.full.pdf, accessed 2 February 2021). 81. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tobacco Control 2013;22(2):144-53 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 21 February 2021). 82. Price and tax. In: TobaccoTactics [website]. Bath: University of Bath; 2020 (https://tobaccotactics.org/ topics/price-and-tax/, accessed 15 December 2020). 83. Vulovic V, Chaloupka FJ. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation—commentary. Tob Control. 24 June 2020. 248 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 84. Verguet S, Kearns PKA, Rees VW. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation. Tob Control. 23 June 2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/08/11/tobaccocontrol-2019-055315. full.pdf, accessed 2 February 2021).. 85. Palipudi KM, Gupta PC, Sinha DN, Andes LJ, Asma S, McAfee T. Social determinants of health and tobacco use in thirteen low and middle income countries: evidence from Global Adult Tobacco Survey. PLoS ONE. 2012;7(3):e33466 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3306395/pdf/ pone.0033466.pdf, accessed 2 February 2021). 86. Systematic review of the link between tobacco and poverty. Geneva: World Health Organization; 2014 (http://www.who.int/tobacco/publications/economics/9789241507820/en, accessed 21 October 2020). 87. Goodchild M, Sinha P, Gill Munish V, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/bitstream/handle/10665/331836/whoseajph2020v9i1p73. pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 88. Faruque GM, Ahmed M, Huq I, Parven R, Wadood SN, Chowdhury SR, et al. The economic cost of tobacco use in Bangladesh: a health cost approach. Dhaka: Bangladesh Cancer Society; 2020 (https://www. researchgate.net/publication/342445467_The_Economic_Cost_of_Tobacco_Use_in_Bangladesh_A_ Health_Cost_Approach, accessed 2 February 2021). 89. John RM, Sung H, Max WB, Ross H. Counting 15 million more poor in India, thanks to tobacco. Tob Control. 2011;20:349–52 (https://www.researchgate.net/publication/49809290_Counting_15_million_ more_poor_in_India_thanks_to_tobacco, accessed 2 February 2021). 90. Liu Y, Rao K, Hu TW, Sun Q, Mao Z. Cigarette smoking and poverty in China. Soc Sci Med. 2006;Dec;63(11):2784–90. 91. Fuchs A, González Icaza F, Paz D. Distributional effects of tobacco taxation – a comparative analysis. Washington (DC): World Bank Group; 2019 (Policy research working paper 8805; https://openknowledge. worldbank.org/bitstream/handle/10986/31534/WPS8805.pdf?sequence=4&isAllowed=y, accessed 2 February 2021). 92. Smith CE, Hill SE, Amos A. Impact of population tobacco control interventions on socioeconomic inequalities in smoking: a systematic review and appraisal of future research directions. Tob Control. 29 September 2020 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/09/28/ tobaccocontrol-2020-055874.full.pdf, accessed 2 February 2021). 93. Furman J. Six lessons from the U.S. experience with tobacco taxes – World Bank conference “Winning the tax wars: global solutions for developing countries”. Washington (DC): Obama White House Archives; 24 May 2016 (https://obamawhitehouse.archives.gov/sites/default/files/page/files/20160524_cea_ tobacco_tax_speech.pdf, accessed 21 October 2020). 94. Önder Z, Yürekli A. Who pays the most cigarette tax in Turkey. Tob Control. 2016;25(1):39–45. 95. Verguet S, Gauvreau CL, Mishra S, MacLennan M, Murphy SM, Brouwer ED, et al. The consequences of tobacco tax on household health and finances in rich and poor smokers in China: an extended cost-effectiveness analysis. Lancet Glob Health. 2015;3:e206–16 (https://www.thelancet.com/action/ showPdf?pii=S2214-109X%2815%2970095-1, accessed 2 February 2021). 96. Fuchs Tarlovsky A, González Icaza F. The welfare and distributional effects of increasing taxes on tobacco in Vietnam. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/ curated/en/758391561477648302/pdf/The-Welfare-and-Distributional-Effects-of-Increasing-Taxes- on-Tobacco-in-Vietnam.pdf, accessed 2 February 2021). 97. Expanding the global tax base: taxing to promote public goods: tobacco taxes. Washington (DC): World Bank Group; 2016 (http://pubdocs.worldbank.org/en/799981480947591462/Taxing-to-Promote- Public-Goods-tobacco-taxation-chapter-conference-25-nov-2016-final-version-pmarquez-002.pdf, accessed 2 February 2021). 98. Blecher E. Increasing tobacco taxes significantly will increase revenues. Chicago: University of Illinois at Chicago; 2018. (Tobacconomics policy brief; https://tobacconomics.org/files/research/470/Laffer- Curve-Policy-Brief_v2.2.pdf, accessed 2 February 2021). 99. Savedoff W, Lopert R. Sugar, rum, and tobacco: domestic resource mobilization for low-income countries through excise taxes. Washington (DC): Centre for Global Development; 2020 (CGD Policy Paper CHAP T ER 4. PO LI T I C AL ECO N OMY 249 164, https://www.cgdev.org/sites/default/files/PP164-Savedoff-Health-Taxes-DRM.pdf, accessed 2 February 2021). 100. Lavado R, Postolovska I, Sotomayor RE. Beyond the perceived regressivity of higher tobacco taxes, turning short-term losers into long-term winners? In: Tobacco reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 101. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 102. Tonga NCD related taxation policy assessment preliminary findings. Presentation made at the innovations and actions against NCDs conference ADB headquarters, Manila, 17 July 2018. 103. Moreno-Dodson B. Raising tobacco taxes: why, what, and how: where public health and economics converge. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e- library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 104. Fuchs A, Del Carmen G, Mukong AK. Long-run impacts of increasing tobacco taxes: evidence from South Africa. Policy Research Working Paper 8369. Washington [DC]: World Bank; 2018 (https://openknowledge. worldbank.org/bitstream/handle/10986/29497/WPS8369.pdf?sequence=4&isAllowed=y, accessed 18 February 2021) . 105. van Walbeek C. Case studies in illicit tobacco trade: South Africa. Tobacconomics fact sheet. Chicago: University of Chicago, Tobacconomics, Health Policy Center; 2020 (https://tobacconomics.org/files/ research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 106. van Loggerenberg, J. Tobacco Wars. Cape Town: NB Publishers; 2019. 107. Marquez P. Global tobacco control: a development priority. In: Tobacco tax reform at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads. pdf, accessed 2 February 2021). 108. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington [DC]: World Bank; 2017 (https:// tobacconomics.org/files/research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 109. Scollo M, Bayly M. Revenue from tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia. org.au/chapter-13-taxation/13-6-revenue-from-tobacco-taxes-in-australia, accessed 21 October 2020). 110. Scollo M, Bayly M. Tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia.org.au/ chapter-13-taxation/13-2-tobacco-taxes-in-australia, accessed 21 October 2020). 111. Breton E, Richard L, Gagnon F, Jacques M, Bergeron P. Fighting a tobacco-tax rollback: a political analysis of the 1994 cigarette contraband crisis in Canada. Journal of Pub Health Policy. 2006;27:77–99. 112. Fanta E, Garcia M. Preventing and controlling illicit tobacco flows. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB- Report-CrossRoads.pdf, accessed 2 February 2021). 113. Goodchild M, Perucic AM, Zheng R, Blecher E, Paul J. The health impact of raising tobacco taxes in developing countries. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/ taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 114. Ross H. Tracking and tracing tobacco products in Kenya. Prev Med. 2017;105:S15–S18 (https://reader. elsevier.com/reader/sd/pii/S0091743517301457?token=424D9EC1D42A9C69BDB55B90090C3F7853AA 20A4970D0E379C91EE64C73A273A0FA949557F4F120A0B636701581EC055, accessed 2 February 2021). 115. Mamudu H, Hammond R, Glantz S. Tobacco industry attempts to counter the world bank report curbing the epidemic and obstruct the WHO framework convention on tobacco control. Soc Sci & 250 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Med. 2008;67(11):1690–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2662513/pdf/nihms81164. pdf, accessed 2 February 2021). 116. Gilmore A. Understanding the vector in order to plan effective tobacco control policies: an analysis of contemporary tobacco industry materials. Tob Control. 2012;21:119–26 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/21/2/119.full.pdf, accessed 2 February 2021). 117. The economics of tobacco taxation and employment in Indonesia: policy implications technical brief. Washington (DC): World Bank Group; 2018 (https://openknowledge.worldbank.org/bitstream/ handle/10986/29814/126158-WP-P154568-PUBLIC-add-series-WBGPNIndoEmploymentUPDATEFin alWeb.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 118. Sahadewo GA, Drope J, Li Q, Nargis N, Witoelar F. Tobacco or not tobacco: predicting farming households’ income in Indonesia. Tob Control. 2020. 119. The economics of tobacco farming in Indonesia: health, population, and nutrition global practice. Washington (DC): World Bank Group, 2017 (http://documents1.worldbank.org/curated/ en/161981507529328872/pdf/120307-REVISED-PUBLIC-WBGIndoEconomicsTobaccoFarming.pdf, accessed 2 February 2021). 120. The revenue and employment outcome of biri taxation in Bangladesh. Dhaka: National Board of Revenue, Government of the People’s Republic of Bangladesh; 2019 (https://www.who.int/docs/ default-source/searo/bangladesh/pdf-reports/cat-2/biri-study-report-03-12-2019.pdf?sfvrsn=b8fea69_2, accessed 2 February 2021). 121. Magati P, Lencucha R, Li Q, Drope J, Labonte R, Appau A, et al. Costs, contracts and the narrative of prosperity: an economic analysis of smallholder tobacco farming livelihoods in Kenya. Tob Control. 2019;28(3):268–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6512316/pdf/nihms-1012064. pdf, accessed 2 February 2021). 122. Ghaus K, Iqbal MA, Sabir M, Saleem W, Ali A, Aamir N. Macroeconomic impacts of tobacco use in Pakistan. Karachi: The Social Policy and Development Centre; 2018 (https://tobacconomics.org/files/ research/483/Macroeconomic-impact-of-Tobacco-in-Pakistan.pdf, accessed 2 February 2021). 123. Ross H. The effects of tobacco taxes on employment and labor productivity. In: Tobacco tax reform: at the crossroads of health and development. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021) 124. Bigawanto M. Tobacco industry interference undermined tobacco tax policy in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance (SEATCA); 2019 (https://seatca.org/dmdocuments/ Indonesia%20TII%20in%20Tax.pdf, accessed 2 February 2021). 125. Ahsan A, Wyono N. Analysis of the impact of higher cigarettes prices on employment in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance; 2007 (https://seatca.org/dmdocuments/An%20 Analysis%20of%20the%20Impact%20of%20Higher%20Cigarette%20Prices%20on%20Employment%20 in%20Indonesia.pdf, accessed 2 February 2021). 126. Higashi H, Khuong TA, Ngo AD, Hill PS. Evidence and decision making: tobacco control policy and legislation in Vietnam. Int J of Health Plann. 2013;2(1):e72–e94. 127. Buck D, Raw M, Godfrey CA, Sutton M. Tobacco and jobs: the impact of reducing consumption on employment in the UK. York: University of York; 1995 (https://www.researchgate.net/ publication/4814523_Tobacco_and_jobs_the_impact_of_reducing_consumption_on_employment_ in_the_UK, accessed 2 February 2021). 128. Petkova R, Ileva R and Rusenov A. Tobacco control policies and tobacco industry employment in Bulgaria. In: Economic, social and health issues in tobacco control. Report of a WHO international meeting, Kobe, Japan, 3-4 December 2001. Geneva: World Health Organization; 2003. 129. Jha V, Narayanan BG, Wadhwa D, Tesche J. Economic and environmental effects of reduction in smoking prevalence in Tanzania. Tob Control; 2020;29:24–8 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/29/1/24.full.pdf, accessed 2 February 2021). 130. Lencucha R, Makoka D, Drope J, Appau A, Labonté R, Li Q, et al. Costs, revenues and profits: an economic analysis of smallholder tobacco farmer livelihoods in Malawi. Tob Control. 2017;26(6)634–40 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC5957074/pdf/nihms965842.pdf, accessed 2 February 2021). 131. Goma FM, Labonté R, Drope J, Li Q, Zulu R, Kangwa E. The economics of tobacco farming in Zambia: CHAP T ER 4. PO LI T I C AL ECO N OMY 251 tobacco farmers survey report. Lusaka: University of Zambia and American Cancer Society; 2019 (https://www.cancer.org/content/dam/cancer-org/research/economic-and-healthy-policy/zambia- report-7-16-19.pdf, accessed 28 October 2020). 132. Chavez JJ, Drope J, Li Q, Aloria MJ. The economics of tobacco farming in the Philippines. Quezon City: Action for Economic Reforms and Atlanta: American Cancer Society, 2016 (https://www.cancer. org/content/dam/cancer-org/research/economic-and-healthy-policy/the-economics-of-tobacco- farming-in-the-philippines.pdf, accessed 22 October 2020). 133. Augstin CP, Cardenas RP, Cortez JB, Gimongala GC, Ramos AR. The effects of the sin tax reform law of 2012 to tobacco farmers of Amulung, Cagayan. IJARMSS. 2017;6(12):57–75. 134. Mercado A. Tobacco tax: good for the tobacco farmers, too. Business World; 2019 [website] (https:// www.bworldonline.com/tobacco-tax-good-for-the-tobacco-farmers-too/, accessed 25 October 2020). 135. Yürekli A, Önder Z, Elibol M, Erk N, Cabuk A, Fisunoglu M, et al. The economics of tobacco and tobacco taxation in Turkey. Paris: International Union Against Tuberculosis and Lung Disease; 2010 (https:// www.who.int/tobacco/en_tfi_turkey_report_feb2011.pdf, accessed 2 February 2021). 136. Latif E, Warner V. Legislation and smoking prevention. In: Loddenkemper R, Kreuter M, editors. The Tobacco Epidemic: progress in Respiratory Research. 2015;42:136–148. 137. Tandon A, Cashin C. Assessing public expenditure on health from a fiscal space perspective. Washington (DC): World Bank Group; 2010 (http://documents1.worldbank.org/curated/en/333671468330890417/ pdf/560530WP0Box341penditureFiscalSpace.pdf, accessed 2 February 2021). 138. Cashin C, Sparkes S, Bloom D. Earmarking for health: from theory to practice. Geneva: World Health Organization; 2018 (https://apps.who.int/iris/bitstream/handle/10665/255004/9789241512206-eng. pdf?sequence=1, accessed 2 February 2021). 139. Wilkinson M. Paying for public spending: is there a role for earmarked taxes? Fiscal Studies. 1994;15(4):119–35 (https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1475-5890.1994.tb00213.x, accessed 2 February 2021). 140. ITAF: impôts et taxes affectées à la protection sociale. Les impôts et taxes affectés (ITAF) [taxes and duties allocated to social protection]. Vie publique: au coeur du débat public [website]. Paris: Directorate of Legal and Administrative Information; 2018 (in French) (https://www.vie-publique.fr/fiches/22010- impots-et-taxes-affectees-itaf-protection-sociale, accessed 31 October 2020). 141. Seventeenth congress of the Philippines, Metro Manila, RA 11346, 23 July 2018 (https://www. officialgazette.gov.ph/downloads/2019/07jul/20190725-RA-11346-RRD.pdf, accessed 31 October 2020). 142. McCleary W. The earmarking of government revenue: a review of some World Bank experience. World Bank Res Obs. 1991;6(1):81–104 (http://documents1.worldbank.org/curated/en/109871468764672767/ pdf/multi-page.pdf, accessed 2 February 2021). 143. Strategic investment of tobacco tax revenue. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/global/pdfs/en/strategic_investment_tobacco_tax_revenue. pdf, accessed 31 October 2020). 144. Bialous SA, Glantz SA. Arizona’s tobacco control initiative illustrates the need for continuing oversight by tobacco control advocates. Tob Control. 1999;8:141–51 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/8/2/141.full.pdf, accessed 2 February 2021). 145. Goldman LK, Glantz SA. The passage and initial implementation of Oregon’s measure 44. Tob Control. 1999;8:311–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1763939/pdf/v008p00311.pdf, accessed 2 February 2021). 146. Heiser PF, Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health. 1997;87(6):968–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380932/pdf/amjph00505-0082. pdf, accessed 2 February 2021). 147. Landman A, Bialick P. Tobacco industry involvement in Colorado. Chicago: American Lung Cancer Association; 2004 (https://escholarship.org/uc/item/9d880448#main, accessed 18 February 2021). 148. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010. (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 149. Vardavas CI, Filippidis FT, Agaku IT, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the 252 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N importance of earmarking the revenue to health care and tobacco control. Tob Induc Dis. 2012;10:21 (http://www.tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the- revenue-to-health-care-and-tobacco,66110,0,2.html, accessed 31 October 2020). 150. Bird RM. Tobacco and alcohol excise taxes for improving public health and revenue outcomes: marrying sin and virtue?. Washington (DC): World Bank Group; 2015 (Policy Research Working Paper 7500; https://openknowledge.worldbank.org/bitstream/handle/10986/23464/ Tobacco0and0al0ing0sin0and0virtue00.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 151. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 152. The establishment and use of dedicated taxes for health. Manila: World Health Organization Western Pacific Regional Office; 2004 (https://www.who.int/health_financing/documents/dedicated_taxes. pdf?ua=1, accessed 2 February 2021). 153. Teja RS. The case for earmarked taxes: theory and example. Washington (DC): International Monetary Fund; 1988 (IMF Working Paper; https://www.elibrary.imf.org/doc/IMF001/14514- 9781451922677/14514-9781451922677/Other_formats/Source_PDF/14514-9781455212002.pdf, accessed 2 February 2021). 154. Keen M. Taxation and development – again. Washington (DC): International Monetary Fund; 2012(IMF Working Paper WP/12/220O; (https://www.imf.org/external/pubs/ft/wp/2012/wp12220.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 253 ANNEX 4.1 METHODS TO ASSESS THE NATURE AND SIZE OF THE ILLICIT TOBACCO TRADE A4.1 DIRECT MEASUREMENT A4.1.1 SMOKER INTERCEPT AND PACK OBSERVATION SURVEYS Illicit trade can be measured directly by examining the cigarette packs of smokers. The smokers themselves can provide information on purchasing patterns, brand preferences and prices paid. Researchers can select individuals or retailers to survey based on a convenience sample (i.e. a sample that may not be representative) or a probability-based sample (i.e. a sample selected to be statistically representative of an underlying population). Data collected from a pack could reveal whether the pack is compliant or non- compliant with the local tax laws. Information can be obtained from objective markings such as brands, public health warning labels, tax stamps, foreign language labels or duty-free labels. During these stops, researchers can record demographic information (e.g. age and gender of the smoker), smoking-related history (e.g. number of cigarettes smoked per day) and price information. This is helpful in understand- ing the profile of smokers who are able and willing to avoid cigarette taxes. Pack observations can be used in conjunction with population-based household surveys to obtain population-based estimates of the illicit tobacco trade (1,2). For example, as part of a regular national health survey, Kaplan et al. conducted a cross-sectional study of smokers in Turkey, using a face-to-face interviewer-administered survey and pack observation (3). They were able to collect sociodemographic, lifestyle and medical details along with pack observations as part of the study protocol. Advantages and disadvantages of smoker intercept and pack observation surveys A primary advantage of conducting pack observation is that it is direct and objective, and smokers are not subject to any value judgements (2). Paired with survey data, pack observation can appropriately account for respondents who are not residents of the area in which they are surveyed (4). Disadvantages include the difficulty of identifying areas that are representative of the tobacco use population and the difficulty of sampling important subpopulations such as elderly and immobile smok- ers. Also, surveys conducted in the daytime may discount the number of youthful smokers who are in school. Another disadvantage is that a sizeable number of smokers may refuse to show their last-purchased pack (2). Kaplan et al. found that 24% of smokers sampled in Turkey did not show their cigarette pack to the study interviewer (3). This issue may be mitigated by asking users to provide information on the brand purchased, whether any public health warnings were posted and the 254 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N price paid (5). Although the responses are based on recall, they may still yield useful information. For example, Joossens et al. allowed smokers who did not show their packs to provide self-reported information and found no statistical differences in illicit packs between these respondents and those who did show their packs (2). Another obstacle to accurate measurement based on smoker intercepts is the inability to determine the tax payment of smokers who purchase single cigarettes, since these smokers are generally not given packs. However, information can still be captured in self-report surveys by asking smokers to report the brand purchased and price paid. Key study for readers to refer to for additional guidance: Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–447. A4.1.2 PACK RETURN AND SWAP SURVEYS Pack return and pack swap surveys fall within the broader category of pack ob- servation studies that use survey sampling techniques to examine smokers’ pack characteristics and to determine whether they are tax compliant. For these surveys, the unit of analysis is the individual. The main differences between pack swap and pack return surveys is that swap surveys offer the smoker a replacement pack, whereas pack returns are built into mail surveys and allow respondents to mail in their unopened packs. Pack swap and pack return surveys use probability and nonprobability sampling procedures. Probability sampling allows researchers to generalize to the broader population. Governments can use this method to rapidly assess the availability of illicit products in a given geographic area or to measure the share of the illicit market- place. Rapid assessment may be performed in instances where there is an emerging tobacco product (e.g. a new cheap white brand) or suspected counterfeiting of tax stamp features. Rapid assessment using a convenience sampling strategy could place researchers near busy intersections where they could ask smokers for permission to look at their cigarette packs or to take photographs that could be analysed later. A population-based study requires a sample that closely mirrors the tobacco use population. Advantages and disadvantages of pack return and pack swap surveys Pack swap and pack return surveys may help to overcome the stigma associated with traditional smoking surveys. For example, when researchers ask smokers to see their cigarette packs (or when they take photographs), no value judgements are made. These surveys are good rapid-assessment tools that can be used to examine CHAP T ER 4. PO LI T I C AL ECO N OMY 255 the effectiveness of physical features of a pack designed to deter illicit trade (e.g. packs that have tracking and tracing technology or high-tech stamps). In addi- tion, they can be supplemented with population-based tobacco use surveys. When coupled with such survey data, these methods allow researchers to obtain relevant information about the context of illicit purchases, including, for example, sources (e.g. street, peer networks, retail stores) and prices. Mail-in surveys are filled out in the comfort of the respondent’s home without the presence of family members or passers-by, which may assure them that responses will be kept confidential. A potential disadvantage is that smokers who purchase both illicit cigarettes and tax-paid cigarettes may disproportionately mail back compliant packs. In addition, in LMICs, this mode of survey distribution may be unreliable because of issues associated with mail delivery systems. Key study for readers to refer to for additional guidance: Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61-66. A4.1.3 LITTERED-PACK SURVEYS Littered-pack surveys, also known as empty discarded pack surveys, are used pre- dominantly in high- and middle-income countries (e.g. the United States, France, Canada, New Zealand, Mexico and Poland). This unobtrusive method relies on the premise that smokers publicly discard packs (e.g. on streets, sidewalks and in public trash cans). The packs bear characteristics that illustrate whether they are tax compliant (e.g. tax stamps, health warnings). For example, an Albanian health warning label on a cigarette pack discarded in Greece provides evidence that the pack was destined for the Albanian market. The pack may have been smuggled into Greece by criminal entrepreneurs or it may have been brought by a visitor. Collecting discarded packs from a representative geographic sample and examining these characteristics can provide estimates of tax compliance. Operationally, this data collection method uses an ecological approach whereby geographies are the units of analysis. Geographical units are meant to represent the smokers in the city/ country and can be administratively defined (e.g. by the country’s census bureau or transit zones) or may reflect researcher-defined neighbourhoods (e.g. half-mile buffer zones near bus stops or activity spaces). Researchers in Canada have expanded the littered-pack method to include col- lection and analysis of cigarette butts on 25 postsecondary campuses. The cigarette butts provide information on the brands sold (or lack thereof) and allow researchers to distinguish between legal and illegal products (6). A recent innovative expansion 256 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the littered-pack methodology is the collection of packs from cigarette retailers. John and Ross collected empty packs of tobacco products from a sample of registered and unregistered retailers in India (7). Collecting packs from retailers was relevant given that single cigarettes dominate the illicit market in India. Smokers who buy single cigarettes would be unable to provide a pack in a pack swap or street intercept survey, so collecting littered packs from the ground would undercount sales of single cigarettes. The feasibility of this method is dependent on the relationship between researchers and retailers (enhanced trust) and the efforts taken to ensure confidentiality. In some countries, retailer compliance with this research method might be strained because of concerns regarding confidentiality, since retailers may face criminal and civil penalties, depending on the research findings. Advantages and disadvantages of littered-pack surveys Littered-pack surveys are generally advantageous for governments because they facilitate comparison with industry estimates. This is one of the most-preferred methods because it yields estimates that are less likely to be biased due to issues of social desirability, recall error and confidentiality that plague survey research, and they are much less expensive than face-to-face interviews used in smoker intercept or household surveys. However, there are some issues regarding these surveys, espe- cially in high-income countries, including the inability to differentiate between tax avoidance and tax evasion (8). For example, a pack in Berlin that bears a Vietnamese tax stamp may have been smuggled in mass quantity or brought in by a temporary visitor. Researchers have circumvented this issue and broadened the umbrella to measure cigarette tax noncompliance considering the potential biases introduced by tourism. Another disadvantage of littered-pack surveys is that larger budgets are needed to employ field researchers to collect, code and analyse the data. Not all countries employ tax stamps on their cigarette packaging, which may make it difficult to measure tax compliance. These surveys also can underestimate the markets in low-income countries such as India, where the main item of illegal trade is single cigarettes (7). The surveys may also overestimate illicit trade if littering behaviour is correlated with willingness to engage in illicit trade. Finally, littered-pack surveys and butt collections provide information on the proportion of butts and packs that are illegal, not the proportion of smokers that purchase illegal cigarettes (6). Key studies for readers to refer to for additional guidance: Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38–i43. Merriman D. The micro-geography of tax avoidance: evidence from littered cigarette CHAP T ER 4. PO LI T I C AL ECO N OMY 257 packs in Chicago. Am Econ J Econ Policy. 2010;2(2):61–84. Stoklosa M., Paraje G., Blecher E., A Toolkit on Measuring Illicit Trade in Tobacco Products. A Tobacconomics and American Cancer Society Toolkit. Chicago, IL:Tobacconomics, Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/621/uic-illicit-trade-tool-kit-eng-v2.0-2. pdf, accessed 18 February 2021). A4.1.4 SELF-REPORT POPULATION SURVEYS Self-report surveys, when distributed to a representative sample of the population, can provide meaningful data on the prevalence of tax noncompliance. The surveys can be distributed to individuals or households in various ways, including face-to-face, telephone, mail and internet. Questions that specifically address illicit purchases can be added as supplementary questions to existing health or tobacco surveys. Some countries include such questions in their adult and youth tobacco surveys to estimate tax evasion/avoidance. For example, Canada’s annual Youth Smoking Survey asks smokers about the frequency of their purchases of First Nations/Native brand cigarettes (9-10). Davis et al. used data from the New York Adult Tobacco Survey to measure the source of purchase of the last cigarette pack purchased (i.e. Native American Reservations, lower-tax neighbouring states or countries, toll-free telephone numbers, the internet, duty-free shops) and the price paid (11). Twenty- eight nations currently use surveys to measure tax noncompliance as part of the ITC Project (12). Similar analyses can be conducted using questions from the Global Adult Tobacco Use Surveys. For example, Iglesias et al. used the Brazil Global Adult Tobacco Use Surveys to compare self-reported prices with a defined threshold retail price to estimate the proportion of illicit cigarette use among smokers in Brazil (13). Countries are encouraged to use existing global health surveys or to incorporate similar types of questions pertaining to illicit trade in their annual health surveys. Asking respondents about price paid per pack (including taxes), brand name and location where cigarettes were purchased (e.g. duty free shop, unlicensed vendor, internet) can contribute to a better understanding of the illicit tobacco trade. Advantages and disadvantages of self-report surveys Self-report surveys can be repeated over time to measure purchasing trends and progress associated with increases in cigarette taxation. Well-designed surveys can also provide generalizable estimates at the national level. Depending on the size of the sample, a self-report survey can provide comparable data across geographies that can help governments target resources. For example, findings that illicit cigarettes are more common in urban areas could lead to additional education campaigns and targeted enforcement. 258 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Limitations of self-report surveys include the possibility of bias due to the social stigma associated with participating in the illicit trade, which could lead survey respondents to underreport participation. Additionally, surveys may be unable to gauge whether individuals are associated with tax avoidance versus tax evasion. Finally, there is evidence that self-report household surveys may underrepresent smokers. Key studies for readers to refer to for additional guidance: Callaghan RC, Veldhuizen S, Ip D. Contraband cigarette consumption among adolescent daily smokers in Ontario, Canada. Tob Control. 2011;20(2):173–174. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany (NY): New York State Department of Health, Tobacco Control Program; 2006. A4.1.5 COVERT-PURCHASES SURVEYS A number of studies in high-, middle- and low-income countries use covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes in public and semi-private spaces (14-17). This method is also used by the tobacco industry in the United States to identify retailers who sell counterfeit cigarettes (18). Covert-purchases surveys do not provide estimates of the size of the illicit trade (i.e. market volume). Instead, they serve as a surveillance tool to identify where illicit cigarettes are sold and the extent to which they have infiltrated legal businesses. For example, a covert-purchases survey can examine whether illicit cigarettes are sold through legal retailers. It can also be used to measure compliance with emerging tobacco control policies that focus on, for example, product standardization or new regulations on flavours (e.g. plain packaging or bans on flavoured tobacco products). Covert-purchases surveys use trained researchers to visit a selected sample of retailers and directly purchase or inquire about the availability of illicit tobacco products. Retailers are not informed about the goals of the studies. Methods for determining the availability of illicit product vary. For example, in some studies, covert buyers do not directly inquire about illicit products. Instead, they purchase packs of tobacco products, paying full price, to determine whether retailers are selling illicit products under the guise that they are licit (14). The research team then examines the packs to determine whether they are legal. In the United States, researchers have observed that some consumers are paying full price for illicit untaxed packs smuggled from lower-tax states (14). Other research protocols directly ask retailers for illicit products (14,16-17). In Guatemala, Arevalo et al. specifically asked retailers for “imported cigarettes” (17). The ways covert buyers ask for illicit products may also vary geographically. For example, in some countries covert buyers may ask for “cheaper” packs or for illicit whites such as Jin Ling. CHAP T ER 4. PO LI T I C AL ECO N OMY 259 Advantages and disadvantages of covert-purchases surveys Covert purchasing allows researchers to directly identify sources of illicit cigarettes. It also allows them to measure and test the dynamics between buyer and seller. For example, researchers can experiment to see if repeated attempts to purchase products increase the likelihood of purchase (known as the familiarity protocol) (16). One methodological challenge associated with covert purchases is that it is difficult to create a sampling frame for illicit sources because some may be unknown (e.g. pubs or homes). The traditional approach is to make purchases in legal outlets, which may bias estimates. Another issue with this method is that it is difficult for buyers (also called raters) to purchase products if they are unfamiliar with the seller or do not fit the typical demographics of purchasers. Therefore, researchers using covert- purchases surveys must have detailed knowledge of the marketplace, including the ways individuals specifically ask for illicit tobacco products, and they must know whether they mirror the demographics of the neighbourhood. For example, in a study of South Bronx smokers, von Lampe et al. found that smokers looked for certain clues to assess whether they were being sold illicit cigarettes (19). Overall, this method can be quite costly because it requires training researchers, travelling to retailers and purchasing product. Covert-purchases surveys do not enable researchers to estimate the level of illicit trade, but they can provide information on availability of supply. Key studies for readers to refer to for additional guidance: Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Global Health. 2016;12(1):78. A4.1.6 SEIZURES OF GOODS Seizures are the result of enforcement activity carried out by local, national and international organizations that confiscate tobacco products that are illegally manu- factured, transported and sold. Seizures are meant to reduce the profits associated with illicit trade by confiscating proceeds (e.g. cash, cars or houses) and the tools of the trade (e.g. print and tobacco machinery). Seizures can occur at various points in the supply chain. Seizure data are often tallied by the responsible agencies and used to measure program effectiveness or as justification for requesting additional resources (e.g. personnel). Some of the data may be supplied to international customs organiza- tions, including the WCO (20). The quality of recordkeeping varies. For example, some agencies may maintain criminal files in databases that detail dates of seizure, brand names and laboratory testing. 260 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Seizures provide preliminary data on the scope of criminal activity and can help identify key trends to guide law enforcement agencies’ efforts. For example, seizures can identify trends on the modus operandi of smugglers, including transporta- tion methods (e.g. sea cargo versus trucks), point of entry and brand preference. Seizure statistics can also be used as a preliminary test to measure the efficacy of interventions. For example, Stoklosa and Ross used seizure data from the Canadian province of Nova Scotia to test the impact of a 2015 menthol ban. He found no statistically significant change in the number of menthol cigarettes seized before and after the ban (1). Advantages and disadvantages of seizures of goods Generally, seizure statistics can be readily obtained from law enforcement agencies through formal requests to agency gatekeepers (e.g. public information officers). Seizure data, however, generally do not provide a representative picture of illicit activity. For example, certain geographies may yield higher seizures because that is where the bulk of operations are being conducted. Police agencies may focus on certain geographies (e.g. locations near borders) rather than randomly inspecting, and their findings may be limited to those specific regions. Seizure data may also be skewed by the type of investigation procedures utilized. Large seizures may be the result of long-term investigations (i.e. wiretaps or culling confidential infor- mants), while smaller seizures may come from anti-smuggling cases that involve cross-border purchases of low quantities of cigarettes (less than 1 000) (2). Seizures can also be skewed by industry cooperation with law enforcement agencies. For example, the tobacco industry may be more likely to support law enforcement on counterfeit seizures rather than smuggling cases because counterfeiting impacts their brand integrity. A4.2 RESIDUAL METHODS Because the illicit tobacco trade is often decentralized, it can be difficult to observe directly. However, researchers are sometimes able to make inferences about its size without direct observation by comparing observed tobacco tax revenues with the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. The difference between observed and hypothetical revenues is called the residual and can be used as an indicator of the magnitude of illicit trade. Even when the residual is only an approximate measure, changes in its size may be a reliable indicator of changes in the size of the trade. When actual tobacco tax revenues are reliably observed, the main challenge for residual methods is that of producing accurate estimates of the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. CHAP T ER 4. PO LI T I C AL ECO N OMY 261 A4.2.1 GAP ANALYSIS Gap analysis is the preferred residual methodology because it is intuitive, straight- forward and relatively easy to explain to policy-makers and the general public, and it has been widely employed in government studies (4). Researchers using gap analysis compare survey-based self-reported consumption data with observed (usually administrative) data on tax-paid sales. The basic premise is that if both self-reports and observed data are accurate, any difference between reported consumption and tax-paid sales can be explained by legal imports of non-taxed cigarettes (such as duty-free sales), exports of taxed cigarettes, tax evasion or tax avoidance. The greatest research challenge in implementing gap analysis – as with most residual methods – is obtaining reliable and accurate estimates of tobacco consump- tion. In its simplest implementation, gap analysis calculates the residual as the difference (which should be a minimum of zero) between the amount of tobacco consumption reported in surveys and tax-paid sales, which are generally available from administrative sources, minus exports. This simple calculation, however, is generally flawed, since surveys of reported tobacco consumption underestimate true consumption. Underreporting of tobacco consumption may result from survey respondents’ reticence about disclosing behaviour that is viewed as unhealthy and potentially socially undesirable. Depending on the legal and cultural context, cer- tain groups (e.g. women or youth) may be more likely than others to underreport consumption. Other groups (e.g. rebellious young men and women) may accurately report or even overestimate consumption. Reuter and Majmundar measured actual consumption by total national taxed sales in the United States, where both legal untaxed imports and exports of taxed tobacco are widely believed to be very small, and found that the ratio of self-reported consumption to actual consumption was only 65% (4). After incorporating this survey underreporting into their analyses and considering the evidence from their gap analysis and the literature, Reuter and Majmundar found that the illicit market in the United States, which largely consists of avoidance or evasion of subnational state taxes, is between 8.5% and 21% of consumption (4). The higher range of the estimate is consistent with prior estimates using population-based pack observation studies (21). Researchers using gap analysis for countries or regions where legal imports of untaxed tobacco (such as duty-free products) or (legal or illegal) exports of taxed cigarettes are more significant should attempt to incorporate data about, or estimates of, these factors into their calculations. Legal untaxed imports of tobacco should be subtracted from reported consumption (after adjustment for underreporting), and exports of taxed tobacco should be subtracted from taxed sales. Obtaining data about legal untaxed imports and exports of taxed tobacco may be challenging, because these imports and exports may be the result of decentralized decisions of individual 262 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N travellers as they cross tax borders. Data on these activities will not necessarily be collected through normal administrative activities. Despite these challenges, gap analyses may still prove useful. For example, if researchers have reason to believe that misreporting of tobacco consumption and the amounts of legal untaxed imports and taxed exports are relatively stable or follow known trends (e.g. are falling) over time, gap analyses can be used to provide estimates or lower (or upper) bounds on illicit trade when several years of data on taxed sales and reported consumption are available. Data sources may be country tax administrators who have access to sales data and health departments that have access to population-level studies of reported tobacco con- sumption. In this context, multiple years of data on reported consumption and tax-paid sales can allow researchers to estimate changes in the size of the illicit trade even when it is difficult to measure the absolute level. Paraje used the 2008 Global Adult Tobacco Survey and the 2013 National Health Survey to measure reported tobacco consumption in Brazil (22-23). Advantages and disadvantages of gap analysis A major advantage of gap analysis is that when quality data are available, it is simple, easily reproduced and explainable to policy-makers and the general public. How- ever, high-quality data on reported consumption may not be available, especially in low-income countries. In many cases, gap analysis does not provide reliable information on the size of the illicit market but only on changes in the size over time (22). Additionally, some low-income countries may not have reliable estimates of tax-paid cigarette sales, and secondary data repositories of cigarette sales may not be transparent about their methodology (24). Another disadvantage of gap analysis is that it generally cannot be used to obtain separate estimates of tax avoidance and tax evasion. Biased estimates may also result if surveys of tobacco consumption are not representative of the population (25). Moreover, it is generally not possible to quantify the precision of the estimates or uncertainty associated with the estimates, because of both statistical uncertainty resulting from the use of samples to imperfectly represent populations (e.g. the share of the population that smokes) and uncertainty about key facts such as the degree to which survey respondents understate their tobacco consumption. Key studies for readers to refer to for additional guidance: Szklo A, Iglesias RM, Carvalho de Souza M, Szklo M, Maria de Almeida L. Trends in illicit cigarette use in Brazil estimated from legal sales, 2012–2016. Am J Public Health. 2018;108(2):265–269. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine and Tob Res. 2019;21(8):1079–86 CHAP T ER 4. PO LI T I C AL ECO N OMY 263 A4.2.2 ECONOMETRIC MODELLING There is a long tradition of using data to estimate parameters of demand functions that relate the quantity of goods consumed to the prices faced by consumers, their incomes and other variables. Because of the addictive nature of tobacco – and because of important public health and public policy concerns relating to tobacco use – economists have paid particular attention to the estimation of cigarette demand functions (26). As the literature on this topic developed, it became apparent that taxed tobacco sales would be a biased indicator of tobacco consumption if some consumers obtained their tobacco in illicit markets. Similarly, the price of cigarettes in the legal market might overestimate the price paid by consumers if some sales were not tax-paid. While economists generally cannot observe sales in the illicit market, they have been able to develop models that predict conditions under which consumers avoid tobacco taxes. They reason that the relative size of illicit tobacco markets depends primarily on two variables: the relative price of taxed and untaxed consumption and the ease of obtaining lower-cost (untaxed) tobacco. Other variables, includ- ing the social stigma from evading tax laws and the perceived relative quality of illicit tobacco, could also influence the demand for it. While illicit trade cannot be directly observed, it can be estimated from the difference between tax-paid sales and predicted consumption. Tax-paid sales can be less than predicted consump- tion when retailers or consumers evade taxes. They can be greater if some tax-paid cigarettes are bought within the jurisdiction and then consumed in areas where after-tax prices are higher. Econometric modelling estimates of illicit trade must be tailored to the situation in the country that is being studied, and therefore the data requirements may differ substantially from case to case. Researchers using this method should be familiar with the literature and should also understand the conditions in the areas they are researching. They must always include some measure of tobacco consumption or sales and some measure of the price of tobacco in the home country, as well as other variables (e.g. income) that are known to affect the demand for tobacco. It is also generally necessary to include variables that measure the availability and relative price of illicit tobacco, which can often be measured by comparing tobacco taxes in the home country with those in areas that are the source of illicit tobacco. Advantages and disadvantages of econometric modelling The major advantage of econometric modelling is that it is consistent with a long tradition of economic theory and practice, and the quality of the modelling techniques and empirical estimates can therefore be evaluated against widely accepted criteria. Empirical analyses provide estimates of price elasticities, income elasticities and price elasticities of tax avoidance. A substantial literature base makes it possible to 264 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N rigorously quantify uncertainty about the estimates and to test their robustness to various assumptions made in the modelling process. Estimates can be compared to other estimates available in the literature, and the results of these analyses can be used to simulate the impacts of policy changes (including tax and/or enforcement increases) on both consumption and tax avoidance. Because this methodology provides direct estimates of the uncertainty of the results, researchers can specify their level of confidence in the findings. A shortcoming of econometric modelling is that it requires high-quality data on a variety of important variables over a period of time, as well as advanced econo- metric modelling expertise. Also, because results from the econometric models are based on statistical inference and economic theory rather than direct observation (e.g. the proportion of packs without tax stamps), it can be difficult to explain to policy-makers and the general public. Key studies for readers to refer to for additional guidance: Becker GS, Grossman M, Murphy KM, (1994). An empirical analysis of cigarette addiction. Amer Econ Review. 1994;84(3):396–418. Schafferer C, Yeh CY, Chen SH, Lee JM, Hsieh CJ. A simulation impact evaluation of a cigarette excise tax increase on licit and illicit cigarette consumption and tax revenue in 36 European countries. Public Health. 2018;162:48–57. A4.2.3 EXPERT OPINION (KEY-INFORMANT SURVEYS AND INTERVIEWS) Insight on illicit trade dynamics can come from experts in the field, including researchers (e.g. in economics, criminal justice and public health), taxation depart- ments, enforcement agencies, product manufacturers, wholesalers and retailers. Other key informants include journalists and academics who have secured confi- dential informants. Experts can provide novel information about emerging trends (e.g. new smuggling routes). In some cases, researchers can obtain interviews with incarcerated or active offenders (27-29). For example, researchers studying cigarette smuggling in eastern Africa conducted interviews with more than 150 Ugandan tobacco smugglers (29). Experts can be queried through surveys or semi-structured interviews. When sampling frames are available (e.g. directories of tax department employees), surveys are more expedient than interviews. However, when experts are hard to find, non- random sampling strategies coupled with interviews are recommended. Identifying experts may require recruiting a gatekeeper who is tasked with helping researchers find additional experts; or purposive sampling, where individuals are identified based on set criteria (e.g. they are taxation experts employed by local governments) (30). CHAP T ER 4. PO LI T I C AL ECO N OMY 265 Advantages and disadvantages of key-informant surveys and interviews Informant interviews can be a useful starting point for identifying trends in the marketplace (e.g. venues where illicit cigarettes are sold or modes of entry). One of the disadvantages of relying on informants is that the information solicited from them may not be generalizable. Expert knowledge may be outdated or limited by the informants’ experience. Furthermore, the opinions of experts are subjective and may be biased by the experts’ employment status and the sampling methods used. For example, persons working in law enforcement may overestimate the extent of bootlegging in order to secure additional funding for future operations. Similarly, manufacturers looking to defeat taxes may overestimate the illicit trade to illustrate the links between taxation and illicit behaviour. Alternatively, tobacco control advocates may underestimate illegal market measures in order to support the argument that taxes do not increase illicit trade. Key studies for readers to refer to for additional guidance: Joossens L, Raw M. Cigarette smuggling in Europe: who really benefits? Tob Control. 1998;7:66–71. doi:10.1136/tc.7.1.66 PMID: 9706757. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20(3):226–232. A4.3 MIXED AND MULTIMETHOD STUDIES Given the shortcomings of the aforementioned methods for assessing the nature and size of the illicit tobacco trade, governments may want to validate their findings by using mixed or multiple methodologies. Mixed methods use two methodological paradigms, qualitative and quantitative, as tools for exploration and explanation. For example, mixed method studies can use littered-pack surveys to measure the size of the market along with self-report surveys of smokers to understand patterns of purchasing, including sources, frequency and social norms. For example, Stoklosa and Ross estimated the share of the illicit market in Poland using a population- based self-report survey and a littered-pack survey (1). Using both types of survey simultaneously enables governments to assess their validity in estimating the size of the illicit market. Alternatively, governments can employ multimethod research, i.e. the use of multiple methods that are similar in tradition (e.g. focus groups and semi-structured interviews) (31). Saenz de Miera et al. used face-to-face interviews (households), litter collection and observation of single-stick sellers, which enabled them not only to cross-validate the two major methodologies, but also to see if the brand of the single stick was a good measure of licit versus illicit trade (33). 266 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Advantages and disadvantages of mixed and multimethod studies Mixed and multimethods studies enable researchers to check the validity of their findings. Multiple methods are preferred in contexts where illicit trade estimates are politicized. For example, low estimates may be challenged by the tobacco indus- try, while high estimates may be challenged by tobacco control researchers and/or proponents. Mixed and multiple methods (e.g. littered-pack surveys and informant interviews) can enable governments to understand the situational context in which the illicit trade operates, including the actors involved and venues of sale. One disadvantages of using mixed and multiple methods is cost. Governments that are constrained by tight budgets may choose to use a single method that provides the most accurate information. However, given the issues faced by each method, this may not be feasible – each method has limitations. Instead, governments can pair a high-cost method with a lower-cost method (e.g. pairing interviews with empty pack surveys, or law enforcement seizure data with face-to-face consumer surveys). Key study for readers to refer to for additional guidance: Zaloshnja E, Ross H, Levy DT. The impact of tobacco control policies in Albania. Tob Control. 2010;19:463–468. CHAP T ER 4. PO LI T I C AL ECO N OMY 267 REFERENCES 1. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 2. Joossens L, Lugo A, La Vecchia C, Gilmore AB, Clancy L, Gallus S. Illicit cigarettes and hand-rolled tobacco in 18 European countries: a cross-sectional survey. Tob Control. 2014;23:e17–e23 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC3812425/pdf/nihms491463.pdf, accessed 2 February 2021). 3. Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–7. 4. Reuter P, Majmundar M. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Washington (DC): National Academies Press; 2015. 5. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 6. Barkans M, Lawrance KA. Contraband tobacco on post-secondary campuses in Ontario, Canada: analysis of discarded cigarette butts. BMC Pub Health. 2013;13(1):335 (https://bmcpublichealth. biomedcentral.com/articles/10.1186/1471-2458-13-335, accessed 2 February 2021). 7. John RM, Ross H. Illicit cigarette sales in Indian cities: findings from a retail survey. Tob Control. 2018;27(6), 684–688. 8. Davis KC, Grimshaw V, Merriman D, Farrelly MC, Chernick H, Coady MH, et al. Cigarette trafficking in five northeastern US cities. Tob Control. 2014;23(e1):e62–e68. 9. Callaghan RC, Veldhuizen S, Leatherdale S, Murnaghan D, Manske S. Use of contraband cigarettes among adolescent daily smokers in Canada. CMAJ. 2009;181(6-7):384–6 (https://www.cmaj.ca/content/ cmaj/181/6-7/384.full.pdf, accessed 2 February 2021). 10. Guindon GE, Burkhalter R, Brown KS. Levels and trends in cigarette contraband in Canada. Tob Control. 2017;26(5):518–25 (https://www.researchgate.net/publication/307890937_Levels_and_trends_in_ cigarette_contraband_in_Canada, accessed 2 February 2021). 11. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany: New York State Department of Health; 2006 (https://www. health.ny.gov/prevention/tobacco_control/docs/cigarette_purchasing_patterns.pdf, accessed 2 February 2021). 12. International Tobacco Control Evaluation Project (2018). Surveys. Waterloo: University of Waterloo; 2018 (http://www.itcproject.org/surveys, accessed 12 October 2020). 13. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 14. Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. 15. Brown J, Welding K, Cohen JE, Cherukupalli R, Washington C, Ferguson J, et al. An analysis of purchase price of legal and illicit cigarettes in urban retail environments in 14 low-and middle-income countries. Addiction. 2017;112:1854–60 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5600117/pdf/ADD- 112-1854.pdf, accessed 2 February 2021). 16. Scollo M, Bayly M, Wakefield M. Availability of illicit tobacco in small retail outlets before and after the implementation of Australian plain packaging legislation. Tob Control. 2015;24(e1):e45–e51. 17. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Globalization and Health. 2016;12(1):78 (https://globalizationandhealth. biomedcentral.com/articles/10.1186/s12992-016-0219-z, accessed 2 February 2021). 18. Phillip Morris USA Inc. v. Shalabi. United States, District Court, C.D. California; 2004 (https://www. casemine.com/judgement/us/5914b6cbadd7b0493477b3da, accessed 2 February 2021). 19. von Lampe K, Kurti M, Johnson J, Rengifo AF. ‘I wouldn’t take my chances on the street’ navigating illegal cigarette purchases in the South Bronx. J Res Crime Delinq. 2016;53(5):654–80 (https://www. 268 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N researchgate.net/publication/298515449_I_Wouldn’t_Take_My_Chances_on_the_Street_Navigating_ Illegal_Cigarette_Purchases_in_the_South_Bronx, accessed 2 February 2021). 20. Customs and Tobacco Report 2009. Brussels: World Customs Organization; 2009. (http://www.wcoomd. org/en/media/newsroom/2010/june/~/media/83967DFEB9F74D388924A4C61F279DC4.ashx, accessed 12 October 2020). 21. Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ, et al. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61–i66 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3984758/pdf/nihms567943.pdf, accessed 2 February 2021). 22. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine Tob Res. 2019;21(8):1079–86. 23. National Survey of Health. Instituto Brasileiro de Geografia e Estatística; 2013 (in Portuguese) (https:// www.ibge.gov.br/en/statistics/social/health/16840-national-survey-of-health.html?=&t=downloads, accessed 27 November 2020). 24. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 25. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 26. Chaloupka FJ, Warner KE. The economics of smoking. In: Arrow KJ and Intriligator MD, editors. Handbook of Health Economics. Amsterdam: Elsiver; 2000. pp1539–1627. 27. Antonopoulos GA. Cigarette smugglers: a note on four ‘unusual suspects’. Glob. Crime. 2007;8(4):393–8 (https://www.researchgate.net/publication/248955367_Cigarette_Smugglers_A_Note_on_ Four_’Unusual_Suspects’, accessed 2 February 2021). 28. Antonopoulos GA. The Greek connection(s): the social organization of the cigarette-smuggling business in Greece. Eur J Criminol. 2008;5(3):263–88 (https://www.researchgate.net/publication/249752218_ The_Greek_ConnectionsThe_Social_Organization_of_the_Cigarette-Smuggling_Business_in_Greece, accessed 2 February 2021). 29. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20:226–32 (https://www.researchgate.net/publication/49809289_ Blood_cigarettes_Cigarette_smuggling_and_war_economies_in_central_and_eastern_Africa, accessed 2 February 2021). 30. Babbie, ER. The basics of social research. Boston: Cengage Learning; 2013. 31. Tashakkori A, Teddlie C., editors. Sage handbook of mixed methods in social & behavioral research. New York: Sage; 2010. 32. Saenz de Miera Juarez B, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 269 ANNEX 4.2 HOW ARE THE TOBACCO TAX REVENUES EARMARKED? The introduction of earmarking of tobacco tax revenue is almost always combined with an increase in excise taxes (or a new surcharge) rather than reallocation of existing revenues (1). Table A4.1 provides examples of the different approaches used by several countries to earmark tobacco tax revenues. Table A4.1 Approaches used to earmark tobacco tax revenues FUNDING SOURCES/ TYPE OF TAX EXAMPLES OF TAX BASE AND RATES As part of the excise system (tobacco, alcohol) Specific Republic of Korea: 841 won (US$ 0.75) per pack or 29% of the specific excise rate Costa Rica: 467.8 cólones (US$ 0.83) per pack or 100% of the specific excise rate Congo: 20 CFA francs (US$ 0.036) per pack or 50% of the specific excise rate Ad valorem Colombia: 10% of retail price (equivalent to 100% of the ad valorem rate) New levy (surcharge on the existing excise or completely new levy) Specific Egypt: additional 0.75 Egyptian pounds (US$ 0.042) per pack Ad valorem Thailand: surcharge of 2% over the excise tax base Botswana: new tobacco levy of 30% of the cost of production or CIF Percentage of excise revenue Cook Islands: 50% of revenues from the excise tax on tobacco Guatemala: 100% of revenues from the excise tax on tobacco Note: Conversions of amounts from the local currency were made using the official exchange rates from the IMF as of 31 July 2018 (date of the data collection). Source: (2). REFERENCES 1. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 2. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 270 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 271 CHAPTER 5. Best practices in tobacco tax policy and administration TAX POLICY Use excise tax increases to achieve the public health goal of reducing the death and diseases caused by tobacco use Extensive research has clearly demonstrated the effectiveness of higher tobacco product taxes and prices in reducing tobacco use and its harmful consequences, particularly among the poor and the young. In fact, tobacco excise tax increases are the single most effective and cost-effective policy for reducing tobacco use. Excise taxes are the most significant taxes applied on tobacco products because of their ability to raise both absolute and relative prices. Tobacco excise tax increases also generate sizeable new revenues that will be sustained in the short to medium term. In the long term, continued increases in tobacco taxes – coupled with implementa- tion of other evidence-based tobacco control policies and programmes – will lead to even larger reductions in tobacco use and its consequences. Include significant tobacco excise tax increases as part of a comprehensive strategy to reduce tobacco use Governments should adopt a comprehensive tobacco control strategy that includes objectives for reducing adult tobacco use and preventing youth tobacco use. Rais- ing excise taxes significantly is the most effective, as well as the most cost-effective, measure for reducing consumption. When combined with other demand reduction interventions, the impact of tax increases on tobacco use is even stronger. Such interventions include comprehensive smoke-free policies in all public spaces, total bans on tobacco advertising, promotion and sponsorship by tobacco companies, large graphic health warnings about the consequences of tobacco use, plain packaging, broad efforts to help current users quit and mass media public education campaigns. Implementation of a comprehensive strategy to reduce tobacco use leads to greater reductions in the harmful consequences of tobacco use, builds public and political support for higher taxes and maximizes the effectiveness of tax increases in achieving public health objectives. 272 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Involve the competent authority from the start when considering the revision of a tax policy Competent authorities such as tax administrations and customs authorities are key partners in the effective implementation of a tax policy. Policy-makers need to ensure that those competent authorities are consulted and involved in the tax policy revision process so that their concerns about the impacts of policy change on enforcement can be taken into account from the beginning. This can also help identify and address possible loopholes early on in the enforcement process. Coordination among relevant bodies, including close cooperation and sharing of information, will optimize enforcement of tax policy and tax collection. To streamline the process of cooperation and exchanges of information, a basis in law needs to be established. Additionally, the involvement of tax administration authorities in the entirety of the tax revision policy process is important to ensure effective implementation of the policy. Promote greater policy coherence across sectors such as agriculture, industry, trade, finance and labour Greater multisectoral integration and policy coherence is needed at the country level to achieve effective health improvements. In particular, it is important to ensure that public policies and interventions in non-health sectors (e.g. agriculture, industry, trade, finance and labour) do not act against the intended public health impact of tobacco control and taxation (such interventions include providing subsidies to tobacco growing or manufacturing). TAX DESIGN Tax structure matters and simpler is better Complex tax structures are difficult to administer, create opportunities for tax avoid- ance and evasion and are less effective than simpler structures in achieving public health and revenue goals. Simplifying the structure of tobacco excise taxes will facilitate tax administration, reduce tax avoidance and evasion, enhance revenues and have a greater impact on tobacco use by reducing incentives to substitute among tobacco products or brands in response to tax increases. Countries with multiple tiers of tobacco tax rates based on product characteristics (e.g. price level, length, weight, type of tobacco) should reduce and eventually eliminate these differential tax rates. An appropriate transition strategy is to reduce the variations in tax rates over time with the aim of implementing a uniform tax (i.e. a single rate applies whether excise is ad valorem or specific) on a given tobacco product. Applying a uniform tax to all brands of a given tobacco product also sends a clear message that they are equally harmful. CHAP T ER 5. B E S T PR AC T I CE S 273 Rely more on specific tobacco excises to drive price increases Greater reliance on specific excise taxes maximizes the impact of tobacco taxes on public health by reducing the gap in prices between premium and low-priced alternatives and limiting opportunities for users to switch down in response to tax increases. For countries that currently rely on an ad valorem tax, an appropriate first step would be to shift to a mixed system by adding a sizeable specific component or introducing a high minimum specific excise tax (an excise tax floor). For countries that rely on a mix of ad valorem and specific taxes, the specific tax component should be increased regularly so that it accounts for a greater share of the total excise tax. Increase tobacco taxes significantly to reduce the affordability of tobacco products To maximize the public health impact of higher tobacco taxes while at the same time generating higher revenues, governments should significantly raise taxes to increase prices and reduce the affordability of tobacco products. In many LMICs, tobacco use increases with incomes, and since incomes rise faster than tobacco product prices, these products are becoming more affordable. To reduce afford- ability, tax increases need to result in real price increases that are higher than the increases in real incomes. Where revenue increases are a goal, rely on regular excise tax increases If governments want to increase tobacco revenues, they must increase excise taxes regularly. From the tax revenue perspective, the important determinant is the tax base elasticity, which has three key components: the price elasticity of demand of tobacco, the share of the tax in the retail price and the degree of pass-through of the excise tax rate increase on to retail price. Tax increases will increase revenues at least in the short to medium term, because demand is price inelastic, tax levels are generally low as a proportion of retail prices and the pass-through of tax increases on to retail prices is unlikely to be higher than the tax increase itself (i.e. there is no overshifting). In addition, increasing tax rates is the only policy measure that can reverse reduced revenues in a declining market that has strong tobacco control policies. Automatically adjust specific tobacco taxes for inflation and income growth Unless specific tobacco taxes are regularly adjusted, their real value will fall over time as general price levels increase. When this happens, their effectiveness in reducing tobacco use will be diminished. Governments should establish a mechanism for automatically adjusting specific taxes to keep pace with inflation. Recently, some governments have begun to extend this indexation to include income growth as well, further ensuring that tobacco does not become more affordable over time. 274 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Pricing regulations cannot be considered an alternative to excise tax. However, in some specific contexts, pricing regulations could be used in conjunction with excise taxes to help ensure the effective implementation of tax increases In certain contexts where increasing taxes is challenging or the tax structure is weak, non-tax policies such as pricing regulation (specifically, minimum mark-ups and price floors/minimum prices) may be seen as a second-best alternative to ensure a high price level and dissuade consumption of tobacco products. These policies, however, do not necessarily lead to the desired price level, nor do they protect consumers and government from industry manipulation. However, in the context of powerful multinationals that sell brands across all market segments and could easily undershift a tax increase to cheaper brands – or where price promotions cannot be banned – minimum price policies may help increase the effectiveness of tax increases, especially if the minimum prices are increased regularly. Implement nontax policies affecting price levels, such as banning promotional discounts for tobacco products and the sale of single sticks of cigarettes The banning of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Do not allow concerns about the inflationary impact of higher tobacco taxes to deter tax increases Given that wages or some government spending may be tied to a price index, govern- ments can reduce concerns about the inflationary impact of a tobacco tax increase by using a price index that excludes tobacco products. TAX PARITY Tax all tobacco products in a comparable way Increasing excise taxes on some tobacco products but not on others results in changes in the relative prices of these other products. This induces substitution towards relatively less-expensive products – for example, from expensive manufactured cigarettes to other, cheaper tobacco products such as RYO tobacco, bidis, cheroots or chewing tobacco. As a result, the overall reduction in tobacco use is smaller than it would have been had all taxes increased by comparable amounts. Comparable increases in the taxes on all tobacco products maximize the public health impact of tobacco tax increases by minimizing opportunities for substitution. Moreover, increases in taxes on all tobacco products will generate larger increases in revenues. CHAP T ER 5. B E S T PR AC T I CE S 275 Strictly regulate new and emerging tobacco and nicotine products where they are not banned and impose an excise tax In recent years, the world has been experiencing the rise of new and emerging tobacco and nicotine products including ENDS, ENNDS and HTPs. The tobacco industry claims these new products are safer than traditional tobacco products, but the evidence so far suggests that they could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of newer products – as well as initiation, smoking cessation and switching behaviour among different socioeconomic groups – are not yet clear. Best practices for taxing new and emerging tobacco and nicotine products, based on current knowledge, are that: 1. HTPs should be taxed at the same level as cigarettes and, in terms of structure, through a specific excise per unit regardless of tobacco content. HTPs contain tobacco and should be treated as a tobacco product. 2. ENDS/ENNDS products should be taxed in a manner that discourages up- take by youth and non-users. Nicotine- and non-nicotine-delivery systems containing e-liquids should be taxed equally. 3. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While these newer products create additional challenges for tobacco control, it is important to remember that cigarettes remain by far the predominant tobacco product and that raising taxes and prices on cigarettes – and thereby reducing their use – should remain the top priority. MONITORING AND EVALUATION Know your market Know your market well. The type of tax structure you choose and the impacts it will have on consumption and tax revenue are shaped by the particular dynamics of your market. Understanding the nature and degree of competition in your market is vital to selecting the appropriate type of tax structure and policies to achieve your public health and revenue objectives. This knowledge will also facilitate more accurate estimates of the impacts of a tax increase, as well as better anticipation of industry responses. Assess the impact of your policies to design and implement the most effective tobacco excise tax policies Monitoring and evaluation are essential for effective tobacco taxation, and they 276 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N should be built into the initial design – or redesign – of tobacco tax policies. A number of tools exist to help policy-makers pre-emptively assess the effects of a proposed tobacco control policy on consumption, smoking prevalence and lives saved. The WHO TaXSiM uses target simulations to assist governments in predict- ing how specific tax changes will impact consumer prices, consumption and tax revenue in their market. Adopt indicators that help you measure improvements in tax policy and its impact Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the improvement of their policies and determine if those policies have an impact on tobacco use over time. The tax share of the retail price for a particular tobacco product is a key indicator that should be used in conjunction with an af- fordability indicator. A recommended target for countries to aspire to is to have an excise tax that represents at least 70% of the retail price of tobacco products. Another useful indicator to assess the performance of the tax policy overall is the use of a tax scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). TAX ADMINISTRATION Implement best practice approaches in general tax administration to make tobacco tax administration more effective and efficient Best practice approaches include (1) defining clearly the roles and responsibilities of competent authorities, (2) ensuring effective coordination among relevant bodies at the national and international levels and (3) undertaking evaluation of performance and accountability against pre-agreed indicators to identify points for improvement. Ensure compliance and accuracy of information on the tax compliance cycle To achieve this, implement the following actions: • Require licences for manufacturing, importing, exporting, retailing, growing, transporting, wholesaling, brokering, warehousing and distributing tobacco products. This will help secure the supply chain while obtaining valuable infor- mation, e.g. through access to companies’ accounting and inventory systems. • Make sure all persons and entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment keep complete and accu- rate records of all relevant transactions and details of materials used in the production of tobacco products. CHAP T ER 5. B E S T PR AC T I CE S 277 • Ensure that tax declarations collect as much information as possible on the taxpayer. • Collect taxes close to the point of production and import to limit the number of taxpayers a competent authority needs to manage. • Maintain a system of authorization for warehousing to carry out controls in production and storage facilities to ensure that taxes are paid. • Use electronic methods, through the best available IT, for declarations and collection of taxes. This allows for cross-check of information provided in dec- larations with information from other government agencies and third parties. Ensure control and enforcement on the supply chain To achieve this, implement the following actions: • Include control and enforcement as a fundamental pillar in the strategic plan of the tax administration overall. • Use a risk-based approach by choosing defined targets for enforcement and control, such as those who have a higher probability of noncompliance. • In the licensing process, ensure that purchases from unlicensed suppliers or sales to unlicensed purchasers are not allowed. Ensure also that the validity of licences is limited in time and require renewals or reapplication to maintain a high level of control. • Use tax stamps with strong security features to reduce the risk of stamp counterfeiting. These markings facilitate the collection of excise taxes, audits and enforcement actions. • Implement a tracking and tracing system for tobacco products. A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, and monitoring and control- ling the movement of tobacco products and their legal status. • Implement anti-forestalling measures so that forestalling does not delay a tax increase and its intended effect on revenues and consumer behaviour. • Control import and export of tobacco products and manufacturing equipment by allowing only duly licensed natural persons or legal entities to conduct such activities. • Strengthen border control, e.g. by utilizing non-invasive tools such as X-ray scanners and dogs to detect tobacco products. • Limit or tightly control and, ideally, ban activities related to production and trade of tobacco products in tax-free zones to avoid opportunities for tax evasion. • Prohibit intermingling of tobacco products with non-tobacco products in a single container or any other similar transportation unit when removed from tax-free zones. 278 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Prohibit the sale to or import by international travellers of tax-free or duty- free tobacco products. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and adversely affect government revenues by creating a loophole in the tax structure. Clearly define procedures to follow after detecting illicit trade of tobacco • Take immediate action to seize and destroy smuggled and/or illicit tobacco and collect due taxes. • Ensure certain, swift and severe sanctions for those caught engaging in illicit trade in tobacco products, such as penalties, fines and withdrawal of licences. It can also be effective to consider illicit trade in tobacco products by law as a source of money-laundering. Become a Party to and/or implement the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products The WHO FCTC Protocol provides a blueprint of best practices and policies for dealing with illicit trade and should be part of any strategy to fight it. Implement, to the extent possible, the same rules and regulations for tax administration and enforcement for all tobacco products, as well as new and emerging nicotine and tobacco products Implement broad policies for ensuring a good tax system that will trickle down to good tax administration of tobacco products by: • ensuring proper resourcing of competent authorities; • having strict rules and regulations to detect corruption and to punish both personnel and taxpayers who are engaged in corrupt practices; and • ensuring a strong judicial system that is independent in fact and in perception, where disputes are solved quickly. The appeal process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially for illicit trade. POLITICAL ECONOMY Beyond the technical soundness of best practices in tax policy and administration, a critical factor in advancing tobacco taxes is the ability to get the political buy-in of the highest instances in the government. One key strategy is to address concerns around the political economy of tobacco taxation, which are often exploited by the tobacco industry to block major reforms. CHAP T ER 5. B E S T PR AC T I CE S 279 SCARE tactics The tobacco industry uses SCARE tactics to dissuade governments from implement- ing tobacco tax increases. These include smuggling and illicit trade (S), court and legal challenges (C), anti-poor rhetoric (A), revenue reduction (R) and employment impact (E). Best practices for countering these tactics are described below. S: Smuggling and illicit trade Do not allow concerns over the impact of increasing excise taxes on illicit trade in tobacco affect your decision to increase them. Rely on your own estimates of the level and nature of illicit trade and not on the industry’s estimates. Illicit trade in tobacco products continues to be a major concern for tax administrators because of the difficulties associated with accurate and independent measurement of it, as well as with its elimination. Industry figures provide a distorted understanding of the extent of the problem, along with a monocausal explanation of the link between illicit trade and tobacco taxation. It is therefore recommended that governments (1) assess independently and with the best statistical practices the size of the illicit trade to assess the scope of the problem; (2) address directly the country-specific institutional and/or governance challenges, including multilateral coordination, and improve tax and customs administrations practices; and (3) implement best practices to fight illicit trade, contained in the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products. Ideally, accede to the Protocol if not yet a Party. C: Court and legal challenges Do not let tobacco industry threats of court and legal challenges to tax increases or reforms prevent you from improving your tax policy. Closely follow legal requirements for design, procedure and consultation to strengthen your legal position and minimize the possibility that any challenge will be raised. Health-protective and non-discriminatory tobacco excise taxes are legally defensible, and industry threats will usually be baseless. Your legal position can be strengthened, however, by exercising care with a tax measure’s procedure, design and consultation: (1) determine the standard of consultation required under domestic law and any applicable international obligations; (2) distance the tobacco industry from the policy-making process to the extent that this is permissible; (3) avoid unnecessary and unjustified discrimination towards foreign tobacco products or investors in the design, implementation or enforcement of a tax measure; and (4) do not offer investment incentives in the form of inducements or contractual undertakings, as these may be binding in and of themselves or grounds for a challenge under an international investment agreement. 280 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A: Anti-poor rhetoric Do not allow concerns about the regressivity of higher tobacco taxes prevent tobacco tax increases. In recent years, there has been an overwhelming increase in the evidence concern- ing the positive distributional impact of tobacco taxes and tax increases. Indeed, tobacco taxation and tax increases are actually a progressive or pro-poor policy once these wider considerations are properly accounted for. In its effort to lobby against tax increases, the tobacco industry often claims that tobacco taxation will hurt the poor. This argument is based on the concept of regressivity in relation to taxation. Conceptually, a tax is regressive if it means lower-income people must pay a relatively greater proportion of their household income to meet the tax liability than wealthy people. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use that are disproportionately experienced by lower socioeconomic groups. Second, higher tobacco taxes and prices can induce behavioural change in the population, as reflected in the price elasticity of demand, which means that lower-income smokers will curtail their smoking the most and thus will benefit disproportionately in terms of health gains from reduced tobacco consumption and use. In fact, these broader considerations make tobacco taxation a progressive, rather than regressive, public health intervention. R: Revenue reduction Do not let fears of potential revenue reductions prevent you from increasing excise taxes on tobacco products. Tax increases, even in countries with already high taxes, bring in additional revenue. Arguments by tobacco control opponents that tax increases will not result in increases in revenue are unfounded. The relatively price inelastic nature of cigarette demand, combined with the low tax share and no overshifting of the tax, means that for most, if not all, countries, increases in revenues will accompany increases in taxes. If tax increases are carefully designed and tax administration is functional, it is extremely unlikely that tax increases will lead to revenue decreases. E: Employment impact Do not allow concerns about employment impact to prevent tobacco tax increases. The tobacco industry often seeks to frame tobacco taxes as an economic issue rather than a public health issue. Particular emphasis is placed on the alleged threat tax increases pose to employment in tobacco farming and manufacturing, as well as related industries. This so-called choice between health and jobs, however, is largely based on exaggeration. The tobacco industry exaggerates the importance of tobacco CHAP T ER 5. B E S T PR AC T I CE S 281 employment relative to total national employment and overstates the impact that domestic demand reduction from local taxes will have on tobacco farmers serving a global market. The argument used by the industry also ignores the fact that expendi- tures on tobacco do not disappear but rather are redistributed to other consumption that can produce a similar or higher number of jobs. Case studies demonstrate the possibility and methods for governments to support farmers in transitioning to other crops that provide similar and often better returns with greater sustainability. Earmarking Consider earmarking tobacco tax revenues for health-focused programmes, especially if it helps advance tobacco control efforts and, more specifically, efforts to implement large tobacco tax increases and tax reforms. This could have the additional benefit of funding health programmes where they are poorly funded or not prioritized. From a tobacco control perspective, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. Earmarking is a tool to improve the political economy of tobacco taxation; it is only a secondary issue, after the primary goal of reducing demand for tobacco. Evidence shows that public support for higher tobacco taxes is greater when at least some of the increased revenues are explicitly used to support health-focused programmes. Current evidence shows that the amounts effectively earmarked for health have been relatively small and therefore unlikely to introduce rigidity in government budgets. At the same time, in some countries, those funds have helped to imple- ment much needed underresourced health programmes. The payoffs will be seen in the future as fewer people fall ill and need less medical care for tobacco-related illnesses. Earmarking tobacco tax revenues to fund high-burden/low-priority health programmes could pave the way for raising awareness about the importance of such programmes and their effectiveness, thereby convincing governments to redefine their priorities and commit to including the programmes in their regular budget. 282 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) Focus on tobacco taxation’s purposes Tobacco tax policy should aim not only to increase revenues but also to decrease consumption and improve health. To both raise revenue and reduce consumption, you need to (1) simplify tobacco tax structures, (2) significantly increase rates to impact price levels, and (3) regularly adjust rates to at least account for inflation and income growth. Analyse your tax structure and identify its weaknesses You need to analyse and identify the problems of your current tax structure to know which steps to consider next. Which tax structure do you have: specific, ad valorem, mixed, or no excise? Identify the improvements to be made to the existing tax policy/structure Your present tax structure and tax situation will provide you with the steps you would ideally take next to achieve the aims in Step 1. Specific: 1. Ensure that the tax automatically adjusts for inflation and income growth effects. 2. Ensure that all price promotions are banned. Ad Valorem: 1. Ensure that the tax base of the ad valorem is retail price. 2. Introduce a high specific excise component (and a minimum specific excise). 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. Mixed: 1. Ensure that the tax base of the ad valorem component is retail price. 2. Ensure that you are using a high specific excise component and a minimum specific excise. 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. No Excise: 1. Introduce a high specific excise. 2. Ensure that the rate automatically adjusts for inflation and income growth effects. 3. Ensure that all price promotions are banned. Assess tobacco taxation’s political economy Reform must begin with an assessment of tobacco taxation’s political economy: (1) learn from past successes and failures – what went wrong, what went right, what you can do differently this time; (2) assess the reform’s strengths and weaknesses, likely opportunities and risks; (3) determine who the main supporters and opponents of reform inside and outside of government have been and may be, based on past reforms and current situation; and (4) anticipate argu- ments that will be used against the reform (refer to SCARE tactics). STEP 1 STEP 2 STEP 3 STEP 4 Prepare a plan for realizing the reform Focusing on the overall aims identified in Step 1, the steps for achiev- ing them identified in Steps 2 and 3 and the political economy around this reform as identified in Step 4, prepare your plan: 1. Be clear on the non-negotiable objectives for the reform and the trade-offs you are prepared to make to realize them. 2. Develop a plan to approach potential allies and win them over to the reform efforts. 3. Develop the counterarguments that will be needed in response to the SCARE arguments identified earlier. 4. Prepare the evidence you will need ahead of time. To do this, get support from academics and relevant intergovernmental agencies. Mobilize a coalition for reform 1. Formulate a strategic communications plan: aim for political support both at the highest levels and among the public (framing tobacco taxation as a health issue has helped win political support in many countries). 2. Identify champions in government: ensure that finance and health officials are on the same page; involve implementing departments, such as enforcement agencies, from the start. 3. Mobilize allies from academia, civil society and the private sector to counter the anticipated pushback from the tobacco industry, its proxies and its allies. Monitor and evaluate To make the most well-informed policy decisions, a reform effort should be monitored to assess its overall impact and its effect on key indicators; this will help identify issues to be fixed while also creating a strong evidence base for further reform efforts. Get and analyse the relevant data to better understand the market situation and its dynamics: 1. Monitor the market and its evolution (e.g. retail prices, duty-paid sales, market shares). 2. Get regular estimates of price elasticity (including cross-price elasticity), income elasticity and tax base elasticity to evaluate any changes in tobacco demand. Use relevant tools to assess the impact of the tax policy on consumption and revenue: 1. Use specific tools on the impact of excise on price, consumption and revenue (e.g. the WHO TaXSiM). 2. Use global tools to assess the tax increase’s impact on prevalence (e.g. the WHO ISPT). Monitor key indicators closely to assess improvements over time: 1. Tax as a percentage of retail price. 2. Change in affordability of tobacco products over time. 3. Change in the tobacco tax scorecard, which combines a mix of best practices in tax policy. 4. Change in sales, prevalence and illicit trade in tobacco products. 5. Improvements in MPOWER package achievement. STEP 5 STEP 6 STEP 7

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION

WHO TECHNICAL MANUAL ON TOBACCO TAX POLICY AND ADMINISTRATION WHO technical manual on tobacco tax policy and administration ISBN 978-92-4-001918-8 (electronic version) ISBN 978-92-4-001919-5 (print version) © World Health Organization 2021 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial- ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/ rules/). Suggested citation. WHO technical manual on tobacco tax policy and administration. Geneva: World Health Organization; 2021. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Document design by Ana Sabino. iii Contents Foreword v Acknowledgements vii Acronyms viii Executive summary xi CHAPTER 1. Why this manual? 1 CHAPTER 2. Tobacco excise tax policy 11 2.1 Global overview of tobacco tax practices 11 2.2 Designing excise tax policy 18 2.3 Domestic and regional policy integration 55 2.4 New and emerging nicotine and tobacco products 59 2.5 Conclusions 75 ANNEX 2.1 Countries that apply different types of excise tax structures 89 ANNEX 2.2 Analytics of the tax base elasticity 90 ANNEX 2.3 Elements of the devices that make up ENDS/ENNDS products 91 CHAPTER 3. Tobacco tax administration 93 3.1 Introduction 93 3.2 Institutional arrangements 94 3.3 The tax compliance cycle 102 3.4 Control and enforcement 115 3.5 Tax administration of other tobacco products 152 3.6 The broader elements of a good tax system 155 3.7 Conclusions 156 ANNEX 3.1 Composition of tobacco products 166 ANNEX 3.2 Example of forestalling and countermeasures 172 CHAPTER 4. Political economy 175 4.1 SCARE tactic S: Smuggling and illicit trade 176 4.2 SCARE tactic C: Court and legal challenges 198 4.3 SCARE tactic A: Anti-poor rhetoric or regressivity 212 4.4 SCARE tactic R: Revenue reduction 217 4.5 SCARE tactic E: Employment impact 228 4.6 Earmarking tobacco tax revenues to fund health 233 ANNEX 4.1 Methods to assess the nature and size of the illicit tobacco trade 253 ANNEX 4.2 How are the tobacco tax revenues earmarked? 269 CHAPTER 5. Best practices 271 TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) 283 iv v Foreword In 1999, the World Bank’s Curbing the Epidemic was the first report by an interna- tional organization to recognize that increasing tobacco excise taxes was the most effective and cost-effective measure to reduce tobacco use and save lives. Over the two decades since, the evidence base supporting this claim, especially in low- and middle-income countries, has been steadily growing. Meanwhile, the credibility of the tobacco industry’s arguments against tobacco taxation has been slowly waning. In short, health-promoting tobacco taxation has come of age, and the evidence has consistently shown that it is a win for public health, a win for revenue and a win for the economy overall. But we must be cautioned against complacency. Although the evidence on tobacco taxation is irrefutable and there are now signs that the tide is turning on the global tobacco epidemic, tobacco taxation was, in 2018, the WHO MPOWER1 measure that was least implemented at the highest level of achievement. Even more concerningly, cigarettes have become more, rather than less, affordable in many low- and middle-income countries over the past decade. Many countries set rates at insufficient levels and increase them too infrequently, while others still use complex and inefficient taxation structures. This failure to advance tobacco taxation able to effect significant price increases constitutes a loss for governments in revenues, a loss for public health and a win for the tobacco industry. To overcome this inertia, this manual charts the way forward for policy-makers, finance officials and others involved in tobacco tax policy development. It equips them with the information and evidence needed for the realization of their coun- tries’ tobacco tax policy objectives. It also analyses the tobacco industry’s tactics for influencing the political economy of tobacco taxation and shows the limitations and exaggerations of the arguments used against tax increases. The manual serves as an update of the 2010 WHO technical manual on tobacco tax administration by adding new evidence on the successes of tobacco taxation in all parts of the world and broadening its scope to capture more material relevant to developing and implementing more effective tobacco tax policy. 1 The WHO MPOWER package of technical measures and resources that comprises (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco. vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Its contributions are particularly timely in a COVID-19-stricken world. As the pandemic has been worsened by the global burden of noncommunicable disease, and revenue is now desperately needed, taxing tobacco should be more palatable than ever. This manual shows policy-makers how to seize this unique opportunity to use tobacco taxation to build back better, save lives and strengthen health systems while increasing revenue. Dr Naoko Yamamoto Assistant Director-General UHC/Healthier Populations Division vii Acknowledgements This manual was developed under the direction of Jeremias Paul Jr and Anne-Marie Perucic. The following contributed to the content of the manual (in alphabetical order): • WHO: Evan Blecher, Annerie Bouw, Mark Goodchild, Roberto Iglesias, Juliette McHardy, Jeremias Paul Jr, Anne-Marie Perucic and Robert Totanes. • External authors: Mauricio Cardenas (Columbia University), Sophia Delipalla (University of Macedonia), Luk Joossens (tobacco control expert, Belgium), Marin Kurti (Eastern Connecticut State University), Enrique Fanta (former World Bank senior specialist), David Merriman (University of Illinois at Chicago) and Jean Tesche (University of Cape Town). WHO would like to thank the following reviewers for their invaluable comments (in alphabetical order): • External: Jo Birckmayer (Bloomberg Philanthropies), Adriana Blanco Mar- quizo (WHO Framework Convention on Tobacco Control), Frank Chaloupka (University of Illinois at Chicago), Yoni Dekker (WHO Framework Conven- tion on Tobacco Control), Jeffrey Drope (University of Illinois at Chicago), Ceren Ozer (World Bank), Corne van Walbeek (University of Cape Town), Chonlathan Visaruthvong (Ministry of Finance, Thailand) and Rodrigo Santos Feijo (WHO Framework Convention on Tobacco Control). • WHO: – Headquarters: Douglas Bettcher, Itziar Belausteguigoitia, Ranti Fayokun, Joseph Kutzin, Benn McGrady, Vinayak Prasad and Susan Sparkes. – Regional offices: Nina Dela Cruz (WPRO), Fatimah El-Awa (EMRO), Charles Frasier (EMRO), Lee Lily Joung-Eun (WPRO), Jagdish Kaur (SEARO), Elizaveta Lebedeva (EURO), Maxime Roche (AMRO) and Rosa Sandoval (AMRO). WHO would also like to thank Amal Amoune-Naal for the administrative support, Alison Goldstein for the technical editing and Janet DeLand for the copy-editing. Production of this document has been supported by a grant from Bloomberg Phi- lanthropies. The contents of this document are the sole responsibility of WHO and should not be regarded as reflecting the position of Bloomberg Philanthropies. viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ACRONYMS AEO authorized economic operator AFRO WHO Regional Office for Africa AMRO WHO Regional Office for the Americas ATO Australian Taxation Office BAT British American Tobacco CCTV closed-circuit television CIF cost, insurance and freight COP Conference of the Parties CPI consumer price index CVA Customs Valuation Agreement (Thailand) DGCE Directorate General of Customs and Excise (Indonesia) DIY do-it-yourself ECBA World Bank Extended Cost-Benefit Analysis EIU Economist Intelligence Unit EMRO WHO Regional Office for the Eastern Mediterranean ENDS electronic nicotine delivery systems ENNDS electronic non-nicotine delivery systems EU European Union EURO WHO Regional Office for Europe FDA Food and Drug Administration (United States) FET fair and equitable treatment GCC Cooperation Council for the Arab States of the Gulf GDP gross domestic product HTP heated tobacco product HMRC Her Majesty’s Revenue and Customs (United Kingdom) IARC International Agency for Research on Cancer IIA international investment agreement IMF International Monetary Fund IRS Internal Revenue Service (United States) ISO International Organization for Standardization IT information technology ITC International Tobacco Control JTI Japan Tobacco International KRA Kenya Revenue Authority LMICs low- and middle-income countries ACRONYMS ix MFN most favoured nation MOP Meeting of the Parties (to the Protocol) MPOWER (M) monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco NCDs noncommunicable diseases NCI National Cancer Institute NT national treatment OECD Organisation for Economic Co-operation and Development OST other smoking tobacco PMI Philip Morris International PPP purchasing power parity QR quick response RGTE WHO Report on the global tobacco epidemic RYO roll-your-own SACU Southern African Customs Union SCARE (S) smuggling and illicit trade; (C) court and legal challenges; (A) anti-poor rhetoric; (R) revenue reduction; and (E) employment impact SDGs Sustainable Development Goals SEARO WHO South-East Asia Regional Office SII Internal Revenue Service of Chile TADAT Tax Administration Diagnostic Assessment Tool TTC transnational tobacco company UAE United Arab Emirates UHC universal health coverage VAT value added tax WAEMU West African Economic and Monetary Union WCO World Customs Organization WHO World Health Organization WHO FCTC WHO Framework Convention on Tobacco Control WHO ISPT WHO interactive smoking projection and target-setting tool WHO TaXSiM WHO tobacco tax simulation model WPRO WHO Regional Office for the Western Pacific WTO World Trade Organization x W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N xi Executive summary This WHO technical manual on tobacco tax policy and administration builds upon the 2010 WHO technical manual on tobacco tax administration by further detailing the strategies for effective tobacco tax policy development, design, implementation and administration. This 2021 edition also serves as an update to the 2010 manual, incorporating the latest developments in science, technology and policy, as well as providing illustrative recent examples from a variety of countries. The best practices laid out in this manual are designed to inform governments on the development of their tobacco taxation policy, facilitating the achievement of their health and revenue objectives while also supporting their overall development strategy. Tobacco taxes have long been seen as a source of revenue for governments, but as evidence of the harms caused by tobacco has accumulated over the years, public perception has evolved. Increasingly, governments, as well as the general public, are recognizing that taxation of tobacco is not only a revenue source but also an effective public health intervention to reduce tobacco consumption and its associated harms. The profile of tobacco taxation as a health policy tool has increased greatly since the publication of the 2010 WHO technical manual on tobacco tax administration. Multiple global commitments have been adopted over the past decade to address tobacco use specifically – as well as noncommunicable diseases (NCDs) and the Sustainable Development Goals (SDGs) more broadly – through tax and price measures to reduce demand for tobacco products, save lives and fund develop- ment. Global development institutions, including the World Bank, the International Monetary Fund (IMF) and major philanthropic foundations, also agree with WHO on the importance of emphasizing and strengthening tobacco taxation as a key health policy tool. The COVID-19 pandemic has further fuelled this shift in the narrative on tobacco taxation by revealing how the global economy is inextricably linked with population and planetary health. Investing in health is fundamental to any economic recovery, and fiscal policy will be a key driver in addressing the socioeconomic consequences of COVID-19. Interventions such as tobacco taxation – which leads to reduced tobacco consumption, improved population health and increased revenues for governments – should be part of a comprehensive strategy for a build back better recovery. The evidence is clear: significant increases in excise taxes that lead to price increases have consistently proven to be the most effective, as well as the most xii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cost-effective, mechanism for reducing tobacco consumption. This manual will guide readers through the necessary steps to create and implement the strongest tobacco taxation policies for their specific countries. There are many factors to consider when developing tobacco taxation policy. Understanding the market is a fundamental step before deciding which form of taxation to use. Specifically, the choice between ad valorem and specific taxation is influenced by the market structure in a given country. At the same time, tax structure also shapes the market structure. Over the past decade, there has been a trend towards countries adopting specific excise taxes or mixed excise systems that rely more heavily on the specific component, which the latest global data associate with the highest average prices. Tobacco tax structures can be simple, with one flat rate across the board, or com- plex, with multiple tiers for products with different characteristics. In 2018, 31 coun- tries used complex, multitiered structures. But evidence demonstrates that simpler excise tax structures – utilized in all high-income countries – leave the least room for industry manipulation or tax avoidance and brand/product switching by consumers. Not only is it important to set taxes at a high level to discourage consumption, specific excise tax policies must include regular adjustments to increase the tax rate so that it keeps up with inflation and income growth in a country over time. Excise tax increases should aim to reduce the affordability of tobacco products. The base on which the tax is applied is also important. For specific taxation, the tax base should be the quantity in clearly defined units. For ad valorem (or mixed) taxation, the best practice is to use the retail price as the tax base and introduce a minimum excise tax. With regard to non-tax regulations that affect the price of tobacco products, pric- ing regulation may be considered to prevent the tobacco industry from exercising differential tax shifting, which it uses to ensure that large price gaps exist between premium and cheap cigarettes. However, pricing policies cannot be used alone. If con- sidered, they should be used only as complements to significant excise tax increases. Other non-tax regulations include banning promotional discounts for tobacco products and banning the sale of single cigarettes. To assuage concerns that tax increases will increase inflation – as well as to reflect the declining trend in consump- tion of tobacco products – it is good practice to exclude tobacco products from the basket of items that are used to develop consumer price indexes. Finally, in order to make excise tax on tobacco products more effective in reducing overall tobacco use, all tobacco products must be taxed in a comparable way. Regular assessment, evaluation and monitoring of the impact of tobacco tax policies over time are essential components of effective tax policy development and analysis. Governments need to have accurate estimates of price, income and tax base elasticities in order to anticipate the impact of a tax increase on consumption and EXECUTIVE SUMMARY xiii tax revenue. Ideally, other factors such as non-price policies should also be taken into account when estimating price and income elasticities for a specific country. A variety of tools and indicators exist to measure impact and monitor progress, and these are described in Chapter 2 of this manual. When developing tobacco tax policy, it is also important to take the broader policy context into consideration at both the domestic and the regional level. Domestically, cooperation is needed across sectors to ensure that policies and interventions in the areas of agriculture, trade, finance and labour do not work against the public health objectives of tobacco control and taxation. For countries that are part of a regional bloc, regional harmonization of tobacco taxation is a useful tool to prevent tax revenue erosion, tax avoidance and tax evasion, as well as to protect population health. Tax harmonization must be designed carefully, however, to be effective. The experience of the European Union (EU) demonstrates that both a declining consumption trend and stable revenues can be achieved with harmonized minimum excise tax rates. Discussions of policy development and implementation for new and emerging nicotine and tobacco products such as heated tobacco products (HTPs) and electronic nicotine and non-nicotine delivery systems (ENDS/ENNDS), are complicated, by their constantly changing technology and market dynamics. Policies and regulations need to be developed carefully and adjusted accordingly. Where HTPs are not banned, the current recommendation is to tax them at the same level as cigarettes on a per-unit basis, regardless of tobacco content. Early evidence from the United States shows that demand for e-cigarettes, a subcategory of ENDS/ENNDS products, is possibly even more price-responsive than the demand for conventional cigarettes, meaning that taxes can be used as an effective deterrent to ENDS/ENNDS products use. While there is preliminary evidence of substitutability between conventional cigarette use and e-cigarette use, further research is needed to understand substitutability effects among users of both conventional cigarettes and ENDS/ENNDS products. It is essential to imple- ment regulation of ENDS/ENNDS products along with any tax policy to safeguard public health. In countries where they are not banned, ENDS/ENNDS products must be regu- lated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS taxation. Nicotine-containing and non-nicotine-containing e-liquids should be taxed equally. Ultimately, while the policy implications of these newer products require careful consideration, the fact remains that conventional tobacco products constitute the overwhelming share of consumption (more than 97% in 2018). Tobacco tax administration must be both efficient and effective to ensure that health objectives are met and the desired level of tax revenue is raised. Since the xiv W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation of tobacco taxation often involves numerous agencies within a country, clearly defined roles and responsibilities are essential to maximize efficiency. Coordination among the different agencies involved, as well as with neighbouring countries, is required for tobacco tax administration to be effective. Performance evaluation and accountability for competent authorities is also necessary, and many tools and indicators exist to facilitate these processes (described in Chapter 3). There are a number of steps authorities should take to ensure efficiency and effectiveness at each stage throughout the tax compliance cycle (see Chapter 3, section 3.3). Control and enforcement are the main functions of tax administration, and these can best be achieved through the use of a strategic plan and a risk-based approach. Controls can be exercised through licensing and due diligence, fiscal markings (e.g. tax stamps), tracking and tracing, implementation of anti-forestalling measures, national audits and specific controls for imports and exports, as well as for free zones and transhipment points. Once smuggling or illicit trade is detected, actions such as seizing and destroying smuggled and/or illicit tobacco and col- lecting due taxes must be taken immediately. To deter further illegal activities, a comprehensive audit must also be carried out, including all those involved in the illicit acts. Penalties and sanctions must be sufficient to deter illegal activities. The Protocol to Eliminate Illicit Trade in Tobacco Products provides invaluable guidance for tobacco tax administration, control and enforcement that is applicable even for countries that are not Parties to it. The broader elements of a good tax system include proper resourcing of competent authorities, strict rules and regulations to detect and punish corruption and a strong judiciary system capable of resolving disputes as soon as possible. In its efforts to oppose tobacco tax increases, the tobacco industry utilizes many SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric, (R) revenue reduction and (E) employment impact – to influence the political economy of tobacco. Chapter 4 provides detailed analyses of these issues with supporting evidence that belies the SCARE tactics, as well as guidance for tax and other relevant authorities on how to anticipate and respond to industry arguments. This manual also provides tools and methodologies to help tax authorities define and evaluate the problem of illicit trade of tobacco products in their countries, inde- pendent of the tobacco industry’s generally inflated estimates. Price (and tax) levels are not a key determinant of illicit trade; rather, the problem is exacerbated by the lack of governance and tax administration capacity. Refraining from increasing taxes is not the solution. Countries should instead respond with a comprehensive strategy to fight illicit trade, including undertaking independent estimates of illicit trade levels and implementing good tax administration practices such as those discussed in Chapter 3 and contained in the WHO Framework Convention on Tobacco Con- trol (WHO FCTC) Protocol to Eliminate the Illicit Trade in Tobacco Products. When it comes to court challenges, the tobacco industry is less likely to challenge excise taxes than other tobacco control measures, because taxation is a comparatively well-established regulatory measure. The industry will, however, exploit the slightest vulnerability in the design, adoption or implementation of tax measures. For this reason, measures to strengthen regulators’ legal position are described that will enable authorities to protect themselves from potential legal challenges. The industry argument of regressivity, or the notion that tobacco tax increases hurt the poor because they have to pay a larger share of their income in taxes than the rich, has two fundamental limitations. First, the notion of regressivity does not take into consideration the broader health and economic harms caused by tobacco use that exacerbate the impoverishment of lower-income smokers. These harms are actually reduced when tobacco consumption decreases following a tax increase. Second, the tobacco industry argument ignores the fact that higher tobacco taxes and prices can induce behaviour change as is reflected in the price elasticity of demand. Evidence consistently shows that lower-income smokers are more sensitive to price and therefore more likely to reduce smoking in response to a tax and price increase. Including these factors shows tobacco taxation to be, in fact, a progressive public health intervention that disproportionately benefits the poor. While essentially admitting that a tobacco tax increase may have the desired effect of reducing consumption, the industry also tries to argue that a tax increase will also reduce revenues. In fact, the price inelastic demand for tobacco makes tobacco tax increases a win-win for both public health and finance. This manual presents several country examples that demonstrate how well-designed and well-implemented tobacco tax increases lead to increases rather than decreases in revenue in the short to medium term. In addition, the reduced consumption resulting from a tax increase results in reductions of other tobacco-related government expenditures as well. The final tactic used by the tobacco industry to challenge proposed tax increases is to frame tobacco taxes as an economic rather than a public health issue. This false choice between health and jobs is based on faulty assumptions that 1) tobacco is a significant source of domestic employment; 2) job creation relies on tobacco consumption and 3) tobacco-related livelihoods are prosperous, sustainable and irreplaceable. Earmarking can be a useful tool for improving the political economy of tobacco tax increases. While the primary goal of tobacco tax increases is to reduce demand for tobacco, setting aside portions of tax revenue to fund other tobacco control efforts or relevant health programmes can help convince the public, politicians and officials of the value of significant tobacco tax increases. Earmarking can also be EXECUTIVE SUMMARY xv x vi W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N used to counter tobacco industry arguments about potential negative effects of tax increases – for example, by supporting tobacco farmers in transitioning to other crops. Tobacco taxes work. This is why the industry invests so much money and effort in blocking large tax increases and other effective tax policy reforms. Policy-makers must not be swayed by industry pressure but need only to follow the facts. This manual provides all the information policy-makers need to make the right deci- sions at each step of the process – from designing, evaluating, implementing and administering tax policy to refuting specious industry attacks and communicating the value of tobacco taxation to legislators and the broader population. An effectively designed and efficiently administered tobacco tax policy will not only produce the direct results of reducing tobacco consumption among smokers and raising revenue for governments, its effects will be felt much more broadly. Indeed, raising tobacco taxes is a SMART policy: it Saves lives; Mobilizes resources; Addresses health inequities; Reduces burdens on health systems; and Targets tobacco use, a major risk factor for NCDs. • Saves lives: Tobacco use is the leading cause of preventable deaths globally – it claims 8 million lives each year. Tobacco taxation is the most effective mechanism for reducing tobacco consumption and its associated health burden worldwide. • Mobilizes resources: Despite being the single most effective tobacco control measure, tobacco taxation is largely underutilized as a policy mechanism. Based on available data on the price and taxation of cigarettes, it is estimated that excise taxes on cigarettes generated a worldwide total of US$ 361 billion in revenues in 2018, including US$ 162 billion in revenues for low- and middle- income countries (LMICs). If all countries were to raise cigarette excise tax rates by the equivalent of US$ 1 per pack, the amount of excise revenue from cigarettes would increase by US$ 178–219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from such tax increases, with excise revenues increasing by 82–103%, providing governments in these countries with an extra US$ 133–167 billion. This shows the substantial revenue potential of tobacco taxes. • Addresses health inequities: Tobacco taxation and tax increases are effectively progressive or pro-poor policies because of their positive distributional impact. Lower-income smokers benefit disproportionately from reduced tobacco consumption and use in terms of health gains and income retention. • Reduces burdens on health systems: The worldwide economic cost of tobacco use was US$ 1.4 trillion in 2012. Tobacco taxes reduce tobacco-related burdens on governments and health systems through population-based preventive measures. • Targets tobacco use: Tobacco taxation directly targets and reduces tobacco use, which is a major risk factor for several deadly NCDs. In summary, significant tobacco tax increases, designed and implemented according to the latest guidance and best practices presented in this technical manual – and as a strong component of a comprehensive tobacco control strategy – will bring about substantial reductions in tobacco use and the health and economic harms it causes. EXECUTIVE SUMMARY xvii x viii W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 1 CHAPTER 1. Why this manual? BACKGROUND Tobacco taxes are not new. Governments around the world have been applying taxes on tobacco and tobacco products practically since the idea of excise was conceived. And rightly so: tobacco is not a necessity, it is easy to tax and the demand for it is relatively inelastic. These characteristics, along with the substantial revenues tobacco taxes generate, have made tobacco a highly appropriate object of taxation. As evidence of the harms of tobacco has accumulated over the years, the public perception of tobacco taxes has evolved. Now tobacco taxes are not only seen as a revenue source, but, more importantly, they are recognized as an effective public health intervention to reduce tobacco consumption. This trend reflects the reasons excise taxes exist in the first place – to discourage harmful behaviour and to mitigate the associated negative externalities (1–2). Many governments view tobacco taxes as a significant and stable source of rev- enue, which may explain why there is often a degree of hesitation whenever tobacco tax reform is proposed. Historically, many governments have relied on revenues from tobacco taxes and have even adjusted the level of taxation according to their revenue needs (3). However, some countries are beginning to recognize the value of applying high tobacco taxes primarily as a public health tool, viewing revenues as a secondary consideration (4). Arguments against tobacco tax hikes or improvements to the tax structure are often economic in nature: such tax changes will allegedly decrease revenues, wipe out jobs, increase illicit trade and harm local industries, among other claims. But the evidence has consistently shown that such claims are simply not true in an overwhelming majority of situations. The tobacco industry, in particular, frequently portrays this conflict as a false dichotomy between public health and the economy – as if prioritizing health comes at the expense of the economy. In fact, studies and real-world experiences have shown that increasing tobacco taxes not only improves public health but also has a net positive impact on the economy and development of a country – a true win-win scenario (5–6). As an update to the first WHO technical manual on tobacco tax administration published in 2010, this manual aims to help readers better navigate the various 2 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N issues surrounding tobacco taxes and their implementation. The primary intended audience includes policy-makers, finance officials, tax authorities, customs officials and other relevant persons/bodies involved in the formulation and implementation of tobacco tax policy. The manual provides a detailed guide to the design of tobacco tax policy and describes how to effectively administer these taxes to maximize impact. Detailed discussions of the political economy considerations and the hurdles that need to be overcome before and during implementation are included as well. The overarching goal is to equip those working in the tax policy and implementation spheres with sufficient information to help realize the health and revenue objectives of a government’s tobacco tax policy in line with its overall development strategy. THE CONSEQUENCES OF TOBACCO USE AND THE NEED FOR INTERVENTION Most people are aware that smoking and tobacco use are harmful to health, but few truly comprehend the scale of this harm. The tobacco epidemic claimed more than 100 million lives in the last century (7), with updated estimates now reaching 8 million deaths annually from tobacco use and exposure to second-hand smoke (8). As much as 80% of these deaths occur in low- and middle-income countries (LMICs) (6), revealing how the developing world carries much of the global burden. Tobacco use is a major risk factor for many chronic conditions, including heart disease, cancer, diabetes and chronic lung disease – collectively known as noncom- municable diseases (NCDs). NCDs account for about 15 million premature deaths (between ages 30 and 69) worldwide, killing people in their most productive years. As the leading cause of preventable deaths, tobacco use remains one of the foremost public health challenges of our time. The consequences of tobacco use also present enormous economic, development and social costs that wreak havoc on families, communities and societies. The annual economic cost of smoking was estimated at US$ 1.4 trillion in 2012, equivalent to 1.8% of the global gross domestic product (GDP) (9). With these figures likely to have increased since then, the massive health and economic burdens of tobacco use provide justification for governments to intervene and strictly regulate the market for tobacco products. The purview of tobacco control extends beyond the strong imperative to protect people’s health and well-being; it should also strive to contain the market failures and negative externalities of tobacco use, particularly since these effects can significantly impact a country’s development trajectory. The mounting evidence of the enduring destruction caused by tobacco in the 20th century provided compelling reasons for a strong global response, which led countries to negotiate the World Health Organization Framework Convention on Tobacco Control (WHO FCTC). The WHO FCTC came into force in 2005 as the CHAP T ER 1. WHY T HIS M ANUAL? 3 first public health treaty under the auspices of WHO. To facilitate its implementation at the country level, WHO packaged a set of demand-reduction measures directly taken from the treaty (7). These interventions, collectively known as MPOWER, are as follows: (M) monitoring tobacco use and prevention policies; (P) protecting people from tobacco smoke (smoke-free laws); (O) offering help to quit tobacco use (cessation services); (W) warning about the dangers of tobacco (including graphic pack warnings and plain packaging); (E) enforcing bans on tobacco advertising, promotion and sponsorship; and (R) raising taxes on tobacco products. Specifically, under Article 6 of the WHO FCTC, Parties recognized that price and tax measures are an effective and important means of reducing tobacco consumption for various segments of the population – in particular, among young persons (10). The severity of the tobacco epidemic and its ongoing damage to health and economies are clear justifications for governments to actively intervene and correct market failures. The scale of the burden and the rate at which lives are being destroyed necessitates urgent and aggressive action on tobacco control, using measures that most countries have committed to implementing and that are proven to be effective in reducing tobacco use. WHY TOBACCO (EXCISE) TAXES ARE CRUCIAL Among the different tobacco control interventions, raising excise taxes has been identified as the most effective as well as the most cost-effective measure to reduce consumption (6). While other interventions are certainly important components of a comprehensive tobacco control strategy, the direct impact of significant tax increases on consumption is by far the strongest. On average, a tax increase that causes prices to go up by 10% reduces consumption by 4% in high-income countries and 5% in LMICs (6). When implemented at scale, this demonstrates the enormous power of tobacco taxation and its potential to save lives. Tobacco taxes differ from other interventions in that their impact can increase and build over time – even if taxes are already relatively high, their rates need to be continuously increased to retain and amplify their effectiveness. However, this should not be taken as a suggestion that governments considering tobacco control interventions should focus solely on taxes. Taxes are even more effective when implemented as part of a comprehensive package of measures such as MPOWER, which covers distinct but complementary intervention points. Among the different taxes applied on tobacco products, excise taxes are the most significant because they raise both absolute and relative prices (6). This is important when considering health objectives, since it is the magnitude of the price increase of tobacco products that determines the reduction in consumption. An excise tax is typically applied on a limited set of products, designed to discourage their use by 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N raising the price significantly over that of other products available in the market. This is in contrast to value added taxes (VAT) or sales taxes, which apply to most goods and services. Seeking to raise the prices of tobacco products through VAT or sales taxes would fail to increase relative prices, making this an ineffective and inefficient method. Customs or import duties on tobacco products are also utilized, but their impact is waning with the global trend towards bilateral and regional agreements aimed at trade facilitation. The application of these duties varies across countries, but overall, they are not applicable to locally produced tobacco products. As a tool to increase prices, import duties cannot substitute for excise taxes, since they are not specifically designed to reduce consumption. When viewed as a public health policy tool, tobacco taxation is highly cost-effective, since it delivers significant impact yet is relatively inexpensive to implement (11). The costs of implementing tobacco taxation are much lower than those of clinical NCD interventions such as cancer treatments or maintenance medications, since the commodity and human capital requirements are less substantial (12). Moreover, increasing tobacco taxes actually generates additional revenue for a government. Tobacco taxes are also very effective in pre-empting or reducing consumption among groups of people who are especially price-sensitive – youth in particular, who are prevented from initiating a lifelong addiction if taxes and prices are sufficiently high (13). This is also true for the poor, who are more prone to catastrophic health expenditures than the wealthy are. Preventing initiation or encouraging cessation by imposing high taxes provides an escape route from the vicious cycle of tobacco use and poverty (6). A DECADE OF PROGRESS AND COMMITMENT TO ACTION Since 2010, when the first WHO technical manual on tobacco tax administration was published, numerous developments have raised the profile of tobacco taxes as an essential public health intervention. The Conference of the Parties (COP) to the WHO FCTC adopted guidelines for implementation of Article 6 of the treaty, which focuses on price and tax measures to reduce the demand for tobacco. Also within this period, three high-level meetings on the prevention and control of NCDs by the United Nations General Assembly, as well as the endorsement of the Global NCD Action Plan in 2013 by the World Health Assembly, have resulted in strong global commitments to implement measures, such as increased tobacco taxes to protect people’s health. The 2030 Agenda for Sustainable Development, which contains 17 goals known as the Sustainable Development Goals (SDGs), describes the global development strategy for the next decade. Within the SDGs, two specified targets are highly relevant for tobacco control: strengthening the implementation of the WHO FCTC (target 3.a) CHAP T ER 1. WHY T HIS M ANUAL? 5 and reducing premature mortality from NCDs by 30% (target 3.4). Furthermore, the Addis Ababa Action Agenda1, which aims to provide a global framework for financing the SDGs, also highlights tax and price measures on tobacco as key mechanisms to reduce demand and save lives while increasing domestic resources for develop- ment. Another important milestone was the 2018 entry into force of the Protocol to Eliminate Illicit Trade in Tobacco Products. These key events, along with several outcome documents and policy declarations in the area of tobacco control and the wider development sphere, have introduced tobacco taxation into the consciousness of a much larger share of policy-makers. As detailed in subsequent chapters, numerous countries have imposed sufficiently high tobacco tax rates while applying best practices in tax policy design and imple- mentation over the past decade (8, 10). For example, sustained and substantial tax increases have reduced tobacco use in LMICs such as Brazil (14), Turkey (15) and the Philippines (16). High-income countries also continued their leadership in this area, comprising 23 of the 38 countries judged to have sufficiently high tobacco taxes in 2018 (8). However, much remains to be done. The 2019 WHO report on the global tobacco epidemic (RGTE) shows that tobacco taxes are still the most underutilized tobacco control policy among the MPOWER measures (8), with only 14% of the world’s population being covered by sufficiently high tobacco taxes. Substantial progress has also been made in building the tools and evidence base for tobacco taxation. Volume 14 of the International Agency for Research on Cancer (IARC) handbooks of cancer prevention, Effectiveness of tax and price policies for tobacco control, published in 2011, is a key review of the literature published as of May 2010 on the effectiveness of tax and price policies in reducing tobacco use. The National Cancer Institute (NCI)-WHO Monograph on the economics of tobacco and tobacco control, published in 2016, details the evidence accumulated over the years from various countries, focusing not only on tax and price policies, but on all aspects of the economics of tobacco and tobacco control. In addition, numerous published studies from LMICs provide a comprehensive picture of the impact of tobacco taxation in different contexts. The updated Appendix 3 of the Global NCD Action Plan explains the cost-effectiveness of tobacco taxation (11), while the Global NCD Business Plan, Saving lives, spending less, built on this work by estimating a dollar figure for the return on investment expected from implementing the best-buy interventions for tobacco control, including taxation (12). The past decade has seen major steps forward for tobacco taxation in terms of global commitments, the number of countries implementing best practices and the 1 The Addis Ababa Action Agenda of the Third Conference on Financing for Development. Third Inter- national Conference, 13-16 July 2015, Addis Ababa, Ethiopia (https://sustainabledevelopment.un.org/ content/documents/2051AAAA_Outcome.pdf, accessed 17 February 2021). 6 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N expansion of the evidence base on tobacco taxes, particularly in LMICs. Within the development sphere, institutions such as the World Bank, the International Monetary Fund (IMF) and many other multilateral agencies are aligned with WHO on the importance of tobacco taxation and the need to improve its implementation (17–18). Considerable challenges remain; although it appears that the world is headed in the right direction, progress needs to accelerate at a much quicker pace in order to achieve the SDG targets by 2030. SHAPING A “NEW NORMAL” FOR TOBACCO TAXATION The global upheaval caused by the COVID-19 pandemic has cast an unprecedented spotlight on how well governments around the world prepared for and responded to the crisis. It has exposed glaring health systems vulnerabilities and highlighted the struggles of many countries to control the spread of the virus. But perhaps more than anything, the pandemic has demonstrated how the economy, trade, science, politics and many other aspects of our societies are very much interdependent and interconnected with the health of the population. It is clear that an individual’s state of health can significantly determine their susceptibility to disease and their ability to overcome it. People with NCDs are more vulnerable to becoming severely ill with a number of conditions, which also appears to be the case with COVID-19 (19). Tobacco use is a major risk factor for NCDs, and available research suggests that smokers are at higher risk of developing severe illness and dying from COVID-19 (20). Just as the different aspects of society are interconnected, so too are people’s health, the existence of health-promoting environments and the government policies and agencies that shape these environ- ments. This critical moment presents a unique opportunity and renewed motivation to discourage the use of harmful products such as tobacco and to further improve tobacco control measures, especially tobacco tax policy. Moving forward, a business-as-usual approach to tobacco taxation will not be sufficient. Responding to this new reality and preparing for the next pandemic entails implementing measures that promote healthier populations. Like the COVID-19 pandemic, any future pandemic will likely exacerbate health inequities, bring about more economic uncertainty and put pressure on governments’ fiscal capacities. Interventions such as higher tobacco taxes, which protect people’s health while generating more revenues and economic benefits, become even more important in such crisis situations. Given this context and the stakes involved, ministries of finance and tax authorities are in a unique and powerful position – one of saving not only livelihoods, but also lives. The importance of increasing tobacco taxes – one of the most effective public health tools available – cannot be overstated. The traditional approach of treating CHAP T ER 1. WHY T HIS M ANUAL? 7 tobacco tax exclusively as a revenue source has no place in the new normal. One cannot deny the scale of the tobacco epidemic, the necessity to correct market failures and the overwhelming evidence of tobacco taxation’s benefits to health and to the economy. The positive trend in the changing narrative around tobacco taxation needs to continue. Tobacco taxation should not be viewed in isolation from the rest of government policies, but rather as an important part of the whole, an essential piece in working towards our common goal of better health for all. OVERVIEW OF SUBSTANTIVE CHAPTERS This manual is primarily designed for policy-makers, finance officials, tax authorities and customs officials. It may also be useful for officials within health ministries or other government agencies, as well as nongovernmental organizations working in this area, including tobacco control advocates. Significant effort is made to present real-world examples and recent experiences from a wide range of countries to demonstrate success stories and lessons learned in raising tobacco taxes. A sub- stantial amount of evidence has been generated in LMICs over the past few years that supports and augments the existing evidence base, providing a much broader body of knowledge than was available when the first WHO technical manual on tobacco tax administration was released. Chapter 2 delves into the theory, practice and empirical evidence on tobacco excise tax policy, including current global trends. The chapter offers a detailed analysis of the various elements that constitute tax structure, aiming to provide policy-makers with a comprehensive understanding of the factors affecting prices, consumption and the market. It describes the key components to keep in mind when designing tobacco tax policy to maximize the impact of tax increases and improve the tax structure. The chapter also includes updated global price and tax data, specific examples from various countries and a discussion of tax base elasticity, automatic excise tax adjustments and pricing regulations, as well as descriptions of new and emerging nicotine and tobacco products, including electronic nicotine- and non- nicotine delivery systems (ENDS/ENNDS) and heated tobacco products (HTPs). Chapter 3 focuses on tobacco tax administration. It provides an in-depth discus- sion of the fundamental components that make tobacco tax collection effective and efficient, ensuring achievement of the health and revenue objectives of tax policy. It highlights the importance of cooperation among the various agencies involved in the implementation of tobacco taxes within countries and across borders. Building on country and regional experiences from previous decades, the chapter outlines specific measures and recommendations to maintain oversight of the whole tax compliance cycle. Also included are actions to facilitate control and enforcement, such as licensing, fiscal markings (e.g. tax stamps), tracking and tracing systems 8 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N and import and export controls. Finally, the discussion pivots towards the broader elements of a good tax system, such as proper resourcing of competent authorities, a strong judiciary and strict rules regarding corruption. Chapter 4 deals with the important challenges in the area of political economy when countries attempt to increase tobacco taxes or simplify the tax structure. The tobacco industry often relies on identified patterns of argumentation and tactics to obstruct such reforms, i.e. SCARE tactics – (S) smuggling and illicit trade, (C) court and legal challenges, (A) anti-poor rhetoric (regressivity), (R) revenue reduc- tion and (E) employment impact – each of which is discussed thoroughly in this chapter. Also included is a detailed discussion of the measurement of illicit trade and a discussion of earmarking tobacco tax revenue for health purposes. Finally, Chapter 5 presents a comprehensive list of the best practices in tobacco tax policy and administration discussed throughout this manual. The list is intended to serve as a practical guide and quick reference to the salient points presented. CHAP T ER 1. WHY T HIS M ANUAL? 9 REFERENCES 1. Ranson K, Jha P, Chaloupka FJ, Nguyen SN. The effectiveness and cost-effectiveness of price and other tobacco control policies. In: Jha P, Chaloupka FJ, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000:427–447 (https://www.paho.org/hq/dmdocuments/2010/ Cost-effectiveness%20of%20price%20increases.pdf, accessed 4 February 2021). 2. Effectiveness of tax and price policies for tobacco control. Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications.iarc.fr/Book-And-Report-Series/Iarc- Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And-Price-Policies-For-Tobacco-Control-2011, accessed 4 February 2021. 3. Tobacco taxation in the United States. In: Lynch BS, Bonnie RJ, editors. Institute of Medicine (US) Committee on Preventing Nicotine Addiction in Children and Youths. Growing up tobacco free: preventing nicotine addiction in children and youths. Washington (DC): National Academies Press (US); 1994 (https://www.ncbi.nlm.nih.gov/books/NBK236771/, accessed 10 November 2020). 4. Sin tax reform. Manila: Department of Finance (Philippines); 2012 (https://www.dof.gov.ph/advocacies/ sin-tax-reform/, accessed 10 November 2020). 5. Goodchild M, Perucic AM, Nargis N. Modelling the impact of raising tobacco taxes on public health and finance. Bull World Health Organ. 2016; 94:250–7 (https://www.who.int/bulletin/ volumes/94/4/15-164707.pdf, accessed 4 February 2021). 6. The economics of tobacco and tobacco control. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute, NIH Publication No. 16-CA-8029A; 2016 (National Cancer Institute tobacco control monograph 21; https://cancercontrol.cancer.gov/ brp/tcrb/monographs/monograph-21, accessed 17 December 2020). 7. WHO report on the global tobacco epidemic, 2008: the MPOWER package. Geneva: World Health Organization; 2008 (https://www.who.int/tobacco/mpower/mpower_report_full_2008.pdf, accessed 10 November 2020). 8. WHO report on the global tobacco epidemic, 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/tobacco/global_report/en/, accessed 10 November 2020). 9. Goodchild M, Nargis N, Tursan d’Espaignet E. Global economic cost of smoking-attributable diseases. Tob Control. 2018;27:58–64 (https://tobaccocontrol.bmj.com/content/27/1/58, accessed 4 February 2021). 10. Guidelines for implementation of Article 6 of the WHO FCTC. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 4 February 2021). 11. Tackling NCDs: ‘best buys’ and other recommended interventions for the prevention and control of noncommunicable diseases. Geneva: World Health Organization; 2017 (https://apps.who.int/iris/ handle/10665/259232, accessed 10 November 2020). 12. Saving lives, spending less: a strategic response to noncommunicable diseases. Geneva: World Health Organization; 2018 (https://www.who.int/publications/i/item/WHO-NMH-NVI-18.8, accessed 4 February 2021). 13. Chaloupka FJ, Warner KE. The economics of smoking. In: Culyer AJ, Newhouse JP, editors. Handbook of health economics. Elsevier; 2000;1(1):1539–1627. 14. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Rev Panam Salud Publica. 2016;40(4):243–9 (https://iris.paho.org/bitstream/ handle/10665.2/31306/v40n4a09_243-9.pdf?sequence=1&isAllowed=y, accessed 17 February 2021). 15. Cetinkaya V, Marquez PV. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 16. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington (DC): World Bank Group; 2016 (http://documents.worldbank.org/curated/en/638391468480878595/pdf/106777-PUB-PUBLIC- PUBDATE-7-26-2016.pdf, accessed 10 November 2020). 17. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 10 November 2020). 10 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 18. Irwin A, Marquez P, Jha P, Peto R, Moreno-Dodson B, Goodchild M, et al. Tobacco tax reform: at the crossroads of health and development – a multisectoral perspective. Washington (DC): World Bank Group; (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report- CrossRoads.pdf, accessed 2 February 2021). 19. Information note: COVID-19 and NCDs. Geneva: World Health Organization; 23 March 2020 (https:// www.who.int/publications/m/item/covid-19-and-ncds, accessed 10 November 2020). 20. WHO statement: tobacco use and COVID-19. Geneva: World Health Organization; 11 May 2020 (https:// www.who.int/news-room/detail/11-05-2020-who-statement-tobacco-use-and-covid-19, accessed 10 November 2020). 11 CHAPTER 2. Tobacco excise tax policy 2.1. GLOBAL OVERVIEW OF TOBACCO TAX PRACTICES A well-designed tax policy is key to having an effective tax policy. Any government that is planning to reform its tax policy must first understand the fundamental components of a good tax policy, as well as consider the strengths and weaknesses of different approaches to taxation, how they impact price and the requirements for tax administration. Understanding how the tobacco market operates in a country is equally important for policy-makers because of the inevitable interaction between the market and tax structures. Beyond the political considerations that strongly influence tobacco tax policy development, this chapter focuses on the technical aspects of tobacco taxation – excise tax in particular. Section 2.1 provides an overview of tobacco tax practices at the global level, focusing on the different ways countries structure excise tax. Section 2.2 emphasizes the importance of carefully designing excise tax policy, highlighting not only the significance of tax increases but also excise tax structure and its impact on prices, taking into account how market structure influences trends. This section also discusses the importance of measuring impact as another aspect of tax policy development, and it presents the crucial elements for performing measurement, as well as the relevant indicators available to monitor progress. Section 2.3 describes external policy considerations in the design phase to ensure that the goals of tobacco control and taxation are achieved. Intersectoral policy integration and coherence at the domestic level is discussed as a strategy to ensure that policies of other sectors do not inhibit or obstruct public health policy objectives. This section also reviews the current state of regional tax harmonization based on the experience of existing regional blocs and draws conclusions on the best policy approaches to preserve the public health interests of individual countries. Section 2.4 discusses new and emerging nicotine and tobacco products, in par- ticular HTPs and ENDS/ENNDS. It reviews the latest evidence on the health impacts of these products and current approaches to regulation. Key policy considerations are identified, and recommendations are provided for adopting an appropriate excise tax policy for these products. Section 2.5 summarizes the issues covered in the chapter and the key takeaways. 12 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.1.1 TAXES APPLIED ON TOBACCO PRODUCTS Taxes are classified as either direct or indirect. Direct taxes are imposed on the profit, income, property or wealth of persons or companies, whereas indirect taxes are imposed on the price of goods and services. Indirect taxes are most relevant to tobacco products taxation, because they directly influence price. A variety of types of indirect taxes can be applied to tobacco products. These include: • excise taxes – taxes that apply to a few selected commodities (they can also be applied to alcohol, fuel, sugar-sweetened beverages, etc.). • VAT or sales taxes – VAT is a multistage tax on all consumer goods and services that is applied proportionally to the price the consumer pays for a product. It is a tax on the amount by which the value of an article has been increased at each stage of its production or distribution. Some countries impose sales taxes instead of VAT. Unlike VAT, which is collected at every stage of the supply chain, sales taxes are generally levied at the point of retail on the total value of goods and services purchased. Ultimately, the consumer ends up paying the tax, whether it is a VAT or a sales tax. • import duties – taxes on selected goods imported into a country to be consumed in that country (i.e. goods that are not in transit to another country). In general, import duties are collected from the importer at the point of entry into the country. • other taxes – other indirect taxes, such as environmental taxes, that do not fall into any of the categories listed above. One of the most well-established and widely understood points in tax policy is that tobacco products should be subject to excise taxation. The focus of this chapter – and of this manual overall – is on excise taxes. They are the most important type of indirect taxes for tobacco control because they are applied directly to tobacco products and contribute the most to increasing the price of tobacco products relative to other goods and, subsequently, to reducing consumption. There are two basic types of excise taxes: • specific – levied as a monetary value per quantity of the product being taxed (e.g. 1 000 cigarettes, pack of 20 sticks, kilogram of tobacco); and • ad valorem – levied as a percentage of the value (e.g. retail price, or the producer/ex-factory price or the cost, insurance and freight [CIF] value1) of the product being taxed. These types of excise tax can be applied at a uniform or a differential (tiered) rate and on their own or in combination (i.e. a mixed system). 1 CIF is the value of an imported product as declared to customs upon entry into a territory. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 13 Given the widespread use of cigarettes – almost the only tobacco product used in some parts of the world – and the scarce availability of data for other tobacco products, this chapter focuses mainly on cigarettes. But there are a few examples and recommendations for other tobacco products, including those that are more prevalent in specific parts of the world (e.g. bidis or smokeless tobacco in South-East Asia and waterpipe tobacco in the Eastern Mediterranean region). 2.1.2 CIGARETTE TAXES AND RECENT TRENDS WORLDWIDE Tax and national income levels: the higher the income level, the higher the taxes and prices At the global level, cigarette price and tax levels correlate positively with a country’s income level: prices and taxes are higher in higher-income countries and lower as income level decreases. This trend has not changed over the years since 2008. Figure 2.1 presents the levels of price and tax by income groups for 2018, using the World Bank classification of income groups. Fig. 2.1 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by income group, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in purchasing power parity (PPP) adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Excise tax per pack Other taxes Retail price minus taxes Pr ic es a nd ta xa tio n pe r p ac k (P PP $ ) High-income Low-income 4.25 0.68 Middle-income 2.06 Total taxes: 5.30 (67.9% of pack price) Total taxes: 2.91 (58.3% of pack price) Total taxes: 1.18 (38.1% of pack price) PPP $ 7.80 PPP $ 4.99 PPP $ 3.09 Global 2.48 PPP $ 5.53 Total taxes: 3.36 (60.8% of pack price) 0 1 2 3 4 5 6 7 8 14 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tax and price levels: the higher the tax share, the higher the price Globally, cigarette prices correlate positively with tax percentage levels: as the total tax share (of which excise represents the largest part) as a percentage of retail price increases, the price of cigarettes generally also increases (see Fig. 2.2 below).2 This indicates that taxes do influence prices. Fig. 2.2 Weighted average retail prices and taxation (excise and total taxes) of most-sold brand of cigarettes, by total tax levels, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Tax level and WHO regional classification: taxes and prices are highest in the European region, followed by South-East Asia, the Americas and the Western Pacific, with the lowest levels in the Eastern Mediterranean and African regions At the regional level (WHO regional classification), average levels of prices and taxes vary greatly. The highest level can be seen in the European region, which includes the European Union (EU) countries. The EU’s unified tax structure includes high levels of minimum taxes – which lead to high prices – and encourage member 2 This is a general trend and does not apply for every country; there are countries that have a large tax share but low prices for cigarettes. Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax ≥ 75% Total tax ≤ 25% 0 7 4 1 2 3 6 5 50% ≤ Total tax < 75% 25% ≤ Total tax < 50% Excise tax Other taxes Retail price minus taxes 4.44 2.24 1.57 PPP $ 5.07 PPP $ 5.33 PPP $ 7.07 PPP $ 2.60 0.17 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 15 countries to regularly increase their taxes to meet their obligations. However, with the current minimum level now being reached by all EU member countries, the motivation to increase excise taxes may wane. Indeed, the minimum cigarette excise amount in the EU has not been adjusted since it went into effect on 1 January 2014, and it is suffering from inflation erosion. Member States of the EU acknowledged this in June 2020 by stating that action at the EU level is required to ensure that minimum excise duty rates regain traction to effectively reduce the consumption of tobacco products and that the minimum rates of excise duties on a number of tobacco products would be increased (2). Excise taxes are lowest in the African and Eastern Mediterranean regions. And China – reported separately due to its size – has lower tax rates than the Western Pacific region (see Fig. 2.3). Fig. 2.3 Weighted average retail prices and taxation (excise and total) of most-sold brand of cigarettes, by region, 2018 Notes: China is represented separately from the Western Pacific Regional Office (WPRO) average because of its exceptionally large number of smokers compared with the number in other countries in the region. AFRO is the African Region, AMRO is the Region of the Americas, EMRO is the Eastern Mediterranean Region, EURO is the European Region, SEARO is the South-East Asia Region. Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Retail price Excise tax Total tax share % AFRO AMRO EMRO EURO SEARO WPRO w/o China China All 3. 80 4. 02 4. 24 4. 89 4. 02 1. 45 2. 89 2. 65 2. 48 7. 53 7. 27 5. 61 2. 51 1. 95 1. 05 42.2% 55.8% 61.2% 72.9% 63.2% 55.7% 60.8% 56.9% Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Total tax share % 5. 53 0 1 2 3 4 5 6 7 8 9 10 40 30 50 60 70 80 20 16 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Global tax structures trend: more countries are adopting specific excise taxes or mixed excise systems that rely more on the specific component Since 2008, the number of countries that rely solely on ad valorem taxes or apply no excise taxes at all has decreased as more countries have adopted specific or mixed systems. More of the countries that have implemented a mixed system have increased the specific component of the tax structure relative to the ad valorem component (see Figs. 2.4 and 2.5).3 Fig. 2.4 Changes in excise tax structure, 2008–2018 Fig. 2.5 Changes in reliance on specific versus ad valorem component in mixed systems, 2008–2018 Source: (1). 3 For information about countries that applied each type of excise tax structure in 2018, see Annex 2.1. N um be r o f c ou nt ri es Specic excise Ad valorem Mixed excise No excise 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 54 56 57 63 50 45 24 23 24 21 19 15 49 47 44 42 41 55 57 59 56 60 63 62 Mixed excise Relying more on specic Relying more on ad valorem N um be r o f c ou nt ri es 2008 2010 2012 2014 2016 2018 10 20 30 40 50 60 70 45 22 23 23 24 22 26 27 27 27 32 35 37 50 54 56 57 63 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 17 The imposition of a minimum specific excise tax: half of the countries that have a mixed or ad valorem structure impose an excise tax floor A minimum specific excise tax ensures that at least a certain minimum amount of tax is paid, irrespective of price level. Almost half of the 101 countries that impose either ad valorem or a mixed excise for which data on minimum excise are available (47 countries) set a minimum specific excise tax.4 Nearly two thirds of those that set a minimum specific excise tax (29 countries) are high-income countries; most of them are in the EU, which requires its members to impose a minimum specific excise tax. The choice of tax base worldwide: almost half of the countries that apply a mixed or ad valorem excise system use retail price as the base Setting the base for applying a specific excise is relatively easy: most countries use a defined quantity of sticks for cigarettes, the weight in kilograms for tobacco and the weight in grams for other tobacco products (1). Different bases for ad valorem excises are applied in different countries. Nearly half of the 105 countries that implement either ad valorem or a mixed excise for which data are available (47 countries) use the retail price5 as the tax base for the ad valorem part, and most of those (28 countries) are high-income countries. Using the retail price as the base for the excise ad valorem tax is more effective than using the producer price or the CIF value. Unlike retail prices, which are easy for tax administrators to ascertain by monitoring the market, the producer price or CIF value is prone to undervaluation by producers or importers, who may pass on their margins to related parties further down the supply chain and successfully reduce their tax burden. This tactic is also known as transfer pricing. Additionally, global-level data show that the excise ad valorem on the retail price seems to lead to higher retail prices on average compared with an excise ad valorem applied on other bases, such as the producer price or CIF value (see Fig. 2.8 below). On complex tiered structures: 31 countries still apply complex, multitiered excise taxes on tobacco products As of 2018, 31 countries imposed excise taxes that varied according to defined char- acteristics of cigarettes, including price level, type of production, type of package and length of cigarette (Table 2.1). Some countries use more than one criterion to differen- tiate the tax rates. Indonesia, for example, imposes differential rates based on volume 4 This means that countries with a mixed system impose an overall minimum specific excise tax (where the yield of the specific plus the ad valorem excise cannot be below the set minimum specific excise tax), in addition to the excise on a specific component. 5 Countries that impose ad valorem on retail price exclusive of VAT are also included, since retail prices are easy to determine and VAT rates are known variables. 18 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N produced, type of cigarette and price level. In other countries, such as Member States of the EU, differential rates for cigarettes are prohibited by law, and the rate of the ad va- lorem tax and the amount of specific excise duty must be the same for all cigarettes (3). Table 2.1 Criteria used by countries for tiered excise taxes, 2018 BASE OF TIERS COUNTRY Retail price Bangladesh, Belarus, Indonesia, Jordan, Mozambique, Myanmar, Pakistan, Thailand Cigarette grade (e.g. premium, mid-grade, economy) Egypt, Japana, Mali Producer price China, Lao People’s Democratic Republic Production volume Indonesia Type filter/non-filter Belarus, Georgiaa, India, Kenya, Republic of Moldova, Nepal, Papua New Guinea hand/machine made India, Indonesia kretek/white cigarette Indonesia tobacco content (dark/ blonde or dark/light) Algeria, Bolivia (Plurinational State of ) Packaging soft/hard Mozambique, Uganda Cigarette length India, Nepal, Sri Lanka Trade (domestic/imported) Iran (Islamic Republic of ), Lebanon, Myanmar, Solomon Islands, Tonga, Uzbekistan Leaf content (domestic/imported) Fiji, United Republic of Tanzania a Japan and Georgia were using a tiered excise tax structure when these data were collected in 2018, but as of 2020, that is no longer the case. Source: (1). 2.2 DESIGNING EXCISE TAX POLICY Significantly increasing the taxes on and prices of tobacco products is the most effective and most cost-effective policy to control tobacco use (4). Increased taxes that are passed on to tobacco users as higher prices reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face challenges ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues such as addressing the SCARE6 tactics of the tobacco industry. This section provides guidance for policy-makers regarding the best tax structure to use from a health perspective, taking into consideration all the appropriate tax designs. It also proposes recommended indicators to consider when formulating policy change. 6 SCARE tactics are the tactics most commonly used by the tobacco industry when countries plan to increase tobacco taxes. They are described, and refuted, in detail in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 19 2.2.1 THE IMPORTANCE OF THE TYPE OF EXCISE TAX STRUCTURE The existing theoretical and empirical evidence on approaches to the choice of (uniform) specific and ad valorem excises is reviewed below, along with their effects on price, consumption, perceived quality and variety of tobacco products, govern- ment revenue and tax administration. The use of the word “quality” in this chapter does not refer in any way to the health impact of a tobacco product. It refers rather to the consumers’ perceptions of quality and their decision to buy a product, which they may evaluate based on the packaging, the blend used for the cigarette or anything that makes the product more appealing to them. Just to be clear, from a public health perspective, all cigarettes are equally harmful even if perceived by consumers as having higher or lower quality. The choice between ad valorem and specific taxation is influenced by the market structure, i.e. the nature and degree of competition in the market for goods and services. Although each country has its own specific characteristics, the tobacco market structure is typically a monopoly or an oligopoly where firms have the power to control prices – and hence exploit the tax structure – to their benefit. For example, China, the largest producer and consumer of tobacco products in the world (5), has a state monopoly. In Viet Nam, foreign brands are produced under licence by the state monopoly. In Thailand and Egypt, despite the presence of foreign companies, the market is dominated by the state-owned company. In Uruguay, the oligopoly is led by a domestically owned company. In Bangladesh, the oligopoly consists of domestically owned companies competing with foreign companies (6). In most of Africa, the market consists of transnational tobacco companies (7). The impact of tax structure on final price: uniform specific versus uniform ad valorem The choice between specific and ad valorem taxes is a long-standing issue in tax policy, as the level and structure of excises have different implications for the interests and goals of various groups. Given the market structure of the tobacco industry – typically a monopoly or oligopoly for most products in most countries – different excises may have different effects on government revenue, manufacturer profit, consumer price, perceived product quality and variety and tax administration (8–16). Consequently, the two types of excise taxes – specific and ad valorem – may have different implications for public health to the extent that they affect individual consumption via their impact on perceived product quality, variety and prices. Moreover, governments have the potential to influence tobacco excises to manage demand, raise revenue and promote public health. 20 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax on prices and in turn on consumption is also influenced by price and income elasticities, as well as consumer perceptions of quality (“perceived quality”) and the variety of available products, which, in turn, are closely linked to the type of tax structure adopted. Tax structure is affected by both the price elasticity of demand and the price elasticity of supply. The price elasticity of demand measures the responsiveness of consumer demand to changes in prices. The price elasticity of supply measures how sensitive producers are to changes in prices. Tobacco tax structure is also influenced by market structure. In a monopoly, the profit-maximizing firm sets the price, considering the price elasticity of demand: the lower the price elasticity (in absolute value) – i.e. the less sensitive the consumer is to price changes – the higher the price the monopolist can set. Profits are typically abnormal in a monopolist market structure, meaning total sales revenue is higher than total cost (where total cost includes a normal profit). A monopolist producer therefore receives more than the minimum reward required to invest its (physical and human) capital and undertake business risks. Economic theory predicts that in a private monopoly, prices are higher than in an oligopolistic market. This is not, however, necessarily true when the monopoly is owned by the state and the government’s objective is not straightforward profit maximization: the government might have other considerations, such as preserving jobs (e.g. in China) or keeping prices low for low-income consumers (e.g. in Egypt). Under a monopoly, an ad valorem taxation structure enables the monopolist producer to set prices lower than would be possible under a specific tax structure. This is feasible because under ad valorem taxation, when supply increases and price falls, the price reduction is not fully borne by the producer. Rather, the price reduc- tion is partly shared by the government since, as supply increases, the tax per unit of product sold falls. In other words, ad valorem taxation leads to lower prices and higher consumption relative to revenue-equivalent specific taxation. Technically, this means that the supply function is less elastic under ad valorem taxation. In contrast, under a specific taxation structure, any increase in the monopolist producer price will go to the producers as revenue, which incentivizes them to increase prices. The same logic also applies to an oligopolistic market structure, where profits again are, in general, abnormal. KEY TAKEAWAY 1 In a monopoly or an oligopoly, specific taxation incentivizes industry to set prices higher than it would with ad valorem taxation. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 21 Understanding oligopolies, however, is more complicated, since they are characterized by strategic interdependence among a few firms. This strategic interdependence extends to the relationship between industry and regulators. Industry anticipates the government’s regulatory policy – whether through tax or other interventions – and acts accordingly. For example, competitors may coordinate and lobby against a certain tax structure reform or tax rate increase. Under an oligopoly market structure, ad valorem taxation is a relatively more efficient tool for transferring part of the profits to the government as tax revenue, since it acts like both an excise and a profit tax. In contrast, a specific tax has a smaller (negative) effect on profits. This explains why we observe multinationals that are leaders in high-priced brands (e.g. Philip Morris International [PMI]) lobbying in favour of specific taxation (17). As an example, in the countries of the Cooperation Council for the Arab States of the Gulf (GCC), the tobacco industry has been trying for a long time to lobby governments to introduce a specific excise (18–21). After years of consideration and discussions on the possible introduction of excise taxes, the GCC adopted the Common Excise Tax Agreement of the States of the Gulf Cooperation Council in November 2016 (21), which introduced an ad valorem excise on tobacco products. Tobacco companies’ support for excise tax structures ultimately depends on the market segments they control in a particular country. A company selling mainly premium brands will favour specific excises, whereas a company that sells mid-priced or economy brands would favour ad valorem excise (17). When oligopolistic firms produce identical products, a specific tax has a stronger positive effect on price and is more likely to be overshifted to consumer prices than an ad valorem tax (13). Overshifting means that the price increases by more than the tax increase itself. Empirical evidence supports this (22–26). KEY TAKEAWAY 2 In an oligopoly, prices are likely to increase by more than the amount of the specific tax increase when demand is relatively inelastic. In general, demand for a product depends not only on prices but also on consumer perceptions of quality and preferences for variety. For example, the most popular brand in GCC countries is Marlboro, a premium brand (1). Consumers differ in their willingness to pay, depending on their respective perceptions of quality, which influence whether they ultimately purchase high- or low-priced brands. A tax-induced price increase can cause the following plausible responses from consumers or users of tobacco products: (1) a group of consum- ers will quit; (2) a group of consumers will reduce their overall consumption; 22 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N (3) another group, most likely high-income users, will switch to an upgraded version of the tobacco product if its relative price (compared to the cheaper brand) has been reduced, which is the case under a specific tax increase; there might also be a group of consumers in the lower income range who switch to lower-priced variants of the tobacco product if the price gap increases, as is the case under an ad valorem tax increase; and (4) another group might switch to the illegal market or buy products in a neighbouring country with a lower tax where possible. Consumers’ decisions to purchase are also affected by their preferences for variety – meaning preferences among products that consumers perceive as equal in quality but are given different characteristics by the producers to account for consumer taste preferences. Thus, it is possible that a tax increase that leads to an increase in average prices will lead to an increase in the total quantity demanded in the market because of an increase in the variety of product choices available to consumers. Variety enables new consumers to be captured, especially in an environment lacking certain regulations (e.g. without plain packaging and flavour bans). The tobacco industry was able to capture a new group of consumers when it introduced menthol cigarettes into the tobacco market. There is more than sufficient evidence that menthol cigarettes increased youth smoking initiation, increased nicotine dependence and reduced adult smoking cessation (27). To prevent this from happening in their countries, Member States of the EU have prohibited characterizing flavours other than tobacco in tobacco products (28). It is therefore important to consider the broader effects that the structure and level of an excise tax can have on average price, perceived quality and the variety of cigarette brands and other emerging substitutes. When consumers make choices based on dimensions other than quantity, the two types of tax structures are not equivalent, even in a perfectly competitive market where firms have no market power (29–30). To illustrate this point, consider a US$ 1 cost to improve consumer perceptions of quality for a tobacco product. This will lead to an equivalent price increase under specific taxation but not under ad valorem taxation. At an ad valorem rate of 20%, the price must increase by more than US$ 1, or by 1/(1 – 0.2) to cover the US$ 1 cost of improvement, due to the multiplier effect. A specific tax induces consumers to reduce the quantity demanded, but they might still choose to pay a higher price in exchange for a product that they perceive to be of better quality. An ad valorem tax, on the other hand, leads to a reduction in both quantity and perceived quality, not a substitution between them. An ad valorem tax has only an income effect and – unlike specific taxation – does not lead to substitution between perceived quality and quantity. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 23 KEY TAKEAWAY 3 Under specific taxation, the industry has incentives to create upgraded variants of tobacco products that attract new consumers and encourage consumption. When firms produce differentiated products, as the tobacco industry does, economic theory provides ambiguous results regarding specific versus ad valorem taxation. The relative effects of the two types of tax are not as straightforward as in the case of oligopolistic firms producing a homogeneous product. With differentiated products, the relative effects of the tax types depend on various assumptions: whether or not firms face symmetric costs, whether the number of firms is fixed or new firms can enter the market and the level of the tax revenue requirement. When firms face different costs, ad valorem taxes exacerbate the absolute differences in marginal costs between them. The high-cost brand is not considered a perfect substitute for the low-cost brand. A sufficiently high ad valorem tax rate may lead to a relative underproduction of the high-cost products (31). The effect of specific and ad va- lorem taxes on consumer perceptions of quality depends on market structure and the price and income elasticities of demand across various qualities. The relative price of the cheapest product does not necessarily remain unchanged or increase; it might fall (32). Empirical evidence showing that increases in the specific tax lead to a lower market share for the cheaper generic brands and an upward shift to premium brands (33–34) usually considers gradual tax increases and ignores income effects. Chaloupka et al. (35) found that in 21 EU countries that impose a mixed tax system, the price gap between premium and low-priced brands – while not reflecting the full distribution of cigarette prices – is smaller when the specific component of the mixed structure dominates.7 Although the price gaps are narrower under specific taxation, there is evidence that firms sometimes respond by introducing new, very cheap (subvalue) brands, or they exercise differential tax shifting. This practice has been evident in India for quite some time, with the Indian Tobacco Company launching a number of cheaper variants of its flagship cigarette brand, Gold Flake, to take advantage of a lower excise tax rate in the so-called microcigarette (< 60 mm length) market (36). Consequently, the cheapest end of India’s cigarette market has expanded significantly in recent years due in part to the marketing of new brand variants like Gold Flake Century. 7 The EU countries impose a mixed tax structure with a minimum tax floor. Some countries rely on the specific component more than others, but they remain within a given range (the specific component must be between 5% and 76.5% of total tax share of the weighted average price). 24 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As another example, transnational tobacco companies, which have sold ultra- low-priced brands in the United Kingdom since 2006, have managed to double their market share in a few years: their real price did not increase, since they absorbed part of the tax increases (37). The share of ultra-low-priced brands increased between 2001 and 2009 from 5% to 10%, while the market share of economy brands increased from 40% to 50% and the market share of premium brands and mid-priced brands decreased during the same period (from 35% to less than 25% for the former and from 15% to 5% for the latter). In order to keep the price of discount brands low and certain consumers in the market, firms may overshift the tax for premium products and undershift it for the lower-priced products (37–40). KEY TAKEAWAY 4 Evidence suggests that the price gap between brands is narrower under a specific tax structure. As the tobacco industry simultaneously consolidates producers and widens its portfolio of products, evidence is emerging that it is introducing cheaper brands while increasing the price of its expensive brands, therefore paradoxically widening the price gap within its products. The extent of the impact is still unclear, however, and this evidence does not negate the overall conclusion that a specific tax structure reduces price gaps. The impact of tax structure on final price: uniform specific, ad valorem and mixed systems Evidence from the 2019 RGTE (1) data suggests that the average price of the most- sold brand of cigarettes – weighted by the number of smokers – is the highest in countries implementing a mixed system that relies more on specific excise, followed by countries applying specific excise taxes only, followed by countries applying a mixed system that relies more on ad valorem and then by countries that apply ad valorem excise only (Fig. 2.6). The price is lowest in countries that have no excise at all. In past WHO reports on the global tobacco epidemic, countries that applied specific excise only had the highest price, on average. The trend may have changed partly because more countries are adopting mixed excise systems. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 25 Fig. 2.6 Weighted average price and excise for a pack of the most-sold brand of cigarettes, by excise tax structure, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Shang et al. (41) describe and compare price distributions, using data from 16 countries of the International Tobacco Control (ITC) Project that impose different cigarette tax structures. Specific uniform taxation tends to result in less variability in prices than all other structures (ad valorem tax, mixed tax, tiered tax). In general, structures other than uniform specific tax give rise to more opportunities for brand switching and tax avoidance. Reliance on complicated systems is likely to be as- sociated with wider price distribution, leading to greater tax avoidance, as there are more opportunities for substitution with cheaper brands when taxes rise. KEY TAKEAWAY 5 Evidence suggests that the tax structures most likely to lead to higher prices are uniform specific excise tax structures or mixed systems that rely more on specific excises. Mixed system relying more on specic excise Pr ic e an d ta xa tio n pe r p ac k of 2 0 st ic ks (P PP $ ) Mixed system relying more on ad valorem excise Ad valorem excise No excise 2.21 Specic excise 2.62 3.26 0 7 6 5 4 3 2 1 1.66 Excise tax Other taxes Retail price minus taxes PPP $ 7.56 PPP $ 5.66 PPP $ 4.84 PPP $ 3.73 PPP $ 2.52 26 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The impact of tax structure on final price: uniform excise versus tiered tax systems Another aspect of tax structure that impacts final price is the use of tiered taxation, i.e. tax rates that vary according to product characteristics. The characteristics can vary, from price level to the type of tobacco leaf contained in the cigarette, the size of production volume, the packaging, etc. Table 2.1 (earlier) lists the criteria used by 31 countries as the basis for different tax rates. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax (see Fig. 2.7). Fig. 2.7 Weighted average price and excise for a pack of the most-sold brand of cigarettes for countries with and without tiered taxation, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries, with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). Uniform specific tax structures are likely to lead to relatively higher prices with less variability in price distribution. Compared with tiered tax structures that have differential rates based on brand characteristics, uniform taxation may reduce consumers’ incentive to switch to cheaper brands (leading to higher quit rates and lower prevalence), as well as decreasing manufacturers’ incentive to reduce their tax liabilities by changing their pricing strategies, production process or size (42–45). In Indonesia, for example, where small producers were taxed more fa- vourably, manufacturers had an incentive to reduce their scale of production but increase the number of affiliated small companies. The issue was resolved when tax authorities considered the aggregate production of all affiliated companies in the application of differential tax rates. By 2017, there were 786 active factories, while Excise uniform Price and taxation per pack of 20 sticks (PPP $) Excise tiers 3.28 1.94 Excise tax Other taxes Retail price minus taxes PPP $ 4.99 PPP $ 6.30 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 27 there had been 4 198 factories in 2006.8 Abolishing the differential tax rates would have been even more effective in removing the incentive for tax avoidance, as well as helping improve both public health and government finances. KEY TAKEAWAY 6 Evidence suggests that applying a uniform excise tax on cigarettes is not only easier to administer than tiered systems but also more likely to lead to higher cigarette prices. The impact of tax structure on final price: the significance of the choice of the tax base It is important for the excise tax to be applied to the base that leads to the greatest possible effect on price and revenue. For specific taxation, the tax base is the quantity of tobacco products. The quantity of cigarettes, cigars and bidis is measured in number of sticks; for other tobacco products, such as smokeless tobacco or roll-your-own (RYO), it is measured in the weight of the tobacco. When the tax is ad valorem, the choice of the tax base is important not only for health considerations – due to its effect on consumption – but also for tax revenue generation and industry profits. An ad valorem tax that is based on the ex-factory price (or CIF value) provides tobacco manufacturers with opportunities to reduce their tax liability, especially when they control the distribution system. Tobacco producers may sell cigarettes to distributors who are related parties at a reduced price, which then serves as the basis for calculating their ad valorem tax liability. Distributors, however, can then set high prices and share the extra profit with the producers (46). Because of the potential for such trade mispricing, the best practice is to use the retail price as the tax base and introduce a minimum excise tax per pack. Data in the 2019 WHO RGTE (1) show that, on average, the price level of a pack of cigarettes and the excise level are both much higher in countries that use retail price as the base for their ad valorem excise (Fig. 2.8). The maximum retail sales price, which includes all taxes, is used as the ad valorem tax base in the EU. That price also forms the tax base for ad valorem taxes in a growing number of LMICs, including Brazil, Egypt, Thailand, Turkey and Rwanda. 8 Indonesian Ministry of Finance, personal communication, 2017. 28 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.8 Weighted average price of the most-sold brand of cigarettes in countries that use retail price as the base for their ad valorem excise, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 7 The base on which the excise is applied is important. For specific excise, the base needs to be clearly defined (for cigarettes, cigars and bidis, it is the number of sticks; for other tobacco products, such as smokeless tobacco or RYO, it is the weight of tobacco). For ad valorem excise – where the base is typically either retail price, CIF value or producer price – evidence suggests that countries that apply the excise tax on the retail price of cigarettes tend to have higher prices than those that apply the tax on other bases. CIF and producer prices are difficult for government authorities to ascertain and are prone to undervaluation. The tax impact on final price: the significance of the minimum excise tax The use of a minimum excise tax in countries with ad valorem or mixed systems is another important factor in determining final price. On average, the price of a pack of cigarettes – as well as the excise level – is much higher in countries that impose a minimum specific excise than in those that do not (see Fig. 2.9). While more than half of the 47 countries that apply a minimum excise are members of the EU, removing EU countries from the average calculations produces the same conclusions. Ad valorem/mixed with retail price as base 3.72 PPP $ 5.01 PPP $ 6.41 Ad valorem/mixed with other base 1.79 Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 29 Fig. 2.9 Weighted average price of the most-sold brand of cigarettes in countries with and without a minimum specific excise tax, 2018 Notes: Averages are weighted by WHO estimates of the number of current cigarette smokers ages 15 and older in each country in 2017. Prices are expressed in PPP adjusted dollars or international dollars to account for differences in the purchasing power across countries – based on 53 high-income, 97 middle-income and 28 low-income countries with data on prices of the most-sold brand, excise and other taxes and PPP conversion factors. Source: (1). KEY TAKEAWAY 8 Among countries that apply an ad valorem or mixed excise tax on cigarettes, evidence suggests that those that impose a minimum specific excise tax tend to have higher prices than those that do not. The minimum excise tax also helps guarantee minimum excise revenues. Summarizing the advantages, disadvantages and impacts of the choice of excise tax structure for tobacco products Table 2.2 summarizes the characteristics of different types of tobacco excise taxes and the advantages and disadvantages of each type in relation to its impact on quantity demanded, perceived quality of brands offered, price, certainty and stability of revenue, administration and enforcement and opportunities for tax avoidance and tax evasion as they are predicted by the economic theory of imperfect competition and observed in real life. Ad valorem/mixed with minimum speci c 1.80 4.23 Ad valorem/mixed without minimum speci c Price and taxation per pack of 20 sticks (PPP $) Excise tax Other taxes Retail price minus taxes PPP $ 4.82 PPP $ 7.41 30 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 2.2 Characteristics of different types of tobacco excise taxes Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) TA X B A SE The unit of product (e.g. 1 000 cigarettes) The value of the product (e.g. retail, wholesale or manufacturer price) Excise is calculated on an ad valorem basis; however, if the calculated tax falls below a specified minimum amount, a specific tax rate applies Unit and value of product Both unit and value, unless the calculated tax falls below a specified minimum, in which case the tax base is the unit A D M IN IS TR AT IV E R EQ U IR EM EN TS The tax should be collected at the point of manufacturing or at the time of importation Low, as only the volume of the products needs to be ascertained Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as with a pure ad valorem regime Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises Requires strong tax administration with technical capacity; otherwise, the administrative burden can be high, as it requires assessing and collecting both ad valorem and specific excises, as well as minimum specific excise tax compliance U N D ER VA LU AT IO N Not an issue Susceptible to undervaluation Provides an easy tool to prevent undervaluation of low-priced brands subject to the minimum specific excise The ad valorem part of the excise collection may be susceptible to undervalua- tion, depending on the choice of tax base The minimum specific excise prevents possible ad valorem tax base undervaluation of low-priced brands CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 31 Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IM PA C T O N P ER C EI V ED PR O D U C T Q U A LI TY Upgrading effect tends to reduce the relative tax on higher-priced brands Multiplier effect provides a disincentive to costly so-called quality improvement No incentive to upgrade higher- priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands Eliminates incentive to upgrade higher-priced brands, while at the same time provides an incentive to upgrade for lower-priced brands IM PA C T O N P R IC E Tends to lead to relatively higher prices, particularly for low-priced cigarettes Tends to lead to relatively lower prices; price reductions will be subsidized if the multiplier effect is strong Tends to lead to relatively higher price increases for low-priced cigarettes An increase in the specific tax will to lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem payment if the base of the ad valorem includes excise An increase in the specific tax will lead to relatively higher prices, particularly for low-priced cigarettes; the increase in the specific excise will also increase the ad valorem tax amount if the base of the ad valorem includes excise. Increases in the ad valorem and /or specific tax will raise the minimum tax paid if the minimum is a percentage of the total tax on, for example, weighted average price; they will reduce price gaps, given impact on perceived quality 32 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Specific excise Ad valorem excise Ad valorem with a minimum specific excise (or excise tax floor) Mixed specific and ad valorem excise Mixed specific and ad valorem excise with a minimum specific excise tax (or excise tax floor) IN FL AT IO N The real value of the excise tax will be eroded unless the tax is adjusted in line with inflation The real value of the excise tax will be preserved as prices increase, at least to the extent that tobacco product prices follow inflation The real value of the minimum specific excise will be eroded over time unless the excise is adjusted in line with inflation The real value of the specific excise will be eroded unless the excise is adjusted in line with inflation The real value of the specific excise and the minimum specific excise will be eroded unless the excises are adjusted in line with inflation H EA LT H B EN EF IT S Will discourage consumption of tobacco products irrespective of the price band May encourage more trading down in favour of cheaper cigarettes, reducing the health benefit The minimum specific excise reduces incentives for trading down May reduce trading down Reduces trading down Source: (47). 2.2.2 OTHER TAX DESIGN CONSIDERATIONS The significance of automatic adjustments and indexation of specific tax to inflation Specific taxation does not depend on price and therefore, unlike the ad valorem tax, is not automatically adjusted for inflation. The real value of a specific tax is eroded over time as the price of the taxed product increases. Therefore, especially in countries with rapid growth in inflation, the nominal value of the specific tax must be increased regularly in order for the tax to maintain its real value. This is of great importance for both public health and public revenues, especially in countries where manufacturers do not increase prices regularly and/or low-priced tobacco products are the dominant products in the market. Table 2.3 lists countries that include automatic adjustments to their excise in order to avoid the erosion of the specific excise over time, using different units of adjustment and based on different frequencies. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 33 Table 2.3 Countries that include automatic adjustments to the specific excise COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Argentina Inflation (consumer price index [ CPI]), on a quarterly basis Armenia Minimum specific excise set to increase in the Tax Code by 15% on average between 2019 and 2021 Australia Wages – excise rates on tobacco and tobacco products increase in March and September each year, based on average weekly ordinary time earnings Bosnia and Herzegovina Specific excise rate is increased annually by at least 7.50 convertible marks per 1 000 cigarettes; minimum excise tax is increased annually to be at least 60% of the weighted average price Canada Inflation – federal tobacco tax rates are to be increased every five years, indexed to Canada’s CPI starting in 2019 Chile Inflation Colombia Specific tax set to 1 400 pesos, increased to 2100 pesos in 2018; starting in 2019, it will increase yearly by the CPI plus 4 points Costa Rica Inflation Dominican Republic Inflation, on a quarterly basis France Increase from 2017 to reach an average price for cigarettes of €10 per pack by 2020 Honduras Inflation, annually to December of the previous year Italy Minimum tax burden calculated every year in March on the basis of the weighted average price of cigarettes sold in the previous year New Zealand Inflation annually plus 10% annually from 2017 to 2020 Nicaragua Updated annually as of 1 January 2017, taking the highest among the annual devaluation of the official exchange rate of Cordoba with respect to the US dollar, published by the Central Bank of Nicaragua, and the annual inflation rate of the CPI published by the National Development Information Institute, observed in the last 12 months available North Macedonia Specific and minimum specific rate increase by 0.2 denars per cigarette on 1 July each year until 2023 Philippines Agreed tax increases and rates for specific excise tax between 2020 and 2023, with a 5% indexation thereafter Romania Inflation, annual (1 January) adjustment of the total excise according to inflation calculated on 1 October of the previous year Serbia Inflation, every six months Southern African Customs Union (SACU) – Botswana, Eswatini, Lesotho, Namibia and South Africa Inflation, on an annual basis9 Sweden Inflation 9 While the adjustment is not strictly automatic in the SACU, it is greatly informed by the inflation rate. The Treasury has some discretion. In recent years, the increases have typically been slightly above inflation. 34 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N COUNTRY UNIT OF ADJUSTMENT OF THE SPECIFIC EXCISE Turkey Producer Price Index Ukraine Agreement to increase the specific component 20% annually between 2019 and 2025 United Kingdom Increase by 2% above the retail price index (measure of inflation) for the 2015–2020 Parliament Source: (1). KEY TAKEAWAY 9 To avoid erosion of their specific excise tax, countries need to regularly – and, ideally, automatically – adjust the excise to inflation. The significance of automatic adjustments and indexation of specific tax to income growth In addition to the risk of erosion due to inflation, the effect of a (specific) tax can be significantly reduced if the tax is not adjusted for increases in consumer income. Income growth makes products more affordable – thereby encouraging consump- tion – especially in countries with rapid income growth. Australia is one of the rare countries that explicitly adjusts its specific excise rates according to wage growth (see Table 2.3). However, a number of countries have adopted automatic adjustments that are higher than inflation and sometimes largely cover income growth as well (see also Table 2.3). Adjusting tax for income growth contributes to increases in prices that make tobacco products less affordable (see section 2.2.3). KEY TAKEAWAY 10 The specific excise tax needs to be adjusted to reflect income growth so that tobacco products do not become more affordable over time. Measures for specific contexts: the role of pricing and other non-tax regulation Emerging evidence indicates that the tobacco industry finds ways to mitigate the impact of higher taxes on prices. For example, despite the heavy reliance on specific taxation in the United Kingdom, a price differential between premium and cheap cigarettes still exists. There is evidence that the tobacco industry does not always pass tax increases on to cheaper products (37, 48). Differential shifting among price categories is also observed in the EU (49–50), New Zealand (38) and the United States (51). Therefore, the public health community has suggested that pricing regulation could be considered as a method of eliminating inexpensive tobacco products that are often used by the young and the poor (52). Three types of pricing regulation are described below: minimum mark-up, price floor and price ceiling. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 35 Pricing regulation • Minimum mark-up – It has been argued that a minimum mark-up of wholesale tobacco prices could be a better strategy to raise prices of tobacco products than excise tax increases. Minimum mark-up laws aim to discourage the sale of products below an assessed cost by imposing a mark-up to the cost declared at different levels of the supply chain. Some studies in the United States have shown that minimum mark-ups do not increase average cigarette prices (53–54). However, a recent study of the impact of minimum mark-up/ price laws has shown that these laws are linked with higher prices, especially for the cheapest brands, and could be used as an effective tool to mitigate the impact of the industry’s price-reducing promotions (55). Another concern related to minimum mark-ups is that they can be manipulated by manufac- turers and are likely to lead to higher profits for the industry, as well as extra administrative costs for the government (56). • Price floor – A few studies suggest that setting a price floor, or a minimum price, is an alternative strategy for increasing tobacco taxes, particularly with respect to reducing health inequities (57–60). A price floor, imposed by the government or as a vertical restraint imposed by the supplier upon retailers, is a price that firms cannot legally undercut. Governments impose price floors to restrain unfair competition or, in the case of services, to increase quality. It is difficult, however, to find the right floor or to anticipate unintended conse- quences or an industry’s adjustments. A study in Malaysia, where a minimum price for cigarettes was imposed in 2010, found that the policy did not seem to have a meaningful impact on prices: licit brand prices remained well above the minimum price, while illicit brands remained well below it. This outcome may be a result of the floor being set too low or the proportion of illicit trade being high, either of which would reduce the effectiveness of the policy (52). In the EU, imposing minimum retail sale prices for cigarettes could be a breach of harmonized legislation concerning the internal market, as minimum prices would distort competition. Therefore, increasing minimum excise duties is recommended instead, to discourage consumption (61). Increasing the minimum excise duties would also result in the additional revenue going to the governments instead of contributing to industry profits. A price floor would probably lead to increased industry profits – giving the industry greater funds for its marketing strategies – and lower tax revenue for governments, reducing their ability to cover costs associated with tobacco use. By reducing price competition, the price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifica- tions). Competition among firms may prevent them from raising their prices, 36 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N but a government that imposes a price floor does this for them. Minimum pricing is likely to create windfall profits for manufacturers and retailers. It can even help manufacturers sustain a cartel. If the industry uses the money to increase promotions, advertising or grant proposals for related research, this could undo some of the potential benefits of the policy.10 Some recent evidence shows that, at least in the case of the United Kingdom, increased concentration of power among a handful of multinational corpora- tions is enabling them to undermine tax increases through increased price segmentation, and that requiring minimum prices might be a good way to address the problem. A longitudinal analysis of price data from the United Kingdom (48) has shown that despite regular excise tax increases over time, average real prices for cheaper segments of the tobacco market (in this case, cigarettes and RYO) did not increase – indicating an undershifting of the tax increases in those segments that resulted in increased sales volume. At the same time, average prices for more-expensive market segments increased, indicating overshifting of the tax increases that resulted in decreased sales volume. This industry strategy ensures that the most price-sensitive consumers remain addicted, while encouraging initiation and discouraging cessation. Furthermore, segmenting the market further by overshifting the tax increase on premium brands while undershifting it for cheaper brands mitigates the impact of declining consumption resulting from higher taxes while increasing overall industry margins and profitability. Another situation where setting minimum prices can be a useful policy is specific to the United States. Banning marketing and promotions11 is not possible under the freedom of expression protections of the Constitution of the United States (Amendment I), and it was estimated in 2008 that more than 82% of all advertising and promotional spending by the tobacco industry was focused on reducing the price of their products at the point of sale (62). This limitation on how government can set policy has paved the way for the implementation of minimum price policies in many states and cities to counter the detrimental impact of price promotions on consumption and on the tax policy itself. Huang et al. (55) found that the presence of minimum price laws was associated with higher cigarette prices. They also noted that cigarette prices were even higher than prices resulting from minimum price laws in states that also prohibit industry from engaging in other price-reduction strategies, 10 See, for example, the PMI strategy of setting up the Foundation for a Smoke-Free World and grant proposals for related research. 11 See section below on banning promotional discounts for tobacco products for further discussion about marketing and promotions. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 37 such as below-cost combination sales, using trade discounts to reduce the base cost of cigarettes and distributing below-cost coupons to consumers. In the contexts of both price segmentation and price promotions, the imposi- tion of minimum prices as a complementary policy to excise tax increases – not as a policy alternative – may help guarantee that taxes do indeed lead to the intended reduction in consumption. Nonetheless, more evidence is needed to support the effectiveness of this policy. • Price ceiling – Concerns about differential tax shifting have led to suggestions that a price cap may benefit public health by limiting the tobacco industry’s ability to reduce average prices by differentially shifting tax increases among various price segments (63–64). Because tobacco manufacturers operate across international markets, however, they could maintain low prices in one coun- try but maintain overall profitability by selling more premium products in another country. Additionally, limiting price increases does not fit the public health purpose of reducing consumption. It is worth noting that maximum retail prices are sometimes used as a base for calculating the ad valorem tax payments in countries with ad valorem or mixed tax systems. In a systematic review of the literature on non-tax policy approaches to raising prices, Golden et al. (63) hypothesized how such policies would influence price dispersion and average prices. Their study found that minimum price policies combined with promotion bans have the potential to increase average prices. This is, of course, relevant in a context where price promotions are present. From either a theoretical or a practical standpoint, however, it is clear that price policies cannot be used alone and should always be considered as complements to excise tax increases. Significantly increasing taxes is the most effective way to dissuade consumption, correcting whatever bias may exist. Significant tax increases also provide the added benefit of raising money for the government rather than profits for the tobacco industry. Nonetheless, a minimum price might help narrow the gap between cheap and pre- mium cigarettes when applied to all tobacco products to avoid product substitution. Other non-tax regulation • Banning promotional discounts for tobacco products – The sale of tobacco products at a discount rate – such as through reduced-price coupons or buy-one-get-one-free offers – encourages consumption and undermines tax increases. Such practices should be completely banned. They often exist outside the realm of the finance sector because they are considered a type of marketing – promotional discounts are usually addressed in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. 38 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N According to the 2019 RGTE (1), 118 countries out of 195 with all levels of income had such a provision implemented as of 31 December 2018. • Banning the sale of single sticks of cigarettes – Article 16 of the WHO FCTC, “Sales to and by minors,” paragraph 3, requires Parties to “prohibit the sale of cigarettes individually or in small packets which increase the affordability of such products to minors”. Some smokers opt for buying single sticks partly because of the lower im- mediate costs of buying cigarettes individually (65). Internal (unpublished) analysis of single-stick prices collected by WHO for the 2012, 2014 and 2016 editions of the WHO RGTE shows that, in fact, the aggregate price of 20 single sticks of cigarettes sold separately is generally higher than the price of a 20-cigarette pack sold in the market of a specific country. Despite this fact, single-stick sales – and sales of small-sized packs – make cigarettes accessible to consumers with limited disposable income. De Ojeda (66) found in a study conducted in Guatemala that single-cigarette sales are associated with increased cigarette accessibility for less-educated, lower-income populations and minors. Single-stick sales are also a feature of many markets in South-East Asia, including most notably Bangladesh and India, but also in other parts of the world, e.g. South Africa. Single-stick sales also reduce the impact of a tax increase, since the in- crease per stick is much smaller than the increase per pack (67). In a study investigating how smokers in New York City responded to a tax increase of US$ 1.25 per pack in 2008, Coady et al. (68) found that 15% of smokers bought more single cigarettes than they had previously.12 By allowing single stick sales, governments risk losing part of the ad valorem taxes if the tax base is the retail selling price; the retail price of single sticks is much more difficult to monitor than the retail price of packs of cigarettes, on which, for example, tax stamps with prices can be applied. An internal WHO analysis of the most recent tobacco control laws in 2018 in 195 countries found that 86 countries impose by law a ban on the sale of single sticks of cigarettes (36% of the countries are high-income, and 64% are LMICs). In addition to banning the sale of single sticks of cigarettes, 67 of the 86 countries specify a minimum size for packs of cigarettes. Most (52 countries) use the 20 cigarettes per pack standard, but minimum sizes 12 Before 2018, the use of single sticks was possible, but it has since been banned. See New York City Administrative Code. chapter 7: regulation of tobacco products, subchapter 1: Tobacco Product Regula- tion Act, §17-704.a-1. New York: New York Legal Publishing Corporation; 2020 (http://library.amlegal. com/nxt/gateway.dll/New%20York/admin/title17health/chapter7regulationoftobaccoproducts?f=tem plates$fn=default.htm$3.0$vid=amlegal:newyork_ny$anc=JD_T17C007, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 39 range from five sticks in Ghana up to 25 sticks in Papua New Guinea. Another 15 countries do not specifically ban the sale of single sticks but do specify the minimum size for packs of cigarettes. According to the WHO FCTC, in order to reduce affordability, single stick sales should be banned and a minimum number of cigarettes contained per pack should also be defined. KEY TAKEAWAY 11 A number of non-tax measures are closely connected to tax policies, including price regulations, bans on promotional discounts for tobacco products and bans on the sale of single sticks of cigarettes. The price policies discussed are (1) minimum mark-up, (2) price floors and (3) price ceilings. Current evidence does not yet demonstrate that minimum mark-ups and price floors lead to increases in average price. Nonetheless, they may be relevant in some specific contexts as complementary policies to excise tax increases. Price ceilings limit price increases, which can mitigate their impact on consumption. Price marketing strategies such as promotional discounts and the sale of single sticks undermine the effect of tax policies and should be banned. A minimum pack size should also be required by regulators. Tax increases and their possible impact on inflation At times, the inflationary impact of tax increases on cigarettes and other tobacco products is raised as an argument for not increasing these taxes. This may be a concern in countries where wages and/or a significant share of government spend- ing is indexed to inflation (e.g. for public pension payments) or where government policy is to keep inflation low. The extent to which tobacco product tax increases lead to increases in inflation depends on several factors, most notably the share of these taxes in prices and the weight tobacco prices are given in computing a price index. For example, if taxes account for 25% of tobacco product prices, a doubling of the tax (100% increase) will increase prices by 25%. If the weight given to tobacco products in the price index is 3%, the index will rise by 0.75% in response to the tax increase. As tobacco taxes account for a larger share of tobacco product prices, the inflationary impact of a tax increase will be greater. Similarly, as tobacco products are given more weight in computing a price index, a given tax increase will have a greater inflationary effect. In general, for most countries, the inflationary impact of tobacco product tax increases would be relatively small (47). 40 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Consumer price indexes have multiple purposes. They are an important economic indicator for most countries and are often a key determinant of monetary policy. Inflation rates have a direct impact on interest rates and exchange rates. In many countries, changes in wages, social security benefits and other payments are tied to inflation as measured by a price index. Price indexes are used to provide more accurate comparisons of changes in expenditures, incomes and prices for specific goods over time, as well as to allow comparisons across countries. Given the many uses of consumer price indexes and the potential inflationary impact of tobacco tax increases, some governments have developed alternatives that exclude tobacco (and sometimes other goods) for some uses. For example, since 1992, France has excluded tobacco products from the price index used for adjust- ing minimum wages (47). However, many countries continue to include tobacco product prices in their consumer price indexes. Excluding tobacco products from the basket of goods used in developing key price indexes would greatly reduce concerns about their impact on inflation. In addition, with declining consumption of tobacco products, the inclusion of their prices in key price indexes results in a distorted measure of price for many consumers. KEY TAKEAWAY 12 If governments are concerned about the potential inflationary impact of a tobacco tax increase because wages or some government spending may be tied to a price index, they can use a price index that excludes tobacco products. The importance of taxing cigarettes and other tobacco products in a comparable way While cigarettes are the most commonly used tobacco product globally, other tobacco products are as prevalent and sometimes more prevalent than cigarettes in some parts of the world. Bidis and smokeless tobacco are the main products consumed in some countries in South-East Asia – Bangladesh and India in particular – and waterpipes are widely used for smoking tobacco in the Eastern Mediterranean region (4). These products, as well as RYO, have historically been taxed much less than cigarettes (see, for example, Fig. 2.10 for Bangladesh and India, where the excise tax and prices of bidis and smokeless tobacco are much lower than those for cigarettes). This differential taxation undermines the health impact of excise taxes on tobacco products because (1) it encourages users to switch from cigarettes to the lower-taxed product (see the case of Thailand below); (2) it is not effective in reducing tobacco use in general, especially if the most widely used product in the country is not cigarettes; (3) it can encourage tax avoidance by companies that may CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 41 redefine products that are similar to cigarettes so that they fall within the lower- taxed product category (see the EU example below) and (4) it reduces the size of government revenues since those products could have been taxed at higher levels. Fig. 2.10 Price and tax of the most-sold brand of cigarettes, bidis and smokeless tobacco in Bangladesh and India, 2018 Source: (1). In Thailand, for example, the price of cigarettes has been raised quite successfully through taxation over a number of years, while taxes and prices of loose or RYO tobacco have until very recently remained unchanged. Indigenous tobacco used for RYO cigarettes has historically been exempt from excise, while foreign tobacco was taxed at a low level relative to that of manufactured cigarettes. Consequently, Thailand experienced growth in the RYO market even though cigarette consumption had been falling.13 The Thai government eventually took strong action to address this issue. First, the exemption for indigenous tobacco was removed in 2018. The Cabinet then approved an increase in the excise rate on small producers (of indigenous tobacco) from 0.005 baht per gram to 0.025 baht per gram in 2020, with another increase to 0.1 baht per gram scheduled for 2021 (69). In the EU, the minimum excise duty levels for cigars and cigarillos is significantly lower than that for cigarettes. The Member States of the EU are required to levy an 13 WHO Country Office for Thailand, personal communication, 2019. 12.8 0 190 54.9 64 24 Price Excise tax amount 80 12.8 3.8 44 Lo ca l C ur re nc y 0 50 100 150 200 Cigarettes (20 sticks) Bangladesh, taka India, rupees Bidis (20 sticks) Bidis (20 sticks) Cigarettes (20 sticks) Smokeless (20 g) 42 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N excise duty of at least €90 per 1 000 cigarettes, which should be 60% of the weighted average retail selling price of cigarettes released for consumption. For cigars and cigarillos, only €12 per 1 000 items, or an excise duty of 5% of the retail selling price, is required. As a result, the excise tax share on cigarettes is much higher in many EU countries than the share for cigars and cigarillos. In response, some companies started to market so-called borderline cigarillos. These products have characteristics similar to cigarettes but can be sold at a lower price because for excise purposes, they are considered as cigarillos. Although this issue seems to be largely solved by amendments to the definitions of these products at the EU level and a change in tax structures in some countries, it is important to be aware of the unintended incentives that can be created by large gaps in excise tax levels between product categories (70–71). For more details on industry tactics to undermine tax increases, see Box 2.1. KEY TAKEAWAY 13 To make excise tax on tobacco products more effective in reducing overall tobacco use and to avoid substitution between products, all tobacco products need to be taxed in a comparable way. The Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce the demand for tobacco) (73) recommend that all tobacco products should be taxed in a comparable way. Box 2.1 Industry tactics used to undermine tax increases Tax increases reduce the demand for tobacco products and present a threat to the tobacco industry’s high profits. The industry responds by using various strategies (17, 46, 48), including the following (46): Stockpiling (forestalling/front-loading) – Before the implementation of an an- nounced tax increase, manufacturers overproduce tobacco products, paying the pre-tax-increase rate. As a consequence, sales and tax revenue decline immediately but temporarily after the tax increase (while sales and revenues had increased sub- stantially just before the tax increase) and the industry attributes this drop in revenue to the emergence or increase of illicit trade. This practice results in tax avoidance if there is no law prohibiting it (see also the discussion on anti-forestalling in Chapter 3). Changing certain product characteristics (for example, weight or length) and/ or adjusting the production process – When tobacco products are taxed at different rates or are subject to different tax increases, the industry can, for example, re-label one type of tobacco product as another product that has a lower tax burden (as in the example of the EU above). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 43 Choosing the time of a price increase announcement strategically – The industry may raise prices in anticipation of a tax rate increase, generating extra profits in the period until the tax is actually implemented. When the tax increase is implemented, consumption and tax revenue will fall, but prices will not change, so the industry can claim that the tax policy was ineffective in reducing demand. Adopting price-discriminating strategies or price-related promotions – The industry may offer discounts, retailer rebates or added value (gifts) to tobacco purchases to minimize the loss of price-sensitive consumers. This, however, is not possible in countries where strict bans on tobacco advertising, promotions and sponsorship are implemented. Using brand proliferation (for example, launching a low-priced brand) and price segmentation – Manufacturers can choose to reduce prices of certain brands or introduce new, even cheaper ones to keep price-sensitive consumers in the market. There is evidence that firms introduce new cheaper products and use price-marking – printing the price directly on packs of tobacco products – to lock in their price (48). Such practices compromise both public health and revenue objectives. Differential shifting of tax increases across different price segments, depending on the market circumstances – The industry may increase the price of a product by more than the amount of the tax increase (tax overshifting) and blame the govern- ment for the total increase. Tax overshifting is profitable when demand is inelastic, that is, when the price increase more than offsets the reduction in sales. The industry may overshift the tax increase for higher-priced brands, which are expected to be more price inelastic than lower-priced brands. Additionally, to keep price-sensitive consumers in the market, the industry may temporarily absorb part (or all) of the tax increase on lower-priced brands. The differential tax shifting will lead to different responses in the demand for the different brands (37, 48). Lobbying government to distort interventions – Government policy might be influenced by tobacco industry lobbying, directly or indirectly. Policy-makers are not simply welfare or revenue maximizers; they also value political support. Industry lobbying might lead to adopting a favourable type of taxation, postponing tobacco tax increases or distorting the tax rate downwards (17). Article 5.3 of the WHO FCTC, “On the protection of public health policies with respect to tobacco control from commercial and other vested interests of the tobacco industry”, and its guidelines provide useful guidance on how to address tobacco industry interference. In fact, all 181 countries that are Parties to the WHO FCTC have a legal obligation to implement the requirements of Article 5.3. Having correct expectations about industry responses is important for estimating the impact of a tax increase on consumption and tax revenue. 4 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 2.2.3 MEASURING IMPACT AND RECOMMENDED INDICATORS Governments need to consider a number of indicators when formulating policy changes. Inappropriate assumptions about consumer behaviour, market structure and industry behaviour can lead to faulty policy analysis. Measuring impact on price and demand Consumption habits, local traditions and industry characteristics – such as the number of different brands offered, the possibilities of cross-border shopping and the presence and level of illicit trade – all affect the shape of the demand and supply of tobacco products, thereby determining the value of the price elasticities. Price elasticity, together with the industry’s pricing strategies – for example, the degree of tax shifting – and the tax share in the retail price, determine the elasticity of the tax base, regardless of whether the base is determined by quantity (for specific taxation) or transaction value (for ad valorem taxation). The importance of elasticity estimates Different types of elasticity should be considered: • price elasticity – own-price elasticity – measures the response of consumers’ demand for a product following a change in the price of the product. – cross-price elasticity – measures the response of consumers’ demand for a product when the price of another product changes. Cross-price elasticity can also occur between different brands or price segments for the same product. • income elasticity – the response of consumers’ demand for a product when their income level changes. Correct estimates of price and income elasticities are important for policy-makers who need to anticipate the impact of a tax increase on consumption and tax rev- enue. Estimates will vary depending on a number of factors, including whether responses are considered in the short run versus the long run, the functional form of the demand function used, whether factors such as addiction or tax evasion are accounted for and the way data are constructed. For example, details such as the degree of aggregation of data, whether gender- or age-specific data are used, the time span covered and which estimation procedures are used (e.g. ordinary least squares, two-stage least squares or generalized method of moments) will all affect the results of the estimate (72). Price elasticities may change over time, as well because of changes in any of the other factors affecting demand, such as income or tobacco control measures, and also because of changes in estimation techniques and the types or sources of data used. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 45 Moreover, what is of most interest is the price elasticity of total demand. A tax increase may reduce tax-paid retail sales but not necessarily total consumption. For example, smuggling can significantly bias price elasticities when the elasticities are estimated using legal sales data; not accounting for possible illicit trade might lead to overestimation. Similarly, when cross-border shopping is included, the price elasticity of demand is lower (in absolute value) (74). Estimating the total price elasticity of demand for legal and illegal consumption can be done by using cross- sectional data from nationally representative household surveys. However, this approach also has its weaknesses. For example, respondents tend to underreport their consumption of tobacco, which leads to bias in the size of demand. Price endogeneity14 is another technical problem that can be challenging to address. To comprehensively estimate the total effect of a tax increase on demand for all tobacco products as well as on tax revenue, the degree of substitutability between them needs to be estimated (55). Cross-price elasticity measures how the quantity demanded of a particular tobacco product changes when the price of another tobacco product increases. When this elasticity is positive, the products are substitutes; the higher the value of the elasticity, the closer substitutes the products are to one another. For example, positive cross-price elasticity between RYO and manufactured cigarettes implies that the demand for RYO increases as the price for cigarettes increases. Substitutability may also arise between different cigarette brands – when the relative price of economy brands increases, demand for premium brands may increase. This effect can be exacerbated when differential (tiered) taxation is ap- plied on different types of cigarettes, further widening the gap in prices between brands and segments and encouraging substitution. The substitutability between traditional and new and emerging tobacco and nicotine products is currently of great interest (see section 2.4 below). In some countries, different tobacco products can also be complementary rather than substitute goods. This means that when the price of a tobacco product increases, the demand for its complement drops because users are unlikely to use the complementary tobacco product alone. For example, a number of studies have found manufactured and indigenous bidi cigarettes to be complementary goods in India (75–76). The sign and magnitude of income elasticity vary across time, countries and demographic groups. For example, in the United States, a high-income country, income elasticity over time has changed from positive to negative, and cigarettes have switched from being a normal good to an inferior good (77–78). On the other hand, among LMICs, where prevalence of smoking tends to be relatively higher, 14 Price is endogenous because it is not an independent variable: it is estimated by dividing expenditure on tobacco by consumption of tobacco, with consumption being a dependent variable in the estimation of price elasticity. 46 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarettes might still be a normal good, with consumption increasing as income levels increase (positive income elasticity) (79–80). There are unobservable characteristics that differentiate higher-income smok- ers from lower-income smokers, such as differences in time and risk preferences, differences in associating a social stigma with smoking and differences in taste for smoking as a pleasurable activity. When these characteristics are ignored, estimates of the correlation between income and smoking-related outcomes are biased. Kenkel et al. (81), using techniques that estimate the causal effect of income on smoking among low-income adults, found that tobacco is a normal (even a luxury) good: higher income is associated with a higher probability of smoking participation and a lower probability of smoking cessation. These results are consistent with those regarding the impact of the business cycle – periods of expansion or recession in economic activity – on health be- haviour and outcomes. Ruhm (82–83), for example, found that smoking declines during temporary economic downturns and increases during economic expansions. Tarantilis et al. (84) found that estimates of income elasticities of demand in Greece were higher after the economic crisis of 2010 than before it. The financial crisis and the austerity measures shifted the demand for cigarettes downwards and turned cigarettes into a more income-elastic good. Interestingly, evidence from Germany suggests that the propensity to become a smoker significantly increases during an economic downturn. However, among those who are already smokers, cigarette consumption actually decreases (85). Ideally, when estimating price and income elasticities, the effect of non-price policies should also be accounted for. A recent study from South Africa shows that failing to take non-price policies into account will overstate the price effect (86). The NCI/WHO Monograph (4) suggests that price elasticity of demand for tobacco is on average -0.4 in high-income countries (ranging from -0.2 to -0.6). Estimates for LMICs are more variable, clustering around -0.5 (ranging from -0.2 to -0.8). A price elasticity of -0.5 means that a 10% increase in price would lead to a 5% reduction in consumption. KEY TAKEAWAY 14 Policy-makers need to know the elasticity of demand – including price elasticity (own-price and cross-price) and income elasticity – for tobacco products in their country in order to correctly assess the impacts of potential policy changes on consumption and subsequent revenues. These estimates need to be made on a regular basis to capture changes in demand over time. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 47 The importance of the tax base elasticity The tax base elasticity measures the sensitivity to a change in the tax rate of the base on which the tax is imposed – the base being tobacco consumption in the case of specific taxation and tobacco expenditure in the case of ad valorem taxation. The magnitude of the elasticity of the tax base depends on price elasticity of demand, the tax structure, the level of the tax rate and its share in price, along with the industry response through its decision to absorb, pass through or overshift the tax on to the retail price. Consumers’ preferences and income, the availability of substitutes and other non-price tobacco control measures also influence the tax base elasticity, essentially through the price elasticity of demand. The magnitude of the elasticity of the tax base also depends on social motivations, including price and tax expectations, which are ultimately impacted by successful tobacco control measures that affect consumers’ willingness to pay taxes or prices. In addition, the tax-base elasticity depends on smokers’ perceptions of the prob- ability of detection and tax enforcement when using illegal products, as well as the availability and accessibility of opportunities for tax evasion and avoidance. Finally, consumers’ willingness to pay taxes depends on their perceptions regarding the use of the tax revenue (87). Therefore, the tax base elasticity is largely influenced by government policy choices. Increasing the tax share in prices is recommended by WHO as a tool to achieve the public health objective of reduced tobacco use: a higher tax share in prices increases the tax base elasticity, all else remaining constant, and therefore increases the reduction in the tax base through the resulting reduction in smoking. However, manufacturers can be expected to attempt to manipulate the tax base elasticity through their pricing policies, such as tax shifting. As discussed earlier in this chapter, industry behaviour is itself affected by government tax policy and regulations. A number of factors need to be taken into account when considering tax pass- through. As discussed earlier, tax is more likely to be overshifted within a specific tax structure than within an ad valorem structure. There is also evidence of industry overshifting the tax for premium or expensive cigarette brands while undershifting the tax for cheaper brands. This indicates that within a given market, the industry’s decision on the extent of tax pass-through will vary based not only on the tax structure but also on the structure of the market. It will also vary by brand. But this does not give an indication about the impact of the tax increase on the average price of a tobacco product. In the context of the tax base elasticity and the impact of tax increases on revenues, it is important to assess how tax increases affect average prices. The example of South Africa is very useful here. Over the past two decades, South Africa has been consistently increasing its specific excise tax on cigarettes, which has led to large price increases. An analysis of the effect of excise tax increases on 48 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N cigarette prices between 2001 and 2015 (26) shows that while there is evidence of tax overshifting, at least within a specified period of time, overall, the tax has been undershifted in real terms. This undershifting is due partly to increased competitive- ness in the market and partly to the introduction of low-priced brands. Of course, this encouraged some consumers to downshift their consumption to cheaper products, but it also pushed manufacturers of more expensive brands to absorb part of the tax increase to reduce the impact on price. A change in the level of the tax rate – with all other factors that influence con- sumption held constant – will result in a change in the tax revenue.15 Estimates of tax base elasticity help governments predict changes in tobacco tax revenues following a tax increase (see details in Annex 2.2). Under specific taxation: • tobacco consumption – the tax base – is expected to be price inelastic (17, 47); • prices increase by less than the tax increase, on average (there is no tax overshifting overall); and • consumption – the tax base – is also expected to be tax inelastic: the quantity of consumption falls less than proportionately to the tax increase, and the tax revenue increases. Under ad valorem taxation: • the tax base is the total consumer expenditure (or, equivalently, the industry sales revenue) on (legal) tobacco consumption – that is, the tax base under ad valorem taxation is determined by both price and quantity, which is itself a function of price; • the sign of the tax base elasticity – which can be either negative or positive – depends on the magnitude of the price elasticity of demand; • since evidence suggests that tobacco demand is price inelastic, the tax base elasticity is positive; • when the ad valorem tax rate increases, both price and quantity adjust, but quantity falls less than proportionately to the price increase, and tax revenue increases; and • a tax rate increase leads to both a higher level of revenue and a lower level of consumption; the value of the elasticity – and hence the tax revenue – increases with the degree of tax shifting. 15 This concept has been used by Laffer to argue that tax increases that are too high will reduce excise tax revenues (the so-called Laffer curve). For a detailed discussion on the Laffer curve, see section 4.4 in Chapter 4. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 49 Taxation serves as an instrument for achieving both fiscal and public health objectives. If, after successful tobacco control interventions, prices reach levels where demand becomes elastic, the tax base is still most likely to be inelastic due to tax undershift- ing, since overshifting is not a good pricing policy when demand is elastic. In other words, a tax rate increase – in combination with non-price tobacco control measures that make consumers more sensitive to price (tax) increases – leads to decelerating but still positive marginal revenues. For an example of a tobacco taxation success story, see Box 2.2. For further details of countries’ experiences with tax increases and their impact on revenues, see section 4.4 in Chapter 4. KEY TAKEAWAY 15 Policy-makers’ key policy tool to control demand is tax. Therefore, it is essential they assess not only the impact of price on demand but, more appropriately, the impact of tax on demand: this is the tax base elasticity. The tax base elasticity is essentially determined by (1) the price elasticity of demand, (2) the degree to which the industry will pass the tax on to the retail price and (3) the tax as a share of the retail price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. Box 2.2 A tobacco taxation success story: Turkey Turkey is an example of a country that has been increasing taxes regularly and sig- nificantly over a relatively short period of time and has reaped the benefits of this policy. As shown in Fig. 2.11, the excise tax per pack of cigarettes more than doubled in real terms over 10 years, with the real price almost doubling as well. In parallel, tobacco excise revenues increased by 67% and cigarette sales decreased by 20%. Since the beginning of the country’s Health Transformation Program in 2003, Turkey has successfully increased public health spending and collected more tobacco tax revenue. According to the latest available figures, in 2015, tobacco tax revenue was equivalent to 42% of the country’s public health expenditure and 1.5% of GDP (88). 50 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 2.11 Tobacco excise revenue and consumption in Turkey (base year 2008), 2008–2018 Sources: Reference 1 for the price of the most-sold brand, Ministry of Finance for the sales and revenue data and IMF world economic outlook, April 2020. See https://www.imf.org/en/Publications/WEO/weo- database/2020/April for the adjustment for inflation. Impact on affordability While price increases clearly have an impact on consumption, when the effects of increasing per capita income of a population are not considered, the price impact may not be as strong as expected. Increases in a population’s income also increase its purchasing power. And, as indicated earlier, tobacco products generally behave like a normal good. Consequently, as income increases, it is expected that tobacco consumption will increase as well. To mitigate this effect, price increases (following tax increases) need to be greater than increases in income. This is where the concept of affordability comes in. Affordability examines the effects of both increasing prices and increasing incomes on consumer behaviour. A common and easy way to calculate affordability, made popular by Blecher and van Walbeek (89), is to use the percentage of GDP per capita required to buy 2000 cigarettes (or 100 packs of 20 cigarettes) in a given year. An increase in this proportion over time will indicate that cigarettes are becoming less affordable and should lead to reductions in consumption. Changes in trends in affordability of cigarettes over time help policy-makers understand how prices are evolving 0 1 2 3 4 5 6 Tu rk is h lir as p er p ac k N um ber of sticks, 10 000 000 Turkish liras 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Real excise tax amount Real price Real tobacco excise revenues Cigarettes, per capita sales 1 100 500 1 700 2 300 CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 51 relative to a population’s ability to purchase cigarettes and enable them to revise their policies accordingly. Recent studies in India, for example, highlight the wide price differential between manufactured cigarettes and indigenous tobacco products such as bidis and chewing tobacco, as well as the propensity for these indigenous products to become more affordable over time due to favourable or more lenient tax policies towards them (90–91). Figure 2.12 shows the change in affordability of a pack of the most-sold brand of cigarettes by country income group between 2008 and 2018. During this time period, affordability declined in almost 70% of high-income countries, while it declined in slightly more than 35% of middle-income countries and only 26% of low-income countries. Fig. 2.12 Number of countries that have experienced a change in affordability of cigarettes between 2008–2018, by income level Note: Change in affordability was computed as the least squares rate of change in the per capita GDP required to purchase 2000 cigarettes of the most-sold brand in local currency in a given year. The trend rate of growth was computed for countries with four or more years of data, including 2018. Affordability was assessed as not having changed if the least squares trend in the per capita GDP required to purchase 2000 cigarettes over the period 2008–2018 was not statistically significant at the 5% level. Source: (1). KEY TAKEAWAY 16 From a health perspective, in addition to examining the impact of a tax increase on the levels of price, demand and revenues, policy-makers should consider a tax hike that will lead to prices rising more than increases in their population’s income; a tax increase should make tobacco products less affordable to consumers so that demand will be effectively reduced. Cigarettes became less aordable Cigarettes became more aordable Aordability did not change Could not be assessed due to insucient data High income Middle income Low income 39 36 9 13 5 7 37 23 6 13 2 5 52 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Projecting impact on consumption, smoking prevalence and lives saved The WHO interactive smoking projection and target-setting tool (WHO ISPT) The WHO ISPT enables national policy-makers and tobacco control experts to explore the potential impact of proposed tobacco control policies. It uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. The WHO ISPT provides projections of (1) tobacco smoking rates and (2) tobacco-smoking-related deaths in a country under different policy settings and for different time periods. It was designed to promote multisec- toral collaboration within countries by enabling experts from various ministries (for example, health, education, finance, national statistics), civil society, academia and media to explore options for medium- and long-term tobacco control planning together with WHO experts. Use of the WHO ISPT enables strong partnerships for policy change advocacy, program development and evaluation.16 In particular, it can help policy-makers in the Ministry of Finance assess the specific contribution of tax policies – within overall tobacco control policies – towards achieving specific targets in tobacco prevalence reduction. Projecting impact on excise revenue The WHO tobacco tax simulation model (WHO TaXSiM) The WHO TaXSiM is a simple but data-intensive Excel-based tool that helps policy- makers analyse their tobacco tax policy and assess the impact of any excise tax increase or change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Using detailed data about the market – including the majority of brands found in the market, their market share and price levels and the applicable tax – and assumptions about price elasticity of demand, the WHO TaXSiM predicts the impact of tax changes on consumer prices, consumption volume and tax revenues generated by each brand and market segment for the following year. The exercise can be done for multiple years.17 By exploring market data in detail, in addition to assessing the potential rev- enue impact of changes in excise tax, the WHO TaXSiM is a useful instrument for highlighting weaknesses and opportunities in an existing tax system and market. It can also encourage policy-makers to create administrative databases that can be periodically updated to monitor the dynamics of the cigarette market. 16 The WHO ISPT is not available publicly, but WHO will work directly with interested countries upon request to use it to produce data-to-action-type plans. 17 For more information about the methodology, see https://www.who.int/tobacco/economics/tax- sim_background.pdf, accessed 29 September 2020. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 53 KEY TAKEAWAY 17 Policy-makers can use available tools to simulate the impact of tax increases on prices, consumption and revenues, as well as smoking prevalence and lives saved. Recommended indicators to monitor tobacco taxation progress MPOWER WHO publishes a biannual RGTE, which monitors global progress in tobacco con- trol. In particular, the report focuses on the implementation of the policy package MPOWER, a set of proven demand-reduction measures in line with the key provi- sions of the WHO FCTC (1). While raising taxes on tobacco (component R) is proven to be the most effective and cost-effective policy to reduce tobacco use (4), implementing the entire MPOWER package at the best practice level will reinforce the impact of R. For example, as mentioned earlier in this chapter, banning promotional discounts as part of the E measure (enforce bans on tobacco advertising, promotion and sponsorship) will favour price increases following a tax increase. If all the MPOWER tobacco control measures except R were implemented at the best practice level, all else remaining constant, revenues would be expected to decline. Thus, in order to maintain revenue levels, it is important to raise excise taxes on tobacco products regularly to compen- sate for the decline in tobacco use from the other four tobacco control measures. Tax share The main indicator in the R policy in the RGTE (1) is the total share of indirect taxes in the retail price of the most-sold brand of cigarettes.18 Countries whose most-sold brand of cigarettes has a total tax that is equal to or greater than 75% of the retail price are considered to be at the highest level of achievement. While total taxes include excise taxes, VAT (or sales taxes), import duties (when applicable) and other indirect taxes (where applicable), it is preferable to focus on excise taxes, since they are the component that most influences the relative price of tobacco. The share of excise tax in the retail price can be extracted from the RGTE database.19 The 2010 WHO technical manual on tobacco tax administration recommended making excise taxes account for at least a 70% share of excise taxes in the retail price of tobacco products (47). 18 For more details about how this indicator was compiled, see Technical Note III of the RGTE 2019 (https://www.who.int/tobacco/global_report/Technical-Note-III.pdf?ua=1). 19 See taxes and retail price for a pack of 20 cigarettes, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.1-Taxes-and-retail-price-for-a-pack- of-20-cigarette-most-sold-brand.xls?ua=1, accessed 29 September 2020). 54 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Affordability As discussed previously, the share of tax in the retail price is not enough to ensure that a policy will be successful in reducing demand. Any tax increase should lead to an increase in price that will effectively discourage consumption. While global trends indicate that a high tax share is positively correlated with a high price level (see Fig. 2.2 in section 2.1.2), this may not necessarily apply to a particular country; a tax share can be high, while at the same time tobacco products remain afford- able. For this reason, it is important to monitor not only tax increases but also whether those increases led to a price increase that is greater than income increases. As described in section 2.2.3 of this chapter, a common indicator is the percentage of GDP per capita required to buy 100 packs of 20 cigarettes in a given year.20 Other indicators As discussed in detail in sections 2.2.1 and 2.2.2, a good tax structure can make a tax policy more effective in increasing prices and decreasing affordability of tobacco products. Indicators can include whether a uniform excise is applied, whether it is a specific excise and whether it is adjusted regularly for inflation. A number of such indicators are also monitored through the RGTE and can be downloaded online.21 A tobacco tax indicator compiled in 2020 combines the various elements that form a good tobacco tax policy. The Tobacconomics Cigarette Tax Scorecard (92) rates a country’s tobacco tax policy performance based on best practices. The four components that determine the level of performance are (1) cigarette price (in PPP), (2) changes in the affordability of cigarettes over time, (3) the share of taxes (total and excise) in retail cigarette prices and (4) the structure of cigarette taxes (i.e. whether excise is applied; whether it is uniform or tiered; whether excise is specific, ad valorem or mixed; and, for the ad valorem component, if the tax is applied on the retail price and if there is a minimum specific excise and, for the specific component, if tax is automatically adjusted upwards). Each of the four components is given a score, using a five-point index, with the total score reflecting an average of the four component scores. The closer the total score is to 5, the bet- ter the tobacco tax policy performance is in a given country. While this published scorecard is currently applied only on cigarettes, it can be easily applied on other tobacco products, provided the needed data are available. 20 This indicator has also been compiled in the RGTE; see (https://www.who.int/tobacco/global_report/ Table-9.6-Affordability.xls?ua=1, accessed 29 September 2020). 21 See supplementary information on taxation, globally, in WHO report on the global tobacco epidemic 2019 (https://www.who.int/tobacco/global_report/Table-9.5-Supplementary-information-on-taxation. xls?ua=1, accessed 29 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 55 KEY TAKEAWAY 18 Tobacco taxation works best if implemented as part of a comprehensive MPOWER package. MPOWER is an overall indicator that incorporates all the key demand-side tobacco control measures. KEY TAKEAWAY 19 In addition to assessing the potential impact of a tax increase, policy-makers need to monitor progress over time. The share of the tax in the retail price is an indicator of progress. However, it is important to remember that an effective tax increase must translate into higher prices in order to make tobacco products less affordable. Combining all the components of a good tax policy into one scorecard can also be useful for assessing tobacco tax policy as a whole. 2.3 DOMESTIC AND REGIONAL POLICY INTEGRATION While it is essential to design tobacco tax policies with the utmost consideration of all the aforementioned factors, it is also important to consider how external factors can impact or even impede public health policy objectives. As Chapter 3 explains, cooperation among the various agencies that are directly involved in tax administration, collection and enforcement is important for effective and efficient tax policy implementation. But in the design phase, it is also essential to engage with agencies and other policy-makers that are not directly involved with taxation. Domestically, coordination is required to ensure that policies in non-health sectors do not negatively impact or even counteract tobacco control initiatives. For countries that are part of a regional bloc, harmonization of tobacco taxation is essential to protect the single market – as well as the health of the population – and to prevent tax revenue erosion, tax avoidance and tax evasion. 2.3.1 INTERSECTORAL COOPERATION ON DOMESTIC POLICY Domestic policies in agriculture, industry, trade, finance and labour have the po- tential to create or support incentives at different stages of tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. For example, subsidies provided to farmers or manufacturers involved in growing or processing tobacco can reduce prices and incentivize continued participation or even increase development in these areas, which is counterproductive to the goals of making tobacco products less affordable and reducing tobacco consumption. Multisectoral integration and policy coherence are needed at the country level to ensure that public policies and interventions in non-health sectors do not act against the intended public health impact of tobacco control and taxation. 56 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 20 Greater policy coherence in agriculture, industry, trade, finance and labour should be promoted to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. 2.3.2 REGIONAL TOBACCO TAX HARMONIZATION Policy integration is driven by the recognition that cooperation on domestic policies can substantially increase the gains from forming a regional bloc. Harmonization is desirable and may be necessary in certain areas with spillover effects, such as tax policy, the possibility of a so-called “race to the bottom” or threats to public health. Harmonization could be as simple as setting minimum standards and requirements based on global norms and best practices. Harmonization of tobacco taxation is required to ensure the establishment and proper functioning of a single market, prevent tax revenue erosion, prevent tax avoidance and tax evasion and protect people. When barriers to trade between countries are removed, harmonized tax rates support the single market because they improve the ability of consumers, producers and investors to make decisions that are not distorted by taxation but reflect real opportunity costs. Tax competition – where countries simply undercut each other’s tax rate – could prevent governments from raising sufficient funds to pursue social policy. To avoid such a race to the bottom, countries can establish minimum tax rates within the customs union (93). Even if tax competition is not present, when substantial tax differences exist in neighbouring countries, there is a clear incentive to trade across borders in order to reduce tax payments legally or illegally. The experiences of established regional economic communities offer important policy lessons, not only in terms of the general integration process but also for the process and extent of tax policy coordination. The EU implemented a successful regional tax harmonization scheme. Over the years, the focus in harmonization of tobacco taxes has broadened from the elimination of tax obstacles to the fight against harmful tax competition, tax avoid- ance and tax evasion and, more recently, to public health protection. Naturally, addressing these issues requires increasing convergence in fiscal policy and tax administration. Although price differentials still exist, setting a minimum on the share of taxes in the final price of tobacco products as well as a minimum excise tax has helped countries reach some level of harmonization. The EU experience confirms that both a declining tobacco consumption trend and stable revenues can be achieved with harmonized minimum excise rates (94). Moreover, the harmonization process has CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 57 offered opportunities for the enforcement authorities (tax or customs) to obtain information that can be used in the fight against fraud and tax evasion. On the other hand, the experience of harmonization efforts in the West African Economic and Monetary Union (WAEMU) shows how the absence of a supranational body (like the EU) or a hegemonic member state (see the SACU example below) can slow down policy integration that would benefit all member countries (95). The eight countries of the WAEMU are bound by a Tax Directive22 that requires them to impose an ad valorem excise on the CIF value or producer price of tobacco products, which is subject to under-declaration and is difficult to ascertain. Additionally, a maximum excise rate is imposed, and some members apply additional taxes to deal with this constraint. The Directive was revised in 2017 (96), but unfortunately the tax structure remains the same, and the maximum rate was not removed but rather has been increased. The Southern African Customs Union (SACU), which has five member countries, is the oldest existing customs union, established in 1910. Thanks to the hegemonic lead of South Africa, a country with a sophisticated administration system and an aggressive tobacco tax policy, SACU adopted a well-integrated tax policy that has benefited all its members (95). The GCC, established in 1981, is a regional intergovernmental political and economic union consisting of six states of the Persian Gulf. Home to one fifth of the global oil supply (97), the GCC has never relied on taxation as a source of revenue; no direct or indirect taxes were applicable in the region. Although there was no excise on tobacco products, as a customs union, the GCC countries have a common external tariff. This common tariff includes harmonized rates but also a harmonized structure. The import duty is 100% of the CIF value of tobacco products imported in the region, with a minimum tax per quantity imported. However, in recent years, to reduce their dependence on income from oil, GCC countries have considered diversifying their sources of income, including by de- veloping reliance on indirect taxes such as excise and VAT. In 2015, a decision was adopted at the 36th GCC summit meeting to implement selective taxes on all imported tobacco products and cultivated raw tobacco grown domestically (GCC Decision number 963/1). A follow-up decision in December 2016 formally agreed to the introduction of an excise tax on tobacco and other products such as sugary and energy drinks, as well as special goods (alcohol and pork meat), in all GCC countries. The decisions at the national level to implement this subregional decision came into force gradually in all GCC countries, starting with Saudi Arabia, which began 22 Directive No. 03/98/CM/WAEMU on the harmonization of Member States’ legislation of excise duties was adopted 22 December 1998. It was amended by Directive No. 03/2009/CM/WAEMU of 27 March 2009 with the objective of harmonizing excise duties within WAEMU. 58 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implementation in June 2017, followed by Bahrain and the United Arab Emirates (UAE) later that year, Qatar in 2018 and Oman in 2019 (1). Only Kuwait has yet to adapt its national laws accordingly. The excise introduced by the GCC countries has a structure somewhat similar to the import duty on tobacco products: the rate is 100%, but the base was changed from the CIF value to the retail price excluding taxes. The introduction of the excise led to large increases in the price of the most- sold brand of cigarettes in member countries between 2016 and 2018 – by 33% in Bahrain, more than 80% in the UAE and more than 100% in Saudi Arabia (1). In federations such as Canada and the United States – where the central govern- ment has real taxing power and some financial and regulatory control over the states or provinces – tobacco taxes are not harmonized (98–99). Even though there are significant interjurisdictional differences in taxes and prices, and tax harmonization holds great potential to reduce the scope of illicit transactions in the tobacco market, there is little evidence that Canadian provinces or individual states in the United States are interested in tobacco tax harmonization. Tax harmonization is most relevant in the context of further economic integration within a group of countries that are already part of a customs union, but it needs to be planned well to be effective. Discrepancies in law interpretation and a lack of standardization of tobacco product definitions and tax base lead to suboptimal situations. Tax rate alignment, or setting minimum rates, should come after tax structure alignment. It is important that governments support the move towards harmonization and are committed to dedicating enough financial resources and skilled personnel to oversee the entire process. KEY TAKEAWAY 21 In the context of regional economic integration and ongoing discussions regarding the possibility of harmonizing tobacco excise taxation among member countries, the experiences of existing groups can be instructive. So far, only the EU, SACU, WAEMU and, more recently, the GCC have effectively implemented a harmonized approach to excise taxation of tobacco products. Lessons learned indicate that harmonization should be planned well and should not come at the expense of tobacco control. Setting a common minimum specific excise tax, adjusted over time, is the best approach. This ensures that taxes and prices are above a minimum level, encouraging equalization of price levels and at the same time reducing affordability across countries. On the other hand, agreeing on maximum tax rates is a bad policy. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 59 Countries that wish to raise their taxes further for revenue purposes, health concerns or both should be given the space to do so. Concerns about illicit trade provoked by higher tax rates are best dealt with by strong cooperation in administration and enforcement, information sharing and adoption of new technology with common or interoperable information systems. 2.4 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In recent years, awareness of tobacco risks and harms, implementation of tobacco control provisions – especially under the WHO FCTC – and tightening of regulations have resulted in declining sales of cigarettes, primarily in high-income economies. This has changed the dynamics of the tobacco market. In response to these effective tobacco control measures, the tobacco industry has diversified its business by promot- ing a new portfolio of products, which they claim to be technological innovations that supposedly reduce the harms and risks associated with conventional tobacco products, particularly cigarettes. So-called novel tobacco products have been promoted by the tobacco industry as “cleaner alternatives,” “safer alternatives” and “reduced harm/risk products” with no smoke and no ash. On the basis of these claims, they negotiate for less-restrictive regulatory environments within countries. Some of the new products are also mar- keted or promoted for smoking cessation, despite the evidence of this outcome being inconclusive. Where these products are not banned, one of the debates in the global health community concerns the issue of their regulation and taxation. 2.4.1 HEATED TOBACCO PRODUCTS (HTPs) HTPs are tobacco products that produce aerosols containing nicotine and toxic chemicals upon heating of the tobacco or activation of a device containing the tobacco. These aerosols are inhaled by users sucking on or smoking the device. They contain the highly addictive substance nicotine (found in tobacco) as well as non-tobacco additives and are often flavoured. The tobacco in HTPs may be in the form of specially designed cigarettes (e.g. so-called heat sticks or Neo sticks) or pods or plugs. These products include IQOS from PMI, Ploom TECH from Japan Tobacco International (JTI), glo from British American Tobacco (BAT) and PAX from PAX Labs. HTPs differ not only from con- ventional cigarettes but also from ENDS – some of which are called e-cigarettes – as ENDS do not contain tobacco but rather a nicotine solution (see next subsection). However, the boundaries between the different products are becoming increasingly 60 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N difficult to define, given the emergence of so-called hybrid tobacco products that contain both nicotine solution and tobacco. HTPs are currently available in more than 40 countries and are banned in fewer than 10 countries. Even in countries where they are regulated, there is significant variation in the approaches taken to regulation. A variety of factors affect a country’s ability to control and regulate the use of HTPs, including national regulatory pow- ers, enforcement capacity regulatory frameworks, country capacity and tobacco industry interference (1). Most countries tax HTPs at a lower rate than cigarettes and on the kilogram of tobacco as a base when applying a specific or mixed excise (see Table 2.4). The use of such a base may be quite challenging for tax collection, especially because of the difficulty of checking the tobacco content in each stick. In the past, some countries taxed cigarettes per kilogram of tobacco, but today it is common practice to tax them per stick regardless of tobacco content. Table 2.4 Excise taxation of HTPs, first collected for July 2018 – updated for July 2020 OVERALL COMPARISON WITH CIGARETTES Type of excise Base unit is kg, overall rate lower than cigarettes Base unit is sticks, rate is the same as cigarettes Base unit is sticks, rate is lower than cigarettes Other Specific excise Albania, Austria, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Greece, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Montenegro, Netherlands, New Zealand, North Macedonia, Romania, Russian Federation, Slovakia, Slovenia, Sweden, United Kingdom Azerbaijan a, Japan, Ukraine b Armenia, Hungary, Jordan, Italy c, Philippines, Republic of Korea d Montenegro e, Republic of Moldova f , Serbia g Ad valorem excise (base is retail price unless specified other- wise between brackets) Spain, Switzerland Saudi Arabia and United Arab Emirates (base is retail price exclusive of excise and VAT) Indonesia h CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 61 Mixed system (ad valorem compo- nent based on retail price unless specified other- wise between brackets) France, Germany, Poland, Portugal Colombia, Georgia, Israel and West Bank and Gaza Strip (ad valorem excise base is wholesale price) a The specific excise rate applied is the same as for imported cigarettes, higher than the rate applied to domestically produced cigarettes. b The rate is the same as the minimum excise on cigarettes per 1 000 pieces. Rate and structure were effective as of 1 January 2021. c The specific excise rate is defined as 25% of the excise tax on cigarettes based on an equivalency used between cigarettes and HTPs. There are planned increases of this proportion to 40% by 2023. d In 2020 the specific excise rate was only 11% lower than cigarettes. e The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. f Specific excise rate is higher than for cigarettes but, unlike HTPs, cigarettes also face an ad valorem excise. Overall effect of excise is a slightly lower for HTPs. g The specific excise tax is based on the weight (kg) of tobacco mixture and is calculated at 40% of the minimum excise tax per 1000 cigarettes. There is a planned phased increase of this proportion aiming equalization with cigarettes by 2025. h While cigarettes face a specific excise tax rate, HTPs face an ad valorem rate, the highest rate as defined by law, on the basis of a pre-defined minimum price. Sources: (1, 100, 101, WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021 and the Campaign for Tobacco-Free Kids website on Taxation and Price for Heated Tobacco Products https://www.tobaccofreekids.org/what-we-do/global/taxation-price/staging-tax-gap). A study by Liber (102) compared prices of HTPs and cigarettes in 34 countries and showed that while taxes have been systematically lower for HTPs than for cigarettes, prices were higher in half of the countries surveyed. KEY TAKEAWAY 22 HTPs, when taxed, are usually taxed lower than cigarettes, although they generally seem to be priced higher than cigarettes. It is important to remember that HTPs are tobacco products, and the same provisions that apply to tobacco products should apply to them as well. This is articulated in WHO’s information sheet on HTPs (103), which provides guidance on how these products should be regulated, as well as Decision FCTC/COP8(22) for novel and emerging tobacco products. Moreover, MPOWER measures, which help WHO Member States to implement the demand-reduction articles of the WHO FCTC, are applicable to HTPs, in particular, Article 6 for taxation. Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. Furthermore, HTPs contain chemicals not found in cigarette smoke, the health effects of which are not yet known. Independent assessment of industry data demonstrates that more than 20 harmful and potentially harmful chemicals 62 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N are significantly higher in HTP emissions than in cigarette smoke (104). Therefore, there is a need to learn more about these products and the health impacts of their emissions, as well as the impacts of exposure to these emissions. KEY TAKEAWAY 23 Currently, there is no evidence demonstrating that HTPs are less harmful than conventional tobacco products. From both public health and tax administration perspectives, HTPs should be taxed at the same level and in the same way as tobacco cigarettes. Some countries have already adopted this approach and are taxing HTPs at the same rate per stick as cigarettes (Azerbaijan, Colombia, Georgia, Israel, Japan, Ukraine and West Bank and Gaza Strip). Saudi Arabia and the UAE, which have recently introduced an excise tax on tobacco products as part of the GCC, are now applying the same import duty rate and excise tax structure for cigarettes and HTPs. Continuing developments in technology and changes in products have led to a recommendation to tax HTPs per unit. The definition of unit may vary by product within the HTP category. For example, one unit of IQOS is one heat stick, for Glo it is one Neo Stick and for Ploom TECH it is one tobacco pod. Governments will need to determine the exact definition of a unit for each product allowed on the market. The potential complexity of the market strongly supports limiting the types of HTPs allowed in a country and setting strict regulations to standardize the products as much as possible. Countries can also consider taxing the devices used to consume HTPs, i.e. the holder and the charger (see product description in Annex 3.1). KEY TAKEAWAY 24 HTPs are tobacco products, and they need to be treated as such. Where they are not banned, HTPs need to be strictly regulated and taxed. The recommendation is to tax them at the same level as cigarettes on a per-unit basis. Countries can also consider taxing the devices used for HTP consumption. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 63 2.4.2 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS)23 Products like ENDS and ENNDS have evolved rapidly over the past decade. ENDS heat a solution (e-liquid) containing nicotine, but not tobacco, and other chemicals that may be toxic to people’s health to create an aerosol, which is inhaled by the user. Examples of ENDS include Juul from Juul Labs, Vype from BAT and blu from Imperial Brands (1). Electronic non-nicotine delivery systems (ENNDS) are essentially the same as ENDS, but the e-liquid used generally does not contain nicotine. Upon testing, however, many so-called zero-nicotine solutions are found to contain nicotine (105–107). While generally considered as a single product class, ENDS products constitute a diverse group with potentially significant differences in the production of toxicants and delivery of nicotine. There are several coexisting types of devices for ENDS/ ENNDS on the market, including first-generation or so-called cigalikes, second- generation tank systems and even-larger third-generation or personal vaporizers. Collectively, they are also often referred to as e-cigarettes, vapes or vape pens. Other categories of ENDS include e-hookahs, e-pipes and e-cigars – hence, ENDS is an all-encompassing term for multiple product categories. Some of the products resemble their conventional tobacco counterparts – cigarettes, cigars, cigarillos, pipes or hookahs – while others are shaped more generically like pens, USB memory sticks or basic cylinders. Different forms of nicotine are also used in these ENDS, the most recent one being nicotine salts, which deliver high levels of nicotine (1). There are two types of ENDS/ENNDs products: open systems and closed systems. Open systems are devices that allow the user to buy e-liquids and fill their device with the mixtures they want (with no nicotine, different nicotine concentrations and/or flavours). Closed systems are products that come with a prefilled container (called a cartridge, pod or tank). For the past decade, divisive debates have been waged over the effectiveness of ENDS as smoking cessation aids – especially for tobacco users who are unable to give up the habit – as well as the possibility of ENDS playing a role in public health. However, the evidence remains inconclusive. Despite the tobacco industry and other related industries promoting these products as tools for quitting smoking, current evidence does not support their use as part of a population-based cessation strategy (108). Accordingly, the United States Surgeon General, in January 2020, concluded that 23 It is worth noting that ENDS are not tobacco products and not exactly new products – the technology has been around since the late 1980s (e.g. Premier, Eclipse and Accord). However, the recent generation of these products is new and has more or less piggybacked on the success of e-cigarettes. 64 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N E-cigarettes, a continually changing and heterogeneous group of products, are used in a variety of ways. Consequently, it is difficult to make generalizations about efficacy for cessation based on clinical trials involving a particular e- cigarette, and there is presently inadequate evidence to conclude that e-cigarettes, in general, increase smoking cessation (109). The evidence on the adverse health effects associated with use of ENDS is mounting, and when ENDS are used in combination with smoking – which is the practice of the majority of ENDS users (110) – the adverse health effects of two or more products are combined. However, there are insufficient data to understand the full breadth of these effects, as ENDS have not been on the market long enough for their long-term effects to be established. Nevertheless, the evidence is clear that the aerosols of the majority of ENDS and ENNDS, some of which are cancer causing chemicals. ENDS also contain nicotine, which is highly addictive. In addition, ENDS are associated with increased risk of cardiovascular diseases and lung disorders, as well as adverse effects on the developing fetus during pregnancy (108, 110). For adolescents, the use of nicotine can lead to dependence and may harm brain development. Use of ENDS could also lead to a new generation of nicotine and tobacco users, as seen in some countries, especially since these products are designed to appeal to young people. Although the specific level of risk associated with ENDS has not yet been determined conclusively, these products are undoubtedly harmful. Therefore they should be strictly regulated if allowed to be sold in domestic markets, and must be kept away from children. Taxation will be a key component of regulation, since it is an effective tool for influencing consumer behaviour. Some countries have taken the bold decision to completely ban these products. Approaches that have been taken range from partial to comprehensive bans, and ENDS/ENNDS products were banned in more than 30 countries in 2018.24 In other countries, they are regulated as, for example, consumer products, pharmaceutical products or tobacco products, or they are completely unregulated. WHO recom- mends that where ENDS/ENNDS are not banned, they should be regulated to achieve the following objectives: 1. prevent the initiation of ENDS/ENNDS by non-smokers and youth, with special attention to vulnerable groups; 2. minimize as much as possible potential health risks for ENDS/ENNDS users and protect non-users from exposure to their emissions; 3. prevent unproven health claims being made about ENDS/ENNDS; and 24 Data collected for the WHO RGTE 2019. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 65 4. protect tobacco control activities from all commercial and other vested interests related to ENDS/ENNDS, including the interests of the tobacco industry. ENDS/ENNDS present a risk to youth, who have taken up their use in high numbers in some countries, including Canada and the United States (1, 111). The Juul brand, for example, has quickly gained a significant e-cigarette market share in the United States (112–113). Its marketing and popularity have led the United States Food and Drug Administration (FDA) to raise serious concerns and to seek solutions to effectively prevent youth from taking up the use of ENDS/ENNDS (114). The city of San Francisco banned the sale of e-cigarettes in June 2019 (115). In addition to posing a risk for initiation by youth, ENDS can attract non- tobacco users or prevent current smokers from quitting. Taxation could play a role in preventing the uptake of these products, specifically among non-smokers, vulnerable groups, children and adolescents. KEY TAKEAWAY 25 The long-term health effects of ENDS/ENNDS products are still unknown, but they are clearly harmful to health. Furthermore, evidence on the effectiveness of ENDS products as a smoking- cessation aid remains inconclusive. Taxing these products could play a role in preventing their uptake, specifically among non-smokers, vulnerable groups, children and adolescents. Price elasticity of demand for ENDS products In the context of taxation, it is important to ask whether demand for ENDS is price-responsive. Preliminary evidence, although almost exclusively focused on e-cigarette data from the United States, indicates that this is the case: demand for e-cigarettes may be even more price-responsive than the demand for conventional cigarettes, so taxes can be used to deter initiation by never-users (116–123). Most of the studies of price elasticity of demand for ENDS products also demonstrate that e-cigarettes and conventional cigarettes are partial substitutes – that is, they show positive cross-price elasticity. The magnitude of the elasticity indicates the degree of substitutability between products: the higher its value, the closer the products are to being substitutes, with higher cigarette prices being associated with increased e-cigarette sales. Some of the studies also show a substitutability effect in the other direction, with increased prices for e-cigarettes leading to an increase in conven- tional cigarette use (117, 120). All of the studies show evidence of substitutability except for one (124), which differentiates between exclusive and dual users and shows no evidence of substitution between e-cigarettes and conventional cigarettes. 66 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The presence of concurrent (or dual) use – consumers using both conventional cigarettes and ENDS products – complicates results and highlights the need for more research in this area to better disentangle the different effects. Liber et al. (125) analysed sales prices in a sample of 45 countries and concluded that comparable units of conventional cigarettes cost less than disposable e-cigarettes. The units considered for pricing e-cigarettes included both the e-liquid and the rechargeable device. Taken alone, the price of e-liquids is on average much lower than that of cigarettes in high-income countries and the same in LMICs. The time needed to buy back a rechargeable device is estimated to be less than two weeks in most countries. One can argue that increasing price differentials by further increasing taxes on regular cigarettes could be effective in driving current smokers of regular cigarettes to e-cigarettes (126) as a potentially lower-risk alternative (127). However, the ef- fectiveness of ENDS as smoking cessation devices is still being debated; a study by Sweet et al. (128) shows that dual use of e-cigarettes as a potential tool for cessation was effective only in the short term. Moreover, significantly more smokers said they would quit if cigarette prices doubled and e-cigarettes were not available (122) or that they would never have become addicted to nicotine if e-cigarettes had not been so readily available (129). Once an e-cigarette user is addicted to nicotine, there is a risk of initiating traditional tobacco products use (130). In general, cessation can be better facilitated by governments via stronger implementation of the other tobacco control policies that have been proven effective at reducing use. KEY TAKEAWAY 26 Few studies are available on the price elasticity of ENDS products, and the available data come almost exclusively from the United States. These early studies indicate that demand for e-cigarettes will go down as the price of e-cigarettes increases. Generally, the results also show that cigarettes and e-cigarettes are partial substitutes, where an increase in cigarette price would increase the demand for e-cigarettes while reducing demand for cigarettes. But these results do not differentiate between people who are exclusive cigarettes or e-cigarette users and those who are users of both products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 67 Tax structure Different countries impose different tax structures on ENDS/ENNDS products25 (see details in Table 2.5). The Republic of Korea, for example, imposes a specific tax per millilitre of ENDS/ENNDS e-liquid (131), while Indonesia imposes an ad valorem tax on the retail price of the e-liquid; the maximum rate allowed by law for tobacco products (132). In the United States, there is no common way to tax e-cigarettes among the states that do tax them (133–134). The situation is similar in the EU, where new and emerging nicotine and tobacco products are not currently covered by the tobacco tax directive, and Member States may apply a national tax as they see fit under their own rules. All the EU countries that tax ENDS products apply a specific excise per millilitre of e-liquid. These different tax treatments have the potential to distort the functioning of the internal market. In February 2020, the European Commission concluded that the current provisions of the harmonized directive are no longer relevant for the taxation of ENDS and HTPs, and this is a source of concern from the internal market perspective (135). In June 2020, the Member States of the EU reiterated that it is urgent and necessary to upgrade the EU regulatory framework by harmonizing defini- tions and the tax treatment of novel products such as ENDS/ENNDS and HTPs (2). Table 2.5 Types of excises applied on ENDS/ENNDS products e-liquids globally and in individual states in the United States, as of July 2019 (updated as of July 2020 for all countries except the United States) TYPE OF EXCISE COUNTRIES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Albania, Kazakhstan, Kyrgyzstan, Pakistan, Portugal, Republic of Korea, Romania, Russian Federation, Slovenia, Sweden Azerbaijan, Cyprus, Estonia, Finland, Georgia, Greece, Hungary, Italyb, Latvia, Lithuania, Montenegro, Morocco, North Macedonia, Philippines, Serbia Ad valorem (% of retail price or import value) Bahraina Indonesia, Jordan, United Arab Emirates, Yemen TYPE OF EXCISE INDIVIDUAL STATES IN THE UNITED STATES Taxing only nicotine- containing e-liquids (ENDS products) Taxing all e-liquids (ENDS and ENNDS products) Specific (based on volume per mL) Delaware, Illinois (Chicago), Cook County, Louisiana, Ohio, Puerto Ricoc, Connecticutc Kansas, North Carolina, Washington, West Virginia, Wisconsin 25 The focus is on the e-liquid used for ENDS/ENNDS products. 68 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ad valorem (% of wholesale/ distributor price) Alaska (Juneau, Matanuska- Susitna Borough), California, Illinois, Maine, Maryland (Montgomery County), Minnesota, Nevada, Pennsylvania, Vermont, Washington DC, Virgin Islandsc New York Mixed New Jersey, New Mexico a Tax applied to e-shisha (or e-hookah) because e-cigarettes are banned in Bahrain. b Italy imposes differential rates for nicotine and non-nicotine containing liquids. c States in which it is unclear if only ENDS or both ENDS and ENNDS products are taxed with an excise. Sources: (1, 135, complementary data from Frank Chaloupka and WHO data collection of price and tax of cigarettes and HTPs in 2020, unpublished as of April 2021). Table 2.6 provides reference material on the pros and cons of different considerations for determining the tax structure and base of ENDS/ENNDS products e-liquids. Table 2.6 Excise tax options for ENDS/ENNDS products e-liquids TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Specific Volume of nicotine- containing e-liquid (regardless of concentration) 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only the volume needs to be determined 1. Difficult to compare if tax equivalencya with cigarettes is sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarettes and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed 5. May encourage more do-it- yourself (DIY) products where e-liquids are mixed by the users themselves, which increases the risk of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 69 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of e-liquid regardless of nicotine presence 1. Reduces the price gap between similar products 2. Simple from a tax administration perspective, as only volume needs to be determined 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Does not require laboratory capacity to detect the presence of nicotine in liquids 1. Difficult to compare if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Potentially favours products with a higher nicotine concentration per mL, which also tend to be the products most responsible for the rapid uptake in youth initiation in countries where these products are available and aggressively marketed Volume of all e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL are affected the most; also reduces the price gap between different products 2. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 3. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 4. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 5. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to user 1. Difficult to compare with cigarettes if tax equivalency with cigarettes is sought 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed 70 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Volume of nicotine- containing e-liquids with an additional tax per unit of nicotine concentration 1. More likely to serve as barrier to youth initiation by affecting the lowest price category, and products with a higher nicotine concentration per mL affected the most; also reduces the price gap between different products 2. Current market observations (2019) within the same product categories indicate that nicotine concentration is not a major determinant of price; increasing taxes as the nicotine concentration becomes higher may change this and would reinforce the health justification that nicotine is addictive and not harmless 3. Reduces manufacturers’ incentive to increase nicotine concentration in order to reduce the tax burden 4. Reduces DIY incentives, as increasing nicotine concentration for personal consumption will increase cost to the user 1. Difficult to compare if tax equivalency with cigarettes is being sought 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids as there is no simple way currently available to determine whether the e-liquid contains nicotine; self-declarations by industry are not sufficient 3. Requires laboratory capacity to measure the amount of nicotine concentration in e-liquid solutions 4. More complicated from a tax administration perspective, as both volume and nicotine content need to be assessed Ad valorem Producer price/ CIF value of nicotine- containing e-liquid (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 71 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Producer price/ CIF value of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Covers non-nicotine-containing e-liquids, which are also harmful when inhaled and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by industry 1. Prone to undervaluation because the true value of the tax base is difficult to ascertain 2. Requires strong tax administration capacity to implement effectively – in particular, capacity to assess the validity of declared tax base value 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Retail price of nicotine- containing e-liquids (regardless of nicotine concentration) 1. Easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotine (self-declarations by industry are not sufficient) 2. Requires capacity to monitor the market to assess market prices 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled in an e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risk of accidents, illness and death 72 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Retail price of all e-liquids regardless of nicotine presence 1. Easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products (especially if tax on cigarettes is ad valorem – although this does not change the long-standing recommendation that specific taxes are better for conventional cigarettes) 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 3. Less prone to undervaluation as tax base is easier to assess (compared with a CIF/producer price base) 1. Requires capacity to monitor the market to assess retail prices 2. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels Ad valorem with minimum specific or mixed Ad valorem + min specific 1. Volume of nicotine- containing e-liquids will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of nicotine- containing e-liquids will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up. b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS products and cigarettes in the context of a large heterogeneity of products 1. Requires capacity to monitor the market to assess retail prices 2. Requires laboratory capacity to detect the presence of nicotine in e-liquids, as there is currently no simple way to determine whether the e-liquid contains nicotin; (self-declarations by industry are not sufficient) 3. Does not cover non-nicotine- containing e-liquids, which are also harmful when inhaled via e-cigarette and during independent tests have often been found to contain nicotine, contrary to product labelling or representations by the industry 4. Difficult to set a minimum specific excise amount/specific excise amount, especially if tax equivalency with cigarettes is sought 5. May encourage more DIY products where e-liquids are mixed by users themselves, which increases the risks of accidents, illness and death CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 73 TYPE OF EXCISE BASE ADVANTAGES DISADVANTAGES Ad valorem + min specific 1. Volume of all e-liquids regardless of nicotine presence will be the base for the minimum specific. 2. Retail price will be the base for the ad valorem excise.b or Mixed 1. Volume of all e-liquids regardless of nicotine presence will be the base for the specific excise. 2. Retail price will be the base for the ad valorem excise.b 1. This system aims to exploit the best characteristics of both the specific and the ad valorem tax systems: a. Specific component guarantees a minimum tax and pushes all prices up b. Ad valorem component is easier to compute and regulate for tax equivalency between ENDS/ENNDS products and cigarettes in the context of a large heterogeneity of products 2. Does not require laboratory capacity to detect the presence of nicotine in all e-liquid solutions sold 1. Requires capacity to monitor the market to assess retail prices 2. Difficult to set a minimum specific excise amount/specific excise amount, especially if equivalency with cigarettes is sought 3. Challenge in detecting and differentiating whether the e-liquids (e.g. propylene glycol and vegetable glycerine) to be used in ENDS/ENNDS are falsely declared as being for other purposes (e.g. food, cosmetics or pharmaceuticals) at import and manufacturing levels a Tax equivalency can be measured in different ways: (1) in terms of tax burden (as % of the retail price) or (2) as the exact amount of tax for equivalent quantities (assuming an equivalency between a certain volume of e-liquid and a pack of cigarettes). b There is also the option to use the producer price/CIF value as a base for the ad valorem component, but it is a weaker option because the base is difficult to ascertain and therefore prone to undervaluation. Note: Table compiled following a WHO Expert Meeting on Taxation of Electronic Nicotine and Non-Nicotine Delivery Systems (ENDS/ENNDS), Geneva, Switzerland, 2–4 September 2019. There is currently a lack of evidence on the practical challenges being faced by countries favouring one approach over the other. Furthermore, such data are difficult to obtain because the nature of the market is constantly changing. However, a clear recommendation can be made with regard to which e-liquids to tax. As indicated in Table 2.5, some countries tax all e-liquids – whether or not they contain nicotine (ENDS and ENNDS products) – while some tax only nicotine- containing e-liquids (ENDS products). As shown in Table 2.6, there is evidence that in a number of instances, ENNDS products do contain some nicotine. Additionally, ENNDS products are not harmless (136–137). It is therefore recommended that all e-liquids be taxed for both ENDS and ENNDs products. The question of whether to employ differential taxation based on nicotine content seems reasonable from a health perspective, since nicotine is a toxic substance. However, this would likely create an additional burden for tax administrators as 74 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N they would need to determine the nicotine concentration of e-liquids on the market. Additionally, this may no longer be relevant, as advancements in technology indicate that other features of the product can influence nicotine delivery beyond the actual concentration of the e-liquid. It is now possible to increase nicotine delivery at low nicotine concentrations by increasing battery power (by reducing resistance or increasing voltage) (138). In terms of implementation, while most countries seem to have adopted a specific excise tax on ENDS/ENNDS e-liquids per millilitre, one benefit of implementing ad valorem taxation is that it seems relatively easier to regulate in the context of a large heterogeneity of products. However, it is essential that the tax be applied on the retail price value of the products, as this base is easier to ascertain than any other value that could be declared by the manufacturer. It is also important to add that regulation of the characteristics of ENDS/ENNDS products is essential, and it should be implemented along with any tax policy adopted. Regulations should include: 1. setting a maximum nicotine concentration per millilitre to safeguard public health, including reducing the risk of dependence, especially among youth; 2. setting a maximum volume for cartridges to reduce toxicants exposure and possibly limit use; 3. setting a maximum capacity for refill containers to reduce toxicants exposure and possibly limit use; 4. setting a maximum battery power to reduce the possibility of influencing nicotine and toxicant delivery; and 5. taxing nicotine regardless of its source (e.g. tobacco, eggplant, synthetic). Countries may choose to impose an excise tax on ENDS and ENNDS devices26 as well. The easiest type of tax would be an ad valorem tax based on the declared retail price. If countries choose not to impose an excise tax on these products, they should at least impose the regular VAT or sales tax rate. Imposing an excise tax on devices can be challenging from an administrative perspective, as all components need to be clearly defined and classified as devices for ENDS/ENNDS consumption. For example, if the device is assembled after importation and some parts may be used for other purposes than ENDS/ENNDS consumption, authorities may face a challenge in detecting and differentiating which component parts would be subject to excise tax and which would not. 26 See Annex 2.3 for an overview of elements of devices used in ENDS/ENNDS products. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 75 KEY TAKEAWAY 27 There is currently not enough evidence to recommend one tax structure over another for ENDS/ENNDS products. However, it is clear that taxing the e-liquids used for consumption is key. The excise tax should be applied on all e-liquids, whether or not they contain nicotine. If the preferred type of excise tax is ad valorem, it should be applied to the retail price. Countries can consider taxing devices as well, but they need to adequately assess their administrative capacity to do so. Policy-makers need to be mindful of the diversity and rapid evolution of ENDS/ ENNDS products and adjust accordingly. Regulation must reflect this reality so that loopholes will not be exploited by the industry. For example, ENDS/ENNDs products include not only e-cigarettes, vapes and vape pens but also other categories such as e-hookahs, e-pipes and e-cigars. Lawmakers need to be clear about how ENDS/ENNDS products are defined so that subcategories do not fall under the radar when regulation comes into effect. Definitions will also be relevant when it comes to taxation. An unclear definition can lead to a seemingly contradictory situation, such as in Bahrain, where e-cigarettes are banned but e-hookahs are not.27 Finally, while policy-makers need to be mindful of the emergence of new products and must take appropriate actions to protect the health of their citizens, it is important to remember that the overwhelming share of nicotine consumption remains that of tobacco products, especially cigarettes. The total market value of ENDS/ENNDS and HTPs sales in 2018 was less than 2.2% of the total market value, while cigarette sales alone accounted for 91% of the same total market value (139–140). 2.5 CONCLUSIONS An overview of excise tax application globally reveals a broad variety of price and tax levels, as well as structures used for taxing tobacco products, in particular, cigarettes. Some trends, however, indicate that tax and price levels are higher among higher- income countries. The rate of taxes also matters: higher tax rates are correlated with higher prices, and higher prices change behaviour, which leads to a reduction in consumption. More countries are moving away from ad valorem taxes and towards either mixed or specific excise systems, and there are few countries that do not impose any excise tax on cigarettes. 27 In Bahrain, the Ministry of Production and Trade Decision 38 of 2013 banned e-cigarettes, while the official list of excisable products from the Ministry of Finance includes e-shishas (or e-hookahs), making them apt to be taxed and therefore considered legal. 76 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Significantly increasing the taxes and prices of tobacco products is the most effective and cost-effective policy to control tobacco use. Increased taxes – which are passed on to smokers as higher prices – reduce consumption. When designing tobacco tax policy or reforming tobacco tax systems, policy-makers face several challenges, ranging from technical issues – such as determining what tax structure and rates to apply – to political economy issues, as well as the perceived contribution of the tobacco sector to economic development. In designing tax policy, the tax structure adopted not only affects consumption overall, it also shapes the market structure. Ad valorem taxation incentivizes industry to set prices lower than specific taxation does. Evidence suggests that under a specific tax, the price gap between premium and lower-priced products is narrower, therefore reducing incentives for substitution to lower-priced products following a tax increase. However, as industry consolidates producers and widens its portfolio of products, new evidence indicates that the industry is introducing cheaper brands while increasing the price of its expensive brands, therefore, paradoxically, widening the price gap between its products. Evidence also suggests that prices are higher under a specific excise tax structure. Additionally, from a tax administration perspective, a specific tax is easier to imple- ment, since only the quantity produced needs to be ascertained rather than the value of the product. Another aspect of tax structure is the use of tiered taxation – that is, tax rates that vary on the basis of different product characteristics. Evidence suggests that the average cigarette price and the average excise level for a pack of cigarettes tend to be much lower in countries that use a tiered excise structure than in countries that use a uniform excise tax. Tiered taxation encourages substitution from premium to cheaper brands, maintaining smoking prevalence and reducing the health im- pact of tax rate increases. In addition to leading to lower prices, tiered taxation is difficult to administer and creates opportunities for the tobacco industry to avoid and evade taxes. The design of a tax structure must also consider the base on which tax is applied. The choice of base should lead to the highest possible effect on price and revenue. For specific taxation, the tax base is the quantity. When the tax is ad valorem, the CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 77 choice of the tax base is important not only for health considerations, through its effect on consumption, but also for tax revenue generation, as well as industry profits. An ad valorem tax based on the producer price, or CIF value, gives tobacco manufacturers opportunities to reduce their tax liability, especially when they control the distribution system through related parties. The best practice in an ad valorem (or mixed) excise structure is to use the retail price as the tax base and introduce a minimum excise tax per pack. Other tax design considerations include the importance of using automatic adjust- ments and indexation to inflation and income growth for the specific excise tax in order to avoid erosion of the tax over time. Emerging evidence indicates that tobacco taxation does not always achieve the intended results, because the tobacco industry finds ways to circumvent it. Non- tax policies such as pricing regulation (in particular, minimum mark-ups or price floors/minimum prices) may be seen as a complementary approach to ensuring a high price level and discouraging consumption of tobacco products. So far, these policies have not proven to increase average prices. A price floor is likely to lead to increased industry profits, giving the industry greater funds for its marketing strategies (such as the introduction of new products), and lower tax revenues for governments. By reducing price competition, a price floor allows firms to compete aggressively for market share in other dimensions (e.g. product specifications). However, where powerful multinationals are operating in certain markets with presence in all market segments and with the capability to overshift a tax on some brands while undershifting the tax on others, or where price promotions cannot be banned, minimum price policies may help increase the effectiveness of tax increases. Other non-tax policies affecting price levels are those relating to promotional dis- counts for tobacco products and the sale of single sticks of cigarettes. Both should be completely banned. The ban of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Higher taxes are the most effective way to dissuade consumption, with the added benefit of raising money for the government – money that can be earmarked for health and education programs, rather than going as profits to the tobacco industry. Additionally, in order to make excise tax on tobacco products more effective in reduc- ing overall tobacco use and in line with the recommendation of the Guidelines for implementation of Article 6 of the WHO FCTC (Price and tax measures to reduce 78 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N the demand for tobacco), all tobacco products need to be taxed in a comparable way; the focus should not be on cigarettes only. Tax choices and reforms have various and sometimes conflicting consequences for the market. For example, there might be a trade-off between quantity and variety or perceived quality implications. It is important for the government to recognize that firms respond strategically to changes in tax policy. Close monitoring of the market is necessary to form correct expectations about industry responses and enable estimates of the impact of a tax increase on consumption and tax revenue. To estimate the total effect of a tax increase on demand for tobacco products and tax revenue, it is important to use correct estimates of the own-price elasticity of demand, the cross-price elasticity and the income elasticity of demand. It is also important to use updated estimates of demand elasticities, as the environment within which consumers make decisions continues to change. For example, financial crises or successful tobacco control interventions can be expected to shift demand and change elasticity. Another key measure of the impact of tax policy is the tax base elasticity. Policy- makers need to be mindful of the three key components of tax base elasticity: (1) the price elasticity of demand of tobacco, (2) the share of the tax in the consumer price and (3) the degree of pass-through of the excise tax rate increase to consumer price. The degree to which these elements are affected by a tax increase will impact demand and revenues. Currently, the three components combined are not high enough in any country for a tax increase to lead to a reduction in excise tax revenues. It is important to acknowledge that if tax increases lead to increases in prices be- low concurrent increases in income levels, they will not be effective in reducing consumption, as tobacco remains a normal good in most countries. Policy-makers need to account for affordability when considering tax increases. They should ensure that tax increases are high enough to increase prices above income growth so that consumption goes down effectively. When designing tax policy and deciding on the right level to impose, policy-makers need to assess and project the impact of their policy decisions. Monitoring and evaluation are important. Tools for measuring impact can be very helpful, and several such tools exist. The WHO ISPT, for example, looks not only at the impact of tax policy but also at a set of tobacco control policies, and this enables national policy-makers and other tobacco control experts to explore the potential impact of CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 79 future tobacco control policies. The tool uses impact factors of selected WHO FCTC demand-reduction measures derived from published literature, trends in tobacco smoking rates, national demographic information and tobacco-related mortality risk. More specific to tobacco tax policy, the WHO TaXSiM assesses the impact of any excise tax increase and change in excise tax structure on price levels, legal sales and revenues from excise and other taxes on tobacco products. Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the effectiveness of their policies and whether those policies have an impact on tobacco use over time. The implementation of the MPOWER package is one useful indicator for assessing tobacco control overall. Tobacco taxation works best if it is implemented as part of a comprehensive MPOWER package. The tax share in the retail price of a selected tobacco product is one indicator of the effectiveness of tax policy, but a more important one is affordability, that is, whether tax increases do lead to price increases that are above income and general price increases. A useful indicator to assess the performance of the tax policy overall is the Tobacconomics Cigarette Tax Scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). Domestic policies in agriculture, industry, trade, finance and labour all have the potential to create or support incentives at different stages in tobacco production, manufacturing and distribution that can be counterproductive to the objectives of tobacco control and taxation. Greater domestic policy coherence should be pro- moted across different sectors of the government to ensure that public policies and interventions in these sectors do not counteract the intended public health impact of tobacco control and taxation. Differential tax structures and rates have the potential to distort the functioning of the internal market. Harmonization of tobacco taxation ensures the establishment and proper functioning of a single market; prevents tax revenue erosion, tax avoidance and tax evasion; and protects people’s health. In this context, tax competition, where countries simply undercut each other’s tax rate, might prevent governments from achieving their tobacco control objectives and raising sufficient funds to pursue public health policies. To avoid such a race to the bottom, countries can establish minimum tax rates on all tobacco products. A common high minimum specific excise tax is the best approach to ensure that taxes and prices are above a minimal level. 80 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N In recent years, the world has experienced the rise of new and emerging tobacco and nicotine products, including ENDS/ENNDS and HTPs, which the industry claims are safer than traditional tobacco products. The evidence so far suggests that these products could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of these products as well as initiation, cessation and switching of tobacco use behaviours among different socioeconomic groups, are not yet clear. Until more evidence for the claimed benefits of these tobacco products is available, caution should be taken in developing tax policy. Therefore, the current recommendation is for HTPs to be taxed at the same level as cigarettes on a per-unit basis regardless of tobacco content. In countries where they are not banned, ENDS/ENNDS products must be regulated and taxed in a manner that discourages uptake by youth and non-users. Taxing e-liquids is a key component of ENDS/ENNDS products taxation. Nicotine- and non-nicotine-containing e-liquids should be taxed equally. If ad valorem excise is chosen as the structure, the base should be applied on the retail price. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While the evolution of new and emerging tobacco and nicotine delivery systems merits the attention of administrators and regulators, it is worth remembering that the bulk of the world’s tobacco consumption is still in conventional cigarettes. CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 81 REFERENCES 1. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 2. Outcome of proceedings: Council conclusions concerning the structure and rates of excise duty applied to manufactured tobacco. Brussels: Council of the European Union; 2 June 2020 (8483/20; https://www.consilium.europa.eu/media/44235/st08483-en20.pdf, accessed 28 September 2020). 3. Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco: Article 7, paragraph 2. Brussels: OJEU. 2011;176:24–36 (https://eur-lex.europa. eu/LexUriServ/LexUriServ.do?uri=OJ:L:2011:176:0024:0036:EN:PDF, accessed 28 September 2020). 4. U.S. National Cancer Institute and World Health Organization.The economics of tobacco and tobacco control. NIH Publication No. 16-CA-8029A. Bethesda, MD: Department of Health and Human Services, National Institutes of Health, National Cancer Institute; 2016 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/brp/tcrb/monographs/monograph-21, accessed 28 September 2020). 5. Drope J, Schluger N, Cahn Z, Drope J, Hamill S, Islami F, et al. The tobacco atlas. Atlanta: American Cancer Society and Vital Strategies; 2018 (https://tobaccoatlas.org/wp-content/uploads/2018/03/ TobaccoAtlas_6thEdition_LoRes_Rev0318.pdf, accessed 24 January 2021). 6. Sector reports, cigarettes. In: GlobalData [website]. London: John Carpenter House; 2019 (restricted access). 7. Vellios N, Ross H, Perucic AM. 2018 trends in cigarette demand and supply in Africa. PLOS ONE. 2018;13(8):e0202467 (https://doi.org/10.1371/journal.pone.0202467, accessed 28 September 2020). 8. Keen MJ. The balance between specific and ad valorem taxation. Fiscal Studies. 1998;19(1):1–37 (https://www.ifs.org.uk/fs/articles/fsmick1.pdf, accessed 24 January 2021). 9. Kay JA, Keen MJ. The structure of tobacco taxes in the European community. IFS report series. Institute for Fiscal Studies; 1982. 10. Kay JA, Keen MJ. How should commodities be taxed?: Market structure, product heterogeneity and the optimal structure of commodity taxes. Euro Econ Rev. 1983;23(3):339–58. 11. Kay JA, Keen MJ. Commodity taxation for maximum revenue. Public Finan Q. 1987;15(4):371–85. 12. Kay JA, Keen MJ. Product quality under specific and ad valorem taxation. Public Finan Q. 1991;19(2):238–47. 13. Delipalla S, Keen MJ. The comparison between ad valorem and specific taxation under imperfect competition. J Public Econ. 1992;49(3):351–67. 14. Suits DB, Musgrave RA. Ad valorem and unit taxes compared. Q J Econ. 1953;67(4):598–604. 15. Skeath SE, Trandel GA. A pareto comparison of ad valorem and unit taxes in noncompetitive environments. J Public Econ. 1994;53(1):53–71. 16. Myles G. Imperfect competition and the optimal combination of ad valorem and specific taxation. Int Tax Public Finance. 1994;3:29–44. 17. Effectiveness of tax and price policies for tobacco control. Geneva: WHO Press, World Health Organization; 2011. (IARC handbooks of cancer prevention: tobacco control: Vol. 14; https://publications. iarc.fr/Book-And-Report-Series/Iarc-Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And- Price-Policies-For-Tobacco-Control-2011, accessed 28 January 2021) 18. Bevan J. GCC taxation. Aylesbury (UK): Rothmans Exports Limited; 1992 (https://www.industrydocuments. ucsf.edu/docs/yxpy0198, accessed 29 September 2020). 19. Morton IT. Note from IT Morton to I Birks regarding agreement on specific taxation; 8 February 1992 (https://www.industrydocuments.ucsf.edu/docs/#id=nxpy0198, accessed 29 September 2020). 20. Galgut R. Oman. BAT (UK and Export) Limited; 16 April 1992 (https://www.industrydocuments.ucsf. edu/docs/xxpy0198, accessed 29 September 2020). 21. Common excise tax agreement of the states of the Gulf Cooperation Council [website]. Abu Dhabi: UAE Ministry of Finance; 2016 (https://www.mof.gov.ae/En/lawsAndPolitics/govLaws/Documents/ Common%20Excise%20Tax%20Agreement.pdf, accessed 29 September 2020). 82 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 22. Brock B, Choi K, Boyle RG, Moilanen M, Schillo B. Tobacco product prices before and after a statewide tobacco tax increase. Tob Control. 2016;25(2):166–73. 23. Hanson A, Sullivan R. The incidence of tobacco taxation: evidence from geographic micro-level data. Natl Tax J. 2009;62(4):677–98. 24. Delipalla S, O’Donnell O. Estimating tax incidence, market power and market conduct: the European cigarette industry. Int J Ind Organ. 2001;19(6):885–908. 25. Keeler TE, Hu TW, Barnett PG, Manning WG, Hai-Yen S. Do cigarette producers price-discriminate by state? An empirical analysis of local cigarette pricing and taxation. J Health Econ. 1996;15:499–512. 26. Linegar DJ, van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2017;0:1–7 (https://tobaccocontrol.bmj.com/content/27/1/65, accessed17 February 2021). 27. Villanti AC, Collins LK, Niaura RS, Gagosian SY, Abrams DB. Menthol cigarettes and the public health standard: a systematic review. Public Health. 2017;17:983 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5747135/, accessed 24 January 2021). 28. Directive 2014/40/EU of the European Parliament and of the Council of 3 April 2014 on the approximation of the laws, regulations and administrative provisions of the member states concerning the manufacture, presentation and sale of tobacco and related products and repealing Directive 2001/37/EC: Article 7. Brussels: OJEU. 2014;57:1–38 (https://eur-lex.europa.eu/legal-content/EN/ TXT/PDF/?uri=OJ:L:2014:127:FULL&from=EN, accessed 24 January 2021). 29. Delipalla S, Keen MJ. Product quality and the optimal structure of commodity taxes. J Public Econ Theory. 2006;8:547–54. 30. Obara T, Tsugawa S. The welfare comparison of ad-valorem tax and specific tax with both quality and quantity choice of a consumer. Theor Econ Lett. 2017;7:1797–1813 (https://www.scirp.org/journal/ paperinformation.aspx?paperid=79862, accessed 24 January 2021). 31. Anderson S, de Palma A, Kreider B. Tax incidence in differentiated product oligopoly. Journal of Public Econ. 2001;81(2):173–92. 32. Razzolini L, Shughart WF, Tollison RD. On the third law of demand. Econ Inq. 2003;41:2. 33. Sobel RS, Garrett TA. Taxation and product quality: new evidence from generic cigarettes. J Polit Econ. 1997;105:880–7 (http://faculty.citadel.edu/sobel/All%20Pubs%20PDF/Generic%20Cigarettes. pdf, accessed 24 January 2021). 34. Nargis N, Fong GT, Chaloupka FJ, Li Q. The choice of discount brand cigarettes: a comparative analysis of International Tobacco Control (ITC) surveys in Canada and the United States (2002-2005). Tob Control. 2014;23:i86–i96 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/suppl_1/i86. full.pdf, accessed 24 January 2021). 35. Chaloupka FJ, Peck R, Tauras JA, Xu X, Yurekli A. Cigarette excise taxation: the impact of tax structure on prices, revenues and cigarette smoking. Cambridge: National Bureau of Economic Research; 2010 (NBER Working Paper 16287; https://www.nber.org/system/files/working_papers/w16287/w16287. pdf, accessed 24 January 2021). 36. Cigarettes in India, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 37. Gilmore AB, Tavakoly B, Taylor G, Reed H. Understanding tobacco industry pricing strategy and whether it undermines tobacco tax policy: the example of the UK cigarette market. Addiction. 2013;108(7):1317–26 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3746125/, accessed 24 January 2021). 38. Marsh L, Cameron C, Quigg R, Hoek J, Doscher C, McGree R, et al. The impact of an increase in excise tax on the retail price of tobacco in New Zealand. Tob Control. 2016;25(4):458–634. 39. Brock B, Schillo BA, Moilanen M. Tobacco industry marketing: an analysis of direct mail coupons and giveaways. Tob Control. 2015;24:505–8. 40. Chaloupka FJ, Warner KE. The economics of smoking. Handbook Health Econ. 2000;1:1539–1627. 41. Shang C, Chaloupka FJ, Fong G, Thompson M, O’Connor RJ. The association between tax structure and cigarette price variability: findings from the ITC Project. Tob Control. 2015;24:iii88–iii93 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/24/Suppl_3/iii88.full.pdf, accessed 24 January 2021) 42. Li Q, Hyland A, Fong GT, Jiang Y, Elton-Marshall T. Use of less expensive cigarettes in six cities in China: findings from the International Tobacco Control (ITC) China Survey. Tob Control. 2010;19(suppl CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 83 2):i63–i68 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/19/Suppl_2/i63.full.pdf, accessed 24 January 2021). 43. Hyland A, Laux FL, Higbee C, Hastings G, Ross H, Chaloupka FJ, et al. Cigarette purchase patterns in four countries and the relationship with cessation: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15(suppl III):iii59–iii64 (https://tobaccocontrol.bmj.com/ content/tobaccocontrol/15/suppl_3/iii59.full.pdf, accessed 24 January 2021). 44. Tsai YW, Yang CL, Chen CS, Liu, TC, Chen PF. The effect of Taiwan’s tax induced increases in cigarette prices and brand switching and the consumption of cigarettes. Health Econ. 2005;14:627–41. 45. Cummings KM, Hyland A, Lewit E, Shopland D. Use of discount cigarettes by smokers in 20 communities in the United States, 1988-1993. Tob Control. 1997;6(suppl 2):S25–S30 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/6/suppl_2/S25.full.pdf, accessed 24 January 2021). 46. Ross H, Tesche J. Undermining government tax policies: common strategies employed by the tobacco industry in response to increases in tobacco taxes. Economics of Tobacco Control Project, School of Economics, University of Cape Town and Tobacconomics, Health Policy Center, Institute for Health Research and Policy. Chicago: University of Illinois at Chicago; 2016 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC5747349/, accessed 26 January 2021). 47. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 48. Hiscock R, Branston JR, McNeill A, Hitchman SC, Partos TR, Gilmore AB. Tobacco industry strategies undermine government tax policy: evidence from commercial data. Tob Control. 2017;0:1–10 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 49. Filippidis F, Laverty A, Hone T, Been JV, Millett C. Association of cigarette price differentials with infant mortality in 23 European Union countries. JAMA Pediatr. 2017;171(11):1100–1106 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5710372/#:~:text=Cigarette%20price%20increases%20across%20 23,priced%20cigarettes%20during%20this%20period, accessed 26 January 2021). 50. López-Nicolás A, Belén Cobacho M, Fernández E. The Spanish tobacco tax loopholes and their consequences. Tob Control. 2013;22(E1):e21. 51. Ballester L, Auchincloss A, Robinson L, Mayne S. Exploring impacts of taxes and hospitality bans on cigarette prices and smoking prevalence using a large dataset of cigarette prices at stores 2001–2011, USA. Int J Environ Res Public Health. 2017;14(3):e318 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5369154/, accessed 24 January 2021). 52. Liber A, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia survey. Tob Control. 2015;24:iii83-iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed on 26 January 2021). 53. Feighery E, Ribisl K, Schleicher N, Zellers L, Wellington N. How do minimum cigarette price laws affect cigarette prices at the retail level? Tob Control. 2005;14:80–85 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/14/2/80.full.pdf, accessed 26 January 2021). 54. Tynan MA, Ribisi KM, Loomis BR. Impact of cigarette minimum price laws on the retail price of cigarettes in the USA. Tob Control. 2012;22:e1. 55. Huang J, Chriqui JF, DeLong H, Mirza M, Diaz MC, Chaloupka FJ. Do state minimum markup/price laws work? Evidence from retail scanner data and TUS-CPS. Tob Control. 2016;25:i52–i59 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i52.full.pdf, accessed 26 Jan 2021). 56. Whitehead R, Brown L, Riches E, Rennick L, Armour G, McAteer J, et al. Strengths and limitations of tobacco taxation and pricing strategies. Edinburgh: NHS Health Scotland; 2018 (http://www. healthscotland.scot/media/1829/rapid-evidence-review-strengths-and-limitations-of-tobacco- taxation-and-pricing-strategies.pdf, accessed 27 January 2021). 57. Golden S, Farrelly M, Luke D, Ribisl K. Comparing projected impacts of cigarette floor price and excise tax policies on socioeconomic disparities in smoking. Tob Control. 2016;25:i60-i66 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i60.full.pdf, accessed 27 January 2021). 58. Doogan NJ, Wewers ME, Berman M. The impact of a federal cigarette minimum pack price policy on cigarette use in the USA. Tob Control. 2018;27(2):203–08. 84 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 59. Marynak K, Xu X, Wang X, Holmes CB, Tynan MA, Pechacek T. Estimating the impact of raising prices and eliminating discounts on cigarette smoking prevalence in the United States. Public Health Rep. 2016;131:536–43 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4937114/#:~:text=We%20found%20 that%20each%20price,smokers%20aged%20%E2%89%A526%20years., accessed 27 January 2021). 60. McLaughlin I, Pearson A, Laird-Metke E, Ribisl K. Reducing tobacco use and access through strengthened minimum price laws. Am J Public Health. 2014;104(10):1844–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4937114/#:~:text=We%20found%20that%20each%20price,smokers%20aged%20 %E2%89%A526%20years., accessed 27 January 2021). 61. European Commission v. French Republic, C-197/08, European Court of Justice 4 March 2019 (https:// eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX%3A62008CC0197, accessed on 27 January 2021). 62. Tobacco price promotion: policy responses to industry price manipulation. Boston: Public Health and Tobacco Policy Center; October 2011 (http://tobaccopolicycenter.org/documents/Price%20 Promotion%20Overview%20FINAL.pdf, accessed 28 September 2020). 63. Golden SD, Holt Smith M, Feighery EC, Roeseler A, Rogers T, Ribisl KM. Beyond excise taxes: a systematic review of literature on non-tax policy approaches to raising tobacco product prices. Tob Control. 2016;25:377–85. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/4/377.full.pdf, accessed 27 January 2021). 64. Branston J, Gilmore S. The case for Ofsmoke: the potential for price cap regulation of tobacco to raise ₤500 million per year in the UK. Tob Control. 2014;23(1):45–50 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3812939/, accesed 27 January 2021). 65. von Lampe K, Kurti M, Johnson J. “I’m gonna get me a loosie” understanding single cigarette purchases by adult smokers in a disadvantaged section of New York City. Prev Med Rep. 2018;12:182–5. (https:// www.sciencedirect.com/science/article/pii/S2211335518302195, accessed 27 January 2021). 66. de Ojeda A, Barnoya J, Thrasher JF. Availability and costs of single cigarettes in Guatemala. Nicotine Tob Res. 2013;15(1):83–7 . 67. Lal P, Kumar R, Ray S, Sharma N, Bhattarcharya B, Mishra D, et al. The single cigarette economy in India - a back of the envelope survey to estimate its magnitude. Asian Pac J Cancer Prev. 2015;16(13):5579-82 (http://journal.waocp.org/article_31295_9ebd3981652934bbf5d78ec066b6a39d.pdf, accessed 27 January 2021). 68. Coady MH, Chan CA, Sacks R, Mbamalu IG, Kansagra SM. The impact of tax on purchasing behaviors among New York City smokers. Am J Public Health. 2013;103(6):e54-e60 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC3698745/, accessed 27 January 2021). 69. Cigarettes in Thailand, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 70. Cigars in Germany, 2019. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). 71. Report from the Commission to the Council on Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2018 (COM(2018) 17 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/report_excise_duty_manufactured_ tobacco_12012018_en.pdf, accessed 28 September 2020). 72. Gallet CA, List JA. Cigarette demand: a meta-analysis of elasticities. Health Econ. 2003;12(10):821–35. 73. Guidelines for implementation of article 6 of the WHO FCTC. Geneva: World Health Organization; 2016 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6.pdf, accessed 29 September 2020). 74. DeCicca P, Kenkel D, Liu F. Excise tax avoidance: the case of state cigarette taxes. J Health Econ. 2013;32(6):1130–41 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3855906/, accessed 27 January 2021). 75. Jawad M, Lee JT, Glantz S, Millett C. Price elasticity of demand of non-cigarette tobacco products: a systematic review and meta-analysis. Tob Control; 2018;27(6):689–95 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6056338/, accessed 27 January 2021). 76. John RM. Price elasticity estimates for tobacco products in India. Health Policy Plan. 2008;23(3):200–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2800993/, accessed on 27 January 2021). 77. Wasserman J, Manning WG, Newhouse JP, Winkler JD. The effects of excise taxes and regulations on CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 85 cigarette smoking. Journal of Health Econ, Elsevier. 1991;10(1):43–64. 78. Cheng K, Kenkel D. U.S. cigarette demand: 1944-2004. B.E. Journal of Econ Analysis & Policy. 2010;10(1):10.2202/1935-1682.2438 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3820165/, accessed 28 January 2021). 79. Bobak M, Jha P, Nguyen S, Jarvis M. Poverty and smoking. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000. 80. Peck, RM. Equity issues, tobacco, and the poor. Yurekli A, de Beyer J, editors. Washington (DC): World Bank Group; 2013. (Economics of Tobacco Toolkit, Tool 6; https://openknowledge.worldbank.org/ handle/10986/16265, accessed 28 January 2021). 81. Kenkel DS, Schmeiser MD, Urban C. Is smoking inferior?: evidence from variation in the earned income tax credit. J Hum Res. 2014;49:1094–1120. 82. Ruhm, CJ. Healthy living in hard times. Journal of Health Econ. 2005;24(2):341–63. 83. Ruhm, CJ. Are recessions good for health? The Quarterly Journal of Econ. 2000;115(2):617-50. 84. Tarantilis F, Athanasakis K, Zavras D, Vozikis A, Kyriopoulos I. Estimates of price and income elasticity in Greece. Greek debt crisis transforming cigarettes into a luxury good: an econometric approach. BMJ Open. 2015;5(1):e004748. (https://bmjopen.bmj.com/content/bmjopen/5/1/e004748.full.pdf, accessed 28 January 2021). 85. Kaiser M, Reutter M, Sousa-Poza A, Strohmayer K. Smoking and the business cycle: evidence from Germany. Discussion Paper Series. Bonn, Germany: IZA Institute of Labour Economics; 2017 (http:// ftp.iza.org/dp10953.pdf, accessed 28 September 2020). 86. Tingum EN, Mukong AK, Mdege N. The effects of price and non-price policies on cigarette consumption in South Africa. Tob Induc Dis. 2020;18(July):62 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC7398599/, accessed 28 January 2021). 87. Vardavas CI, Filippidis F, Agak, I, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the importance of earmarking revenue to health care and tobacco control. Tob Induc Dis. 2012;10(1):21 (http://www. tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the-revenue-to- health-care-and-tobacco,66110,0,2.html, accessed 28 January 2021). 88. Marquez P, Catinkaya V. Tobacco taxation in Turkey: an overview of policy measures and results. Washington (DC): World Bank Group; 2017 (https://openknowledge.worldbank.org/handle/10986/26387, accessed 10 November 2020). 89. Blecher EH, van Walbeek CP. An international analysis of cigarette affordability. Tob Control. 2004;13:339– 46 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/13/4/339.full.pdf, accessed 28 January 2021). 90. Goodchild M, Sinha P, Munish VG, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/handle/10665/331836, accessed 28 January 2021). 91. John RM, Dauchy E. Trends in affordability of tobacco products before and after the transition to GST in India. Tob Control Online First. 20 March 2020. 92. Chaloupka FJ, Drope J, Siu E, Vulovic V, Stoklosa M, Mirza M, et al. Tobacconomics cigarette tax scorecard. Chicago: Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/636/uic-tobacco-scorecard-report-final.pdf, accessed 28 January 2021). 93. Kanbur R, Keen MJ. Jeux sans frontieres: tax competition and tax coordination when countries differ in size. Am Econ Rev. 1993;83(4):877–92 (https://www.researchgate.net/publication/4901220_Jeux_ Sans_Frontieres_Tax_Competition_and_Tax_Coordination_When_Countries_Differ_in_Size, accessed 28 January 2021). 94. Bouw A. Tobacco taxation in the European Union: an overview (English). Washington (DC): World Bank Group; 2017 (http://documents.worldbank.org/curated/en/493581492415549898/Tobacco- taxation-in-the-European-Union-an-overview, accessed on 28 September 2020). 95. Blecher E, Drope J. The rewards, risks and challenges of regional tobacco tax harmonization. Tob Control. 2014;0:1–5. 96. Directive No.n° 01/2017/CM/UEMOA of 22 December 2017. Paris: Droit-Afrique; 2017 (https://www. 86 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N droit-afrique.com/uploads/UEMOA-Directive-2017-01-droits-accise-tabacs.pdf, accessed 29 September 2020). 97. Mirzoev T, Zhu L, YangY, Pescatori A, Matsumoto A, Callen T. The future of oil and fiscal sustainability in the GCC region. Middle East and Central Asia Research Department. Washington (DC): International Monetary Fund; 2020 (https://www.imf.org/-/media/Files/Publications/DP/2020/English/FOFSGCCEA. ashx, accessed 28 January 2021). 98. State cigarette excise tax rates & rankings. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/factsheets/0097.pdf, accessed 28 September 2020). 99. A map and table comparing cigarette prices in Canada (March 2018). Toronto: Non-Smokers’ Rights Association/Smoking and Health Action Foundation; 2018 (https://nsra-adnf.ca/wp-content/ uploads/2018/03/180302_map_and_table.pdf, accessed 4 February 2021). 100. European Commission excise duty rates on manufactured tobacco. Brussels: European Commission; 31 December 2017 (https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco- energy/excise-duties-tobacco_en, accessed 1 July 2018). 101. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2019 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study-on-the-tobacco-taxation-directive-2019_en.pdf, accessed 28 September 2020). 102. Liber AC. Heated tobacco products and combusted cigarettes: comparing global prices and taxes. Tob Control; 2018;28(6). 103. Heated tobacco products: information sheet - 2nd edition. Geneva: World Health Organization; 2020 (https://www.who.int/publications/i/item/WHO-HEP-HPR-2020.2, accessed 28 January 2021). 104. Kass Lempert L, Glantz S. Analysis of FDA’s IQOS marketing authorisation and its policy impacts. Tob Control 2020;0:1–9. 105. Chivers E, Janka M, Franklin P, Mullins M, Larcombe A. Nicotine and other potentially harmful compounds in “nicotine-free” e-cigarette liquids in Australia. Med J Aust. 2019;210(3):118–9 (https://onlinelibrary. wiley.com/doi/epdf/10.5694/mja2.12059, accessed 28 September 2020). 106. Wenzl T, Zelinkova Z, Regueiro J, Giri A. Scientific-technical support activities to DG TAXUD-C-2 on the option to include e-cigarettes within the scope of excisable goods for the Impact Assessment on a possible revision of Directive 2011/64/EU. Administrative Arrangement TAXUD/2015/DE/329. Luxembourg: Publications Office of the European Union; 2017 (https://publications.jrc.ec.europa.eu/ repository/bitstream/JRC105828/kjna29062enn.pdf, accessed 28 September 2020). 107. Lisko JG, Tran H, Stanfill SB, Blount BC, Watson CH. Chemical composition and evaluation of nicotine, tobacco alkaloids, pH, and selected flavors in e-cigarette cartridges and refill solutions. Nicotine Tob Res. 2015;17(10):1270–78. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4573955/, accessed 28 January 2021). 108. Ghebreyesus TA. Progress in beating the tobacco epidemic. Lancet. 17 August 2019;394:548–49 (https:// www.thelancet.com/pb-assets/Lancet/pdfs/S0140673619317301.pdf?hss_channel=tw-27013292, accessed 29 September 2020). 109. Smoking cessation, a report of the Surgeon General. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2020. (https://www.hhs.gov/sites/default/ files/2020-cessation-sgr-executive-summary.pdf, accessed 8 October 2020). 110. E-cigarettes are harmful to health. Geneva: World Health Organization. 5 February 2020 (https://www. who.int/news-room/detail/05-02-2020-e-cigarettes-are-harmful-to-health, accessed 29 September 2020). 111. Shan L, Manzione LC, Azagba S. Psychological well-being and dual-use of cigarettes and e-cigarettes among high school students in Canada. J Affect Disord. 2020;265:357–63 112. King BA, Gammon DG, Marynak KL, Rogers T. Electronic cigarettes sales in the United States, 2013- 2017. JAMA. 2018;320(13):1379-80 (https://jamanetwork.com/journals/jama/fullarticle/2705175, accessed 28 January 2021). 113. LaVito A. Popular e-cigarette Juul’s sales have surged almost 800 percent over the past year [website]. CNBC. 2 July 2018 (https://www.cnbc.com/2018/07/02/juul-e-cigarette-sales-have-surged-over-the- past-year.html, accessed 28 September 2020). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 87 114. Statement from FDA Commissioner Scott Gottlieb, M.D., on new enforcement actions and a youth tobacco prevention plan to stop youth use of, and access to, JUUL and other e-cigarettes. Washington (DC): US Food and Drug Administration; 23 April 2018 (www.fda.gov/NewsEvents/Newsroom/ PressAnnouncements/ucm605432.htm, accessed 29 September 2020). 115. Paul K. San Francisco becomes first US city to ban sale of e-cigarettes. The Guardian. 25 June 2019 (https://www.theguardian.com/us-news/2019/jun/25/san-francisco-e-cigarettes-vaping-juul-ban, accessed 29 September 2020). 116. Pesko MF, Courtemanche CJ, Maclean JC. The effects of traditional cigarette and e-cigarette tax rates on adult tobacco product use. J Risk Uncertain. 2020;60:229–258. 117. Cotti CD, Courtemanche CJ, Maclean JC, Nesson ET, Pesko MJ, Tefft N. The effects of e-cigarettes taxes on e-cigarettes prices and tobacco product sales: evidence from retail panel data. Cambridge: National Bureau of Economic Research; 2020 (NBER Working Paper 26724; https://www.nber.org/ system/files/working_papers/w26724/w26724.pdf, accessed 28 January 2021). 118. Saffer H, Dench DL, Grossman M, Dave DM. E-cigarettes and adult smoking: evidence from Minnesota. Cambridge: National Bureau of Economic Research; 2019 (NBER Working Paper 26589; https://www. nber.org/system/files/working_papers/w26589/w26589.pdf, accessed 28 January 2021). 119. Huang J, Gwarnicki C, Xu X, Caraballo RS, Wada R, Chaloupka FJ. A comprehensive examination of own- and cross-price elasticities of tobacco and nicotine replacement products in the U.S. Prev Med. 2018;117:107–14 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6195827/, accessed 28 September 2020). 120. Zheng Y, Zhen C, Dench D, Nonnemaker JM. U.S. demand for tobacco products in a system framework. Health Econ. 2017;26:1067–86. 121. Stoklosa M, Drope JM, Chaloupka FJ. Prices and e-cigarette demand: evidence from the European Union. Nicotine Tob Res. 2016;18(10):1973–80. 122. Grace RC, Kivell BM, Laugesen M. Estimating cross-price elasticity of e-cigarettes using a simulated demand procedure. Nicotine Tob Res. 2015;17(5):592–8. (https://www.researchgate.net/ publication/270291421_Estimating_Cross-Price_Elasticity_of_E-Cigarettes_Using_a_Simulated_ Demand_Procedure, accessed 28 January 2021). 123. Huang J, Tauras J, Chaloupka FJ. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob Control. 2014;23:iii41–iii47 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/23/suppl_3/iii41.full.pdf, accessed 17 February 2021). 124. Cheng KW, Chaloupka FJ, Shang C, Ngo A, Fong GF, Borland R, et al. Prices, use restrictions and electronic cigarette use—evidence from wave 1 (2016) US data of the ITC Four Country Smoking and Vaping Survey. Addiction. 2019;114(suppl 1):115–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC6661221/, accessed 28 January 2021). 125. Liber AC, Drope JM, Stoklosa M. Combustible cigarettes cost less to use than e-cigarettes: global evidence and tax policy implications. Tob Control. 2016;0:1–6 (https://www.researchgate.net/ publication/299473181_Combustible_cigarettes_cost_less_to_use_than_e-cigarettes_Global_ evidence_and_tax_policy_implications, accessed 28 January 2021). 126. Chaloupka FJ, Sweanor D, Warner KE. Differential taxes for differential risks — toward reduced harm from nicotine yielding products. N Engl Journal Med. 2016;373(7):594-7. 127. Goniewicz ML, Knysak J, Gawron M, Kosmider L, Sobczak A, Kurek J, et al. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob Control. 2014;23:133–9. 128. Sweet L, Brasky TM, Cooper S, Doogan N, Hinton A, Klein EG, et al. Quitting behaviors among dual cigarette and e-cigarette users and cigarette smokers enrolled in the tobacco user adult cohort. Nicotine Tob Res. 2019;21(3):278–84 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6379027/, accessed 28 September 2020). 129. Wills TA, Knight R, Williams RJ, Pagano I, Sargent JD. Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics. 2014;135(1):e43–e51 (https://www. ncbi.nlm.nih.gov/pmc/articles/PMC4279062/, accessed 28 January 2021). 130. Electronic Nicotine Delivery Systems: report by WHO. Geneva: World Health Organization; 2014. (FCTC/ COP/6/10 Rev.1; https://apps.who.int/gb/fctc/PDF/cop6/FCTC_COP6_10Rev1-en.pdf, accessed 28 January 2021) 88 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 131. WHO report on the global tobacco epidemic 2017: monitoring tobacco use and prevention policies. Geneva: World Health Organization; 2017 (https://www.who.int/tobacco/global_report/2017/en/, accessed 28 January 2021). 132. Regulation of the Minister of Finance of the Republic of Indonesia, number 146, PMK.010/2017 on excise tax rates of tobacco products. Jakarta: Ministry of Finance Indonesia; 2017. 133. Gourdet CK, Chriqui JF, Chaloupka FJ. A baseline understanding of state laws governing e-cigarettes. Tob Control. 2014;23:iii37-iii40 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4078672/, accessed 28 January 2021). 134. An act to ensure that nicotine products are taxed equally. 127th Maine Legislature: first regular session-2015. H.P. 670, Legislative Document No. 973. 2015 (https://www.mainelegislature.org/legis/ bills/getPDF.asp?paper=HP0670&item=1&snum=127&PID=, accessed 28 January 2021). 135. Evaluation of the Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco. Commission Staff Working Document. Brussels: European Commission; 2020 (SWD(2020) 32 final; https://ec.europa.eu/taxation_customs/sites/taxation/files/10- 02-2020-tobacco-taxation-report.pdf, accessed 28 September 2020). 136. Electronic nicotine delivery systems and electronic non-nicotine delivery systems (ENDS/ENNDS): report by WHO. Geneva: World Health Organization; 2016. (FCTC/COP/7/11; https://www.who.int/ fctc/cop/cop7/FCTC_COP_7_11_EN.pdf, accessed 28 September 2020). 137. E-cigarette use among youth and young adults: a report of the Surgeon General. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2016 (https://e-cigarettes. surgeongeneral.gov/documents/2016_SGR_Full_Report_non-508.pdf, accessed 28 September 2020). 138. Wagener TL, Floyd EL, Stepanov I, Driskill LM, Frank SG, Meier E, et al. Have combustible cigarettes met their match? The nicotine delivery profiles and harmful constituent exposures of second-generation and third-generation electronic cigarette user. Tob Control 2017;26:e23–e28 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/26/e1/e23.full.pdf, accessed 28 September 2020). 139. Vapour products data. In: Euromonitor International [website]. London: Euromonitor International; 2019 (restricted access). 140. Global market data: segment insights. In: GlobalData [website]. London: John Carpenter House; 2020 (restricted access). CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 89 ANNEX 2.1 Table A2.1 Countries that apply different types of cigarette excise tax structures, 2018 SPECIFIC EXCISE (65 COUNTRIES) AD VALOREM EXCISE (42 COUNTRIES) MIXED EXCISE (63 COUNTRIES) NO EXCISE (15 COUNTRIES) Albania, Andorra, Australia, Azerbaijan, Barbados, Belarus, Belize, Bolivia (Plurinational State of ), Burundi, Canada, Cook Islands, Dominica, Ecuador, Eswatini, Fiji, Gambia, Honduras, Iceland, India, Indonesia, Jamaica, Japan, Jordan, Kazakhstan, Kenya, Kiribati, Kyrgyzstan, Lesotho, Malaysia, Mauritius, Mongolia, Mozambique, Myanmar, Namibia, Nepal, New Zealand, Nicaragua, Norway, Pakistan, Palau, Papua New Guinea, Peru, Philippines, Republic of Korea, Saint Lucia, Saint Vincent and the Grenadines, Samoa, Seychelles, Singapore, Solomon Islands, South Africa, Sri Lanka, Suriname, Tajikistan, Timor-Leste, Tonga, Trinidad and Tobago, Uganda, United Republic of Tanzania, USA, Uruguay, Uzbekistan, Vanuatu, Yemen, Zimbabwe Argentina, Armenia, Bahrain, Bangladesh, Benin, Burkina Faso, Cabo Verde, Cambodia, Cameroon, Chad, Comoros, Côte d’Ivoire, Cuba, Democratic Republic of the Congo, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Ghana, Grenada, Guatemala, Guinea-Bissau, Liberia, Madagascar, Mali, Mauritania, Niger, Panama, Paraguay, Saint Kitts and Nevis, Saudi Arabia, Senegal, Sierra Leone, Sudan, Syrian Arab Republic, Togo, Turkmenistan, Tuvalu, United Arab Emirates, Venezuela (Bolivarian Republic of ), Viet Nam, Zambia Algeria, Austria, Belgium, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Central African Republic, Chile, China, Colombia, Congo, Costa Rica, Croatia, Cyprus, Czechia, Denmark, Dominican Republic, Egypt, El Salvador, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iran (Islamic Republic of ), Ireland, Israel, Italy, Lao People’s Democratic Republic, Latvia, Lebanon, Lithuania, Luxembourg, Malta, Mexico, Montenegro, Morocco, Netherlands, Nigeria, North Macedonia, Poland, Portugal, Republic of Moldova, Romania, Russian Federation, Rwanda, Sao Tome and Principe, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Thailand, Tunisia, Turkey, Ukraine, United Kingdom, West Bank and Gaza Strip Afghanistan, Angola, Antigua and Barbuda, Democratic People’s Republic of Korea, Iraq, Kuwait, Libya, Maldives, Marshall Islands, Micronesia (Federated States of ), Nauru, Niue, Oman,a Qatar,a Somalia a This table shows the status of cigarette excise tax structures as of July 2018 and does not account for changes occurring after that date, in particular for the cases of Qatar and Oman, which introduced excise on tobacco in January 2019 and June 2019, respectively. Source: WHO RGTE. 90 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 2.2 THE ANALYTICS OF THE TAX BASE ELASTICITY Assume tax revenue R = tsQ or R = tvPQ , where Q is the quantity consumed, ts is the specific tax, tv is the ad valorem tax and P is the consumer price. The following equations can help to illustrate the different components of the tax base elasticity. Under a specific excise regime, change in revenue depends essentially on the change in consumption: where R is the tobacco tax revenue, is the specific excise tax and is the tobacco tax base elasticity. The tax base elasticity is made of: where ε, the price elasticity = , is the degree of pass-through of the specific excise tax rate increase on consumer price and is the tax-price ratio. Under an ad valorem excise regime, change in revenue depends essentially on the change in tobacco expenditure: where R is the tobacco tax revenue, tav is ad valorem excise tax and ηav is the tobacco tax base elasticity. The tax base elasticity here is made of: where is the degree of pass-through of the ad valorem excise tax rate increase on consumer price, is the tax-price ratio and ε the price elasticity = . ∂R = Q (1+ηs)∂ ts ηs = ε ts P ∂P ∂ts tav P ηav = (1 + ε) tav P ∂P ∂tav∂P ∂tav ∂P ∂ts ΔQ ΔP P Q ΔQ ΔP P Q ts P ∂R = P Q (1+ηav)∂tav CHAP T ER 2. TO BACCO E XCISE TA X PO LI C Y 91 ANNEX 2.3 ELEMENTS OF THE DEVICES THAT MAKE UP ENDS/ENNDS PRODUCTS The main components of any ENDS/ENNDS kit include essentially: • USB charger (not a car charger) • Inbuilt battery Additionally, For open systems • Tanks (refillable containers) with removable atomizer (often sold bundled with atomizers) • Clearomizers/refillable pods (no removable atomizer) • E-liquid For closed systems • Disposable e-cigarettes: not rechargeable, thrown away after e-liquid is finished • Nondisposable e-cigarettes: – Pre-filled cartomizers (cartridges designed to go with the cigalike kit) – Pre-filled tank refills/pods (pods or cartridges designed to go with the prefilled tank/pod kits) Some definitions: • Atomizer: uses a heating element to vaporize the e-liquid • Cartomizer: combines the cartridge/tank and the atomizer • Clearomizer: same as cartomizer, uses different technology • Cartridge/tank/pod: container that includes the e-liquid In summary, ENDS/ENNDS product devices include the following: • USB charger (not a car charger) • Inbuilt battery • Disposable e-cigarettes • Atomizer • Cartomizer/clearomizer • Cartridge/tank/pod with or without atomizer • Pre-filled cartridge/tank/pod (for closed systems, includes e-liquid) • E-liquid (added in the cartridge/tank/pod in open systems) Source: ECigIntelligence, 2020. Information also obtained from vaping websites, including https://www. misthub.com/blogs/vape-tutorials/76788357-tutorial-atomizer-vs-cartomizer-vs-clearomizer, http:// www.bestclearomizer.com/clearomizer-vs-cartomizer-vs-atomizer/, https://wayofleaf.com/accessories/ vapes/atomizer-vs-clearomizer-vs-cartomizer, https://wayofleaf.com/accessories/vapes/atomizer-vs- clearomizer-vs-cartomizer, accessed 15 July 2020. 92 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 93 CHAPTER 3. Tobacco tax administration 3.1 INTRODUCTION Imposing excise taxes on tobacco products usually serves more than one purpose. Governments often find themselves balancing interests between financial and public health objectives. Both objectives can best be achieved by an efficient and effective competent authority with strong technical capacity to enforce and collect taxes. A competent authority is the agency, organization or department that is legally as- signed to complete a particular activity; in the case of administering tobacco taxes, the competent authority is often a tax administration, revenue authority, customs department or ministry of finance. Article 6 of the WHO FCTC (1), along with its guidelines (2), provides a solid foundation for sound tax administration. As stated under section 1.5 in the guidelines, tobacco tax systems should be efficient and effective. They should be structured to minimize the costs of compliance and administration, while ensuring that the desired level of tax revenue is raised and health objec- tives are achieved. Efficient and effective administration of tobacco tax systems enhances tax compliance and collection of tax revenue while reducing tax evasion and the risk of illicit trade. Efficiency in tax administration refers to minimizing the costs per unit of tax revenue collected. It is measured by comparing the resources used with the revenues gener- ated. Effectiveness in tax administration refers to a high level of compliance – also described as taxpayers meeting their obligations. Thus, an efficient and effective competent authority collects the tax at a minimum cost while ensuring conformity to the rules. Tobacco taxation is the single most effective tobacco control measure for re- ducing tobacco use and is best implemented as part of a comprehensive tobacco control plan (3). Illicit trade – including smuggling and illicit manufacturing – and tax avoidance undermine the effectiveness of tax policies and their objectives (4). The impact on illicit trade is often cited by opponents of tax increases, who argue that increasing taxes increases illicit trade. They contend that illicit trade can lead 94 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N to lower revenues for governments and lower prices. The challenge faced by the competent authorities is to ensure that due taxes are declared and collected on all tobacco products that are manufactured in and/or imported into its jurisdiction, while at the same time detecting tobacco products that are illegally manufactured in and/or imported into its jurisdiction, stopping such activity and prosecuting the responsible parties. This chapter describes the shared characteristics of good tax administrations, including best practices based on country experiences. It regularly refers to the WHO FCTC, and – given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade – draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (5). Any practice or conduct prohibited by law and related to production, shipment, receipt, possession, distribution, sale or purchase of tobacco products – including any practice or conduct intended to facilitate such activity – is considered as illicit trade (Article 1). The objective of the Protocol is to eliminate and prevent all forms of illicit trade in tobacco products. At the same time, the Protocol includes measures for tobacco tax administration based on international best practices, which makes it relevant for all countries, even those that are not Parties to it. The Protocol was adopted at the fifth session of the COP to the WHO FCTC in 2012 and entered into force on 25 September 2018. As indicated in the Preamble, it was developed in response to the increasing international illicit trade in tobacco products (5). The Protocol covers three main areas: (1) measures to control the supply chain (Part III); (2) measures dealing with offences, including sanctions (Part IV) and (3) international cooperation (Part V). Different provisions of the Protocol are discussed in detail throughout this chapter, and section 3.4 is devoted specifically to control and enforcement. 3.2 INSTITUTIONAL ARRANGEMENTS Competent authorities that collect taxes effectively in an efficient way share a number of attributes. The organizational structures of these authorities contain clearly defined roles, responsibilities and rules for coordination among relevant bodies. Moreover, competent authorities collect data regularly and manage information needed for assessing risks. The key to successful risk management is to share this information among relevant authorities both within a country and between countries. Effective and efficient competent authorities also regularly evaluate their performance and accountability according to key performance indicators to identify areas for improve- ment. These characteristics are discussed in greater detail in the following sections. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 95 3.2.1. CLEARLY DEFINED ROLES AND RESPONSIBILITIES OF COMPETENT AUTHORITIES The designation of competent authorities for the implementation and enforcement of tax laws – including clear definitions of the boundaries of authority among numerous agencies within a country – is essential for efficient collection of taxes. Areas where different agencies need to cooperate and share data must also be defined. Overlap of activities by different authorities leads to inefficient use of resources, whereas gaps create opportunities for fraud, leading to ineffective tax laws. The importance of clearly defined roles and responsibilities applies not only to tax authorities and customs but also to law enforcement agencies, including police and border control forces. The implementation and enforcement of taxation is organized differently in various countries. The most common structure separates customs and tax administration. The trend since the 1990s, however, has been to combine these functions into one agency, such as Her Majesty’s Revenue and Customs (HMRC) in the United King- dom, SUNAT in Peru1 and AFIP in Argentina.2 Several countries have increased coordination between tax and customs by creating a revenue secretariat and also implementing systems to share tax records as a single taxpayer account. Coordina- tion between tax policy and tax administration authorities has also increased. One can think of combining both into one department within the ministry of finance or ensure that tax administration authorities are consulted during the tax policy process. Some tasks, such as licensing, may be handled by other ministries such as the ministries of health, agriculture or trade. For example, the Ministry of Health of Brunei and the Health Science Authority of Singapore are responsible for the licensing of importers of tobacco products (6). In some federal countries, including Colombia and the United States, excise taxes – including tobacco taxes – are collected and enforced by local or state tax administrations. Other countries have organized the administration of national taxes by establishing a single unified revenue body. Particularly in larger economies, that body is often responsible for both direct and indirect taxes, including excise taxes, and reports to the ministry of finance. All the functions needed for effective and efficient tax administration are established within these bodies (7). No matter what the institutional arrangements may be, it is vital that the agencies cooperate and exchange information and that their competencies find their basis in law. More information on this topic is provided in section 3.2.2. 1 Law Decreto Supremo 061-2002-PCM - Disponen fusión por absorción de la Superintendencia Na- cional de Administración Tributaria – SUNAT con la Superintendencia Nacional de Aduanas - Aduanas [Supreme decree year 2002 about the merger between Tax and Customs Administration]. Lima: El Peruano, 12, July 2002 (in Spanish) (http://www.sunat.gob.pe/legislacion/sunat/ds061-2002-PCM.pdf, accessed 13 November 2020). 2 Administracion Federal de Ingresos Publicos, Decreto 618/1997 [Federal Administration of Public Revenue, Decree 618] (in Spanish) (http://servicios.infoleg.gob.ar/infolegInternet/an- exos/40000-44999/44432/norma.htm, accessed 13 November 2020). 96 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Many countries, however, have separate bodies for the collection of taxes and customs duties. A 2015 survey of 135 tax administrations worldwide found that only 36% of them were responsible for both tax administration and customs ad- ministration (8). In most countries, customs authorities are more likely to collect excise duties on imports, and in many countries, VAT or sales tax is collected jointly with tobacco tax, particularly for imported products. This simplifies controls and creates synergy by unifying common processes and procedures, resulting in cost savings for tax administrations and taxpayers. The involvement of multiple bodies in tax collection requires especially good collaboration and information-sharing to ensure efficient and effective collection of taxes and duties. KEY TAKEAWAY 1 Institutional arrangements with clearly defined roles and responsibilities – designed to prevent overlaps and voids – contribute to effective and efficient tax administration. 3.2.2. EFFECTIVE COORDINATION AMONG RELEVANT BODIES Coordination at the national level Coordination among relevant bodies is key to effective tobacco tax administration. This means not only clearly defined roles and responsibilities, as described in the previous section, but also coordination among the competent authority, customs and those responsible for formulating, analysing and implementing tax policy. Regardless of the institutional arrangements – whether the responsible parties are all within the ministry of finance or in separate government agencies – all parties need to cooperate and exchange information to optimize tax collection and enforcement of tax policy. In practice, this means that information should be shared among, for example, customs, local government units that issue licences and health authori- ties – particularly those that regulate the sale of tobacco products. For tax authorities, the most relevant information concerning excise taxes in- cludes the identity of taxpayers and those involved in the trade of tobacco (import and export data, licences, criminal records, tax returns, bank statements, etc.); the category, quantity, value and location of manufactured goods; and the movement of those goods until all taxes are paid. Legal impediments to obtaining this informa- tion – such as bank secrecy or privacy regulations – should be kept in mind, and where needed, exceptions for fiscal procedures should be incorporated into law. Seizure data are also a valuable source of information; more details on this are provided in section 3.4. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 97 Tax authorities should regularly coordinate with law enforcement agencies – such as the police and border control forces, depending on a country’s laws – to properly monitor tobacco-related activities and enforce the tax laws. Often, the competent tax authority and customs authorities work in close cooperation with anti-fraud teams.3 Coordination and sharing of information can be required in legislation or regulations to ensure a streamlined process and avoid confusion. This can be done on an ad hoc basis as needed or with formal planned exchanges of information and regular meetings. It is recommended that at least a legal basis for exchange or access to information among government bodies be established to prevent claims during legal procedures that evidence was obtained unlawfully. Some countries go beyond exchanging information and cooperation. In the Neth- erlands, for example, customs authorities not only carry out work for the Ministry of Finance, they also carry out non-fiscal tasks for seven other departments, including the Ministry of Agriculture, Nature and Food Quality; the Ministry of Justice and Security; and the Ministry of Foreign Affairs (9). These activities are often based on bilateral agreements between the Ministry of Finance and the other departments. In other countries, such as the United States and Canada, Customs and Border Protection are not part of the Ministry of Finance; they are part of the Department of Homeland Security in the United States and the Ministry of Public Safety and Emer- gency Preparedness in Canada. These agencies also carry out many non-fiscal tasks. Along with the implementation of new tobacco control and tax laws, several countries have also created high-level committees to ensure good coordination and implementation of the laws. Led by health and finance ministries, committees ensure coordination and fine-tuning to achieve desired results. Botswana, Chile, Colombia, Indonesia and Senegal, among other countries, have successfully started with coordina- tion, planning and monitoring of tobacco laws’ implementation through periodic com- mittee meetings. The committees usually include representatives from the ministries of health, finance, tax and customs, police, transport and, in some cases, education. Coordination across borders Effective approaches to control smuggling in tobacco products require interventions at the borders of jurisdictions and therefore must involve the border agencies. However, with the globalization of trade, there is a need for close coordination not only between tax and border control authorities but also between different jurisdictions. Recent cases have demonstrated that an absence of formal cooperation frameworks may expose a market to financial crime, including money-laundering and financing of terrorism (10). 3 See, for example, Focus on tax fraud. Customs administration of the Netherlands, tax and customs administration. 2017;2 (https://customsnl-insight.nl/article/309563676, accessed 3 October 2020). 98 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Accession to international cooperation agreements such as the Protocol, the Organisation for Economic Co-operation and Development (OECD) Multilateral Convention on Mutual Administrative Assistance in Tax Matters and other regional arrangements will contribute greatly to the effective exchange of information and cooperation among enforcement agencies. An effective exchange of market data and information from participating jurisdictions can prevent potential cross-border crimes and loss of domestic revenue. International cooperation reinforces domestic measures to stop illicit trade and raise much-needed revenues. Parties to the Protocol have a commitment to cooperate with one another and to share information to meet their obligations under the Protocol (Article 20). The Protocol itself is the legal instrument that allows Parties to cooperate and share information across borders. Authorities of governments that are not Parties to the Protocol or another coopera- tion agreement that represents a legal instrument to exchange information could conclude a mutual assistance agreement or exchange of information agreement to guide the procedures under which information exchange can take place effectively. The Revised Kyoto Convention of 2010 promulgated by the World Customs Organization (WCO) recommends that jurisdictions that enter into bilateral agree- ments require the other jurisdiction to provide pre-arrival information on goods bound for their customs territory. A survey of 87 WCO members in 2013 found that the vast majority of customs administrations had the legal authority to share information related to the supply chain of tobacco products with other administrations (11). Some economic blocs have also established harmonized legislation applying to all of their Member States to provide administrative cooperation to efficiently cooperate on tax matters (12). Coordination can include the establishment of a special agency to ensure the safety and proper functioning of external borders, such as the European Border and Coast Guard Agency, also known as Frontex (from the French frontières extérieures, “external borders”). In some of the Frontex-led operations, EU and non-EU countries cooperate together with international organizations to target cross-border crime, including the smuggling of cigarettes and raw tobacco (13). Criminals who engage in illicit trade of tobacco products are usually also en- gaged in related criminal activities such as bribery, money laundering, corruption, obstruction of justice and even financing of terrorist organizations (14). A number of international treaties provide the legal framework for addressing such conduct through mechanisms that tackle illicit trade from a criminal justice perspective, such as the United Nations Convention against Transnational Organized Crime, the United Nations Convention against Corruption and the International Convention for the Suppression of the Financing of Terrorism. Table 3.1 summarizes the types of structures available for such coordination. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 99 Table 3.1 Structures for coordinating mechanisms TYPE OF COORDINATION BASIS INVOLVED ACTORS National coordination Agreements with a basis in law between national agencies Customs authorities, ministries of finance and those responsible for formulating, analysing and implementing tax policy; law enforcement agencies, such as police and border control forces; and anti-fraud teams Agreements between ministries or a basis in law or regulation on the establishment of high- level committees Ministry of health, finance, revenue, justice, transport and sometimes education and enforcement entities such as customs and police Bilateral coordination Bilateral cooperation agreements National governments Regional coordination Regional arrangements such as • Harmonized legislation applying to all Member States of an economic bloc to provide administrative cooperation in taxation to efficiently cooperate on tax matters • Regulation to jointly establish a special agency to ensure the safety and functioning of external borders EU Member States, the European Border and Coast Guard Agency (Frontex), customs, law and border enforcement agencies International coordination International treaties or conventions such as • The Protocol • OECD multilateral Convention on Mutual Administrative Assistance in Tax Matters • United Nations Convention against Transnational Organized Crime • United Nations Convention against Corruption • International Convention for the Suppression of the Financing of Terrorism Parties to international treaties and conventions, law and border enforcement agencies KEY TAKEAWAY 2 Regardless of differing institutional arrangements, coordination and cooperation within a country and across jurisdictions are essential to optimize tax collection and enforcement of tax policy. 100 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 3.2.3. EVALUATION OF PERFORMANCE AND ACCOUNTABILITY Key strategic indicators are useful for assessing the performance of a competent authority. Performance indicators can include measures such as net revenue col- lected, total expenditures compared with budgeted amounts, the ratio of costs to collection, measures of filing and payment compliance and taxpayer satisfaction (15). Several international organizations, including the IMF, the World Bank, the Inter-American Development Bank and OECD have developed tools to evaluate tax and customs with key performance indicators. This section provides information on some of the indicators that are particularly useful for measuring performance related to tobacco taxes, including the cost of collection ratio, tax gap analysis and tax revenue targets. Cost of collection ratio Collection costs vary among countries. The cost of collection ratio is the total ex- penditure as a percentage of the total net taxes collected. This ratio is often used as a measure of efficiency and effectiveness of competent authorities. In Table 3.2, the cost of collection ratio is calculated for country groups by income level, based on an annual IMF survey. The numbers in the table give an indication of resources used and revenues collected for taxes in general. The same definition of cost of collection was used for all countries. The tax revenue excludes VAT and excise taxes on imported products, so it reflects internal taxes only: personal and corporate income taxes, VAT and excise on domestic production. Customs duties are also not included. The results show the differences among countries at various income levels. Other contributing factors include differences between tax systems, economic situations and compliance levels. Table 3.2 Cost of collection ratio in 2015 per 100 units (ratio of average recurrent budget to revenue collecteda) GROUP (SAMPLE SIZE) 2015 Low-income countries (6) 1.3 Lower-middle-income countries (15) 1.6 Upper-middle income countries (18) 0.9 High-income countries (36) 0.9 All (76) 1.1 a Does not include VAT or excise on imports Source: (Reference 8, Appendix Table 12). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 101 As one would expect, given lower levels of automation and resources, the ratio is higher for low- and lower-middle-income countries, greater than 1.0 (more than 1.0 currency unit needed to collect 100 currency units). The ratios for upper-middle income and high-income countries are below 1.0, indicating more efficient and/or effective collection systems. The cost of collection might be less relevant for taxes that are introduced with other than solely financial objectives, such as influencing a change in behaviour. In particular, in the case of excise taxes applied on tobacco products, the cost of tax collection does not reveal the full picture. If excise tax rates are increased substan- tially – or at least increased above inflation and income growth – consumption will be reduced. As a result, health care costs will be reduced due to reduced tobacco- related mortality and morbidity and increased productivity. These savings are not factored into the ratio of cost of collection to revenue, but governments do benefit from these lower expenses overall. Nevertheless, the cost of collection can be used as an indicator of the efficiency of a competent authority. Tax gap analysis Tax gap analysis is another method of determining how effectively taxes on tobacco products are collected. The tax gap is the difference between the tax due and the tax that is collected. For example, the theoretical tax due under an ad valorem tax on the retail price of cigarettes would be the average price of a pack of cigarettes multiplied by the number of packs sold (estimated from household expenditure surveys, for example) multiplied by the tax rate. This outcome can then be compared to the actual revenues collected (16). The effectiveness of tax collection can also be determined by using the macro- economic input-output matrix, measuring the added value of the economic sector – tobacco in this case – and the theoretical VAT due and then comparing the result with the real VAT collection. This methodology is valid for measuring domestic tax evasion (more information on the use of this method to estimate illicit trade is provided in Chapter 4, section 4.1). Tax revenue target The performance of a competent authority can also be evaluated by determining whether the tax revenue target has been met, if mandated, for a given tax period. Although revenue forecasts are often used as targets, caution is advised. Forecast revenues could include assumptions such as economic growth, inflation and amount collected. Forecasting is a good practice, however, and competent authorities should provide input to the government for the forecasting of revenues to improve the quality of the estimates. Competent authorities should monitor the actual collections 102 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N in comparison with the forecasted revenues, but the theoretical base may not be attainable for a variety of reasons. In addition, a revenue target could provide an incentive for some customs and competent authorities to simply aim to reach the target amount, rather than making efforts to collect the maximum amount possible with the available resources. 3.3 THE TAX COMPLIANCE CYCLE For any tax, there are associated compliance, control and enforcement processes. The compliance cycle usually includes registration and licensing, tax declarations, recordkeeping, storage in warehouses, duty suspension, collection of tax and tax refunds. Figure 3.1 illustrates the typical stages of the tax compliance cycle. Fig. 3.1 Tax compliance cycle 3.3.1 REGISTRATION AND LICENSING Along with regulating and ensuring the integrity of those who deal with controlled substances or goods, the main objective of licensing is to regulate the supply chain. Licensing is a powerful tool for obtaining more information and securing the supply chain of tobacco products. Parties to the Protocol are committed to licensing the manufacturing, import and export of tobacco products and manufacturing equipment (Article 6). In addition, Parties are committed to endeavouring to license – as considered appropriate – the persons involved in the growing of tobacco and the retailing, transporting, wholesal- ing, brokering, warehousing and distribution of tobacco products or manufacturing Registration and licensing (renewal) Tax declaration Authorities: audit and control Taxpayers: recordkeeping Audit and control of information provided in tax declaration Payment and collection of tax Refund Licensed activities, for example: import, production, transport, storage, export, etc. 1 2 3 456 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 103 equipment (Article 6). To ensure an effective licensing system, Parties shall monitor and collect, where applicable, any licence fees that may be levied and consider using them in effective administration and enforcement of the licensing system, for public health or for any other related activity in accordance with national law. If feasible, each Party shall require that retailers and tobacco growers – except for traditional growers working on a noncommercial basis – maintain complete and accurate records of all relevant transactions in which they engage, in accordance with its national law (Article 9.4). Article 6.3(b) of the Protocol provides a list of information to be requested from the applicant of the licence, including: • relevant identity information on the applicant • business location of the manufacturing unit or warehouse and production capacity • detailed list of tobacco products and equipment used • description of where the manufacturing equipment will be installed and used • documentation or declaration of any criminal records • information on bank accounts to be used for transactions and payments • description of intended use and intended market of sale of the tobacco products. To make it easier for authorities to collect all the information they need, rules of confidentiality could be exempted in the licensing process. Licences can be general – covering all activities requiring a licence – or issued for each activity separately, such as different licences for manufacturing, importing and retail. A general licence is less burdensome for the licensing authority, whereas licences for each type of activity offer greater control but at the cost of more adminis- tration (17). The cost of implementing the licensing system should be proportionate to the potential impact of the system. Not only should the type of licences be taken into consideration, the process and information needed to obtain a licence should be carefully considered to ensure proportionality. The more stringent the process is – in terms of the information required and the obligations the system imposes on licensees – the more burdensome the regime will be on both businesses and the authorities who must administer and enforce it. The more information is collected, the higher the compliance and administrative burden will be. It is recommended that the added value of the information be balanced with the additional compliance, administrative and/or enforcement burden. The level of stringency should be decided with consideration of factors such as the level of risk of the activity and the availability of enforcement capacity. A more stringent regime might be justifiable for activities that pose a higher risk for the 104 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government in terms of potential loss of tax revenues – such as the import, production and handling of excisable products on which the excise taxes have not yet been paid. Authorities could consider setting licence fees at a high enough level to cover the costs of administering and enforcing the system. For an example of a system that relies mainly on licensing and permissions, see the case study of Australia in Box 3.1. Wholesalers, distributors and retailers of tobacco products could also be required to obtain a licence before they can engage in the trade of those products. This would enable the competent authority to require reports on, for example, transactions relating to the purchase and sale of tobacco products. Moreover, it would allow the authorities to complete the audit trail of the entire supply chain and to obtain data that will help tax and health policy-makers properly and effectively monitor tobacco products. Governments could also require a licence for entities dealing with raw materials or growing tobacco, including farmers. If licensing of tobacco farmers is deemed appropriate and subsequently required in a country, the farmers have to identify and register their farm areas and location to obtain a licence. The benefit of requiring licences for farmers is that the control of the legitimate supply chain is extended to the identification of the source of the raw material for tobacco products. It also makes it more difficult to divert raw tobacco from the licit to the illicit supply chain. Licences are issued by different agencies across the world. In Brazil, for example, the Health Surveillance Agency is responsible for providing licences. Operators need to obtain approval of the layout of manufacturing and warehousing facilities before they can operate. In addition, they must demonstrate how they will comply with other laws and regulations – for example, by showing the design of product packaging, including the pack, carton and master case. The factory location must be identified before manufacturers can obtain a licence. Finally, a licence is required for the importation of machinery to produce tobacco products (18). Licences can be a source of useful information if authorities establish the informa- tion that applicants must supply in order to obtain the licence. Such information could include the quantity, price and how the tobacco harvests are disposed, as well as the identity of the buyers. It is recommended that an effective licensing regime collect information to establish both the identity and characteristics of applicants by requiring criminal records on relevant offences, such as previous noncompliance with tobacco licences or fraud. To avoid loopholes for monitoring raw tobacco, importers of tobacco leaf could also be licensed or at least required to register and report information on quanti- ties, sources and sales. In some countries, this information is already collected by a government agency other than tax authorities, for example, by the ministry of agriculture. Duplication of requirements and reporting should be avoided through CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 105 legislation and coordination among agencies. If licensing of (small-scale) farmers is difficult to implement, subsequent purchasers (first processors) in the supply chain could be licensed and regulated instead. Licensing first processors is often less burdensome to enforce for competent authorities because, in general, there are far fewer first processors than there are growers. For example, in the EU, between 50 and 100 first processors have been identified, compared with 55 000 farmers (19). Countries could also consider requiring registration of persons or entities engaged in the manufacture and import or sale of materials used for the manufacturing of tobacco products, such as cigarette papers, tobacco leaves, additives, adhesives, acetate or any other type of filters used for cigarettes, tipping paper and cellophane or plastic wraps, as well as materials for packing the cigarettes into packs, reams and master cases. In addition, tobacco manufacturers could be required to obtain a licence before they can purchase these materials. The Parties to the Protocol should decide on appropriate measures, depending on research as to whether key inputs that are essential for manufacturing of tobacco products exist and can be identified and subject to effective controls. KEY TAKEAWAY 3 The objective of licensing is to regulate and secure the supply chain. It is a powerful tool for obtaining information for verification, further investigation and audits. Ideally, all persons involved in the growing of tobacco and retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. Licensing requisites Based on case studies and best practices – including experiences from managing bonded warehouses where the value of merchandise or suspended duties or taxes is high – the following kinds of information could be required to obtain a licence, in particular, for producers, warehouses and distributors of tobacco products: • certification of safety of installations, perimeter security for production and storage (may include CCTV [closed-circuit television]4 access for tax administration) • certification of financial solvency • detailed online, real-time inventory of tobacco products and main raw materi- als, accessible by tax administration • electronic accounting systems 4 The term “closed-circuit television” is used generically to describe surveillance camera systems. 106 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • detailed lists of owners and managers • banking and other financial records • periodic electronic reports of transactions for tobacco products • anytime tax administration right of entry for inventories • mandatory electronic tax returns and payments • mandatory prior-to-arrival customs declarations for tobacco products • declarations of compliance with the tax stamp system (if applicable) • for those involved in import or export, authorized economic operator (AEO) certification • proof of compliance with the bond or guarantee regime • agreement to finance reasonable cost of inspections and tracking and tracing. Box 3.1 Case study of licensing in Australia Australia has taken an approach to controlling tobacco taxes that differs from that in many other countries. It has not used fiscal marks or tracking and tracing.5 Instead, it administers tobacco taxes through licensing and permission-based systems aimed at facilitating operations by lower-risk entities while preventing or tightly controlling commerce involving higher-risk entities. The domestic tax agency, the Australian Taxation Office (ATO), is responsible for most of the controls. These controls cover tobacco that is grown or manufactured and imported as finished goods or as leaf for manufacturing in Australia. In fact, the legal tobacco market in Australia consists only of imported finished tobacco products. In 2006, all tobacco-growing licences were cancelled by the ATO because manufacturers switched to cheaper leaf from external suppliers. The last domestic cigarette manufacturers closed in 2015 and 2016, and there has been no legal domestic tobacco growing or manufacture since then. The ATO administers all other functions relating to the import of tobacco and tobacco products, including licensing of bonded warehouses used to store imported products and issuance of permissions to undertake movement of bonded tobacco products between licensed bonded warehouses or to places of export. Importers must apply for a licence for a bonded warehouse to store imported tobac- co. The applicant must meet general criteria such as fitness, recordkeeping and security.6 These criteria are designed to ensure that only low-risk entities are able to enter the ex- cise tax system. Risk levels are also kept at an acceptable level through provisions allow- ing the suspension or cancellation of licences, subject to appeal. Licences are valid for a 5 See sections 3.4.4 and 3.4.5 for detailed discussions on fiscal marks and tracking and tracing. 6 The entity must not have been charged with an offence under the Excise Act or any Commonwealth, State or Territory Act that carries a penalty in excess of US$ 105 000 in the previous 12 months (or convicted in the previous 10 years), has shown a history of compliance with tax law in the previous four years, has had no previous cancellation of a licence, has adequate financial resources and is not in receivership. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 107 three-year period, with automatic renewal for licensees with demonstrated compliance. The permission system relies on post-transaction audits of commercial records. Criteria used to assess risk include the size of the duty liability, the compliance record of both parties and the possibility of diversion into the market. When there is a perceived risk of revenue loss, the application can be denied or a financial security deposit can be required. Exports of tobacco products are also subject to an export declaration process with the Australian Border Force. An approved export declaration is required for the products to be able to leave the country. Following recommendations from a government task force in 2017, the status of tax-suspended, bonded tobacco was eliminated as of 1 July 2019. In addition, an import licensing regime was introduced, and commercial tobacco imports without a licence are banned. Importers are required to identify their duty liabilities at import and make immediate payment; there are no credit terms available. Full payment of duties and taxes to the Australian Border Force are required prior to a release of tobacco products into the country. Sources: (20–21). As mentioned above, certification as an AEO could be requested as part of the licensing process. Most customs authorities are familiar with the concept of AEOs. Created by the WCO, AEO principles were initially focused on security concerns (22). Having a special licensing regime for operators of the tobacco supply chain is recom- mended due to the special nature of the product. For countries that have no system in place, AEO certification could be a starting point for setting up such a regime. An AEO is defined by the WCO SAFE Framework of Standards (22) as a party involved in the international movement of goods – in whatever function – that has been approved by, or on behalf of, a national customs administration as complying with WCO or equivalent supply chain security standards. AEOs include, inter alia, manufacturers, importers, exporters, brokers, carriers, consolidators, intermediaries, ports, airports, terminal operators, integrated operators, warehousers and distributors. For many years – in some cases, even since the 1970s – customs administrations have been increasingly involved in the security of the international trade supply chain. More recently, customs administrations have developed security programmes in a global context. The AEO is part of these programmes, and in 2005, the WCO adopted the SAFE Framework of Standards. Since then, a number of traders have been required to make substantial investments in order to obtain AEO status and must continue to invest to maintain that status. The AEO program is also recognized by the Trade Facilitation Agreement, a multilateral agreement signed by 174 countries (23). 108 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Some regional blocs have further specified the standards for AEOs and provide clear and well-structured information on their websites to guide and encourage operators to apply for AEO status. A good example of this practice is the website of the Revenue Commissioners of the Republic of Ireland, which contains the in- formation shown in Box 3.2. Box 3.2 AEO: Republic of Ireland Tax and Customs What are AEOs? AEO status is a certified standard authorization issued by customs administrations in the European Union (EU). It certifies that an economic operator has met certain standards in relation to: • safety and security • systems to manage commercial records • compliance with customs rules • financial solvency • practical standards of competence or professional qualifications. This is primarily a trade facilitation measure that recognizes reliable operators and encourages best practices in the international supply chain. As an AEO, an operator could benefit from: • recognition worldwide as a safe, secure and compliant business partner in international trade; • lower risk scores in risk analysis systems when profiling; • priority treatment if physical controls are conducted; • mutual recognition of AEO programmes under Joint Customs Cooperation Agreements, which could result in faster movement of goods through third- country borders; • reduced data sets for entry and exit summary declarations (this applies only to AEO safety and security); • easier access to simplified procedures; • reduction or waiver of comprehensive guarantees. The conditions for AEO status apply to all businesses regardless of size. Manufacturers, exporters, freight forwarders, warehouse keepers, clearance agents, carriers and importers may all apply for AEO status. Source: (24). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 109 3.3.2 DATA COLLECTION, DECLARATIONS AND ACCOUNTING The effectiveness of risk analysis depends on the quality and reliability of the available data. This is also the case for risk analysis in relation to tobacco taxes. Obtaining reliable data can be a challenge in many countries, but the use of electronic sys- tems to collect and manage data is increasing in most competent authorities. The introduction of VAT in many countries around the world has greatly improved the availability of data that can be used for tobacco tax analysis, since reporting is done along the supply chain on, for example, the value, quantity of goods and transaction date. Most countries applying excise duties also have a VAT system in place. In addition, more countries are becoming Parties to the Protocol. With the implementation of the Protocol, more data will become available because countries will be obliged to implement, among other measures, licensing systems with report- ing requirements and tracking and tracing systems. More information on tracking and tracing systems is provided in section 3.4. The obligations of the Protocol will also assist in monitoring the stock of tobacco products. Ideally, all entities involved in the tobacco product supply and distribution chains should be licensed and required to record every transaction that occurs. As this might be burdensome for both tax authorities and taxpayers, the use of automated and electronic systems is recommended in order to decrease the costs of compliance. An accurate inventory system for all raw materials, machinery, goods in process and finished products can be required. It is even more important to have good recordkeeping of the required data. As the volume of reported data increases, a good information technology (IT) system will be needed. The use of IT for periodic tax declarations, accounting, inventory and financial data is critical for obtaining accurate information and decreasing costs for the entire reporting system. Most countries now have some level of automation that can facilitate data analysis. An emerging trend is the use electronic invoices, issued by traders, as part of online real-time information for tax administration. Countries generally start by using electronic invoices at public utility companies and then later expand the use to large companies. Electronic invoices minimize the use of paper, contribute to automated recordkeeping and give accurate and timely information about transactions for tax administration. Several countries began using electronic invoices for companies on a voluntary basis and later made their use mandatory, especially for large companies with a high number of transactions – including the tobacco industry. Electronic invoices have been implemented successfully in EU countries and almost all Latin American countries, as well as several Asian countries.7 7 Electronic Invoicing in Latin America: English Summary of the Spanish Document; Inter-American Development Bank, Inter-American Center of Tax Administrations, 2018 (https://publications.iadb.org/ publications/english/document/Electronic-Invoicing-in-Latin-America.pdf ). 110 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N To verify that information is accurate, competent authorities could systematically cross-check declared information against third-party information (e.g. from banks, financial institutions, employers) or match the data with the information in registers of other government agencies. Processes of cross-checking and data matching could also be automated to minimize the administrative burden (25). KEY TAKEAWAY 4 Reliable data are essential for effective risk analysis. While obtaining these data can be challenging, electronic systems can help reduce the burden by automating procedures of data collection and cross-checking of information with different sources. 3.3.3 RECORDKEEPING Parties to the Protocol are committed to requiring, as appropriate, that all persons or entities engaged in the supply chain of tobacco, tobacco products and manu- facturing equipment keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of tobacco products (Article 9). Relevant information includes market volumes, trends, forecasts of tobacco products and quantities of tobacco products and manufacturing equipment kept in stock in tax and customs warehouses in transit, transhipment and under duty suspension. This information should be required from the persons and entities engaged in the supply chain and submitted to the competent authority on a regular basis, as provided for in the law. The competent authority can use the submitted information to monitor compliance with tobacco regulations and payment of taxes. A registry with this level of detail can realistically be kept only in electronic form. Records must provide full accountability for materials used in the production of tobacco products. The intention is that tax authorities and manufacturers should be able to reconcile the production quantities with the inputs used in production – thereby providing confidence that no unrecorded or illicit production has occurred. Obligations should also be imposed on suppliers of key inputs to show that supply is commensurate with demand (17). KEY TAKEAWAY 5 To monitor compliance and payment of taxes, all persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions, as well as details of materials used in the production of tobacco products. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 111 3.3.4 WAREHOUSING, STORAGE AND DISTRIBUTION According to Article 6.2 of the Protocol, all Parties shall endeavour to license persons involved in any wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Maintaining a system of authorization allows the authorities to carry out controls in production and storage facilities to ensure that taxes are paid (2). The approval process to obtain an authorization could include an evaluation of the layout of the plant or warehouse, the machinery that will be used and the flow of production, warehousing and shipping, including the points of entry and exit of raw materials and finished products. The basic method of monitoring production and ensuring that only tax-paid products are released to the market from the premises is to identify the production facilities and to control the entry and exit points. From time to time, the competent authority should conduct a physical inventory of the goods contained therein to check whether all documentation was duly prepared and approved and to determine the accuracy and completeness of the records kept. If the jurisdiction requires tax stamps to be placed on the tobacco products, only products with the proper stamps affixed can be withdrawn. Generally, tobacco products for which the required taxes have not been paid and, if required, fiscal marks have not been affixed should not be allowed into warehouses. For practical reasons, many countries allow suspension of excise duties, meaning that prior authorized persons can produce, send, receive and store tobacco products on which the excise duty has not yet been paid. The relevant authorities could also require that products on which the taxes have been paid should not be stored in the same areas as the products under duty suspension. Obviously, products under suspension of payment of excise duties are at high risk, which could justify stricter requirements for production, trade, storage and handling. Australia, which has a strict system of licensing and requirements for permission to move tobacco products, has migrated to a new system that eliminates bonded warehouses from the supply chain as of 1 July 2019. Importers are required to have an import licence and to pay excise taxes on cigarettes immediately upon import (see Box 3.1 in section 3.3.1). 3.3.5 DUTY SUSPENSION Many countries require authorization of natural or legal persons (as authorized warehouse keepers) to produce, process, hold, receive and dispatch products sub- ject to excise duty during their business. Producing, processing, holding, receiving and dispatching excise goods often take place under suspension of the excise duty. Guarantees can be requested from authorized persons to secure the payment of taxes. Features of such a system may include strict criteria for granting authorization, 112 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N warehouse pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. The use of a computerized system for monitoring movements of excise goods under suspension of excise duty can be a control as well. Different licences for products under duty suspension could also be considered. This would make enforcement easier and less burdensome for both authorities and operators. In general, it is recommended to allow the handling of excise goods under suspension of duties only if strict criteria are met. Such criteria could include pre-authorization visits, adequate stock control measures, checking the origin of excise products and the entire production process and coding and marking products. In principle, the movements of tobacco products should also be covered by the tracking and tracing system. Considering the high risk related to these products, additional monitoring could be considered appropriate, such as a computerized system monitoring the movements of excise goods under suspension of excise duty. In the design of such a system, it is recommended that close attention be paid to customs procedures for import and export to ensure alignment and avoid a vacuum in monitoring. An example of a computerized system is the EU’s Excise Movement and Control System, which follows the movement of all excise products – including manufactured tobacco products – for which excise taxes have not been paid. The system records the movement in real time and is thereby an important tool for combatting fraud. In addition, this system is indispensable for the exchange of information and co- operation between the relevant authorities of Member States of the EU (26). Finally, authorization is required before tobacco products can be produced, imported or stored under suspension of excise duties (27). KEY TAKEAWAY 6 Products under duty suspension of excise taxes are at a higher risk of tax evasion, which can justify stringent measures such as requesting guarantees to ensure the payment of taxes, additional licensing requirements, compliance with computerized systems to monitor the movement of excise goods under suspension and on-site authorization and audits. 3.3.6 COLLECTION OF TAXES To reduce the complexity of tax collection systems, it is recommended that excise taxes be imposed at the point of manufacture, import or release from storage or production warehouses for consumption. This is common practice in the majority of countries that impose excise taxes. Collecting taxes at this level of the supply chain CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 113 greatly limits the number of taxpayers and thus the resources needed to control them. Encouraging taxpayers to use electronic payment methods can also increase the chances of collecting all taxes. The same applies to requiring guarantees for certain high-risk activities, such as the handling of goods under duty suspension. Many countries decide on a case- by-case basis the level of the guarantee, depending on the situation of the requestor and the level of risk (quantity or value and potentially due excise taxes) that the regular business activities represent in a given time frame. Some countries allow a reduction of guarantees for operators with a track record of good compliance. It should be noted that a guarantee is not a limitation of the liability; taxpayers can still be requested and liable to pay an amount far above the level of the guarantee. Tax payments should be required by law to be remitted at fixed intervals after sales or on a fixed date each month (2). Many countries have a specialized collection enforcement unit that works full-time on the collection of taxes. It is important to have a stop-filer or payment control that can act immediately when noncompliance occurs, by sending a message and phone call of late declaration or late payment to the taxpayer. This increases the likelihood of keeping taxpayers compliant. If nondeclaration or nonpayment persists, the bond or guarantee could be executed. Another reason for collecting excise taxes around the time of production or import is that quantities can be monitored more effectively at these points. There are different options for monitoring the supply chain of tobacco products. The decision about what kind of monitoring system to use depends on the country’s financial, technical and human resources. The weakest form of monitoring is in- dustry self-declaration. Activities to verify compliance and ensure the collection of the full amount of taxes due can include, for example, physical checks, audits, cross-checking of declared information with third-party data and inspection of administration and recordkeeping. In general, in countries with poor administration systems, enforced compliance is carried out by imposing physical control over the production or manufacturing process. The cost of physical control increases when there is a potential for fraud by excise officers. However, fraud can be diminished significantly when excise officers are rotated frequently among different locations and supervisors make surprise visits. Historically, some countries (e.g. India) have posted tax administration staff at production facilities to monitor production and removals. In India, a staff member of the competent authority is placed in cigarette and large bidi manufacturing facili- ties around the clock. Each officer records the daily production and the quantity of cigarettes/bidis that leaves the factory and reports to the next officer.8 8 Ministry of Finance India, personal communication, 2009. 114 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A better option is to monitor production remotely. The competent authority can require the installation of CCTV cameras in strategic places throughout the manufacturing and warehousing facilities. With these, the authority can establish a central command post from which the facilities and activities can be continu- ously monitored and documented. In addition, the competent authority can carry out physical inventory controls from time to time and – if electronic invoices are implemented – cross-checking between invoices and declared inventory. This is also an effective way to prevent collusion between staff of a competent authority and manufacturers or importers. For example, in 2015, the Bureau of Internal Revenue of the Philippines required all tobacco companies to install CCTV cameras in their production lines and warehouses. This decision was taken in response to large seizures of untaxed cigarettes, with the objective of monitoring production to ensure the payment of all taxes. The collection process must also be supported by IT systems. These systems must provide for transparency and accuracy to ensure a safe process for the flow of payments from taxpayers to the tax treasury. Most countries have implemented automated electronic systems for tax payments linked to each declaration, for both domestic and import. It is key for tax administrations to have a comprehensive agreement with the banking system in order to obtain lower transaction costs, if applicable. Some countries have implemented a state payment web portal that allows citizens to pay their taxes and other fees such as county fees, fees for car permits and licences and agricultural, health and environmental fees online. KEY TAKEAWAY 7 Excise taxes should be imposed at the point of manufacture, import or release from storage or production warehouses for consumption, to ensure that quantities can be monitored effectively. This also reduces the complexity of tax collection systems by limiting the number of taxpayers and thus the resources needed to control them. 3.3.7 TAX REFUNDS Refunds for VAT, excise taxes and customs duties are a common process in most countries, under the principle that consumption taxes are not exported. Frequency and methods of refund vary by country. It is common to have monthly refunds (if there are exports during the period), and the reimbursements may be sent directly to the exporter or reserved as a credit to pay other taxes. An alternative used by some countries that have a high volume of exports is a so-called zero rate, or suspension, meaning that indirect taxes (VAT, excise taxes and customs duties) are suspended for the whole chain – from import of raw materials to production and packing until CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 115 export. This regime requires a special licensing process. Since the tobacco industry has an export component, the refund process for this sector requires special atten- tion for tax administration. 3.4 CONTROL AND ENFORCEMENT Control and enforcement are the main functions of tax administration. In fact, most tax laws include the objective “to control and enforce tax compliance” and, for customs, “to control and enforce tax and duty payments at the border” or similar phrases. The Protocol provides guidance for control and enforcement of tobacco taxes. Efficient and effective competent authorities often have a strategic plan to ensure compliance, a risk-based approach to identify the problematic points in the chain and the ability to direct resources accordingly to high-risk or high-value areas. Tasks that can play a role in control and enforcement include controlling the registration and licensing process, due diligence, verifying declarations and collec- tion of taxes. Production and distribution controls including tracking and tracing, fiscal markings, audits and import and export controls all play a role in control and enforcement. This section describes the main activities for improving control and enforcement, focusing on the tobacco supply chain. The procedures and penalties that can be enacted once illicit trade in tobacco has been detected are also discussed. 3.4.1 CONTROL AND ENFORCEMENT PLANNING Strategic plan In modern tax administrations, it is common to have a strategic plan, with control and enforcement as pillars. Appropriate control of the compliance cycle is key to keeping taxpayers in compliance and preventing illicit trade and tax avoidance. For this reason, most tax administrations focus a majority of their resources on preventive policy. Some examples of this can be found in the strategic plans of the United Kingdom and the United States’ Internal Revenue Service (IRS). In the United Kingdom, HMRC has had a well-developed strategic plan for years. A key pillar of the plan focuses on keeping taxpayers compliant. This is the concept of prevention: controlling initial minor noncompliant behaviour for the majority of taxpayers, while using strong enforcement for the minority on the noncompliant side (28). The strategic plan of the IRS has a similar approach, with a focus on control. If noncompliance is detected, data analysis and behavioural insights are used to identify the best way to address noncompliance. Early intervention or self-correction are examples of ways to address detected noncompliance. The IRS also highlights the importance of resolving noncompliance to ensure taxpayer confidence in the tax system and protecting the integrity of the system (29). 116 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Risk-based approach Following the establishment of a strategic plan, an enforcement and control plan must be drafted. This plan should include definitions of the activities that will be enforced, the taxpayers upon whom they will be enforced and the circumstances under which they will be enforced, as well as allocating resources for staffing, audit- ing, infrastructure and IT. Targets must also be defined, including the number of interventions and the amount of additional collected revenue or reduction of tax evasion. Several tax administrations elaborate annual plans with periodic perfor- mance reviews aimed at improving results, and they correct allocations and targets as needed. Clear targeting of interventions is needed for better results, more efficient use of resources, lower costs for taxpayers and more effective collection. In other words, the point is to focus interventions on those who have a higher probability of noncompliance. Using a risk-based approach can be particularly beneficial. Tax risk management is a key element of control strategy in modern tax admin- istration. A risk is a possible threat to reaching objectives such as collecting taxes in an effective and efficient way for competent authorities. Risk assessment is the process of analysing risks and deciding on the best way to manage an identified risk. The responses can vary from acceptance to mitigation to avoidance. Proper risk assessment allows competent authorities to use their available resources most efficiently and to become more effective in dealing with risks. It can be used to improve compliance by identifying taxpayers or types of activities with a high risk of noncom- pliance. Groups of taxpayers with the same characteristics often have similar risks. Groups with a high risk of noncompliance could then be subject to greater review. Areas of potentially greater risk of noncompliance in the tobacco supply chain include import, export and transfers to and from warehouses, particularly when they take place under duty suspension. Gathering risk-related information from internal and external sources is a best practice in compliance risk management. Such sources could include third-party information (e.g. from banks, credit card companies, transport companies), studies on taxpayer behaviour and research on compliance issues, tax gap analysis, tax audits and declarations (30). Risk management uses these different sources of data along with algorithms to find patterns of high noncompliance. Risk analysis can indicate reduced risk as well. Lower-risk areas are likely to need less governance to ensure compliance, which allows for resources to be directed elsewhere. Risk assessment can therefore help with strategic allocation of limited resources to the areas of greatest risk while at the same time reducing the burden on lower-risk taxpayers. Risk management has always been done by competent authorities, but data availability and statistical methods to identify patterns have changed the way risks CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 117 can be assessed. Although many risk assessment systems are still done manually or include manual elements, the use of intensive data techniques allows systematic, deeper and more targeted analysis (31). Modern risk assessment makes use of electronic data on taxpayers, tax payments, declarations from other taxes, such as VAT, and third-party information. With these data, tax authorities can identify indicators that suggest where further activities might be required to ensure compliance. For example, VAT invoices can be used to match reported purchases of inputs of tobacco leaf to sales invoices of tobacco leaf wholesalers. For taxpayers (i.e. those who are licensed and provide required reports), competent authorities can create a business analytics program to determine whether the data reported are consistent on each side of the transactions. Moreover, in countries with a VAT system, competent authorities can compare the data reported by taxpayers under the VAT system with data reported under the tobacco excise tax system to detect any inconsistency. VAT invoices can also be used to verify inputs and sales data. If VAT is collected at all levels of the supply chain, it is easier for govern- ments to monitor the supply chain for the enforcement of excise duty obligations. Regular surveys on tobacco consumption that use the same methodology can also provide indications about the level of compliance with excise tax policy. A sudden drop in revenue that is not reflected in consumption data could be an indication of illicit manufacturing, illegal imports, cross-border shopping or forestalling. In addition, seizure data can provide valuable information on areas and activities at high risk of noncompliance. The structure of tobacco tax policies should also be taken into account when con- ducting risk analyses. If excise tax rates are increased, there might be a greater risk of forestalling or front-loading (see the discussion on anti-forestalling later in this section). Differentiated excise tax rates based on product or packaging characteristics – such as distinctions between soft and hard packs or filter and nonfilter cigarettes – are also prone to manipulation by operators, which could affect tax revenues. One of the options to mitigate these risks is to amend the excise tax policy and apply a uniform tax rate. For customs transactions, the use of risk management is a key element in target- ing merchandise and support declarations to be inspected. Before the 1990s, most customs agencies used random criteria for selecting targets for inspection. Since that time, many countries have implemented risk-based approaches for selecting inspections. Historical data on importers and trade communities, complemented by artificial intelligence technology, show that risk management tools dramatically increased the effectiveness of physical inspections. Most modern customs agencies have implemented such techniques, allowing for more effective control processes while facilitating smoother processes for those transactions that are in compliance. Box 3.3 details some of the recent changes in risk management processes. 118 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.3 Changes in risk management The OECD developed Fig. 3.2 to show the framework and key steps for understanding compliance risks in 2004 (32). The same approach is still used to identify, assess and prioritize risk. However, many competent authorities now use new technologies and advanced data analytics, along with more information sources, including external data from banks, employers and sales invoices for VAT, for example. Fig. 3.2 Compliance risk management process Source: (32). The methods of identifying risks and the analysis of compliance behaviour have also changed. Traditionally, competent authorities used audits to identify high-risk cases. With more diverse and better data, competent authorities can now use more evidence-based approaches to examine risk patterns. Success of compliance activi- ties is now more often measured in terms of their impact on the overall compliance environment, rather than only on increased revenues.9 9 For more information on effective risk management with several indicators and a checklist of questions, see the Tax Administration Diagnostic Assessment Tool (68). OPERATING CONTEXT Assess and prioritize risks Evaluate compliance outcomes: - Registration - Filing - Reporting - Payment Monitor performance against plan Analyse compliance behaviour (causes, options for treatment) Determine treatment strategies Identify risks Plan and implement strategies CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 119 In Indonesia, use of the compliance risk management process reduced the share of illicit trade in total consumption of cigarettes from 12% to 3%. More information on this can be found in the case study of Indonesia in Box 3.13 later in this chapter. Understanding the products – as well as the supply and distribution chains – al- lows competent authorities to identify which areas along a chain pose the greatest risk and therefore require more resources. Detailed information on the composition of selected tobacco products is given in Annex 3.1. KEY TAKEAWAY 8 Risk analysis helps identify the points of intervention that have higher probabilities of noncompliance. A risk-based approach with targeted interventions allows for better results and more efficient use of resources to ensure effectiveness of tax collection. 3.4.2 CONTROLS OVER THE TOBACCO SUPPLY CHAIN As defined in Article 1 of the Protocol, the supply chain covers the manufacture of tobacco products and manufacturing equipment – as well as their import or export – and may be extended, where relevant, to one or more of the following activities when so decided by a Party: 1. retailing of tobacco products 2. growing of tobacco, with the exception of traditional small-scale growers, farmers and producers 3. transporting of commercial quantities of tobacco products or manufacturing equipment 4. wholesaling, brokering, warehousing or distribution of tobacco and tobacco products or manufacturing equipment. Article 4.1 of the Protocol requires parties to “adopt and implement effective mea- sures to control or regulate the supply chain of tobacco products to prevent, deter, detect, investigate and prosecute illicit trade in such goods and to cooperate with one another to this end”. Concrete measures to regulate the supply chain, as well as best practices in this regard, are discussed further below. Figure 3.3 shows the main places for reporting and monitoring along the supply chain: import, ex-factory and removals from warehouses. Manufacturers could be required to report imported inputs at the border, as importers of finished products do. If components are subject to licensing, information can be required as part of the licensing process. The arrows in Fig. 3.3 represent transporting, which is also part of the supply chain. 120 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 3.3 Cigarette supply chain from manufacture or import to retail sale 3.4.3 LICENSING10 AND DUE DILIGENCE A licensing system is effective only if it is properly controlled. Most tax administra- tions have experience with licensing processes for excise taxes on products such as alcoholic beverages and energy products. It is strongly recommended that lessons learned with the licensing process of such products be applied when implementing and enforcing tobacco-related licensing. Licensing provides timely and accurate data that can serve as the basis for audits because it identifies and controls legitimate operators. For new operators, the process to obtain a licence could include visits and verification of production factories, storage facilities and distribution premises. Countries that have no licensing system in place and would like to start applying licences could allow a transitional period for existing operators to comply with the new licensing requirements. The process of licensing control must be carried out and updated periodically, in particular by controlling the validity of bonds or guarantees, the proper functioning of the required systems (CCTV, for example) and recordkeeping. 10 Licensing is discussed here in the context of due diligence and enforcement. Details about how licensing can be set up and what information could be requested are presented in section 3.3.1. Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blend d. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: wher he tobac o will be roll d into st cks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 Cigarette supply chain from manufacture or import to retail sale PRIMARY MANUFACTURING: locally grown or imported leaves are shredded, cured and blended. Distributors/wholesalers/warehouses Retailers IMPORTED CIGARETTES: from cigarette importers Cigarette production: where the tobacco will be rolled into tobacco sticks. Packing of cigarettes: • Cigarette sticks are put into packs • Fiscal marking, such as tax stamps, are axed • Packs are wrapped in cellophane • Packs are placed into cartons then master cases 1 2 CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 121 Where licences are required, the law should include a provision specifying that purchases from unlicensed suppliers – or sales to unlicensed purchasers – are not allowed. This means that both suppliers and purchasers would need to verify those with whom they are doing business. This requirement provides enforcement au- thorities with an entry point to enforce the licensing system at both ends. Also, a licensing requirement for manufacturing equipment assists authorities in identifying and prosecuting illegal manufacturing of tobacco products, reducing the burden of proof substantially. In many countries, the presence of manufacturing equipment is not sufficient proof that illegal manufacturing is taking place; the machinery has to be in operation and producing illegal tobacco products when authorities inspect the location. With a licensing requirement, however, the presence of machinery without a licence is sufficient for authorities to act. The validity of licences should be time-limited, requiring renewals or reapplica- tion, to maintain a high level of control. Adherence to the conditions required for a licence should be controlled by the authorities, and penalties for noncompliance – for example, suspension or withdrawal of a licence – should be severe enough to act as a deterrent (33). Regulations for licensing should provide for inspection of the licensee’s products and premises, with penalties for noncompliance, which could include criminal and civil prosecution for serious or repeated offences. As stated in Article 6.3(a) of the Protocol, Parties need to establish or designate a single authority or multiple authorities to issue, renew, suspend, revoke and/or cancel licences. In accordance with Article 7 of the Protocol, persons engaged in the supply chain are required by law to conduct due diligence before and during business relationships. They also must report to the competent authorities any evidence that a customer is engaged in activities in contravention of its obligations arising from the Protocol. This requirement includes customer identification, monitoring of sales to ensure that the quantities are commensurate with demand for such products within the intended market and taking measures to ensure compliance. Knowledge of the demand of a market is indispensable for determining if there is a case of oversupplying. If the supply of tobacco products to a lower-taxing foreign market exceeds the demand, it creates a higher risk that these products will be smuggled back into a higher-taxing country, undermining the objectives and effectiveness of the higher-taxing jurisdiction. In the past, some governments decided to impose a fine on tobacco companies if the quantities supplied were significantly higher than the demand and the risk of being smuggled back into their jurisdiction was judged to be high (34,35).11 11 Excise duty rates applied in all the EU countries can be found on the European Commission’s webpage: https://ec.europa.eu/taxation_customs/business/excise-duties-alcohol-tobacco-energy/ excise-duties-tobacco_en. 122 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 9 Licensing helps to identify and control legitimate operators. The data obtained from licensing can serve as a basis for audits. Licences should be controlled on a regular basis and updated periodically to ensure their validity. 3.4.4 FISCAL MARKINGS (E.G. TAX STAMPS) Fiscal markings are another important tool for controlling and monitoring pro- duction and import of tobacco products. Their use is generally considered to be appropriate for increasing compliance with tax laws. Fiscal markings can also be of help in distinguishing between genuine and illicit tobacco products. Tax stamps or other fiscal markings affixed to packs of cigarettes or tobacco products facilitate the collection of excise taxes, as well as audits and enforcement actions. The pres- ence of fiscal markings enables both the competent authority and the public to monitor whether the taxes on tobacco products were properly paid. It thus assists the competent authority in investigating illicit trade and prosecuting violations. Fiscal markings include tax stamps, enhanced tax stamps (banderols) and digital tax stamps. Examples of fiscal marks are tobacco stamps, tax stamps, excise stamps, tax stickers and banderols. Box 3.4 presents details on the different types and features of tax stamps. Tobacco products for export are often required to be marked that they are for export. Box 3.5 provides useful information regarding the International Organization for Standardization (ISO) standard for excise tax stamps. The terminology “fiscal mark” holds no indication of the characteristics of the mark. A fiscal marking is affixed to each pack of tobacco product. Requiring a standard package size can facilitate the application of the markings (2). Fraudsters can be deterred from attempting to re-use fiscal markings (in particular stamps) by having the marking affixed to each pack of cigarettes (or other tobacco product) before the pack is wrapped with cellophane (36). In most cases, tax stamps are purchased by the producer or importer and applied to each product sold as proof of excise tax payment (33). Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Box 3.4 Types and features of tax stamps Over time, tax stamps and markings have become more sophisticated. In the past, tax stamps were often paper-based and easy to counterfeit. New tax stamps use additional security features to make them more difficult to counterfeit. Authentica- tion solutions against counterfeiting can utilize various security features, including: • overt features – features that can be verified by the naked eye; CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 123 • covert features – features that can be authenticated only by using dedicated and specialized electronic readers; • semi-covert features – features requiring a simple tool that does not involve extensive training; and • forensic features – features that can be identified through laboratory analysis. Tax markings can be either physical or digital: • physical markings – the information is contained in the document or device attached to the package. • digital markings – information is obtained through a link with a database and by decrypting with the tools and keys used for creation of the data. The term “digital tax stamp” sometimes leads to confusion, as some paper-based stamps with digital components are also described as digital tax stamps. Tax markings that are fully digital do not contain information in the document or device attached to the package. It is probably too simplistic to say that digital tax stamps are more secure than paper- based tax stamps. Both types have advantages and weaknesses. For instance, both physical and digital tax stamps can be weak or strong on security features. Neverthe- less, the management, production, sales, transport and monitoring of physical tax stamps require increased attention. Tax stamps have the same value as banknotes and are a possible attraction for theft, loss and fraud. In Belgium, for example, the Court of Audit severely criticized the lack of control of the production and stock management of tax stamps in 2015. The Court of Au- dit concluded that tax stamps issuance should operate under recognized security practices and procedures relative to the security risk associated with the various production, distribution and issuance processes. Moreover, it was noted that new printing technology of digital tax stamps on packs may facilitate stock management and lead to less fraud. In March 2016, Belgium changed its stamps. The printing became an in-house process by the financial federal government department. The new stamps are still printed with a watermark, but they also have a digital component. The change resulted in a cost reduction by standardizing the sizes and optimizing the production process. Some of the more advanced fiscal marking technologies include embedded threads and watermarks; special inks and coatings, such as so-called invisible inks, holograms and foils; and calculated or changeable content. Because of their enhanced security features, these stamps can be more expensive than traditional stamps. In the state of California in the United States, the traditional stamps cost US$ 0.42 per 1 000 stamps. The cost of the first generation of high-tech stamps was 10 times higher, at US$ 4.77 per 1 000 stamps. This price nearly doubled for the second-generation encrypted stamp, to US$ 8.20 per 1 000 stamps. Nevertheless, California collected 124 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N about US$ 450 million of additional tax revenue in the first decade following the implementation of encrypted tax stamps. This additional revenue was far greater than the costs of implementation and enforcement. Other jurisdictions have also revised their tax stamps to incorporate new technolo- gies. The state of Michigan, for instance, replaced heat-applied cigarette tax stamps with digital pressure-applied stamps in 2015. Michigan deployed a tax stamp with several overt and covert security features and a unique quick response (QR) code and serial number. QR codes (machine-readable codes consisting of an array of black and white squares, typically used for storing URLs or other information) can have purposes beyond tracking and tracing. The QR code can be read by consumers with a smartphone or tablet application to access information on smoking-cessation programs, report violations of the state’s youth access policies, connect to a tip line to report noncompliant packs and learn about the harms from illicit tobacco sales and purchases. Enforcement authorities can validate stamps using the smartphone- based eTRACS (Electronic Tax Reporting and Audit Compliance System). As part of the system’s implementation, the Michigan State Police department created teams of enforcement officers in each of the state’s seven districts and the state Department of Treasury created its own enforcement team. Sources: (37–42). Box 3.5 The ISO standard on excise tax stamps The ISO published its excise tax stamp standard (ISO/TC 292/SC) in October 2018. The purpose of the ISO standard is to assist tax and finance authorities in enhancing compliance with excise tax regulations. A tax stamp is defined as a visible tax stamp, label or mark placed on certain types of consumer goods to show that the applicable excise tax has been paid. The ISO standard applies to tax stamps that are physical in nature – not to digital markings, which are directly printed on to packs without a physical component. “Authentication” in this standard refers to the authentication of the tax stamp, not the product on which the tax stamp is affixed. In other words, authentication of a tax stamp on a cigarette pack means that the tax stamp is authentic but does not guarantee that the pack is authentic. In addition, control measures are needed at the time of the application of the stamp to verify the conformity of the tax stamp with the corresponding product. The standard provides guidance on the content, security, issuance and examination of physical tax stamps used to indicate that the required taxes have been paid and that the tax stamp is authentic. The use of stamps to facilitate tracking and tracing within the supply chain is not described. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 125 Specifically, the ISO standard deals with the following issues: • defining the functions of a tax stamp • identifying and consulting with stakeholders • planning the procurement process and selection of suppliers • the design and construction of tax stamps • the overt and covert security features that provide protection of the tax stamp • the finishing and application processes for the tax stamp • security of the tax stamp supply chain • serialization and unique identifier codes for tax stamps • examination of tax stamps • monitoring and assessing tax stamp performance. A stamp may fulfil many functions, but the core business of tax stamps is to ensure and facilitate the collection of revenue. The tax stamp must use a combination of security features. The tax authority should ensure that the tax stamp can be authen- ticated and that counterfeit, altered, tampered or otherwise fraudulent tax stamps can be detected. The standard provides detailed information on the different components of the tax stamp such as the substrate, inks, adhesives, laminate, authentication or security features and the unique identifier that should enable checks on the payments of the required tax. The process of procurement is discussed in detail in the standard. The tax author- ity should ensure that the procurement process is open, transparent and meets the sustainability objectives. The tax authority should set out the goals and requirements to give tendering organizations more leeway in proposing optimum solutions that might be different from those the authority would specify. The standard is not prescriptive; rather, it provides a catalogue of options. It does not, for instance, recommend specific security features, but it does describe the different types of features that are necessary for a tax stamp to be secure. Tax officials still need to make decisions and choose the option that suits them best, but the standard remains recommended reading for those who would like to introduce tax stamp programmes in their jurisdiction. ISO standards are not freely available but can be purchased at the ISO Store (www.ISO.org) or from an ISO national member body. Source: (43). According to Article 8 of the Protocol, each Party shall require that unique, secure and nonremovable identification markings – such as codes or stamps – are affixed to or form part of all unit packets, packages and any outside packaging of cigarettes for the implementation of the tracking and tracing system within a period of five 126 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N years, and of other tobacco products within a period of 10 years, of entry into force of the Protocol for that Party. The Protocol specifies that at least the following information shall form part of the unique marking: • date and location of manufacture • manufacturing facility • product description • where available, the intended market of retail sale. In several countries, QR codes are used as fiscal markings for tobacco and alcohol tax control. Each stamp has a unique identifier code and a QR code. The data stored in the QR code provide the following product information: • manufacturer • production location • stamp order date • tax status and class • brand • intended market • unique identifier (serial number). KEY TAKEAWAY 10 The use of fiscal markings is generally considered to be an appropriate tool for increasing compliance with tax laws. Fiscal markings can also be helpful for distinguishing between genuine and illicit tobacco products. 3.4.5 TRACKING AND TRACING A tracking and tracing system assists authorities in determining the origin of tobacco products – and the point of diversion, if applicable – as well as monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to enable authorities to have information on all transactions through the entire tobacco product supply chain until duties are paid or other obligations are discharged. Traceability is not used only for tobacco products. It is also used to improve the supply chain function, as in the case of parcel services, as well as for product safety reasons, to manage potential product recalls and for regulatory reasons. Tracking is the process that monitors where a product is at all times while also creating a time and location record for all movements. Tracing is the ability to identify the past locations of a product, so that the product’s route can be followed CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 127 back to its origin (44). In other words, traceability is “the ability to trace the history, application or location of an object” (45). A tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it is possible to unambiguously register that product’s movements. Other necessary characteristics include the ability to share the registered move- ment information and to authenticate products. This enables a product’s status to be captured through the supply chain and its history to be identified and verified retrospectively. According to Article 8.4.1 of the Protocol, Parties should require the following information to be available: • the date and location of manufacture • the manufacturing facility • the machine used • the production shift or time of manufacture • the name, invoice, order number and payment records of the first customer not affiliated with the manufacturer • the product description and intended market of retail sale • any warehousing and shipping • the identity of any known subsequent purchaser • the intended shipment route, date, destination, point of departure and consignee. A good tracking and tracing system enables the government to properly monitor the supply chain, improves its ability to ensure collection of the proper duties and taxes, provides it with the ability to authenticate whether the identification marking is genuine and matches the product and improves its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. The following elements are required for an effective tracking and tracing system (46): • A serialized unique identification marking for each package of product. These identifiers are a distinctive combination of numbers, letters or both. They cannot be predictable or used more than once. The representation of the identifier on the package can be human-readable (letters or numbers) or machine-readable (barcodes). Generation of codes and encryption that are part of a tobacco industry patent should be excluded. • A data carrier with the serialized unique identifier and other information such as date and location of manufacture, manufacturing facility, product description and, where available, the intended retail market. This informa- tion should be readable by authorized agencies of any Party to the Protocol. The data carrier should comply with quality standards and be suitable for 128 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N high-speed production lines. Two-dimensional barcodes, for example, meet these standards and are readable with inexpensive equipment. • A link and parent-child relationship (called aggregation) between different packaging units that offers the option to trace a pallet without the need to scan all the packs and master cases of that pallet. • Recordkeeping of all shipping and receiving events along the supply chain. This includes, for example, the departure location and the arrival location, as well as the involved operators. International standards from the ISO are recommended for the capture and exchange of data and events. • The use of international standards for key information that is encoded in the data carrier (5). An example of a unique and internationally recognized identifier for products is a Global Trade Item Number. The following details on information storage and sharing are drawn from various sections of the Protocol. Data and events along the supply chain must be stored in an independent database that is controlled by competent government authorities. At the global level, national and/or regional databases can be interconnected to facilitate international inquiries by competent authorities. Parties to the Protocol agree to establish a global information-sharing focal point located at the Conven- tion Secretariat of the WHO FCTC, accessible to all Parties, enabling them to make enquiries and receive relevant information. Each Party shall ensure that the information recorded under paragraph 5 of Article 8 of the Protocol is accessible to the global information-sharing focal point on request, subject to paragraph 9, through a standard electronic secure interface with its national and/or regional central point. The global information-sharing focal point shall compile a list of the competent authorities of Parties and make the list available to all Parties. The cost of tracking and tracing systems is a concern for many countries, but as indicated in paragraph 14 of Article 8 of the Protocol, jurisdictions may require the tobacco industry to bear any costs associated with putting in place the tracking and tracing system in a country (46). In Brazil, the cost for cigarette manufacturers was US$ 0.0185 per pack (42). In Kenya, the cost for manufacturers was US$ 0.024 per pack (42). Along with considering the characteristics of a tracking and tracing system in selecting a particular one, it is important to avoid conflicts of interest, ensure fair and transparent dealing with suppliers, implement a zero-tolerance policy for corruption or anti-competitive behaviour and ensure compliance. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, which deals with industry interference, and Article 8 of the Protocol. Article 8.13, which states that “each Party shall ensure that its competent authorities, in participating CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 129 in the tracking and tracing regime, interact with the tobacco industry and those representing the interests of the tobacco industry only to the extent strictly necessary in the implementation of this Article”. Box 3.8 provides a cautionary example of a tracking and tracing system that is not compliant with Article 5.3 of the WHO FCTC. While the Protocol contains a great deal of information on the requirements that a tracking and tracing system should meet, questions come up in relation to the implementation of such systems. To achieve the objectives of the Protocol, the Meeting of the Parties (MOP) to the Protocol, as the governing body of the treaty, has the prerogative to establish subsidiary bodies, such as expert groups and working groups. In decision FCTC/MOP1(6), the MOP established a working group for the development and implementation of tracking and tracing systems in accordance with Article 8 of the Protocol, including the global information-sharing focal point (Article 8.1) and unique identification markings for cigarette packets and pack- ages (Article 8.3), to further elaborate on the next steps. The working group will produce a comprehensive report compiling good practices and experiences on the implementation of tracking and tracing systems, as well as unique identification markings for cigarette packets and packages at national or regional levels. The working group was also given a mandate to prepare a conceptual analysis of how a global information-sharing focal point could be set up. Implementing a complete tracking and tracing system with fiscal markings takes time. In most of the countries that have already implemented tracking and tracing, it took several years from starting with the legal framework to final implementation. Several hurdles need to be overcome: • Legal framework approval is usually delayed by the tobacco industry. • Knowledge of tracking and tracing and associated technologies is scarce at tax administrations. • Tender and bidding processes are complex. • Coordination between domestic tax authorities and customs is weak. Even though the process might be lengthy, the investment in a tracking and tracing system will be repaid with the amount of tobacco taxes that are not lost due to evasion. When implementing a new tracking and tracing system, tax administration should ask for collaboration and technical assistance from intergovernmental organizations and countries that have successfully implemented such systems, in order to speed up and ensure success of the process. Examples of tracking and tracing systems implementation in Chile, Kenya and the EU are detailed in Boxes 3.8, 3.9 and 3.10. For countries that already have fiscal markings in place, the potential interaction between the markings and the implementation of a tracking and tracing system should be taken into account. Further information on this interaction is presented in Box 3.6. 130 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.6 Tracking and tracing and fiscal markings It is becoming increasingly common for stamps to contain some tracking and trac- ing features, such as unique identification markings and basic information on the product that describes the company, tax status or the product itself. The intention is to mark each pack with a unique identification marking so it can be monitored from the point of production to the retailer, including each step in between, thereby creating a complete time and location history. Although a tax stamp could meet the requirements of Article 8 of the Protocol and have tracking and tracing features, in general, the focus of tax stamp systems differs from that of tracking and tracing systems. Tracking and tracing is more than the unique, secure and nonremovable identification markings on the packages of tobacco products. It implies reading or scanning the codes; linking the codes between packs, cartons, master cases and pallets; uploading the information to a database; recording of any shipping and receiving events along the supply chain; and interconnecting the different databases. While new tax stamp programs contain tracking and tracing features, they are primarily intended to facilitate tax collection on the domestic market and not to track duty-suspended cross-border trade or the export of products. The focus of tax stamp systems is on authenticity and the proof that taxes are paid. The focus of tracking and tracing systems is on unique identification and on control of the movements in the supply chain by monitoring and investigating the past and future location of products. Tax stamp programs focus on stock management, verification (that the stamps correspond to the product) and authentication (that the stamps are genuine), while the focus of tracking and tracing systems is on the origin, intended route, first customer and final destination. The focus of tax stamps is primarily on individual packs intended for the duty-paid domestic market, while the focus of tracking and tracing systems is on all packaging (packs, cartons, master cases, pallets) and certainly – but not exclusively – for the duty-suspended export market. Nevertheless, sometimes there can be synergies. For example, the EU countries that require a tax stamp or national identification mark for fiscal purposes have the option to use it as the security feature for tracking and tracing purposes, provided that the requirements are met. In summary, tax stamps can be converted to or be part of a tracking and tracing system when the converted system provides aggregation between packs, cartons and master cases and records all movement along the sup- ply chain. For the export market, a unique identification marking should be added. Sources: (39–42). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 131 Box 3.7 What not to do: use the industry solution to tracking and tracing Codentify is a serialization system used to produce cigarette pack markers for the purpose of verifying whether cigarette packs are legal. It was patented by PMI but subsequently licensed at no cost to other major cigarette manufacturers. In 2016, Codentify was transferred to Inexto, which is an affiliate of the French group Impala. What is the problem with Codentify/Inexto? The main issue is that Codentify/Inexto’s links to the tobacco industry make it incom- patible with the Protocol, which came into force in September 2018. The Protocol specifies that obligations assigned to a Party shall not be performed by or delegated to the tobacco industry. Additionally, many elements indicate it is an ineffective means of authentica- tion. For example, the 12-character digital codes generated by Codentify can be easily duplicated or cloned and used as originals on either a counterfeit or genuine pack, which can then pass the system’s basic verification test. The codes are also produced by relatively unsecured, commercially available equipment and do not include high-security features capable of protecting the authenticity of identifier numbers. Systems that use multilayered, advanced security solutions that enable distributors, retailers, customers and authorities to identify noncompliant products are more secure. Another problem is that Codentify/Inexto cannot track products as efficiently as other available systems. It requires a much larger enforcement capacity to achieve the same detection rates as other systems that are not linked to the tobacco indus- try. Authorities would have to inspect significantly more packs marked under the Codentify system than is necessary under some other systems to achieve the same certainty of not missing a fraudulent pack. In addition, not all stakeholders will be able to verify that a pack marked under the Codentify system is genuine, while other available systems do offer this possibility. Source: (47). 132 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.8 A successful tobacco traceability system: SITRAF, Chile The Servicio de Impuestos Internos (SII, Internal Revenue Service of Chile) has success- fully coordinated and implemented a tracking and tracing system for tobacco products. Application of a compliance management model In the framework of a compliance management model, a traceability system for tobacco products utilizes a structural measure to reduce tax evasion. It is estimated that evasion of taxes (VAT and excise taxes) in the cigarette market in Chile amounts to 16.6% of the country’s total market annually – approximately US$ 300 million. The traceability system implemented in Chile (SITRAF (TAB2)) allows authorities to know in a certain and timely manner the quantities of cigarettes produced or imported into the country. Moreover, it helps authorities to distinguish between counterfeit products and original products that did not comply with payment of the tax. In 2018, the implementation of the traceability system was awarded to a company through public bidding. The company is in charge of the implementation and operation of the system for five years, according to a contract signed with the SII, and must maintain a team of 20 people available for the project. Direct markings are applied to items produced in Chile for national consump- tion, and stamps are used for imported products. For both types of product, the marking is based on a data matrix code, which is printed using security ink that is distinguishable from any other type of ink with specific devices that are provided by the awarded company. Although products for export are not subject to marking, they are controlled and accounted for by the traceability system. For national production, devices are installed on each production line that rec- ognize the type of pack being produced, print a unique code on each pack and then read it (activation) to save all the information on servers located in the production plant. This information is transmitted to the central servers of the system and then to the SII. It is also available for on-site inspection. In the case of imported products, the stamps must be acquired in Chile by each importer and then sent to its producer abroad, which is responsible for adhering them to each pack of cigarettes prior to wrapping the packs with cellophane, us- ing applicators on the production lines. Once the cigarettes enter Chile, the tax determination process has been completed in the service (Provisional Free Transit Guide) and the corresponding taxes have been paid to customs (Import Declara- tion), the importer must enter the data on the stamps used by the importer on the platform of the traceability system. After validation, the stamps can be activated in the system – that is, they are recognized as valid for commercialization. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 133 In addition to the devices provided by the company, SII has developed a smart- phone application for verification by citizens. Although the application cannot verify the authenticity of the ink used, it is able to verify whether a code is correctly gener- ated and display the information contained in the traceability system for the code (brand, variety, quantity of cigarettes, products/importer) so that the taxpayer can verify its consistency. Progress of the compliance management model Implementation of the traceability system has required a coordinated effort both within the service – for the generation of instructions, procedures and computer developments – and with other institutions, such as the National Customs Service and the Ministry of Health. It is an unprecedented project in terms of coordinating the implementation of the system in the production lines of the different tobacco companies in the country and the provider company. Some of the main milestones of the project are: September 2014: Law 20,780 on Tax Reform establishes an obligation to implement the system within a term of six months, after the publication of the resolution determining the obligated tax payers. May 2015: Resolution No. 47 determines obligated taxpayers. June 2015: Circular No. 47 describes obligation to incorporate stamps or distinctive marks as a traceability mechanism. February 2016: Law 20,899 on Tax Reform simplifies the definition of the system, allowing the system to be outsourced or provided by the SII, in addition to making the type of traceability more flexible. August 2016: Traceability system regulation D.S. 1,027 is issued (published on 28 December 2016). March 2017: Exempt Resolution No. 49 of the Ministry of Finance authorizes the SII to outsource all or part of the traceability system. June 2017: Bidding bases in public market are published. February 2018: Tender is awarded to selected company. June 2018: Decision of contract is made by General Comptroller of the Republic. August 2018: Resolution No. 61 determines taxpayers obliged to apply the trace- ability system. August 2018: Holding of first workshops for detailed definitions of the project, with the participation of Customs, Ministry of Health, provider and SII. September – October 2018: Visits of plants to coordinate with producers and define adaptations to production lines for system implementation. January 2019: Resolutions No. 6, 7 and 8 passed, with response to a request for an 134 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N extension of producers; start of system implementation in all production lines in the country. February 2019: Resolution No. 16 establishes a term to commercialize the remaining stock without marking. 2019: Resolution No. 24, with request for extension to importer, includes training of customs staff. March 2019: The traceability system is started up. June 2019: Stamp process begins for all imported cigarettes. Currently, the system is installed and operating in all production lines in the coun- try, placing traceability markings on virtually 100% of the cigarettes produced and imported. During 2019, the traceability system enabled controlling approximately 1 175 million packs: 744 million produced for national consumption, 409 million produced for export and 22 million imported yearly. Source: (48). Box 3.9 Case study of Kenya’s implementation of a tracking and tracing system Kenya’s current tracking and tracing system was preceded by a series of reforms in both tax structure and administration of excise taxes. The reforms included electronic cargo monitoring of exports, which allowed for automatic monitoring and reporting. The system appears to be highly effective because it requires less capacity and is less prone to manipulation than earlier systems. The experience of Kenya shows that a lower-middle-income country can successfully implement a sophisticated system capable of decreasing illicit trade. It also shows the importance of other measures such as strengthening enforcement, increasing cooperation and communication among different agencies and increasing penalties for noncompliance. Illicit trade, as measured by the Kenya Revenue Authority (KRA), was estimated to be around 15% of total consumption in the market during the initial reform period. After the introduction of the new system in 2015, it dropped to 5%. Timeline of the major reforms: 2003: In this period, the paper tax stamps used had a unique identifier and were colour-coded to indicate the type of product. Regular compliance checks were in- troduced. In 2007, the cost of a stamp was 2.124 Kenyan shillings or US$ 0.023 per pack. However, the stamps were found to be easily counterfeited and could not be linked to specific brands. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 135 2008: The KRA proposed a tracking and tracing system and increasing tax rates. The new system was introduced gradually. 2010: Enhanced security features, including ultraviolet markings, were added to the paper stamps. The stamps were to be clearly visible when packs were displayed for sale and placed so that opening a pack would destroy the stamp. The stamps were verified at four different points along the supply chain. The costs were just slightly higher than those of the previous stamps at US$ 0.024 per pack. Licensing was introduced for domestic manufacturers, subject to annual renewal. Importers were required to register with the KRA. Licences required submission of details on the company directors, inventories and equipment, accounting systems, input-to-production ratios and brands produced. Penalties for noncompliance were increased and included up to three years in prison. An electronic cargo tracking system was launched. Electronic seals were affixed on containers or trucks, and GPS technology was used for tracking. A bond was payable on exports to cover excise and VAT taxes. The bond was released only when the goods reached the final destination and taxes were paid. Verification involving both countries of the business deal takes place at the bor- ders. The electronic system provides information about the departure and arrival of the goods and the disarming of the seals. Authorities in the importing country are notified before the shipment leaves the domestic production facility. The system reduces the number of checkpoints and staff needed and generates arrival reports that can be verified with VAT refund requests. As a result of these changes, three factories and seven of the 10 importers were shut down due to noncompliance. Exports to Côte d’Ivoire, Eritrea, Mali and Sudan stopped because companies could not provide evidence that the goods reached the final destination and taxes were paid. More than US$ 11 million in excise tax losses was recovered in 2011. The KRA estimated that illicit trade dropped to 8%. 2013: A contract was signed to introduce a tracking and tracing system for tobacco and alcohol, the Excisable Goods Management System, in April. The system added production counting, tracking and tracing, stock control, processing and other data collection to the existing system. Infrastructure requirements included high-speed broadband internet at production facilities, warehouses, the KRA and ports, along with reliable power or backup generators at those points. Implementation was planned in three stages: • Stage 1 – A new electronic digital stamp with a unique identifier was introduced. It included a data matrix code plus overt markings (holograms, fluorescent fibres, a security link for KRA authentication and visible two-dimensional codes for verification and activation), semi-covert markings (UV features, fluorescent prints detectable by specialized devices, mini text printing for retailers and 136 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N distributors) and forensic taggants for use in prosecutions. The stamp also included human-readable codes for verification by short message service using the KRA web portal. • Stage 2 – Control and monitoring systems were automated in February 2014. Manufacturers had to install photosensitive readers on production lines, with data automatically sent to the KRA in real time. Each stamp was activated and associated with a brand and package size on the line. The KRA database is automatically updated every 15 minutes. • Stage 3 – Market surveillance began, with 83 officers given powers to seize illicit cigarettes and make arrests. The officers were equipped with hand-held devices that transmitted data to the KRA for authentication. Distributors and retailers became liable for selling products without an excise stamp and were subject to fines plus prison sentences of up to three years for noncompliance. In 2016, a smartphone application became available with which the public could authenticate cigarette packs. Importers must now buy digital stamps and send them to export facilities in other countries to be affixed. Tax liability is due at removal from a factory or at import. The electronic cargo monitoring system is still in effect. 2016: The Excise Duty and Tax Procedures Acts clarified new obligations and penalties. 2017: A new integrated customs management system was launched. The KRA estimates that illicit trade levels are now around 5%. The current, more comprehensive digital system is cheaper than the previous paper tax stamp system. Manufacturers pay for the production monitoring system, but it counts as a busi- ness expense on corporate tax returns. In 2018, two manufacturers and 10 licensed importers were operating in Kenya. In 2018, aggregation between the markings of packs, cartons and master cases had not yet been implemented but was expected to be forthcoming. Source: (36) Box 3.10 Case study of the new EU tracking and tracing system, May 2019 Cigarette smuggling and other forms of illicit trade in the EU is estimated to cause a loss of €10 billion in revenue annually. In 2018, 4.2 million packs (20 sticks per pack) of illegal cigarettes were seized by customs in the EU. Illicit tobacco production was also increasing: an illegal factory in Ireland, dismantled in 2018, was capable of producing 250 000 cigarettes per hour. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 137 The EU tobacco control policy is described in the Tobacco Products Directive and is influenced by the Protocol. Article 15 of the Directive calls for the traceability of cigarettes and RYO tobacco products by May 2019 and of other tobacco products by May 2024. The EU tracking and tracing system is sufficiently flexible to be imple- mented at both the regional and the single-country level. Countries can choose among providers as long as the basic requirements are met. The policy provides a high level of protection against any attempts at manipulating the data. The report- ing obligations cover all the economic operators involved in the manufacture and distribution of tobacco products. The EU system requires all unit packets of tobacco products to be marked with a unique identifier Information on the movements of those products is to be stored by third-party data storage providers. The suppliers of the unique identifiers and data stor- age are to be financially and legally independent from the tobacco industry. The data are to be fully accessible to authorities of EU Member States for enforcement purposes. The generation of unique identifiers, as well as all other codes required for pre- registration of economic operators, facilities and machines, will be done at the Member State level by designated identifier issuers. Manufacturers and importers are required to supply information relating to the product and production lines when requesting unique identifiers from the issuers. The issuers will then generate and deliver batches of unique identifiers. On the production line, manufacturers of tobacco products will complete each unique identifier with a marking indicating the date and time. The unique identifier will be a machine-readable, optical, one- or two-dimensional barcode. An anti-tampering device, capable of creating an unalterable independent record of the verification process, must have been installed previously. This additional record will be accessible to public authorities for potential investigation and inspection. Unit packets, as well as aggregated packages such as cartons, master cases or pallets, can be tracked and traced throughout the supply chain. Tracking is also allowed at an aggregated packaging level as long as unit packets remain traceable. During transport, each dispatch and arrival up to the final dispatch to the first retail outlet must be recorded and reported. All recorded information must be submitted to the independent third-party data storage facility, generally within three hours, and 24 hours before dispatch and transloading. Costs, including operational costs, are shifted to the tobacco industry, in line with Article 8 of the Protocol. The EU system of tobacco traceability and security features became operational on 20 May 2019. Sources: (36, 49–50). 138 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 11 A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as monitoring and controlling the movement of tobacco products and their legal status. 3.4.6 ANTI-FORESTALLING “Forestalling” is a term that describes increases in the production or stock of products in anticipation of a tax increase (2). Other terms referring to this practice include “stockpiling” and “front-loading”. Forestalling occurs when manufacturers or importers increase their tax-paid stock or oversupply the market by increasing production or imports in order to pay the previous lower rate. It reduces and delays the effective- ness of tax measures. The effective starting date of the new tax rate will be delayed, revenues will be lower and the possible effect on prices and thus consumer behaviour will also be postponed. To illustrate how anti-forestalling measures function, an example is presented in Annex 3.2. A legal basis must exist for anti-forestalling measures; otherwise, the govern- ment cannot prevent the industry from forestalling. Legal measures to deal with forestalling include (51–52): 1. Limiting the amount of tobacco products that can be released subject to the old tax rate and levying the new tax on the products exceeding that limit. 2. Levying the new tax rate on all goods that are still in stock and not yet sup- plied to the final consumer. 3. Limiting the number of tax stamps issued at the rate that was in effect before the increase or limiting the time that products with a tax stamp with the old rate can be sold. 4. Requiring producers and importers to buy new tax stamps annually or after a tax increase. Under the first three measures, the competent authority determines the limit for taxation at the previous (lower) rate. The quantity allowed may be based on the shelf life of tobacco products – around six months for cigarettes – or normal inventory levels, such as an average over the previous three years. The first measure, limiting the amount of tobacco products that can be released, requires resources from the competent authority for enforcement. Authorities may decide to post inspectors in each production facility, but even without posting inspectors, procedures are necessary for determining when the allowed quantity has been exceeded and what subsequent actions to take. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 139 The second measure, levying the new tax on goods in stock and not yet supplied to the final consumer, might be difficult to implement. The competent authority is required to monitor the manufacturing process and, at the very least, to conduct a stock-taking of all players in the supply chain – including the various manufacturers, importers, wholesalers and retailers – in a very short window of time. If monitoring covers only the stock of manufacturers and importers, this measure could easily be circumvented by ensuring the stock is sold to others in the supply chain or by setting up separate distribution companies to purchase the stock. Controlling stock at the retail level is burdensome and might not be administratively feasible given the large number of cigarette retailers. It becomes even more burdensome if there is no licensing requirement for retailers because they will first have to be identified. The third measure accomplishes the same thing as the first if a country uses tax stamps. The fourth also requires tax stamps and is simpler for the competent authority but somewhat more burdensome for the tobacco companies, since stamps must be purchased every year. Box 3.11 provides examples of anti-forestalling measures in EU countries. Box 3.11 Examples of anti-forestalling measures in EU countries Several EU Member States have taken measures to limit forestalling. A cautious ap- proach seems to be required for designing such measures to ensure that they comply with EU legislation and the general principles of EU law – in particular, the principle of proportionality. No disputes have occurred concerning the right of initiative of EU Member States to implement anti-forestalling measures. Nevertheless, several EU Member States had to defend their measures in front of the Court of Justice of the European Union, the institution that ensures all national legislation is in line with EU law and a consistent application of that law (53). The Court acknowledged that anti-forestalling measures are appropriate to combat tax evasion and tax avoidance. Moreover, the Court emphasized that fiscal legislation is an important and effective instrument for discouraging consumption of tobacco products and therefore for protecting public health (54–55). However, the measures taken should be proportion- ate to the objectives. The principle of proportionality means that only the action needed to achieve the objective should be taken, and it should not exceed what is necessary. This principle regulates the measures taken within the EU and is included in the Treaty on the European Union. The Court demanded that Portugal amend its legislation to ensure compliance with the principle of proportionality. Belgium, Estonia and Hungary were also urged to change their anti-forestalling measures to bring them in line with EU legislation (56–57). 140 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The majority of the EU Member States have anti-forestalling measures in place, but there is no harmonization of these measures. The following are some examples:12 • Portugal limits the quantity of cigarettes that can be released in the last four months of a year to the average of the previous 12 months plus 10%. In addition, manufacturers and importers must sell cigarette packages with a tax marking of the preceding calendar year within three months. For other tobacco products, longer limits apply. • Denmark limits the number of tax stamps issued before a tax increase at the old rate to 20% more than are usually purchased in the two months before the end of the year. • In Poland, tax stamps are valid only for the current calendar year, and cigarettes with the old stamp can be sold only through February of the following year. • In Romania, companies must apply for approval to release for consumption from the customs office. Source: (58). KEY TAKEAWAY 12 Forestalling reduces and delays the effectiveness of tax measures. Implementing anti-forestalling measures can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. 3.4.7 ADDITIONAL NATIONAL AUDITS AND CONTROLS In addition to the previously described measures, several periodic audits and controls could be implemented to increase compliance with tax laws. The most common audits and controls are the following: • Cost audit – The cost audit method provides expected VAT and tobacco tax collection by simulating the intermediate and final cost of cigarettes. It starts with the inventories of raw materials and estimates added values and final cost, then matches the results with real collection from the tobacco supply chain. Annex 3.1 provides more information about the components that make up some selected (tobacco) products. • Transfer pricing audit – To ensure companies pay their fair share of tax, prices of transactions between related companies should be assessed, and when prices are not in line with the market conditions, they should be corrected. Companies that operate at the international level (transnational companies), including many tobacco companies, can manipulate import or export prices 12 Considering the frequency of court cases concerning anti-forestalling incidents, these measures might have been replaced or amended. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 141 of merchandise or raw material to related companies or branches in other countries, with the objective of lowering profits in countries with higher tax rates, and can transfer those profits to countries with lower taxes. • Price and market monitoring – Retail price surveys can provide information about variance from the market price in certain locations, highlighting areas of potential tax avoidance or illicit trade. Physical control of such locations requires rapid response teams, as implemented in the Philippines. To monitor the tax compliance of its taxpayers, the competent authority in the Philippines needs to understand the tobacco market; it must have information on brands, market segments and prices of products. This information enables authori- ties to estimate the impact of tax and price changes on consumer behaviour and revenues. Market data can be analysed as part of risk management and anti-fraud analysis to determine whom to investigate for noncompliance and when to do so. Sales data can be triangulated to validate other data sources, such as household surveys on prevalence. Market data and trends are also useful indicators for determining whether there is a case of oversupplying. See also section 3.4.3. • Consumer control – Involving the public via awareness campaigns has also been shown to be effective. Consumers have the right to be assured that the products available in the market are authentic and come from legitimate sources. Thus, it is in the consumers’ interest to understand and be able to verify that they are buying genuine products. The features of the fiscal marks on tobacco products should help consumers distinguish between genuine and illicit products. Some countries – Kenya, for example (see the case study in Box 3.9) – use a smartphone application to allow anyone to check both covert and overt features and to report any cigarettes with incorrect markings. Other countries, such as the Netherlands, have developed a smartphone application that allows anyone to report a suspected case of excise tax fraud. • Cross-check controls – Competent authorities should consider using multiple sources to obtain market data and determine if these data are consistent with tax declarations. VAT declarations can be used to verify that suppliers and purchasers of raw materials and final products are reporting the same amounts. Bank information can be used to verify both sides of transactions along the supply chain. Any discrepancy can alert the competent authority to conduct further investigation for possible illicit trade of tobacco or tax evasion. 142 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 13 Several different types of periodic audits and controls that can be carried out to increase compliance, including cost audits, transfer pricing audits, price and market monitoring, consumer controls and cross-check controls. 3.4.8 IMPORT AND EXPORT CONTROLS Parties to the Protocol should allow import and export of tobacco products and manufacturing equipment only by duly licensed natural persons or legal entities (Article 6.1). A well-known strategy used by fraudsters is to declare products for export so that no duties are due to the country of export. These products are subse- quently transported through other countries, using the in-transit regime that allows temporary suspension of duties until the goods arrive at their final destination. Before arriving at their final destination – where the excise duties would be due – the goods disappear or are lost while being diverted to the illegal supply chain. The goods may never leave the country, or they may be smuggled back into the country from which they were exported without declaring or paying duties. This risk of loss of revenue can be mitigated by requiring a guarantee or bond, which will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless the required fiscal marking (such as a tax stamp or export label) is affixed on the pack, according to the national law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Illicit tobacco trade could be decreased significantly if the various competent authorities that have jurisdiction over manufacturers and exporters of tobacco prod- ucts and manufacturing equipment would provide the competent authority at the destination with prior information when a shipment has been authorized and is about to take place. The information could include the name of the consignee, a description of the item shipped and the quantity. Also, the competent authority at the destination should inform the competent authority having jurisdiction over the shipper that a shipment was received along with the pertinent information relating to the shipment. A good IT system is also required for import and export. Electronic processing of prior-to-arrival manifest and import declarations is recommended. Most countries have implemented an online customs system to process import and export declara- tions, including all required data such as country of origin or export, description of merchandise, value, weight, cargo insurance, carrier, importer or exporter and broker identification, detailed tax duties to pay and final destination. The World Trade Organization (WTO) Trade Facilitation Agreement provides sev- eral tools for better controls, including collaboration between customs administrations, CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 143 exchange of information, use of non-invasive devices and prior-to-arrival import declaration (23). Non-invasive detection equipment at customs posts is highly capable of detect- ing contraband merchandise. The most common tools are X-ray scanners that are used for small parcels, containers, trucks and trains. Most modern ports have also implemented the use of X-ray scanners, and such technology is improving the speed of controls to as little as two minutes per container. Although the cost of scan- ners is declining, it remains inaccessibly high for countries with limited resources. Fortunately, scanners are often available for lease, making them accessible for tax administrations in those countries. Less sophisticated and less costly detection equipment includes endoscopes, mirrors, night vision equipment, cameras and automatic licence plate readers (33). A still less expensive alternative is the use of dogs, which can be trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied in order to reduce contra- band. These include the separation of processing operations from the sealed storage of taxed and untaxed products. Physical and direct control by officials of the excise authority during a part or the whole of an operation can be applied (for example, physical escort of the transit consignment from border to border by individual trucks or in a convoy, or application of radio or satellite tracking systems such as GPS-enabled devices to goods, conveyances, vehicles or containers). Control at borders is essential and should include integrated technology and cooperation with agencies at the border station. Frontline officers should be sup- ported by appropriate intelligence, guidance and supervision from management, as well as technical aids to enforcement. Within a country, mobile excise control units are helpful for verifying excisable goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. These operations require close coordination between police, border guards and other public services. Exports also require special attention, in particular if VAT and tax refunds are granted to the export of tobacco products. Validating the real exit in the declared amount is essential to avoid illegal re-entry to a territory and the improper refund of taxes. For any tax refund, an audit including tax credit information must be carried out. The audit may include the invoices for the whole chain involved in the export, including tobacco farmers, first processors, manufacturers, wholesalers, storage and transport. 14 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N KEY TAKEAWAY 14 To ensure control of import and export, it is recommended that only duly licensed persons and entities be allowed to import and export tobacco products and manufacturing equipment. 3.4.9 FREE ZONES AND TRANSHIPMENT POINTS The term “free zone” is very broad and can refer to a number of different types of areas. The Financial Action Task Force listed the following types in its 2010 report (59): free trade, export processing, enterprise, free ports, foreign trade, special economic zones and bonded warehouses. A number of these areas can include tobacco manufacturing and trade. By definition, controls such as regulation and oversight within free zones are less strict than in other areas. This can make them appealing to persons involved in illegal cigarette manufacturing or trade (17). In fact, illicit activities related to free zones (not limited to tobacco) are regularly documented by organizations that recognize the linkage. These activities include money laundering, tax evasion and trade in counterfeit goods or other illicit goods (60). A report from the European Parliament (61) referring to free ports in particular, mentioned that the motivation for using them included a “high degree of secrecy and deferral of import duty and indirect taxes”. The report even proposed the “urgent phasing out of free ports”. In the European Parliament report, free ports are free zones that function as (semi-) permanent storage areas for high-value goods. The Protocol includes a time-bound provision of effective controls on all manu- facturing and transactions of tobacco products in free zones (Article 12). Free zones are defined as a part of the territory of a Party where goods are considered to be outside the customs territory for import duties and taxes (Article 1.5). This is the same definition used in the International Convention on the Simplification and Harmonization of Customs Procedures (Revised Kyoto Convention) (62). Parties to the Protocol must implement effective controls in free zones within three years of entry into force of the Protocol. For countries not yet Parties to the Protocol, stringent controls of manufacturing and transactions involving tobacco products in free zones are an important component of an effective and efficient tax administration. One of the measures for dealing with free zones within the Protocol includes implementing “effective controls on all manufacturing of, and transactions in, tobacco and tobacco products, in free zones, by use of all relevant measures as provided in this Protocol”. As indicated in an Interpol report (17), a significant vulnerability of free zones is the fact that different economic operations (e.g. manufacturing, assembly, re-packaging and warehousing) take place outside the control of authorities. It is therefore essential for customs administrations to exercise their authority in free zones CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 145 to effectively identify and fight illicit trade in tobacco products. Relevant measures listed in the Protocol should be applied. These include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a track- ing and tracing regime. Removing exemptions on excise taxes is an additional way to increase control and remove incentives for using free zones as a means for tax evasion. Indonesia, for example, imposes excise taxes on cigarette manufacturing in its free trade zone.13 Parties to the Protocol shall also prohibit the intermingling of tobacco products with non-tobacco products in a single container or any other such similar transporta- tion unit when removed from free zones. Finally, each Party is to “adopt and apply control and verification measures to the international transit or transhipment of tobacco products and manufacturing equipment in conformity with the provisions of the Protocol”. Article 13 of the Protocol, which covers all duty-free sales of tobacco products, requires Parties to the WHO FCTC to consider prohibiting or restricting the sale to or import by international travellers of tax-free or duty-free tobacco products, as mentioned in Article 6 of the WHO FCTC. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect government revenues by creating a loophole in the tax structure (2). KEY TAKEAWAY 15 Customs administrations should exercise their authority in free zones to prevent different economic operations from taking place outside the control of authorities. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing a tracking and tracing regime. 3.4.10 PROCEDURES AFTER DETECTING ILLICIT TRADE OF TOBACCO The procedures described in previous sections are intended to increase compliance and to prevent illicit trade. When smuggling or illicit trade is detected – through, for example, audits, tracking and tracing systems, verification of declarations or border control – actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. To deter further illegal behaviour, a comprehensive audit of everyone and everything involved in the illicit acts must also be carried out. Assets and vessels involved in the illicit activity can be seized, and financial accounts can be frozen. Some countries, including the United Kingdom, Canada and Chile, have also adopted a strategy known as “follow the 13 Indonesian Ministry of Finance, personal communication, January 2020. 146 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N money” to obtain more information on those who finance illicit trade. This strategy is intended to have a further-reaching effect by targeting those who finance the transport, production and storage of illicit products. The United Kingdom, Canada and Chile all created special teams tasked with identifying and targeting the offenders. Furthermore, Article 18 of the Protocol provides for the confiscation and destruction of tobacco, tobacco products and manufacturing equipment. One of the difficulties faced by competent authorities in exercising the authority to seize and forfeit products and/or equipment used in the manufacture or distribution of tobacco products is the cost of keeping or storing the goods and/or machinery before destruction. Thus, the law should also provide for a mechanism and timetable for the disposal and/or destruction of seized and forfeited goods or machinery, while prescribing a mechanism by which these properties can still be presented as admissible evidence in a judicial proceeding. Boxes 3.12 and 3.13 provide examples of successful efforts to combat illicit trade in the United Kingdom and Indonesia. KEY TAKEAWAY 16 As soon as smuggling or illicit trade in tobacco products is detected, actions such as collecting taxes and seizing and destroying smuggled and/or illicit tobacco must be taken. Box 3.12 The United Kingdom’s experience in fighting illicit trade in tobacco products In 2000, illicit cigarettes accounted for 22% of the cigarette market in the Unit- ed Kingdom. To deal with the problem, Her Majesty’s Customs and Excise14 implemented a major anti-smuggling effort. The strategy was refreshed with ad- ditional resources and measures in 2011 and reviewed in 2015. The result was a steady decline in the illicit cigarette market to 10% by 2013/2014. The measures taken were comprehensive and included hiring 1 000 new customs officers and investigators. In addition, tobacco supply chain legislation was introduced, aimed at discouraging tobacco manufacturers from facilitating smuggling. Tougher sanctions included increased fines of up to £5 million levied on a manu- facturer, criminal prosecution with sentences up to seven years, confiscation of assets as part of the proceeds of the crime, payment of duty on the confiscated goods plus penalties up to 100% of the duty, prohibition of the sale of tobacco products for 14 By the time of the renewed strategy, the respective bodies were the HMRC and the United Kingdom Border Agency. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 147 up to six months, unlimited fines for selling tobacco without the United Kingdom duty-paid fiscal mark after 13 March 2015, referral for withdrawal of the transporter’s licence, use of immigration sanctions to refuse entry to the United Kingdom for tobacco smugglers and civil action, including bankruptcy. The cost of these measures was £209 million over the first three years of the program and around £100 million annually by 2008/2009. This figure covers only HMRC and excludes any costing of the United Kingdom Border Agency. In 2013/2014, tobacco tax revenues were £9.5 billion. After a review of the strategy in 2015, the controls on raw tobacco were strength- ened by the introduction of an approval system in 2017. Anyone who manufactures, purchases, acquires, owns or is in the possession of a tobacco products manufacturing machine must be licensed with customs as of 1 August 2018 (63). The United Kingdom ratified the Protocol on 27 June 2018. It was the 40th country to ratify, which was the trigger point for the Protocol to enter into force. Fig. 3.4 Estimate of the illicit cigarette market and United Kingdom tax-paid consumption Sources: (16, 64). Bi lli on c ig ar et te s UK tax paid consumption Illicit market 20 00 -01 20 01 -02 20 02 -03 20 03 -04 20 04 -05 20 05 -06 20 06 -07 20 07 -08 20 08 -09 20 09 -10 20 10 -11 20 11 -12 20 12 -13 20 13 -14 20 14 -15 20 15 -16 20 16 -17 20 17 -18 0 10 20 30 40 50 60 148 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Box 3.13 A success story: Indonesia reduced illegal cigarettes from 12% to 3% of the market In some countries, most of the illegal cigarettes are imported, but in Indonesia most of them are produced within the country by unregistered manufacturers that are usually home-based and relatively small. In several territories, specifically on the Island of Java, illegal cigarettes have been produced for generations. This practice is supported by the availability of raw tobacco materials and cloves, as well as cheap labour costs, especially for female workers. Indonesia also produces cigarette products that are not available in most other countries. These products – known as handmade clove cigarettes (sigaret kretek tangan) – contain cloves, and the production process covers blending, rolling and packing. Parts of the process are done by hand, and 99% of the labourers are women. The strategy for combatting the illegal cigarette trade in Indonesia is divided into two main parts: preventive actions and responsive actions. Indonesia’s success in effectively tackling illicit trade is attributed to the following key factors. Monitoring and surveillance Preventive actions consist of administrative measures – such as issuance of permits and the excise stamp purchasing mechanism – that use risk management by optimizing the Excise Service Information System (ExSis). With this IT system, the Directorate General of Customs and Excise (DGCE) can oversee both daily transactions and daily production from factories. When information of suspicious activities is obtained, DGCE can suspend the purchase of excise stamps. The efforts to fight the illegal cigarette trade also invite stakeholders to be involved by supplying information regarding high-risk areas and regional governments. Strategic communications and community involvement DGCE continuously disseminates information and conducts public education to fight illegal cigarettes. These efforts are conducted every year, using a special campaign slogan. In 2019, the slogan was “Gempur Rokok Ilegal” (“Fight Illegal Cigarettes”). Key performance indicators for DGCE units and offices Parallel with the above-mentioned preventive actions, DGCE also continuously conducts responsive actions: enforcement, investigation and audit activities in cigarette factories. Enforcement activities are planned and measured by consider- ing the limited human resources and the large scale of the monitored territories. To demonstrate the effectiveness of administrative and enforcement measures in curbing the trade of illegal cigarettes, both types of activities are translated into key CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 149 performance indicators for all DGCE working units and offices, including regional offices and personnel. Use of technology and intelligence Coordination between different DGCE offices responsible for monitoring the produc- tion and marketing of cigarettes is maintained by sophisticated IT applications that enable efficient distribution of information and investigation activities. The applica- tions are the Customs Intelligence and Tactical Centre for data and analysis and the Centre for Command and Control (Pusat Komando dan Pengendalian/Puskodal) for ensuring that sea patrols work effectively and efficiently. Independent evaluation To evaluate the efforts and activities to reduce the circulation of illegal cigarettes in Indonesia (e.g. cigarettes without stamps, with fake stamps or with used stamps), in 2016, the government commissioned the University of Gadjah Mada in Yogyakarta to conduct a survey using a stratified random sampling method. To maintain objectiv- ity and independence, an independent body from this well-known university was appointed to conduct the survey. The survey results showed that the level of illegal cigarette circulation in Indonesia was 12.1% of total consumption. In 2018, the DGCE commissioned the University of Gadjah Mada to conduct another survey. Results showed that circulation of illegal cigarettes had been reduced to 7.0%. In 2019, using the same method the university used, DGCE conducted a survey that showed a reduction to 3.0%. Fig. 3.5 shows the results of the surveys. Fig. 3.5 Share of illicit trade in total cigarette consumption in Indonesia, 2016–2019 Source: Customs and excise department, Ministry of Finance, Indonesia, personal communication, 2020. 2016 2017 2018 2019 12.1% 10. 9% 7% 3% 150 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The results of the actions taken can be used as feedback for DGCE in developing excise policies for both service and enforcement, including policies for excise tariffs. DGCE will provide recommendations for an optimum policy format for controlling consumption, maintaining labour protections, optimizing revenue and, most impor- tantly, constantly reducing the consumption of illegal cigarettes on a national level. 3.4.11 PENALTIES Penalties and sanctions must be sufficient to deter illegal activities. Otherwise, finan- cial penalties may simply be paid as a cost of doing business while the illegal activity continues. The Protocol specifies commitments for Parties and provides information on best practices for non-Parties. Article 14.1 in Part IV of the Protocol requires each Party to establish unlawful activities, including manufacturing, wholesaling, broker- ing, selling, transporting, distributing, storing, shipping and importing or exporting tobacco products or manufacturing equipment without the payment of applicable duties or taxes or without using fiscal stamps or other required markings or labels. Articles 14.2 and 15 of the Protocol mandate Parties to determine which of the types of unlawful conduct set out in Article 14.1 shall be criminal offences. Parties must adopt legislative and other measures to give effect to such determinations, as well as to define whether the liability for committing illicit trade in tobacco is a criminal, civil or administrative offence. Article 17 further provides that the Parties shall consider adopting measures as needed to authorize competent authorities to levy penalties in an amount proportionate to lost taxes and duties resulting from the commission of illicit trade. Box 3.14 provides a case study of how Colombia used penalties to fight illicit trade. Box 3.14 The use of penalties to combat illicit trade in Colombia In 2017, the specific tax on cigarettes in Colombia was doubled, increasing from COL$ 700 per pack in 2016 to COL$ 1 400 in 2017. The tax rate was tripled from 2016 levels in 2018, reaching COL$ 2 100 per pack. A provision was added to increase taxes annually after 2018 at the rate of inflation plus 4%. In 2015, before the tax increase, Law 1762 introduced a number of measures to fight illicit trade more effectively. The length of imprisonment for dealing in contraband cigarettes was increased from 3–5 years to 4–12 years. Moreover, government officials who facilitate illicit trade – or anyone involved in transporting or retail sales of illicit tobacco – face similar prison terms. The law allows vehicles used for smuggling to be confiscated, and penalties were increased for illicit trade that is conducted through areas such as special economic zones. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 151 Under the law, illicit trade is considered to be a source of money laundering, which means that the Financial Intelligence Unit can use the same methods it uses to investigate other illegal financial activities. This practice is not common. The law specifically created new sanctions related to alcohol and tobacco excise tax evasion, including the seizure of goods, fines, closure of retail outlets and the suspension or cancellation of licences, authorizations or registries. Arrests and seizures have increased under the new law. Indeed, since its enactment, law authorities reported that between 2016 and 2018, five criminal organizations were dismantled, 53 individuals were apprehended and 72 assets were confiscated. In addition, 2 236 individuals were apprehended and 503 vehicles transporting smuggled goods were confiscated, as transport of such goods is now also considered a crime under the law. More importantly, thanks to the large tax increases, consumption decreased while revenues increased substantially in 2017 and 2018 (see Fig. 3.6). It is estimated that illicit trade in cigarettes in five Colombian cities in 2016 constituted 3.5% of total consumption, a much lower estimate than the industry data suggest. In 2017, after nine months of the tax increase implementation, a similar study found that illicit cigarettes remained low, at 6.4% of total consumption. Fig. 3.6 Packs sold and tobacco tax revenue before and after the tax increase in Colombia, 2016–2018 Sources: (65–67 and Ministry of Finance, Colombia (Direccion de Apoyo Fiscal), personal communication, 2020). 2016 2017 2018 Cigarette sales, million packs Cigarette excise revenue, million US$ current 0 100 200 300 400 500 600 700 800 673.7 521 446.4 195.1 301.2 386.2 33.7% reduction 97.9% increase 152 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N For consumers in possession of illicit tobacco, the minimum penalty should be confiscation and destruction of the illicit tobacco products found in their possession, with payment required for the unpaid tax and duties on those products. In the state of California in the United States, it is illegal to possess a tobacco product on which taxes are due and not yet paid. The burden of proving that taxes have been paid is on those who have the products in their possession. The provi- sion is enforced by the California Department of Tax and Fee Administration and local law enforcement agencies. A violation is a misdemeanour, with a maximum fine of US$ 5 000 and/or up to one year in prison. Illegal packages are subject to seizure and forfeiture. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are using this type of legislation to address illicit trade of tobacco. Asset confiscation and increased penalties for involvement in illicit trade are becoming more common as well. Withholding or even confiscation of trucks involved in smuggling is also common in several countries. KEY TAKEAWAY 17 Penalties and sanctions imposed should be sufficient to deter illegal tobacco trade activities. Penalties should be levied in amounts proportionate to lost taxes and duties resulting from illicit trade. 3.5 TAX ADMINISTRATION OF OTHER TOBACCO PRODUCTS In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. However, there is a lack of standard- ization of other products and sometimes large informal markets. For example, it is estimated that two thirds of waterpipe tobacco in the EU is non-duty-paid (19). Other tobacco products – such as bidis in South-East Asia, waterpipe tobacco in the Eastern Mediterranean region and snus in Sweden – are considered part of a country’s traditions. This sometimes leads to situations where governments are hesitant to strongly regulate and tax these products. Some products, such as kreteks (clove cigarettes) in Indonesia and bidis in India and Bangladesh, are mainly sold in one market. Other tobacco products are more likely to be produced by hand on a small scale, making it difficult to detect and collect taxes on them. The same applies to RYO tobacco, which can be produced on a small scale by hand or with the use of small machinery. The trade in raw tobacco and small-scale home production of tobacco often take place outside of monitoring and control systems (19). As mentioned in section 3.3.1, countries have found various solutions to address this problem, including prior approval for purchase or sale of raw materials and reg- istering, authorizing or licensing of all operators and growers that handle raw tobacco. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 153 3.5.1 NEW AND EMERGING NICOTINE AND TOBACCO PRODUCTS In principle, adding a product to an existing tax framework is not likely to impose significant costs. It is reasonable to expect that challenges similar to those faced in dealing with conventional tobacco products will be faced in the collection of taxes on new products, as market players will attempt to use loopholes in tax regulation to avoid or evade taxes whenever possible. However, new challenges are expected to arise when those new products involve rapidly changing technology and where their market dynamics are widely unknown. Furthermore, taxation of new tobacco products may require additional capacity, as a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. 3.5.2 HEATED TOBACCO PRODUCTS (HTPs) Many countries apply a specific excise on tobacco products according to tobacco weight (see Table 2.4 in Chapter 2). With HTPs, this is likely to impose a challenge, since assessing the content of tobacco in a heated tobacco stick will be an additional burden. From a tax administration perspective, it will be easier for authorities to apply taxes per stick or per unit, as is done for cigarettes. 3.5.3 ELECTRONIC NICOTINE AND NON-NICOTINE DELIVERY SYSTEMS (ENDS/ENNDS) PRODUCTS Some countries tax only nicotine-containing e-liquids while others tax both nicotine- and non-nicotine-containing e-liquids. Taxing only nicotine-containing e-liquids re- quires laboratory capacity to detect the presence of nicotine (see Table 2.5 of Chapter 2). Self-declarations by industry are not sufficient, since some e-liquids labelled as nicotine-free have been found to contain nicotine (see section 2.4.2). Therefore, it is simpler to tax both nicotine- and non-nicotine-containing e-liquids. One likely challenge of taxing all e-liquids will be the capacity to detect and differentiate whether the e-liquids used in ENDS/ENNDS are falsely declared as being for other purposes at the import and manufacturing levels. More information on advantages and disadvantages of different excise tax policies is given in Table 2.6 of Chapter 2. The challenge in taxing the other components of ENDS/ENNDS products is their diversity (see section 2.4.2 of Chapter 2) and the possibility that some parts may be used for other purposes (e.g. in batteries). As indicated earlier, rapidly changing technology and the lack of control and knowledge of the market make taxation of ENDS/ENNDS devices challenging. It may be for this reason that the majority of countries that tax those products address only the e-liquids. When applying a tax on these newer products, countries should be aware that many customers buy their products online. It is therefore recommended that countries 154 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N deciding to tax these products draw up a proper implementation plan, including how taxes will be collected on imported products and online sales. Online cross- border sales of tobacco products are not permitted in all countries. Several EU countries have banned such sales, which include online sales from retail outlets to consumers in another country. Of course, such bans makes sense only if there is also capacity to enforce them. As in the case of taxing tobacco products, the following actions will be important to more effectively impose taxes on these products: 1. implementing strong enforcement mechanisms such as licensing, recordkeep- ing and control of the supply chain, which can include but is not limited to: a) imposing strict licensing of retailers, importers and manufacturers; ideally, licensing of all those involved in the supply chain and developing a tracking and tracing regime for ENDS/ENNDS products and HTPs (to share cost, this can be done in tandem with the system developed for cigarettes); b) exercising the right to set the frequency and type of audits or controls; c) exercising the right to confiscate goods; and d) imposing sanctions such as penalties, fines and/or withdrawal of licences (if applicable) if legislation is not respected. Specific to ENDS/ENNDS products: 2. implementing highly consequential sanctions for producers who declare nicotine-containing e-liquids as “non-nicotine-containing”; and 3. requiring a fee (contribution to the costs) for laboratory tests when a new product is brought on the market or when there are significant modifications to an existing one. More information on the policy options to apply excise taxes on ENDS/ENNDS is given in Chapter 2, section 2.4.2, Table 2.6. KEY TAKEAWAY 18 In principle, the administration of taxes on new and emerging nicotine products and tobacco products should be similar to that for cigarettes. Due to the lack of standardization of these products, however, a rapid and constantly evolving understanding of them and their supply chain will be required to achieve effective and efficient administration of taxes. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 155 3.6 THE BROADER ELEMENTS OF A GOOD TAX SYSTEM 3.6.1 PROPER RESOURCING OF COMPETENT AUTHORITIES In addition to legal tools and a legal basis on which to act and enforce, the authori- ties in charge of implementing excise tax laws should be provided with sufficient resources to hire the necessary staff to properly implement and enforce them. The necessary staffing could encompass multiple agencies and will often require coop- eration between agencies, since some aspects – such as regulation, licensing and border control – may be performed by agencies other than the competent authority. Staff of competent authorities need the necessary tools, equipment, training and supplies to carry out their functions. This requirement includes the means to build or purchase and maintain a software system that will allow taxpayers to submit required information electronically. Electronic filing has benefits for both taxpayers and authorities. It minimizes the compliance cost for taxpayers and can therefore support voluntary compliance (68). To identify risks of noncompliance, the software system should offer competent authorities the ability to analyse the data submitted by taxpayers and cross-check it with data from other taxes – such as VAT – and third-party sources, such as banks and household surveys. Another option for authorities is to make more efficient use of existing resources. For example, an authority could optimize the risk management system by switching to a risk-based approach: resources could be saved by auditing taxpayers who are more likely to be noncompliant based on risk analysis rather than auditing all of them. Other problems that challenge the effective functioning of a competent authority are lack of a coherent strategy and problems with professionalism related to lack of training or corruption (69). Having a strategy avoids directing resources towards less- important areas. The strategy should always be aligned with the objectives, so that com- petent authorities can identify which steps they should take and in which order they should take them to reach these objectives. A strategy is indispensable to prioritizing and organizing resources so that identified issues or risks can be addressed efficiently. 3.6.2 CORRUPTION Competent authorities should implement tax laws with integrity and have strict rules and regulations for detecting corruption. Strict rules and regulations should also be in place for the punishment of both agency personnel and taxpayers who engage in corrupt practices. Corruption within a competent authority results in the improper monitoring of tax compliance and is one of the causes of the proliferation of illicit trade in tobacco products. It also erodes confidence in competent authorities and ultimately in governments overall. In addition to effective laws and regulation, strong internal audits covering prevention, investigation and sanctions should be 156 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N implemented. To improve prevention, a risk map should be created that highlights areas of misconduct and possible leakages. An action plan to update controls should be established to improve areas of weakness detected in procedures and systems. The audits should also be scheduled regularly. Prepared internal auditors with pow- ers to conduct investigations are necessary. Sanctions for corruption, including administrative sanctions and criminal prosecution, must be strong. 3.6.3 A STRONG JUDICIARY The judicial system should be honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. The appeals process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially in the context of illicit trade. KEY TAKEAWAY 19 Broader elements of a good tax system include (1) proper resourcing of competent authorities to hire staff and obtain necessary equipment and systems, (2) strict rules and regulations to detect and punish corruption among both agency personnel and taxpayers and (3) ensuring that the judicial system is honest and independent, with disputes being solved as quickly as possible. 3.7 CONCLUSIONS Policies are more effective if they are properly implemented and enforced. Com- petent authorities have a key role in the achievement of financial and public health objectives of excise taxes. Given the close linkages between tax administration and efforts to fight tax evasion resulting from illicit trade, this chapter draws extensively from the Protocol to Eliminate Illicit Trade in Tobacco Products (i.e. the Protocol). The Protocol provides a blueprint of measures to address the problem of illicit trade and can be used as a model even by countries that are not Parties to it. Qualities of an effective and efficient tax administration include institutional ar- rangements where roles and responsibilities of competent authorities are clearly defined to avoid overlap and voids. Additionally, effective collaboration among relevant bodies must be facilitated. At the national level, within any organizational arrangement, it is vital that agencies cooperate and exchange information and that their competences find their basis in law. A legal basis for exchange or access to information between government bodies should be ensured. At the international level, especially for border control, the role of customs is key, and access to international CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 157 cooperation agreements such as the Protocol is very useful. An organizational tax administration structure must include a system of performance evaluation and accountability through pre-defined key indicators. To ensure compliance, the accuracy of information for the tax compliance cycle is key, including clear and straightforward taxpayer registration and licensing, declara- tion, recordkeeping, warehousing, distribution, collection and tax refund processes. • Licensing is a powerful tool for obtaining information and securing the supply chain of tobacco products. Ideally, all persons involved in the growing of tobacco and the retailing, transporting, wholesaling, brokering, warehousing and distribution of tobacco products or manufacturing equipment should be licensed. • Collecting as much information as possible on the business of tobacco and recording all transactions are key to reducing tax evasion, but this may be burdensome for authorities. The use of IT for periodic tax declarations, ac- counting, inventory and financial information is critical for obtaining accurate information and can help decrease the cost of the whole reporting system. • Recordkeeping should be ensured. All persons or entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment should keep complete, detailed and accurate records of all relevant transactions and details of materials used in the production of those products. • Maintaining a system of authorization for warehousing allows the authori- ties to carry out controls in production and storage facilities to ensure that taxes are paid. Ideally, bonded warehouses should be eliminated from the supply chain. • Duty suspension – which is often applied during the producing, processing, holding, receiving and dispatching of excise goods – should be granted only if strict criteria are met (e.g. for granting authorization, warehouse pre-authoriza- tion visits, adequate stock control measures, checking the origin of excise prod- ucts and the entire production process and coding and marking of products). • To limit the number of taxpayers a competent authority has to manage, tax collection should take place close to the point of production and import. • Refunds for VAT, excise taxes and customs duties are common in most coun- tries, under the principle that taxes are not exported. The refund process must be closely monitored to avoid opportunities for tax evasion. Control and enforcement – key components of tax administration – include a number of measures to secure the supply chain: licensing and due diligence, fiscal markings, 158 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N track and trace, anti-forestalling measures, audits and controls, import and export control and attention to free zones and transhipment points. Control and enforce- ment need to be included as pillars in the strategic plan of the tax administration. Enforcement and control plans must be designed to define the activities and taxpayers that are subject to enforcement and to allocate staffing, auditing, infrastructure and IT resources. Targets must be defined, including the number of interventions and any additional collection or reduction of tax evasion. This includes choosing interventions for those who have a higher probability of noncompliance (the risk- based approach). In the tobacco supply chain, import, export and transfers to and from warehouses may be areas at greater risk of noncompliance. • Licensing provides timely and accurate data that can serve as the basis for audits, since it identifies and controls legitimate operators. The process of licensing control must be carried out and updated periodically – in particular, by controlling the validity of bonds or guarantees and the proper functioning of the required systems and recordkeeping. Where licences are required, the law should include a provision that disallows purchases from unlicensed suppliers or sales to unlicensed purchasers. This means that both suppliers and purchasers will need to verify those with whom they are doing business. This will substantially help to reduce the burden of proof for authorities. In addition, to maintain a high level of control, the validity of licences should be limited in time, making renewals or reapplication required. • Another important measure for controlling and monitoring production and import of tobacco products is the use of fiscal markings (e.g. tax stamps). In addition to increasing compliance with tax laws, fiscal markings can help distinguish between genuine and illicit tobacco products. The use of fis- cal marks enables both the competent authority and the public to monitor whether the taxes on tobacco products have been properly paid. In addition to locally produced and imported products, tobacco products for export should also be required to be marked, but with an indication that they are for export. Requiring a standard package size can facilitate the application of fiscal markings. To lower the chance that fraudsters attempt to re-use fis- cal markings (in particular, stamps) the marking should be applied to each pack of cigarettes (and other tobacco products) before the pack is wrapped with cellophane. Fiscal markings should be issued to the manufacturer or importer of tobacco products only when excise taxes for the products have been fully paid or a guarantee is established. Fiscal markings should include several security features to make them more difficult to counterfeit. These can include overt, covert, semi-covert and/or forensic features. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 159 • Tracking and tracing systems assist authorities in determining the origin of tobacco products and the point of diversion, if applicable, as well as in monitoring and controlling the movement of tobacco products and their legal status. The objective of a tracking and tracing system is to provide authorities with information on all transactions throughout the entire tobacco product supply chain until duties are paid or other obligations are discharged. Any tracking and tracing system must be able to uniquely identify individual products. By marking a product with a unique code or identifier, it becomes possible to unambiguously register that product’s movements. A good tracking and tracing system enables the government to properly monitor the supply chain, improve its ability to ensure collection of the proper duties and taxes, authenticate whether the identification marking is genuine and matches the product, improve its ability to enforce the law and provide sufficient evidence to prove noncompliance by a violator. To reduce the financial burden of implementing such a system, jurisdictions could require the tobacco industry to bear the cost. Any tracking and tracing system should be compliant with Article 5.3 of the FCTC, and governments should ensure that the system is independent from the tobacco industry. While the objectives of fiscal markings and tracking and tracing systems are different, stamps increasingly contain tracking and tracing features. • Implementing legal measures to prevent forestalling can limit the delay of a tax increase and its intended effect on revenues and consumer behaviour. Forestalling, stockpiling or front-loading occur when manufacturers or im- porters increase their tax-paid stock or oversupply the market by increasing production or imports before a tax increase in order to pay the previous lower rate. • Periodic audits and controls can be implemented to increase compliance. These include cost audits, transfer price audits, price and market monitoring, consumer controls and cross-check controls. • Import and export of tobacco products and manufacturing equipment should be allowed only for duly licensed natural persons or legal entities. The risk of loss of revenue can be mitigated by requiring a guarantee or bond that will be released only if payment of duties in another country is proven. No tobacco product should be allowed into a jurisdiction unless required fiscal markings (such as tax stamps or export labels) are affixed on the pack, according to the law. Tobacco products for export should bear a marking indicating that the product is destined for the export market. Exchange of information between jurisdictions on the movement of goods can also reduce the risk of evasion. 160 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Non-invasive detection equipment (such as X-ray scanners) can be used at customs posts to detect contraband merchandise. A cheaper alternative is the use of dogs that are trained to detect cigarettes and other organic products. Many countries use both scanners and dogs to detect contraband tobacco products. Special physical control measures can also be applied to reduce contraband. Such measures include the separation of processing operations from the sealed storage of taxed and untaxed products. Within a country, mobile excise control units are helpful in verifying excis- able goods as they are transported domestically. These units should be dispatched to important transport corridors, communication centres and areas of congestion, such as bridges, ferries and passes. Physical control operations require close coordination between police, border guards and other public services. • Controls such as regulation and oversight are usually less strict in free zones and transhipment points. This can make free zones appealing to persons involved in illegal cigarette manufacturing or trade. Customs administrations should exercise their authority in free zones to effectively identify and fight illicit trade in tobacco products. Relevant measures include licensing, due diligence and recordkeeping for all operators within free zones, as well as implementing tracking and tracing regimes and removing exemptions from excise taxes. Other actions include the prohibition of intermingling of tobacco products with non-tobacco products in a single container or other similar transportation unit when the products are removed from free zones. Sale of tax-free or duty-free tobacco products to international travellers should be prohibited, as these sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and also adversely affect govern- ment revenues by creating a loophole in the tax structure. Procedures after detection of illicit trade of tobacco products should be clearly defined. If smuggling or illicit trade is detected through audits, tracking and tracing systems, verification of declarations or border control, actions such as seizing and destroying smuggled and/or illicit tobacco and collecting due taxes must be taken immediately. It is also important that penalties and sanctions be sufficient to deter illegal activities. Low financial penalties may simply be paid as a cost of doing business while the illegal activity continues. The minimum penalty for consumers in possession of illicit tobacco products should be confiscation and destruction of the products found in their possession and required payment for the unpaid tax and duties on those products. Most countries have adopted legislation to combat organized crime and money laundering. Some countries are taking advantage of this type of legislation and using it to address illicit trade of tobacco as well. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 161 In principle, administering tobacco taxes on tobacco products other than cigarettes is similar to administering them on cigarettes. The challenge for taxation of other products includes the lack of standardization of those products and sometimes large informal markets. Knowledge of the product and the supply chain greatly help to facilitate effective tax administration. The trade in raw tobacco and small-scale home production of RYO and other products such as bidis often takes place outside of monitoring and control systems. The best way to address this challenge is to enforce prior approval for purchase or sale of raw materials and a requirement to register, obtain an authorization or license all operators and growers handling raw tobacco. In principle, adding new and emerging nicotine and tobacco products to an exist- ing tax framework is not expected to impose significant costs. It is reasonable to expect that similar challenges will be faced in the collection of taxes on these newer products, as market players will attempt to use the current loopholes in tax regulation to avoid or evade taxes on these products whenever possible. However, challenges are expected to arise, as newer products involve rapidly changing technology, and their market dynamics are widely unknown. Furthermore, additional capacity may be required, since a country may need to identify new agencies or authorities, given the different characteristics of the taxable items and the tax base. Because the newer nicotine and tobacco products are widely purchased online, countries deciding to tax these products should draw up a proper implementation plan that includes rules on how taxes will be collected on imported products and online sales. Online cross-border sales are not permitted in some countries. The elements of a good tax system include (1) proper resourcing of competent authorities sufficient for hiring the necessary staff to properly implement and en- force excise tax laws; (2) implementation of tax laws with integrity and with strict rules and regulations to detect corruption and for the punishment of both agency personnel and taxpayers who are engaged in corrupt practices; and (3) ensuring that the judicial system is honest and independent in fact and in perception. Disputes should be solved rapidly – not in years, as is the case in some countries. 162 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N REFERENCES 1. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 2. WHO FCTC guidelines for implementation of Article 6 Price and tax measures to reduce the demand for tobacco. Geneva: World Health Organization; 2014 (https://www.who.int/fctc/treaty_instruments/ Guidelines_Article_6_English.pdf?ua=1, accessed 5 October 2020). 3. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 4. Joossens L, Raw M. From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–234 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 5 October 2020). 5. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 6. Reyes IP. Measures to control the tobacco supply chain in the ASEAN: a review of excise tax stamps, track and trace mechanisms, and licensing systems in the ASEAN region. Bangkok: Southeast Asia Tobacco Control Alliance; 2014 (https://seatca.org/dmdocuments/SEATCA%20MEASURE%20BOOK%20. pdf, accessed 5 October 2020). 7. Italy’s tax administration: a review of institutional and governance aspects. Paris: Organisation for Economic Co-operation and Development; 2016 (https://www.oecd.org/tax/administration/italy- tax-administration-a-review-of-institutional-and-governance-aspects.pdf, accessed 5 October 2020). 8. Crandall W, Gavin E, Masters A. ISORA 2016: understanding revenue administration. Washington (DC): International Monetary Fund; 2019 (Departmental Paper Series No. 19/05; (https://www.imf.org/en/ Publications/Departmental-Papers-Policy-Papers/Issues/2019/03/07/ISORA-2016-Understanding- Revenue-Administration-46337, accessed 5 October 2020). 9. Dutch customs in 2018. Heerlen: Customs Administration of the Netherlands; 2019 (https://over-ons. belastingdienst.nl/wp-content/uploads/2019/04/Dutch-Customs-in-2018.pdf, accessed 8 October 2020). 10. ESAs transform the way competent authorities cooperate with each other on AML/CFT matters [press statement]. European Banking Authority; 16 December 2019 (https://eba.europa.eu/esas-transform- way-competent-authorities-cooperate-each-other-amlcft-matters, accessed 5 October 2020). 11. Illicit trade report 2013. Brussels: World Customs Organization; 2014 (http://www.wcoomd.org/-/ media/wco/public/global/pdf/topics/enforcement-and-compliance/activities-and-programmes/ illicit-trade-report/itr_2013_en.pdf?la=en, accessed 5 October 2020). 12. Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation and repealing Directive 77/799/EEC. Brussels: OJEU; 2011. (L 64/1; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:32011L0016&from=en, accessed 29 January 2021). 13. 439 stolen cars seized in 12 days as part of Frontex-led operation. Frontex News Release; 14 October 2019 (https://frontex.europa.eu/media-centre/news-release/439-stolen-cars-seized-in-12-days-as- part-of-frontex-led-operation-vvvirI, accessed 5 October 2020). 14. Focus on: the illicit trafficking of counterfeit goods and transnational organized crime. Vienna: United Nations Office on Drugs and Crime; 2014 (https://www.unodc.org/documents/counterfeit/FocusSheet/ Counterfeit_focussheet_EN_HIRES.pdf, accessed 5 October 2020). 15. Crandall W. Revenue administration: performance measurement in tax administration. Washington (DC): International Monetary Fund; 2010 (Technical Notes and Manuals 10/11; https://www.imf.org/ external/pubs/ft/tnm/2010/tnm1011.pdf, accessed 7 October 2020). 16. Measuring tax gaps 2019 edition – tax gap estimates for 2017-18. London: HM Revenue and Customs; 2019 (https://webarchive.nationalarchives.gov.uk/20200701215139/https://www.gov.uk/government/ statistics/measuring-tax-gaps, accessed 29 January 2021). 17. Countering illicit trade in tobacco products: a guide for policy-makers. Lyon: INTERPOL; 2014 (Legal handbook series; https://www.aph.gov.au/DocumentStore.ashx?id=704d9168-2aa7-4df4-bb81- CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 163 0fe1f357dfa6&subId=408832, accessed 29 January 2021). 18. Resolução- RDC No. 226 de 30 de Abril de 2018 [Resolution of April 30, 2018]. Ministério da Saúde/ Agência Nacional de Vigilância Sanitária/Diretoria Colegiada; 2018 (in Portuguese) (https://www.in.gov. br/materia/-/asset_publisher/Kujrw0TZC2Mb/content/id/12527411/do1-2018-05-02-resolucao-rdc- n-226-de-30-de-abril-de-2018-12527407, accessed 17 February 2021). 19. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 20. Preece R. Addressing the illicit flow of tobacco products in Australia. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 21. Excise on tobacco [website]. Canberra: Australian taxation office; 2019 (https://www.ato.gov.au/ business/excise-on-tobacco/, accessed 5 February 2021). 22. Safe Framework of Standards. Brussels: World Customs Organization; 2018 (http://www.wcoomd. org/-/media/wco/public/global/pdf/topics/facilitation/instruments-and-tools/tools/safe-package/ safe-framework-of-standards.pdf?la=en, accessed 29 January 2021). 23. The Trade Facilitation Agreement. Geneva: World Trade Organization; 2017 (https://www.wto.org/ english/docs_e/legal_e/tfa-nov14_e.htm, accessed 29 January 2021). 24. Authorised economic operator (AEO) [website]. Kildare: Irish Tax and Customs; 2020 (https://www. revenue.ie/en/customs-traders-and-agents/authorised-economic-operators/index.aspx, accessed 7 October 2020). 25. Performance outcome area 1: integrity of the registered taxpayer base. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www. tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 26. Commission implementing regulation (EU) 2016/323 of 24 February 2016 laying down detailed rules on cooperation and exchange of information between member states regarding goods under excise duty suspension pursuant to Council Regulation (EU) No 389/2012. Brussels: OJEU; 2018 (http://data. europa.eu/eli/reg_impl/2016/323/2018-02-15, accessed 7 October 2020). 27. Council Directive 2008/118/EC concerning the general arrangement for excise duty repealing Directive 92/12/EEC. Brussels: OJEU; 2009 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:3 2008L0118&from=en, accessed 29 January 2021). 28. Our strategy. London: HM Revenue & Customs, 2017 (https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/629941/HMRC-Strategy.pdf, accessed 31 January 2021). 29. Strategic plan FY2018-2019. Washington (DC): Internal Revenue Service, 2017 (https://www.irs.gov/ pub/irs-pdf/p3744.pdf, accessed 31 January 2021). 30. Performance outcome area 2: effective risk management. In: Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/ assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 31. Tax administration 2017: comparative information on OECD and other advanced and emerging economies. Paris: Organisation for Economic Co-Operation and Development; 2017. 32. Compliance risk management: managing and improving tax compliance. Paris: Organisation for Economic Co-Operation and Development; 2017 (https://www.oecd.org/tax/administration/33818656. pdf, accessed 7 October 2020). 33. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 34. Evans P. U.K. Fines British American Tobacco for Oversupplying Cigarettes in Belgium. The Wall Street Journal. 13 November 2014 (https://www.wsj.com/articles/u-k-fines-british-american-tobacco-for- oversupplying-cigarettes-in-belgium-1415873470, accessed 31 January 2021). 35. Doward J. BAT fined for oversupplying tobacco in low-tax European jurisdictions. The Guardian. 164 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 16 November 2014 (https://www.theguardian.com/business/2014/nov/16/bat-fined-for-oversupplying- tobacco-in-low-tax-european-jurisdictions, accessed 31 January 2021). 36. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021) 37. Cour des comptes. Accises sur les produits du tabac, Bruxelles [Court of Auditors, Excise duties on tobacco products, Brussels]. Bruxelles: Cour des comptes; 2015 (in French) (https://www.ccrek.be/ docs/2015_27_AccisesProduitsDuTabac.pdf, accessed 7 October 2020). 38. Chaloupka FJ, Edwards SM, Ross H, Diaz M, Kurti M, Xu X, et al. Preventing and reducing illicit tobacco trade in the United States. Atlanta: Centers for Disease Control and Prevention; 2015 (https://www. cdc.gov/tobacco/stateandcommunity/pdfs/illicit-trade-report-121815-508tagged.pdf, accessed 29 January 2021). 39. The Tax Stamp Forum 2015. Hexham: Reconnaissance International (https://10times.com/tax-stamp- forum-miami, accessed 10 September 2020). 40. Tax stamps: a technical study and market report. London: Reconnaissance international; 2012. 41. ЕАНС/2013/Health/ll concerning the provision of an analysis and feasibility assessment regarding EU systems for tracking and tracing of tobacco products and for security features. Brussels: European Commission, 2015 (https://ec.europa.eu/health/sites/health/files/tobacco/docs/2015_tpd_tracking_ tracing_frep_en.pdf, accessed 29 January 2021). 42. Ross H. Controlling illicit tobacco trade: international experience. Economic research informing tobacco control policy. Cape Town: University of Cape Town; 2015 (https://tobacconomics.org/ uploads/misc/2015/05/Ross_International_experience_05.28.15.pdf, accessed 18 December 2020). 43. Security and resilience — authenticity, integrity and trust for products and documents — guidelines for the content, security, issuance and examination of excise tax stamps. Geneva: International Organization for Standardization; 2018. 44. Borkowski F, Twomey C. European Union: an update on EU policies. In: Dutta S (editor). Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959-REPL-PUBLIC-6- 2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 29 January 2021). 45. Quality management systems — Fundamentals and vocabulary. ISO 9001:2015(en). Geneva: International Organization for Standardization; 2015 (https://www.iso.org/obp/ui/#iso:std:iso:9001:ed- 5:v1:en, accessed 16 February 2021). 46. Protocol to Eliminate Illicit Trade in Tobacco Products: questions & answers. Geneva: World Health Organization; 2019 (https://www.who.int/fctc/protocol/faq/en/, accessed 7 October 2020). 47. Ross H, Eads M, Yates M. Why governments cannot afford Codentify to support their tracking and tracing solutions. Tob Control. 2018;27(6):706–08. 48. Ser vicio de Impuestos Internos, Chi le, Noticias, 28 May 2019. (http://www.si i .c l/ noticias/2019/280519noti01er.htm, accessed 20 October 2020). 49. Directive 2014/40/EU of the European Parliament and of the Council. OJEU. 2014;127:1–38 (https:// ec.europa.eu/health/sites/health/files/tobacco/docs/dir_201440_en.pdf, accessed 7 October 2020). 50. Report on the EU customs enforcement of intellectual property rights: results at the EU border. Luxembourg: Publications Office of the European Union, 2019 (https://ec.europa.eu/taxation_customs/ sites/taxation/files/2019-ipr-report.pdf, accessed 7 October 2020). 51. Matthews P. Forestalling ahead of property tax changes. London: Office for Budget Responsibility; 2016 (OBR Working Paper No. 10; https://obr.uk/docs/dlm_uploads/Working-paper-No.10-1.pdf, accessed 29 January 2021). 52. Ross H, Tesche J, Vellios N. Undermining government tax policies: common strategies employed by the tobacco industry in response to tobacco tax increases. Prev Med. 2017;105S:S19–S22 (https://www. sciencedirect.com/science/article/pii/S0091743517302165?via%3Dihub, accessed 17 February 2021). 53. European Commission v Portugal supported by Belgium, Estonia and Poland, Case C-126/15. Judgement of the Court (Fifth Chamber); 29 June 2017 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?u ri=CELEX:62015CJ0126&from=EN, accessed 31 January 2021). CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 165 54. Amalia Valesko v Zollamt Klagenfurt, Case C-140/05. Judgement of the Court (Second Chamber); 5 October 2006 (http://curia.europa.eu/juris/showPdf.jsf;jsessionid=02DEE17CC11845BAE0F42FC7 CFD24C85?text=&docid=65599&pageIndex=0&doclang=en&mode=lst&dir=&occ=first&part=1&c id=2591459, accessed 29 January 2021). 55. European Commission v French Republic, Case C-197/08. Judgement of the Court (Third Chamber); 4 March 2010 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:62008CJ0197&qid =1612223322830&from=EN, accessed 31 January 2021). 56. Commission requests that Belgium amend its legislation on tobacco: letter of formal notice. Brussels: European Commission; 8 November 2018. (Memo 18/6247; https://ec.europa.eu/commission/ presscorner/detail/EN/MEMO_18_6247, accessed 29 January 2021). 57. Commission requests that Belgium amend its legislation on tobacco, MEMO/18/6247 of 8 November 2018. 58. Pederson H, Floristean A, Iseppi L, Dawkins R, Smith C, Mørup C, et al. Study on the measuring and reducing of administrative cost for economic operators and tax authorities and obtaining in parallel a higher level of compliance and security in imposing excise duties on tobacco products. Brussels: European Commission; 2014 (https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/ ramboll-tobacco-study.pdf, accessed 3 October 2020). 59. Money laundering vulnerabilities of free trade zones. Paris: Financial Action Task Force; 2010. https:// www.fatf-gafi.org/media/fatf/documents/reports/ML%20vulnerabilities%20of%20Free%20Trade%20 Zones.pdf, accessed 7 October 2020). 60. Omi K. ‘Extraterritoriality’ of free zones: the necessity for enhanced customs involvement. World Customs Organization; 2019 (WCO research paper no. 47; http://www.wcoomd.org/-/media/wco/public/global/ pdf/topics/facilitation/ressources/permanent-technical-committee/225-226/itemixb_fzs_wco_e. pdf?la=en, accessed 29 January 2021). 61. European Parliament resolution of 26 March 2019 on financial crimes, tax evasion and tax avoidance. European Parliament; 2019 (2018/2121(INI)); http://www.europarl.europa.eu/doceo/document/TA- 8-2019-0240_EN.html, accessed 7 October 2020). 62. International Convention on the Simplification and Harmonization of Customs Procedures. Brussels: World Customs Organization; 2008 (http://www.wcoomd.org/Topics/Facilitation/Instrument%20 and%20Tools/Conventions/pf_revised_kyoto_conv/Kyoto_New, accessed 7 October 2020). 63. Evidence Notice 2004: tobacco duty – tobacco products manufacturing machine licensing scheme. London: HM Revenue & Customs, 2018 (https://www.gov.uk/government/publications/excise-notice- 2004-tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme/excise-notice-2004- tobacco-duty-tobacco-products-manufacturing-machine-licensing-scheme, accessed 31 January 2021). 64. Tackling illicit tobacco: from leaf to light: the HM Revenue & Customs and Border Force strategy to tackle tobacco smuggling. London: HM Revenue and Customs; 2015 (https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/418732/Tackling_illicit_ tobacco_-_From_leaf_to_light__2015_.pdf, accessed 7 October 2020). 65. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 66. Cardenas M. Case studies in illicit tobacco trade: Colombia. Chicago: University of Illinois at Chicago; 2020 (Tobacconomics Fact Sheet; https://tobacconomics.org/files/research/606/UIC_Colombia-Illicit- Trade-Fact-Sheet_v1.4.pdf, accessed 29 January 2021). 67. Maldonado N, Llorente BA, Escobar D, Iglesias RM. Smoke signals: monitoring illicit cigarettes and smoking behaviour in Colombia to support tobacco taxes. Tob Control. 2019;29:s243-s248 ( https:// tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/s243.full.pdf, accessed 29 January 2021). 68. Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat; April 2019 (https://www.tadat.org/assets/files/IMF_TADAT-FieldGuide_web2.pdf, accessed 29 January 2021). 69. Pellechio A, Tanzi V. The reform of tax administration. Washington (DC): International Monetary Fund; 1995 (Working Paper no. 95/22; https://www.elibrary.imf.org/view/IMF001/07242- 9781451843941/07242-9781451843941/07242-9781451843941_A001.xml?language=en&redirect=true, accessed 29 January 2021). 166 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.1 COMPOSITION OF TOBACCO PRODUCTS To implement and enforce tobacco taxes in the most efficient way, competent authori- ties should be familiar with all of the components of tobacco products, including each of the raw materials used in their manufacture, production inputs and tobacco manufacturing machinery. Knowledge of the components of excisable products and machinery provides valuable information to identify activities at high risk for non- compliance, implement measures to ensure all taxes are paid and prevent illicit trade. Nicotine, non-nicotine and tobacco products and their component parts In most countries, the ministry of finance determines tax policy, including which tobacco products are taxed, while the ministry of health is responsible for product and use regulation. This could lead to different definitions of the same product, depending on which ministry is responsible for a given law or regulation. Wherever possible, a clear and common definition should be developed to simplify procedures and avoid confusion. Tobacco products take various forms, and not all may be regulated or subject to excise tax in a specific jurisdiction. In addition to cigarettes, other traditional tobacco products include smokeless tobacco – such as chewing tobacco, snuff and snus – as well as bidis and kreteks (clove cigarettes), which can be hand-rolled or manufactured, pipes, hookah or waterpipe and cigars. Cigarettes It is important to understand the materials and component parts of the tobacco products most commonly used in a particular country. Cigarettes are the most common and significant tobacco products in terms of volume and tax revenues in most jurisdictions. A cigarette stick is composed of: • the tobacco blend of various types of tobacco plant (leaves and stem and other plant parts) and additives (including flavours); • the cigarette paper used to wrap the tobacco blend to make up the tobacco rod; • the acetate filter that forms the white portion at the tip of a filtered cigarette, which is in direct contact with the smoker’s mouth; • the tipping paper or wraps around the filter; and • the adhesive that secures the cigarette paper around the tobacco blend and the tipping (1). Each manufacturer follows a specific process to produce cigarettes. Aside from the tobacco blend, manufacturers also vary the size of cigarette paper and tipping paper and the length of acetate filter used per stick (1). In some countries, these elements are standardized. In an ideal regulatory framework, a manufacturer would CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 167 be required to submit information on the specific process for each brand and variant of the tobacco product that it manufactures to the competent authorities as part of the licensing requirement (see section 3.4.3). Authorities could, for example, require manufacturers to submit this information in order to obtain a licence. The minimum requirements of the administration and the information that should be included could be laid down in law or lower regulation to ensure that authorities have the information in their possession for all licensed manufacturers. This information contributes to verifying whether a company is reporting the actual quantity of cigarettes manufactured for sale and sold cigarettes by comparing the amount of materials used for production and the quantities used per cigarette with the total number of manufactured cigarettes. Eventually, this information also contributes to validating whether the taxes are properly paid. Figure A3.1 shows the component parts of a typical machine-made traditional cigarette. Fig. A3.1 Component parts of a machine-made cigarette Source: Author’s compilation. Photo by Walter Klerx. Not all parts of tobacco products are subject to the same level of control. According to Article 6.5 of the Protocol, five years following the entry into force of this Protocol, the MOP shall ensure at its next session that evidence-based research is conducted to ascertain whether any key inputs exist that are essential to the manufacture of tobacco products, are identifiable and can be subject to an effective control mechanism. On the basis of such research, the MOP shall consider appropriate action. In addition to the component parts of tobacco products, materials needed for packaging a specific number of sticks into a pack of cigarettes, usually 20 per pack, can be monitored. These materials include the foil paper, the package paper (which could bear the brand name, design and health warnings), the fiscal marking (if required) and the plastic or cellophane wrap. A fixed number of packs of cigarettes, Filter Tipping paper Cigarette paper Tobacco Tobacco rodFiltration zone 168 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N normally 10 packs, are packed into cartons, also called reams. These cartons are usu- ally made of soft paperboard or cardboard, possibly with branding, and are wrapped in plastic wrap or cellophane. Fifty cartons are packed in master cases, which are made of sturdier and thicker paperboard and stacked on pallets (usually 50 master cases to a pallet). An effective regulatory framework would require manufacturers and importers to provide information to the competent authorities on packaging and design, as well as the number of sticks per pack, carton and master case. Prior approval for purchase or sale of materials used in the cigarette production process can also be required. In the Philippines, suppliers of such raw materials, including those providing tobacco papers and filter components, are required to have a licence (2). In some of the Member States of the EU, raw tobacco is also subject to fiscal and legal requirements. For example, in Slovakia and Poland, raw tobacco can be handled only by authorized operators. While authorized operators do not have to pay excise duties on raw tobacco, if raw tobacco is detected by an unauthorized operator, excise duties will be due. Hungary, Italy and the United Kingdom require registration or authorization for all operators and growers handling raw tobacco (3). In addition to knowing the quantities of inputs required to produce a specific amount of a regulated product (e.g. cigarettes), the competent authority also needs to understand the supply, manufacturing and distribution chains to be able to properly monitor, regulate and determine whether taxes have been paid (see also Fig. 3.3). Novel and emerging nicotine, non-nicotine and tobacco products In recent times, new products have been introduced to several markets, namely, ENDS, ENNDS and HTPs. ENDS usually comprise a nicotine-containing e-liquid but do not contain to- bacco. ENNDS are essentially the same but do not (ostensibly) contain nicotine. The WHO COP requested the Convention Secretariat to invite Parties to monitor and report on scientific, regulatory and market developments such as initiation, cessation, advertising and promotion of ENDS and ENNDS. Furthermore, the COP requested WHO to report on the development of methods by regional and international standards-development organizations for the testing and measuring of contents and emissions of these products (4). There are different types of e-cigarettes – the most common type of ENDS and ENNDS – and currently there are four generations of products. However, they can be divided into two broad categories: open systems and closed systems. Both types of e-cigarette use a wick and a heat source to generate an aerosol. The wick is saturated with e-liquid, and a microprocessor is used to control operations (not all include this). Some e-cigarettes also have an LED light to imitate the burning end of a conventional cigarette (5). Fig. A3.2 presents examples of open and closed systems. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 169 E-liquid pod cover Rechargeable battery Heating element (heats solution, aerosolizing nicotine) Mouthpiece E-liquid pod Fig. A3.2 Examples of open and closed systems of ENDS/ENNDs products Open ENDS/ENNDS system (e-cigarette) Closed ENDS system (e-cigarette) Source: (6). Unlike ENDS/ENNDS, HTPs do contain tobacco. HTPs produce aerosols containing nicotine and toxic chemicals when tobacco is heated or when a device containing tobacco is activated (7). HTPs are composed of two elements: the sticks or pods that contain the tobacco and the device used to heat the tobacco. Both are necessary for the product to be used. Fig. A3.3 shows an example of a heated tobacco product. HTPs are tobacco products and are therefore subject to the regulatory measures contained in the WHO FCTC. Rechargeable battery Power button (to start vaping) Mouthpiece Atomizer / Heating element (heats solution, aerolizing nicotine) E-liquid tank (refillable e-liquid nicotine tank) 170 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. A3.3 Components of an HTP Holder: Tobacco stick: Note: PLA: polyactic acid, MPF: mouthpiece filter. Sources: (8-9). More information on tax administration of other tobacco products is presented in section 3.5. Tobacco stick Holder Charger Casing Control Electronics Battery Heating Blade Heatstick total lenght: 45mm MPF (7 mm) PLA (18 mm) Tobacco plug (12 mm) Hollow acetate tube (8 mm) Outer paper Diameter max 7.42 mm Tipping paper CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 171 REFERENCES 1. Memorandum by Gallaher Group Plc: the tobacco industry and the health risks of smoking. Select Committee on Health Minutes of Evidence. London: UK Parliament, 2000 (https://publications. parliament.uk/pa/cm199900/cmselect/cmhealth/27/0011323.htm, accessed 7 October 2020). 2. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 3. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco.. Brussels: European Commission; 2017 (https://ec.europa.eu/taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 4. Electronic nicotine delivery systems and electronic non-nicotine delivery systems. Geneva: World Health Organization; 2016 (Decision FCTC/COP7(9) of the Conference of the Parties to the WHO Framework Convention on Tobacco Control; https://www.who.int/fctc/cop/cop7/FCTC_COP7_9_EN.pdf?ua=1, accessed 7 October 2020). 5. Brown CJ, Cheng JM. Electronic cigarettes: product characterization and design considerations. Tob Control. 2014;23:ii4-ii10 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3995271/pdf/ tobaccocontrol-2013-051476.pdf, accessed 31 January 2021). 6. E-cigarettes and vapor products [webpage]. King County, Washington (USA); 2019 (https://www. kingcounty.gov/depts/health/tobacco/data/e-cigarettes.aspx, accessed 9 October 2020). 7. Heated tobacco products. Geneva: World Health Organization; 2020 (WHO/HEP/HPR/2020.2 Information sheet; https://apps.who.int/iris/bitstream/handle/10665/331297/WHO-HEP-HPR-2020.2- eng.pdf?sequence=1&isAllowed=y, accessed 7 October 2020). 8. Tobacco heating system (IQOS) briefing document. Silver Spring: US Federal Drug Administration; 2018 (https://www.fda.gov/media/110377/download, accessed 31 January 2021). 9. Premarket tobacco product application: technical project lead review. Silver Spring: US Federal Drug Administration; 2017 (https://www.fda.gov/media/124247/download, accessed 7 October 2020). 172 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N ANNEX 3.2 EXAMPLE OF FORESTALLING AND COUNTERMEASURES This example is hypothetical but inspired from the situation in the Philippines. The amounts and prices have been altered, however, and it is assumed that the normal inventory kept by a tobacco manufacturer is two months.  The excise tax imposed on a pack of cigarettes in the current year is US$ 3.00; it will be increased to US$ 3.30 at the beginning of the new fiscal year (January in this example). The monthly production of Brand Y cigarettes of company X, which it declares for tax purposes, is as follows: MONTH OF THE CURRENT YEAR PACKS OF CIGARETTES January 10 000 000 February 10 500 000 March 9 900 000 April 11 000 000 May 10 200 000 June 10 600 000 July 9 700 000 August 10 100 000 September 9 900 000 October 10 100 000 November 20 000 000 December 25 000 000 Since the normal inventory is two months, the quantity in the two months prior to the implementation of the new excise tax rate is disregarded. The shelf life of tobacco products is approximately six months. The average of the six months prior to November is computed to obtain the quantity presumed to be produced or imported if there was no tax increase. The quantity from May to October (inclusive) divided by 6 is 10 100 000 packs. Thus, any quantity produced beyond 10 100 000 packs for the months of November and December (the months prior to the implementation of the new tax rate) is assessed using the new tax rate. In this example, 10 100 000 of the packs produced in November will be taxed at the old rate of $3.00, and 9 900 000 packs will be taxed at the new rate of $3.30. For December, 10 100 000 packs will be taxed at $3.00, while 14 900 000 packs will be taxed at $3.30. Without imposing these measures, the government would have been deprived of the excise tax increase on 24 800 000 packs. In addition, the effect of the increase on prices and consumers would have been delayed by approximately two months. CHAP T ER 3. TO BACCO TA X ADM INIS T R AT I O N 173 In countries using tax stamps, the withholding of the issuance of stamps is a well-known approach to counter forestalling. Another practical solution is to allow the competent authorities to request advances from the industry to cover revenue shortfalls, provided there is a legal basis for such requests. 174 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 175 CHAPTER 4. Political economy As with any proposed government action, policy-makers need to navigate the political environment of tobacco taxation at every stage of policy development, implementa- tion and administration. While every country’s distinct history, culture, systems and structural forces shape its unique political landscape, there are some universal themes when it comes to tobacco control and particularly tobacco taxation. These themes boil down to the distribution of money, power and resources. The tobacco industry, as both a political and economic player, understands these themes well. The industry has been effective in using principles of the political economy of tobacco taxation in its efforts to block important advancements in tobacco control. Nevertheless, the savvy policy-maker can see through the industry arguments by considering who benefits from industry-favoured policy measures and interven- tions. The industry’s challenges to tobacco tax policies can be organized into the five categories of SCARE tactics. This chapter provides a road map to help policy-makers navigate the political economy of tobacco taxation through each of these themes. The first five sections dissect the tobacco industry framing of each issue, pinpoint- ing the flaws in each argument, identifying the extent to which each concern has merit and suggesting how a responsible government can address each one. These discussions are supported by unbiased evidence from independent, peer-reviewed research, as well as specific examples from country experiences. Sections 4.1 through 4.5 on SCARE tactics will equip policy-makers with the tools they need to proceed with confidence that their tobacco tax policy – developed and implemented fol- lowing the guidelines spelled out in this technical manual – will bring about the greatest health and economic benefits for their constituents, regardless of industry attempts to thwart them. Section 4.6 further buoys policy-makers’ efforts to ensure the beneficial impacts of their policies, as it describes how earmarking can improve the political economy of tobacco taxation by funding programmes and initiatives that promote and support the health and well-being of the population. 176 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.1 SCARE TACTIC S: SMUGGLING AND ILLICIT TRADE 4.1.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases will inevitably result in an increase in the illicit trade in tobacco products (1–2). They claim that higher tax rates and higher prices strengthen the financial incentives for criminal enterprises to supply cigarettes from lower-tax jurisdictions, boost domestic tax evasion and encourage smokers to seek cheaper illegal cigarettes. The industry also challenges the argument that tobacco tax hikes increase government revenue by claiming that the presence of an illicit tobacco market will actually reduce revenue collection following a tax increase. More recent versions of this argument – adapted to address public health concerns about tobacco use – claim that illicit market growth also offsets reductions in smoking prevalence that would otherwise be brought about by tobacco tax increases. In summary, the tobacco industry and its allies claim that raising tobacco taxes are ineffective – and even counterproductive – because they are circumvented by illicit markets, which prevents the government from achieving its public health objective and reduces rather than increases tax revenues. When a country considers a proposal to increase tobacco excises, the tobacco industry and its allies frequently make exaggerated claims about the size and scope of illicit tobacco trade in that country. Opponents of tobacco tax increases argue that price differentials are the exclusive – or at least the dominant – cause of illicit trade. Influenced by this fear-inducing faulty diagnosis, tax authorities frequently find it difficult to make decisions about tobacco taxes. However, the industry diagnosis always contains the same erroneous elements. First, the illicit trade in a country is frequently less than the industry portrays it to be, and the country’s tax enforcement policy towards tobacco products is rarely unique or in any way different from the norm in the country (3). Second, the scale of illicit trade in tobacco is not exclusively or even primarily determined by tax or price differentials. Typically, it results from a set of governance problems characterized by government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, lack of dissuasive sanctions and/or weak norms regarding participation in illegal and informal markets (4–7). This section provides guidance for tax and other relevant authorities on how to respond to the tobacco industry SCARE tactic that increasing tobacco taxes will lead to smuggling and illicit trade in their countries. Tax authorities need to know the nature, causes and extent of illicit trade so that they can define the problem properly and formulate an appropriate response. This chapter addresses the available tools for better defining and understanding specific illicit trade problems, particularly tools that facilitate independent assessment of the magnitude of that trade. Improvements CHAP T ER 4. PO LI T I C AL ECO N OMY 177 to governance within the realm of tax authorities – such as best practices in tobacco tax administration and policies to improve the effectiveness of fiscal regulations or norms regarding participation in informal and illegal markets – are discussed in Chapter 3. This section first describes the nature of the illicit tobacco trade to highlight some of its complexities and identify complementary policies for tackling the problem. Next, evidence that calls into question the link between illicit trade and high prices or tax rate changes is discussed. Finally, to help tax authorities assess their own situation, several different methodologies are presented to estimate the scope of the illicit tobacco trade and to evaluate estimates of that trade for a particular country or tax jurisdiction. 4.1.2 THE NATURE AND EXTENT OF THE ILLICIT TOBACCO TRADE The WHO FCTC defines illicit trade as: any practice or conduct prohibited by law and which relates to production, ship- ment, receipt, possession, distribution, sale or purchase, including any practice or conduct intended to facilitate such activity (8). Non-duty-paid tobacco products found in a jurisdiction (i.e. through littered-pack surveys) could be the result of either of two related but distinct activities: tax evasion and tax avoidance. Tax evasion is a set of unlawful actions seeking the non-payment of tobacco taxes and duties, whereas tax avoidance comprises legal actions with the purpose of avoiding payment of some or all taxes, such as bringing an amount of cigarettes up to the legal allowance from a lower- into a higher-tax jurisdiction. Tax avoidance is not illegal and is therefore not considered part of illicit trade in tobacco products.1 The focus of this section is on tax evasion activities, which can occur in the movement across borders or in domestic production and distribution. When tax evasion happens across borders, it is known as smuggling (9) and can be done on a large scale or a small scale. Tax evasion in the domestic market can be partial, when licensed and authorized producers or distributors comply with only part of their tax obligations, or total, when the whole production and distribution system is illegal and out of sight of tax administrators (5, 10–14).2 Large-scale tax evasion schemes can be run by different types of producers and their associated distributors, such as the transnational tobacco companies (TTCs) and their national subsidiaries, 1 Tax avoidance practices –common among states in the United States and countries in the EU – are not analysed in this section. 2 Tax evasion is normally considered as illicit manufacturing in the literature (14). 178 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N other local tobacco companies producing their own brands and illegal factories that normally counterfeit other brands or produce illicit brands. Large-scale smuggling involves, for example, taking advantage of tax-free zones and mislabelling shipping products prior to or during transit (11) or using so- phisticated clandestine networks. This form of tax evasion is systemic and can be carried out by TTCs (12), by local companies producing in countries with low tax enforcement, e.g. Paraguay (15) that feed neighbouring countries and regional illicit hubs through a network of clandestine distributors (15) or by companies located in tax-free zones like Jebel Ali and Dubai in the UAE (16) or such zones in Russia or Cyprus (17). The origins, routes and quantities of large-scale smuggling frequently change as affected countries or markets react by strengthening enforcement and seeking bilateral cooperation with the jurisdictions of origin. For example, the 2013 comprehensive strategy of the EU (10) enhanced bilateral cooperation with major source and transit countries of illicit cigarettes coming to Europe, including Russia, Belarus and Ukraine. These agreements improved day-to-day cross-border coopera- tion, reduced illegal flows and introduced gradual tobacco excise rate adjustments in those three countries to bring them to European levels. Small-scale smuggling (also known as ant smuggling or bootlegging) is the cross-border trafficking of cigarettes in quantities that are larger than the allowable limits (e.g. two cartons) but smaller than large shipments (e.g. truckloads, cargo containers), normally for the purpose of selling at a profit (11). This type of illicit trade may exist in places where there are opportunities within neighbouring tax jurisdictions. For example, small-scale smuggling is commonly done by individuals living in French and German provinces near lower-taxed countries (e.g. Belgium, Luxembourg, Switzerland, Spain, Poland and Czechia) (18). Counterfeiting is a form of illicit manufacturing that involves the production of tobacco products (including packaging and tobacco filler) without the approval of the trademark holder (13). Another product of illicit manufacturing is so-called cheap or illicit whites. Cheap whites are branded (e.g. Jin Ling) or unbranded cigarettes that are legally or illegally produced3 and knowingly sold in the illicit market (17). Cheap whites are not usually produced by TTCs (17, 19).4 They are produced by small tobacco produc- tion companies in one country and often sold in illegal markets of neighbouring 3 Ross et al. (17) analysed this issue and found that the sale to the first purchaser is usually legal. Their analysis covers the production in free zones (i.e. in the UAE, Russia and Cyprus) and production exported from Viet Nam, Indonesia and China. In those cases, there is no need to make the first sale illegally. However, cheap white production in Paraguay is sold to domestic distributors, and most of those sales are completely illegal. 4 Ross et al. (17) and Gilmore et al. (19) identify some cheap white brands sold by TTCs, such as President (PMI), produced in Ukraine, and Esse (Korea Tobacco & Ginseng Company, KT&G), produced in Indonesia. CHAP T ER 4. PO LI T I C AL ECO N OMY 179 countries. For example, in Paraguay, cheap whites are produced on a large scale by a few companies under the guise that they are marketed domestically, but a large share is smuggled into Uruguay and Brazil (4). Iglesias et al. (20) showed how TTCs’ cheap brands were illicitly shipped through Paraguay to be sold in the Brazilian and Argentine markets in the 1990s. This contributed to increased production of cheap whites in Paraguayan firms, which continued the illicit business even after Brazilian legislation obstructed the illegal activity of the TTCs. Domestic tax evasion is a pervasive phenomenon, particularly in LMICs. Partial tax evasion in tobacco products can be found at any level of tax rates or prices and is generally the result of defective legislation or weak tax enforcement.5 Complete or total tax evasion occurs when producers and distributors are clandestine or when there are serious institutional challenges to tax enforcement between two tax jurisdictions, such as between the United States and Native American Reservations. Evidence of illicit manufacturing has increased in recent years in several places in the world, including the EU (10) and Brazil (21). TTCs were predominant in illicit trade activity until the end of the 20th century, and even with the entrance of new actors into the illicit business, TTCs have not entirely exited. Gilmore et al. (22) analysed industry-funded data and seizure data and concluded that TTCs are still involved in illicit trade in Europe, despite the Anti-Contraband and Anti-Counterfeit Agreements (the “Agreements”) signed between the four TTCs and the EU (23).6 Using industry-funded data, Gilmore et al. show that 58% of illicit EU cigarettes can be attributed to the four main TTCs. When seizure data are used, 69% to 73% of illicit EU cigarettes can be attributed to these firms (22). It is always difficult to assess the extent of the global illicit tobacco trade because of its illegality, its global and changing nature and problems with data collection (24). Before the 21st century, when TTCs were almost unique actors in the large-scale smuggling of well-known cigarette brands, the difference between global exports and imports of cigarettes could provide a good approximation of the size of this problem globally (7). However, with the growth of illicit manufacturing in general, the manufacturing of cheap whites and the illegal movements of those products 5 This occurs when licensed and authorized producers underreport actual quantities and sell the non- duty-paid produced quantities through illegal channels. It can also include instances when producers do not report quantities at all, as in many ad valorem systems of LMICs. 6 From 2000 on, the European Commission and 10 Member States launched court cases regarding smuggling and money laundering against several TTCs. To end the court cases, the Anti-Contraband and Anti-Counterfeit Agreements were signed, which required the TTCs to exercise stringent control over their supply chain (through tracking and tracing, due diligence and anti-money-laundering and reporting obligations), share operational intelligence with Member States and the EU and pay penalties for seizures, as well as annual payments over a period of 12 years. The agreement with PMI has ended, the one with JTI will end in 2022 and the others with Imperial Tobacco and BAT will run until 2030 (23). 180 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N over the past two decades, trade statistics are no longer as useful as they were in the past. Joossens et al. (25) tried to estimate the size of the global illicit cigarette market by adding different types of national estimations prepared around 2007. They found that the estimated size was 657 billion cigarettes per year, or 11.6% of the global cigarette market. According to Joossens et al., illicit trade costs govern- ments US$ 40.5 billion in tax revenue worldwide, and eliminating illicit tobacco trade would recover US$ 13 billion in immediate revenue in high-income countries and US$ 18.3 billion in LMICs.7 Descriptions of the types of illicit trade are useful for developing the first com- ponent of a strategy to fight it: assess the nature and size of the problem. Table 4.1 presents all the main components of a strategy to fight illicit trade. To make progress in this first component – knowing the problem – authorities could use and adapt existing instruments of health surveillance or seek partnerships with academia and independent specialists to investigate the issues involved, using different methodolo- gies (see subsection 4.1.4 and Annex 4.1 on methodologies to assess the nature and size of the problem). Knowing the nature of the problem requires the cooperation of different government actors – for example, to investigate both the financial and criminal operations of organized crime behind the illicit trade. The gathering of qualitative information on the nature of the illicit trade should start simultaneously with the statistical work of measuring the magnitude of the problem. Table 4.1 Components of a strategy to fight the illicit tobacco trade 1. Assess the nature and size of the problem Use and adapt existing health surveillance and other existing national surveys to assess the problem Seek partnerships with academia and independent specialists to find ways to rigorously study illicit trade Use financial and police investigations to identify and fight organized crime operating in illicit trade 2. Start identifying and implementing appropriate country-specific policies and strategies to address illicit trade Improve tax and customs administration to close the legal and administrative loopholes facilitating illicit trade Implement other appropriate policies to deal with country-specific problems 3. Become a Party and/or implement the Protocol to Eliminate Illicit Trade in Tobacco Products Adapt the Protocol supply-chain control obligations Adjust national penalties for illicit trade offences Seek and build international cooperation 7 The WCO publishes an Illicit Trade Report annually, with the main characteristics and trends of illicit flows in key products, including tobacco, using data based on customs seizures. CHAP T ER 4. PO LI T I C AL ECO N OMY 181 Methodologies available to estimate the nature and size of illicit trade are discussed in subsection 4.1.4. This is the first step for dealing with SCARE tactic S. Chapter 3 discusses at length the relevant tax administration measures and best practices to minimize opportunities for illicit trade in tobacco products. Table 4.2 presents examples of appropriate policies and strategies targeted to address specific types of illicit trade in addition to the best practices described in Chapter 3. After completing the first step of this strategy, tax, health and justice authorities should discuss how to face country-specific problems, considering not only tax and customs administration measures but also social, law enforcement and international cooperation policies and strategies. Table 4.2 Suggested policies and strategies to address country-specific illicit trade problems MAIN TYPE OF ILLICIT TRADE IN THE JURISDICTION PROBLEMS POLICIES/STRATEGIES TO USE Bootlegging Neighbouring low-tax jurisdiction Bilateral negotiations to harmonize tobacco tax systems Difficulty of controlling people’s movements in countries with extensive land borders Identify and establish suitable social protection or employment policies for targeted populations in border regions Extensive land border with multiple accesses Bilateral cooperation with law enforcement and border control forces, monitoring of access routes to main consumption markets Large-scale smuggling from neighbouring jurisdiction Neighbouring low-tax jurisdiction and difficulties in controlling borders Bilateral negotiations to harmonize tobacco tax systems and bilateral law enforcement cooperation Producers and distributors in the lower-tax jurisdiction aiming to supply the high-tax jurisdiction Bilateral cooperation to harmonize tax systems and control producers and distributors in the origin country, create conditions for legal exports and taxed imports Large-scale smuggling from a third country or tax-free zones Producers and distributors aiming to supply non-duty-paid tobacco products wherever possible Customs and other forms of international cooperation to control and monitor exports from identified areas Domestic tax evasion Existence of many small informal or semi-formal producers Encourage business concentration through producer associations and cooperatives, create incentives for formalization and establish licensing rules and basic electronic information systems for raw material and production 182 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Underreporting from formal producers Improve tax administration with policies such as basic electronic information systems for inputs and production, establish neutral procedures to verify production, improve audit systems, increase third-party information on inputs and production of tobacco products Clandestine factories Law enforcement investigation of commercial associations with raw- material and machine producers and distributors Governments should try to identify the incentives and governance problems that encourage and allow illicit trade movement inside their country. As seen in Table 4.2, the design and implementation of policies to deal with those problems do not depend exclusively on tax and customs authorities; they also depend on the efforts of the police and law enforcement, the Justice Department and the judicial apparatus. In other words, a great deal of coordination and consultation among different types of government bodies and expertise is needed to produce an adequate response. It is also clear from Table 4.2 that domestic tax evasion by formal producers can be tackled by tax authorities and is mainly related to the supply-chain-control provisions of the Protocol. The Protocol (26) builds upon and complements Article 15 of the WHO FCTC, which addresses means of countering illicit trade in tobacco products as a key aspect of a comprehensive tobacco control strategy. The Protocol is a blueprint of measures to deal with this problem, and its provisions should be part of any strategy for fighting the illicit market. It is a legally binding treaty in its own right that entered into force on 25 September 2018. As described in detail in Chapter 3, the Protocol has three main lines of action: supply-chain controls, recommendations on how to treat unlawful conduct related to the illicit tobacco trade and suggested mechanisms to seek and build international cooperation to fight that trade. Countries can start implementing Article 15 of the WHO FCTC and the appropriate polices or strate- gies recommended by the Protocol even before acceding to it, selecting those most suitable to the nature and extent of their particular problem. Such transitional work will facilitate the eventual implementation of the Protocol, because any plan to correct loopholes in tax and customs practices will bring government authorities closer to the best practices recommended in the Protocol. 4.1.3 DETERMINANTS OF TAX EVASION: THE ROLE OF PRICE LEVELS The argument that price and tax rates are the main determinants of the illicit tobacco trade has persuaded some governments (e.g. Uruguay and Georgia in the past) to avoid policies that may lead to cigarette price increases (e.g. excise tax rate increases) (4). CHAP T ER 4. PO LI T I C AL ECO N OMY 183 Some governments (e.g. Canada in 1994, Brazil in 1999 and Pakistan in 2017) (20, 27) have even reduced tax rates in attempt to reduce the illicit trade. The wider scholarly literature demonstrates that illicit trade is not a monocausal phenomenon (7) but is the result of many factors, most of them related to gover- nance issues. Government corruption, weak regulatory frameworks, poor tobacco tax administration, ineffective criminal justice systems, weak norms regarding participation in illegal and informal markets and conflicts between neighbouring countries (5) all contribute to the existence and growth of the illicit tobacco trade. It is difficult to isolate the role of price from each of the other factors because (1) obtaining prices and quantity measures of illicit trade is inherently challenging; (2) in most countries, there are many cigarette brands, and prices vary between and even among brands; and (3) there is a lack of good measures to deal with nonprice factors affecting illicit trade, such as government corruption and ineffective criminal justice. These constraints make it challenging to develop rigorous empirical evidence about how price and other factors affect illicit trade. Despite these fundamental challenges, the economic literature has produced credible evidence that price is always only one factor – and often not the most important factor – determining the extent of illicit trade. Many econometric studies about the influence of price and other factors have focused on cross-border shopping (or small-scale bootlegging from low- to high-tax jurisdictions), given the availability in the United States and Europe8 of sales data for low- and high-tax jurisdictions, classified in a convenient way by geographical zones – i.e. close to or far from the borders. Those studies attempted to explain the illicit trade flows or the relatively higher sales in low-tax jurisdictions as a function of price and tax differentials between the lower-tax and surrounding higher-tax jurisdictions, after controlling for other important factors affecting cross-border sales such as proximity to borders and levels of corruption (6, 7, 11, 28).9 The main conclusion of the studies is that illicit trade flows are not linked solely to price (29). Some show a significant effect of price differentials together with other factors, but others do not find significant price differential effects. The important policy implication of these analyses is that decreasing tobacco tax rates and real prices in higher-tax jurisdictions could have minimal or no effect on illicit market shares.10 8 This was a traditional strand of the literature in the United States on trade among states, and to a lesser extent in European countries, most of which used conventional but inaccurate illicit trade measurements. 9 Recently, PMI-Altria financed some studies of factors affecting cross-border sales. One of those studies, Prieger and Kulic (28), criticized Merriman et al. (2000) (9) and arrived at the conclusion that in cross-border shopping, price differentials are important for determining the magnitude of illicit trade. 10 Brazil decreased tax rates and real prices at the beginning of the 21st century to fight illicit trade coming from lower-tax jurisdictions. After this action, however, the government lost revenues, and the size and scope of illicit trade remained unaltered, according to industry sources (20). 184 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Observational and case studies provide information that may improve public policy even when they are unable to produce compelling evidence of causal relation- ships. Some observational studies have correlated price levels with illicit market shares, using large samples of countries. Joossens et al. (25) found that countries with high taxes and prices normally have lower shares of illicit trade than countries with lower tax shares and prices. In their sample, high-income countries generally have relatively high cigarette prices and tax shares, but their favourable results (i.e. lower levels of illicit trade) are related to effective tax administration and lower corruption levels. In contrast, LMICs generally have lower prices and tax shares, along with significant illegal market shares. Joossens et al. attribute difficulties in fighting illicit trade to weak tax and customs administrations and, in most cases, institutional and legal challenges (25). Figure 4.1.1 illustrates the relationship between price and illicit trade, using the price (in US$) per pack of the most-sold brand of cigarettes and the estimated level of illicit trade for 94 countries in 2018.11 There is no apparent unique association between the two variables. Running a linear regression with retail price as the explana- tory variable and share of illicit trade as the dependent variable shows an inverse, but not statistically significant, relationship between price and illicit market share.12 Figure 4.1.1 illustrates some particular cases: • Many countries with low prices (i.e. lower than US$ 2 per pack) have the highest levels of illicit trade in the sample, e.g. Brazil (BRA) ($1.33 and 46.3% illicit share), Pakistan (PAK) ($0.39 and 40%), Ethiopia (ETH) ($0.55 and 32.9%), Ghana (GHA) ($1.06 and 29%) and Cameroon (CMR) ($0.89 and 25%). • In contrast, many of the countries with prices between US$ 4 and $8 – which could be considered high enough for financial incentives to operate – have illicit trade shares of less than 10% of total consumption. These countries include the Republic of Korea (KOR) ($4.02 and 0.8%), Czechia (CZE) ($4.31 and 2.9%) and Sri Lanka (LKA) ($6.89 and 1.6%). • All countries that have very high prices – higher than US$ 8 – except for Ireland, register illicit trade shares below 20%. These countries include France (FRA) ($9.39 and 17.8%), Switzerland (CHE) ($8.71 and 5.5%), Singapore (SGP) ($10.35 and 3.7%) and Norway (NOR) ($14.51 and 9.6%). 11 National estimates of the magnitude of illicit trade are controversial. The tobacco industry’s numbers overestimate the problem and are based on questionable methodologies. Estimates with a rigorous and transparent methodology are not available for a large sample of countries for the same year. In order to compare price levels with illicit market shares, Euromonitor’s estimations of illicit market share were selected, for two reasons: they are comparable estimates for a large sample of countries in a given year, and no one could argue that they are biased towards tobacco control’s points of view. The use of Euromonitor data does not imply that WHO fully agrees with all the details and methodologies used to obtain them. 12 Other factors must be taken into account to transform this observational analysis into a rigorous analysis of cause and effect. CHAP T ER 4. PO LI T I C AL ECO N OMY 185 Fig. 4.1.1 Share of illicit trade versus retail price of the most-sold brand of cigarettes in US$, by country, 2018 Note: The extent of illicit trade in cigarettes is measured by Euromonitor as the estimated quantity of illegal cigarettes consumed in a country divided by the estimated total consumption of cigarettes in that country. Sources: (27, 30). 51 2 3 4 6 7 8 9 10 11 12 13 14 1615 0 10 20 25 15 5 30 35 40 45 50 MYS BRA ECU ECUPAK ETH PAN CRI URY GRCIND GTM SLV ARE LVAHND NGA ZAF CAN FRA LBNAGO IRL DOM GHA CMR Retail price, USD per pack Ili ci t t ra de % IRQ VNM COL MMR AUT SWE EST PER ESP ROM POL CHL ISR GBR KEN LAO KHM EGY MKD TZA THA TUN BGD UZB UKR GEO KAZ CHN KWT HRVJOR CZE SVK KOR JPN BLR SRB TUR OMN FIN NLD CHE SGP SAU USA LKA DNK DEU ITA HUN SVN PRT BOL DZAIDN CIV BEL MAR ARG AUS NOR NZL LTU BIH MEX BGRPHLAZE RUS 186 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N As indicated in numerous studies and analyses of illicit trade in tobacco products (4–5), the most effective way to tackle the problem is not to forgo tax increases but rather to strengthen the capacity to fight the trade. Therefore, it is important to consider the relationship between good governance and illicit trade. The more capacity a country has to counter illicit trade in general, the lower the level of that trade will be. An index compiled by the Economist Intelligence Unit (EIU), the Global Illicit Trade Environment Index, measures countries’ structural capacity to fight illicit trade overall. The EIU indicator is a combination of four indicators or categories designed to assess countries’ performance in those areas; the closer the overall indicator is to 100, the better the country’s capacity to fight illicit trade. The four categories are:13 1. government policy, which measures the government’s commitment to pro- actively monitoring and preventing illicit trade; 2. supply and demand, which measures the extent to which the domestic en- vironment discourages or encourages supply and demand for illicit goods; 3. transparency and trade, which measures transparency and the degree of governance applicable to free-trade zones and transhipments; and 4. customs environment, which measures how effectively customs services facilitate legitimate trade while at the same time preventing illicit trade. Figure 4.1.2 illustrates the relationship between the EIU indicator and the estimated level of illicit trade in cigarettes in a set of countries. There is an inverse and statisti- cally significant relationship between the indicator and the estimated level of illicit trade in cigarettes. This suggests that as the capacity to fight illicit trade in general increases, the illicit trade in cigarettes falls.14 13 For more details about this indicator, visit http://illicittradeindex.eiu.com/. 14 The association was significantly different from zero at a 90% confidence level, using a linear regression between the two variables. CHAP T ER 4. PO LI T I C AL ECO N OMY 187 Fig. 4.1.2 Share of illicit trade versus the EIU indicator in 70 countries, by country, 2018 Sources: (30–31). % Il ic it tr ad e ci ga re tt es MYS 0 10 15 5 20 25 30 35 40 45 50 55 60 EIU GIT indicator 20 3010 40 50 60 70 80 90 JPN KOR HRV SAU UKR BLR KHM LAO MMR IRQ GTM DOM VNM MAR SRB PER BGR MEX ROM COL ZAF BIH PHL IDN ITA PRT HUN BEL DEU ESP ARG AUS SWECHL TW POL ISR AUTLTU FRA CAN IRL LVA ARE GRC IND URY CRI PAN PAK ECU BRA EST GBR FIN SVN NLD SVK TUN RUS DZA TUR CHNKAZ THA CZE USA SGP NZL DNK 188 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Rigorous independent research has established that despite the challenges of illicit trade, taxation of tobacco products is an effective public health intervention that substantially reduces tobacco use and generates government revenue (5). Further, when cigarette taxes increase, governments generate higher revenue and consumption is reduced (32–33). However, ineffective tax administration can allow illicit trade to grow and can undermine some of the benefits of tobacco taxation by making cheaper cigarettes available. For example, the average street price of smuggled cigarettes in Malaysia is 55% lower than its legal tax-paid equivalent (34). Illicit tobacco trade also reduces government tax revenue and may increase health costs associated with smoking and costs associated with policing. 4.1.4 MEASURING ILLICIT TRADE IN TOBACCO PRODUCTS The magnitude of illicit trade is a powerful argument in tax policy discussions, and for this reason the tobacco industry funds estimation of illicit trade in countries or regions of particular interest to itself (i.e. Project Sun and Project Star in the EU and Oxford Economics in East Asia). However, a recent systematic review of industry data on illicit trade finds substantial methodological weaknesses in industry-commissioned reports (24). Furthermore, Blecher et al. (35) argue that industry-funded studies tend to systematically overestimate the size of illicit trade to persuade authorities to abandon tobacco tax reforms. Independent researchers have also uncovered inconsistencies in tobacco-industry-funded estimates (36). Some examples of inflated industry-linked illicit trade estimates are given in Table 4.3, which compares peer-reviewed and independent studies with estimates funded by the tobacco industry. Because some countries have several industry estimates from different sources or years, Table 4.3 presents the estimate included in the article that published the independent study, because it was considered as representative and adequate to illustrate the overestimation. In all cases, the industry estimates exceed those of the independent studies. Measuring the scale of illicit trade can be a daunting task for governments because different methods are employed by independent researchers, governments and the tobacco industry. Nonetheless, it is worth investing in these studies because they drive policy discussions and can be used to evaluate the impact of policies (e.g. tax increases, plain packaging and health warnings). CHAP T ER 4. PO LI T I C AL ECO N OMY 189 Table 4.3 Illicit market share estimated in independent studies compared with estimates in tobacco-industry-funded studies COUNTRY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDEPENDENT STUDIES SOURCE OF THE INDEPENDENT STUDY SIZE OF THE ILLICIT MARKET AND YEAR OF THE ESTIMATION – INDUSTRY- FUNDED STUDIES INSTITUTION RESPONSIBLE FOR THE INDUSTRY- FUNDED STUDIES Colombia 3.5% of the total market in five cities, 2016 Maldonado et al., 2018 (37) 13% of the total market, 2014 FND and INVAMER, 2015 Chile 16.3% of the total market in the Metropolitan Region of Santiago, 2017 Paraje et al., 2020 (38) 24.3% of the total market, 2017 Observatorio del Comercio Ilícito BATC, 2017 Brazil 28.8% of the total market, 2014 Iglesias et al, 2017 (39) 34%, of the total market, 2014 BAT public statement, 2015 Mexico 8.8% of the total market in eight major cities, 2017 Saenz de Miera Juarez et al., 2020 (40) 16.6% of the total market, 2012 Confederación de Cámaras Industriales, 2012 As shown in Table 4.4, methodologies to measure illicit trade can be grouped into three types: (1) direct measurement; (2) residual methods and (3) expert opinion (12). Direct measurements rely on evidence directly linked to actual illicit behaviour and pack observation; residual methods infer evasion based on theory and evidence about consumption and legal sales; and expert opinion distills information garnered from talking to individuals with the most direct knowledge of the tobacco market. Each method has advantages and disadvantages. No single method is unambigu- ously superior to others, but direct measurement and residual methods are more conducive to determining the size of the illicit market, whereas expert opinion could provide insight into the details of the market’s operations.15 Table 4.4 presents the relative amount of resources and the degree of expertise required to implement each main measurement method, as well as the primary purpose, data collection characteristics, sampling features and unit of analysis. A brief description of each of the methods is presented in Annex 4.1. Merriman (11) and Ross (9) provide more expansive details. There is no simple selection rule for deciding what measurement method to use. The major factors to consider when selecting a method or methods include (1) the nature and characteristics of the illicit trade problem (i.e. where and how the 15 In interviews with experts from the tobacco industry, provisions of Article 5.3 of the FCTC and its Guidelines need to be followed. 190 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N problem manifests and whether domestic tax evasion or illegal inflows of foreign brands or a combination of both predominates), (2) previously collected data, (3) available budget and (4) expertise of available analysts. Available budget and staff skills are often the main restrictions that governments face. Therefore, Table 4.4 orders the measurement methods according to resources needed and available expertise. For example, residual methods and expert opinion can provide crude but useful estimates at low cost and require the lowest levels of technical sophistication. Another low-cost option for countries that employ popula- tion health surveillance surveys is to add questions to measure illicit trade, such as brand name, value and quantities of the last purchase. In contrast, the direct measurement approach often requires sophisticated research designs and expensive (and time-consuming) field research. Table 4.4 Overview of resources and expertise needed and main purpose of measurement methods METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Seizures (D) $ Low Identify trends in types of products, transporta- tion methods, points of entry and brand names Secondary data use Non- probability Shipments Use of existing health sur- veillance surveys – self- reported consump- tion (D) $ Low, only additional questions Size of illicit trade, adding or improving questions on brands, value and quantities of the last purchase Additional primary data collection Probability Individuals Gap analysis (R) $ Medium Provides a measure of changes in illicit trade Secondary data use Universe Nations Econo- metric modelling (R) $ High Estimation of price elasticity of substitution from tax- paid to illicit products Secondary data use Universe Geography Expert interviews (E) $ Low Characteristics of the illicit trade Primary data collection Non- probability Individuals Smoker intercepts and pack observa- tion surveys (D) $$$ Medium Size and characteristics of illicit trade, probability-based sample to be representative of population Primary data collection Probability Individuals CHAP T ER 4. PO LI T I C AL ECO N OMY 191 METHOD LEVEL OF RE- SOURCES EXPER- TISE MAIN PURPOSE OF MEASUREMENT METHOD DATA COLLEC- TION SAM- PLING UNIT OF ANALYSIS Pack return and swap surveys (D) $$$ Medium Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Littered- pack surveys (D) $$$ Medium Size and characteristics of illicit trade, compa- rability with industry estimation using empty-pack surveys Primary data collection Probability Individuals Covert purchases (D) $$$ Medium Type of products and trade channels of illicit trade Primary data collection Probability Geography Self-report consumer surveys (D) $$$ High Size and characteristics of illicit trade, probabili- ty-based sample to be representative of population Primary data collection Probability Individuals Notes: Universe includes total population; D = direct measurement, R = residual method, E = expert opinion. Scale for resource costs assuming a moderately sized study (e.g. a representative study of a region of several million): $ (cheapest) – weeks of skilled labour hours; $$ (moderately expensive) – 1 to 2 months of skilled labour hours; and $$$ (most expensive) – 6 to 12 months of skilled and unskilled labour hours. A more detailed description of the different measurement methods is given in Annex 4.1. To further assist responsible authorities in deciding which method to select, Table 4.5 presents the key characteristics of each of the measurement methods, along with the main advantages and disadvantages of each. Countries may begin with methods that require fewer resources and less skills to obtain an overview of the problem. Seizures – which are a by-product of law enforcement efforts – pro- vide a first step, and countries can analyse the information obtained (origin of the products, brands, location, etc.) and report the results to increase public awareness of the problem.16 Alternatively, countries can add questions related to illicit trade to existing and funded health surveillance surveys conducted regularly by health surveillance authorities and statistical authorities. In that way, cooperation in using existing measurement methods between health authorities – the tobacco control office and health surveillance unit – tax and customs authorities and the national 16 Seizures are useful for obtaining qualitative information about the illegal activity, but they have to be treated very cautiously in projecting the size of the problem. Countries may think they have a very large problem because they have competent authorities doing an extraordinary job at finding illicit goods. On the other hand, countries can have less-efficient authorities making few seizures, and in these environments, seizures tell nothing about the size and nature of the problem. 192 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N statistical office could be a starting point for identifying the nature and size of the illicit trade problem in the country. Direct observation of packs has been increasingly implemented in many LMICs, through different types of surveys such as intercepts of smokers or retailers, pack return, littered-pack inspections and covert purchases of cigarettes. These activities have expanded the skills of independent researchers and academia and increased knowledge of these methods. Also, increasingly cheaper digital technologies allow interviewers to take pictures and record pack characteristics in direct observation surveys or in larger national self-report consumer surveys. Table 4.5 Key characteristics, advantages and disadvantages of illicit trade measurement methods METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Seizures (D) Statistics of tobacco products confiscated by local and national authorities Readily available from law enforcement agencies May not provide a representative picture of the size and/or nature of illicit trade Using existing health surveillance surveys to obtain self-reported consumption (D) Adding or improving questions about brand names, quantities, prices, locale of purchase and other factors Produces good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Gap analysis (R) Compare self-reported consumption data with observed (usually administrative) data about tax-paid sales When quality data are available, is simple and easily reproduced (providing for measurements over time) and explainable Data on tax-paid sales and/or consumption are frequently inaccurate and in many cases do not provide information on the size of the illicit market, but only on changes over time Econometric modelling (R) Estimated according to the difference between tax-paid sales and predicted consumption given by the model Because it is consistent with a long tradition of economic theory, empirical estimates can be evaluated Requires high-quality data on a variety of important variables over a period of time and advanced econometric modelling expertise CHAP T ER 4. PO LI T I C AL ECO N OMY 193 METHOD KEY CHARACTERISTICS MAIN ADVANTAGE MAIN DISADVANTAGE Expert interviews (E) Experts include researchers (e.g. in economics, criminal justice and public health), journalists, tax and enforcement specialists, product manufacturers and wholesalers Useful for identifying the nature of and trends in the marketplace (e.g. venues where illicit cigarettes are sold, modes of entry), and the interviews can be useful for defining the method to assess the size of the illicit trade Information obtained may not be generalizable, and expert knowledge may be outdated or limited by the experts’ experience; also, experts often have strong biases Smoker/retailer intercepts and pack observation surveys (D) Examining the packs of smokers and cigarette retailers, convenience or probability-based sample Is direct and objective, and smokers do not suffer from any value judgements The difficulty of identifying areas representative of the tobacco use population and sampling important subpopulations such as elderly and immobile smokers, but household surveys could overcome sampling issues Pack return and swap surveys (D) Also a pack observation study using survey sampling techniques to examine smokers’ pack characteristics May decrease the stigma associated with traditional smoking surveys In LMICs, survey distribution may be unreliable because of the mail delivery system Littered pack surveys (D) Also known as empty-discarded-pack surveys; publicly discarded packs bear characteristics (e.g. tax stamps, public health warnings) that indicate whether they are tax compliant Yields estimates that are less likely to be biased from issues of social desirability, recall error and confidentiality Significant budgets could be needed to employ field researchers to collect, code and analyse the data; surveys do not provide information about the smoker and the price paid Covert purchases (D) Uses covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes Directly identifies sources of illicit cigarettes It is difficult to create a sampling frame of retailers for illicit sources or to know what smokers are actually buying and how much Self-report consumer surveys (D) Surveys can be distributed to individuals or households, using various modes of distribution Good estimates at national level Self-reported data may be biased due to the social stigma of consuming illicit tobacco products Notes: D = direct measurement, R = residual method, E = expert opinion. 194 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ultimately, when capacity allows, more solid estimates will need to be made using more than one methodology at a given point in time. Ideally, estimates will be made on a regular basis in order to assess the evolution of illicit trade over time and its possible connection to policy changes. 4.1.5 CRITIQUING STUDIES THAT MEASURE THE SIZE OF THE ILLICIT MARKET Measuring illicit trade is a challenge for researchers, industry and governments, because the trade is, by definition, hidden from plain sight. Buyers and dealers sometimes go to great lengths to ensure that their participation in illicit activity is concealed. Over the years, researchers and government agencies have been increas- ingly interested in estimating the size of illicit markets and identifying effective interventions. As consumers of research, governments should critically examine available studies and evaluate them on their scientific rigor and methodological transparency. Measurement issues are particularly acute with respect to the illicit tobacco trade because it is a politicized topic. High estimates may raise questions about the tobacco industry’s ability to control the supply chain, its involvement in illicit diversion, the impact of taxation policies and the effectiveness of enforcement strategies. While the industry has portrayed itself as taking an active stance in measuring and fighting illicit trade (e.g. Project Star, conducted by KPMG LLC but paid for by PMI, later followed by Project Sun), in the past it has used smuggling as a strategy to enter closed markets – for example, in China and Russia (19, 41). Govern- ments should carefully scrutinize evidence about the illicit tobacco trade produced by industry or quasi-industry sources and are advised to seek alternative evidence. Quasi-industry reports are studies commissioned by the industry but published by private research companies (e.g. Ernst and Young, Oxford Economics) (11). Characteristics of good analyses One of the main characteristics of a good analysis is scientific rigor, which involves the use of relevant theoretical frameworks, sound statistical methods and examination of the robustness of findings (e.g. sensitivity analyses). High-quality research reports provide transparent explanations about their methodology and statistical analysis steps undertaken, as well as supplementary analyses that established the robustness of the findings. For example, Joossens et al. (25) clearly describe the data sources used (limitations and advantages and where they can be found) and calculations performed on the number of lives that would be saved if the global market share of illicit cigarettes was eliminated. Explanations should be detailed enough to allow future researchers to scrutinize the analysis and replicate the findings. Replicability is another hallmark of good science. For example, littered-pack studies should detail where and when data collection took place, how many packs CHAP T ER 4. PO LI T I C AL ECO N OMY 195 were collected per geographical unit, the protocol of identifying the illicit packs (e.g. characteristics of the warning labels, brands, tax stamps, etc.) and details of statistical analyses. There should also be explanations of the representativeness of the selected geographical areas. Failure to provide this depth of information may call into question the generalizability of a study and whether there are faults with the chosen method. In the context of policy decisions regarding illicit trade, the most useful data provide information about a representative sample of individuals and geographies. Studies that are limited to, for example, one group of individuals based on specific characteristics or a given geography may yield biased information. Research reports also should be clear about the study’s limitations. For example, studies that measure illicit trade often do not measure product counterfeiting and do not include non- cigarette tobacco products in their estimates (42). When statistical estimates are included, they should provide confidence intervals as well as point estimates to account for uncertainty resulting from simple random chance (11). Characteristics of flawed analyses Flawed analyses can convolute and distort scientific knowledge about illicit trade. Flaws usually manifest in the data, methodology, statistical analysis and/or interpreta- tion of the results (11). Studies may be purposefully designed with methodological flaws to yield high or low estimates of the trade. For example, research showing that illicit trade constitutes a large share of the total market may be used to support arguments that taxes cause sharp increases in illicit trade, whereas lower estimates may be used to support arguments that certain governmental interventions (e.g. increased retail inspections) are effective. Pressures to skew data may also be tied to funding. For example, high estimates can sway governments to provide more resources for law enforcement activities. Analyses can be purposefully skewed by using data sources or data collection methods that will provide biased estimates. Flawed studies sometimes provide incomplete or inaccurate descriptions of their methodology. They may lack detail regarding the quality of the data used or information about how the data were collected and analysed. For example, a common weakness in industry-funded research on discarded packs is that the methods of collection and forensic analysis are not reported, ostensibly because they are “proprietary” information (43). However, these methodological details are key to assessing whether a study’s findings are biased by sampling error, model misspecification, measurement error, non-response or other flaws. It may be impossible to assess measurement error if researchers fail to disclose questions included in a survey instrument. Survey items used to measure the illicit tobacco trade may be imprecise. For example, asking respondents the frequency with 196 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N which they purchase “cheap” cigarettes may yield biased estimates, particularly if consumers can purchase cigarettes at discounted prices by using coupons. To more ac- curately measure tax evasion, surveys must include questions about the location of last purchase, purchase price, presence of public health warning labels and brand names. There are other ways that flawed studies can inadvertently or purposefully distort estimates of illicit trade. For example, data collectors can intentionally oversample areas known to be hot spots of illicit sales or sites that residents from lower-tax jurisdictions visit. Researchers can collect discarded cigarette packs close to the borders of countries with lower taxes to (inadvertently or purposefully) demonstrate the undesired side effects of tax policies. Studies published in non-peer-reviewed or lightly peer-reviewed outlets such as edited book volumes or policy briefs should be viewed with more scepticism than those published in highly regarded peer- reviewed outlets. 4.1.6 CONCLUSIONS Globally, the illicit tobacco trade continues to be a major concern for tax admin- istrators because of the challenges it generates to collecting higher revenues as well as the challenges to accurate and independent measurement. Industry figures provide distorted conclusions regarding the extent of the problem – frequently with a monocausal explanation of the link between illicit trade and tobacco taxation. Illicit trade comprises multisystemic issues and requires multiple strategies. Worldwide, countries at different levels of economic development have implemented a variety of effective measures to combat the illicit trade in tobacco products. The Philippines and the United Kingdom, for example, have addressed illicit trade as part of their overall tobacco tax reform (4). Price (and tax) levels are not a key determinant of illicit trade, the presence of which is exacerbated by the lack of tax administration capacity. Refraining from increasing taxes is not the solution; countries should instead respond with a com- prehensive strategy that includes at least these three main components: 1. It should identify – independently from the industry – the nature and dimen- sions of the problem. It is necessary to assess scientifically and with the best statistical practices the size of the illicit trade to understand the characteristics and scope of the problem. 2. It should identify and implement appropriate policies and strategies targeted at addressing the specific type of illicit trade the country is experiencing. It should address directly the country-specific institutional and/or governance challenges – as well as the lack of multilateral coordination that can exacer- bate illicit trade – and improve tax and customs administration practices as described in Chapter 3. CHAP T ER 4. PO LI T I C AL ECO N OMY 197 3. It should implement best practices contained in the WHO FCTC Protocol to Eliminate the Illicit Trade in Tobacco Products and accede to the Protocol if the country is not yet a Party. There are proper methods and policies with which to address the illicit tobacco trade. If countries start implementing the appropriate policies, they can raise tobacco taxes and reap health and revenue benefits even in the presence of illicit trade. 198 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.2 SCARE TACTIC C: COURT AND LEGAL CHALLENGES 4.2.1 INTRODUCTION The tobacco industry views well-designed and significant tax increases as a threat to the profit, growth and long-term sustainability of its business. As noted by PMI in 1985: Of all the concerns there is one – taxation – that alarms us the most. While [other restrictions] … do depress volume, in our experience taxation depresses it much more severely (44). The industry is, however, less likely to launch direct legal challenges to excise taxes than to other tobacco control measures (see Box 4.2.1 for details), because taxation – and excise tax in particular – is a comparatively well-established regulatory measure; in many jurisdictions, taxes have been levied on tobacco products for more than a century. There is also less unanimity in opposition to taxes among tobacco industry actors, because differences in the market position of different tobacco companies affect their interests in tax policy. This, in turn, decreases the likelihood that they will act collectively on the issue (45). BAT’s stated strategy in the early 1990s was to influence governments with regard to the level and structure of tobacco taxation in order to promote market growth and to secure competitive advantage (46). Nevertheless, tobacco industry actors will still legally challenge, or at least legally threaten, significant tax measures when vulnerabilities in their design, adoption or implementation are apparent. Box 4.2.1 Court and legal challenges to tobacco tax measures Evidence suggests that the tobacco industry and its allies instigate fewer legal actions against tax measures than against other tobacco control measures: 1. The Campaign for Tobacco-Free Kids’ tobacco control laws database contains only a handful of cases concerning tobacco tax measures, but hundreds on other tobacco control topics. This pattern can also be seen in a 2018 review of tobacco control legal challenges that examined this and two other databases to select 96 cases relevant to the question of the WHO FCTC’s usefulness in litigation (47). Only 6 of these 96 cases were challenges related to tax measures. CHAP T ER 4. PO LI T I C AL ECO N OMY 199 2. A 2013 systematic review of empirical studies on tobacco industry interference with tobacco tax policy found that only 9 of 36 relevant articles reported the specific use of litigation as a tobacco industry tactic (1). All 9 concerned constitutional challenges to earmarking provisions for tobacco tax initiatives in the United States (1). 3. A 2015 study on industry interference in LMICs cited legal challenges to tobacco control measures in 15 countries as examples of industry interference, but none of the challenges concerned a tobacco tax measure (48). 4. A 2016 analysis of papers published in systematic reviews of industry inter- ference with tax and marketing measures found that only 5 of 65 papers concerning tobacco tax related to the use of litigation or threats of litigation to interfere with tobacco tax measures (49). The tobacco industry makes extensive use of legal experts (1, 50–52) who study all relevant laws and regulations closely to determine their likely and arguable boundar- ies for the purpose of manipulating regulations and regulators (1, 50–52). Based on this expert advice, tobacco companies know when regulations remain within the bounds of both international and domestic obligations but can still argue that legally permissible tobacco control measures would be defeated in litigation if passed (48, 51–52). As the threat of a legal challenge alone can be used to the industry’s advantage, recourse to litigation is seldom needed or desirable (1, 45, 48, 51, 53–56). Even when litigation is launched, the objective may be to delay or weaken a measure rather than to win on the merits of the case (1, 45, 48, 53). To counter actual and threatened legal challenges, policy-makers need to be aware of relevant legal obliga- tions when preparing and implementing tobacco control measures. Fortunately, the tobacco industry playbook is relatively predictable. Tax and other tobacco control measures can thus be designed to strengthen the regulators’ legal position against genuine threats and enable them to dismiss baseless industry threats. 4.2.2 COUNTRY EXPERIENCES WITH LEGAL CHALLENGES TO TOBACCO TAXATION Legal obligations that are relevant to tobacco taxation include those under do- mestic law and international instruments such as international trade agreements and international investment agreements (IIAs).17 Some of the legal issues that a tax measure may encounter are outlined in Table 4.6. Case studies from various countries illustrate how these legal issues have and have not been avoided in the 17 Relevant international trade agreements include the WTO Agreement and custom unions such as the EU, the East African Customs Union and Mercosur. Relevant IIAs include bilateral investment treaties and the investment chapters in free trade agreements and within custom unions. 200 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N passage, design and implementation of tobacco taxes. These issues are not the norm, however, and should not give rise to undue apprehension. The case studies are rated as positive, mixed or negative based on the extent to which the legal decision upheld the taxation measure in question. Table 4.6 Potential legal issues for tobacco tax measures VULNERABILITIES LEGAL OBLIGATIONS CASE STUDIES Inadequate consultation and other procedural vulnerabilities Domestic procedural law 1, 2 Due process protections for investors under IIAs None Procedural requirements under WTO Agreements and Custom Unions 3 Discrimination against imports or investors Nondiscrimination obligations under WTO Agreements and Customs Unions 8, 9, 10 Nondiscrimination obligations under IIAs 11 Investment incentives or inducements Arbitration mechanisms under investor-state contracts 12 Fair and equitable treatment clauses of IIAs None Other substantive breaches Constitutional rights and restrictions on taxation 4 Statutory restrictions on the imposition of taxation 6 Expropriation clauses of IIAs 5 Ultra vires (the scope of legal authority) 7 Avoiding procedural vulnerabilities in tax laws Procedural defects can be avoided by taking great care in progressing and imple- menting regulatory or legislative provisions. Procedural concerns pose a dilemma for tobacco control regulators. Article 5.3 of the WHO FCTC and the COP guide- lines for its implementation state that policy-makers and regulators should interact with the tobacco industry only when and to the extent strictly necessary (57). For taxation measures, interaction might be necessary because consultative and de- liberative processes could be prescribed under domestic constitutional provisions and procedures for good governance, due process requirements of IIAs and some international trade agreements. The tobacco industry may use these requirements as leverage to delay, distort or hijack the rule-making process in contravention of Article 5.3. Accordingly, interactions with the tobacco industry should be limited to strictly necessary consultation conducted in a transparent or public manner but with care that this does not come at the expense of a measure’s defensibility. The proper balance will depend on the jurisdiction in question, since constitutional, statutory and applicable international legal obligations vary. CHAP T ER 4. PO LI T I C AL ECO N OMY 201 CASE STUDY 1 (MIXED): Industry manipulation of legislative procedures In 2012, a bill stipulating, among other things, the creation of a new specific excise tax on cigarettes passed its final reading in Costa Rica’s Legislative Assembly. Passage of the bill had, however, proceeded under “urgency” and notwithstanding a pending constitutional enquiry (a constitutional query is meant to prevent passage of a bill).18 ISSUE MAJORITY DECISION MINORITY DECISION LESSON Whether the court could consider the enquiry despite passage of the bill and the effect the bill’s passage could have despite the enquiry. The enquiry was taken up by the Supreme Court’s Constitutional Division’s majority (58). The signing and publication of the bill by the executive was suspended by the Constitutional Division pending their decision on the merits of the case – which, in the end, found any question of the bill’s constitutionality baseless (58). The enquiry was inadmissible by reason of having been filed too late and notice of its filing having not been received by the legislature prior to the reading of the bill (58). In disagreement with the majority, the minority held that the court could not consider the enquiry or suspend the bill’s signing by the executive – the final step in becoming law. This challenge demonstrates how the tobacco industry’s defenders may attempt to frustrate and impede a tax measure’s passage. In this case, the challenge seemed to have been a delaying tactic, as it was posted on the same day as the final reading of the bill. Its authors may have either wanted its pending nature to cause the legislature to delay or, as occurred, to create conditions for a procedural and constitutional challenge in the absence of delay. All the grounds of the challenge itself were found to be without merit. Although such frivolous challenges cannot be prevented, they can and should be anticipated to ensure that they do not lead to a tax measure’s defeat. 18 “Urgency” is a procedure under which a bill is progressed through a legislature in an expedited fashion. 202 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N CASE STUDY 2 (POSITIVE): Adhering to domestic procedural requirements Kenya’s tobacco control regulations required the tobacco industry to pay a levy to compensate the state for health care and other negative externalities of smoking. In a 2016 challenge brought against these regulations, the plaintiff, BAT, was unsuccessful on every count (59–60). Even though the levy was not considered a tax measure by the court, the case study is instructive on how regulators may safeguard tax measures against procedural challenges. ISSUES LAWS AND ARGUMENTS DECISIONS LESSONS Whether the government’s consultations on the measure were adequate. Asserting that the Constitution and the Statutory Instruments Act together meant that “appropriate consultations with persons who are likely to be affected” were required because of the measure’s likely substantial effect on business. BAT claimed that this standard was not met. Kenya’s government claimed that it was under no obligation to undertake special or extensive consultation with the tobacco industry. The judge found in favour of Kenya’s government, noting that (1) the requirement to consult does not imply that any particular view needs to prevail; (2) dissatisfaction with the level of consultation is not decisive; (3) on the facts, industry was allowed, and often invited, to send representatives to all relevant public consultative meetings and parliamentary committee hearings; and (4) consultation on the regulations was adequate (59). The tobacco industry carefully scrutinizes legislative and regulatory processes for defects. In this case, Kenyan government officials appropriately distanced themselves from the tobacco industry by not permitting its representatives special consideration but did permit their attendance at public meetings and the ability to submit their views under usual procedures. In this way, both the principles behind WHO FCTC Article 5.3 and the requirement for consultation under Kenyan law were observed. CHAP T ER 4. PO LI T I C AL ECO N OMY 203 Avoiding procedural issues in tax administration CASE STUDY 3 (NEGATIVE): Contravening procedural requirements in international obligations In 2010, a WTO panel held that Thailand violated the Customs Valuation Agreement (CVA) by the process it used to value cigarettes that Phillip Morris (PM) Thailand imported into the country from a related party, PM Philippines. Customs values are important as they are the tax base for tariffs and can feed into the base for other taxes levied against the value of the good, such as ad valorem excise taxes and VAT. Transaction values declared by PM Thailand were rejected by Thai tax authorities as influenced by the relationship between the parties and a customs value determined by deduction was substituted (61). ISSUE LEGAL OBLIGATION DECISION LESSONS Whether Thailand adequately consulted with PM Philippines before rejecting its declared transaction value (61). The CVA requires good faith exchange of reasons and information, with opportunities for response (61). Thailand had failed to properly explain its reasons for rejecting the transaction value, as well as its belief that price was influenced by the relationship between the two parties (61). This was a violation of the CVA. Thailand did not appeal these findings. Thailand’s authorities needed to take greater care in their dealings with the tobacco industry to ensure they met the pertinent procedural obligations. In this instance, a specific and high standard of consultation – the provision of detailed reasons and an opportunity for response – was prescribed by the CVA and Thailand failed to meet it. 204 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Ensuring compliance with substantive requirements Rules found in domestic and international law also establish substantive obliga- tions. This subsection focuses on substantive obligations found in domestic law. International obligations concerning discrimination and investment incentives are considered in the next subsections. CASE STUDY 4 (POSITIVE): Tax measure found to be consistent with the Constitution The Chilean government introduced a substantial increase in tobacco and fuel excise, and in 1995, a coalition of taxpayers brought a Constitutional challenge to the measure (62). ISSUE DECISION LESSON Whether the tax was, per article 19 of the Chilean Constitution, “obviously disproportionate or unjust” (62). The excise tax increase did not violate the Constitution, as it was neither confiscatory nor manifestly irrational. Generally applicable excise taxes are not vulnerable to challenges for being excessive, unfair or disproportionate. CASE STUDY 5 (POSITIVE): Failure to grant tax rebates not an expropriation under an IIA This case study is an example of a claim for breach of an expropriation clause in an investment treaty. Such clauses protect foreign investors against measures that can be construed as directly or indirectly seizing an investment or depriving it of its value (63). In the case, an investor was, for more than a decade, denied tax rebates by the Mexican government. This affected the profitability of the business of purchasing and reselling Mexican cigarettes abroad, and the investor brought the claim to an investment agreement arbitral tribunal in 2002 (64). ISSUE DECISION LESSONS Whether Mexico’s failure to grant rebates to the investor exceeded the bounds of valid regulation to constitute indirect expropriation of the investor’s investment (65). There was no expropriation. The arbitral tribunal noted that not all business problems are violations: the investor had no right to participate in the “grey market” export of cigarettes and there were sound reasons to restrict that market (65). Further, the investor was able to participate in other business ventures and actually continued to have business success (65). Claims of indirect expropriation made under IIAs are unlikely to be successful, as generally applicable tax measures are a legitimate form of regulation. A mere loss of profit will not suffice. Claims of expropriation will not succeed unless a substantial or significant deprivation of the investment results. CHAP T ER 4. PO LI T I C AL ECO N OMY 205 CASE STUDY 6 (NEGATIVE): A regulation contrary to superior domestic legislation In 2011, an Indonesian tobacco industry association group, FORMASI, challenged a new excise regulation. Since 2009, the government had been implementing a tiered specific excise tax system based on a set of characteristics (size of production, type of cigarettes and price levels). In 2011, excise rates were increased in nearly all of the 19 tiers, but the reference prices were not accordingly adjusted. This gave rise to a legal issue. ISSUE DECISION LESSONS Whether new excise regulations breached a 57% ceiling for the rate of excise on the retail sale price of tobacco products under the superior Excise Law (66–70). The challenge specified that excise exceeded this ceiling for hand-rolled domestic clove cigarettes (kreteks) (68, 71). The Court found in favour of the tobacco industry association, and the government was required to immediately revoke the 2011 regulation. (69–70). It is advisable to stay within the rules and be aware of legal hierarchies – including superior domestic legislation. The tobacco industry scrutinizes all increases in tobacco taxes. In this case, a breach of a legislative requirement for a single category of tobacco product resulted in Indonesia suffering lost revenue and a setback in its efforts to reduce tobacco consumption. Ensuring a tax measure is within an authority’s legal power A tax measure is ultra vires when it goes beyond the legal power of the enacting body. As with case study 6, this is a legal issue that involves legal hierarchies. In ultra vires cases, however, instead of centring on conflict between inferior and superior law, the issue is whether an authority that enacts a tax measure is authorized to do so. This issue may arise when a tax measure is enacted by a subnational jurisdiction or by an executive acting under a statutory delegation. CASE STUDY 7 (NEGATIVE): Tobacco taxation contrary to the Australian Constitution ISSUE DECISION LESSONS Whether New South Wales’ licensing and penalty fees regime constituted an excise tax by other means contrary to the Australian Constitution’s exclusive grant of that power to the federal government (72). The court found that state licensing fees were excise taxes and that this was contrary to the Australian Constitution (72). Authorities enacting tobacco tax measures must act within the scope of their legal power. 206 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Avoiding discrimination against imports and foreign investments Although inherently discriminatory, customs duties may be used subject to the agreed upper limits in a country’s trade agreements. Excise and other taxes designed with the aim of raising tobacco prices to reduce demand and advance human health should be origin-neutral: they should not seek to tax local products less than foreign products or aim to treat foreign products differently from one another. Tobacco tax measures are pursuing objectives other than health when they aim to raise the price of imports more than that of local products or seek to burden favoured market participants less than others. Solely health-protective tobacco taxes will not ordinarily violate Articles III:2 and I:1 of the GATT (the WTO’s General Agreement on Tariffs and Trade), which prohibit discriminatory taxation (in light of general exceptions). Nor will solely health-protective tobacco taxes directly violate anti-discrimination protections for investors found in the national-treatment (NT), most-favoured-nation (MFN), expropriation and fair-and-equitable-treatment (FET) clauses of IIAs (63, 73–74). It is possible to make claims for breach of international obligations on grounds other than discrimination, but such claims are generally highly unlikely to succeed. CASE STUDY 8 (NEGATIVE): BAT v Uganda (2017 East African Court of Justice) DISCRIMINATION LESSON Uganda established a higher level of excise taxes on imported cigarettes – including those from Partner states of the East African Customs Union (75) – than on local cigarettes. Its implementation was discrimination contrary to Article 15 of the Customs Union Protocol (75). Differential taxation explicitly based on origin can be construed as protectionist discrimination in violation of international obligations. The tobacco industry can also turn to international trade agreements outside of the WTO – in particular, customs union mechanisms. Difficulties arise when ostensibly origin-neutral and health-protective tobacco taxes result in dissimilar taxation of tobacco products (73). Discrimination does not exist simply because there is dissimilar taxation – the taxation must adversely impact imported goods more than local products, the imports of one nation more than another or a particular investor’s products more than comparable products. Where dissimilar taxation between product categories results in discrimination, the tax will ordinarily still be lawful if the dissimilar taxation is based solely on a legitimate regulatory distinction between the product categories in question.19 19 The precise applicable rules vary depending on the nature of the legal obligations in question. Under the GATT, dissimilar taxation of like or directly competitive products can be justified based on scientifically grounded distinctions between products under Article III:2 and, in the alternative, discrimination that is necessary under the explicit carve-out for health-protective measures, Article XX(b) (73). For the MFN and NT clauses of IIAs, differential taxation can be argued as nondiscriminatory on the basis that difference in harm means the products are not “alike” or, in the alternative, discrimination is justified based on scientific evidence of differences in harm and rational reasons for the health-protective role of differential taxation (63). CHAP T ER 4. PO LI T I C AL ECO N OMY 207 Where discrimination is inadvertent, lack of an intention to discriminate is not sufficient as a defence for breach of obligations under IIAs or the GATT (73, 76). Policy-makers should carefully scrutinize measures to determine: 1. whether an aspect of a tax measure’s design or implementation may be more to the detriment of imports or foreign investors than of local products or domestic investors; 2. whether the potentially discriminatory aspect of the tax measure serves any useful purpose in supporting the tax measure (i.e. it is needed to achieve the health goal); 3. whether there is any reasonable alternative that could achieve the same effect without the potential for discrimination; (i.e. it is indispensable) and 4. when it is needed and indispensable there is a good chance that it will be defensible. The case studies below provide examples of discrimination arising in connection with a tobacco tax measure. CASE STUDY 9 (NEGATIVE): Thailand – Customs and Fiscal Measures on Cigarettes from the Philippines (2010 WTO panel) The facts of this case are presented in case study 3. This case study examines claims of discrimination rather than the procedural issues. DISCRIMINATION EXPLANATION LESSONS Thailand implemented its policy for determining the tax base for VAT on cigarettes inconsistently (61). Thailand applied a methodology in fixing the tax base, in particular a marketing cost component, of imported cigarettes that differed from that for local products (61). This resulted in the marketing cost component for the imported cigarettes being higher than it would have been under the general methodology. This difference in treatment was insufficiently justified and therefore considered discriminatory. As there is potential for inadvertent discrimination when the base for an ad valorem tax is fixed, tax base determinations must be consistent and well- reasoned (61). This case study demonstrates how policy- makers need to take care in designing and implementing ad valorem taxes to ensure they are nondiscriminatory and legally defensible. 208 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Thailand’s VAT rebate policy imposed a potentially higher tax burden and also created more burdensome administrative requirements for imported cigarettes (61). Resellers of cigarettes produced by a government entity were granted an exemption from VAT (61). Although resellers of imported cigarettes would be eligible for a tax credit on their VAT, this was not an automatic process (61). The distinct treatment of resellers of imported cigarettes and those of local cigarettes resulted in the risk that there would be a higher VAT burden for the former (61). The distinct treatment also imposed an additional administrative burden on resellers of imported cigarettes and altered conditions of competition (61). Rules for the collection and enforcement of tax obligations should be the same, or as similar as practicable, in both form and effect for domestic and imported tobacco products. CASE STUDY 10 (NEGATIVE): Dominican Republic – Measures Affecting the Importation and Internal Sale of Cigarettes (2004 WTO panel; 2005 WTO Appellate Body) Under article XX(d) of the GATT, discrimination that is necessary to secure com- pliance with a legitimate tax measure will be justified provided there is no less- discriminatory alternative. In this case, this justification was used unsuccessfully. DISCRIMINATION EXPLANATION LESSONS The Dominican Republic’s tax stamp regulations were discriminatory towards imported goods (77). Under the regulations, all cigarette packs had to be affixed with tax stamps, but imported cigarettes were to be affixed with tax stamps under the supervision of local tax authorities following importation, while locally manufactured cigarettes could be affixed with a tax stamp in the course of production. This de facto distinction between local and imported products modified the conditions of competition to the detriment of imported cigarettes by (1) increasing costs for importers and (2) impairing the aesthetics of imported products (77). The panel did not consider this discrimination justified: it was not necessary for the enforcement of tax measures, because less restrictive alternatives were available such as permitting importers to affix tax stamps during the course of production (Dominican Republic – measures affecting) (77). The panel’s findings were upheld on appeal (78). Policies crafted to ensure compliance with tax measures need to also be nondiscriminatory. Discrimination claims can arise when compliance costs are higher for imports than for local products and this de facto distinction is avoidable. It is important to consider whether less burdensome alternatives may achieve the same objective. CHAP T ER 4. PO LI T I C AL ECO N OMY 209 CASE STUDY 11 (NEGATIVE): Feldman Karpa v Mexico (2002 ICSID [International Centre for Settlement of Investment Disputes] Arbitral Tribunal) Arbitral tribunals have accepted differences in treatment accorded to investors protected by IIAs when there is a legitimate connection between the distinctions drawn and public welfare objectives (76). The facts of this case are presented in case study 5. This case study examines aspects of the case involving the investor’s claim of discrimination, rather than the substantive issue of expropriation. Claims of discrimination are made on different grounds than claims for expropriation, which is why the case was decided differently on this claim. DISCRIMINATION LESSONS Denial of foreign investors’ claims for tax rebates. Tax rebate claims were granted to similar local investors (65), which was a violation of an IIA’s national treatment clause (65). Foreign and local investors must be treated similarly, and consistent and well-documented policies must be used to guide administrative decisions. The denial of the rebates may have been justified, but the government was unable to establish this due to a lack of documentation. Avoiding the investment incentives trap Investor-state contracts between the tobacco industry and governments should be avoided. They are not merely “contractual” in the domestic law sense, as even in the absence of an applicable IIA, they can be internationalized to provide inves- tors the right to (1) remove dispute settlement from the state’s court in favour of independent arbitration and (2) remove the dispute from the state’s legal framework in favour of general principles of law (63, 76). Commitments under these clauses cannot, therefore, be legislatively moderated or extinguished, nor can liability be limited within domestic courts that may be more likely to favour the state’s right to regulate in favour of public health (76). Investor-state contracts and other noncontractual inducements can be further internationalized by umbrella clauses within IIAs. Such clauses make reneging on undertakings assumed towards investors a breach of the IIA (76). Moreover, even in the absence of an umbrella clause, contracts and inducement can underpin a claim for legitimate expectation and breach of fair and equitable treatment and can also strengthen an investor’s claim for indirect expropriation (63). Arbitral awards make clear that although taxes can be expected to vary and tobacco will be regulated, investors can have the legitimate expectation that states will abide by formal inducements and written contractual undertakings. A common clause within investor-state contracts, the stabilization clause, is ruinous to evidence-based tobacco control’s most effective measure: excise tax in- creases. Stabilization clauses purport to freeze specific domestic law from the time 210 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of investment (63). Seemingly less onerous, economic equilibrium clauses require contracting states to compensate for regulatory changes that negatively affect an investment’s value (63). There is little difference in effect between these two types of clauses: liability for the cost of breaching an equilibrium economic clause can be onerous enough to make it fiscally challenging and politically unpalatable. CASE STUDY 12: An investor-state contract A state entered into an investment agreement with a TTC in 2001 on the privatiza- tion of its state-owned tobacco enterprise and creation of a joint venture. This investment was to provide economic benefits under the agreement: the joint venture would increase exports and profit using the TTC’s cash and expertise while also ensuring prioritization of local employment, manufacturing and resources. The final investor-state contract included a form of economic equilibrium clause under which any increase in the excise tax rates applied to the company’s tobacco products before a set date would be compensable. While the agreement was not removed from the state’s law, it provided for independent arbitration in case of a dispute over its compensation. In addition, there is a bilateral investment treaty between the host state and another state in which the TTC’s subsidiary has residence that includes a FET clause – this could buttress, if needed, the protection provided by the stand-alone arrangements of the investor-state contract. There were similar less-formal inducements offered to a separate TTC. The extent to which incentives have been granted to the tobacco industry is unknown, but contracts and inducements are likely to be offered in the context of the privatization of state-owned tobacco interests and in dealings between investors and state-owned tobacco enterprises (63). Although countries have been entrapped by their incentives to industry, the investor-state contract provides the clearest example of how undertakings and inducements with the tobacco industry under- mine tobacco control (56, 79–80). States should avoid offering industry incentives and, in particular, entering into contractual undertakings with the industry. More systematically, government should consider avoiding IIAs that elevate incentives and inducements above sensible and reasonable regulation. 4.2.3 CONCLUSIONS Health-protective and origin-neutral tobacco excise taxes are legally defensible, and industry threats are usually baseless. There are, however, certain rules governing procedure, design and consultation that governments may need to consider: 1. Governments should be aware of the standard of consultation required under do- mestic law and any applicable international obligations (case studies 1, 2 and 3). CHAP T ER 4. PO LI T I C AL ECO N OMY 211 It is important to distance the tobacco industry from the policy-making process to the extent that this is permissible. Do not grant the industry special consideration, but do ensure that it is consulted with as required – for example, by providing public meetings, timely information and the ability to submit industry views – while being aware of potential procedural manipulation (case studies 1, 2 and 3). 2. Excise tax is generally safe from challenges that claim it is confiscation or expropriation under domestic or international law (case studies 4 and 5). But express limits on taxation can be found in other laws or a country’s constitution or in the limits of the power to tax granted to an authority (case studies 6 and 7). 3. Explicit and de facto discrimination against foreign tobacco products or investors must be avoided in the design, implementation or enforcement of tax measures (case studies 8, 9, 10 and 11). Legal issues may arise not from the tax measure itself, but rather from ancillary measures that support its implementation (case studies 9 and 10). 4. Explicit differentiation between products based on their effect on health may be challenged as discrimination if it falls heaviest on imported products and has to be justified on the basis of evidence of impact on health and a lack of alternatives. 5. Investment incentives in the form of inducements or contractual undertak- ings should not be offered, as these may be binding (case study 12) or may ground a challenge under an IIA; they are also contrary to the WHO FCTC Article 5.3 Guidelines. 212 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.3 SCARE TACTIC A: ANTI-POOR RHETORIC (REGRESSIVITY) 4.3.1 INTRODUCTION In their efforts to lobby against tax increases, the tobacco industry and its affiliates often claim that increases in tobacco taxation will hurt the poor (81–82). This argu- ment is based on the concept of regressivity in relation to taxation. Conceptually, a tax can be regressive if it means that lower-income people pay a greater proportion of their household income to meet the tax burden than do wealthy people. In other words, the tax burden tends to be relatively higher for lower-income households than for middle- and high-income households. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use. Second, higher tobacco taxes and prices can induce behaviour change among the population, as reflected in the price elasticity of demand (83–84). In combination, these broader considerations effectively make tobacco taxation a progressive – rather than regressive – public health intervention. 4.3.2 REGRESSIVITY AND THE BROADER PERSPECTIVE In a narrow sense, tobacco taxation can be seen as regressive because lower-income people must allocate a relatively greater proportion of their household income than wealthy people to pay for tobacco products when those products become more expensive following a tax increase. In many countries, people from lower-income groups use tobacco more than other people (85). A systematic literature review by WHO found a robust association between lower income and a higher prevalence of current smoking among adults, both men and women (86). This finding was consistent across three decades of studies, across most geographic regions and across countries of different income classifications. For example, in India, high rates of tobacco use – i.e. use by more than 30% of the adult population – are found only in lower-income states such as Assam and Odisha, where net state domestic product is still below 100 000 rupees per capita (see Fig. 4.3.1) (87). CHAP T ER 4. PO LI T I C AL ECO N OMY 213 Fig. 4.3.1 Relationship between adult tobacco use and net state domestic product per capita in states and union territories of India, 2016–2017 Source: (87). However, this finding does not account for broader health and economic factors that determine the full impact on households. Tobacco taxation can in fact be viewed as a progressive – or pro-poor – policy when these wider considerations are properly ac- counted for and explained. In terms of health concerns, the relatively high use of tobac- co among low-income populations translates into a much greater burden of tobacco- attributable diseases for these populations, including higher morbidity and mortality. Low-income groups are also less able to afford medical care to treat tobacco- attributable diseases, and large out-of-pocket medical expenditures can further impoverish many families. Consequently, many poor individuals do not get or even seek the medical care they need. One study found that in Bangladesh, 55% of patients diagnosed with a tobacco-attributable illness did not seek further medical care. This lack of health care utilization was attributed in part to prohibitively high out-of-pocket treatment costs (88). The combination of high rates of tobacco use and lack of access to affordable medical care means that tobacco use measurably contributes to the poverty rate in a number of high-tobacco-burden countries, including China and India (89–90). A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India A du lt (> 15 y ea rs ) t ob ac co u se (% ) Net state domestic product per capita (thousands of rupees) 0 10 20 30 40 50 500 100 150 200 250 300 350 400 The dierent states and union territories of India 214 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N The poor are also known to be more price-sensitive than the wealthy; lower-income smokers exhibit higher price elasticities than their higher-income counterparts. This is demonstrated in recent studies by the World Bank, findings of which are shown in Fig. 4.3.2 (91). The poor respond more strongly to higher tobacco taxes and prices by reducing their use of tobacco products more than others, and thus they benefit disproportionately in terms of avoiding tobacco-related deaths, diseases and associated medical costs. A similar conclusion was drawn in a systematic review of the population impact of tobacco control policies on socioeconomic inequities in high-income countries at the late stage of the tobacco epidemic (92). The review found 16 relevant studies relating to taxation, only one of which found a regressive association between tax and the social economic gradient (seven found a progressive impact, while the others produced mixed results). Fig. 4.3.2 Price elasticity of tobacco consumption, medium estimate, by decile Source: (91). This wider economic perspective is explained in the World Bank’s Extended Cost- Benefit Analysis (ECBA) framework, which assesses the distributional impact of tobacco tax increases on health, among other factors (82, 83). That is, the ECBA framework looks beyond the simple or partial definition of regressivity (i.e. impact El as tic it y Deciles -1.2 1 2 3 4 5 6 7 8 9 10 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina CHAP T ER 4. PO LI T I C AL ECO N OMY 215 on household expenditure by income levels) to capture the full distribution of benefits, including improved health and income. The ECBA framework has been applied in studies of various countries, including Bangladesh, Bosnia and Herzegovina, Chile, Indonesia, Republic of Moldova, South Africa, the Russian Federation, Ukraine and Viet Nam. The evidence from these studies supports the view that effective tobacco tax policies can generate pro-poor and welfare-improving outcomes. When reductions in medical expenditures and additional years of working life that result from lower smoking-related mortality are taken into account, the overall policy of tobacco tax increases becomes progressive rather than regressive (see Fig. 4.3.3) (84). A similar conclusion has been reached in studies of high-income countries, such as the United States, where a tobacco tax increase was enacted in 2009 (93). Fig. 4.3.3 Impact of a 100% price increase, with medium elasticities, by deciles Source: (91). Tobacco tax increases will also often lead wealthier smokers to contribute relatively more than poorer smokers to the overall amount of tax revenue collected. This is because poorer smokers reduce their consumption the most, since they are more price- sensitive and wealthier smokers also tend to purchase premium (higher-priced and In co m e ga in s (% ) Deciles -1 2 3 54 6 7 8 9 10 0 1 2 3 4 5 1 Ukraine Republic of Moldova South Africa Bangladesh Indonesia Russian Federation Bosnia and Herzegovina 216 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N taxed) tobacco products (94). Hence, higher tobacco taxes can be seen as progressive in terms of additional revenue collection and health gains resulting from cessation, as well as from preventing the young from taking up smoking in the first place. One study from China suggests that a 50% tax increase would raise US$ 703 billion over 50 years, with just 14% of this increase being borne by smokers in the lowest income quintile (95). In addition, the tax increase would yield a savings of US$ 24 billion in expenditures on tobacco-related diseases, with about 28% of these savings being enjoyed by smokers in the lowest income quintile. The ECBA framework actually presents a rather conservative profile of the net benefits of raising tobacco taxes, since it does not include other sources of gain, such as reduced harm from exposure to second-hand smoke, increased productivity and the potential for poor households to benefit from social programmes funded through increased tax revenues (96). Assessments of the distributive impact of the 2009 tobacco tax increase in the United States found that the overall progressivity of the increase was enhanced by the tax rev- enue being used to expand health insurance coverage for children of low- and middle- income families (97). Accounting for this expanded coverage added to the progressiv- ity of the overall legislative package, the bottom line being that the impacts are positive for lower-income quintiles and greatest, on average, for low-income households (93). Similarly, a large proportion of the tobacco tax revenues from the Philippines’ so-called Sin Tax Reform was used to subsidize universal health coverage (UHC) for poor and near-poor families. Globally, 37 countries are known to earmark some tobacco tax revenues for health programs, with many of these programs indirectly benefiting the poor and less-advantaged disproportionately more than other groups (27) (for details on earmarking, see section 4.6). 4.3.3 CONCLUSIONS Contrary to the perception of tobacco taxation being regressive, it is a strong pro- poor policy when the broader economic impacts are taken into consideration. The tax burden is not a complete indicator of regressivity, since it does not include the negative health and economic impacts of tobacco-attributable diseases or the positive impacts of behaviour change in response to tax and price increases. The health and economic burdens of tobacco-attributable diseases fall dispropor- tionately on the poor, who tend to have higher tobacco use and are also the least able to afford the necessary medical care. Because the poor tend to be more price-sensitive, they curtail their use and consumption more significantly than wealthier smokers in response to tax increases, which in turn reduces their downstream health and economic costs. Tobacco taxation can be made even more progressive by earmarking or allocating tobacco tax revenues for social goods and services that benefit the poor (see section 4.6). CHAP T ER 4. PO LI T I C AL ECO N OMY 217 4.4 SCARE TACTIC R: REVENUE REDUCTION 4.4.1 INTRODUCTION The tobacco industry and its allies argue that tobacco tax increases result in reduced tax revenues for the government. According to them, the reduction in revenues is caused either by substitution to cheaper, lower-taxed or smuggled tobacco products or by reductions in consumption overall (98–99). The tobacco industry often refers to the Laffer curve to make this argument. According to this curve, revenues increase along with tax rates up to a certain point, after which further increasing tax rates leads to declining revenues. When considering tobacco taxes, the tobacco industry assumes that countries are already approaching or are even beyond the critical tax rate level (98). However, the argument rests on a narrow theoretical and empirically unsubstanti- ated foundation (98–100). The price inelastic demand for tobacco and the relatively low tax share in prices in many countries explain the win-win for public health and finance, i.e. that declines in consumption and increases in revenues can occur simultaneously (98, 101). Furthermore, many country examples (see case studies below) demonstrate that well-designed and well-implemented tobacco tax increases lead to increases in revenue, at least in the short to medium term (98, 100). Although consumption will diminish with a tobacco tax increase, the percentage increase in excise tax per unit is greater than the percentage decrease in tobacco consumption, cancelling out at least some of the effect of reduced consumption on revenue (98–99). A change in the tax rate, with all other factors influencing consumption kept constant, corresponds to a change in the tax revenue and is represented by a move- ment along the Laffer curve. As the tax rate changes, so does the elasticity of the tax base; each point on the Laffer curve corresponds to a different tax base elasticity. When one or more of the other factors changes, this affects the position of the curve, and the tax base elasticity changes at a given tax rate. For example, a successful smoke-free policy or advertising ban that reduces the demand for tobacco shifts the curve down, reducing the tax revenue potential for each tax rate. To demonstrate that few, if any, countries are beyond the revenue-maximizing point on the Laffer curve, Table 4.7 shows the revenue impact of increasing excise taxes under different scenarios, using different price elasticities of demand, different levels of tax increases and different starting tax shares, depending on country income levels. This is the tax base elasticity approach from which the Laffer curve is derived (for more details, see section 2.2.3 and Annex 2.2). The total and excise tax shares shown are weighted averages for each country income group, calculated from the RGTE dataset. The revenue gains were simulated using progressive levels of excise tax increases (25%, 50%, 75% and 100%) and varying price elasticities of demand (-0.4 to -1.2). 218 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.7 Percent increase in excise revenues under different scenarios of tax levels, tax increases and price elasticities20 Income group Total tax as % of retail price Excise tax as % of retail price Increase in excise tax Increase in excise revenue when price elasticity of demand is: -0.4 -0.6 -0.8 -1.0 -1.2 LOW INCOME 38% 22% 25% 22% 20% 19% 17% 16% 50% 43% 39% 36% 33% 29% 75% 63% 57% 52% 46% 41% 100% 82% 74% 66% 59% 51% MIDDLE INCOME 58% 41% 25% 19% 17% 14% 11% 9% 50% 37% 31% 26% 20% 15% 75% 54% 45% 36% 27% 19% 100% 71% 57% 45% 34% 23% HIGH INCOME 68% 55% 25% 18% 15% 11% 8% 5% 50% 35% 27% 21% 14% 8% 75% 50% 39% 29% 19% 10% 100% 65% 50% 36% 23% 11% Source: Authors’ calculations using data from the RGTE (27).21 Substantial revenue increases occurred in all the scenarios that were considered in the simulation. These results reaffirm much of what is already known, i.e. that higher tax increases generate higher revenue gains, and that these gains increase with the increasing inelasticity of demand. Even when demand is relatively price elastic (-1.2), the simulation predicts a gain in revenue. The tax share in price also affects revenue potential. The lower the tax share in price, the larger the revenue potential. This suggests that revenue reductions as a result of an excise tax increase will occur only if the scenario is extreme (i.e. a very elastic demand coupled with a very high current tax share). It is important to note that the vast empirical literature 20 These projections use 2018 data from 185 countries. The countries were classified according to World Bank income group, with the average total tax share, excise tax share and VAT/sales tax share for each country weighted according to the number of current adult cigarette smokers. To calculate the projected revenue for each stated elasticity, it was assumed that there would be full pass-through of the excise tax increase, along with constant percentages of non-excise taxes (VAT/sales tax) as a share of the retail price. The consequent changes in price were multiplied against the respective elasticities to derive the expected change in consumption. The projected revenues could be easily computed by multiplying the new consumption figures against the increased excise tax rates. 21 These calculations do not take into account brand substitution (cross-price elasticities), income ef- fects or illicit trade. The excise tax was assumed to be a specific tax, while the non-excise taxes (VAT and others) were bundled and treated as an ad valorem tax with retail price as the tax base. The difference between retail price minus all taxes was also assumed to be constant, with full pass-through of the tax increase to consumers. CHAP T ER 4. PO LI T I C AL ECO N OMY 219 shows tobacco to be universally inelastic; thus the extreme scenario should not be given credence by policy-makers. Furthermore, as demonstrated by the data in Table 4.7, tax shares in most countries are relatively low and reinforce the revenue potential of tobacco tax increases. The revenue potential of tobacco taxes is indeed quite significant. It is estimated that in 2018, excise taxes on cigarettes generated a total of US$ 361 billion in revenues worldwide, including US$ 162 billion in LMICs. If all countries were to raise excise rates by the equivalent of US$ 1 per pack of cigarettes, the amount of excise revenue would increase by between US$ 178 billion and US$ 219 billion, or by 49–61% at 2018 levels. LMICs would gain the most from these tax increases, with excise revenues in these countries increasing by US$ 133 billion to US$ 167 billion, or by 82–103%.22 Revenue reduction in the countries examined was due to other causes, not the tax increase per se. For example, Tonga significantly increased its excise tax on cigarettes in 2016 and saw a very sharp decrease in its consumption (40% decrease), followed by a revenue decrease. This occurred because 20% of smokers switched to an untaxed, cheap local loose tobacco product called Tapaka Tonga (102). The lesson learned was that Tonga needed to tax all its tobacco products at the same level to avoid substitution to lower-price/untaxed tobacco products. Another example of revenue decrease that was not related to tax increases but rather to tax administration mismanagement is the case of South Africa (see explanation in the case study later in this section). Finally, declines in revenue due to long-term declining trends in tobacco use should not be confused with being beyond the revenue-maximizing point of the Laffer curve. For example, in the United Kingdom, where long-term declines in tobacco use are being experienced, a nominal decline in revenues occurred between 2017 and 2018 even though excise taxes remained unchanged. Conversely, even countries with very high tobacco excise rates experience increases in revenues as a result of tobacco tax increases (see the case study of Australia below) (98, 100). This suggests that few countries, if any, are beyond the revenue-maximizing point on the Laffer curve. Tobacco consumption is expected to be tax inelastic, even if demand becomes effectively price elastic as a result of successful tobacco control interventions. Taxation serves as an instrument for both fiscal and public health objectives. If after successful tobacco control interventions, prices reach levels where demand is elastic, the tax base is still most likely to be inelastic due to tax undershifting, since overshifting is not a good pricing policy when demand is elastic (for a more detailed discussion on the shifting of tax, see section 2.2.2). In other words, a tax rate increase in combination 22 Goodchild M, Perucic AM, Paul J. Tobacco taxation as a strategy to achieve global targets for smoking prevalence. Unpublished manuscript. October 2020. 220 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N with non-price tobacco control measures, which make consumers more sensitive to price (tax) increases, leads to declining but still positive marginal revenues. In the long run, tobacco control policies, including price and tax measures, may be so successful in reducing consumption that revenues will plateau or fall. This is ultimately the long-term policy goal. Reducing the impact of the policies and ending the global tobacco epidemic is the aim of tobacco control and not something to be avoided. However, ending the global tobacco epidemic is unfortunately not foreseen in the short to medium term; therefore, governments can currently rely upon tobacco taxes as a reliable source of revenue (103). 4.4.2 THE REVENUE IMPACT OF EXCISE TAX INCREASES: CASE STUDIES The following case studies illustrate four key points: (1) large and regular tax increases result in large and consistent revenue increases; (2) countries with high taxes and falling prevalence of tobacco use can still increase revenue with tax increases; (3) countries that reduce taxes experience revenue declines; and (4) countries that increase taxes in the face of illicit trade still increase revenue. Large and regular tax increases usually mean large and consistent revenue increases South Africa’s experience shows how successive tax increases, well above inflation and year after year, generate additional revenues even after taxes have been increased substantially. After two decades of declining real revenue in the 1970s and 1980s as real excise per pack declined, South Africa implemented successive excise tax increases from 1994 until 2011 (Fig. 4.4.1) (98, 104). After adjusting for inflation, this resulted in a real excise tax revenue increase of 245% (98). Revenues began to plateau from 2012 as tax increases stalled. They began to decline after 2015 – not due to tax increases, however, but due to a dramatic decline in administrative capacity and enforcement measures exacerbated by large-scale corruption in the government, including the tax administration authority (105). The rapid and catastrophic decline in tax administration and enforcement has been the subject of much attention (106). CHAP T ER 4. PO LI T I C AL ECO N OMY 221 Fig. 4.4.1 Real excise tax per pack of cigarettes and real excise tax revenue in South Africa, 1961–2020 Source: Data shared by University of Cape Town, 2020. Similarly, the Philippines provides a compelling example of how large and regular tax increases alongside reforms to tax structure can lead to large and consistent revenue increases – in this case, also through an accompanying reform to the tax structure (Fig. 4.4.2) (98). The 2012 Sin Tax Law consolidated the country’s four tax tiers into two by 2013 and established a uniform structure by 2017. The same law provided for large, progressive increases across the board, but in particular for the lowest tax categories (98). Not only were the revenue gains substantial, they exceeded all the projections for 2013–2017 made prior to the law’s passage (98). Excise tax per pack Excise tax revenue Ra nd s pe r p ac k (c on st an t 2 02 0 ra nd s) Excise revenue (constant 2020 rands) 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11 20 13 20 15 20 17 20 19 0 0 5 6 10 10 15 14 20 18 222 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.2 Real excise tax per pack of cigarettes (multitier, two-tier, unitary) and total tobacco real excise tax revenue in the Philippines, 2012–2018 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2012 as the base year. Sources: (107, 108 and data shared by the Philippines Department of Finance, September 2020). Ukraine is another example of a country that has regularly increased taxes over the past 10 years and has experienced increased revenues along with decreases in consumption and the number of smokers. Figure 4.4.3 the shows the trends in excise tax, revenues, cigarette sales and number of smokers in 2008–2017. Increases in excise rates were consistently accompanied by increases in revenues. In 2014–2015, excise tax was not increased above inflation (and inflation, especially in 2015, was very high, at 48.7%), so real values of excise and revenues went down. But it is evident from the data that revenues closely follow the path of excise levels even when sales go down. 12 Re al e xc is e ta x pe r p ac k, P hi lip pi ne p es os (2 01 2 ba se ) Billions (Philippine pesos) in tobacco real excise tax revenue (2012 base) 00 20 60 100 140 2012 HIGH PREMIUM SIN TAX LAW RA 10354 TRAIN LAW RA 10963 UNITARY RATE 2013 2014 2015 2016 2017 2018 5 10 15 20 25 30 32 .1 9 68 .6 6 71 .0 9 93 .5 2 87 .2 0 96 .0 1 11 4. 39 24.4 25.4 26.2 26.8 26.9 27.5 11.7 16 23.1 19.6 LOW Real tobacco excise revenues Dierent levels of real tobacco excise tax per pack MEDIUM 7.6 2.7 28.3 CHAP T ER 4. PO LI T I C AL ECO N OMY 223 Fig. 4.4.3 Average real cigarette excise tax rates, real cigarettes excise tax revenues (base year 2008) and cigarette sales and number of cigarette smokers in Ukraine, 2008–2017 Note: Data were adjusted for inflation, using annual percentage change of average consumer prices from the IMF World Economic Outlook, April 2020, and using 2008 as the base year. Source: Data provided by Konstantin Krasovsky, July 2020. Countries with high tax and falling prevalence of tobacco use can still increase revenue with tax increases Countries with already high tobacco taxes and rapidly diminishing tobacco use can still increase revenue by increasing taxes (98, 109). Australia has implemented comprehensive tobacco control policies and enacted consistent tobacco tax increases on top of what were already some of the highest tax rates in the world (see Fig. 4.4.4). Between 2001 and 2010, revenue increased with increasing tax rates, but in real terms (inflation-adjusted) it remained static (109). Then, in 2010, a 25% excise tax increase was introduced, with large annual increases scheduled from 2013 onward (98, 109). The result of this tax policy has been consistent and large increases in revenue year after year for nearly a decade, even when the increases were being made on already high tax rates.23 23 The apparent reduction in revenues in 2012 and 2013 was due to a change in the source of the data for 2001–2011 and 2012–2016. Data for 2012 and 2013 do not include customs duty, while all other years do. 73 95 125 112 88 82 75 76 67 11.8 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Number of daily smokers, in millions Real tobacco excise revenue, in billions of Ukrainian hryvnia Real average excise per cigarette packs of 20, Ukrainian hryvnia Number of taxed cigarettes (sales), in billions of sticks 0.6 3.6 10.1 9.2 8.7 8.6 8.4 8.1 7.3 6.2 6.5 6.3 7.8 10.3 11.2 12.1 13 1.4 2.2 2.5 2.9 3.5 3.2 2.7 3.3 4 12.7 13.4 74 9.7 224 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.4 Real excise tax and customs duty per stick and real total revenue (all tobacco products) in Australia, 2001–2019 Notes: Rates published by Australian Taxation Office and Australia Department of Immigration and Border Protection, adjusted using Australian Bureau of Statistics Consumer Price Index rates. The 2011–2012 and 2012–2013 figures do not include customs duty, which explains the apparent decline in revenue. Using official disclosures, Scollo and Bayly estimate that duties in these years were $7397.2 and $7687.2 respectively (110). Sources: (109, 110). Countries that reduced taxes and saw revenues decline Prior to 1982, Canada lowered taxes on cigarettes and experienced declining revenues as well as increased smoking – particularly among youth. Subsequent fivefold in- creases in cigarette taxes between 1982 and 1992 resulted in more revenue, increases in retail price and substantial reductions in consumption, with teenage smoking declining by nearly two thirds (5). In the early 1990s, a growing illicit trade in ciga- rettes emerged in which Canadian cigarettes exported to the United States were then smuggled back into Canada (5). The tobacco industry – which was later found to be complicit in and profiting from this illicit trade – sought to frame Canada’s high tax rates as the cause of smuggling (111–112) and succeeded in convincing the federal government, as well as six provincial governments, to make massive reductions in the tobacco tax (111–112). As a result, federal tax revenues fell significantly – more than twice as much as the government had predicted – and smoking rates among both adults and youth began to increase (5, 112). The Canadian government later changed its strategy, and the federal excise tax was restored, resulting in increased Specic excise/ duty per stick (cigarettes/cigars less than 0.8 g) Total revenue (all tobacco products) To ta l c us to m s/ du ty re ve nu e (in a tio n, a dj us te d, 20 19 A us tr al ia n do lla rs , m ill io ns ) Total excise/duty per stick (ination, adjusted, 2019 A ustralian dollars) 0 3 000 6 000 9 000 12 000 $0.8 $0.6 $0.4 $0.2 $0 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 CHAP T ER 4. PO LI T I C AL ECO N OMY 225 revenues and decreased smoking (5). Canada’s focus then shifted to using customs enforcement, rather than tax rates, as the best means of countering illicit trade (5). Countries that increased taxes in the face of illicit trade and still increased revenue As discussed in section 4.1, the tobacco industry exploits illicit trade as a strategy to undermine tobacco tax policy, with the goal of deterring governments from increasing tobacco taxes. The narrative that has been created is that higher tobacco tax rates result in increased illicit trade and undermine the policy goals by resulting in lower (or no) declines in tobacco use or lower (or no) increases or even decreases in revenue. However, as shown in section 4.1, the empirical evidence does not sup- port the industry arguments. Furthermore, the evidence shows that the industry and its allies have consistently overstated and exaggerated the scale and extent of illicit trade (see section 4.1). As was the case in Canada, Brazil’s tobacco tax policy suffered from a fear that the illicit market would expand unless it was undercut by price competition in the legal market, which it was thought could be best encouraged through tax cuts (20). Real excise tax rates declined from 1999 until the mid 2000s, as nominal increases were below the rate of inflation. This resulted in declines in real tax revenues (20). In these years, the tobacco industry used the tax cuts to increase profit margins rather than decrease prices and outcompete the illicit market, while also exaggerating the size and scope of the illicit trade problem (20). This caused the industry’s argument on illicit trade and revenue to lose credibility and resulted in increases in tax rates from 2007 onwards, with a major reform passed in 2011 (20). Tobacco excise rates and minimum prices were scheduled by the law to increase at levels above expected inflation from 2011 until 2015 (20). This resulted in substantial increases in the tobacco excise per pack, as well as overall revenue, which by 2015 had more than doubled from its low point in 2013 – equating to more than 50% in real terms (see Fig. 4.4.5). The success of this reform shows that revenues can be increased by higher rates despite the presence of a sizeable illicit market (113). More recent data show that revenues in Brazil declined in 2015 and 2016, coinciding with an increase in illicit trade, but also with an exceptionally bad economic recession that saw GDP decline by more than 3% in those years. 226 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Fig. 4.4.5 Average real excise tax per pack, real excise tax revenues and domestic cigarette sales in Brazil, 1999–2014 Notes: Data based on domestic sales and tobacco excise revenues, Federal Revenue Secretariat, indexed to 2013 Brazilian reals, using Consumer Price Index. Revenue collection indexed to 2013 reals, using Consumer Price Index. Source: (20). Improvements in tax administration and enforcement can also generate increases in revenues. In Kenya, several measures, including fiscal markings and, later, an advanced tracking and tracing system, improved collection, resulting in increases in both legal sales and tax revenues and a reduction in illicit sales (114). Moreover, these examples of poor governance indicate that attention should be focused on countries where a significant loss in administrative and enforcement capacity un- dermined revenue collection. 4.4.3 CONCLUSIONS The tobacco industry uses revenue concerns as a SCARE tactic to avoid, dilute and/or delay tobacco tax increases. The argument that higher taxes will decrease revenue is theoretically plausible, but real-world examples have demonstrated that this has not occurred. Furthermore, simulations show that even large tax increases in current average tax shares yield substantial revenue gains. The use of the Laffer curve by the tobacco industry should be challenged and refuted. The relatively price inelastic nature of cigarette demand combined with the Real excise tax amount per pack Domestic sales Real excise tax revenue D om es tic s al es (b ill io n pa ck s) Re al e xc is e ta x re ve nu es (b ill io n 20 13 re ai s) Real excise tax am ount per pack (2013 reais) 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 0 1 2 3 4 5 6 0 1 2 3 CHAP T ER 4. PO LI T I C AL ECO N OMY 227 low tax share and no overshifting of the tax means that most – if not all – countries are still far from the revenue-maximizing point, indicating that increases in taxes will lead to increases in revenues. The case studies in this section refute each of the tobacco industry’s arguments regarding alleged potential revenue loss due to tax increases. The experiences of South Africa, the Philippines and Ukraine demonstrate that large and regular tax increases result in large and consistent revenue increases. Well-designed tax structures have also proven to play an important role in generating revenues. The experience of Australia shows that even countries with already high tax rates and declining prevalence of tobacco use can increase revenues with regular, large tax increases. The experience of Canada warns against following the advice of the tobacco industry to decrease taxes as a way to fight illicit trade. It demonstrates clearly that decreasing tobacco taxes will decrease revenue and encourage consumption, rather than counteract illicit trade. The experience of Brazil shows that countries with substantial illicit trade issues can still increase revenue by increasing taxes. Finally, in the few cases where revenue decreases were seen, the reasons for the decreases were not strictly linked to tax increases. This was the case in Tonga, where the increase in tax was applied only to cigarettes and not to their close substitute, loose tobacco – leading smokers to switch products. In South Africa, a decrease in revenue was the result of the weakening of government institutions. And in Ukraine, real revenues decreased only during the two years when taxes were not increased above inflation. 228 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 4.5 SCARE TACTIC E: EMPLOYMENT 4.5.1 INTRODUCTION In opposing tax increases, the tobacco industry often seeks to frame tobacco taxes as an economic rather than a public health issue (5, 48, 53, 115). Particular emphasis is placed on the alleged threat that tax increases pose to employment in tobacco farm- ing and manufacturing, as well as other related industries (5). This so-called choice between health and jobs is, however, based on several false premises, including (5): 1. tobacco is a significant source of jobs within the context of broader labour markets, and domestic tobacco tax increases will have a drastic effect on domestic employment (48, 53, 116–117); 2. tobacco consumption is an indispensable engine for job creation (5, 48, 54); and 3. tobacco provides highly prosperous, sustainable and irreplaceable livelihoods (5, 53, 118). In reality, the relationship between tobacco taxation and employment is consider- ably more complex than the industry makes it out to be. In fact, there is ample evidence to show that tobacco taxes are a win-win for public health and the fiscal space, without measurable risks to employment. 4.5.2 THE LINK BETWEEN TOBACCO EMPLOYMENT AND TOBACCO TAX RATES Tobacco farming, production and manufacturing (including hand-rolling in some countries, most of them in South-East Asia) constitute a small proportion of the labour force, even in countries where the industry is most heavily concentrated (5, 48, 103, 116). Employment in tobacco farming and manufacturing has been declining globally due to advances in technology, trade liberalization, market consolidation and the privatization of formerly state-owned tobacco companies (5, 103, 119). These same trends have led to the heavy concentration of tobacco growing and manufacturing in only a handful of countries – and within these countries, often in only a small number of regions (103, 119–121). Even in those countries that lead in tobacco growing and manufacturing, tobacco’s overall share of total agricultural and manufacturing employment is relatively small and is often decreasing as efficiencies in production reduce labour intensity (5, 117–118, 122). Similarly, the industry’s claim that tobacco taxes reduce employment is exag- gerated and typically overlooks wider trends driving tobacco industry employment. Indeed, tobacco industry developments and innovations have played a greater role in the reduction of employment in the tobacco industry than have tobacco control policies (103). Despite industry claims that tobacco taxes can affect employment, CHAP T ER 4. PO LI T I C AL ECO N OMY 229 characteristics of the location of production – such as market size, labour costs, growing conditions and leaf preferences – have much more to do with tobacco industry interests than with the tobacco tax rate (103, 123). Moreover, jobs in countries that produce tobacco primarily for export are not greatly affected by reductions in local consumption resulting from tax increases (5, 103, 116, 123). Finally, it has been demonstrated that tobacco tax increases do not have a significant effect on employment in the retail sector, as most retail businesses sell other goods (103). Estimates of the gross employment impact of tobacco tax hikes demonstrate that job losses that do occur can be more than compensated for by increases in revenue. A 2018 World Bank study estimated that in Indonesia, for example, an ambitious tax reform that would simplify tiers and increase prices by close to 50% would reduce gross employment in the tobacco manufacturing sector by less than 0.5% (a loss of 2 914 jobs). The government could provide income support to the displaced workers (for example, through training, temporary transport/mobility or income support) with less than 2% of the revenue gained from the tax increase (117). Similarly, a 2019 study by Bangladesh’s National Board of Revenue estimated that a substantial increase in tobacco taxation would cause 7 012 lost jobs, but that the total income associated with these job losses in the bidi industry would amount to only 3.5% of the revenue gained (120). Accordingly, increased revenue can more than compensate for the expenditure of supporting those who lose jobs and need to acquire new skills before transitioning to new employment (120). Box 4.5.1 Employment fears deployed to frustrate tobacco tax reform in Indonesia In 2017, the Indonesian Ministry of Finance decided to implement tobacco tax in- creases by 2019 and tier simplification by 2021 (124). This resolution was, however, abandoned within a year, after a concerted campaign by tobacco industry actors and their allies to reframe the increase as an economic issue with a focus on, among other things, the effect the tobacco tax increase would have on employment (124). This defeat for the tobacco tax initiative came despite estimations of how the loss of income associated with lost jobs would be dwarfed by the additional revenue gained by the tax (117). Earlier analysis had estimated an overall large net positive impact on employment from tobacco tax increases (125), which illustrates how evidence that challenges assumptions around the negative socioeconomic impacts of tobacco control tends to be discounted (126). In this case, tobacco industry arguments seem to have resonated strongly with politicians from the electoral districts of West Java, East Java, Central Java and West Nusa Tenggara, where employment in tobacco farming and manufacturing is concentrated (124). Although tobacco manufacturing 230 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N represented only 5.13% of total manufacturing employment, the concentration of the job and economic activity meant that arguments regarding employment were particularly salient (117,119). Accordingly, tobacco industry arguments that tobacco taxation would negatively impact employment and farmer livelihoods prevailed, despite strong opposing evidence (119). Concentration of tobacco industry activity within countries poses particular obstacles to overcoming industry arguments on employment and needs to be given careful attention. 4.5.3 THE EVIDENCE ON THE NET EFFECT OF TOBACCO TAX INCREASES ON EMPLOYMENT A proper analysis of the effect of tobacco tax increases on employment must examine their impact on net or economywide employment. Decreases in expenditures on tobacco associated with tobacco control do not mean that expenditures simply disappear; rather, they are redistributed towards consumption of other goods and services, thereby generating employment elsewhere in the economy (5, 103, 123). Similarly, though the effect of higher tobacco taxes on net consumption is arguably more ambiguous, revenues from this intervention do generate spending, invest- ment and employment in public services such as health and education (5). Tobacco control polices usually have a marginal neutral or positive effect on net employment, particularly in countries that are net importers of raw or manufactured tobacco products, as expenditures on these imported items tend to flow out of the country (5). Export-oriented tobacco producers are less sensitive to local demand and are not significantly affected by domestic tobacco tax measures, which likely have a near-neutral net impact (5). In some cases, the net employment impact is a very small negative number, typically less than 1% (127–128). A recent study estimated that in the United Republic of Tanzania – a large tobacco-producing and exporting country – a 30% reduction in smoking prevalence would result in a net employment decline of just 0.5% across the economy as a whole (129). A similar study of Pakistan found that, with some variance depending on where spending was redistributed from tobacco consumption, the overall net effect on employment from a significant reduction in expenditure on cigarette employment – 1 billion rupees – would be a gain of between 6 651 and 5 803 jobs (122). This increase would occur because expenditure on cigarettes produces much less employment in the broader economy than expenditure on food and education (122). In the United Republic of Tanzania, as elsewhere, increased revenue could be used to assist those who lose employment with transitioning to new livelihoods. CHAP T ER 4. PO LI T I C AL ECO N OMY 231 4.5.4 THE VIABILITY OF BETTER LIVELIHOODS In arguing against tobacco tax increases, the tobacco industry advances the myth that people employed in tobacco production – particularly tobacco farming, but also manufacturing – lack any other prospect for a comparably attractive livelihood. However, studies based on extensive survey data in Indonesia, Kenya, Malawi, the Philippines and Zambia have shown that despite needing to commit significant amounts of labour to their crop, tobacco farmers often suffer losses rather than gain profits (119, 121, 123, 130–132). Furthermore, the Indonesian studies demonstrate that tobacco farming has a negative impact on household income and opportunity compared with the experience of other farming households that have given it up (119). Declines in consumption as a result of tobacco tax increases are gradual and susceptible to the same progressive adaptation that has occurred for decades (5, 103). While there will be a need in some countries for the government to help farmers transition to other crops or industries in the longer term, this process will not be a major short-term shock to employment or the wider economy (123). Because tobacco growing and manufacturing can be concentrated in just a few locations within a country, job losses within the tobacco industry might have a disproportionate effect in one location, while employment gains from reduced consumption may be spread across the whole country (120, 123). A study of the employment effects of tobacco tax increases in Bangladesh estimated that up to 60% of all job losses would occur in only two districts – among the poorest in the country – due to the high level of industry concentration (120). Studying the need for support, as well as the means of delivery and funding of support, is particularly necessary in these circumstances. Beyond the need to ensure equity and support employment, a failure to provide for targeted relief can exacerbate fear of job losses and may prove fatal to a tobacco tax proposal (120). Box 4.5.2: Supporting alternative livelihoods in the Philippines The Philippines earmarked 15% of the revenue from a 2012 increase in tobacco taxes to supporting economically viable alternative livelihoods for tobacco farmers and workers (5). Tobacco farming in the Philippines is regionally concentrated, and the tobacco industry had previously been successful in deploying concern for smallholder tobacco farmers to undermine tobacco control measures (136). The provision of economic support was a politically effective countermeasure to tobacco industry SCARE tactics and eased the tax increase’s passage. Given the Philippines’ integration with global tobacco markets and demand, tobacco farmers have not been seriously affected by the tobacco tax increase and 232 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N reduced domestic demand (108, 132). Nevertheless, transfers to tobacco-growing regions from the earmarked tax have been substantial (108). These funds are support- ing gradual transitions to alternative livelihoods, with farmers being encouraged to take up alternative crops, as well as establishing infrastructure, such as market-to-farm roads, that will make these alternative crops more economically viable (108, 133–134). Supporting alternative livelihoods for farmers and other tobacco workers is important because it can offset the political effect of industry arguments, even though domestic tax increases usually have only a modest and gradual effect on employment. There are various models for supporting alternative livelihoods when employment in the tobacco sector gradually diminishes due to decreases in either global or national demand. The Philippines is exemplary, but many other countries have either implemented or experimented with supporting crop transitions. Turkey’s alternative crop pro- gramme, implemented in anticipation of the privatization of the country’s cigarette monopoly, has proven effective in supporting many tobacco farmers’ move to other crops (135). Smaller-scale crop substitution projects in Kenya and Yunnan Province in China have shown how financial, regulatory and infrastructure support from government can contribute to crop transitions (5, 53). Argentina, Bangladesh, Mexico and the state of Maryland in the United States provide additional case studies of how governments can support these transitions (5, 136). 4.5.5 CONCLUSIONS The tobacco industry exaggerates the importance of tobacco employment and over- states the impact that domestic demand reduction due to local taxes will have on tobacco farmers serving a global market. The industry also simplifies employment’s relationship with taxation by focusing only on gross employment in tobacco, which ignores the reality that expenditures on tobacco do not disappear but rather are redistributed for other consumption that can produce a similar or higher number of jobs. Many detailed studies have found that tobacco growing is much less profitable and sustainable than the tobacco industry claims. Tobacco farmers throughout the world have successfully transitioned to other crops, although the transition often requires temporary or additional support from the government or other stakeholders. The extent of such support is moderated by the reality that transition from tobacco to other crops is a long-term consideration. CHAP T ER 4. PO LI T I C AL ECO N OMY 233 4.6 EARMARKING TOBACCO TAX REVENUES TO FUND HEALTH 4.6.1 INTRODUCTION Earmarking tax revenues involves the separation of all or a portion of revenue from a tax or group of taxes to be put aside for a specific purpose (137). Globally, more than 80 countries earmark for health (138), and 37 earmark tobacco tax revenues for health (27). There are two main types of earmarks: hard – also called substantive – and soft, or symbolic (139). Hard earmarks link the expenditure with a revenue source in legislation. This can limit funding if the earmarked revenues are the main source of funding, or it can cause surpluses to accrue wastefully when more revenues are raised than may be expended for the earmarked purpose. Soft earmarks include dedicated funds or commitments to use funds for a particular purpose. They are not necessarily legally binding. For example, in France, the ma- jority of tobacco tax revenue is used to fund social security (which includes health insurance and health care), but there is no hard, formal earmark (140). Earmarks can also be some combination of hard and soft. In the Philippines, tobacco tax earmarks are legally binding, but earmarked revenues go to the general fund, and the Department of Health must submit an annual budget for covered programs as part of its budget request. Earmarking is a broad and contentious topic that goes beyond the specifics of tobacco tax earmarking. Discussions on the topic fall within the ambit of public financial management, and earmarking generally is not encouraged. From a tobacco control perspective, however, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. It is a tool to improve the political economy of tobacco taxation; it is a secondary issue only, after the primary goal of reducing demand for tobacco. One way to use earmarking to improve the political economy of tobacco taxation is to link the payment of tax by tobacco users to benefits they will receive through the funding of complementary tobacco control programmes, such as cessation support, or through increased funding for health programmes on which they will rely disproportionately. This is known as the benefit principle. Earmarking for tobacco control makes sense, as its financial cost is relatively small and tobacco tax reduces demand more effectively when implemented within a package of complementary tobacco control measures. Another way earmarking improves the political economy of tobacco taxation is by safeguarding against any perceived or potential negative ramifications of the tax itself. This is important for neutralizing erroneous but often convincing tobacco industry arguments against effective tobacco tax policies. For example, the Philippines earmarks the bulk of the additional revenues from sin taxes for the health insurance premiums of the poor. In addition, a portion 234 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the country’s tobacco tax revenues is earmarked to provide for the economic well-being of tobacco growers and tobacco growing regions, with the general aim of promoting economically viable alternatives to tobacco farming and manufacturing as a safeguard against the potential for reduced domestic tobacco demand (141). Tobacco tax earmarks are complex, however, and care is required when determin- ing whether a particular earmark is needed on the grounds of political economy and justified on the grounds of equity and economic efficiency. In assessing tobacco tax earmarks, many of the same criteria that have been used to assess the appropriateness of generic earmarks also apply. In the rest of this section, these criteria are set out and matched with reasons for the ability of well-designed tobacco tax earmarks to fulfil them. The types and structures of tobacco tax earmarking are explored alongside descriptions of country experiences to provide guidance on how tobacco tax earmarks are used, when they are justified and the best ways to design them. 4.6.2 CIRCUMSTANCES IN WHICH EARMARKS MAY BE SUITABLE Scepticism about earmarking is both long-standing and justified, but much of the debate concerns earmarking generally and is not specifically concerned with the merits of tobacco tax earmarking (138). The main concerns raised about earmark- ing are listed in Table 4.8, accompanied by suggestions for how earmarks may be structured to address these concerns. Table 4.8 Concerns about earmarking and suggested safeguards to avoid the concerns MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Democratic accountability and oversight: earmarks undermine democratic processes by impeding legislative and executive oversight over expenditure. Establishing proper oversight and accountability procedures is important to ensure funds are not mismanaged (138). Additionally, if a soft earmark structure, which transfers revenue to the general fund from which it is then allocated, is adopted, this will not be a concern. Budget rigidity: earmarking may create budget rigidity that can lead to inefficient allocation of resources (138). An earmark’s particular design determines how much rigidity is introduced (138). Flexible soft earmarks are less prone to introducing rigidity than hard earmarks. Concerns about rigidity can be reduced by the inclusion of a sunset clause that ensures that the earmark is automatically discontinued or reviewed after a set period of time has elapsed (138). A further safeguard is to establish the earmark as a waterfall account, with any excess revenue over a set amount being allocated to the general fund. CHAP T ER 4. PO LI T I C AL ECO N OMY 235 MAJOR CONCERNS RAISED ABOUT EARMARKING HOW DESIGNING EARMARKING SAFEGUARDS CAN ADDRESS THE CONCERN Fragmentation: earmarking can result in fragmented and uncoordinated expenditures. This means policies complementary to the earmarked purpose but outside of its purview may be unfunded (138, 142). This is a legitimate concern. The negatives of fragmentation cannot be entirely eliminated, but they may be outweighed by the other merits of tobacco tax earmarking. That said, proposals for tobacco tax earmarks should be scrutinized to ensure that the funded purpose is at least cost-effective. Decreased equity: equity will decrease if individual access to benefits is narrowly defined according to payments made. This issue is not likely to arise with tobacco tax earmarks but is conceivable and something that should be guarded against in an earmark’s design. Capture by special interests: because earmarks are often the result of political expediency, an earmarked purpose may be determined by powerful special interests promoting a tax’s passage rather than careful prioritization of resources (138). Well-designed earmarks will guarantee funding for underresourced programmes and high- priority programmes. While the above concerns may be valid and design does matter, tobacco tax and other health-promoting taxes are not subject to the same concerns when it comes to the justifiability of earmarking their revenue (138, 143). Some of the factors that distinguish tobacco tax earmarks from more general critiques of earmarking are listed in Table 4.9 (138). Table 4.9 Concerns about earmarking and distinguishing factor for tobacco tax earmarks GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Procyclicality: earmarked revenues are often procyclical and susceptible to booms and busts (138–139, 142). Tobacco tax revenues are generally not cyclical (they are recession-proof ), and revenue is predictable relative to most other indirect and direct taxes (103). Budget rigidity Tobacco tax earmarks necessarily involve only a relatively small proportion of the budget; therefore, the effect of any rigidity will be relatively insignificant. Partly because of the relatively small amounts involved, there is only limited real-world evidence of tobacco tax earmarks having introduced harmful rigidity (143).24 24 See also the subsection on the amount of money associated with tobacco tax earmarks in section 4.6.3. 236 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N GENERAL CONCERNS RAISED BY EARMARKS DISTINGUISHING FACTOR FOR TOBACCO TAX EARMARKS Capture by special interests All earmarks should be scrutinized to ensure that their funded purpose is cost-effective. In the case of tobacco tax earmarks, however, political economy considerations may mean that it is sufficient for a low-priority purpose to be funded if the funding will unlock the political will needed for effective tobacco tax increases. In these cases, special interests are being purposefully catered to in order to ensure that tobacco tax increases occur. Of course, arguments against tobacco taxes and tobacco tax earmarking are led by special interests such as the tobacco industry (144–147). Insufficient revenue: the earmarked revenue source may become insufficient for funding its purpose (139, 142). Even though revenue may decrease in the long term when more tobacco users quit, such decrease is expected to be gradual. See Section 4.4 for details on how, with effective design, revenue will generally increase even with declining consumption. In addition to these reasons why general concerns about earmarking do not fully apply to well-designed tobacco tax earmarks, there are a number of compelling reasons for tobacco tax earmarking to finance tobacco control or public health that argue in favour of its implementation: • Significant increases in excise taxes are the most effective, as well as the most cost-effective mechanism for reducing consumption, but they are best implemented as a part of a package of complementary tobacco control measures, such as the WHO MPOWER package. Earmarking tobacco tax revenue for interventions that may not be funded otherwise can strengthen overall tobacco demand reduction (148). • The political economy of tobacco tax increases also makes earmarks attractive: – People have been shown to be more supportive of tobacco tax increases when they know the revenues will be used for targeted social programmes (143, 149–150). Earmarking tax revenue for health or tobacco control frames tobacco tax as a public health intervention in the minds of the public, which may otherwise view it as merely a revenue source (138). Research has shown that using earmarking to link a tobacco tax to health can also help raise awareness about the dangers of tobacco use (143). – When tobacco tax revenue is earmarked for programmes that benefit vulnerable groups, the tax becomes more equity-enhancing. Although lower socioeconomic groups and young adults receive disproportionate health and economic benefits from tobacco tax increases over the medium term, these groups will expend a greater share of their income in the CHAP T ER 4. PO LI T I C AL ECO N OMY 237 short term because of tobacco taxes. Earmarking tobacco tax revenue for programmes such as UHC or cessation services that provide immediate benefits to these groups neutralizes some critiques of tobacco taxation (e.g. the 2009 United States federal excise tax increase and the 2012 Philippines Sin Tax Reform illustrate how equity-enhancing earmarking facilitated passage of substantial tax rises) (93, 138, 143, 150). 4.6.3 EARMARKING PRACTICES AND COUNTRY EXAMPLES Earmarking tax revenues for health is a common practice in 80 countries. In 2018, 37 countries from all regions of the world earmarked tobacco tax revenues for health purposes.25 Case studies in the political economy of tobacco tax earmarking In 2012, the Philippines comprehensively reformed tobacco and alcohol excise taxes. Tobacco taxes were increased significantly, and numerous tax tiers were reduced to only one tier by 2017. Although increasing revenue was a foremost motive for some officials, the reform was explicitly framed around boosting UHC funding and advancing public health by reducing alcohol and tobacco consumption. Earmark- ing of tax revenue for UHC was essential to the political compromise that made this trailblazing tax increase a reality. It ensured that the increase, which may have otherwise been perceived as regressive, was framed as a progressive public health measure in the public imagination, while also appeasing tobacco growers and their political representatives. Earmarking was also important because the earmark en- sured high-level support for the tax by achieving a key political priority (151). Its soft-earmark structure meant it was not a blank cheque to the Ministry of Health, and this addressed concerns within the Ministry of Finance. Similarly, in Australia, earmarking of revenue helped overcome community objections to tobacco taxes and tobacco control more generally that resulted from the tobacco industry’s sponsorship of sports and the arts in the 1980s. Attempts to completely ban tobacco advertising and sponsorship had been unsuccessful due to strong pressure from sports, arts and racing lobbies that claimed that a ban would harm these activities. States, starting with Victoria, responded by earmarking funding for Health Promotion Foundations that took over the tobacco industry’s sponsorship activities and also paid for antismoking campaigns. In 1997, these earmarks ended after a High Court ruling that the Constitution did not allow states to collect excise taxes. However, in recognition of the successful work of the Foundations, the federal 25 Details about earmarked taxes by country are provided at https://www.who.int/tobacco/global_report/ Table-9-4-Use-of-earmarked-tobacco-taxes.xls?ua=1. 238 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N government began funding them directly from the federal budget (152). Although this example may be difficult to replicate precisely, it shows how earmarks with built-in sunset clauses for piloting cost-effective interventions can potentially graduate to funding from the general budget once they have proven their effectiveness. It also more generally shows how earmarks can disarm community objections, reframe tobacco tax increases and unlock the political will needed to advance effective tobacco control measures. Structures for managing earmarked tobacco taxes A 2016 review of nine countries’ tobacco tax revenue earmarking experiences identified three arrangements for governance and allocating revenue (151). Table 4.10 presents some examples of these allocation arrangements. In some countries, earmarked tobacco tax revenues are combined with alcohol tax revenues. Table 4.10 Illustrative arrangements for allocating earmarked tobacco tax revenues POSSIBILITIES FOR ALLOCATING TOBACCO TAX EARMARK REVENUE Forms of budget allocation Revenue goes to the general fund and is later assigned to the official actor(s) specified in the earmark. Revenues do not go through the general budget but are instead paid into a separate account belonging to the official actor(s) specified in the earmark. Earmarked tax revenue is paid directly to the account of the entity managing an autonomous or semi- autonomous fund. Examples In the Philippines, revenue goes to the general fund before being allocated to the Ministry of Health following submission of a budget for its use (140). In Romania, revenue goes directly into a Ministry of Health account that is distinct from the general fund. In Panama, revenue is paid into subaccounts of the three recipient agencies (the Ministry of Health, the National Cancer Institute and the Customs Authority). In Thailand, ThaiHealth directly receives the earmarked revenues in its own account. In Viet Nam, the Viet Nam Tobacco Control Fund receives the revenues directly into a subaccount it manages but that belongs to the Ministry of Health. Source: (151). Where is the money being spent? Earmarked tobacco tax revenues are used for a variety of health purposes, including tobacco control, health promotion and UHC. A wide variety of other programmes have also been funded with earmarks from tobacco taxes, including disaster relief (e.g. hospital medical supplies and equipment to treat COVID-19 in India), youth pro- grams, sports and craft jobs in Yemen, social cohesion in Morocco, health and social programs in areas dependent on tobacco growing in Argentina, health promotion CHAP T ER 4. PO LI T I C AL ECO N OMY 239 and tobacco control in Thailand and alternative livelihood programs for tobacco farmers as well as economic projects in tobacco-growing provinces in the Philippines. Table 4.11 shows the three main categories of health programmes to which ear- marked tobacco tax revenue is allocated, as well as a fourth miscellaneous category, with country-specific examples for each.26 Table 4.11 Programmes to which earmarked tobacco tax revenue is allocated TOBACCO CONTROL NCD PREVENTION AND CONTROL PROGRAMMES (otherwise indicated between brackets) HEALTH COVERAGE EXPANSION (e.g. through health insurance coverage) OTHER, MORE GENERAL OR UNSPECIFIED HEALTH PROGRAMMES Costa Rica, Côte d’Ivoire, the Islamic Republic of Iran, Madagascar, Panama (tobacco cessation and fighting illicit trade), Switzerland, Viet Nam Cook Islands, Costa Rica, Mauritania (anti-cancer research), Palau (NCD prevention only), Panama (National Institute of Oncology), Paraguay Colombia, Congo, Egypt, Palau, Philippines Algeria, Argentina, Bangladesh, Botswana, Cabo Verde, Chad (programmes delivering antiretroviral drugs), Colombia (sports), Comoros (sports, hospital emergencies), Congo (sports), Côte d’Ivoire (AIDS programme), El Salvador, Estonia (sports), Guatemala, Indonesia, the Islamic Republic of Iran (sports), Ireland, Jamaica, Lithuania (sports), Madagascar (sports), Morocco, Nepal, Paraguay (sports), Republic of Korea (health promotion), Romania, Thailand (health promotion), United States, Yemen (sports) Note: Countries appear in more than one column when their earmarked tax revenues are used in more than one specific health programme. Source: (27). The amount of money associated with tobacco tax earmarks Case studies of the experiences of nine countries in tobacco tax earmarking show that earmarked funds are relatively small in comparison with government spend- ing on health (see Table 4.12) and, consequently, even smaller in terms of GDP. Therefore, the argument that tobacco tax earmarks would introduce rigidity into public financial management may not apply. 26 Details about how the tobacco tax revenues are earmarked are given in Annex 4.2. 240 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Table 4.12 Proportion of earmarked tobacco tax funds in government expenditures COUNTRY ESTIMATED ANNUAL TOTAL FUNDS FROM EARMARKED TAX ANNUAL FUNDS FROM TOBACCO TAX EARMARKS as a % of general government expenditure on health in 2013 Botswana 2014–2015: 4 million pula (US$ 0.48 million) NA Egypt 2013–2014: 392 million Egyptian pounds (US$ 52.06 million); earmarked taxes only 1.8% of total taxes on cigarettes 1.086% Iceland 2014: 108.3 million kronor (US$ 0.89 million) 0.083% Panama 2014: US$ 27.8 million 1.322% Philippines 2014: 50.18 billion Philippine pesos (US$ 1.18 billion) NA Poland 2013: 1 million złoty (US$ 0.316 million) from general budget  0.001% Romania 2014: 1.1 million lei (US$ 0.33 million); 14.4% of total health budget 0.004% Thailand 2014: 4064.74 million baht (US$ 125.15 million); 1.78% of Ministry of Health budget and 1.84% of National Health Security Fund 0.932% Viet Nam 2014: 299.171 billion dong (US$ 13.91 million); 0.5% of national health budget 0.335% Source: (151). 4.6.4 CONCLUSIONS Despite the initial principled resistance to earmarking by some ministries of finance, experience has shown that the use of revenue from tobacco taxes and other taxes on the consumption of products that have negative externalities can ensure political as well as public support. Successful earmarking needs a well-developed structure for the use of funds for health purposes. Even intergovernmental organizations that are opposed to earmarking (e.g. the IMF) have acknowledged the justifiability of well-designed tobacco tax earmarks when revenue is directed to specific cost- effective programmes (153–154). The amounts of tobacco tax revenue effectively earmarked for health have been relatively small and could hardly introduce the feared rigidity in government budgets. Moreover, in some countries, those funds have helped to implement much-needed health programmes (e.g. Australia, the Philippines, Thailand). More governments are considering this option as a stable medium-term source of secure funding for programmes such as tobacco control. The payoffs will be seen in the future as fewer people fall ill and less medical care for tobacco-related illnesses is needed. In Australia, CHAP T ER 4. PO LI T I C AL ECO N OMY 241 an earmarked tax was used to fund a needed and underresourced programme that proved to be successful, effective and impactful; the programme is now sustainably funded, embedded in the federal budget. Earmarking is desirable in a particular political economy when it enables the implementation of effective tobacco taxation that will increase price and reduce consumption. It will, however, also be rational as a matter of public financial man- agement, economic efficiency and democratic governance when concerns such as the following are considered. Although not every question needs an affirmative answer, policy-makers who can answer yes to many of the following questions will likely be considering an effective and rational tobacco tax earmark:27 • Does the tobacco tax earmark’s purpose rationally connect with the recipient programme’s purpose? Earmarks that fund tobacco control or other health programmes are more economically rational under the benefit principle than those that fund unrelated programmes such as childhood education, even when the popularity of the unrelated programmes may make a tax increase politically palatable. • Does the tobacco tax earmark’s amount rationally connect with the needs of the recipient programme? Earmarked funds that cannot be absorbed by the recipient programme are, in effect, money taken away from other needs. • When a tobacco tax earmark funds health programmes, is this clearly com- municated to the public to ensure that the framing of the tobacco tax increase as a health measure reinforces the demand-reduction effect? • Is the programme being funded by the tobacco tax earmark a politically neglected but highly cost-effective or crucially needed programme that, once established as a proof of concept, has a chance of being funded out of the general budget? • Does the tobacco tax earmark’s purpose rationally connect with the effects of the tax itself? Earmarks that fund programmes that disproportionately benefit lower socioeconomic groups or that fund alternative livelihoods for former tobacco workers and farmers will have equity-enhancing effects that will reinforce the already progressive nature of tobacco taxes. • Is the scope of the earmark’s purpose narrow enough that it can be funded mostly from the tobacco tax earmark, to ensure that the revenue is additive and does not merely substitute for spending that would otherwise come from the general fund? • Does the design of the tobacco tax earmark provide for flexibilities that ensure that windfall revenue collection is not squandered on a purpose already saturated with overfunding? 27 Adapted from and informed by References 138–139, 143, 150. 242 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Does the design of the tobacco tax earmark include a sunset clause that triggers its automatic end or review? The presence of such a clause will guard against inefficiency in allocation arising and being left uncorrected due to political inertia. CHAP T ER 4. PO LI T I C AL ECO N OMY 243 REFERENCES 1. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tob Control. 2013; 22(2): 144–153 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 31 January 2021). 2. LaFaive MD, Drenkard S, Nesbit T. Strong link between cigarette tax and illegal smuggling rates [website]. Midland: Mackinak Center for Public Policy; 2018 (https://www.mackinac.org/strong- link-between-cigarette-tax-and-illegal-smuggling-rates#:~:text=An%20increase%20of%2011%20 percentage,packs%20to%2030.6%20million%20packs.&text=Tax%20evasion%20and%20avoidance%20 is,high%20excise%20taxes%20on%20cigarettes., accessed 31 January 2021). 3. Ross H, Blecher E. Illicit trade in tobacco products need not hinder tobacco tax policy reforms and increases. Chicago: University of Illinois at Chicago; 2019 (Tobacconomics White Paper; https:// tobacconomics.org/research/illicit-trade-in-tobacco-products-need-not-hinder-tobacco-tax- policy-reforms-and-increases-white-paper/#:~:text=additional%20related%20research-,Illicit%20 Trade%20in%20Tobacco%20Products%20Need%20Not%20Hinder%20Tobacco,Reforms%20 and%20Increases%20( White%20Paper)&text=The%20paper%20cites%20research%20 that,consumption%2C%20and%20improves%20overall%20health, accessed 31 January 2021). 4. Confronting illicit tobacco trade: a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/curated/en/677451548260528135/pdf/133959- REPL-PUBLIC-6-2-2019-19-59-24-WBGTobaccoIllicitTradeFINALvweb.pdf, accessed 12 October 2020). 5. The economics of tobacco and tobacco control. Bethesda: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute; and Geneva: World Health Organization; 2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf, accessed 28 September 2020). 6. Joossens L, Chaloupka FJ., Merriman D, Yürekli A. Issues in smuggling of tobacco products. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac00Box 370064B00PUBLIC0.pdf, accessed 18 February 2021). 7. Merriman D, Yürekli A., Chaloupka, FJ. How big is the worldwide cigarette smuggling problem. In: Jha P, Chaloupka F, editors. Tobacco control in developing countries. Oxford: Oxford University Press; 2000 (http://documents1.worldbank.org/curated/en/602821468330954036/pdf/709670WP0tobac 00Box370064B00PUBLIC0.pdf, accessed 18 February 2021). 8. WHO Framework Convention on Tobacco Control. Geneva: World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf;jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B?sequence=1, accessed 5 October 2020). 9. Merriman D. Understand, measure, and combat tobacco smuggling. Washington (DC): World Bank Group; 2003 (Economics of tobacco toolkit 7; https://openknowledge.worldbank.org/bitstream/ handle/10986/16267/805800NWP070Sm0Box0379807B00PUBLIC0.pdf?sequence=1&isAllowed=y, accessed 31 January 2021). 10. Progress report on the implementation of the Commission Communication. Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products – a comprehensive EU strategy. Brussels: European Commission; 2017 (COM(2017) 235 final; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/?uri=CELEX:52017DC0235&from=en, accessed 31 January 2021). 11. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 12. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Reuter P, Majmundar M. (editors). Washington (DC): National Academies Press; 2015. 13. von Lampe K. The illegal cigarette trade. In: Natarajan M, editor. International crime and justice. New York: Cambridge University Press. 2010:148–54. 14. Joossens L, Raw M, From cigarette smuggling to illicit tobacco trade. Tob Control. 2012;21:230–34 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.full.pdf, accessed 31 January 2021). 15. Gomis B, Lee K, Carrillo Botero N, Shepherd P, Iglesias RM. “We think globally”: the rise of Paraguay’s 24 4 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Tabacalera del Este as a threat to global tobacco control. Global Health. 2018;14(1):110 (https:// globalizationandhealth.biomedcentral.com/articles/10.1186/s12992-018-0412-3, accessed 31 January 2021). 16. Communication from the commission to the Council and the European Parliament: Stepping up the fight against cigarette smuggling and other forms of illicit trade in tobacco products - A comprehensive EU Strategy. Brussels: European Commission; 2013 (https://eur-lex.europa.eu/legal-content/EN/TXT/ PDF/?uri=CELEX:52013DC0324&from=EN, accessed 31 January 2021). 17. Ross H, Vellios N, Smith KC, Ferguson J, Cohen JE. A closer look at ‘cheap white’ cigarettes. Tob Control. 2016;25(5):527–31 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/527.full.pdf, accessed 31 January 2021). 18. Nagelhout GE, van den Putte B, Allwright S, Mons U, McNeill A, Guignard R, et al. Socioeconomic and country variations in cross-border cigarette purchasing as tobacco tax avoidance strategy. Findings from the ITC Europe Surveys. Tob Control. 2014;23:i30–i38 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/23/suppl_1/i30.full.pdf, accessed 1 February 2021). 19. Gilmore AB, Rowell A, Gallus S, Lugo A, Joossens L, Sims M. Towards a greater understanding of the illicit tobacco trade in Europe: a review of the PMI funded ‘Project Star’ report. Tob Control. 2014;23:e51–e61 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/23/e1/e51.full.pdf, accessed 1 February 2021). 20. Iglesias RM. Increasing excise taxes in the presence of an illegal cigarette market: the 2011 Brazil tobacco tax reform. Revista Panam Salud Publica. 2016;40(4):243–9 (https://www.scielosp.org/pdf/ rpsp/2016.v40n4/243-249, accessed 1 February 2021). 21. Operação Huno combate mercado clandestino de cigarros em sete Estados [Operação Huno fights clandestine cigar market in seven states]. Brasilia: Federal Police of Brazil; 2015 (in Portuguese) (http:// www.pf.gov.br/agencia/noticias/2015/11/operacao-huno-combate-mercado-clandestino-de-cigarros- em-seis-estados, accessed 12 October 2020). 22. Gilmore AB, Gallagher AWA, Rowell A. Tobacco industry’s elaborate attempts to control a global track and trace system and fundamentally undermine the Illicit Trade Protocol. Tob Control. 2019;28:127–40 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/28/2/127.full.pdf, accessed 1 February 2021). 23. Technical assessment of the experience made with the Anti-Contraband and Anti-Counterfeit Agreement and general release of 9 July 2014 among Philip Morris International and affiliates, the Union and its Member States. Brussels: European Commission; 2016 (SWD (2016) 44 final; https://ec.europa.eu/anti- fraud/sites/antifraud/files/technical_assessment_pmi_24022016_en.pdf, accessed 1 February 2021). 24. Gallagher AWA, Evans-Reeves KA, Hatchard JL, Gilmore AB. Tobacco industry data on illicit tobacco trade: a systematic review of existing assessments. Tob Control. 2019;28(3):334–45 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/28/3/334.full.pdf, accessed 1 February 2021). 25. Joossens L, Merriman D, Ross H, Raw M. How eliminating the global illicit cigarette trade would increase tax revenue and save lives. Paris: International Union Against Tuberculosis and Lung Disease; 2009 (https://theunion.org/sites/default/files/2020-08/How%20Eliminating%20the%20Global%20Illicit%20 Cigarette%20Trade%20Would%20Increase%20Tax%20Revenue%20and%20Save%20Lives%20-%20 EN.pdf, accessed 1 February 2021). 26. Protocol to Eliminate Illicit Trade in Tobacco Products. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf?sequence=1, accessed 5 October 2020). 27. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 28. Prieger JE, Kulic J. Cigarette taxes and illicit trade in Europe. Econ Inquiry. 2018;56(3):1706–23 (https:// papers.ssrn.com/sol3/papers.cfm?abstract_id=2718519, accessed 18 February 2021). 29. Stoklosa M. Prices and cross-border cigarette purchases in the EU: evidence from demand modelling. Tob Control. 2020;29:55–60. 30. Passport: illicit trade in tobacco products. In: Euromonitor International [website]. London: Euromonitor International; 2018 (restricted access). 31. The Global Illicit Trade Environment Index. London: The Economist Intelligence Unit; 2018 (http:// illicittradeindex.eiu.com/documents/EIU%20Global%20Illicit%20Trade%20Environment%20Index%20 CHAP T ER 4. PO LI T I C AL ECO N OMY 245 2018%20-%20Overall%20Results%20White%20Paper%20June%206%20FINAL.pdf, accessed 1 February 2021). 32. Chaloupka FJ, Yürekli A, Fong GT. Tobacco taxes as a tobacco control strategy. Tob Control. 2012;21:172– 80 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/172.full.pdf, accessed 1 February 2021). 33. Chernick H, Merriman D. Using littered pack data to estimate cigarette tax avoidance in NYC. Natl Tax J. 2013;66(2):635–68 (https://www.ntanet.org/NTJ/66/3/ntj-v66n03p635-668-littered-pack-data- cigarette-tax.pdf, accessed 1 February 2021). 34. Liber AC, Ross H, Omar M, Chaloupka FJ. The impact of the Malaysian minimum cigarette price law: findings from the ITC Malaysia Survey. Tob Control. 2015;24:iii83–iii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/iii83.full.pdf, accessed 1 February 2021). 35. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 36. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 37. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 38. Paraje G, Araya D, Drope J. Illicit cigarette trade in metropolitan Santiago de Chile. Tob Control 2020;29:68–73. 39. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 40. Saenz de Miera Juarez BS, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020(0):1-7 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 1 February 2021). 41. Lee K, Collin J. ‘Key to the future’: British American Tobacco and cigarette smuggling in China. PLoS Medicine. 2006;3(7):e228 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1502159/pdf/pmed.0030228. pdf, accessed 2 February 2021). 42. Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38– i43 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i38.full.pdf, accessed 2 February 2021). 43. Aziani A, Kulick J, Norman N, Prieger JE. Empty discarded pack data and the prevalence of illicit trade in cigarettes; 2017. (https://ssrn.com/abstract=2906015, accessed 12 October 2020). 44. Philip Morris International. The Perspective of PM International on Smoking and Health Issues. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1985. Bates no. 2023268328-2023268349 (https://www.industrydocuments.ucsf.edu/tobacco/docs/#id=pxpb0040, accessed 5 February 2021). 45. Mandal S, Gilmore A, Collin J, Weishaar H, Smith K, McKee M. Block, amend, delay: tobacco industry efforts to influence the European Union’s Tobacco Products Directive (2001/37/EC). Smokefree Partnership/Cancer Research UK; 2009 (http://www.smokefreepartnership.eu/images/Reports-and- Position-Papers/EU_TI_TPD_report_May_2012.pdf, accessed 15 October 2020). 46. Bishop DJ. Key area paper: indirect taxation – British American Tobacco. In: Legacy Tobacco Documents Library [website]. San Francisco: University of California; 1994. Bates no. 503911771-503911780 (http:// bat.library.ucsf.edu/data/h/f/k/hfk11a99/hfk11a99.pdf, accessed 15 October 2020). 47. Zhou S, Liberman JD, Ricafort E. The impact of the WHO Framework Convention on Tobacco Control in defending legal challenges to tobacco control measures. Tob Control. 2009:28:s113-s118 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/28/Suppl_2/s113.full.pdf, accessed 2 February 2021). 48. Gilmore AB, Fooks G, Drope J, Bialous SA, Jackson RR. Exposing and addressing tobacco industry 246 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N conduct in low-income and middle-income countries. Lancet. 2015;385:1029–43. 49. Ulucanlar S, Fooks G, Gilmore A. The policy dystopia model: an interpretive analysis of tobacco industry political activity. PLoS Med. 2016:13(9):e1002125 (https://journals.plos.org/plosmedicine/ article?id=10.1371/journal.pmed.1002125, accessed 2 February 2021). 50. Apollonio D, Glantz S. Tobacco manufacturer lobbying to undercut minimum price laws: an analysis of internal industry documents. Tob Control. 2020;29:e10-e17 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC7374022/pdf/nihms-1590806.pdf, accessed 2 February 2021). 51. Crosbie E, Glantz S. Tobacco industry argues domestic trademark laws and international treaties preclude cigarette health warning labels, despite consistent legal advice that the argument is invalid. Tob Control. 2012;23:e7 (https://www.researchgate.net/publication/233769722_Tobacco_industry_ argues_domestic_trademark_laws_and_international_treaties_preclude_cigarette_health_warning_ labels_despite_consistent_legal_advice_that_the_argument_is_invalid, accessed 2 February 2021). 52. Puig S. Tobacco litigation in international courts. Harv Int Law J. 2016;57(2):1–68. 53. Crooked nine: nine ways the tobacco industry undermines health policy. New York: Vital Strategies; 2019 (https://exposetobacco.org/wp-content/uploads/2019/09/Crooked-9-STOP.pdf, accessed 2 February 2021). 54. Balwicki L, Stokłosa M, Balwicka-Szczyrba M, Tomczak W. Tobacco industry interference with tobacco control policies in Poland: legal aspects and industry practices. Tob Control. 2016;25(5):521–26 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/521.full.pdf, accessed 2 February 2021). 55. Sud SR, Brenner JE, Shaffer ER. Trading away health: the influence of trade policy on youth tobacco control. J Pediatrics. 2015;166(5):1303–07 (https://www.jpeds.com/action/showPdf?pii =S0022-3476%2815%2900017-7, accessed 2 February 2021). 56. Lee S, Ling PM, Glantz SA. The vector of the tobacco epidemic: tobacco industry practices in low and middle-income countries. Cancer Cause Control. 2012;23:117–29 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC3332051/pdf/nihms-367996.pdf, accessed 2 February 2021). 57. Zhou S, Liberman J. The global tobacco epidemic and the WHO Framework Convention on Tobacco Control—the contributions of the WHO’s first convention to global health law and governance. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018 (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3339718, accessed 18 February 2021). 58. Legislative Consultation with Constitutional Division of the Supreme Court, No. 2012-003918. Constitutional Division of the Supreme Court of Costa Rica; 20 March 2012 (https://www. tobaccocontrollaws.org/files/live/litigation/830/CR_Legislative%20Consultation%20with%20.pdf, accessed 2 February 2021). 59. British American Tobacco Kenya Ltd v Ministry of Health, Petition No. 143 of 2015. High Court of Kenya; 2016 (https://www.tobaccocontrollaws.org/files/live/litigation/2643/KE_British%20American%20 Tobacco%20Kenya.pdf, accessed 2 February 2021). 60. British American Tobacco v Ministry of Health, et al, Civil Appeal No. 112 of 2016. Kenyan Court of Appeal; 2017 (https://www.tobaccocontrollaws.org/files/live/litigation/2521/KE_British%20 American%20Tobacco%20Ltd%20v.pdf, accessed 2 February 2021). 61. Thailand – customs and fiscal measures on cigarettes from the Philippines: Report of the Panel. Geneva: World Trade Organization; 2010 (WT/DS371/R; https://docs.wto.org/dol2fe/Pages/SS/directdoc. aspx?filename=Q:/WT/DS/371R.pdf&Open=True, accessed 2 February 2021). 62. Tribunal Constitucional de Chile [Constitutional Court of Chile] (1995 Rol No. 219-95; https://www. tribunalconstitucional.cl/descargar_sentencia2.php?id=3074, accessed 4 April 2020.) 63. McGrady B. International investment law. In: Voon T, Mitchell AD, Liberman J, editors. Regulating tobacco, alcohol and unhealthy foods: the legal issues. New York: Routledge; 2014. 64. Marvin Roy Feldman Karpa v United Mexican States, Case No. ARB(AF)/99/1 International Centre for Settlement of Investment Disputes; 2002 (https://www.italaw.com/sites/default/files/case-documents/ ita0319.pdf, accessed 2 February 2021). 65. Feldman Karpa v Mexico (2002 ICSID Arbitral Tribunal) (https://www.italaw.com/sites/default/files/ case-documents/ita0319.pdf, accessed 18 February 2021). 66. Undang-Undang Nomor 11 Tahun 1995 tentangCukai (Lembaran Negara Republik Indonesia Tahun1995 CHAP T ER 4. PO LI T I C AL ECO N OMY 247 Nomor 76) (Excise Law,in Indonesian). Article 5. 67. Excise Regulation (2011) Reference No. PMK Number 167/PMK.011/2011 (in Indonesian) (https:// peraturan.bpk.go.id/Home/Details/145897/pmk-no-167pmk0112011, accessed 3 February 2021). 68. Mahkamah Agung. Indonesia Supreme Court, Case No. 73/PDT/2018/PT SBY (https://putusan3. mahkamahagung.go.id/direktori/putusan/74efff9cfe18d60990fcde19583ecd5f.html, accessed 6 February 2021). 69. Maya D. Analisis kebijakan kenaikam tarif cukai dan penyesuaian batasan harga jual eceran hasil tembakau [Policy analysis on excise tax increases and price limit adjustments retail tobacco products]; 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-09/S45937-Deacy%20Maya accessed 3 March 2020). 70. Indriawati EN, Susanti L, Qurnia MA. Kebijakan tarif cukai hasil tembakau terhadap perekonomian [Tobacco excise tariff policy on the economy]; 2014 (in Indonesian). (https://www.academia. edu/6611977/MAKALAH_CHT_kel_6, accessed 15 October 2020). 71. Arroyani DW. Evaluasi Kebijakan Penetapan Tarif Cukai Hasil Tembakau Berdasarkan PMK [Evaluation of Tobacco Excise Tariff Determination Policy Based on PMK].167/PMK.011/2011. 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-08/S44664-Dwira%20Wanti%20Arroyan, accessed 15 October 2020). 72. Ngo Ngo Ha and Another v. State of New South Wales and Others (Matter No S 45 of 1996); Walter Hammond & Associates Pty Ltd v. State of New South Wales and Others (Matter No S 165 of 1996). Constitutional law; (1997) 146 ALR 355: 5 August 1997 (https://staging.hcourt.gov.au/assets/ publications/judgments/1997/029--NGO_NGO_HA_AND_ANOR_v_STATE_OF_NEW_SOUTH_WALES_&_ ORS_(Matter_No_S_45_of_1996);_WALTER_HAMMOND--(1997)_146_ALR_355.html, accessed 2 February 2021). 73. McGrady B. Trade and public health: the WTO, tobacco, alcohol, and diet. New York: Cambridge University Press; 2011. 74. Giorgetti C. Health and international investment law. In: Burci GL, Toebes B, editors. Research handbook on global health law. Cheltenham: Edward Elgar Publishing; 2018. 75. British American Tobacco v Attorney General and Center for Human Rights and Development. Constitutional Court of Uganda Petition No.46 of 2016; 2017 (https://media.ulii.org/files/judgments/ ugcc/2019/4/2019-ugcc-4_0.pdf, accessed 18 February 2021). 76. Lim CL, Ho J, Paparinskis M. International investment law and arbitration: commentary, awards and other materials. Cambridge: Cambridge University Press; 2018. 77. Dominican Republic – measures affecting the importation and internal sale of cigarettes : Report of the Panel. Geneva: World Trade Organization; 2004 (WT/DS302/R; https://docs.wto.org/dol2fe/ Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302R.pdf&Open=True, accessed 21 February 2021) 78. Dominican Republic — measures affecting the importation and internal sale of cigarettes: Report of the Appellate Body. Geneva: World Trade Organization; 2005 (WT/DS302/AB/R; https://docs.wto. org/dol2fe/Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302ABR.pdf&Open=True, accessed 21 February 2021). 79. Gilmore A. Tobacco and transition: understanding the impact of transition on tobacco use and control in the former Soviet Union [Thesis]. London: London School of Hygiene and Tropical Medicine, June 2005 (https://escholarship.org/uc/item/3rw2c04c, accessed 2 February 2021). 80. MacKenzie R, Collin J, Sopharo C, Sopheap Y. ‘Almost a role model of what we would like to do everywhere’: British American Tobacco in Cambodia. Tob Control. 2004;13(suppl II): ii112–ii117 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/13/suppl_2/ii112.full.pdf, accessed 2 February 2021). 81. Smith KE, Savell E, Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tobacco Control 2013;22(2):144-53 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 21 February 2021). 82. Price and tax. In: TobaccoTactics [website]. Bath: University of Bath; 2020 (https://tobaccotactics.org/ topics/price-and-tax/, accessed 15 December 2020). 83. Vulovic V, Chaloupka FJ. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation—commentary. Tob Control. 24 June 2020. 248 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 84. Verguet S, Kearns PKA, Rees VW. Questioning the regressivity of tobacco taxes: a distributional accounting impact model of increased tobacco taxation. Tob Control. 23 June 2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/08/11/tobaccocontrol-2019-055315. full.pdf, accessed 2 February 2021).. 85. Palipudi KM, Gupta PC, Sinha DN, Andes LJ, Asma S, McAfee T. Social determinants of health and tobacco use in thirteen low and middle income countries: evidence from Global Adult Tobacco Survey. PLoS ONE. 2012;7(3):e33466 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3306395/pdf/ pone.0033466.pdf, accessed 2 February 2021). 86. Systematic review of the link between tobacco and poverty. Geneva: World Health Organization; 2014 (http://www.who.int/tobacco/publications/economics/9789241507820/en, accessed 21 October 2020). 87. Goodchild M, Sinha P, Gill Munish V, Tullu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018: a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73–81 (https://apps.who.int/iris/bitstream/handle/10665/331836/whoseajph2020v9i1p73. pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 88. Faruque GM, Ahmed M, Huq I, Parven R, Wadood SN, Chowdhury SR, et al. The economic cost of tobacco use in Bangladesh: a health cost approach. Dhaka: Bangladesh Cancer Society; 2020 (https://www. researchgate.net/publication/342445467_The_Economic_Cost_of_Tobacco_Use_in_Bangladesh_A_ Health_Cost_Approach, accessed 2 February 2021). 89. John RM, Sung H, Max WB, Ross H. Counting 15 million more poor in India, thanks to tobacco. Tob Control. 2011;20:349–52 (https://www.researchgate.net/publication/49809290_Counting_15_million_ more_poor_in_India_thanks_to_tobacco, accessed 2 February 2021). 90. Liu Y, Rao K, Hu TW, Sun Q, Mao Z. Cigarette smoking and poverty in China. Soc Sci Med. 2006;Dec;63(11):2784–90. 91. Fuchs A, González Icaza F, Paz D. Distributional effects of tobacco taxation – a comparative analysis. Washington (DC): World Bank Group; 2019 (Policy research working paper 8805; https://openknowledge. worldbank.org/bitstream/handle/10986/31534/WPS8805.pdf?sequence=4&isAllowed=y, accessed 2 February 2021). 92. Smith CE, Hill SE, Amos A. Impact of population tobacco control interventions on socioeconomic inequalities in smoking: a systematic review and appraisal of future research directions. Tob Control. 29 September 2020 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/09/28/ tobaccocontrol-2020-055874.full.pdf, accessed 2 February 2021). 93. Furman J. Six lessons from the U.S. experience with tobacco taxes – World Bank conference “Winning the tax wars: global solutions for developing countries”. Washington (DC): Obama White House Archives; 24 May 2016 (https://obamawhitehouse.archives.gov/sites/default/files/page/files/20160524_cea_ tobacco_tax_speech.pdf, accessed 21 October 2020). 94. Önder Z, Yürekli A. Who pays the most cigarette tax in Turkey. Tob Control. 2016;25(1):39–45. 95. Verguet S, Gauvreau CL, Mishra S, MacLennan M, Murphy SM, Brouwer ED, et al. The consequences of tobacco tax on household health and finances in rich and poor smokers in China: an extended cost-effectiveness analysis. Lancet Glob Health. 2015;3:e206–16 (https://www.thelancet.com/action/ showPdf?pii=S2214-109X%2815%2970095-1, accessed 2 February 2021). 96. Fuchs Tarlovsky A, González Icaza F. The welfare and distributional effects of increasing taxes on tobacco in Vietnam. Washington (DC): World Bank Group; 2019 (http://documents1.worldbank.org/ curated/en/758391561477648302/pdf/The-Welfare-and-Distributional-Effects-of-Increasing-Taxes- on-Tobacco-in-Vietnam.pdf, accessed 2 February 2021). 97. Expanding the global tax base: taxing to promote public goods: tobacco taxes. Washington (DC): World Bank Group; 2016 (http://pubdocs.worldbank.org/en/799981480947591462/Taxing-to-Promote- Public-Goods-tobacco-taxation-chapter-conference-25-nov-2016-final-version-pmarquez-002.pdf, accessed 2 February 2021). 98. Blecher E. Increasing tobacco taxes significantly will increase revenues. Chicago: University of Illinois at Chicago; 2018. (Tobacconomics policy brief; https://tobacconomics.org/files/research/470/Laffer- Curve-Policy-Brief_v2.2.pdf, accessed 2 February 2021). 99. Savedoff W, Lopert R. Sugar, rum, and tobacco: domestic resource mobilization for low-income countries through excise taxes. Washington (DC): Centre for Global Development; 2020 (CGD Policy Paper CHAP T ER 4. PO LI T I C AL ECO N OMY 249 164, https://www.cgdev.org/sites/default/files/PP164-Savedoff-Health-Taxes-DRM.pdf, accessed 2 February 2021). 100. Lavado R, Postolovska I, Sotomayor RE. Beyond the perceived regressivity of higher tobacco taxes, turning short-term losers into long-term winners? In: Tobacco reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 101. Petit P, Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund; 2016 (International Monetary Fund How To Notes, No.3/2016; https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 102. Tonga NCD related taxation policy assessment preliminary findings. Presentation made at the innovations and actions against NCDs conference ADB headquarters, Manila, 17 July 2018. 103. Moreno-Dodson B. Raising tobacco taxes: why, what, and how: where public health and economics converge. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e- library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 104. Fuchs A, Del Carmen G, Mukong AK. Long-run impacts of increasing tobacco taxes: evidence from South Africa. Policy Research Working Paper 8369. Washington [DC]: World Bank; 2018 (https://openknowledge. worldbank.org/bitstream/handle/10986/29497/WPS8369.pdf?sequence=4&isAllowed=y, accessed 18 February 2021) . 105. van Walbeek C. Case studies in illicit tobacco trade: South Africa. Tobacconomics fact sheet. Chicago: University of Chicago, Tobacconomics, Health Policy Center; 2020 (https://tobacconomics.org/files/ research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 106. van Loggerenberg, J. Tobacco Wars. Cape Town: NB Publishers; 2019. 107. Marquez P. Global tobacco control: a development priority. In: Tobacco tax reform at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads. pdf, accessed 2 February 2021). 108. Kaiser K, Bredenkamp C, Iglesias R. Sin tax reform in the Philippines: transforming public finance, health, and governance for more inclusive development. Washington [DC]: World Bank; 2017 (https:// tobacconomics.org/files/research/607/UIC_South-Africa-Illicit-Trade-Case-Study-v1.3.pdf, accessed 18 February 2021). 109. Scollo M, Bayly M. Revenue from tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia. org.au/chapter-13-taxation/13-6-revenue-from-tobacco-taxes-in-australia, accessed 21 October 2020). 110. Scollo M, Bayly M. Tobacco taxes in Australia. In: Scollo M, Winstanley M, editors. Tobacco in Australia: facts and issues. Melbourne: Cancer Council Victoria; 2016 (https://www.tobaccoinaustralia.org.au/ chapter-13-taxation/13-2-tobacco-taxes-in-australia, accessed 21 October 2020). 111. Breton E, Richard L, Gagnon F, Jacques M, Bergeron P. Fighting a tobacco-tax rollback: a political analysis of the 1994 cigarette contraband crisis in Canada. Journal of Pub Health Policy. 2006;27:77–99. 112. Fanta E, Garcia M. Preventing and controlling illicit tobacco flows. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB- Report-CrossRoads.pdf, accessed 2 February 2021). 113. Goodchild M, Perucic AM, Zheng R, Blecher E, Paul J. The health impact of raising tobacco taxes in developing countries. In: Tobacco tax reform: at the crossroads of health and development: a multisectoral perspective. Washington (DC): World Bank Group; 2017 (https://untobaccocontrol.org/ taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021). 114. Ross H. Tracking and tracing tobacco products in Kenya. Prev Med. 2017;105:S15–S18 (https://reader. elsevier.com/reader/sd/pii/S0091743517301457?token=424D9EC1D42A9C69BDB55B90090C3F7853AA 20A4970D0E379C91EE64C73A273A0FA949557F4F120A0B636701581EC055, accessed 2 February 2021). 115. Mamudu H, Hammond R, Glantz S. Tobacco industry attempts to counter the world bank report curbing the epidemic and obstruct the WHO framework convention on tobacco control. Soc Sci & 250 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Med. 2008;67(11):1690–9 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2662513/pdf/nihms81164. pdf, accessed 2 February 2021). 116. Gilmore A. Understanding the vector in order to plan effective tobacco control policies: an analysis of contemporary tobacco industry materials. Tob Control. 2012;21:119–26 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/21/2/119.full.pdf, accessed 2 February 2021). 117. The economics of tobacco taxation and employment in Indonesia: policy implications technical brief. Washington (DC): World Bank Group; 2018 (https://openknowledge.worldbank.org/bitstream/ handle/10986/29814/126158-WP-P154568-PUBLIC-add-series-WBGPNIndoEmploymentUPDATEFin alWeb.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 118. Sahadewo GA, Drope J, Li Q, Nargis N, Witoelar F. Tobacco or not tobacco: predicting farming households’ income in Indonesia. Tob Control. 2020. 119. The economics of tobacco farming in Indonesia: health, population, and nutrition global practice. Washington (DC): World Bank Group, 2017 (http://documents1.worldbank.org/curated/ en/161981507529328872/pdf/120307-REVISED-PUBLIC-WBGIndoEconomicsTobaccoFarming.pdf, accessed 2 February 2021). 120. The revenue and employment outcome of biri taxation in Bangladesh. Dhaka: National Board of Revenue, Government of the People’s Republic of Bangladesh; 2019 (https://www.who.int/docs/ default-source/searo/bangladesh/pdf-reports/cat-2/biri-study-report-03-12-2019.pdf?sfvrsn=b8fea69_2, accessed 2 February 2021). 121. Magati P, Lencucha R, Li Q, Drope J, Labonte R, Appau A, et al. Costs, contracts and the narrative of prosperity: an economic analysis of smallholder tobacco farming livelihoods in Kenya. Tob Control. 2019;28(3):268–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6512316/pdf/nihms-1012064. pdf, accessed 2 February 2021). 122. Ghaus K, Iqbal MA, Sabir M, Saleem W, Ali A, Aamir N. Macroeconomic impacts of tobacco use in Pakistan. Karachi: The Social Policy and Development Centre; 2018 (https://tobacconomics.org/files/ research/483/Macroeconomic-impact-of-Tobacco-in-Pakistan.pdf, accessed 2 February 2021). 123. Ross H. The effects of tobacco taxes on employment and labor productivity. In: Tobacco tax reform: at the crossroads of health and development. Washington (DC): World Bank Group; 2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf, accessed 2 February 2021) 124. Bigawanto M. Tobacco industry interference undermined tobacco tax policy in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance (SEATCA); 2019 (https://seatca.org/dmdocuments/ Indonesia%20TII%20in%20Tax.pdf, accessed 2 February 2021). 125. Ahsan A, Wyono N. Analysis of the impact of higher cigarettes prices on employment in Indonesia. Bangkok: Southeast Asia Tobacco Control Alliance; 2007 (https://seatca.org/dmdocuments/An%20 Analysis%20of%20the%20Impact%20of%20Higher%20Cigarette%20Prices%20on%20Employment%20 in%20Indonesia.pdf, accessed 2 February 2021). 126. Higashi H, Khuong TA, Ngo AD, Hill PS. Evidence and decision making: tobacco control policy and legislation in Vietnam. Int J of Health Plann. 2013;2(1):e72–e94. 127. Buck D, Raw M, Godfrey CA, Sutton M. Tobacco and jobs: the impact of reducing consumption on employment in the UK. York: University of York; 1995 (https://www.researchgate.net/ publication/4814523_Tobacco_and_jobs_the_impact_of_reducing_consumption_on_employment_ in_the_UK, accessed 2 February 2021). 128. Petkova R, Ileva R and Rusenov A. Tobacco control policies and tobacco industry employment in Bulgaria. In: Economic, social and health issues in tobacco control. Report of a WHO international meeting, Kobe, Japan, 3-4 December 2001. Geneva: World Health Organization; 2003. 129. Jha V, Narayanan BG, Wadhwa D, Tesche J. Economic and environmental effects of reduction in smoking prevalence in Tanzania. Tob Control; 2020;29:24–8 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/29/1/24.full.pdf, accessed 2 February 2021). 130. Lencucha R, Makoka D, Drope J, Appau A, Labonté R, Li Q, et al. Costs, revenues and profits: an economic analysis of smallholder tobacco farmer livelihoods in Malawi. Tob Control. 2017;26(6)634–40 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC5957074/pdf/nihms965842.pdf, accessed 2 February 2021). 131. Goma FM, Labonté R, Drope J, Li Q, Zulu R, Kangwa E. The economics of tobacco farming in Zambia: CHAP T ER 4. PO LI T I C AL ECO N OMY 251 tobacco farmers survey report. Lusaka: University of Zambia and American Cancer Society; 2019 (https://www.cancer.org/content/dam/cancer-org/research/economic-and-healthy-policy/zambia- report-7-16-19.pdf, accessed 28 October 2020). 132. Chavez JJ, Drope J, Li Q, Aloria MJ. The economics of tobacco farming in the Philippines. Quezon City: Action for Economic Reforms and Atlanta: American Cancer Society, 2016 (https://www.cancer. org/content/dam/cancer-org/research/economic-and-healthy-policy/the-economics-of-tobacco- farming-in-the-philippines.pdf, accessed 22 October 2020). 133. Augstin CP, Cardenas RP, Cortez JB, Gimongala GC, Ramos AR. The effects of the sin tax reform law of 2012 to tobacco farmers of Amulung, Cagayan. IJARMSS. 2017;6(12):57–75. 134. Mercado A. Tobacco tax: good for the tobacco farmers, too. Business World; 2019 [website] (https:// www.bworldonline.com/tobacco-tax-good-for-the-tobacco-farmers-too/, accessed 25 October 2020). 135. Yürekli A, Önder Z, Elibol M, Erk N, Cabuk A, Fisunoglu M, et al. The economics of tobacco and tobacco taxation in Turkey. Paris: International Union Against Tuberculosis and Lung Disease; 2010 (https:// www.who.int/tobacco/en_tfi_turkey_report_feb2011.pdf, accessed 2 February 2021). 136. Latif E, Warner V. Legislation and smoking prevention. In: Loddenkemper R, Kreuter M, editors. The Tobacco Epidemic: progress in Respiratory Research. 2015;42:136–148. 137. Tandon A, Cashin C. Assessing public expenditure on health from a fiscal space perspective. Washington (DC): World Bank Group; 2010 (http://documents1.worldbank.org/curated/en/333671468330890417/ pdf/560530WP0Box341penditureFiscalSpace.pdf, accessed 2 February 2021). 138. Cashin C, Sparkes S, Bloom D. Earmarking for health: from theory to practice. Geneva: World Health Organization; 2018 (https://apps.who.int/iris/bitstream/handle/10665/255004/9789241512206-eng. pdf?sequence=1, accessed 2 February 2021). 139. Wilkinson M. Paying for public spending: is there a role for earmarked taxes? Fiscal Studies. 1994;15(4):119–35 (https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1475-5890.1994.tb00213.x, accessed 2 February 2021). 140. ITAF: impôts et taxes affectées à la protection sociale. Les impôts et taxes affectés (ITAF) [taxes and duties allocated to social protection]. Vie publique: au coeur du débat public [website]. Paris: Directorate of Legal and Administrative Information; 2018 (in French) (https://www.vie-publique.fr/fiches/22010- impots-et-taxes-affectees-itaf-protection-sociale, accessed 31 October 2020). 141. Seventeenth congress of the Philippines, Metro Manila, RA 11346, 23 July 2018 (https://www. officialgazette.gov.ph/downloads/2019/07jul/20190725-RA-11346-RRD.pdf, accessed 31 October 2020). 142. McCleary W. The earmarking of government revenue: a review of some World Bank experience. World Bank Res Obs. 1991;6(1):81–104 (http://documents1.worldbank.org/curated/en/109871468764672767/ pdf/multi-page.pdf, accessed 2 February 2021). 143. Strategic investment of tobacco tax revenue. Washington (DC): Campaign for Tobacco-Free Kids; 2020 (https://www.tobaccofreekids.org/assets/global/pdfs/en/strategic_investment_tobacco_tax_revenue. pdf, accessed 31 October 2020). 144. Bialous SA, Glantz SA. Arizona’s tobacco control initiative illustrates the need for continuing oversight by tobacco control advocates. Tob Control. 1999;8:141–51 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/8/2/141.full.pdf, accessed 2 February 2021). 145. Goldman LK, Glantz SA. The passage and initial implementation of Oregon’s measure 44. Tob Control. 1999;8:311–22 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1763939/pdf/v008p00311.pdf, accessed 2 February 2021). 146. Heiser PF, Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health. 1997;87(6):968–73 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380932/pdf/amjph00505-0082. pdf, accessed 2 February 2021). 147. Landman A, Bialick P. Tobacco industry involvement in Colorado. Chicago: American Lung Cancer Association; 2004 (https://escholarship.org/uc/item/9d880448#main, accessed 18 February 2021). 148. Technical manual on tobacco tax administration. Geneva: World Health Organization; 2010. (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf?sequence=1, accessed 26 January 2021). 149. Vardavas CI, Filippidis FT, Agaku IT, Mytaras V, Bertic M, Connolly GN, et al. Tobacco taxation: the 252 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N importance of earmarking the revenue to health care and tobacco control. Tob Induc Dis. 2012;10:21 (http://www.tobaccoinduceddiseases.org/Tobacco-taxation-the-importance-of-earmarking-the- revenue-to-health-care-and-tobacco,66110,0,2.html, accessed 31 October 2020). 150. Bird RM. Tobacco and alcohol excise taxes for improving public health and revenue outcomes: marrying sin and virtue?. Washington (DC): World Bank Group; 2015 (Policy Research Working Paper 7500; https://openknowledge.worldbank.org/bitstream/handle/10986/23464/ Tobacco0and0al0ing0sin0and0virtue00.pdf?sequence=1&isAllowed=y, accessed 2 February 2021). 151. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 152. The establishment and use of dedicated taxes for health. Manila: World Health Organization Western Pacific Regional Office; 2004 (https://www.who.int/health_financing/documents/dedicated_taxes. pdf?ua=1, accessed 2 February 2021). 153. Teja RS. The case for earmarked taxes: theory and example. Washington (DC): International Monetary Fund; 1988 (IMF Working Paper; https://www.elibrary.imf.org/doc/IMF001/14514- 9781451922677/14514-9781451922677/Other_formats/Source_PDF/14514-9781455212002.pdf, accessed 2 February 2021). 154. Keen M. Taxation and development – again. Washington (DC): International Monetary Fund; 2012(IMF Working Paper WP/12/220O; (https://www.imf.org/external/pubs/ft/wp/2012/wp12220.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 253 ANNEX 4.1 METHODS TO ASSESS THE NATURE AND SIZE OF THE ILLICIT TOBACCO TRADE A4.1 DIRECT MEASUREMENT A4.1.1 SMOKER INTERCEPT AND PACK OBSERVATION SURVEYS Illicit trade can be measured directly by examining the cigarette packs of smokers. The smokers themselves can provide information on purchasing patterns, brand preferences and prices paid. Researchers can select individuals or retailers to survey based on a convenience sample (i.e. a sample that may not be representative) or a probability-based sample (i.e. a sample selected to be statistically representative of an underlying population). Data collected from a pack could reveal whether the pack is compliant or non- compliant with the local tax laws. Information can be obtained from objective markings such as brands, public health warning labels, tax stamps, foreign language labels or duty-free labels. During these stops, researchers can record demographic information (e.g. age and gender of the smoker), smoking-related history (e.g. number of cigarettes smoked per day) and price information. This is helpful in understand- ing the profile of smokers who are able and willing to avoid cigarette taxes. Pack observations can be used in conjunction with population-based household surveys to obtain population-based estimates of the illicit tobacco trade (1,2). For example, as part of a regular national health survey, Kaplan et al. conducted a cross-sectional study of smokers in Turkey, using a face-to-face interviewer-administered survey and pack observation (3). They were able to collect sociodemographic, lifestyle and medical details along with pack observations as part of the study protocol. Advantages and disadvantages of smoker intercept and pack observation surveys A primary advantage of conducting pack observation is that it is direct and objective, and smokers are not subject to any value judgements (2). Paired with survey data, pack observation can appropriately account for respondents who are not residents of the area in which they are surveyed (4). Disadvantages include the difficulty of identifying areas that are representative of the tobacco use population and the difficulty of sampling important subpopulations such as elderly and immobile smok- ers. Also, surveys conducted in the daytime may discount the number of youthful smokers who are in school. Another disadvantage is that a sizeable number of smokers may refuse to show their last-purchased pack (2). Kaplan et al. found that 24% of smokers sampled in Turkey did not show their cigarette pack to the study interviewer (3). This issue may be mitigated by asking users to provide information on the brand purchased, whether any public health warnings were posted and the 254 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N price paid (5). Although the responses are based on recall, they may still yield useful information. For example, Joossens et al. allowed smokers who did not show their packs to provide self-reported information and found no statistical differences in illicit packs between these respondents and those who did show their packs (2). Another obstacle to accurate measurement based on smoker intercepts is the inability to determine the tax payment of smokers who purchase single cigarettes, since these smokers are generally not given packs. However, information can still be captured in self-report surveys by asking smokers to report the brand purchased and price paid. Key study for readers to refer to for additional guidance: Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–447. A4.1.2 PACK RETURN AND SWAP SURVEYS Pack return and pack swap surveys fall within the broader category of pack ob- servation studies that use survey sampling techniques to examine smokers’ pack characteristics and to determine whether they are tax compliant. For these surveys, the unit of analysis is the individual. The main differences between pack swap and pack return surveys is that swap surveys offer the smoker a replacement pack, whereas pack returns are built into mail surveys and allow respondents to mail in their unopened packs. Pack swap and pack return surveys use probability and nonprobability sampling procedures. Probability sampling allows researchers to generalize to the broader population. Governments can use this method to rapidly assess the availability of illicit products in a given geographic area or to measure the share of the illicit market- place. Rapid assessment may be performed in instances where there is an emerging tobacco product (e.g. a new cheap white brand) or suspected counterfeiting of tax stamp features. Rapid assessment using a convenience sampling strategy could place researchers near busy intersections where they could ask smokers for permission to look at their cigarette packs or to take photographs that could be analysed later. A population-based study requires a sample that closely mirrors the tobacco use population. Advantages and disadvantages of pack return and pack swap surveys Pack swap and pack return surveys may help to overcome the stigma associated with traditional smoking surveys. For example, when researchers ask smokers to see their cigarette packs (or when they take photographs), no value judgements are made. These surveys are good rapid-assessment tools that can be used to examine CHAP T ER 4. PO LI T I C AL ECO N OMY 255 the effectiveness of physical features of a pack designed to deter illicit trade (e.g. packs that have tracking and tracing technology or high-tech stamps). In addi- tion, they can be supplemented with population-based tobacco use surveys. When coupled with such survey data, these methods allow researchers to obtain relevant information about the context of illicit purchases, including, for example, sources (e.g. street, peer networks, retail stores) and prices. Mail-in surveys are filled out in the comfort of the respondent’s home without the presence of family members or passers-by, which may assure them that responses will be kept confidential. A potential disadvantage is that smokers who purchase both illicit cigarettes and tax-paid cigarettes may disproportionately mail back compliant packs. In addition, in LMICs, this mode of survey distribution may be unreliable because of issues associated with mail delivery systems. Key study for readers to refer to for additional guidance: Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61-66. A4.1.3 LITTERED-PACK SURVEYS Littered-pack surveys, also known as empty discarded pack surveys, are used pre- dominantly in high- and middle-income countries (e.g. the United States, France, Canada, New Zealand, Mexico and Poland). This unobtrusive method relies on the premise that smokers publicly discard packs (e.g. on streets, sidewalks and in public trash cans). The packs bear characteristics that illustrate whether they are tax compliant (e.g. tax stamps, health warnings). For example, an Albanian health warning label on a cigarette pack discarded in Greece provides evidence that the pack was destined for the Albanian market. The pack may have been smuggled into Greece by criminal entrepreneurs or it may have been brought by a visitor. Collecting discarded packs from a representative geographic sample and examining these characteristics can provide estimates of tax compliance. Operationally, this data collection method uses an ecological approach whereby geographies are the units of analysis. Geographical units are meant to represent the smokers in the city/ country and can be administratively defined (e.g. by the country’s census bureau or transit zones) or may reflect researcher-defined neighbourhoods (e.g. half-mile buffer zones near bus stops or activity spaces). Researchers in Canada have expanded the littered-pack method to include col- lection and analysis of cigarette butts on 25 postsecondary campuses. The cigarette butts provide information on the brands sold (or lack thereof) and allow researchers to distinguish between legal and illegal products (6). A recent innovative expansion 256 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N of the littered-pack methodology is the collection of packs from cigarette retailers. John and Ross collected empty packs of tobacco products from a sample of registered and unregistered retailers in India (7). Collecting packs from retailers was relevant given that single cigarettes dominate the illicit market in India. Smokers who buy single cigarettes would be unable to provide a pack in a pack swap or street intercept survey, so collecting littered packs from the ground would undercount sales of single cigarettes. The feasibility of this method is dependent on the relationship between researchers and retailers (enhanced trust) and the efforts taken to ensure confidentiality. In some countries, retailer compliance with this research method might be strained because of concerns regarding confidentiality, since retailers may face criminal and civil penalties, depending on the research findings. Advantages and disadvantages of littered-pack surveys Littered-pack surveys are generally advantageous for governments because they facilitate comparison with industry estimates. This is one of the most-preferred methods because it yields estimates that are less likely to be biased due to issues of social desirability, recall error and confidentiality that plague survey research, and they are much less expensive than face-to-face interviews used in smoker intercept or household surveys. However, there are some issues regarding these surveys, espe- cially in high-income countries, including the inability to differentiate between tax avoidance and tax evasion (8). For example, a pack in Berlin that bears a Vietnamese tax stamp may have been smuggled in mass quantity or brought in by a temporary visitor. Researchers have circumvented this issue and broadened the umbrella to measure cigarette tax noncompliance considering the potential biases introduced by tourism. Another disadvantage of littered-pack surveys is that larger budgets are needed to employ field researchers to collect, code and analyse the data. Not all countries employ tax stamps on their cigarette packaging, which may make it difficult to measure tax compliance. These surveys also can underestimate the markets in low-income countries such as India, where the main item of illegal trade is single cigarettes (7). The surveys may also overestimate illicit trade if littering behaviour is correlated with willingness to engage in illicit trade. Finally, littered-pack surveys and butt collections provide information on the proportion of butts and packs that are illegal, not the proportion of smokers that purchase illegal cigarettes (6). Key studies for readers to refer to for additional guidance: Barker DC, Wang S, Merriman D, Crosby A., Resnick EA, Chaloupka FJ. Estimating cigarette tax avoidance and evasion: evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38–i43. Merriman D. The micro-geography of tax avoidance: evidence from littered cigarette CHAP T ER 4. PO LI T I C AL ECO N OMY 257 packs in Chicago. Am Econ J Econ Policy. 2010;2(2):61–84. Stoklosa M., Paraje G., Blecher E., A Toolkit on Measuring Illicit Trade in Tobacco Products. A Tobacconomics and American Cancer Society Toolkit. Chicago, IL:Tobacconomics, Health Policy Center, Institute for Health Research and Policy, University of Illinois at Chicago, 2020 (https://tobacconomics.org/files/research/621/uic-illicit-trade-tool-kit-eng-v2.0-2. pdf, accessed 18 February 2021). A4.1.4 SELF-REPORT POPULATION SURVEYS Self-report surveys, when distributed to a representative sample of the population, can provide meaningful data on the prevalence of tax noncompliance. The surveys can be distributed to individuals or households in various ways, including face-to-face, telephone, mail and internet. Questions that specifically address illicit purchases can be added as supplementary questions to existing health or tobacco surveys. Some countries include such questions in their adult and youth tobacco surveys to estimate tax evasion/avoidance. For example, Canada’s annual Youth Smoking Survey asks smokers about the frequency of their purchases of First Nations/Native brand cigarettes (9-10). Davis et al. used data from the New York Adult Tobacco Survey to measure the source of purchase of the last cigarette pack purchased (i.e. Native American Reservations, lower-tax neighbouring states or countries, toll-free telephone numbers, the internet, duty-free shops) and the price paid (11). Twenty- eight nations currently use surveys to measure tax noncompliance as part of the ITC Project (12). Similar analyses can be conducted using questions from the Global Adult Tobacco Use Surveys. For example, Iglesias et al. used the Brazil Global Adult Tobacco Use Surveys to compare self-reported prices with a defined threshold retail price to estimate the proportion of illicit cigarette use among smokers in Brazil (13). Countries are encouraged to use existing global health surveys or to incorporate similar types of questions pertaining to illicit trade in their annual health surveys. Asking respondents about price paid per pack (including taxes), brand name and location where cigarettes were purchased (e.g. duty free shop, unlicensed vendor, internet) can contribute to a better understanding of the illicit tobacco trade. Advantages and disadvantages of self-report surveys Self-report surveys can be repeated over time to measure purchasing trends and progress associated with increases in cigarette taxation. Well-designed surveys can also provide generalizable estimates at the national level. Depending on the size of the sample, a self-report survey can provide comparable data across geographies that can help governments target resources. For example, findings that illicit cigarettes are more common in urban areas could lead to additional education campaigns and targeted enforcement. 258 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Limitations of self-report surveys include the possibility of bias due to the social stigma associated with participating in the illicit trade, which could lead survey respondents to underreport participation. Additionally, surveys may be unable to gauge whether individuals are associated with tax avoidance versus tax evasion. Finally, there is evidence that self-report household surveys may underrepresent smokers. Key studies for readers to refer to for additional guidance: Callaghan RC, Veldhuizen S, Ip D. Contraband cigarette consumption among adolescent daily smokers in Ontario, Canada. Tob Control. 2011;20(2):173–174. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany (NY): New York State Department of Health, Tobacco Control Program; 2006. A4.1.5 COVERT-PURCHASES SURVEYS A number of studies in high-, middle- and low-income countries use covert purchases of packs and single cigarettes to gauge the availability of illicit cigarettes in public and semi-private spaces (14-17). This method is also used by the tobacco industry in the United States to identify retailers who sell counterfeit cigarettes (18). Covert-purchases surveys do not provide estimates of the size of the illicit trade (i.e. market volume). Instead, they serve as a surveillance tool to identify where illicit cigarettes are sold and the extent to which they have infiltrated legal businesses. For example, a covert-purchases survey can examine whether illicit cigarettes are sold through legal retailers. It can also be used to measure compliance with emerging tobacco control policies that focus on, for example, product standardization or new regulations on flavours (e.g. plain packaging or bans on flavoured tobacco products). Covert-purchases surveys use trained researchers to visit a selected sample of retailers and directly purchase or inquire about the availability of illicit tobacco products. Retailers are not informed about the goals of the studies. Methods for determining the availability of illicit product vary. For example, in some studies, covert buyers do not directly inquire about illicit products. Instead, they purchase packs of tobacco products, paying full price, to determine whether retailers are selling illicit products under the guise that they are licit (14). The research team then examines the packs to determine whether they are legal. In the United States, researchers have observed that some consumers are paying full price for illicit untaxed packs smuggled from lower-tax states (14). Other research protocols directly ask retailers for illicit products (14,16-17). In Guatemala, Arevalo et al. specifically asked retailers for “imported cigarettes” (17). The ways covert buyers ask for illicit products may also vary geographically. For example, in some countries covert buyers may ask for “cheaper” packs or for illicit whites such as Jin Ling. CHAP T ER 4. PO LI T I C AL ECO N OMY 259 Advantages and disadvantages of covert-purchases surveys Covert purchasing allows researchers to directly identify sources of illicit cigarettes. It also allows them to measure and test the dynamics between buyer and seller. For example, researchers can experiment to see if repeated attempts to purchase products increase the likelihood of purchase (known as the familiarity protocol) (16). One methodological challenge associated with covert purchases is that it is difficult to create a sampling frame for illicit sources because some may be unknown (e.g. pubs or homes). The traditional approach is to make purchases in legal outlets, which may bias estimates. Another issue with this method is that it is difficult for buyers (also called raters) to purchase products if they are unfamiliar with the seller or do not fit the typical demographics of purchasers. Therefore, researchers using covert- purchases surveys must have detailed knowledge of the marketplace, including the ways individuals specifically ask for illicit tobacco products, and they must know whether they mirror the demographics of the neighbourhood. For example, in a study of South Bronx smokers, von Lampe et al. found that smokers looked for certain clues to assess whether they were being sold illicit cigarettes (19). Overall, this method can be quite costly because it requires training researchers, travelling to retailers and purchasing product. Covert-purchases surveys do not enable researchers to estimate the level of illicit trade, but they can provide information on availability of supply. Key studies for readers to refer to for additional guidance: Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Global Health. 2016;12(1):78. A4.1.6 SEIZURES OF GOODS Seizures are the result of enforcement activity carried out by local, national and international organizations that confiscate tobacco products that are illegally manu- factured, transported and sold. Seizures are meant to reduce the profits associated with illicit trade by confiscating proceeds (e.g. cash, cars or houses) and the tools of the trade (e.g. print and tobacco machinery). Seizures can occur at various points in the supply chain. Seizure data are often tallied by the responsible agencies and used to measure program effectiveness or as justification for requesting additional resources (e.g. personnel). Some of the data may be supplied to international customs organiza- tions, including the WCO (20). The quality of recordkeeping varies. For example, some agencies may maintain criminal files in databases that detail dates of seizure, brand names and laboratory testing. 260 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Seizures provide preliminary data on the scope of criminal activity and can help identify key trends to guide law enforcement agencies’ efforts. For example, seizures can identify trends on the modus operandi of smugglers, including transporta- tion methods (e.g. sea cargo versus trucks), point of entry and brand preference. Seizure statistics can also be used as a preliminary test to measure the efficacy of interventions. For example, Stoklosa and Ross used seizure data from the Canadian province of Nova Scotia to test the impact of a 2015 menthol ban. He found no statistically significant change in the number of menthol cigarettes seized before and after the ban (1). Advantages and disadvantages of seizures of goods Generally, seizure statistics can be readily obtained from law enforcement agencies through formal requests to agency gatekeepers (e.g. public information officers). Seizure data, however, generally do not provide a representative picture of illicit activity. For example, certain geographies may yield higher seizures because that is where the bulk of operations are being conducted. Police agencies may focus on certain geographies (e.g. locations near borders) rather than randomly inspecting, and their findings may be limited to those specific regions. Seizure data may also be skewed by the type of investigation procedures utilized. Large seizures may be the result of long-term investigations (i.e. wiretaps or culling confidential infor- mants), while smaller seizures may come from anti-smuggling cases that involve cross-border purchases of low quantities of cigarettes (less than 1 000) (2). Seizures can also be skewed by industry cooperation with law enforcement agencies. For example, the tobacco industry may be more likely to support law enforcement on counterfeit seizures rather than smuggling cases because counterfeiting impacts their brand integrity. A4.2 RESIDUAL METHODS Because the illicit tobacco trade is often decentralized, it can be difficult to observe directly. However, researchers are sometimes able to make inferences about its size without direct observation by comparing observed tobacco tax revenues with the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. The difference between observed and hypothetical revenues is called the residual and can be used as an indicator of the magnitude of illicit trade. Even when the residual is only an approximate measure, changes in its size may be a reliable indicator of changes in the size of the trade. When actual tobacco tax revenues are reliably observed, the main challenge for residual methods is that of producing accurate estimates of the amount of tax revenue that hypothetically would have been collected had all tobacco consumption been taxed. CHAP T ER 4. PO LI T I C AL ECO N OMY 261 A4.2.1 GAP ANALYSIS Gap analysis is the preferred residual methodology because it is intuitive, straight- forward and relatively easy to explain to policy-makers and the general public, and it has been widely employed in government studies (4). Researchers using gap analysis compare survey-based self-reported consumption data with observed (usually administrative) data on tax-paid sales. The basic premise is that if both self-reports and observed data are accurate, any difference between reported consumption and tax-paid sales can be explained by legal imports of non-taxed cigarettes (such as duty-free sales), exports of taxed cigarettes, tax evasion or tax avoidance. The greatest research challenge in implementing gap analysis – as with most residual methods – is obtaining reliable and accurate estimates of tobacco consump- tion. In its simplest implementation, gap analysis calculates the residual as the difference (which should be a minimum of zero) between the amount of tobacco consumption reported in surveys and tax-paid sales, which are generally available from administrative sources, minus exports. This simple calculation, however, is generally flawed, since surveys of reported tobacco consumption underestimate true consumption. Underreporting of tobacco consumption may result from survey respondents’ reticence about disclosing behaviour that is viewed as unhealthy and potentially socially undesirable. Depending on the legal and cultural context, cer- tain groups (e.g. women or youth) may be more likely than others to underreport consumption. Other groups (e.g. rebellious young men and women) may accurately report or even overestimate consumption. Reuter and Majmundar measured actual consumption by total national taxed sales in the United States, where both legal untaxed imports and exports of taxed tobacco are widely believed to be very small, and found that the ratio of self-reported consumption to actual consumption was only 65% (4). After incorporating this survey underreporting into their analyses and considering the evidence from their gap analysis and the literature, Reuter and Majmundar found that the illicit market in the United States, which largely consists of avoidance or evasion of subnational state taxes, is between 8.5% and 21% of consumption (4). The higher range of the estimate is consistent with prior estimates using population-based pack observation studies (21). Researchers using gap analysis for countries or regions where legal imports of untaxed tobacco (such as duty-free products) or (legal or illegal) exports of taxed cigarettes are more significant should attempt to incorporate data about, or estimates of, these factors into their calculations. Legal untaxed imports of tobacco should be subtracted from reported consumption (after adjustment for underreporting), and exports of taxed tobacco should be subtracted from taxed sales. Obtaining data about legal untaxed imports and exports of taxed tobacco may be challenging, because these imports and exports may be the result of decentralized decisions of individual 262 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N travellers as they cross tax borders. Data on these activities will not necessarily be collected through normal administrative activities. Despite these challenges, gap analyses may still prove useful. For example, if researchers have reason to believe that misreporting of tobacco consumption and the amounts of legal untaxed imports and taxed exports are relatively stable or follow known trends (e.g. are falling) over time, gap analyses can be used to provide estimates or lower (or upper) bounds on illicit trade when several years of data on taxed sales and reported consumption are available. Data sources may be country tax administrators who have access to sales data and health departments that have access to population-level studies of reported tobacco con- sumption. In this context, multiple years of data on reported consumption and tax-paid sales can allow researchers to estimate changes in the size of the illicit trade even when it is difficult to measure the absolute level. Paraje used the 2008 Global Adult Tobacco Survey and the 2013 National Health Survey to measure reported tobacco consumption in Brazil (22-23). Advantages and disadvantages of gap analysis A major advantage of gap analysis is that when quality data are available, it is simple, easily reproduced and explainable to policy-makers and the general public. How- ever, high-quality data on reported consumption may not be available, especially in low-income countries. In many cases, gap analysis does not provide reliable information on the size of the illicit market but only on changes in the size over time (22). Additionally, some low-income countries may not have reliable estimates of tax-paid cigarette sales, and secondary data repositories of cigarette sales may not be transparent about their methodology (24). Another disadvantage of gap analysis is that it generally cannot be used to obtain separate estimates of tax avoidance and tax evasion. Biased estimates may also result if surveys of tobacco consumption are not representative of the population (25). Moreover, it is generally not possible to quantify the precision of the estimates or uncertainty associated with the estimates, because of both statistical uncertainty resulting from the use of samples to imperfectly represent populations (e.g. the share of the population that smokes) and uncertainty about key facts such as the degree to which survey respondents understate their tobacco consumption. Key studies for readers to refer to for additional guidance: Szklo A, Iglesias RM, Carvalho de Souza M, Szklo M, Maria de Almeida L. Trends in illicit cigarette use in Brazil estimated from legal sales, 2012–2016. Am J Public Health. 2018;108(2):265–269. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine and Tob Res. 2019;21(8):1079–86 CHAP T ER 4. PO LI T I C AL ECO N OMY 263 A4.2.2 ECONOMETRIC MODELLING There is a long tradition of using data to estimate parameters of demand functions that relate the quantity of goods consumed to the prices faced by consumers, their incomes and other variables. Because of the addictive nature of tobacco – and because of important public health and public policy concerns relating to tobacco use – economists have paid particular attention to the estimation of cigarette demand functions (26). As the literature on this topic developed, it became apparent that taxed tobacco sales would be a biased indicator of tobacco consumption if some consumers obtained their tobacco in illicit markets. Similarly, the price of cigarettes in the legal market might overestimate the price paid by consumers if some sales were not tax-paid. While economists generally cannot observe sales in the illicit market, they have been able to develop models that predict conditions under which consumers avoid tobacco taxes. They reason that the relative size of illicit tobacco markets depends primarily on two variables: the relative price of taxed and untaxed consumption and the ease of obtaining lower-cost (untaxed) tobacco. Other variables, includ- ing the social stigma from evading tax laws and the perceived relative quality of illicit tobacco, could also influence the demand for it. While illicit trade cannot be directly observed, it can be estimated from the difference between tax-paid sales and predicted consumption. Tax-paid sales can be less than predicted consump- tion when retailers or consumers evade taxes. They can be greater if some tax-paid cigarettes are bought within the jurisdiction and then consumed in areas where after-tax prices are higher. Econometric modelling estimates of illicit trade must be tailored to the situation in the country that is being studied, and therefore the data requirements may differ substantially from case to case. Researchers using this method should be familiar with the literature and should also understand the conditions in the areas they are researching. They must always include some measure of tobacco consumption or sales and some measure of the price of tobacco in the home country, as well as other variables (e.g. income) that are known to affect the demand for tobacco. It is also generally necessary to include variables that measure the availability and relative price of illicit tobacco, which can often be measured by comparing tobacco taxes in the home country with those in areas that are the source of illicit tobacco. Advantages and disadvantages of econometric modelling The major advantage of econometric modelling is that it is consistent with a long tradition of economic theory and practice, and the quality of the modelling techniques and empirical estimates can therefore be evaluated against widely accepted criteria. Empirical analyses provide estimates of price elasticities, income elasticities and price elasticities of tax avoidance. A substantial literature base makes it possible to 264 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N rigorously quantify uncertainty about the estimates and to test their robustness to various assumptions made in the modelling process. Estimates can be compared to other estimates available in the literature, and the results of these analyses can be used to simulate the impacts of policy changes (including tax and/or enforcement increases) on both consumption and tax avoidance. Because this methodology provides direct estimates of the uncertainty of the results, researchers can specify their level of confidence in the findings. A shortcoming of econometric modelling is that it requires high-quality data on a variety of important variables over a period of time, as well as advanced econo- metric modelling expertise. Also, because results from the econometric models are based on statistical inference and economic theory rather than direct observation (e.g. the proportion of packs without tax stamps), it can be difficult to explain to policy-makers and the general public. Key studies for readers to refer to for additional guidance: Becker GS, Grossman M, Murphy KM, (1994). An empirical analysis of cigarette addiction. Amer Econ Review. 1994;84(3):396–418. Schafferer C, Yeh CY, Chen SH, Lee JM, Hsieh CJ. A simulation impact evaluation of a cigarette excise tax increase on licit and illicit cigarette consumption and tax revenue in 36 European countries. Public Health. 2018;162:48–57. A4.2.3 EXPERT OPINION (KEY-INFORMANT SURVEYS AND INTERVIEWS) Insight on illicit trade dynamics can come from experts in the field, including researchers (e.g. in economics, criminal justice and public health), taxation depart- ments, enforcement agencies, product manufacturers, wholesalers and retailers. Other key informants include journalists and academics who have secured confi- dential informants. Experts can provide novel information about emerging trends (e.g. new smuggling routes). In some cases, researchers can obtain interviews with incarcerated or active offenders (27-29). For example, researchers studying cigarette smuggling in eastern Africa conducted interviews with more than 150 Ugandan tobacco smugglers (29). Experts can be queried through surveys or semi-structured interviews. When sampling frames are available (e.g. directories of tax department employees), surveys are more expedient than interviews. However, when experts are hard to find, non- random sampling strategies coupled with interviews are recommended. Identifying experts may require recruiting a gatekeeper who is tasked with helping researchers find additional experts; or purposive sampling, where individuals are identified based on set criteria (e.g. they are taxation experts employed by local governments) (30). CHAP T ER 4. PO LI T I C AL ECO N OMY 265 Advantages and disadvantages of key-informant surveys and interviews Informant interviews can be a useful starting point for identifying trends in the marketplace (e.g. venues where illicit cigarettes are sold or modes of entry). One of the disadvantages of relying on informants is that the information solicited from them may not be generalizable. Expert knowledge may be outdated or limited by the informants’ experience. Furthermore, the opinions of experts are subjective and may be biased by the experts’ employment status and the sampling methods used. For example, persons working in law enforcement may overestimate the extent of bootlegging in order to secure additional funding for future operations. Similarly, manufacturers looking to defeat taxes may overestimate the illicit trade to illustrate the links between taxation and illicit behaviour. Alternatively, tobacco control advocates may underestimate illegal market measures in order to support the argument that taxes do not increase illicit trade. Key studies for readers to refer to for additional guidance: Joossens L, Raw M. Cigarette smuggling in Europe: who really benefits? Tob Control. 1998;7:66–71. doi:10.1136/tc.7.1.66 PMID: 9706757. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20(3):226–232. A4.3 MIXED AND MULTIMETHOD STUDIES Given the shortcomings of the aforementioned methods for assessing the nature and size of the illicit tobacco trade, governments may want to validate their findings by using mixed or multiple methodologies. Mixed methods use two methodological paradigms, qualitative and quantitative, as tools for exploration and explanation. For example, mixed method studies can use littered-pack surveys to measure the size of the market along with self-report surveys of smokers to understand patterns of purchasing, including sources, frequency and social norms. For example, Stoklosa and Ross estimated the share of the illicit market in Poland using a population- based self-report survey and a littered-pack survey (1). Using both types of survey simultaneously enables governments to assess their validity in estimating the size of the illicit market. Alternatively, governments can employ multimethod research, i.e. the use of multiple methods that are similar in tradition (e.g. focus groups and semi-structured interviews) (31). Saenz de Miera et al. used face-to-face interviews (households), litter collection and observation of single-stick sellers, which enabled them not only to cross-validate the two major methodologies, but also to see if the brand of the single stick was a good measure of licit versus illicit trade (33). 266 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Advantages and disadvantages of mixed and multimethod studies Mixed and multimethods studies enable researchers to check the validity of their findings. Multiple methods are preferred in contexts where illicit trade estimates are politicized. For example, low estimates may be challenged by the tobacco indus- try, while high estimates may be challenged by tobacco control researchers and/or proponents. Mixed and multiple methods (e.g. littered-pack surveys and informant interviews) can enable governments to understand the situational context in which the illicit trade operates, including the actors involved and venues of sale. One disadvantages of using mixed and multiple methods is cost. Governments that are constrained by tight budgets may choose to use a single method that provides the most accurate information. However, given the issues faced by each method, this may not be feasible – each method has limitations. Instead, governments can pair a high-cost method with a lower-cost method (e.g. pairing interviews with empty pack surveys, or law enforcement seizure data with face-to-face consumer surveys). Key study for readers to refer to for additional guidance: Zaloshnja E, Ross H, Levy DT. The impact of tobacco control policies in Albania. Tob Control. 2010;19:463–468. CHAP T ER 4. PO LI T I C AL ECO N OMY 267 REFERENCES 1. Stoklosa M, Ross H. Contrasting academic and tobacco industry estimates of illicit cigarette trade: evidence from Warsaw, Poland. Tob Control. 2014; 23(e1), e30–e34 (https://www.researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_estimates_of_illicit_cigarette_ trade_Evidence_from_Warsaw_Poland, accessed 1 February 2021). 2. Joossens L, Lugo A, La Vecchia C, Gilmore AB, Clancy L, Gallus S. Illicit cigarettes and hand-rolled tobacco in 18 European countries: a cross-sectional survey. Tob Control. 2014;23:e17–e23 (https:// www.ncbi.nlm.nih.gov/pmc/articles/PMC3812425/pdf/nihms491463.pdf, accessed 2 February 2021). 3. Kaplan B, Navas-Acien A, Cohen JE. The prevalence of illicit cigarette consumption and related factors in Turkey. Tob Control. 2018;27(4):442–7. 4. Reuter P, Majmundar M. Understanding the US illicit tobacco market: characteristics, policy context, and lessons from international experiences. Washington (DC): National Academies Press; 2015. 5. Maldonado N, Llorente BA, Iglesias RM, Escobar D. Measuring illicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_4/ s260.full.pdf, accessed 29 January 2021). 6. Barkans M, Lawrance KA. Contraband tobacco on post-secondary campuses in Ontario, Canada: analysis of discarded cigarette butts. BMC Pub Health. 2013;13(1):335 (https://bmcpublichealth. biomedcentral.com/articles/10.1186/1471-2458-13-335, accessed 2 February 2021). 7. John RM, Ross H. Illicit cigarette sales in Indian cities: findings from a retail survey. Tob Control. 2018;27(6), 684–688. 8. Davis KC, Grimshaw V, Merriman D, Farrelly MC, Chernick H, Coady MH, et al. Cigarette trafficking in five northeastern US cities. Tob Control. 2014;23(e1):e62–e68. 9. Callaghan RC, Veldhuizen S, Leatherdale S, Murnaghan D, Manske S. Use of contraband cigarettes among adolescent daily smokers in Canada. CMAJ. 2009;181(6-7):384–6 (https://www.cmaj.ca/content/ cmaj/181/6-7/384.full.pdf, accessed 2 February 2021). 10. Guindon GE, Burkhalter R, Brown KS. Levels and trends in cigarette contraband in Canada. Tob Control. 2017;26(5):518–25 (https://www.researchgate.net/publication/307890937_Levels_and_trends_in_ cigarette_contraband_in_Canada, accessed 2 February 2021). 11. Davis K, Farrelly M, Li Q, Hyland A. Cigarette purchasing patterns among New York smokers: implications for health, price, and revenue. Albany: New York State Department of Health; 2006 (https://www. health.ny.gov/prevention/tobacco_control/docs/cigarette_purchasing_patterns.pdf, accessed 2 February 2021). 12. International Tobacco Control Evaluation Project (2018). Surveys. Waterloo: University of Waterloo; 2018 (http://www.itcproject.org/surveys, accessed 12 October 2020). 13. Iglesias RM, Szklo AS, de Souza MC, de Almeida LM. Estimating the size of illicit tobacco consumption in Brazil: findings from the global adult tobacco survey. Tob Control. 2017;26(1):53–9. 14. Silver D, Giorgio MM, Bae JY, Jimenez G., Macinko J. Over-the-counter sales of out-of-state and counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25(5):584–586. 15. Brown J, Welding K, Cohen JE, Cherukupalli R, Washington C, Ferguson J, et al. An analysis of purchase price of legal and illicit cigarettes in urban retail environments in 14 low-and middle-income countries. Addiction. 2017;112:1854–60 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5600117/pdf/ADD- 112-1854.pdf, accessed 2 February 2021). 16. Scollo M, Bayly M, Wakefield M. Availability of illicit tobacco in small retail outlets before and after the implementation of Australian plain packaging legislation. Tob Control. 2015;24(e1):e45–e51. 17. Arevalo R, Corral JE, Monzon D, Yoon M, Barnoya J. Characteristics of illegal and legal cigarette packs sold in Guatemala. Globalization and Health. 2016;12(1):78 (https://globalizationandhealth. biomedcentral.com/articles/10.1186/s12992-016-0219-z, accessed 2 February 2021). 18. Phillip Morris USA Inc. v. Shalabi. United States, District Court, C.D. California; 2004 (https://www. casemine.com/judgement/us/5914b6cbadd7b0493477b3da, accessed 2 February 2021). 19. von Lampe K, Kurti M, Johnson J, Rengifo AF. ‘I wouldn’t take my chances on the street’ navigating illegal cigarette purchases in the South Bronx. J Res Crime Delinq. 2016;53(5):654–80 (https://www. 268 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N researchgate.net/publication/298515449_I_Wouldn’t_Take_My_Chances_on_the_Street_Navigating_ Illegal_Cigarette_Purchases_in_the_South_Bronx, accessed 2 February 2021). 20. Customs and Tobacco Report 2009. Brussels: World Customs Organization; 2009. (http://www.wcoomd. org/en/media/newsroom/2010/june/~/media/83967DFEB9F74D388924A4C61F279DC4.ashx, accessed 12 October 2020). 21. Fix BV, Hyland A, O’Connor RJ, Cummings KM, Fong GT, Chaloupka FJ, et al. A novel approach to estimating the prevalence of untaxed cigarettes in the US: findings from the 2009 and 2010 International Tobacco Control Surveys. Tob Control. 2014;23:i61–i66 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3984758/pdf/nihms567943.pdf, accessed 2 February 2021). 22. Paraje G. Illicit cigarette trade in five South American countries: a gap analysis for Argentina, Brazil, Chile, Colombia and Peru. Nicotine Tob Res. 2019;21(8):1079–86. 23. National Survey of Health. Instituto Brasileiro de Geografia e Estatística; 2013 (in Portuguese) (https:// www.ibge.gov.br/en/statistics/social/health/16840-national-survey-of-health.html?=&t=downloads, accessed 27 November 2020). 24. Blecher E, Liber A, Ross H, Birckmayer J. Euromonitor data on the illicit trade in cigarettes. Tob Control. 2015;24:100–1. (https://tobaccocontrol.bmj.com/content/tobaccocontrol/24/1/100.full.pdf, accessed 1 February 2021). 25. Ross H. Understanding and measuring cigarette tax avoidance and evasion: a methodological guide. Tobacconomics; 2015 (https://tobacconomics.org/wp-content/uploads/2015/03/Ross_Methods_to_ Measure_Illicit-Trade_03-17-15.pdf, accessed 12 October 2020). 26. Chaloupka FJ, Warner KE. The economics of smoking. In: Arrow KJ and Intriligator MD, editors. Handbook of Health Economics. Amsterdam: Elsiver; 2000. pp1539–1627. 27. Antonopoulos GA. Cigarette smugglers: a note on four ‘unusual suspects’. Glob. Crime. 2007;8(4):393–8 (https://www.researchgate.net/publication/248955367_Cigarette_Smugglers_A_Note_on_ Four_’Unusual_Suspects’, accessed 2 February 2021). 28. Antonopoulos GA. The Greek connection(s): the social organization of the cigarette-smuggling business in Greece. Eur J Criminol. 2008;5(3):263–88 (https://www.researchgate.net/publication/249752218_ The_Greek_ConnectionsThe_Social_Organization_of_the_Cigarette-Smuggling_Business_in_Greece, accessed 2 February 2021). 29. Titeca K, Joossens L, Raw M. Blood cigarettes: cigarette smuggling and war economies in central and eastern Africa. Tob Control. 2011;20:226–32 (https://www.researchgate.net/publication/49809289_ Blood_cigarettes_Cigarette_smuggling_and_war_economies_in_central_and_eastern_Africa, accessed 2 February 2021). 30. Babbie, ER. The basics of social research. Boston: Cengage Learning; 2013. 31. Tashakkori A, Teddlie C., editors. Sage handbook of mixed methods in social & behavioral research. New York: Sage; 2010. 32. Saenz de Miera Juarez B, Reynales-Shigematsu LM, Stoklosa M, Welding K, Drope J. Measuring the illicit cigarette market in Mexico: a cross validation of two methodologies. Tob Control;2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/03/31/tobaccocontrol-2019-055449. full.pdf, accessed 2 February 2021). CHAP T ER 4. PO LI T I C AL ECO N OMY 269 ANNEX 4.2 HOW ARE THE TOBACCO TAX REVENUES EARMARKED? The introduction of earmarking of tobacco tax revenue is almost always combined with an increase in excise taxes (or a new surcharge) rather than reallocation of existing revenues (1). Table A4.1 provides examples of the different approaches used by several countries to earmark tobacco tax revenues. Table A4.1 Approaches used to earmark tobacco tax revenues FUNDING SOURCES/ TYPE OF TAX EXAMPLES OF TAX BASE AND RATES As part of the excise system (tobacco, alcohol) Specific Republic of Korea: 841 won (US$ 0.75) per pack or 29% of the specific excise rate Costa Rica: 467.8 cólones (US$ 0.83) per pack or 100% of the specific excise rate Congo: 20 CFA francs (US$ 0.036) per pack or 50% of the specific excise rate Ad valorem Colombia: 10% of retail price (equivalent to 100% of the ad valorem rate) New levy (surcharge on the existing excise or completely new levy) Specific Egypt: additional 0.75 Egyptian pounds (US$ 0.042) per pack Ad valorem Thailand: surcharge of 2% over the excise tax base Botswana: new tobacco levy of 30% of the cost of production or CIF Percentage of excise revenue Cook Islands: 50% of revenues from the excise tax on tobacco Guatemala: 100% of revenues from the excise tax on tobacco Note: Conversions of amounts from the local currency were made using the official exchange rates from the IMF as of 31 July 2018 (date of the data collection). Source: (2). REFERENCES 1. Earmarked tobacco taxes: lessons learnt from nine countries. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng.pdf?sequence=1, accessed 2 February 2021). 2. WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: World Health Organization; 2019 (https://www.who.int/teams/health-promotion/tobacco-control/who- report-on-the-global-tobacco-epidemic-2019, accessed 24 January 2021). 270 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N 271 CHAPTER 5. Best practices in tobacco tax policy and administration TAX POLICY Use excise tax increases to achieve the public health goal of reducing the death and diseases caused by tobacco use Extensive research has clearly demonstrated the effectiveness of higher tobacco product taxes and prices in reducing tobacco use and its harmful consequences, particularly among the poor and the young. In fact, tobacco excise tax increases are the single most effective and cost-effective policy for reducing tobacco use. Excise taxes are the most significant taxes applied on tobacco products because of their ability to raise both absolute and relative prices. Tobacco excise tax increases also generate sizeable new revenues that will be sustained in the short to medium term. In the long term, continued increases in tobacco taxes – coupled with implementa- tion of other evidence-based tobacco control policies and programmes – will lead to even larger reductions in tobacco use and its consequences. Include significant tobacco excise tax increases as part of a comprehensive strategy to reduce tobacco use Governments should adopt a comprehensive tobacco control strategy that includes objectives for reducing adult tobacco use and preventing youth tobacco use. Rais- ing excise taxes significantly is the most effective, as well as the most cost-effective, measure for reducing consumption. When combined with other demand reduction interventions, the impact of tax increases on tobacco use is even stronger. Such interventions include comprehensive smoke-free policies in all public spaces, total bans on tobacco advertising, promotion and sponsorship by tobacco companies, large graphic health warnings about the consequences of tobacco use, plain packaging, broad efforts to help current users quit and mass media public education campaigns. Implementation of a comprehensive strategy to reduce tobacco use leads to greater reductions in the harmful consequences of tobacco use, builds public and political support for higher taxes and maximizes the effectiveness of tax increases in achieving public health objectives. 272 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Involve the competent authority from the start when considering the revision of a tax policy Competent authorities such as tax administrations and customs authorities are key partners in the effective implementation of a tax policy. Policy-makers need to ensure that those competent authorities are consulted and involved in the tax policy revision process so that their concerns about the impacts of policy change on enforcement can be taken into account from the beginning. This can also help identify and address possible loopholes early on in the enforcement process. Coordination among relevant bodies, including close cooperation and sharing of information, will optimize enforcement of tax policy and tax collection. To streamline the process of cooperation and exchanges of information, a basis in law needs to be established. Additionally, the involvement of tax administration authorities in the entirety of the tax revision policy process is important to ensure effective implementation of the policy. Promote greater policy coherence across sectors such as agriculture, industry, trade, finance and labour Greater multisectoral integration and policy coherence is needed at the country level to achieve effective health improvements. In particular, it is important to ensure that public policies and interventions in non-health sectors (e.g. agriculture, industry, trade, finance and labour) do not act against the intended public health impact of tobacco control and taxation (such interventions include providing subsidies to tobacco growing or manufacturing). TAX DESIGN Tax structure matters and simpler is better Complex tax structures are difficult to administer, create opportunities for tax avoid- ance and evasion and are less effective than simpler structures in achieving public health and revenue goals. Simplifying the structure of tobacco excise taxes will facilitate tax administration, reduce tax avoidance and evasion, enhance revenues and have a greater impact on tobacco use by reducing incentives to substitute among tobacco products or brands in response to tax increases. Countries with multiple tiers of tobacco tax rates based on product characteristics (e.g. price level, length, weight, type of tobacco) should reduce and eventually eliminate these differential tax rates. An appropriate transition strategy is to reduce the variations in tax rates over time with the aim of implementing a uniform tax (i.e. a single rate applies whether excise is ad valorem or specific) on a given tobacco product. Applying a uniform tax to all brands of a given tobacco product also sends a clear message that they are equally harmful. CHAP T ER 5. B E S T PR AC T I CE S 273 Rely more on specific tobacco excises to drive price increases Greater reliance on specific excise taxes maximizes the impact of tobacco taxes on public health by reducing the gap in prices between premium and low-priced alternatives and limiting opportunities for users to switch down in response to tax increases. For countries that currently rely on an ad valorem tax, an appropriate first step would be to shift to a mixed system by adding a sizeable specific component or introducing a high minimum specific excise tax (an excise tax floor). For countries that rely on a mix of ad valorem and specific taxes, the specific tax component should be increased regularly so that it accounts for a greater share of the total excise tax. Increase tobacco taxes significantly to reduce the affordability of tobacco products To maximize the public health impact of higher tobacco taxes while at the same time generating higher revenues, governments should significantly raise taxes to increase prices and reduce the affordability of tobacco products. In many LMICs, tobacco use increases with incomes, and since incomes rise faster than tobacco product prices, these products are becoming more affordable. To reduce afford- ability, tax increases need to result in real price increases that are higher than the increases in real incomes. Where revenue increases are a goal, rely on regular excise tax increases If governments want to increase tobacco revenues, they must increase excise taxes regularly. From the tax revenue perspective, the important determinant is the tax base elasticity, which has three key components: the price elasticity of demand of tobacco, the share of the tax in the retail price and the degree of pass-through of the excise tax rate increase on to retail price. Tax increases will increase revenues at least in the short to medium term, because demand is price inelastic, tax levels are generally low as a proportion of retail prices and the pass-through of tax increases on to retail prices is unlikely to be higher than the tax increase itself (i.e. there is no overshifting). In addition, increasing tax rates is the only policy measure that can reverse reduced revenues in a declining market that has strong tobacco control policies. Automatically adjust specific tobacco taxes for inflation and income growth Unless specific tobacco taxes are regularly adjusted, their real value will fall over time as general price levels increase. When this happens, their effectiveness in reducing tobacco use will be diminished. Governments should establish a mechanism for automatically adjusting specific taxes to keep pace with inflation. Recently, some governments have begun to extend this indexation to include income growth as well, further ensuring that tobacco does not become more affordable over time. 274 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N Pricing regulations cannot be considered an alternative to excise tax. However, in some specific contexts, pricing regulations could be used in conjunction with excise taxes to help ensure the effective implementation of tax increases In certain contexts where increasing taxes is challenging or the tax structure is weak, non-tax policies such as pricing regulation (specifically, minimum mark-ups and price floors/minimum prices) may be seen as a second-best alternative to ensure a high price level and dissuade consumption of tobacco products. These policies, however, do not necessarily lead to the desired price level, nor do they protect consumers and government from industry manipulation. However, in the context of powerful multinationals that sell brands across all market segments and could easily undershift a tax increase to cheaper brands – or where price promotions cannot be banned – minimum price policies may help increase the effectiveness of tax increases, especially if the minimum prices are increased regularly. Implement nontax policies affecting price levels, such as banning promotional discounts for tobacco products and the sale of single sticks of cigarettes The banning of promotional discounts is usually dealt with in tobacco control laws under the Tobacco Advertising, Promotion and Sponsorship provision. Do not allow concerns about the inflationary impact of higher tobacco taxes to deter tax increases Given that wages or some government spending may be tied to a price index, govern- ments can reduce concerns about the inflationary impact of a tobacco tax increase by using a price index that excludes tobacco products. TAX PARITY Tax all tobacco products in a comparable way Increasing excise taxes on some tobacco products but not on others results in changes in the relative prices of these other products. This induces substitution towards relatively less-expensive products – for example, from expensive manufactured cigarettes to other, cheaper tobacco products such as RYO tobacco, bidis, cheroots or chewing tobacco. As a result, the overall reduction in tobacco use is smaller than it would have been had all taxes increased by comparable amounts. Comparable increases in the taxes on all tobacco products maximize the public health impact of tobacco tax increases by minimizing opportunities for substitution. Moreover, increases in taxes on all tobacco products will generate larger increases in revenues. CHAP T ER 5. B E S T PR AC T I CE S 275 Strictly regulate new and emerging tobacco and nicotine products where they are not banned and impose an excise tax In recent years, the world has been experiencing the rise of new and emerging tobacco and nicotine products including ENDS, ENNDS and HTPs. The tobacco industry claims these new products are safer than traditional tobacco products, but the evidence so far suggests that they could pose a threat to public health, especially if they attract new or young users or prevent current smokers from quitting. The market and demand dynamics of newer products – as well as initiation, smoking cessation and switching behaviour among different socioeconomic groups – are not yet clear. Best practices for taxing new and emerging tobacco and nicotine products, based on current knowledge, are that: 1. HTPs should be taxed at the same level as cigarettes and, in terms of structure, through a specific excise per unit regardless of tobacco content. HTPs contain tobacco and should be treated as a tobacco product. 2. ENDS/ENNDS products should be taxed in a manner that discourages up- take by youth and non-users. Nicotine- and non-nicotine-delivery systems containing e-liquids should be taxed equally. 3. Countries can also consider taxing the devices used for ENDS/ENNDs and HTP consumption, but they need to adequately assess their administrative capacity to do so. While these newer products create additional challenges for tobacco control, it is important to remember that cigarettes remain by far the predominant tobacco product and that raising taxes and prices on cigarettes – and thereby reducing their use – should remain the top priority. MONITORING AND EVALUATION Know your market Know your market well. The type of tax structure you choose and the impacts it will have on consumption and tax revenue are shaped by the particular dynamics of your market. Understanding the nature and degree of competition in your market is vital to selecting the appropriate type of tax structure and policies to achieve your public health and revenue objectives. This knowledge will also facilitate more accurate estimates of the impacts of a tax increase, as well as better anticipation of industry responses. Assess the impact of your policies to design and implement the most effective tobacco excise tax policies Monitoring and evaluation are essential for effective tobacco taxation, and they 276 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N should be built into the initial design – or redesign – of tobacco tax policies. A number of tools exist to help policy-makers pre-emptively assess the effects of a proposed tobacco control policy on consumption, smoking prevalence and lives saved. The WHO TaXSiM uses target simulations to assist governments in predict- ing how specific tax changes will impact consumer prices, consumption and tax revenue in their market. Adopt indicators that help you measure improvements in tax policy and its impact Building and monitoring indicators of tax and tobacco control policies helps policy- makers assess the improvement of their policies and determine if those policies have an impact on tobacco use over time. The tax share of the retail price for a particular tobacco product is a key indicator that should be used in conjunction with an af- fordability indicator. A recommended target for countries to aspire to is to have an excise tax that represents at least 70% of the retail price of tobacco products. Another useful indicator to assess the performance of the tax policy overall is the use of a tax scorecard, which synthesizes best practices in tobacco taxation by combining the four key components of tax policy (price level, change in affordability over time, total and excise tax share in the retail price and tobacco tax structure). TAX ADMINISTRATION Implement best practice approaches in general tax administration to make tobacco tax administration more effective and efficient Best practice approaches include (1) defining clearly the roles and responsibilities of competent authorities, (2) ensuring effective coordination among relevant bodies at the national and international levels and (3) undertaking evaluation of performance and accountability against pre-agreed indicators to identify points for improvement. Ensure compliance and accuracy of information on the tax compliance cycle To achieve this, implement the following actions: • Require licences for manufacturing, importing, exporting, retailing, growing, transporting, wholesaling, brokering, warehousing and distributing tobacco products. This will help secure the supply chain while obtaining valuable infor- mation, e.g. through access to companies’ accounting and inventory systems. • Make sure all persons and entities engaged in the supply chain of tobacco, tobacco products and manufacturing equipment keep complete and accu- rate records of all relevant transactions and details of materials used in the production of tobacco products. CHAP T ER 5. B E S T PR AC T I CE S 277 • Ensure that tax declarations collect as much information as possible on the taxpayer. • Collect taxes close to the point of production and import to limit the number of taxpayers a competent authority needs to manage. • Maintain a system of authorization for warehousing to carry out controls in production and storage facilities to ensure that taxes are paid. • Use electronic methods, through the best available IT, for declarations and collection of taxes. This allows for cross-check of information provided in dec- larations with information from other government agencies and third parties. Ensure control and enforcement on the supply chain To achieve this, implement the following actions: • Include control and enforcement as a fundamental pillar in the strategic plan of the tax administration overall. • Use a risk-based approach by choosing defined targets for enforcement and control, such as those who have a higher probability of noncompliance. • In the licensing process, ensure that purchases from unlicensed suppliers or sales to unlicensed purchasers are not allowed. Ensure also that the validity of licences is limited in time and require renewals or reapplication to maintain a high level of control. • Use tax stamps with strong security features to reduce the risk of stamp counterfeiting. These markings facilitate the collection of excise taxes, audits and enforcement actions. • Implement a tracking and tracing system for tobacco products. A tracking and tracing system assists authorities in determining the origin of tobacco products and the point of diversion, if applicable, and monitoring and control- ling the movement of tobacco products and their legal status. • Implement anti-forestalling measures so that forestalling does not delay a tax increase and its intended effect on revenues and consumer behaviour. • Control import and export of tobacco products and manufacturing equipment by allowing only duly licensed natural persons or legal entities to conduct such activities. • Strengthen border control, e.g. by utilizing non-invasive tools such as X-ray scanners and dogs to detect tobacco products. • Limit or tightly control and, ideally, ban activities related to production and trade of tobacco products in tax-free zones to avoid opportunities for tax evasion. • Prohibit intermingling of tobacco products with non-tobacco products in a single container or any other similar transportation unit when removed from tax-free zones. 278 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N • Prohibit the sale to or import by international travellers of tax-free or duty- free tobacco products. These sales erode the effects of tax and price measures aimed at reducing the demand for tobacco products and adversely affect government revenues by creating a loophole in the tax structure. Clearly define procedures to follow after detecting illicit trade of tobacco • Take immediate action to seize and destroy smuggled and/or illicit tobacco and collect due taxes. • Ensure certain, swift and severe sanctions for those caught engaging in illicit trade in tobacco products, such as penalties, fines and withdrawal of licences. It can also be effective to consider illicit trade in tobacco products by law as a source of money-laundering. Become a Party to and/or implement the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products The WHO FCTC Protocol provides a blueprint of best practices and policies for dealing with illicit trade and should be part of any strategy to fight it. Implement, to the extent possible, the same rules and regulations for tax administration and enforcement for all tobacco products, as well as new and emerging nicotine and tobacco products Implement broad policies for ensuring a good tax system that will trickle down to good tax administration of tobacco products by: • ensuring proper resourcing of competent authorities; • having strict rules and regulations to detect corruption and to punish both personnel and taxpayers who are engaged in corrupt practices; and • ensuring a strong judicial system that is independent in fact and in perception, where disputes are solved quickly. The appeal process should have limits so that appeals cannot continue for years. The use of criminal rather than civil charges should also be considered, especially for illicit trade. POLITICAL ECONOMY Beyond the technical soundness of best practices in tax policy and administration, a critical factor in advancing tobacco taxes is the ability to get the political buy-in of the highest instances in the government. One key strategy is to address concerns around the political economy of tobacco taxation, which are often exploited by the tobacco industry to block major reforms. CHAP T ER 5. B E S T PR AC T I CE S 279 SCARE tactics The tobacco industry uses SCARE tactics to dissuade governments from implement- ing tobacco tax increases. These include smuggling and illicit trade (S), court and legal challenges (C), anti-poor rhetoric (A), revenue reduction (R) and employment impact (E). Best practices for countering these tactics are described below. S: Smuggling and illicit trade Do not allow concerns over the impact of increasing excise taxes on illicit trade in tobacco affect your decision to increase them. Rely on your own estimates of the level and nature of illicit trade and not on the industry’s estimates. Illicit trade in tobacco products continues to be a major concern for tax administrators because of the difficulties associated with accurate and independent measurement of it, as well as with its elimination. Industry figures provide a distorted understanding of the extent of the problem, along with a monocausal explanation of the link between illicit trade and tobacco taxation. It is therefore recommended that governments (1) assess independently and with the best statistical practices the size of the illicit trade to assess the scope of the problem; (2) address directly the country-specific institutional and/or governance challenges, including multilateral coordination, and improve tax and customs administrations practices; and (3) implement best practices to fight illicit trade, contained in the WHO FCTC Protocol to Eliminate Illicit Trade in Tobacco Products. Ideally, accede to the Protocol if not yet a Party. C: Court and legal challenges Do not let tobacco industry threats of court and legal challenges to tax increases or reforms prevent you from improving your tax policy. Closely follow legal requirements for design, procedure and consultation to strengthen your legal position and minimize the possibility that any challenge will be raised. Health-protective and non-discriminatory tobacco excise taxes are legally defensible, and industry threats will usually be baseless. Your legal position can be strengthened, however, by exercising care with a tax measure’s procedure, design and consultation: (1) determine the standard of consultation required under domestic law and any applicable international obligations; (2) distance the tobacco industry from the policy-making process to the extent that this is permissible; (3) avoid unnecessary and unjustified discrimination towards foreign tobacco products or investors in the design, implementation or enforcement of a tax measure; and (4) do not offer investment incentives in the form of inducements or contractual undertakings, as these may be binding in and of themselves or grounds for a challenge under an international investment agreement. 280 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N A: Anti-poor rhetoric Do not allow concerns about the regressivity of higher tobacco taxes prevent tobacco tax increases. In recent years, there has been an overwhelming increase in the evidence concern- ing the positive distributional impact of tobacco taxes and tax increases. Indeed, tobacco taxation and tax increases are actually a progressive or pro-poor policy once these wider considerations are properly accounted for. In its effort to lobby against tax increases, the tobacco industry often claims that tobacco taxation will hurt the poor. This argument is based on the concept of regressivity in relation to taxation. Conceptually, a tax is regressive if it means lower-income people must pay a relatively greater proportion of their household income to meet the tax liability than wealthy people. However, there are two limitations to the industry’s argument. First, the concept of regressivity based solely on tax burden does not consider the wider health and economic harms caused by tobacco use that are disproportionately experienced by lower socioeconomic groups. Second, higher tobacco taxes and prices can induce behavioural change in the population, as reflected in the price elasticity of demand, which means that lower-income smokers will curtail their smoking the most and thus will benefit disproportionately in terms of health gains from reduced tobacco consumption and use. In fact, these broader considerations make tobacco taxation a progressive, rather than regressive, public health intervention. R: Revenue reduction Do not let fears of potential revenue reductions prevent you from increasing excise taxes on tobacco products. Tax increases, even in countries with already high taxes, bring in additional revenue. Arguments by tobacco control opponents that tax increases will not result in increases in revenue are unfounded. The relatively price inelastic nature of cigarette demand, combined with the low tax share and no overshifting of the tax, means that for most, if not all, countries, increases in revenues will accompany increases in taxes. If tax increases are carefully designed and tax administration is functional, it is extremely unlikely that tax increases will lead to revenue decreases. E: Employment impact Do not allow concerns about employment impact to prevent tobacco tax increases. The tobacco industry often seeks to frame tobacco taxes as an economic issue rather than a public health issue. Particular emphasis is placed on the alleged threat tax increases pose to employment in tobacco farming and manufacturing, as well as related industries. This so-called choice between health and jobs, however, is largely based on exaggeration. The tobacco industry exaggerates the importance of tobacco CHAP T ER 5. B E S T PR AC T I CE S 281 employment relative to total national employment and overstates the impact that domestic demand reduction from local taxes will have on tobacco farmers serving a global market. The argument used by the industry also ignores the fact that expendi- tures on tobacco do not disappear but rather are redistributed to other consumption that can produce a similar or higher number of jobs. Case studies demonstrate the possibility and methods for governments to support farmers in transitioning to other crops that provide similar and often better returns with greater sustainability. Earmarking Consider earmarking tobacco tax revenues for health-focused programmes, especially if it helps advance tobacco control efforts and, more specifically, efforts to implement large tobacco tax increases and tax reforms. This could have the additional benefit of funding health programmes where they are poorly funded or not prioritized. From a tobacco control perspective, tobacco tax earmarking is best understood as a way of selling significant tobacco tax increases to the public, politicians and officials. Earmarking is a tool to improve the political economy of tobacco taxation; it is only a secondary issue, after the primary goal of reducing demand for tobacco. Evidence shows that public support for higher tobacco taxes is greater when at least some of the increased revenues are explicitly used to support health-focused programmes. Current evidence shows that the amounts effectively earmarked for health have been relatively small and therefore unlikely to introduce rigidity in government budgets. At the same time, in some countries, those funds have helped to imple- ment much needed underresourced health programmes. The payoffs will be seen in the future as fewer people fall ill and need less medical care for tobacco-related illnesses. Earmarking tobacco tax revenues to fund high-burden/low-priority health programmes could pave the way for raising awareness about the importance of such programmes and their effectiveness, thereby convincing governments to redefine their priorities and commit to including the programmes in their regular budget. 282 W H O T ECHNI C AL M ANUAL O N TO BACCO TA X PO LI C Y AND ADM INIS T R AT I O N TOBACCO TAX REFORM CHECKLIST (FOR TAX POLICY-MAKERS) Focus on tobacco taxation’s purposes Tobacco tax policy should aim not only to increase revenues but also to decrease consumption and improve health. To both raise revenue and reduce consumption, you need to (1) simplify tobacco tax structures, (2) significantly increase rates to impact price levels, and (3) regularly adjust rates to at least account for inflation and income growth. Analyse your tax structure and identify its weaknesses You need to analyse and identify the problems of your current tax structure to know which steps to consider next. Which tax structure do you have: specific, ad valorem, mixed, or no excise? Identify the improvements to be made to the existing tax policy/structure Your present tax structure and tax situation will provide you with the steps you would ideally take next to achieve the aims in Step 1. Specific: 1. Ensure that the tax automatically adjusts for inflation and income growth effects. 2. Ensure that all price promotions are banned. Ad Valorem: 1. Ensure that the tax base of the ad valorem is retail price. 2. Introduce a high specific excise component (and a minimum specific excise). 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. Mixed: 1. Ensure that the tax base of the ad valorem component is retail price. 2. Ensure that you are using a high specific excise component and a minimum specific excise. 3. Ensure that the specific excise and/or the minimum specific excise automatically adjusts for inflation and income growth effects. 4. Ensure that all price promotions are banned. No Excise: 1. Introduce a high specific excise. 2. Ensure that the rate automatically adjusts for inflation and income growth effects. 3. Ensure that all price promotions are banned. Assess tobacco taxation’s political economy Reform must begin with an assessment of tobacco taxation’s political economy: (1) learn from past successes and failures – what went wrong, what went right, what you can do differently this time; (2) assess the reform’s strengths and weaknesses, likely opportunities and risks; (3) determine who the main supporters and opponents of reform inside and outside of government have been and may be, based on past reforms and current situation; and (4) anticipate argu- ments that will be used against the reform (refer to SCARE tactics). STEP 1 STEP 2 STEP 3 STEP 4 Prepare a plan for realizing the reform Focusing on the overall aims identified in Step 1, the steps for achiev- ing them identified in Steps 2 and 3 and the political economy around this reform as identified in Step 4, prepare your plan: 1. Be clear on the non-negotiable objectives for the reform and the trade-offs you are prepared to make to realize them. 2. Develop a plan to approach potential allies and win them over to the reform efforts. 3. Develop the counterarguments that will be needed in response to the SCARE arguments identified earlier. 4. Prepare the evidence you will need ahead of time. To do this, get support from academics and relevant intergovernmental agencies. Mobilize a coalition for reform 1. Formulate a strategic communications plan: aim for political support both at the highest levels and among the public (framing tobacco taxation as a health issue has helped win political support in many countries). 2. Identify champions in government: ensure that finance and health officials are on the same page; involve implementing departments, such as enforcement agencies, from the start. 3. Mobilize allies from academia, civil society and the private sector to counter the anticipated pushback from the tobacco industry, its proxies and its allies. Monitor and evaluate To make the most well-informed policy decisions, a reform effort should be monitored to assess its overall impact and its effect on key indicators; this will help identify issues to be fixed while also creating a strong evidence base for further reform efforts. Get and analyse the relevant data to better understand the market situation and its dynamics: 1. Monitor the market and its evolution (e.g. retail prices, duty-paid sales, market shares). 2. Get regular estimates of price elasticity (including cross-price elasticity), income elasticity and tax base elasticity to evaluate any changes in tobacco demand. Use relevant tools to assess the impact of the tax policy on consumption and revenue: 1. Use specific tools on the impact of excise on price, consumption and revenue (e.g. the WHO TaXSiM). 2. Use global tools to assess the tax increase’s impact on prevalence (e.g. the WHO ISPT). Monitor key indicators closely to assess improvements over time: 1. Tax as a percentage of retail price. 2. Change in affordability of tobacco products over time. 3. Change in the tobacco tax scorecard, which combines a mix of best practices in tax policy. 4. Change in sales, prevalence and illicit trade in tobacco products. 5. Improvements in MPOWER package achievement. STEP 5 STEP 6 STEP 7

图书在版编目(C I P)数据 世界卫生组织烟草税政策和管理技术手册 /郑榕译. 合肥: 安徽科学技术出版社,2023.6 ISBN 978-7-5337-8701-1 Ⅰ.①世… Ⅱ.①郑… Ⅲ.①世界卫生组织-烟草 工业-税收管理-技术手册 Ⅳ.①F811.4-62 中国版本图书馆CIP数据核字(2022)第254917号 SHIJI EWEISHENG ZUZHI YANCAO SHUI ZHENGCE HE GUANLI JISHU SHOUCE 世 界 卫 生 组 织 烟 草 税 政 策 和 管 理 技 术 手 册 郑 榕 译 ………………………………………………………………………………………… 出 版 人: 丁凌云 选题策划: 王 勇 责任编辑: 王 勇 责任校对: 戚革惠 责任印制: 梁东兵 装帧设计: 冯 劲 出版发行: 安徽科学技术出版社 http://www.ahstp.net (合肥市政务文化新区翡翠路1118号出版传媒广场,邮编:230071) 电话: (0551)63533330 印 制: 合肥创新印务有限公司 电话:(0551)64321190 (如发现印装质量问题,影响阅读,请与印刷厂商联系调换) ………………………………………………………………………………………… 开本: 787×1092 1/16 印张: 14.5 字数: 290千 版次: 2023年6月第1版 印次: 2023年6月第1次印刷 ………………………………………………………………………………………… ISBN 978-7-5337-8701-1 定价: 45.00元 版权所有,侵权必究 对外经济贸易大学国际贸易经济学院 (SITE) ,世界卫生组织烟草控制与经济政 策合作中心,2023。 此翻译并非由世界卫生组织 (WHO) 完成。世界卫生组织不对本翻译的内容或准 确性负 责。世 界 卫 生 组 织 [WHO Technical Manual on Tobacco Tax Policy and Administration Geneva: World Health Organization, 2021]英文原版许可证:CC BY- NC-SA 3.0 IGO 应为具有约束力的正版版本。此翻译作品可在 CC BY-NC-SA 3.0 许可 下使用。 前 言 1999 年,世界银行发布了《遏制烟草流行》报告,在这个报告中,国际组织首次提出提 高烟草消费税是减少烟草使用和挽救生命的最有效和最具成效的措施。此后的二十年 来,支持这一主张的证据一直在稳步增长,尤其是在低收入和中等收入国家。与此同时, 烟草业反对烟草税的论点的可信度一直在慢慢减弱。简而言之,促进健康的烟草税政策 环境日臻成熟,越来越多的证据表明提高烟草税是一项促进公共卫生、增加政府税收和促 进经济整体发展的“三赢”策略。 但我们也需要保持清醒的头脑。尽管关于烟草税的证据无可辩驳,而且现在有迹象 表明全球烟草流行趋势正在得到遏制,但在2018年世卫组织关于各国 MPOWER 政策 措施的评估中,全面实施烟草税政策措施却是所有措施中得到实施最少的措施。更令人 担忧的是,在过去十年中,许多低收入和中等收入国家的卷烟的可负担性变得更强了,即 卷烟的价格随时间推移变得更便宜了。许多国家将烟草税率定得太低,而且很少提高税 率。另有一些国家仍在使用复杂且效率低下的烟草税制,未能推进能够带来卷烟价格大 幅上涨的烟草税,不仅导致了政府税收收入的损失和公共卫生的损失,还促成了烟草业的 “胜利”局面。 为了改观这一局面,本手册旨在为政策制定者、财政官员和其他参与烟草税收政策制 定的人员指明前进的道路,为他们提供实现其国家烟草税政策目标所需要的信息和证据。 本手册还分析了烟草业抵制烟草税收政策和一国政治经济的惯用策略,揭示了烟草业反 对加税的论点的局限性和夸大性。本手册是对2010年出版的《世界卫生组织烟草税管理 技术手册》的更新,并增加了世界各国烟草税收成功的新证据,并扩大其范围,以获取更多 与制定和实施更有效的烟草税收政策相关的材料。 在当前受新型冠状病毒袭击的世界中,本手册的贡献尤其及时。新冠肺炎疫情的大 流行加重了全球非传染性疾病的负担,使得各国政府比以往任何时候都更迫切需要税收 收入,因此对烟草征税比以往任何时候都更受欢迎。本手册向政策制定者展示了如何抓 住当前这一特殊时机,利用征收烟草税来更好地重建和加强卫生系统以挽救生命,并增加 政府的税收收入。 Naoko Yamamoto 博士 全民健康覆盖/健康人口司 助理总干事 世卫组织 MPOWER 一揽子技术措施包括:(M)监测烟草使用和预防政策;(P)保护人们远离烟草烟雾;(O)提供 戒烟帮助;(W)警告烟草危害;(E)禁止烟草广告、促销和赞助;(R)提高烟草税。 致 谢 这本手册是在Jeremias Paul Jr和Anne-Marie Perucic的指导下由以下人士编写的, 他们是(按字母顺序): • 世卫组织:Evan Blecher、Annerie Bouw、Mark Goodchild、Roberto Iglesias、 Juliette McHardy、Jeremias Paul Jr、Anne-Marie Perucic和Robert Totanes。 • 外部作者:Mauricio Cardenas(哥伦比亚大学)、Sophia Delipala(希腊马其顿大 学)、LukJoossens(比利时烟草控制专家)、Marin Kurti(东康涅狄格州立大学)、Enrique Fanta(前世界银行高级专家)、David Merriman(伊利诺伊大学芝加哥分校)和Jean Tesche(开普敦大学)。 世卫组织感谢以下审查员提供的宝贵意见(按字母顺序): • 外部:Jo Birckmayer(彭博慈善基金会)、Adriana Blanco Mar-quizo(世卫组织烟草 控制框架公约)、Frank Chaloupka(伊利诺伊大学芝加哥分校)、Yoni Dekker(世卫组织烟 草控制框架公约)、Jeffrey Drope(伊利诺伊大学芝加哥分校)、Ceren Ozer(世界银行)、 cornne van Walbee k(开普敦大学)、ChonlathanVisaruthvong(泰国财政部)和 Rodrigo Santos Feijo(世卫组织烟草控制框架公约)。 • 世卫组织: 总 部:DouglasBettcher,ItziarBelausteguigoitia,RantiFayokun,JosephKutzin, BennMcGrady,VinayakPrasad和SusanSparkes. 区域办事处:Nina Dela Cruz(WPRO),Fatimah El-Awa(EMRO),Charles Frasier (EMRO),Lee Lily Joung-Eun(WPRO),Jagdish Kaur(SEARO),Elizaveta Lebedeva (EURO),Maxime Roche(AMRO) andRosaSandoval(AMRO). 世卫组织还要感谢 Amalamoun-naal的行政支持、Alison Goldstein的技术编辑和 Janet DeLand的文字编辑的辛勤付出。 本手册的编写及出版得到了彭博慈善基金会的资助。手册的内容由世卫组织负全 责,不应被视为代表了彭博慈善基金会的立场。 缩写 AEO authorized economic operator 经授权的经济经营者 AFRO WHO Regional Office for Africa 世界卫生组织非洲区办事处 AMRO WHO Regional Office for the Americas 世界卫生组织非洲区办事处 ATO Australian Taxation Office 澳大利亚税务办公室 BAT British American Tobacco 英美烟草集团 CCTV closed-circuit television 闭路电视 CIF cost, insurance and freight 到岸价(成本、运费加保险费价) COP Conference of the Parties 缔约方会议 CPI consumer price index 消费者物价指数 CVA Customs Valuation Agreement (Thailand) 海关估价协议(泰国) DGCE Directorate General of Customs and Excise(Indonesia) 海关总局(印度 尼西亚) DIY do-it-yourself 自制 ECBA World Bank Extended Cost-Benefit Analysis 世界银行扩展成本效益 分析 EIU Economist Inteligence Unit 经济学人智库 EMRO WHO Regional Office for the Eastern Mediterranean 世界卫生组织东 地中海办事处 ENDS electronic nicotine delivery systems 电子尼古丁输送系统 ENNDS electronic non-nicotine delivery systems 电子非尼古丁输送系统 EU European Union 欧洲联盟 EURO WHO Regional Office for Europe 世界卫生组织欧洲区域办事处 FDA Food and Drug Administration(United States) 食品和药物管理局 (美国) FET fair and equitable treatment 公正与公平待遇 GCC Cooperation Council for the Arab States of the Gulf 海湾阿拉伯国家合 作委员会 GDP gross domestic product 国内生产总值 HTP heated tobacco product 加热不燃烧烟草制品 HMRC Her Majesty􀆳s Revenue and Customs(United Kingdom) 英国税务及 海关 IARC International Agency for Research on Cancer 国际癌症研究总署 IIA international investment agreement 国际投资协定 IMF International Monetary Fund 国际货币基金组织 IRS Internal Revenue Service(United States) 美国国税局 ISO International Organization for Standardization 国际标准化组织 2 世界卫生组织烟草税政策和管理技术手册 IT information technology 信息技术 ITC International Tobacco Control 国际烟草控制 JTI Japan Tobacco International 日本烟草国际公司 KRA Kenya Revenue Authority 肯尼亚税务总局 LMICs low-and middle-income countries 中低收入国家 MFN most favoured nation 最惠国 MOP Meeting of the Parties (to the Protocol) (议定书)缔约方会议 MPOWER(M)monitor tobacco use and prevention policies; (P) protect people from tobacco smoke; (O) offer help to quit tobacco use; (W) warn about the dangers of tobacco; (E) enforce bans on tobacco advertising, promotion and sponsorship; and (R) raise taxes on tobacco (M) 监测烟草使用和预防政策;(P)保护人们远离烟草烟雾;(O)提供戒 烟帮助;(W)警告烟草危害;(E)禁止烟草广告、促销和赞助;(R)提高烟 草税 NCDs noncommunicable diseases 非传染性疾病 NCI National Cancer Institute 美国国家癌症研究所 NT national treatment 国民待遇 OECD Organisation for Economic Co-operation and Development 经济合作与 发展组织 OST other smoking tobacco 其他吸入性烟草 PMI Philip Morris International 菲利普·莫里斯国际 PPP purchasing power parity 购买力平价 QR quick response 快速响应 RGTE WHO Report on the global tobacco epidemic 世界卫生组织全球烟草流 行报告 RYO rol-your-own 自制手卷烟 SACU Southern African Customs Union 南部非洲关税联盟 SCARE(S) smuggling and ilicit trade; (C) court and legal chalenges; (A) anti- poor rhetoric; (R) revenue reduction; and (E) employment impact (S) 走私和非法贸易;(C)法庭和法律挑战;(A)反贫困言论;(R)减少收入; (E)就业影响 SDGs Sustainable Development Goals 可持续发展目标 SEARO WHO South-East Asia Regional Office 世界卫生组织东南亚区办事处 SII Internal Revenue Service of Chile 智利国税局 TADAT Tax Administration Diagnostic Assessment Tool 税务管理诊断评估 工具 TTC transnational tobacco company 跨国烟草公司 UAE United Arab Emirates 阿拉伯联合酋长国 3致谢 UHC universal health coverage 全民健康保险 VAT value added tax 增值税 WAEMU West African Economic and Monetary Union 西非经济和货币联盟 WCO World Customs Organization 世界海关组织 WHO World Health Organization 世界卫生组织 WHO FCTC WHO Framework Convention on Tobacco Control 世界卫生组 织烟草控制框架公约 WHO ISPT WHO interactive smoking projection and target-setting tool 世卫 组织的交互式吸烟预测和目标设定工具 WHO TaXSiM WHO tobacco tax simulation model 世界卫生组织烟草税模拟 模型 WPRO WHO Regional Office for the Western Pacific 世卫组织西太平洋区域 办事处 WTO World Trade Organization 世界贸易 4 世界卫生组织烟草税政策和管理技术手册 执 行 摘 要 《世界卫生组织烟草税政策和管理技术手册》以2010年出版的《世界卫生组织烟草税 管理技术手册》为基础,进一步详细说明了设计、制定有效的烟草税收政策、实施和管理的 战略。本版作为2010年版手册的修订版,纳入了科学、技术和政策的最新发展现状,并提 供了来自多个国家的最新例证。本手册中列出的最佳做法旨在向各国政府通报他国烟草 税收政策的制定情况,促进实现其健康目标和收入目标,同时支持其总体发展战略。 烟草税长期以来一直被视为政府的收入来源,但随着多年来关于烟草造成危害的证 据不断积累,公众的看法也在改变。政府和公众日益认识到,对烟草征税不仅是一项收入 来源,而且是一项有效的公共卫生干预措施,能减少烟草消费带来的相关危害。 自2010年《世界卫生组织烟草税管理技术手册》出版以来,以征收烟草税作为一种卫 生政策工具的做法得到大幅加强。在过去十年中通过了多项全球承诺:运用税收和价格 措施来减少烟草制品需求,拯救生命和利用专项发展资金的方式解决烟草的具体使用问 题等。该承诺也包括更广泛的非传染性疾病(NCDs)和可持续发展目标(SDGs)。包括世 界银行、国际货币基金组织(IMF)和主要慈善基金会在内的多个全球发展机构也认可世 界卫生组织关于强调和加强烟草税作为一项关键卫生政策工具的重要性的观点。 新型冠状病毒疾病的暴发揭示了全球经济如何与人类健康密不可分,从而进一步推 动了烟草税重要地位的确立。投资健康是任何经济复苏的基础,财政政策将成为解决新 型冠状病毒造成的社会经济后果的关键驱动因素。烟草税等干预措施将会减少烟草消 费、改善人类健康和增加政府收入,一定会成为后疫情时代关于经济复苏的综合战略之一。 可以明确的是,显著增加消费税以提高烟草价格,是减少烟草消费的最有效,也是最 具成本效益的机制。本手册将指导读者采取必要步骤,为本国家制定和实施最强有力的 烟草税收政策提供帮助。 在制定烟草税收政策时需要考虑许多因素。在决定使用哪种税收形式之前,了解市 场是一个基本步骤。具体而言,从价税和从量税之间的选择受特定国家市场结构的影响。 同时,税收结构也影响着市场结构。在过去十年中,有一种趋势是各国采用特定消费税或 混合消费税制度,这些制度更加依赖特定市场成分。而最新的全球数据表明,这些特定市 场成分与最高平均价格指数相关联。 烟草税制结构可以是简单的、一刀切的统一税率,也可以是复杂的、针对不同类型的 产品实行的多层税率。2018年,31个国家使用了复杂的多层税率结构。但有证据表明, 所有高收入国家都采用了更简单的消费税结构,这使得烟草行业进行行业操纵、避税和品 牌或产品转换的空间变得更小。 重要的不仅是要制定较高水平的税率以抑制消费,对于从量税而言,还要定期调整税 率以跟上一个国家一段时间内的通货膨胀和收入增长。消费税的增加旨在降低烟草制品 的负担能力。计税方法也很重要:对于从量税,税基应以明确定义的单位表示的数量;对 于从价税(或混合税),最佳做法是使用零售价作为税基并引入最低消费税。 对于影响烟草制品价格的非税收政策,可以考虑采用定价法规以防止烟草行业实行 税负转移的差别策略,以确保优质卷烟和廉价卷烟之间存在巨大的价格差距。但是,价格 政策不能单独使用,只能作为大幅增加消费税的补充。 其他非税收法规包括禁止烟草制品的促销、打折和禁止销售单支卷烟。为了缓解相 关部门对增税会增加通货膨胀的担忧,以及更好地反映烟草制品消费下降的趋势,应将烟 草制品从制定消费者价格指数的商品篮子中剔除。最后,为了使烟草制品的消费税能更 有效地减少烟草使用,所有烟草制品都必须以类似的方式征税。 定期评估、评价和监测烟草税收政策的长期影响是制定有效税收政策和进行政策分 析的重要保障措施。政府需要准确估计价格、收入和税基弹性,以便预测增税对消费和税 收的影响。理想情况下,在估计特定国家的价格和收入弹性时,还应考虑其他因素,例如 非价格政策因素。对此有多种工具和指标可以用来衡量税收政策的影响和监测其进展, 本手册第2章对这些工具和指标进行了介绍。 在制定烟草税政策时,还必须在国内和地区两级考虑更广泛的政策背景。在国内,需 要跨部门合作,以确保农业、贸易、金融和劳工领域的政策和干预措施不违背烟草控制和 税收的公共卫生目标。对于属于区域集团的国家来说,烟草税的区域协调是防止税收流 失、避税和逃税以及保护人口健康的有力工具。然而,税收协调必须仔细设计才能有效。 欧盟(EU) 的经验表明,统一的最低消费税税率可以实现消费下降趋势和维持收入稳定 的目标。 由于不断变化的技术和市场动态,有关对于新出现的尼古丁和烟草制品[如加热型烟 草制品(HTPs)以及电子尼古丁和非尼古丁输送系统(ENDS/ENNDS)]的政策制定和实 施的讨论变得十分复杂,对这些产品的政策和法规需要仔细制定并适时调整。 在加热型烟草制品(HTP)尚未被禁止的情况下,当前对这类产品征税的政策建议是 按每单位对它们征收与卷烟相同水平的税,而不论其烟草含量的多少。 来自美国的早期证据表明,对电子烟(ENDS/ENNDS产品的一个子类)的需求相较 于传统卷烟可能对价格更为敏感,这意味着税收可以作为遏制ENDS/ENNDS产品使用 的有效政策手段。虽然有初步证据表明传统卷烟和电子烟的使用具有可替代性,但需要 进一步研究传统卷烟和 ENDS/ENNDS产品之间的替代效应,以保障公众健康,对 ENDS/ENNDS产品实施监管并采用税收政策遏制其流行至关重要。 在电子烟未被禁止的国家,必须对 ENDS/ENNDS 产品进行监管并征税,以阻止年 轻人和不吸烟的人尝试使用该类产品。对电子烟的烟油征税是制定对ENDS/ENNDS 征税政策的关键组成部分,即含尼古丁的烟油和不含尼古丁的烟油应同等征税。最终,虽 然对这些新型烟草制品的政策影响尚需仔细考虑,但事实仍然是传统烟草制品占据了烟 草消费的压倒性市场份额(2018年超过97%)。 烟草税收管理必须既高效又有效,以确保实现大众健康目标并确保国家税收收入。 由于烟草税的实施通常涉及一个国家内的众多的部门和机构,因此明确各部门和机构的 职责对于最大限度地提高效率至关重要。要使烟草税收管理有效,需要在所涉及的不同 机构之间以及与邻国之间进行协作。主管机关的绩效评估和问责制也是必要的,在第 3 2 世界卫生组织烟草税政策和管理技术手册 章中描述了许多相应的工具和指标,主管机关应采取多项措施来确保在一个税务遵从周 期的每个阶段的效率和效果(见第 3 章第 3.3 节)。 控制和执行是税务管理机构的主要职能,为此,职能部门最好通过制定战略计划和加 强风险控制来实现。可以通过许可证发放和尽职调查、财政标记(如税票)、跟踪和溯源、 实施反垄断措施、国家审计和对进出口以及自由区和转运点的具体管制来实施监管。一 旦发现走私或违法贸易,必须立即采取行动,如扣押和销毁走私和(或)非法烟草制品,征 收应缴税款。为了阻止进一步的违法活动,还必须对所有参与违法行为的人员进行全面 管控,并施以严厉的处罚和制裁以阻止违法活动。《消除烟草制品违法贸易议定书》为烟 草税的管理、控制和执法提供了宝贵的指导,也适用于非缔约的国家。 合理税收制度的更广泛要素还包括:主管机关的适当资源配置、侦查和惩治腐败的严 格规则和条例,以及能够尽快解决争端的强大司法系统。 在反对增加烟草税的措施中,烟草行业运用了许多恐吓策略来影响烟草的政治经济 地位,具体包括:(S)走私和非法贸易;(C)法庭和法律挑战;(A)反贫困言论;(R)收入减 少;(E)就业影响。第4章对这些问题进行了详细分析,并提供了与恐吓策略不符的证 据,以及税务和其他相关部门如何预测和应对烟草行业反对措施的指南。 本手册还提供了工具和方法,以帮助税务机关定义和评估本国烟草制品的非法贸易 问题,评估结果独立于烟草业普遍夸大的估计数。价格(和税收)水平不是导致非法贸易 的决定因素;相反,缺乏治理和税收管理效率才是使问题更加严重的原因。避免增税不是 解决办法。相反,各国政府应采取全面的综合策略来打击烟草非法贸易,包括对非法贸易 水平进行独立估算,并实施合理的税收管理政策。例如第3章中所讨论的措施,以及《世 界卫生组织烟草控制框架公约(WHO FCTC)》非法贸易议定书中所包含的措施,以消除 烟草制品的非法贸易。 当涉及法庭质疑时,与其他烟草控制措施相比,烟草业不太可能挑战消费税,因为税 收是一项相对完善的监管措施。尽管如此,烟草行业仍将极尽可能地挖掘税制设计、采用 或实施中的漏洞。因此,本手册描述了加强监管机构法律地位的措施,这些措施将使监管 机构能够保护自己免受潜在的法律挑战。 烟草业主张烟草税具有累退性,或者说增加烟草税伤害了穷人,因为增加烟草税使穷 人比富人缴纳占其收入更大比例的税收。烟草业关于税收累退性的论点存在两个基本的 局限:首先,累退性的概念没有考虑到烟草使用所造成的更广泛的健康和经济危害,这些 危害加剧了低收入吸烟者的贫困。当烟草消费在增税后减少时,这些危害实际上就减少 了。其次,烟草业的论点忽视了一个事实,即较高的烟草税和价格会导致消费行为的改 变,这反映在价格需求弹性上。证据始终表明,低收入烟民对价格更为敏感,因此也更有 可能因税收和价格上涨而减少吸烟。以上这些因素表明,事实上,烟草税作为一种累进的 公共卫生干预措施,对穷人而言利远大于弊。 虽然烟草业基本上承认增加烟草税可能会带来减少消费的预期效果,但该行业也试 图辩称提高烟草税也会减少政府的税收收入。事实上,烟草价格的非弹性需求使得烟草 税的增加在公共卫生和财政两方面是双赢的。本手册介绍了几个国家的例子,说明了精 心设计并良好实施的烟草税增加如何在中短期内增加了政府的收入而不是减少收入。此 3执行摘要 外,由于增税而导致的消费减少也导致与烟草有关的其他政府支出减少。 烟草业用来反对增税的最后一种策略是将烟草税视为经济问题而不是公共卫生问 题。这种在健康和就业之间错误的定位是基于错误的假设:①烟草业是国内就业的重要 来源;②就业岗位依赖于烟草消费;③与烟草相关的就业岗位持续繁荣且不可替代。 将烟草税专款专用可以用来改善增加烟草税的政治经济环境。虽然增加烟草税的主 要目的是减少社会对烟草的需求,但留出一部分税收用于资助其他烟草控制工作或相关 的健康计划,有助于说服公众、政治家和官员相信大幅增加烟草税这一政策的价值所在。 也可以是专款专用于反驳烟草业关于增税潜在负面影响的论点,例如,支持烟农从种植烟 草转移到种植其他农作物。 烟草税的作用有目共睹,这也是为什么烟草行业投入如此多的资金和精力来阻止大 幅增税和其他有效的税收政策改革的原因。政策制定者不应被烟草行业的施压所左右, 而只需遵循事实。本手册提供了决策者在流程的每一步做出正确决策所需的所有信 息———从如何设计、评估、实施和管理税收政策,到怎样反驳似是而非的行业攻击,以及向 立法者和广大民众宣传烟草税的价值的各种方法。有效制定并有效实施的烟草税政策在 减少烟草消费、增加政府收入的同时,还会带来更加广泛的影响。事实上,提高烟草税是 一项明智(Smart)的政策:它可以挽救生命(Save);调动社会资源(Mobilize);解决健康不 平等问题(Address);减轻卫生系统的负担(Reduce);减少烟草使用(Target)这一非传染 性疾病的主要风险因素。 • 减少拯救生命:烟草使用是全球可预防死亡的最主要手段。烟草每年夺走800万 人的生命。烟草税是全球减少烟草消费及其相关健康负担的最有效机制。 • 调动资源:尽管烟草税是目前最有效的烟草控制措施,但作为一种政策机制,烟草 税在很大程度上并没有得到充分利用。根据有关卷烟价格和税收的现有数据,2018年卷 烟消费税在全世界产生了3610亿美元的收入,其中包括低收入和中等收入国家(LMIC) 1620亿美元的收入。如果所有国家都将卷烟消费税税率提高到相当于每包1美元的水 平,那么卷烟消费税收入将增加1780亿~2190亿美元,或者在2018年的水平上增加 49%~61%。低收入和中等收入国家将从此类增税中获益最多,消费税收入将增长 82%~103%,为这些国家的政府提供额外的1330亿~1670亿美元收入。这表明烟草 税额具有巨大的增收潜力。 • 解决健康不平等问题:烟草税和增值税是有效的累进或惠贫政策,因为它们具有 积极的分配影响。低收入吸烟者从减少烟草消费和使用中的受益取决于其健康改善和省 下的买烟钱。 • 减轻卫生系统的负担:2012年,全球烟草使用的经济成本为1.4万亿美元。烟草 税减轻了政府和卫生系统由于烟草使用而带来的相关负担。 • 遏制烟草使用:烟草税直接遏制了烟草使用,减少了烟草消费。烟草制品是导致 几种致命的非传染性疾病的主要风险因素。 总之,根据本手册中提供的最新指南和最佳做法设计和实施的大幅增加烟草税的方 案,是综合烟草控制策略的重要组成部分,将大幅减少烟草使用以及烟草制品对健康和经 济造成的危害。 4 世界卫生组织烟草税政策和管理技术手册 目 录 前言 v…………………………………………………………………………………………… 致谢 vi………………………………………………………………………………………… 缩写 vii………………………………………………………………………………………… 执行摘要 xi…………………………………………………………………………………… 第一章 为什么制作本手册 1………………………………………………………………… 1.1 背景………………………………………………………………………………… 1.2 烟草使用的危害和政策干预的必要性…………………………………………… 1.3 为什么烟草税(消费税)至关重要………………………………………………… 1.4 十年的进步和行动承诺…………………………………………………………… 1.5 塑造烟草税的“新常态”…………………………………………………………… 1.6 各章节主要内容概述……………………………………………………………… 第二章 烟草消费税政策 11………………………………………………………………… 2.1 烟草税全球概况 11………………………………………………………………… 2.2 消费税政策设计 18………………………………………………………………… 2.3 国内和区域政策一体化 55………………………………………………………… 2.4 新型尼古丁和烟草制品 59………………………………………………………… 2.5 总结 75……………………………………………………………………………… 附录2.1 2018年采用不同类型圈烟消费税结构的国家 89………………………… 附录2.2 税基弹性的分析 90………………………………………………………… 附录2.3 ENDS/ENNDS产品设备的构成 91……………………………………… 第三章 烟草税的管理 93…………………………………………………………………… 3.1 简介 93……………………………………………………………………………… 3.2 制度安排 94………………………………………………………………………… 3.3 税收征管流程 102………………………………………………………………… 3.4 控制和执行 115…………………………………………………………………… 3.5 其他烟草制品的税收管理 152…………………………………………………… 3.6 良好税收制度的更广泛要素 155………………………………………………… 3.7 总结 156…………………………………………………………………………… 附录3.1 烟草制品组成 166…………………………………………………………… 附录3.2 买断和对策实例 172………………………………………………………… 第四章 烟草税收的政治经济学 175………………………………………………………… 4.1 威胁策略S:走私和非法贸易 176………………………………………………… 4.2 威胁策略C:法庭和法律挑战 198………………………………………………… 4.3 威胁策略A:反贫困言论(累退性) 212………………………………………… 4.4 威胁策略R:收入减少 217………………………………………………………… 4.5 威胁策略E:影响就业 228………………………………………………………… 4.6 将烟草税收用于资助卫生事业 233……………………………………………… 附录4.1 评估非法烟草贸易性质和规模的方法 253………………………………… 附录4.2 如何指定烟草税收入的用途 269…………………………………………… 第五章 烟草税收政策和管理的最佳实践 271……………………………………………… 5.1 税收政策…………………………………………………………………………… 5.2 税收设计…………………………………………………………………………… 5.3 监控和评估………………………………………………………………………… 5.4 税务管理…………………………………………………………………………… 5.5 烟草税改革清单(适用于税收政策制定者) 283………………………………… 2 世界卫生组织烟草税政策和管理技术手册 第一章 为什么制作本手册 1.1　背景 烟草税并不是个新事物。自有消费税(Excise tax)以来,世界各国政府实际上一直在 对烟草和烟草制品征收消费税。之所以如此是因为烟草不是必需品,对其征税容易,而其 需求相对缺乏弹性。这些特点以及烟草税所产生的巨额收入使烟草成为非常合适的征税 对象。随着多年来烟草危害的证据不断增多,公众对烟草税的看法也发生了变化。现在 烟草税不仅被视为一种政府收入来源,更重要的是,它被认为是减少烟草消费的有效公共 卫生干预措施。这一趋势反映了消费税最初存在的原因,即“寓禁于征”并减轻相关的负 面外部效应[12]。 许多政府将烟草税视为重要而稳定的收入来源,这也可以解释为什么政府在提议烟 草税改革时往往会有一定程度的犹豫。历史上,许多政府财政依赖烟草税收入,甚至根据 其收入需求调整对烟草制品的征税程度[3]。然而,一些国家开始认识到将征收高额烟草 税作为主要公共卫生政策工具的价值,而将烟草税筹措的收入视为次要考虑因素[4]。 反对提高烟草税或改善烟草税制结构的争论通常围绕经济影响:在众多的争论中,最 主要的争论包括烟草税改革将减少政府税收收入、减少工作岗位、增加非法贸易并损害当 地产业等。但证据一直表明,以上论点在绝大多数情况下是不正确的。尤其是烟草业,经 常将这些争论描述为公共卫生与经济之间的错误二分法,认为优先考虑健康是以牺牲经 济为代价的。事实上,研究和现实世界的经验均表明,提高烟草税不仅可以改善公共健 康,而且也会对一个国家的经济和发展产生积极的影响,能呈现出真正的双赢政策效 果[56]。 本手册是对2010年出版的《世界卫生组织烟草税管理技术手册》的更新,旨在帮助读 者更好地了解围绕烟草税及其实施过程中的各种问题。本书的主要目标受众包括政策制 定者、财政官员、税务机关、海关官员和其他参与制定和实施烟草税收政策的相关人员或 机构。本手册为烟草税收政策的设计提供了详细指南,并描述了如何有效管理烟草税以 最大限度地发挥其政策效果。手册还涉及对政治经济因素的考量,讨论了烟草税政策实 施之前和实施过程中需要克服的障碍。手册的总体目标是为烟草税政策制定和实施领域 的专业人士提供足够的信息,以帮助各国政府在其总体发展战略下,实现烟草税收政策的 健康目标和经济目标。 1.2　烟草使用的危害和政策干预的必要性 大多数人都知道吸烟和烟草使用对健康有害,但很少有人真正了解这种危害的程度。 烟草的流行在20世纪夺去了 1 亿多人的生命[7],最新估计现在每年有 800 万人死于烟 草使用和接触二手烟[8]。其中多达 80% 的死亡发生在低收入和中等收入国家 (LMIC)[6],这意味着发展中国家承担了大部分烟草使用带来的全球危害。烟草使用是 许多慢性病的主要致病因素,包括心脏病、癌症、糖尿病和慢性肺病———统称为非传染性 疾病(NCD)。非传染性疾病导致全球约1500万人过早死亡(年龄在30~69岁),导致人 们在最富生产力的青壮年时期死亡。作为可预防的死亡的主要原因,烟草使用仍然是我 们这个时代最重要的公共卫生挑战之一。 烟草使用还造成了巨大的经济、发展和社会成本提升,对家庭、社区和社会也产生了 严重破坏。2012 年,吸烟的年度经济成本估计为 1.4 万亿美元,相当于当年全球生产总 值(GDP) 的 1.8%[9]。这一数字很有可能会继续增加,因此烟草使用造成的巨大健康和 经济负担为政府干预和严格监管烟草制品市场提供了依据。关于烟草控制的考量不应局 限于对人类社会健康福祉的强烈诉求;还应着力于对烟草使用的市场失灵和负外部性的 纠正,尤其是当这种市场失灵和负外部性会显著影响一个国家的发展轨迹时。 越来越多的证据表明,烟草使用在20世纪造成了持久的破坏,这些令人信服的证据 促使全球对烟草使用做出了强烈反应,各国就《世界卫生组织烟草控制框架公约(WHO FCTC)》进行谈判。《世界卫生组织烟草控制框架公约》(以下简称《公约》)于2005年生 效,是世界卫生组织主持下的第一个全球公共卫生条约。为了促进各国落实公约,世卫组 织从《公约》条款中总结了一套减少需求措施的一揽子政策干预措施包[7]。这些干预措施 统称为 MPOWER,具体包括:(M)监测烟草使用和预防政策;(P)保护人们远离烟草烟雾 (无烟立法);(O)提供戒烟帮助(戒烟服务);(W)烟草警示包装(包括图形包装警示和简 单包装);(E)禁止烟草广告、促销和赞助;(R)提高烟草制品的税率。具体而言,根据《公 约》第6条,各缔约方一致认为,价格和税收措施是减少各类人群,特别是年轻人烟草消费 的有效和重要手段[10]。 烟草使用的广泛流行及其对健康和经济的持续损害为各国政府积极采取措施加以干 预以纠正市场失灵提供了决策的依据。鉴于烟草使用所摧残的生命数量庞大、增速惊人 的现状,有必要立即采取积极的烟草控制措施,即采用大多数国家已承诺实施并被证明可 有效减少烟草使用的措施。 2 世界卫生组织烟草税政策和管理技术手册 1.3　为什么烟草税（消费税）至关重要 在各种烟草控制干预措施中,提高烟草消费税被认为是减少消费的最有效且最具成 本效益的措施[6]。虽然其他干预措施也是综合烟草控制战略的重要组成部分,但显著增 加税收对减少烟草消费的直接影响是所有措施中最有效的。平均而言,使得价格上涨 10% 的增税会使高收入国家的烟草消费减少 4%,使中低收入国家的烟草消费减少 5%[6]。当烟草税得以大规模实施时,其作用和挽救生命的潜力将得到很好的体现。烟草 税与其他政策干预措施的不同之处在于,烟草税的作用随着时间的推移而增加和积累,即 使征税程度已经相对较高,也需要不断提高税率以保持和扩大其有效性。然而,这并不意 味着政府考虑烟草控制干预措施应仅关注税收。把税收作为综合烟草控制措施(如 MPOWER一揽子政策措施,各项措施既有独特针对性又互补)的有机组成部分来实施, 则烟草税的政策效果会更好。 在对烟草制品征收的不同税种中,消费税至关重要,因为消费税会同时提高烟草制品 的绝对价格和相对价格[6]。在考虑健康目标时,这一点尤为重要,因为决定消费减少的是 烟草制品价格上涨的幅度。消费税通常对特定的产品征收,通过显著提高这类产品相对 市场上其他产品的价格来减少对这类特定商品的消费。这一点与对大多数商品和服务征 收的增值税(VAT)或销售税不同。寻求通过增值税或销售税来提高烟草制品的价格将 无法提高烟草制品相对于其他商品的价格,因而是一种无效或低效率的方法。烟草制品 也被征收关税或进口税,但随着旨在贸易便利化的双边和区域协定的全球流行趋势的发 展,关税的影响正在减弱。关税的征收因国家而异,但总体而言,关税不适用于本国生产 的烟草制品。作为提高价格的工具,进口关税不能替代消费税,因为它不是专门为减少消 费而设计的。 当被视为一种公共卫生政策的工具时,烟草税具有很高的成本效益,因为它产生了显 著的影响,但实施起来相对简单[11]。实施烟草税的成本远低于实施临床非传染性疾病干 预措施(如癌症治疗或维持性药物)的成本,因为烟草税的实施对商品和人力资本的要求 不高[12]。此外,提高烟草税实际上也为政府带来了额外的收入。烟草税在预防或减少对 价格特别敏感的人群,尤其是青少年的消费方面也非常有效,如果烟草的税收和价格足够 高,青少年就不会尝试吸烟而成为一个终身烟民[13]。烟草税对穷人也同样有效,因为穷 人比富人更难以承担因吸烟导致疾病的巨额医疗支出。通过征收高额烟草税来阻止人们 吸烟或鼓励人们戒烟,为摆脱吸烟致病致贫的恶性循环提供了有效方法[6]。 1.4　十年的进步和行动承诺 自2010年《世界卫生组织烟草税管理技术手册》出版以来,烟草税作为一项重要的公 共卫生干预措施得以长足发展。《公约》缔约方会议通过了《公约》第6条执行指南,该条 侧重于减少烟草需求的价格和税收措施。同样在过去的10年间,联合国大会召开了3次 关于预防和控制非传染性疾病的高级别会议,世界卫生大会于2013年批准了《全球非传 3第一章 为什么制作本手册 染性疾病行动计划》,这些举措在全球范围做出了采取措施以预防和控制非传染性疾病的 坚定承诺,比如增加烟草税以保护人们的健康。 2030年可持续发展议程包含了17个可持续发展目标(SDGs),这17个可持续发展 目标描述了未来10年的全球发展战略。在这17个可持续发展目标中,有两个目标与烟 草控制高度相关:一是加强《公约》的实施(目标3.a);另一个是将非传染性疾病的过早死 亡率降低30%(目标3.4)。此外,旨在为可持续发展目标融资提供全球框架的亚的斯亚 贝巴行动纲领 也强调了烟草税收和价格措施是减少烟草消费和拯救生命的关键机制, 同时也会增加国内发展资源。具有程碑意义的《消除烟草制品非法贸易议定书》于2018 年生效。这些关键事件,连同烟草控制领域和更广泛的发展领域的若干成果文件和政策 宣言,使更多的决策者认识到了烟草税的重要性。 如后续章节所述,在过去10年中,许多国家通过对烟草税的合理设计和有力实施对 烟草制品征收高额税收,提供了堪称烟草税的最佳实践的范例[8,10]。例如,持续大幅增税 减少了巴西[14]、土耳其[15]和菲律宾[16]等低收入国家的烟草使用。高收入国家在这方面 也继续发挥领头羊的作用,2018年,38个高收入国家中的23个都被认定为烟草税足够 高[8]。然而,仍有许多后续工作要做。2019年世界卫生组织《关于全球烟草流行 (RGTE)的报告》显示,烟草税仍然是 MPOWER 一揽子政策措施中最未被充分使用的烟 草控制政策[8],只有 14% 的世界人口被足够高的烟草税所覆盖。 在建立烟草税的工具和证据基础方面也取得了重大进展。2011年出版的国际癌症 研究机构(IARC)癌症预防手册《烟草控制税收和价格政策的有效性》第14卷是对截至 2010年5月出版的关于税收和价格政策在减少烟草使用方面的有效性的文献的一个重 要回顾。美国国家癌症研究所(NCI)2016年出版的《世界卫生组织烟草和烟草控制经济 学》专著,详细介绍了多年来多个国家积累的证据,不仅侧重于税收政策和价格政策,还侧 重于烟草和烟草控制经济学的各个方面。来自中低收入国家的大量已发表研究提供了烟 草税在不同背景下的影响的全面图景。更新后的《全球非传染性疾病行动计划》附录3解 释了烟草税的成本效益[11],而《拯救生命、减少支出 全球非传染性疾病业务计划》则以 这项工作为基础,通过预期投资回报的美元数字建立在这项工作的基础上实施最合理的 烟草控制干预措施,包括税收措施[12]。 在过去的10年中,全球承诺采用税收政策、实施最佳烟草税做法的国家数量以及在 烟草税证据基础的扩大(特别是在低收入国家)方面都取得了重大的提高和进展。在这一 进程中,世界银行(WB)、国际货币基金组织(IMF)和许多其他多边机构等与世卫组织一 起就烟草税的重要性和改进并实施的必要性方面达成了一致[17,18]。尽管世界正朝着正 确的方向前进,但仍然存在相当大的挑战,为了到2030年实现可持续发展战略目标,需要 加快各项政策推进的速度。 4 世界卫生组织烟草税政策和管理技术手册 第三次发展筹资会议亚的斯亚贝巴行动议程。联合国第三次国际会议,2015 年 7 月 13 16 日,埃塞俄比亚亚的 斯亚贝巴(https://sustainabledevelopment.un.org/content/documents/2051AAAA_Outcome.pdf,2021 年 2 月 17 日访问)。 1.5　塑造烟草税的“新常态” 新冠病毒大流行引起了全球动荡,也使人们前所未有地关注世界各国政府如何为有 效应对危机所做的准备。这场危机冲击了许多国家脆弱的卫生防疫系统,并凸显了这些 国家控制病毒传播的力不从心。但也许最重要的是,这场病毒大流行让我们清楚地认识 到,经济、贸易、科技、政治和社会的许多其他方面与人类的健康是如此地相互依存又相互 关联。 很明显,个人的健康状况取决于自身的免疫力。患有非传染性疾病的人更容易因多 种基础性疾病导致健康状况很差,如果感染新冠病毒[19],也更容易成为重症患者。烟草 使用是患上非传染性疾病的主要风险因素,现有研究也表明,吸烟者患严重疾病和死于新 冠病毒[20]的风险更高。正如社会的不同方面是相互关联的,人们的健康、促进健康的环 境以及塑造这些环境的政府政策和法规也是如此。这也为我们持续阻止烟草等有害产品 的使用,并进一步改进烟草控制措施,特别是烟草税收政策提供了理论依据。 展望未来,对烟草税采取一切照旧的做法是不够的。应对疫情后这一新的现实并为 下一次病毒大流行做好准备,需要采取措施促进人类健康。与新冠肺炎疫情一样,未来的 任何疫情都可能加剧人类健康不平等,从而带来更多的经济不确定性,并给政府的财政能 力带来压力。提高烟草税等干预措施既保护了人们的健康,同时也创造了更多的收入和 经济利益,这在危机情况下变得尤其重要。鉴于这一背景和所涉及的利害关系,财政部门 和税务部门占据了独特而强大的有利地位并且要利用好,因为其不仅拯救国家生计,而且 拯救国民生命。 增加烟草税,是现阶段最有效的公共卫生政策工具之一,它的重要性怎么强调都不为 过。将烟草税仅仅视作为政府创造收入的传统做法在新常态中没有立足之地。人们不能 否认烟草流行的规模之大、纠正市场失灵的阶段性以及烟草税对健康和经济有利的压倒 性证据。围绕烟草税不断变化的必要性的积极讨论需要继续下去。烟草税不应孤立于政 府其他政策之外,而应被视为整体政策的一个重要组成部分,是实现我们改善所有人健康 的共同目标的重要组成部分。 1.6　各章节主要内容概述 本手册主要面向决策者、财务官员、税务机关和海关官员。它也可能对卫生部或其他 政府机构的官员以及在这一领域工作的非政府组织,包括烟草控制倡导者有参考价值。 为了展示提高烟草税方面的成功案例和经验教训,我们做出了重大努力,介绍了现实世界 中的例子和来自许多国家的最新经验。在过去几年中,中低收入国家积累了大量支持和 补充现有证据库的最新证据,相较于世界卫生组织第一本烟草税收管理技术手册,本手册 提供了更为广泛的知识体系。 第二章深入研究了烟草消费税政策的理论、实践和实证证据,包括当前烟草消费税的 全球发展趋势。本章详细分析了烟草税制结构的各种构成因素,以便让决策者全面了解 5第一章 为什么制作本手册 影响价格、消费和市场的因素。本章描述了为最大限度地发挥增税的作用,在设计烟草税 政策和改善税制结构时应关注的要点。本章还包括最新的全球价格和税收数据、各国的 具体例子、税基弹性讨论、自动消费税调整和定价规则,以及对新出现的尼古丁、烟草制品 的说明,包括电子尼古丁、非尼古丁输送系统(ENDS/ENNDS)和加热烟草制品(HTPs)。 第三章侧重于烟草税收管理。本章深入讨论了使烟草税征收有效和高效的基本要 素,确保实现税收政策的健康目标和收入目标。本章强调了参与国家内部和跨境实施烟 草税的各机构之间开展合作的重要性。以过去几十年的国家和地区经验为基础,概述了 维持对整个税收合规周期的监督的具体措施和建议。还包括便于管制和执法的措施,如 许可证、财政标记(如税票)、追踪和溯源系统以及进出口管制等。最后,讨论的重点是合 理税收制度的更广泛要素,如主管机构的适当资源配置、强有力的司法手段和严格的反腐 败法律。 第四章论述了当各国试图增加烟草税或简化税收结构时,在政治经济领域面临的重 要挑战。烟草行业通常依赖已确定的论证模式和策略来抵制此类改革,本书总结为恐吓 (SCARE)策略,表现为以下几个方面:(S)走私和非法贸易;(C)法庭和法律挑战;(A)反 贫困言论;(R)收入减少;(E)就业影响。本章将详细讨论针对这些方面的策略,并就非法 贸易规模的衡量标准以及烟草税专款专用展开详细讨论。 第五章提供了本手册中讨论的烟草税收政策和管理最佳实践的综合清单列表,该列 表为实用指南和所提出的要点提供了快速查询参考。 6 世界卫生组织烟草税政策和管理技术手册 本章参考文献 1.Ranson K,Jha P,Chaloupka FJ,Nguyen SN. The effectiveness and cost-effectiveness of price and other tobacco control policies. In:Jha P,Chaloupka FJ,editors. Tobacco control in developing countries. Oxford:Oxford University Press;2000:427-447 (https://www.paho. org/hq/dmdocuments/2010/ Cost-effectiveness%20of%20price%20increases.pdf,accessed 4 February 2021). 2.Effectiveness of tax and price policies for tobacco control. Organization;2011. (IARC handbooks of cancer prevention:tobacco control:Vol. 14;https://publications.iarc.fr/Book- And-Report-Series/Iarc-Handbooks-Of-Cancer-Prevention/Effectiveness-Of-Tax-And-Price- Policies-For-Tobacco-Control-2011,accessed 4 February 2021). 3.Tobacco taxation in the United States. In:Lynch BS,Bonnie RJ,editors. Institute of Medicine (US) Committee on Preventing Nicotine Addiction in Children and Youths. Growing up tobacco free:preventing nicotine addiction in children and youths. Washington (DC):National Academies Press (US);1994 (https://www.ncbi.nlm.nih.gov/books/ NBK236771/,accessed 10 November 2020). 4.Sin tax reform. Manila:Department of Finance (Philippines);2012 (https://www.dof.gov. ph/advocacies/ sin-tax-reform/,accessed 10 November 2020). 5.Goodchild M,Perucic AM,Nargis N. Modeling the impact of raising tobacco taxes on public health and finance. Bul World Health Organ. 2016;94:250-7 (https://www.who.int/ buletin/ volumes/94/4/15-164707.pdf,accessed 4 February 2021). 6.The economics of tobacco and tobacco control. Bethesda,MD:Department of Health and Human Services,National Institutes of Health,National Cancer Institute,NIH Publication No. 16-CA-8029A;2016 (National Cancer Institute tobacco control monograph 21;https:// cancercontrol.cancer.gov/ brp/tcrb/monographs/monograph-21,accessed 17 December 2020). 7.WHO report on the global tobacco epidemic,2008:the MPOWER package. Geneva:World Health Organization;2008 (https://www.who.int/tobacco/mpower/mpower_report_ful_ 2008.pdf,accessed 10 November 2020). 8.WHO report on the global tobacco epidemic,2019:offer help to quit tobacco use. Geneva: World Health Organization;2019 (https://www.who.int/tobacco/global_report/en/, accessed 10 November 2020). 9.Goodchild M,Nargis N,Tursan d􀆳Espaignet E. Global economic cost of smoking-attributable diseases. Tob Control. 2018;27:58-64 (https://tobaccocontrol.bmj.com/content/27/1/ 58,accessed 4 February 2021). 10.Guidelines for implementation of Article 6 of the WHO FCTC. Geneva:World Health Organization;2014 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6. pdf,accessed 4 February 2021). 7第一章 为什么制作本手册 11.Tackling NCDs:“best buys” and other recommended interventions for the prevention and control of noncommunicable diseases. Geneva:World Health Organization;2017 (https:// apps.who.int/iris/ handle/10665/259232,accessed 10 November 2020). 12.Saving lives,spending less:a strategic response to noncommunicable diseases. Geneva: World Health Organization;2018 (https://www.who.int/publications/i/item/WHO- NMH-NVI-18.8,accessed 4 February 2021). 13.Chaloupka FJ,Warner KE. The economics of smoking. In:Culyer AJ,Newhouse JP, editors. Handbook of health economics. Elsevier;2000;1(1):1 539-1 627. 14.Iglesias RM. Increasing excise taxes in the presence of an ilegal cigarette market:the 2011 Brazil tobacco tax reform. Rev Panam Salud Publica. 2016;40(4):243-9 (https://iris. paho.org/bitstream/ handle/10665.2/31306/v40n4a09 _ 243-9.pdf? sequence = 1&isAlowed=y,accessed 17 February 2021). 15.Cetinkaya V,Marquez PV. Tobacco taxation in Turkey:an overview of policy measures and results. Washington (DC):World Bank Group;2017 (https://openknowledge.worldbank. org/handle/10986/26387,accessed 10 November 2020). 16.Kaiser K,Bredenkamp C,Iglesias R. Sin tax reform in the Philippines:transforming public finance,health,and governance for more inclusive development. Washington (DC):World Bank Group;2016 (http://documents.worldbank.org/curated/en/638391468480878595/ pdf/106777-PUB-PUBLIC-PUBDATE-7-26-2016.pdf,accessed 10 November 2020). 17.Petit P,Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund;2016 (International Monetary Fund How To Notes,No.3/ 2016;https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 10 November 2020). 18.Irwin A,Marquez P,Jha P,Peto R,Moreno-Dodson B,Goodchild M,et al. Tobacco tax reform:at the crossroads of health and development -a multisectoral perspective. Washington (DC):World Bank Group;(https://untobaccocontrol.org/taxation/e-library/ wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf,accessed 2 February 2021). 19.Information note:COVID-19 and NCDs. Geneva:World Health Organization;23 March 2020 (https:// www.who.int/publications/m/item/covid-19-and-ncds,accessed 10 November 2020). 20.WHO statement:tobacco use and COVID-19. Geneva:World Health Organization;11 May 2020 (https://www.who.int/news-room/detail/11-05-2020-who-statement-tobacco-use- and-covid-19,accessed 10 November 2020). 8 世界卫生组织烟草税政策和管理技术手册 第二章 烟草消费税政策 2.1　烟草税全球概况 设计合理的税收政策是制定有效税收政策的关键。任何计划改革税收政策的政府都 必须首先了解合理税收政策的基本组成要素,并考虑不同税收政策的优缺点、税收如何影 响价格以及对税收管理的要求。由于市场和税制结构之间存在不可避免的相互作用机 制,了解一个国家的烟草市场如何运作对决策者而言同样重要。 除了强烈影响烟草税收政策制定的政治因素之外,本章重点关注烟草税收的技术方 面———特别是消费税。 第 2.1 节概述了烟草税的全球实践,重点关注各国构建消费税的不同方式。 第 2.2 节从增加税收、消费税结构及其对价格的影响,以及烟草市场结构如何影响 税制结构变化趋势的角度强调了精心设计烟草消费税制的重要性。本节还讨论了衡量标 准是影响税收政策制定的原因,并介绍了执行衡量标准的关键要素,以及可用于监测进展 的相关指标。 第 2.3 节描述了烟草税制设计阶段的外部政策考虑,以确保实现烟草控制目标和税 收收入目标。为此,本节从战略层面的角度讨论了国内跨部门政策的整合和连贯性,以确 保其他部门的政策不会干扰或阻碍公共卫生政策目标。本节还根据现有区域一体化的经 验回顾了区域税收协作的现状,并就维护一体化成员国家的公共卫生利益总结了最佳政 策方法。 第 2.4 节讨论了新出现的尼古丁和烟草制品,特别是加热不燃烧制品( HTP)和电 子尼古丁和非尼古丁传输装置(ENDS/ENNDS)。本节回顾了有关这些产品对健康影响 的最新证据以及当前对这类产品的监管措施。确定了关键的政策考虑因素,并就对这些 产品采取适当的消费税政策提供了建议。 第 2.5 节总结了本章中涉及的议题和关键要点。 2.1.1　对烟草制品所征收的税 税收分为直接税和间接税。对个人或公司的利润、收入、财产(或财富)征收的是直接 税,而对商品和服务的价格征收的是间接税。间接税与烟草制品税收最为相关,因为它直 接影响烟草制品的价格。对烟草制品征收的间接税可以有很多税种,具体包括: • 消费税———适用于对少数特定商品征收(通常也适用于酒精、燃油、含糖饮料等)。 • 增值税或销售税———增值税是对所有商品和服务在多环节征收的税,以消费者购 买商品支付的价格为税基按比例征收。它是对商品在其生产或流通的每一阶段的价值增 加额征税。有些国家征收销售税而不是增值税。与在供应链的每个阶段都征收增值税不 同,销售税通常在零售环节就消费者购买的商品和服务的零售价格为税基按比例征收。 无论是征收增值税还是消费税,消费者都是最终的负税人。 • 进口关税———对进口到一个国家并在该国消费的特定商品(即未过境到另一个国 家的商品)征税。一般而言,进口关税在商品进入该国时向进口商征收。 • 其他税———不属于上述类别的其他间接税,例如环境税。 关于烟草制品的税收政策中最被广为人知和深入人心的观点之一是烟草制品应缴纳 消费税。因此,本章以及整个手册关注的重点就是消费税。消费税是控制烟草制品消费 最重要的税种,因为消费税直接对烟草制品征收,并且消费税的征收显著地拉大了烟草制 品相对于其他商品的价格差距,进而减少了烟草制品的消费。 消费税的税率设计有两种基本类型: • 从量税———以被征税产品的物理单位量为计税依据征收定额税(例如以每1 000 支卷烟、20 支一包、每千克的烟草制品为计税依据); • 从价税———以被征税产品的价格作为计税依据[例如零售价、生产商/出厂价或成 本、保险和运费 (CIF,到岸价) ]征收一定百分比的税。 无论是从量税还是从价税,都可以对所有烟草制品实行单一税率,也可以实行多重税 率;从量税和从价税既可以单独使用,也可以混合使用。 鉴于卷烟在世界上很多地方几乎是唯一使用的烟草制品,且缺乏其他烟草制品的数 据,本章主要关注卷烟的税收政策,同时也适当纳入了一些关于其他烟草制品的例子和政 策建议,如在世界一些特定地区流行的烟草制品(例如东南亚的比迪烟或无烟烟草以及东 地中海地区的水烟)。 2.1.2　卷烟税和最近的全球趋势 税收和国民收入水平:收入水平越高,烟草制品的税收和价格水平就越高。 在全球层面,卷烟价格和税收水平与一个国家的收入水平呈正相关:高收入国家卷烟 的价格和税收较高,而随着收入水平的降低,卷烟的价格和税收水平也相应较低。自 2008年以来,这种趋势一直没有改变。图2.1使用世界银行对国家收入规模的分类,显 01 世界卫生组织烟草税政策和管理技术手册 CIF,到岸价,是指进口产品在进入一国时向海关申报的价值。 示了2018年按国家收入规模划分的卷烟价格和税收水平。 零售价减去总税收每包消费税 其他税收 PPP$7.80 每 包 卷 烟 的 价 格 和 税 收 /P PP $ 8 7 6 5 4 3 2 1 0 高收入国家 低收入国家 中等收入国家 全球平均 总税收： 5.30 （67.9% of pack price） 总税收： 2.91 （58.3% of pack price） 总税收： 1.18 （38.1% of pack price） 总税收： 3.36 （60.8% of pack price） PPP$4.99 PPP$3.09 PPP$5.53 4.25 0.68 2.06 2.48 图2.1 2018年按国家收入规模计算的卷烟加权平均零售价格以及 最畅销品牌卷烟的税收(消费税和总税收) 注:平均值为世卫组织根据 2017 年每个国家当前 15 岁及以上吸烟人数的估计进行加权平均后的值。 价格以购买力平价(PPP)美元或国际美元表示,以考虑到各国购买力的差异———具体计算基于 53个高收入国 家,97个中等收入国家和28个低收入国家最畅销品牌的价格数据、消费税以及其他税收和 PPP 转换因素。 资料来源:参考文献1。 税收和价格水平:税负越高,价格越高 在全球范围内,卷烟价格与税收百分比水平呈正相关:随着总税收(其中消费税占最 大头)占零售价格的比重增加,卷烟价格通常也会增加(图2.2) 。这说明税收确实影响 卷烟价格。 税收水平和世卫组织区域分类:欧洲区税收和价格最高,其次是东南亚、美洲区和西 太平洋区,最低的是东地中海和非洲区。 在区域层面(按世卫组织区域分类),烟草制品的价格和税收的平均水平差异很大。 欧洲区的价格和税收水平最高,其中包括欧盟(EU) 国家。欧盟统一的税收结构要求成 员国对烟草制品征收较高水平的最低税,从而直接导致了烟草制品的高价格并鼓励成员 国定期增加烟草税以达到欧盟的相关要求。然而,随着所有欧盟成员国现在都达到了目 前的最低要求水平,增加消费税的动力可能会减弱。事实上,欧盟的最低卷烟消费税自 2014年1月1日生效以来就没有调整过,并且正遭受通胀侵蚀。 欧盟成员国在2020年6月承认了这一点,并表示需要在欧盟层面采取行动,以确保 最低消费税重新获得关注,以有效减少烟草制品的消费,并确保对一些烟草制品增加最低 消费税[2]。 11第二章 烟草消费税政策 这是一个普遍的趋势,并不适用于每个国家;有些国家卷烟有很大的税收份额,但其价格很低。 每包消费税 其他税收 零售价减去总税收 PPP$7.07 每 包 20 根 卷 烟 的 价 格 和 税 收 /P PP $ 7 6 5 4 3 2 1 0 总税收≥75% 50%≤总税收<75% 25%≤总税收<50% 总税收≤25% PPP$5.33 PPP$5.07 PPP$2.60 1.57 4.44 2.24 0.17 图2.2 2018年按总税收水平计算的最畅销品牌卷烟的加权平均零售价格和税收(消费税和总税收) 注:平均值为世卫组织根据 2017 年每个国家当前 15 岁及以上吸烟人数的估计进行加权平均后的值。 价格以购买力平价(PPP)美元或国际美元表示,以考虑到各国购买力的差异———具体计算基于 53个高收入国 家,97个中等收入国家和28个低收入国家最畅销品牌的价格数据、消费税以及其他税收和 PPP 转换因素。 资料来源:参考文献1。 非洲区和东地中海地区的烟草消费税最低,而中国(因其烟草市场规模庞大而单独报 告)的烟草税率则低于其所处的西太平洋地区(图2.3)。 零售价 消费税 总税收份额% 8 7 6 5 4 3 2 1 0 9 10 每 包 20 支 卷 烟 的 价 格 和 税 收 /P PP $ 非洲地区 美洲地区 东地中海地区 欧洲地区 东南亚地区 西太平洋地区 不含中国 中国全部 3. 80 1. 05 5. 61 2. 51 4. 02 1. 95 7. 53 4. 24 7. 27 2. 89 4. 89 2. 65 4. 02 1. 45 5. 53 2. 48 20 30 40 50 60 70 80 42.2% 55.8% 61.2% 72.9% 56.9% 63.2% 55.7% 60.8% 税占 零 售 价 格 比 重 /% 图2.3 2018年按地区划分最畅销品牌卷烟的加权平均零售价格和税收(消费税和总税收) 注:之所以将中国与西太平洋地区(WPRO)的平均数分开计算,是因为与该地区其他国家的吸烟人数相比,中 国的吸烟人群庞大。 价格以购买力平价(PPP)美元或国际美元表示,以考虑到各国购买力的差异———具体计算基于53个高收入国 家,97个中等收入国家和28个低收入国家最畅销品牌的价格数据、消费税以及其他税收和PPP转换因素。 资料来源:参考文献1。 21 世界卫生组织烟草税政策和管理技术手册 全球税收结构趋势:越来越多的国家采用从量税或更多依赖于从量税的混合消费税制 自2008年以来,随着越来越多的国家采用从量税或混合消费税制度,仅依赖从价税 或根本不征收消费税的国家数量有所减少。更多实施混合消费税制的国家增加了从量税 在混合消费税制中的比重(图2.4、图2.5) 。 从量税 从价税 混合税 无消费税 70 60 50 40 30 20 10 2008 2010 2012 2014 2016 2018 国 家 的 数 量 /个 24 23 24 21 19 15 42 41 44 4749 55 45 50 57 59 56 60 63 63 62 54 56 57 图2.4 2008—2018年消费税结构的变化 混合税 更多地依赖于从量税 更多地依赖于从价税 70 60 50 40 30 20 10 2008 2010 2012 2014 2016 2018 国 家 的 数 量 /个 45 50 54 56 57 63 23 27 27 32 35 37 22 23 27 24 22 26 图2.5 混合消费税制中倚重从量税与倚重从价税比较(2008—2018) 征收最低从量消费税:半数采用混合或从价结构的国家征收最低消费税 最低从量消费税确保至少缴纳一定的最低税额,而与价格水平无关。在可提供最低 消费税数据的101个征收从价税或混合消费税的国家(47个国家/地区)中,几乎一半的 31第二章 烟草消费税政策 有关2018年各国烟草消费税制结构的信息,请参见附件2.1。 国家/地区设定了最低从量消费税 。其中近2/3(29个国家/地区)是高收入国家,且大 多数都是欧盟成员国,因为欧盟要求其成员国对烟草制品征收最低从量税。 全球范围烟草消费税税基的选择:在实行混合或从价消费税制的国家中,几乎有一半 以零售价为征税的税基 从量税的税基设定相对容易:大多数国家对卷烟以规定的支数为税基、对烟草以千克 为单位,以及对其他烟草制品以克为单位作为从量税的税基[1]。 各国采用从价税时,税基的适用则表现出差异。在105个实施从价税或可获得数据 的实行混合消费税的国家(47个国家)中,近一半的国家使用零售价作为从价税的税基 , 其中大多数是高收入国家(29个国家)。使用零售价作为从价消费税的税基比使用生产 者价格或海关到岸价格(CIF)更有效率。以零售价作为税基便于税务管理人员通过监测 市场价格核定征税,而如果以生产者价格或进口到岸价格(CIF)作为消费税从价税的税 基,则有可能被生产商或进口商故意压低出厂价或进口价格,并将利润转移给供应链下游 的相关方,从而降低税收负担。这种策略也被称为转移定价。 此外,全球层面的数据还表明,与在其他环节的价格(如生产者价格或进口价格)作为 消费税税基相比,以零售价为税基的从价消费税似乎能够使平均零售价更高。 关于复杂的、多层税率的消费税制:仍有31个国家对烟草制品征收复杂、多层税率结 构的消费税 截至2018年,有31个国家根据卷烟的不同特征征收不同税率的消费税,这些特征包 括价格水平、生产类型、包装类型以及卷烟的长度(表2.1)。有些国家甚至采用不止一种 标准来适用差异税率。例如,印度尼西亚根据卷烟的产量、卷烟的类型和价格水平(对生 产商)按不同税率征税。而在有些国家,例如欧盟,法律则禁止对卷烟实行差别税率,所有 卷烟的从价税或从量税税率都必须相同[3]。 表2.1 各国适用多重差异税率的标准(2018) 差异税率的税基标准 国 家 零售价 孟加拉国、白俄罗斯、印度尼西亚、约旦、莫桑比克、缅甸、巴 基斯坦、泰国 卷烟等级(例如高级、中级、经济型) 埃及、日本a、马里 生产价格 中国、老挝人民民主共和国 产量 印尼 41 世界卫生组织烟草税政策和管理技术手册 这意味着实行混合税率的国家除了不仅征收从量消费税外,还设置了最低价从量税(即从量消费税加从价消费税 之和不能低于设定的最低从量消费税)。 包括对不含增值税的零售价征收从价税的国家,由于零售价格很容易确定,而增值税税率是已知的变量。 续表 差异税率的税基标准 国 家 类型 带过滤嘴/不带过滤嘴 白俄罗斯、格鲁吉亚、印度、肯尼亚、摩尔多瓦共和国、尼泊 尔、巴布亚新几内亚 手制烟/机制烟 印度、印度尼西亚 丁香烟/白肋烟 印尼 烟草成分(深色/金黄色或 深色/浅色) 阿尔及利亚、玻利维亚 外包装 软/硬 莫桑比克、乌干达 卷烟长度 印度、尼泊尔、斯里兰卡 贸易(国产/进口) 伊朗、黎巴嫩、缅甸、所罗门群岛、汤加、乌兹别克斯坦 烟叶来源(国产/进口) 斐济、坦桑尼亚 a 日本和格鲁吉亚在2018年收集这些数据时正在使用分层消费税结构,但到2020年,这两个国家已不再使用 差异税率。 资料来源:参考文献1。 2.2　消费税政策设计 大幅提高烟草制品的税收和价格是控制烟草使用的最有效和最具成本效益的政 策[4]。增加的税收通过价格机制传导给消费者,进而价格上涨能够减少烟草消费。在设 计烟草税收政策或改革烟草税收制度时,政策制定者面临着从技术问题(例如确定适用的 税收结构和税率)到政治经济问题(例如应对烟草业的SCARE 策略)的挑战。本节从 健康角度为政策制定者提供关于烟草税制设计需要考虑的方方面面的问题,包括最佳税 制结构的设计。本节还提出了烟草税政策和制度调整时需要考虑的指标。 2.2.1　消费税结构类型的重要性 下面回顾了关于选择(统一)从量消费税和从价消费税的现有理论和经验证据,以及 它们对价格、消费、感知质量和烟草制品种类、政府税收收入以及税收管理方面的影响。 本章中“质量”一词的使用绝不是指烟草制品对健康的影响,是指消费者对质量的看 法影响着他们购买产品的决定,他们可能会根据包装、卷烟使用的混合物或任何使产品对 他们更有吸引力的因素来评估卷烟质量。需要明确的是,即便消费者认为卷烟质量有高 低,但从公共健康的角度来看,我们必须明确所有卷烟都同样有害。 从价税或从量税的选择取决于烟草市场结构,即商品和服务市场的竞争性质和程度。 尽管每个国家都有自己的特点,但烟草市场结构通常是垄断或寡头垄断,企业有权控制价 51第二章 烟草消费税政策 恐吓策略是烟草业在各国政府计划增加烟草税时最常遇到的抵制策略。在第四章中将加以详细地描述和驳斥。 格———从而利用税收结构为自己谋取利益。例如,作为世界上最大的烟草制品生产国和 消费国[5],中国实行烟草国家垄断。在越南,外国品牌是在国家垄断的许可下生产的。在 泰国和埃及,尽管有外国公司存在,但市场仍由国有企业主导。在乌拉圭,烟草行业则是 由国内公司实行寡头垄断。在孟加拉国,则是由本国企业与几家外国公司竞争形成寡头 垄断[6]。非洲大部分地区的烟草市场由跨国烟草公司把持[7]。 税收结构对最终价格的影响:单一从量税与单一从价税 烟草税收政策中关于从量税和从价税的选择是一个影响深远的议题,因为消费税的 水平和结构设计对不同群体的利益和目标有着不同的影响。鉴于烟草行业的市场结构在 大多数国家,就大多数烟草制品而言通常都是垄断或寡头垄断,即不同的消费税可能对政 府收入、制造商利润、消费者价格、感知的产品质量和品种以及税收管理产生不同的影 响[816]。因此,从量税和从价税这两种消费税,可能对公共卫生产生不同的影响,因为它 们通过影响消费者感知到的产品质量、品种和价格来影响个人消费。此外,政府有可能通 过影响烟草消费税从而管理烟草使用需求、增加政府收入和促进大众公共健康。 税收对价格的影响进而对消费的影响还受到价格和收入弹性以及消费者对质量的看 法(感知质量)和可用产品种类的影响,而这些又与采用的税收结构类型相关。 税收结构受需求价格弹性和供给价格弹性的影响。需求价格弹性衡量消费者需求对 价格变化的反应程度。供给的价格弹性衡量生产者对价格变化的敏感程度。 烟草税收结构也受市场结构的影响。在垄断市场中,公司以利润最大化为目标设定 价格,考虑到需求的价格弹性:价格弹性(绝对值)越低,即消费者对价格变化的敏感性越 低,垄断者可以设定的价格就越高。在垄断市场结构中,利润通常是超额的,这意味着总 销售收入高于总成本(其中总成本包括正常利润)。因此,垄断生产商获得的回报高于投 资资本(物质和人力)和承担商业风险所需的最低回报。 经济学理论认为,在私人垄断市场中的产品价格高于寡头垄断市场。然而,当存在国 家垄断且政府的目标不是直接的利润最大化时,就不一定是这样了。政府可能有其他考 虑,例如保留工作岗位(例如在中国)或为低收入者提供低价产品(例如在埃及)。 在垄断体制下,相较于采用从量税,垄断生产商采用从价税能够将价格压得更低。之 所以会如此是因为在从价税率下,当供应增加而价格下降时,价格下降并不完全由生产者 承担而是部分由政府承担了,因为随着供应的增加,每单位销售产品的税负下降。换句话 说,相对于收入等价的从量税,从价税会导致更低的价格和更高的消费。从技术上讲,这 意味着从价税率下的供给弹性较小。相反,在从量税率结构下,垄断生产者价格的任何上 涨都会作为其收入,这将会激励他们提高价格。 同样的逻辑也适用于寡头垄断市场结构,寡头垄断市场通常也拥有超额利润。 关键要点1 在垄断或寡头垄断市场结构中,从量税比从价税更能激励烟草商设定较高的价格。 然而,理解寡头垄断更为复杂,因为它们的特点是少数公司之间的战略相互依赖。这 种战略上的相互依赖延伸到行业和监管机构之间的关系。无论是通过税收还是其他干预 61 世界卫生组织烟草税政策和管理技术手册 措施,行业会对政府的监管政策进行预判,并相应地采取行动。例如,竞争者可能会协调 和游说反对某种税收结构改革或税率提高。 在寡头垄断市场结构下,从价税是将部分利润作为税收转移给政府的一种相对更有 效的工具,因为它的作用类似于消费税和利润税。相比之下,从量税对利润的(负面)影响 较小。这解释了为什么我们观察到主打高价品牌的跨国烟草公司(例如菲利普莫里斯国 际[PMI])游说支持从量税[17]。例如,海湾阿拉伯国家合作委员会(GCC)中的国家,烟草 业长期以来一直试图游说政府引入从量消费税[1821]。经过多年关于可能征收消费税的 考虑和讨论,海湾合作委员会于2016年11月[21]通过了海湾合作委员会各国的共同消费 税协定,该协定对烟草制品实行从价消费税政策。 烟草公司对消费税结构的支持最终取决于它们在特定国家控制的细分市场的地位。 主要销售优质品牌的公司会倾向于支持从量税,而销售中等价位或经济型品牌的公司会 倾向于从价消费税[17]。 当寡头垄断企业生产相同的产品时,从量税对价格具有更强的积极影响,并且比从价 税更有可能过度转移到消费者价格[13]。过度转移意味着价格上涨幅度超过税收上涨幅 度本身。这一点得到实际的证据支持[2226]。 关键要点2 在寡头垄断中,当需求相对缺乏弹性时,价格的上涨幅度可能会超过从量税的增加 幅度。 一般来说,对产品的需求不仅取决于价格,还取决于消费者对质量的看法和对品种的 偏好。例如,在海湾合作委员会国家最受欢迎的品牌是万宝路,它是一个高端品牌[1]。 消费者依据他们各自对烟草制品质量的感知而表现出不同的支付意愿,并进而决定 他们最终购买高价品牌还是低价品牌。税收引起的价格上涨可能会引起消费者或烟草制 品使用者的以下反应:①有些烟民会戒烟;②有些烟民会减少他们的整体消费;③在提高 从量税的情况下,高价烟相对于低价烟的价格差缩小了,这时高收入烟民可能消费升级, 即转而购买更高价格的烟,而在提高从价税的情况下,高价烟相对于低价烟的价格差扩大 了,低收入烟民就有可能会消费降级,即转而购买更便宜的烟;④有些烟民可能转向进入 非法烟草市场或在可能的情况下在税收较低的邻国购买烟草产品。 消费者的购买决定还受到他们对产品类型偏好的影响,即消费者认为质量相同,但生 产商根据消费者的口味偏好赋予了不同特征的产品以不同的口味,增加了消费者可选择 的产品类型。因此,由于消费者可选择的产品种类增加,增税引起的平均价格上涨可能会 导致市场需求总量的增加。产品的多样化能够吸引新的消费者,尤其是在相关法律法规 不严格的市场环境中(例如,没有平面包装和对烟草口味限制的相关法律法规)。当烟草 业将薄荷烟引入烟草市场时,就吸引了新的消费者群体。有充分的证据表明薄荷烟会吸 引更多青少年尝试吸烟,同时增加了吸烟者的尼古丁依赖并减少了成人戒烟意愿[27]。为 防止这种情况的发生,欧盟成员国已禁止在烟草制品中使用烟草以外的口味[28]。因此, 考虑消费税的结构和水平可能对平均价格、感知质量以及卷烟品牌和其他新型替代品的 71第二章 烟草消费税政策 多样性产生更广泛的影响,这一点也很重要。 当消费者根据数量以外的维度做出选择时,从价税和从量税这两种税率结构就会表 现出不同的政策效果,即使在企业没有市场支配力的完全竞争市场中也是如此[2930]。为 了说明这一点,我们来考察在提高1美元成本的情况下消费者对烟草制品质量的看法。 在从量税率结构下,提高1美元将导致所有烟草制品等幅价格上涨,但从价税率结构下则 不会如此。在从价税率为20%的情况下,由于乘数效应,价格必须增加1美元以上,或增 加1/(1-0.2)美元以支付1美元的改进成本。从量税促使消费者减少需求量,但他们仍 可能选择支付更高的价格以换取他们认为质量更好的产品。另一方面,从价税会导致数 量和感知质量的下降,而不是两者之间的替代。从价税仅具有收入效应,与从量税不同的 是,它并不会导致感知质量和数量之间的替代。 关键要点3 从量税率激励烟草行业投放高端烟草制品,以吸引新的消费者并刺激消费。 当企业生产价格差异化的产品时,如烟草业所做的那样,采用从量税与从价税在经济 学意义上会产生较为模糊的结果。这两种税率形式的相对影响并不像生产同质产品的寡 头垄断企业那样直接。对于差异化产品,税率形式的相对影响取决于多种假设:企业是否 面临对称成本,企业数量是否固定或新企业是否可以进入市场以及税收收入水平。当企 业面临不同的成本时,从价税会加剧它们之间边际成本的绝对差异。高价品牌不被认为 是低价品牌的完美替代品。足够高的从价税率可能会导致高成本产品的相对生产不 足[31]。从量税和从价税对消费者质量观念的影响取决于市场结构以及不同质量需求的 价格弹性和收入弹性。最便宜产品的相对价格不一定保持不变或上涨,它可能会下 降[32]。 实践经验表明,在逐步增加税收而忽略收入效应的情况下,增加从量税会导致低端品 牌的市场份额下降,而高端品牌的市场份额则会增加[3334],Chaloupka等人[35]发现,在实 行混合税率的21个欧盟国家中,当混合税率制的从量税组成部分占主导地位时,高档和 低价品牌之间的价格差距(尽管没有反映卷烟的全部分销价格)就会缩小 。虽然从量税 下价格差距更小,但有证据表明,企业有时会通过引入新的、非常便宜(较便宜)的品牌来 应对增税,或者实行差别税收转嫁策略。 这种做法在印度很长一段时间内都体现得很明显,由于对所谓的微型卷烟(长度<60 mm)适用较低的税率,印度烟草公司在其旗舰卷烟品牌Gold Flake下推出了许多更便宜 的卷烟品种[36]。因此,近年来印度卷烟市场低端品牌显著扩大,部分原因就是推出像 Gold Flake Century 这样的新品的市场营销策略。 再举一个例子,自2006年以来跨国烟草公司在英国推出的超低价品牌卷烟在几年内 成功地将其市场份额翻了一番:它们的实际价格并没有上涨,因为它们消化了部分增加的 81 世界卫生组织烟草税政策和管理技术手册 欧盟国家实行设有最低税收下限的混合税率结构。有些国家比其他国家在混合税率结构中更依赖从量税,但是 需要保持在限定的范围内(从量税须占加权平均价格的5%~76.5%)。 税收[37]。在2001—2009年间,超低价品牌的市场份额从5%上升到10%,而经济品牌的 市场份额从40%上升到50%,高端品牌和中等价位品牌的市场份额则同期下降(前者从 35%下降为低于25%,后者从15%下降到5%)。为了保持折扣品牌的低价和留住市场 上购买低价产品的消费者,烟草公司可能会实施差异化的税收转嫁策略,即对高档产品多 转嫁增加的税收(加价大于加税),对低价产品则少转嫁增加的税收(加价小于加 税)[3740]。 关键要点4 有证据表明,在从量税率结构下,品牌之间的价格差距会缩小。随着烟草业同时整 合生产商并扩大其产品组合,越来越多的证据表明,烟草业一方面正在引进更便宜的品 牌,另一方面又同时提高高档品牌的价格,这样看起来自相矛盾的操作实际上扩大了产 品的价格差距。然而,如此这般操作的影响程度仍不清楚,而且这一证据并不能否定从 量税收结构缩小价格差距的总体结论。 税率结构对最终价格的影响:单一从量税、从价税和混合税率制度 来自2019年RGTE[1]的数据表明,在实施更多依赖从量税的混合税率制度的国家 中,(按吸烟者人数加权)最畅销品牌卷烟的平均价格最高,其次是只征收从量消费税的国 家,再次是更多依赖从价消费税的混合税率制度的国家,最后是仅采用从价消费税的国家 (图2.6)。而对于那些不征收烟草消费税的国家,最畅销品牌卷烟的平均价格则是最低 的。在世卫组织过去关于全球烟草流行的报告中,采用从量税的国家卷烟平均价格最高。 然而这种趋势可能已经改变,部分原因是越来越多的国家正在采用混合消费税制度。 零售价减去税收消费税 其他税 7 6 5 4 3 2 1 0 PPP$7.56 PPP$5.66 PPP$4.84 PPP$3.73 PPP$2.52 1.66 2.21 2.62 3.26 每 包 20 支 卷 烟 的 价 格 和 税 收 /P PP $ 更依赖于从量税 的混合税率 从量税 更依赖于从价 税的混合税率 从价税 不征消费税 图2.6 2018年按消费税率结构分类的最畅销品牌卷烟的加权平均价格和消费税 注:平均值根据世卫组织对2017年各国当前15岁及以上吸烟人数的估计进行加权。价格以购买力平价(PPP) 调整后美元或国际美元表示,以说明各国购买力的差异———基于53个高收入国家,97个中等收入国家和28个 低收入国家最畅销品牌的价格数据、消费税以及其他税收和PPP转换因素。 资料来源:参考文献1。 91第二章 烟草消费税政策 Shang等人[41]描述和比较了价格分布,他们使用了来自国际烟草控制(ITC)项目的 16个征收卷烟税率结构不同的国家的数据。与所有采用其他税率结构(从价税、混合税、 分层税)的国家相比,采用统一从量税率结构的国家的烟草价格波动通常较小。一般而 言,除了统一从量税以外,其他税率结构为品牌转换和避税提供了更多机会。复杂的税率 结构可能会导致更广泛的价格分布,从而导致更严重的避税行为,因为当税率提高时,便 宜的品牌会有更多替代的机会。 关键要点5 有证据表明,单一从量税率或更多依赖于从量税的混合税率结构最有可能导致价 格上涨。 税率结构对最终价格的影响:单一税率与差异税率 税收结构中影响最终价格的另一个方面是差异税率的使用,即根据产品不同的特点 而征收不同的税率。这些特点包括价格水平、卷烟所含烟叶的类型、生产量的大小、包装 等。表2.1(见前文)列出了31个国家采用差异税率所依据的标准。有证据表明,使用差 异消费税率的国家的卷烟平均价格和平均消费税水平往往比使用单一消费税率的国家低 得多(图2.7)。 零售价减去税收消费税 其他税 单一消费税率 PPP $6.30 PPP $4.99 差异消费税率 每包 20 支卷烟的价格和税收/PPP$ 1.94 3.28 图2.7 最畅销卷烟品牌的加权平均价格和消费税在实行单一消费税率和 差异消费税率国家的比较(2018年) 注:平均值根据世卫组织对2017年每个国家当前15岁及以上吸烟人数的估计进行加权。 价格以购买力平价调整美元或国际美元表示,以说明各国购买力的差异———基于53个高收入国家、97个中等 收入国家和28个低收入国家最畅销品牌的价格数据、消费税和其他税收以及PPP转换因素。 资料来源:参考文献1。 统一的从量税率结构可能会导致价格相对较高,而价格分布的可变性较小。与基于 品牌特征的不同税率的分层税收结构相比,统一税收可能会减少消费者转向更便宜品牌 的动力(导致更高的戒烟率和更低的流行率),并降低制造商减少税收负担的动力,改变他 们的定价策略、生产过程或规模[4245]。例如,在印度尼西亚,小生产商的税收更优惠,制 造商有动力减小生产规模,但增加附属小公司的数量。当税务机关在应用差别税率时考 虑所有关联公司的总产量时,这个问题得到了解决。到2017年,活跃工厂有786家,而 02 世界卫生组织烟草税政策和管理技术手册 2006年有4 198家工厂 。取消差别税率将会更有效地消除避税诱因,并有助于改善公 共卫生和政府财政。 关键要点6 有证据表明,对卷烟采用单一的消费税率不仅比差异消费税率更容易管理,而且更 有可能导致卷烟价格的上涨。 税率结构对最终价格的影响:税基选择的意义 选择能对烟草价格和税收收入带来最大化影响的消费税税基是很重要的。对于从量 税而言,税基是烟草制品的数量。卷烟、雪茄和比迪烟的数量以支数来衡量;对于其他烟 草制品,例如无烟烟草或自卷烟(RYO),税基是以烟草的重量衡量。当采用从价税时,税 基的选择不仅要考虑到公共健康(通过对消费的影响而传导),而且要考虑到对税收收入 和行业利润的影响。 基于出厂价(或到岸价格,CIF)的从价税为烟草制造商提供了减少纳税义务的机会, 尤其是当他们控制分销系统时。烟草生产商可以以较低的价格将卷烟出售给关联方的分 销商,然后以此作为从价税的税基。然而,分销商可以设定高价并与生产商分享额外的利 润[46]。由于存在这种投机定价的可能性,最佳做法是使用零售价作为从价税的税基并引 入每包最低消费税。 2019年 WHO RGTE[1]中的数据显示,平均而言,在以零售价作为从价消费税税基 的国家,一包卷烟的价格水平和消费税水平均要高得多(图2.8)。欧盟采用包括了所有 其他流转税的最高零售价格作为烟草消费税的税基,越来越多的中低收入国家(包括巴 西、埃及、泰国、土耳其和卢旺达)也采用类似的做法。 零售价减去税收消费税 其他税 以零售价为税基的从价税或混合税率 不以零售价作为税基的从价税或混合税率 每包 20 支卷烟的价格和税收/PPP$ PPP $5.01 PPP $6.41 3.72 1.79 图2.8 (2018年)以零售价格为从价税税基的国家最畅销卷烟品牌的加权平均价格 注:平均值根据世卫组织对2017年每个国家当前15岁及以上吸烟人数的估计进行加权。 价格以PPP调整后的美元或国际美元表示,以说明各国购买力的差异———基于53个高收入国家、97个中等收 入国家和28个低收入国家的最畅销品牌的价格、消费税以及其他税收和PPP转换因素的数据。 资料来源:参考文献1。 12第二章 烟草消费税政策 印度尼西亚财政部,个人通讯,2017年。 关键要点7 消费税的税基很重要。对于从量消费税,需要明确定义税基(卷烟、雪茄和比迪烟 通常以支数为税基;对于其他烟草制品,例如无烟烟草或RYO,通常以烟草的重量为税 基)。对于从价消费税(税基通常是零售价、到岸价格或出厂价格),有证据表明,以卷烟 零售价作为消费税税基的国家,往往比采用其他环节卷烟价格作为税基的国家卷烟价 格更高。以到岸价格和生产商的出厂价作为税基会给税务当局确认税基带来困难,并 且容易低估。 税收对最终价格的影响:最低消费税的意义 对于实行从价税或混合税制的国家而言,设置最低消费税是决定最终零售价格的另 一个重要因素。平均而言,在征收最低从量消费税的国家中,一包卷烟的价格以及消费税 水平,远高于没有征收最低从量消费税的国家(图2.9)。虽然设置了最低消费税的47个 国家中有一半以上是欧盟成员国,但在平均计算中如果去除欧盟国家也会得出相同的结 论。 零售价减去税收消费税 其他税收 设置了最低从量税的从价或混合消费税制 没有设置最低从量税的从价或混合消费税制 PPP $7.41 PPP $4.82 4.23 1.80 一包 20 支卷烟的价格与税收/PPP$ 图2.9 (2018年)有或没有设置最低从量税的国家最畅销品牌卷烟的加权平均价格的比较 注:平均值根据世卫组织对2017年每个国家当前15岁及以上吸烟人数的估计进行加权。 价格以PPP调整后的美元或国际美元表示,以说明各国购买力的差异———基于53个高收入国家、97个中等收 入国家和28个低收入国家最畅销品牌的价格数据、消费税以及其他税收和PPP转换因素。 资料来源:参考文献1。 关键要点8 有证据表明,对于采用从价税或混合消费税的国家,征收最低从量消费税的国家中 一包卷烟的价格远高于不征收最低从量消费税的国家。征收最低消费税还能够保证实 现最低消费税收入。 总结烟草制品消费税结构不同选择的优缺点及影响 表2.2总结了根据不完全竞争经济理论预测的及在现实生活中所观察到的不同类型 烟草消费税的特点,以及每种类型的优点和缺点,涉及其对需求数量、品牌感知质量、价 22 世界卫生组织烟草税政策和管理技术手册 格、收入的确定性和稳定性、税务征管及逃避税方面的影响。 表2.2 不同类型的烟草消费税的特点 从量消费税 从价消费税 设置最低从量 税(或 消 费 税 下 限 )的 从 价税 混合从量消费 税 和 从 价 消 费税 具有最低从量消费税 (或最低消费税)的混 合从量消费税和从价 消费税 税基 单位产品(例 如1 000支卷 烟) 产品价 值 (例 如零 售 价、批 发价或制造商 出厂价) 消费税按从价 计算;但是,如 果计算的税额 低于指定的最 低金 额,则 适 用最低从量税 产 品 单 位 及 价值 单位和价值,除非计算 的税额低于指定的最 小值,在这种情况下, 税基是单位(采用设置 的最低从量消费税) 实施 要求 应在生产环节或进口环节征收消费税 低,因为只需 要确定产量 需要强大的税 务管理和技术 能力,否 则 税 收征管成本会 很高 需要强大的税 务管理和技术 能力;否 则 税 收征管成本会 很高,就 像 实 行从价税制度 一样 需要强大的税 务管理和技术 能力;否 则 税 收征管成本会 很高,因 为 需 要同时评估和 征收从价消费 税 和 从 量 消 费税 需要强大的税务管理 和技术能力;否则税收 征管成本会很高,因为 需要同时评估和征收 从价消费税和从量消 费税,以及最低从量消 费税的税收遵从 是否 低估 不是问题 容易被低估 最低从量消费 税能防止低价 品牌在从价税 率中被低估征 税 消费税从价征 收的部分可能 会被 低 估,具 体取决于税基 的选择 最低从量消费税防止 低价品牌可能出现的 从价税基被低估 对产品 质量感 知的影响 升级 效 应 往 往会 降 低 高 价品 牌 的 相 对税收 乘数效应阻碍 了代价高昂的 所谓质量改进 没有动力升级 高价品牌 消除升级高价 品牌 的 动 力, 同时为低价品 牌 提 供 升 级 动力 消除升级高价品牌的 动力,同时为低价品牌 提供升级动力 32第二章 烟草消费税政策 续表 从量消费税 从价消费税 设置最低从量 税(或 消 费 税 下 限 )的 从 价税 混合从量消费 税 和 从 价 消 费税 具有最低从量消费税 (或最低消费税)的混 合从量消费税和从价 消费税 对价格 的影响 往往 导 致 卷 烟价 格 相 对 较高,尤其是 低价卷烟 倾向于导致相 对 较 低 的 价 格;如 果 乘 数 效应 强,降 价 将得到补贴 往往导致低价 卷烟价格上涨 幅度相对较高 从量税的增加 将导致卷烟价 格相 对 较 高, 尤其是低价卷 烟;如 果 从 价 税的基数包括 消费 税,从 量 税的增加也将 增加从价税 从量税的增加将导致 卷烟价格相对较高,尤 其是低价卷烟;如果从 价税的基数包括消费 税,则从量消费税的增 加也会增加从价税额。 如果最低税额占总税 额的百分比,例如加权 平均价格,则从价税和 从量税的增加将提高 支付的最低税额;考虑 到对感知质量的影响, 它们将减少价格差距 通货膨胀 除非 根 据 通 货膨 胀 调 整 税收,否则消 费税 的 实 际 价 值 将 被 侵蚀 随着价格的上 涨,消 费 税 的 实际价值也相 应增 加,至 少 在 某 种 程 度 上,烟 草 制 品 价格跟随通货 膨胀保持同步 增长 除非根据通货 膨胀适时调整 消费 税,否 则 最低从量税的 实际价值将随 着时间的推移 而受到侵蚀 除非根据通货 膨胀适时调整 消费 税,否 则 从量税的实际 价 值 将 受 到 侵蚀 除非根据通货膨胀适 时调整消费税,否则从 量税和最低从量税的 实际价值将被侵蚀 健康收益 无论 价 格 范 围如何,都会 抑制 烟 草 制 品的消费 可能会刺激向 低价烟转移的 降级 消 费,从 而 减 少 健 康 收益 最低从量税会 抑制降级消费 可能会减少降 级消费 减少降级消费 资料来源:参考文献47。 2.2.2　其他税制设计考虑 以从量税自动调整和指数化应对通货膨胀 由于从量税的税基不取决于价格,因此与从价税不同,它不会根据通货膨胀进行自动 调整。随着应税产品价格的上涨,从量税的实际价值会随着时间的推移而被侵蚀。因此, 42 世界卫生组织烟草税政策和管理技术手册 特别是在通货膨胀率快速增长的国家,从量税的名义价值必须定期增加,以保持税收的实 际价值。这对公共卫生和财政收入都非常重要,特别是在那些烟草制造商不定期提高产 品价格或低价烟草制品占据市场主导地位的国家。 表2.3列出了对消费税进行自动调整以避免从量税随着时间的推移而被侵蚀的国 家,这些国家采用不同的调整标准并确定了不同的调整频率。 表2.3 对从量税进行自动调整的国家 国 家 从量税调整依据的标准 阿根廷 通货膨胀率[消费者价格指数(CPI)],按季度计算 亚美尼亚 在2019年至2021年期间,税法中的最低从量税将平均增加15% 澳大利亚 工资———烟草和烟草制品的消费税在每年3月和9月增加,基于每周 正常平均时薪 波斯尼亚和黑塞哥维那 每1 000支卷烟的从量税每年至少增加7.50可兑换马克;最低消费税 每年增加至少达到加权平均价格的60% 加拿大 通货膨胀率———联邦烟草税率每5年提高1次,从2019年开始与加拿 大的CPI挂钩 智利 通货膨胀率 哥伦比亚 从量税定为1 400比索,2018年增至2 100比索;2019年开始按CPI逐 年增加4个百分点 哥斯达黎加 通货膨胀率 多米尼加共和国 通货膨胀率,按季度调整 法国 从2017年开始增加,到2020年达到每包10欧元的卷烟平均价格 洪都拉斯 通货膨胀率,每年根据至上一年12月的通货膨胀率进行调整 意大利 每年3月份按上年销售卷烟加权平均价格计算最低税负 新西兰 从2017年到2020年,每年通货膨胀率再加10% 尼加拉瓜 截至2017年1月1日,每年更新,取尼加拉瓜中央银行公布的科多巴 对美元的官方汇率年度贬值和国家发展信息研究所公布的在过去12 个月内观察到的CPI的年通货膨胀率的最高值 北马其顿 从每年7月1日到2023年,每支卷烟的从量和最低从量税增加0.2第 纳尔 菲律宾 2020年至2023年按已批准的方案增加从量税,此后按每年5%指数化 调整 罗马尼亚 通货膨胀率,根据上一年10月1日计算的通货膨胀对总消费税进行年 度(1月1日)调整 塞尔维亚 通货膨胀率,每6个月调整1次 52第二章 烟草消费税政策 续表 国 家 从量税调整依据的标准 南 部 非 洲 关 税 同 盟 (SACU):博茨瓦纳、埃斯瓦 蒂尼、莱索托、纳米比亚和 南非 通货膨胀率,按年计算调整 瑞典 通货膨胀率 土耳其 生产价格指数 乌克兰 准许在2019年至2025年对从量税部分每年增加20% 英国 议会批准在2015—2020年,以比零售价格指数(通胀衡量指标)高2% 的标准进行调整 资料来源:参考文献1。 关键要点9 为避免从量税受到侵蚀,各国需要定期且最好根据通货膨胀率对消费税进行自动 调整。 从量税自动调整和指数化管理对收入增长的意义 除了通货膨胀造成的税收侵蚀风险之外,如果不根据消费者收入的增加及时对税收 进行调整,那么从量税的税收效应会显著降低。消费者收入的增长会使其对商品的价格 接受程度更高,从而刺激消费,尤其是在收入快速增长的国家。澳大利亚是少数明确根据 工资增长调整从量税的国家之一(表2.3)。然而,也有一些国家采用幅度高于通货膨胀 水平的自动调整,这种自动调整的幅度有时甚至会高于收入增长(表2.3)。根据收入增 长调整烟草消费税会直接导致价格上涨,从而使得消费者对烟草制品的可负担能力下降 (第2.2.3节)。 关键要点10 从量消费税需要根据收入的增长进行调整,这样烟草制品就不会随着时间的推移 而变得越来越便宜。 针对特定情况的措施:定价和其他非税收监管的作用 新出现的证据表明,烟草业找到了减轻高税收对价格影响的方法。例如,尽管英国严 重依赖从量税,但优质卷烟和廉价卷烟之间的价格差异仍然存在。有证据表明,烟草业并 不总是将增税转嫁给更便宜的产品[3748]。欧盟[49,50]、新西兰[38]和美国[51]也观察到价格 62 世界卫生组织烟草税政策和管理技术手册 虽然SACU中的调整不是完全自动的,但它很大程度上受通货膨胀率的影响。各国有一定的自由裁量权。近年 来,涨幅通常略高于通货膨胀率。 类别之间的差异转移。因此,公共卫生界建议可以将定价监管视为消除年轻人和穷人经 常使用的廉价烟草制品的一种方法[52]。下面描述了3种定价规则:最低加价、价格下限 和价格上限。 定价监管 • 最低加价:有人认为,烟草批发价格的最低加价可能是比增加消费税更好的提高 烟草制品价格的策略。最低加价法旨在通过对供应链不同级别宣布的成本进行加价来阻 止低于评估成本价格的产品销售。美国的一些研究表明,最低加价不会增加平均卷烟价 格[53,54]。然而,最近一项关于最低加价/价格法影响的研究表明,这些法律与更高的价格 有关,尤其是对于最便宜的品牌而言,可以作为一种有效的工具来减轻行业降价促销的影 响[55]。与最低加价相关的另一个问题是,它们可能被烟草制造商操纵,并可能为该行业 带来更高的利润,以及政府的额外管理成本[56]。 • 价格下限:一些研究表明,设定价格下限或最低价格是增加烟草税的替代策略,特 别是在减少健康不平等方面[5760]。由政府施加的或由供应商对零售商施加的纵向供应 链的价格下限使得烟草制品不得随意打折出售。政府设置价格下限可以抑制不公平竞 争,对服务业而言意味着提高质量。然而,要找到合适的价格底线或预测意外后果或行业 的调整都是非常困难的。一项研究发现,马来西亚2010年实施了卷烟最低价格政策,该 政策似乎对价格没有产生预期的影响:合法品牌价格仍远高于设定的最低价格,而非法品 牌价格仍远低于设定的最低价格。这可能是下限设置过低或非法贸易比例过高的结果, 这两种情况都会降低设置最低价格政策的有效性[52]。 在欧盟,对卷烟实行最低零售价可能违反有关内部市场的立法统一,因为最低价格会 扭曲竞争。因此,建议改为提高最低消费税,以抑制消费[61]。提高最低消费税还会使政 府获得额外收入,而不是为烟草行业利润做出贡献。 设置最低价格下限可能会增加烟草行业的利润,因其为行业提供更多的营销策略资 金,并降低政府的税收,降低他们支付与烟草使用相关成本的能力。通过减少价格竞争, 设置最低价格促使烟草公司在其他方面(例如产品规格)积极争夺市场份额。企业之间的 竞争可能会阻止他们提高价格,但设定价格下限的政府会为他们做到这一点。最低定价 可能会为制造商和零售商带来暴利,甚至可以帮助制造商维持卡特尔垄断联盟。如果该 行业用这笔钱来增加相关研究的促销、广告或拨款提案,可能会抵消该政策的一些潜在好 处 。 最近的一些证据表明,至少在英国所表现出的情况是,少数跨国公司权力集中度的提 高使它们能够通过增加价格分割来破坏税收增加,而要求最低价格可能是一个解决这个 问题的好方法。对英国价格数据的纵向分析[48]表明,尽管随着时间的推移定期增加消费 税,但烟草市场中较便宜的部分(在本例中为机卷烟和手卷烟)的平均实际价格并未增加, 这表明转移不足以导致销售额增加的那些细分市场的税收增加。与此同时,价格较高的 细分市场的平均价格上涨,表明税收上调过度导致销量下降。这一行业战略确保对价格 最敏感的消费者仍然有依赖,同时开始鼓励和劝阻戒烟。此外,通过对高端品牌的税收增 72第二章 烟草消费税政策 例如,参阅PMI建立无烟世界基金会的战略和相关研究的资助提案。 加过度而对更便宜的品牌减少税收来进一步细分市场,可以减轻因为税收增加而导致的 消费下降的影响,同时还能提高行业整体利润率和营利能力。 设定最低价格可能是一项有效政策的另一种情况是美国所特有的。根据美国宪法 (修正案I)的言论自由保护法案,禁止营销和促销 是不可能的。据估计,2008年美国烟 草业82%以上的广告和促销支出都集中在减少他们的产品在销售点的价格[62]。这种对 政府如何制定政策的限制为许多州和城市实施最低价格政策铺平了道路,以应对价格促 销对消费和税收政策本身的不利影响。Huang等人[55]发现最低价格法的出现与较高的 卷烟价格有关。他们还指出,卷烟价格甚至高于各州最低价格法所规定的价格,这些法律 还禁止行业采取其他降价策略,例如低于成本的组合销售、使用贸易折扣来降低卷烟的基 本成本和向消费者分发低于成本的优惠券。 在价格细分和价格促销的背景下,实施最低价格作为消费税增加的补充政策,而不是 作为替代政策,将会有助于保证税收导致预期的消费减少。尽管如此,还需要更多的证据 来支持这项政策的有效性。 • 价格上限:对差别税收转移的担忧导致有人提出,价格上限可能会通过在不同价 格段之间差别转移税收增加来限制烟草业降低平均价格的意愿,从而有益于公共健 康[63,64]。然而,由于烟草制造商在国际市场上运营,他们可以在一个国家保持低价,但通 过在另一个国家销售更多优质产品来保持整体盈利能力。此外,限制价格上涨不符合减 少消费的公共健康目的。值得注意的是,在实行从价税制或混合税制的国家,最高零售价 有时被用作计算从价税的税基。 在对有关提高价格的非税收政策方法的文献进行系统回顾时,Golden等人[63]假设 了这些政策将如何影响价格分散和平均价格。他们的研究发现,最低价格政策与促销禁 令相结合有可能提高平均价格。当然,这与存在价格促销的情况下是相关的。然而,无论 是从理论还是实践的角度来看,价格政策显然都不能单独使用,而应始终被视为对消费税 增加的补充。大幅增加税收是劝阻消费、纠正可能存在的任何偏差的最有效方法。大幅 度的税收增加还为政府筹集资金而不是为烟草业的利润提供了额外的好处。尽管如此, 当适用于所有烟草制品以避免产品替代时,设置最低价格可能有助于缩小廉价卷烟和高 档卷烟之间的差距。 其他非税收法规 • 禁止烟草制品的促销折扣,即禁止以折扣销售烟草制品。例如通过降价券或买一 送一优惠来鼓励消费者购买并削弱税收增加。这种做法应该被完全禁止。它们通常存在 于金融部门之外,因为它们被视为一种营销,促销折扣通常在烟草广告、促销和赞助条款 下的烟草控制法中得到解决。根据2019年的RGTE[1],截至2018年12月31日,195个 不同收入水平的国家中有118个实施了这项规定。 禁止销售单支卷烟———WHO FCTC第16条“向未成年人和由未成年人销售”第3 款要求缔约方“禁止单独或小包销售卷烟,这会增加未成年人对此类产品的购买能力”。 一些吸烟者选择购买单支卷烟的部分原因是单支购买卷烟的即时成本较低[65]。世 82 世界卫生组织烟草税政策和管理技术手册 有关营销和促销的进一步讨论,请参阅下面关于禁止烟草产品促销打折的部分。 卫组织为2012年、2014年和2016年版的 WHO RGTE收集的单支卷烟价格的内部分析 (未公布)表明,实际上,单独出售20支包装的单支卷烟的总价格普遍高于整包出售的价 格。尽管如此,单支销售,以及小包装的销售使可支配收入有限的消费者能够实现消费卷 烟的目的。De Ojeda[66]在危地马拉进行的一项研究中发现,单支卷烟销售与受教育程度 较低、收入较低的人群和未成年人的卷烟可及性增加有关。单支销售也是东南亚许多国 家市场的一个特点,最显著的包括孟加拉国和印度,但也存在于世界上其他一些地区,例 如南非。 单支销售也减少了对税收增加的影响,因为每支税收的增长远小于每包的税收增 长[67]。在一项调查纽约市吸烟者如何应对2008年每包1.25美元的税收增加的研究中, Coady等人[68]发现15%的吸烟者购买的单支卷烟比以前多 。一旦允许单支销售,如果 税基是零售价,政府就有可能失去部分从价税,因为单支卷烟的零售价比一包卷烟的零售 价更难监控,例如,可以使用带有价格的印花税票。 世卫组织对195个国家2018年最新烟草控制法律的内部分析发现,86个国家通过 法律禁止销售单支卷烟(其中36%的国家是高收入国家,64%是中低收入国家)。除了禁 止销售单支卷烟外,86个国家中有67个国家规定了卷烟的最小包装尺寸。大多数国家 (52个国家)实行每包20支卷烟的标准,每包所含支数范围从加纳的5支到巴布亚新几 内亚的25支。另有15个国家没有明确禁止销售单支卷烟,但明确规定了每包卷烟的最 少支数。 根据 WHO FCTC,为了降低可负担性,应禁止单支卷烟销售,还应定义每包卷烟包 含的最少卷烟数量。 关键要点11 许多非税收措施与税收政策密切相关,包括价格监管、禁止烟草制品促销折扣和禁 止销售单支卷烟。 讨论的价格政策:①最低加价;②价格下限;③价格上限。目前的证据尚未证明最低 加价和价格下限会导致平均价格上涨。尽管如此,它们在某些特定情况下可能是相关的, 作为增加消费税的补充政策。价格上限限制了价格上涨,这可以减轻其对消费的影响。 促销打折、单支销售等价格营销策略有损税收政策效果,应予以取缔。监管机构还 应要求最小包装卷烟数量。 税收增加及其对通货膨胀的可能影响 有时,卷烟和其他烟草制品增税对通货膨胀的影响被提出作为反对这些税收政策的 论据。在工资和政府支出的很大一部分与通货膨胀挂钩(例如公共养老金支付)或政府政 策要保持低通货膨胀率的国家,这可能是一个问题。烟草制品税收增加导致通货膨胀增 92第二章 烟草消费税政策 在2018年之前,可以使用单支销售,但此后已被禁止。请参阅纽约市行政法典第7章:烟草制品管制,第1分章: 烟草制品管制法,§17 704.a 1。纽约:纽约法律出版公司;2020(http://library.amlegal.com/nxt/gateway. dl/New%20York/admin/title17health/chapter7regulationoftobaccoproducts? f=templates$fn=default.htm $3.0$vid=amlegal:newyork_ny$anc=JD_T17C00,2020年9月29日访问)。 加的程度取决于几个因素,最显著的是这些税收在价格中的份额以及计算价格指数时烟 草价格的权重。例如,如果税收占烟草制品价格的25%,那么税收增加一倍(增加100%) 将使价格上涨25%。如果价格指数中烟草制品的权重为3%,则该指数将因税收增加而 上升0.75%。由于烟草税占烟草制品价格的比重较大,因此增加税收对通货膨胀的影响 将更大。同样,由于烟草制品在计算价格指数时被赋予更大的权重,给定的税收增加将产 生更大的通货膨胀影响。一般对于大多数国家而言,烟草制品税增加对通货膨胀的影响 相对较小[47]。 消费者价格指数有多种用途,它是大多数国家的重要经济指标,通常是货币政策的关 键因素。通货膨胀率对利率和汇率有直接影响。在许多国家,工资、社会保障福利和其他 支付的变化与以价格指数衡量的通货膨胀有关。价格指数用于更准确地比较特定商品的 支出、收入和价格随时间的变化情况,并允许进行国家间的比较。 鉴于消费者价格指数的多种用途以及增加烟草税对潜在通货膨胀的影响,一些政府 制定了将烟草(有时是其他商品)排除在某些用途之外的替代方案。例如,自1992年以 来,法国已将烟草制品排除在用于调整最低工资的价格指数之外[47]。然而,许多国家继 续将烟草制品价格纳入其消费者价格指数。将烟草制品从用于制定关键价格指数的一揽 子商品中排除,将大大减少其对通货膨胀影响的担忧。此外,随着烟草制品消费量的下 降,将其价格纳入关键价格指数会导致对许多消费者的价格衡量产生扭曲。 关键要点12 如果政府担心烟草税增加对潜在通货膨胀的影响(因为工资或某些政府支出可能 与价格指数挂钩),他们可以使用不包括烟草制品的价格指数。 以类似方式对卷烟和其他烟草制品征税的重要性 虽然卷烟是全球最普遍的烟草制品,但在世界某些地区,其他烟草制品与卷烟一样普 遍,有时甚至比卷烟更普遍。比迪烟和无烟烟草是东南亚一些国家(特别是孟加拉国和印 度)消费的主要产品,而水烟则在东地中海地区广泛用于吸烟消费[4]。这些产品以及 RYO的税收历来远低于卷烟(例如,参见图2.10的孟加拉国和印度,比迪烟和无烟烟草 的消费税和价格远低于卷烟)。这种差别税收削弱了烟草制品消费税对健康影响的原因: ①它鼓励吸烟者从卷烟转向税收较低的产品(见下文泰国的案例);②总体上对减少烟草 使用无效,特别是如果该国使用最广泛的产品不是卷烟;③它会鼓励那些可能重新定义类 似于卷烟的产品的公司避税,使它们被纳入低税产品类别(见下面的欧盟示例);④它减少 了政府原本收入的规模,因为对那些产品本来可以征收更高的税。 例如,在泰国,多年来通过征税成功地提高了卷烟的价格,而散装或RYO烟草的税 收和价格直到最近还保持不变。用于RYO卷烟的本土烟草历来免征消费税,而外国烟 草相对于卷烟的税率较低。因此,尽管卷烟消费量一直在下降,泰国的RYO市场仍在增 长 。泰国政府最终采取了强有力的政策来解决这个问题。首先,2018年取消了对本地 03 世界卫生组织烟草税政策和管理技术手册 世界卫生组织泰国代表处,个人通讯,2019年。 烟草的豁免;政府部门随后批准将(本地烟草)小型生产商的消费税从每克0.005泰铢提 高到2020年的每克0.025泰铢,2021年预计将再次增至每克0.1泰铢[69]。 消费税额价格 200 150 100 50 0 本 国 货 币 孟加拉国/塔卡 印度/卢比 卷烟 （20 支） 比迪烟 （20 支） 无烟烟草 （20 克） 卷烟比迪烟 （20 支） 无烟烟草 （20 支） 80 2412.8 64 54.9 44 12.8 3.8 0 190 图2.10 2018年,在孟加拉国和印度,最畅销的卷烟、比迪斯和无烟烟草品牌的价格和税收 来源:参考文献1。 在欧盟,雪茄和小雪茄的最低消费税水平明显低于卷烟。欧盟成员国必须对每1 000 支卷烟征收至少90欧元的消费税,该税应为用于消费的卷烟加权平均零售价的60%。 对于雪茄和小雪茄,每1 000件商品只需支付12欧元,或零售价的5%的消费税。因此, 在许多欧盟国家,卷烟的消费税份额远高于雪茄和小雪茄的份额。作为回应,一些公司开 始销售所谓的边缘小雪茄。这些产品具有与卷烟相似的特性,但可以以较低的价格出售, 出于消费目的,它们被视为小雪茄。虽然这个问题似乎在很大程度上通过欧盟层面对这 些产品的定义的修正和一些国家税收结构的变化得到了解决,但重要的是要意识到产品 类别之间消费税水平的巨大差距可能会产生的意外激励措施[70,71]。有关阻碍增税的行 业策略的更多详细信息请参见方框2.1。 关键要点13 为了使烟草制品的消费税能更有效地减少总体烟草使用并避免不同产品之间的消 费替代,应对所有烟草制品以同等方式征税。 世卫组织烟草控制框架公约第6条(减少烟草需求的价格和税收措施)的实施指 南[73]建议对所有烟草制品都应以同等方式征税。 13第二章 烟草消费税政策 方框2.1 烟草行业破坏增税的策略 税收增加减少了对烟草制品的需求,并对烟草业高额利润构成威胁。烟草业通 过使用各种策略[17,46,48]应对提税,具体包括[46]: 囤货(预售/前期装载):在政府宣布增税实施之前,烟草制造商超额生产烟草制 品,并以增税前的税率缴税。结果,销量和税收收入在增税后立即阶段性下降(而销 量和税收收入在增税之前大幅增加),烟草行业将这种收入下降归因于(由提税引起 的)非法贸易的出现或增加。如果没有法律对此明文禁止,则这种做法会导致避税 (另见第3章关于反囤货的讨论)。 改变某些产品特性(例如重量或长度)和/或调整生产过程:当烟草制品以不同的 税率征税或受到不同的增税时,行业可能会重新推出一种税负较低的烟草产品(如上 面欧盟的例子)。 战略性地选择涨价公告的时间:烟草行业可能会在预计政府提税前先行提高价 格,从而在提税实施之前的这段时间内产生高额的利润。加税实施后,烟草的消费和 税收会下降,但价格仍保持不变,如此烟草业会宣称提税对减少烟草需求无效。 采用价格歧视策略或与价格相关的促销活动:烟草行业可能会为烟草购买提供 打折、零售商折扣或增值(礼品),以尽量减少对价格敏感的消费者的损失。然而,这 在严格禁止烟草广告、促销和赞助的国家是不可能的。 使用品牌扩散(例如,推出低价品牌)和价格细分:烟草制造商可以选择降低某些 品牌的价格或推出新的甚至更便宜的品牌,以留住市场上对价格敏感的消费者。有 证据表明,公司推出新的更便宜的产品并使用价格标记,比如直接在烟草制品包装上 印刷价格来锁定价格[48],这样的做法会损害公共卫生和收入目标。 根据市场情况,对不同价位的产品实施差异税收转嫁:烟草行业可能会将产品价 格增加超过税收增加的金额(税收超额转嫁),并将价格上涨归咎于政府提税。当需 求缺乏弹性时,即当价格上涨超过抵消销量减少时,税收超额转嫁是有利可图的。烟 草行业通常会针对高价品牌实施超额转嫁,因为高价品牌比低价品牌更缺乏弹性。 此外,为了留住市场上对价格敏感的消费者,烟草行业还可能会针对低价品牌暂时自 己消化部分增加的税收。差异税收转嫁策略将导致市场对不同品牌价位的产品做出 不同的反应[37,48]。 游说政府扭曲干预措施:政府政策可能直接或间接受到烟草业游说的影响。政 策制定者关注的不仅仅是社会福利或税收收入的最大化;他们也很在意政治支持。 烟草行业会游说政策制定者采用有利于烟草业的税收类型,推迟提税或降低税 率[17]。世界卫生组织烟草控制框架公约第5.3条“关于保护与烟草控制有关的公共 卫生政策免受烟草业行业和其他既得利益方的影响”及其指南为如何解决烟草业干 扰问题提供了有用的指导。事实上,作为 WHO FCTC缔约方的所有181个国家都 有执行第5.3条要求的法律义务。 对烟草行业反应的正确预期对于评估增税对消费和税收的影响很重要。 23 世界卫生组织烟草税政策和管理技术手册 2.2.3　衡量增税影响的推荐指标 政府在制定政策时需要考虑一系列变化的指标。对消费者行为、市场结构和行业行 为的不恰当假设可能会导致错误的政策分析。 衡量增税对价格和需求的影响 消费习惯、当地传统和行业特征(例如提供的不同品牌的数量、跨境购物的可能性以 及非法贸易的存在和水平)都会影响烟草制品的供需形态,从而决定价格弹性的价值。价 格弹性,连同行业的定价策略(例如,税收转移的程度)以及零售价格中的税收份额,决定 了税基的弹性,无论基数是否由数量决定(对于从量税)或交易价值决定(对于从价税)。 弹性估计的重要性 应考虑到不同类型的弹性: • 价格需求弹性 ﹣自身价格弹性:衡量消费者对产品价格变化后对产品需求的反应。 ﹣交叉价格弹性:衡量消费者对一种产品的需求在另一种产品的价格发生变化时的 反应。同一产品的不同品牌或价格段之间也可能出现交叉价格弹性。 • 收入需求弹性:消费者在收入水平发生变化时对产品需求的反应。 对价格和收入弹性的正确估计对于预测增税对消费和税收收入的影响的决策者很重 要。弹性估计值会受多种因素影响,包括是否考虑短期与长期响应、所用需求函数的函数 形式、是否考虑了成瘾或逃税等因素以及数据的收集方式。例如,诸如数据聚合程度、是 否使用特定性别或特定年龄数据、涵盖的时间跨度以及使用哪些估计程序(例如普通最小 二乘法、两阶段最小二乘法或广义矩量法)等细节都会影响估计[72]的结果。 价格弹性可能会随着时间的推移而变化,也可能因为影响需求的任何其他因素发生 变化而变化,例如收入或烟草控制措施,以及估计技术和所用数据类型或来源的变化。 此外,最令人感兴趣的是总需求的价格弹性。增加税收可能会减少纳税零售额,但不 一定会减少总消费。例如,当使用合法销售数据估计弹性时,走私会显著影响价格弹性; 不考虑可能的非法贸易可能会导致高估弹性。同样,当包括跨境购物时,需求的价格弹性 较低(绝对值)[74]。可以通过使用具有全国代表性的住户调查的横截面数据来估计合法 和非法消费需求的总价格弹性。然而,这种方法也有它的缺点。例如,受访者倾向于少报 他们的烟草消费量,从而导致需求规模存在偏差。价格内生性 是另一个棘手的技术问 题。 为了全面估计增税对所有烟草制品的需求以及税收的总体影响,需要估计它们之间 的可替代程度[55]。交叉价格弹性衡量当一种烟草制品的价格上涨时,特定烟草制品的需 求量如何变化。当这种弹性为正时,产品是替代品;弹性值越高,产品之间的替代品就越 接近。例如,手卷烟(RYO)和成品卷烟之间的正交叉价格弹性意味着对手卷烟的需求随 着卷烟价格的上涨而增加。不同卷烟品牌之间也可能出现可替代性:当经济品牌的相对 33第二章 烟草消费税政策 价格是内生的,因为它不是一个自变量:它是通过将烟草支出除以烟草消费来估计的,消费是估计价格弹性的因 变量。 价格上涨时,对高档品牌的需求可能会增加。当对不同类型的卷烟实施差别(分层)税收 时,这种影响会加剧,将进一步扩大品牌和细分市场之间的价格差距并鼓励替代。传统与 新型烟草和尼古丁产品之间的可替代性目前备受关注(见下文第2.4节)。在一些国家, 不同的烟草制品也可以是互补品而不是替代品。这意味着当烟草制品价格上涨时,对其 补充品的需求下降,因为使用者不太可能单独使用补充烟草制品。例如,一些研究发现, 在印度,工业制造的卷烟和本土的比迪烟是互补品[75,76]。 收入弹性的符号和大小因时间、国家和人口群体而异。例如,在高收入国家美国,随 着时间的推移,收入弹性从正变为负,卷烟从正常商品变为劣等商品[77,78]。另一方面,在 中低收入国家中,吸烟率往往相对较高,但卷烟可能仍然是一种正常商品,消费随着收入 水平的增加而增加(正收入弹性)[79,80]。 有一些无法观察到的特征将高收入吸烟者与低收入吸烟者区分开来。例如,时间和 风险偏好的差异、将社会污名与吸烟联系起来的差异以及将吸烟作为一种愉快活动的品 位差异。当这些特征被忽略时,收入与吸烟相关结果之间相关性的估计就会出现偏差。 Kenkel等人[81]使用估计收入对低收入成年人吸烟因果影响的技术,发现烟草是一种正 常(甚至是奢侈品)商品:较高的收入和较高的吸烟率与较低的戒烟率相关。 这些结果与商业周期(经济活动的扩张或衰退时期)对健康行为和结果的影响一致。 例如,Ruhm[82,83]发现,在暂时的经济衰退期间吸烟者会减少,而在经济扩张期间吸烟者 会增加。Tarantilis等人[84]发现,在2010年经济危机之后,希腊需求收入弹性的估计值 高于之前。金融危机和紧缩措施降低了对卷烟的需求,使卷烟成为收入弹性更大的商品。 有趣的是,来自德国的证据表明,在经济衰退期间,吸烟的倾向会显著增加。然而,在那些 已经吸烟的人中,卷烟消费实际上减少了[85]。 理想情况下,在估计价格和收入弹性时,还应考虑非价格政策的影响。南非最近的一 项研究表明,不考虑非价格政策会夸大价格效应[86]。 NCI/WHO专著[4]表明,高收入国家的烟草需求价格弹性平均为-0.4(范围 是-0.2~-0.6)。对中低收入国家的估计变化更大,集中在-0.5左右(范围是-0.2~ -0.8)。-0.5的价格弹性意味着价格上涨10%将导致消费减少5%。 关键要点14 政策制定者需要了解本国烟草制品的需求弹性,包括价格弹性(自身价格弹性和交 叉价格弹性)和收入弹性,以便正确评估潜在政策变化对消费和后续收入的影响。这些 估计需要定期进行,以捕捉需求随时间的变化。 税基弹性的重要性 税基弹性衡量征税税基对税率变化的敏感性:从量税的情况下税基为烟草消费量,从 价税的情况下税基为烟草支出量。税基弹性的大小取决于需求的价格弹性、税收结构、税 率水平和其在价格中的份额,以及行业对其决定吸收、传递或将税收转移到零售价格的反 应。消费者的偏好和收入、替代品的可用性和其他非烟草价格控制措施也影响税基弹性, 主要是影响需求的价格弹性。 43 世界卫生组织烟草税政策和管理技术手册 税基弹性的大小还取决于社会动机,包括价格和税收预期,这些动机最终会受到消费 者纳税意愿或成功的烟草价格控制措施的影响。此外,税基弹性取决于吸烟者对使用非 法产品时被发现和被征税的可能性的看法,以及逃税和避税机会的可用性和可及性。最 后,消费者的纳税意愿取决于他们对税收用途的看法[87]。因此,税基弹性在很大程度上 受政府政策选择的影响。 世卫组织建议将增加价格中的税收占比作为实现减少烟草使用的公共卫生目标的一 种工具:当其他条件保持不变时,价格中较高的税收占比会增加税基弹性,并进而减少吸 烟。然而,制造商可能会试图通过他们的定价策略来操纵税基弹性,例如税收转移。正如 本章前面所讨论的,烟草行业行为本身受政府税收政策和法规的影响。 在考虑税收转移时,需要考虑许多因素。如前所述,与从价税结构相比,从量税收结 构中的税收更有可能过度转移。还有证据表明,行业对高档或昂贵卷烟品牌的税收转移 过度,而对廉价品牌的税收转移不足。这表明,在特定市场内,行业对税收转移程度的确 定不仅会因税收结构而异,还会因市场结构而异,也会因品牌而异。但这并没有说明增税 对烟草制品平均价格的影响。在税基弹性和税收增加对收入产生影响的背景下,评估税 收增加如何影响平均价格很重要。 南非的例子在这里非常有用。在过去的二十年里,南非一直在增加对卷烟的从量消 费税,这导致卷烟价格大幅上涨。就2001年至2015年消费税增加对卷烟价格影响的分 析[26]表明,虽然有证据表明税收过度转移,至少在特定时期内,总体而言,税收实际上转 移不足。这种转变不足的部分原因是市场竞争力增强,部分原因是引入了低价品牌。当 然,这鼓励了一些消费者将消费转向更便宜的产品,但也促使更昂贵品牌的制造商吸收部 分增税,以减少对价格的影响。 税率水平的变化:在影响消费的所有其他因素保持不变的情况下,将导致税收收入的 变化 。税基弹性的估计有助于政府预测税收增加后烟草税收的相应变化(详见附件2.2)。 实行从量税: • 烟草消费(税基)预计价格无弹性[17,47]; • 平均而言,价格上涨幅度小于税收上涨幅度(总体上没有税收过度转移); • 消费(税基)也预计是无税收弹性的:消费量的下降与税收增加的比例不符,税收 增加。 实行从价税: • 税基是(合法)烟草消费的总消费者支出(或等价的行业销售收入),也就是说,从 价税下的税基由价格和数量决定,其本身是价格的函数; • 税基弹性的符号可以是负的也可以是正的,其大小取决于需求价格弹性的大小。 • 因为有证据表明烟草需求是无价格弹性的,所以税基弹性是正的。 • 当从价税率提高时,价格和数量都会调整,但数量下降的幅度小于价格上涨的幅 度,税收增加。 53第二章 烟草消费税政策 Laffer曾用这个概念来论证过高的税收增加会减少消费税收入(所谓的拉弗曲线)。有关拉弗曲线的详细讨论, 请参阅第4章中的第4.4节。 • 提高税率会导致更高的收入水平和更低的消费水平;弹性的价值,以及税收收入 会随着税收转移的程度而增加。 税收是实现财政目标和公共卫生目标的工具。如果在成功的烟草控制干预之后,价 格达到需求变得有弹性的水平,那么税基仍然很可能由于税收转移不足而缺乏弹性,因为 当需求有弹性时过度转移并不是一个好的定价策略。换句话说,提高税率,结合非烟草价 格控制措施使消费者对价格(税收)增长更加敏感,会导致边际收入减速但仍为正。有关 烟草税成功案例的示例,请参见方框2.2。有关各国增税经验及其对收入影响的更多详 细信息,请参见第4章第4.4节。 方框2.2 烟草税成功案例:土耳其 土耳其就是一个例子,该国在相对较短的时间内定期显著增加了税收,并从该政 策中获益。如图2.11所示,每包卷烟的消费税在10年内实际增加了一倍以上,价格 也几乎翻了一番。与此同时,烟草消费收入增长了67%,卷烟销售额下降了20%。 自该国2003年开始实施健康转型计划以来,土耳其成功地增加了公共卫生支出并征 收了更多的烟草税收。根据最新数据,2015年,土耳其烟草税收相当于该国公共卫 生支出的42%和GDP的1.5%[88]。 关键要点15 政策制定者控制需求的关键政策工具是税收。因此,他们不仅要评估价格对需求 的影响,还要更恰当地评估税收对需求的影响:这就是税基弹性。税基弹性主要取决 于:①需求的价格弹性;②行业将税收转嫁给零售价格的程度;③作为零售价格份额的 税收。这些元素受税收增加影响的程度取决于需求和收益。 目前,这三个组成部分的总和在任何国家都不够高,不足以通过增加税收导致消费 税收入减少。 对负担能力的影响 虽然价格上涨显然对消费有影响,但如果不考虑人均收入增长的影响,价格影响可能 没有预期的那么强烈。人口收入的增加也会增加其购买力。而且,如前所述,烟草制品通 常表现得像正常商品。因此,随着收入的增加,预计烟草消费量也会增加。为了减轻这种 影响,价格上涨(在税收增加之后)需要大于收入的增加。这就是负担能力概念的由来。 负担能力研究价格上涨和收入增加对消费者行为的影响。 由Blecher和van Walbeek(89)推广的一种常见且简单的计算负担能力的方法是使 用特定年份购买2 000支卷烟(或100包20支卷烟)所需的人均 GDP百分比。随着时间 的推移,这一比例的增加将表明卷烟变得越来越难以负担,应该会导致消费量减少。随着 时间的推移,卷烟可负担性趋势的变化有助于政策制定者了解价格相对于人口购买卷烟 的能力如何演变,并提醒他们能够相应地修改政策。例如,最近在印度进行的研究强调, 卷烟、比迪烟和嚼烟等本土烟草制品之间存在巨大的价格差异,以及由于优惠或更宽松的 税收政策,这些本土产品随着时间的推移对他们有变得更加负担得起的倾向[90,91]。 63 世界卫生组织烟草税政策和管理技术手册 每 包 土 耳 其 里 拉 6 5 4 3 2 1 0 10 000 000 支 卷 烟 土 耳 其 里 拉 1 100 1 700 500 2 300 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 实际消费税额 实际价格 实际烟草消费税收入 人均卷烟消费量 图2.11 土耳其烟草消费税收入和消费量(2008年基准年),2008—2018年 资料来源:最畅销品牌价格参考文献1,财政部销售收入数据和IMF世界经济展望,2020年4月。通货膨胀调 整参见https://www.imf.org/en/Publications/WEO/weo-database/2020/April. 图2.12显示了2008—2018年按国家收入组划分的一包最畅销品牌卷烟的可负担性 变化。在此期间,近70%的高收入国家的可负担性下降,而超过35%的中等收入国家和 仅26%的低收入国家的可负担性略有下降。 卷烟可负担性越来越强卷烟可负担性越来越差 由于数据不足，无法评估卷烟负担能力没有改变 高收入国家 中等收入国家 低收入国家 9 13 75 36 37 23 6 39 13 2 5 图2.12 2008—2018年,按收入水平计算,卷烟可负担性发生变化的国家数量 注:可负担性的变化计算为在给定年份以当地货币购买2 000支最畅销品牌卷烟所需的人均GDP的最小二乘 变化率。计算了具有4年或更多年数据(包括2018年)的国家的趋势增长率。如果2008—2018年购买2 000 支卷烟所需的人均GDP的最小二乘趋势在5%水平上不显著,则被评估为负担能力没有变化。 资料来源:参考文献1。 73第二章 烟草消费税政策 关键要点16 从健康的角度来看,除了检查加税对价格、需求和收入水平的影响外,政策制定者 也应考虑加税导致价格上涨是否超过其人口收入的增加;增加税收应该使消费者负担 不起烟草制品,从而有效减少需求。 预测税收对消费、吸烟流行率和挽救生命的影响 WHO交互式吸烟预测和目标设定工具(WHO ISPT): WHO ISPT使国家政策制定者和烟草控制专家能够探索拟议烟草控制政策的潜在 影响。它使用 WHO FCTC选定的减少需求措施的影响因素,这些措施来自已发表的文 献、吸烟率趋势、国家人口信息和烟草相关死亡风险。WHO ISPT提供了一个国家在不 同政策环境下和不同时间段内的吸烟率[1]和与吸烟相关的死亡人数的预测[2]。它旨在通 过使来自各个部门(例如卫生、教育、财政、国家统计)、民间组织、学术界和媒体的专家能 够与世卫组织专家一起探索中长期烟草控制规划的备选方案,来促进国家内部的多部门 合作。世卫组织ISPT的使用为政策变革倡导、规划制定和评估提供了强有力的支持 。 特别是它可以帮助财政部门的决策者评估税收政策在总体烟草控制政策中实现减少烟草 流行的特定目标的具体贡献。 预测对消费税收入的影响 WHO烟草税模拟模型(WHO TaXSiM)是一个简单但数据密集的Excel工具,可帮 助政策制定者分析其烟草税政策,并评估任何消费税增加或消费税结构变化对价格水平、 合法销售、消费税收入以及烟草制品其他税收的影响。使用有关市场的详细数据,包括市 场上发现的大多数品牌、它们的市场份额和价格水平及适用的税收,以及关于需求价格弹 性的假设。WHO TaXSiM预测了税收变化对每个品牌和市场部门的消费者价格、消费 量和所产生的税收收入的影响。该预测可以进行多个年份的分析 。 通过详细探索市场数据,除了评估消费税变化对收入的潜在影响外,WHO TaXSiM 还是一种有用的工具,可以突出现有税收体系和市场中的缺点和机会。它还可以鼓励政 策制定者创建可以定期更新的管理数据库,以监测卷烟市场的动态。 关键要点17 政策制定者可以使用现有工具来模拟增税对价格、消费和收入以及吸烟率和挽救 生命的影响。 监测烟草税进展的推荐指标 MPOWER:世卫组织每两年发布一次RGTE,监测全球烟草控制进展。特别是该报 告侧重于 MPOWER一揽子政策的实施,这是一套行之有效的减少需求措施,符合 WHO FCTC[1]的关键规定。 83 世界卫生组织烟草税政策和管理技术手册 WHO ISPT不公开,但 WHO将根据请求直接与感兴趣的国家合作,使用它来制定数据到行动类型的计划。 有关该方法的更多信息,请参见https://www.who.int/tobacco/economics/tax-sim_background.pdf,2020年9 月29日访问。 虽然提高烟草税(R部分)被证明是减少烟草使用的最有效和最具成本效益的政 策[4],但在最佳实践层面实施整个 MPOWER一揽子计划将加强R的影响。例如,本章 前面提到,作为 E 措施(强制禁止烟草广告、促销和赞助)的一部分,禁止促销打折将有利 于增加税收后的价格上涨。如果除 R 之外的所有 MPOWER 烟草控制措施都在最佳实 践水平上实施,所有其他项保持不变,预计收入将下降。因此,为了维持收入水平,重要的 是定期提高烟草制品的消费税,以弥补其他四项烟草控制措施造成的烟草使用量下降带 来的收入下降。 税收份额 RGTE[1]中 R 政策的主要指标是间接税在最畅销品牌卷烟零售价格中的总份额 。 最畅销卷烟品牌的税收总额等于或超过零售价格75%的国家被认为是最高水平。 虽然总税款包括消费税、增值税(或销售税)、进口关税(如适用)和其他间接税(如适 用),但最好将重点放在消费税上,因为消费税是对烟草相对价格影响最大的组成部分。 消费税在零售价格中的份额可以从 RGTE 数据库中提取 。2010年《世卫组织烟草税收 管理技术手册》中建议应使消费税至少占烟草制品零售价格的70%[47]。 可负担性 如前所述,零售价格中的税收份额不足以确保政策在减少需求方面取得成功。任何 增税都会导致价格上涨,从而有效抑制消费。虽然全球趋势表明高税收份额与高价格水 平正相关(参见第 2.1.2 节中的图 2.2),但这不一定适用于特定国家;税收份额可能很 高,同时消费者对烟草制品仍然可以负担得起。出于这个原因,重要的是不仅要监控税收 增加,还要监控这些增加是否导致价格上涨大于收入增加。如本章第 2.2.3 节所述,一 个通用指标是在特定年份购买 100 包 20 支卷烟所需的人均 GDP 百分比 。 其他指标 正如第2.2.1和2.2.2节中详细讨论的那样,良好的税收结构可以使税收政策更有 效地提高烟草制品的价格和降低消费者的负担能力。指标可以包括是否适用统一消费 税、是否为从量税以及是否根据通货膨胀定期调整。许多此类指标也通过RGTE进行监 测,并可在线下载 。 2020年编制的烟草税指标结合了构成良好烟草税政策的各种要素。烟草税指标学 卷烟税记分卡[92]根据最佳实践对一个国家的烟草税收政策绩效进行评级。决定绩效水 平的四个组成部分:①卷烟价格(以购买力平价计);②卷烟可负担性随时间的变化;③税 收(总额和消费税)在零售卷烟价格中的份额;④卷烟税的结构(是否征收消费税;是统一 93第二章 烟草消费税政策 有关如何编制该指标的更多详细信息,请参阅 RGTE 2019 的技术说明 III(https://www.who.int/tobacco/ global_report/Technical-Note-III.pdf? ua=1)。 参见世卫组织 2019 年全球烟草流行报告(https://www.who.int/tobacco/global_report/Table-9.1-Taxes-and- retail-price-for-a-pack-of-20-cigarette-most-sold-brand.xls? ua=1,于2020年9月29日访问)。 该指标也已编入 RGTE;参见https://www.who.int/tobacco/global_report/Table-9.6-Affordability.xls? ua= 1,2020 年 9 月 29 日访问。 参见 世 卫 组 织 2019 年 全 球 烟 草 流 行 报 告 (https://www.who.int/tobacco/global_report/Table-9.5- Supplementary-information-on-taxation.xls? ua=1,2020年9月29日访问)。 的还是分层的;消费税是具体的、从价的还是混合的;对于从价部分,是否对零售价格征 税;有最低从量税,对于特定组成部分,如果税收自动向上调整)。四个组成部分中的每一 个都有一个分值,使用5分制,总分反映四个组成部分分数的平均值。总分越接近 5,说 明特定国家的烟草税收政策表现越好。尽管此已发布的记分卡目前仅适用于卷烟,但只 要具备所需数据,就可以轻松应用于其他烟草制品。 关键要点18 如果将烟草税作为综合 MPOWER 包的一部分实施,则效果最佳。MPOWER 是 一个综合指标,包含了所有关键的需求方烟草控制措施。 关键要点19 除了评估增税的潜在影响外,政策制定者还需要监测随着时间的推移进展情况。 税收占零售价格的比重是衡量烟草税进展程度的一个指标。然而,重要的是要记住,有 效的增税必须转化为更高的价格,以降低消费者对烟草制品的负担能力。将良好税收 政策的所有组成部分合并到一张记分卡中,也有助于从整体上评估烟草税收政策。 2.3　国内和区域政策一体化 虽然设计烟草税政策时必须充分考虑上述所有因素,但考虑外部因素如何影响甚至 阻碍公共卫生政策目标也很重要。正如第3章所解释的,直接参与税收征管、征收和执法 的各个机构之间的合作对于有效和高效地实施税收政策非常重要。但在设计阶段,与不 直接参与税收的机构和其他政策制定者的沟通也很重要。在国内,需要协调以确保非卫 生部门的政策不会对烟草控制政策产生负面影响甚至抵消。对属于区域集团的国家而 言,统一烟草税收对于保护单一市场、人口健康以及防止税收流失、避税和逃税至关重要。 2.3.1　国内政策跨部门合作 农业、工业、贸易、金融和劳动力方面的国内政策有可能在烟草生产、制造和分销的不 同阶段创造或支持对烟草控制和税收目标产生反作用的激励措施。例如,向参与种植或 加工烟草的农民或制造商提供补贴可以降低价格并激励其继续参与甚至促进这些领域的 发展,这与降低烟草制品的可负担性和减少烟草消费的目标背道而驰。国家层面需要多 部门整合和政策一致性,以确保非卫生部门的公共政策和干预措施不会违背烟草控制和 税收对预期公共卫生的影响。 关键要点20 应促进农业、工业、贸易、金融和劳动力方面的更大政策一致性,以确保这些部门的 公共政策和干预措施不会抵消公众烟草控制和税收对健康的预期影响。 04 世界卫生组织烟草税政策和管理技术手册 2.3.2　区域烟草税的协调 政策一体化的驱动力是认识到国内政策合作可以大大增加区域集团的收益。协调是 可取的,并且在某些具有溢出效应的领域可能是必要的,例如税收政策、所谓的“逐底竞 争”的可能性或对公共健康造成威胁。可以简单地根据全球规范和最佳做法协调制定最 低标准和要求。 需要统一烟草税收,以确保单一市场的建立和正常运作,防止税收流失,防止避税和 逃税并保护人民。当国家之间的贸易壁垒被消除时,统一税率将支持单一市场,因为它们 提高了消费者、生产者和投资者做出不会被税收扭曲但反映真实的机会成本的决策能力, 反映真实的机会成本。各国税收竞争只是简单地降低彼此的税率,可能会阻止政府筹集 足够的资金来推行社会政策。为了避免这种逐底竞争,各国可以在关税同盟内制定最低 税率[93]。即使不存在税收竞争,当邻国之间存在巨大的税收差异时,跨国贸易也有明显 的动机,以便合法或非法地减少纳税。 已建立的区域经济共同体的经验提供了重要的政策教训,不仅在总体一体化进程方 面,而且在税收政策协调的进程和范围方面。 欧盟实施了一项成功的区域税收协调计划。多年来,统一烟草税的重点已从消除税 收障碍扩大到打击有害的税收竞争、避税和逃税,以及最近的公共卫生保护。当然,解决 这些问题需要加强财政政策和税收管理的融合。 尽管价格差异仍然存在,但设定烟草制品最终价格中的最低税收份额以及最低消费 税有助于各国达到某种程度的协调。欧盟的经验也证实,通过统一的最低消费税率可以 实现烟草消费趋势的下降和稳定的收入[94]。此外,协调过程为执法部门(税务或海关)提 供了获取可用于打击欺诈和逃税的信息的机会。 另一方面,西非经济和货币联盟(WAEMU) 协调的经验表明,没有超国家机构(如欧 盟)或霸权成员国(见下面的 SACU 示例)将延缓有利于所有成员国[95]的政策一体化。 WAEMU 的8个国家受税收指令 的约束,该指令要求它们对烟草制品的 CIF 价值或生 产者价格征收从价消费税,这会导致申报不足且难以确定。此外,还规定了最高消费税, 一些成员国会征收额外的税来应对这一限制。该指令于 2017 年修订[96],但不幸的是税 收结构保持不变,最高税率并未取消,而是增加了。 拥有5个成员国的南部非洲关税同盟(SACU) 是现存最古老的关税同盟,成立于 1910 年。得益于南非的领导地位,南非拥有完善的管理体系和激进的烟草税收政策, SACU 采取了使其所有成员受益的综合税收政策[95]。 海湾合作委员会成立于 1981 年,是一个由波斯湾6个国家组成的区域性政府间政治 和经济联盟。海湾合作委员会拥有全球1/5的石油供应[97],它从未将税收作为收入来 源。该地区不适用直接税或间接税。虽然烟草制品没有消费税,但作为关税同盟,海湾合 作委员会国家有共同的对外关税。这种共同关税包括统一费率,但也包括统一结构。进 14第二章 烟草消费税政策 1998 年 12 月 22 日通过了关于统一成员国消费税立法的第 03/98/CM/WAEMU 号指令。它由 2009 年 3 月 27 日的第 03/2009/CM/WAEMU 号指令修订,目的是统一 WAEMU 内的消费税。 口关税为该地区进口烟草制品 CIF 价值的 100%,按进口数量征收最低税。 然而,近年来,为减少对石油收入的依赖,海湾合作委员会国家已考虑使收入来源多 样化,包括发展对消费税和增值税等间接税的征收。2015 年,第 36 届 GCC 峰会通过了 一项决议,对所有进口烟草制品和在国内种植的种植烟草实行选择性征税(GCC 决定编 号 963/1)。2016 年 12 月的后续决议正式同意在所有海湾合作委员会国家对烟草和其 他产品(如含糖和能量饮料)以及特殊商品(酒精和猪肉)征收消费税。 该次区域决定在所有海湾合作委员会国家逐渐生效,首先是沙特阿拉伯,于 2017 年 6 月开始实施;其次是巴林和阿拉伯联合酋长国(阿联酋)在同年晚些时候实施;卡塔尔于 2018 年实施、阿曼于 2019 年[1]实施;只有科威特尚未相应地调整其国家法律。海湾合作 委员会国家引入的消费税结构有点类似于烟草制品的进口关税:税率为 100%,但基数从 CIF 价格改为不含税的零售价。2016 年至 2018 年,消费税的引入导致成员国最畅销品 牌卷烟的价格大幅上涨:巴林上涨了 33%,阿联酋上涨了 80% 以上,沙特阿拉伯上涨了 100% 以上[1]。 加拿大和美国等联邦制国家(中央政府拥有真正的税收权力并对州或省有一定的金 融监管控制)烟草税没有统一[98,99]。尽管在税收和价格方面存在显著的跨辖区差异,并 且税收协调在缩小烟草市场非法交易范围方面具有巨大潜力,但几乎没有证据表明加拿 大各省或美国的个别州对烟草税收协调感兴趣。 在已经成为关税同盟一部分的一组国家内进一步经济一体化的背景下,税收协调最 为相关,但需要精心地规划才能有效。法律解释上的差异以及烟草制品定义和税基缺乏 标准化导致次优状况。税率调整,或设定最低税率,应该在税收结构调整之后进行。重要 的是,政府支持朝着统一方向迈进,并承诺投入足够的财政资源和技术人员来监督整个过程。 关键要点21 在区域经济一体化和正在进行的关于在成员国之间协调烟草消费税可能性的讨论 背景下,现有集团的经验可能具有指导意义。迄今为止,只有欧盟、SACU、WAEMU 以及最近的 GCC 有效地实施了统一的烟草制品消费税政策。吸取的教训表明,应该 很好地规划协调,不应以牺牲烟草控制为代价。 设置一个共同的最低从量税,随着时间的推移进行调整是最好的方法。这确保税 收和价格高于最低水平,鼓励价格水平的均等化,同时降低各国的负担能力。另一方 面,同意最高税率是一个糟糕的政策。 希望出于收入目的、健康问题或两者兼而有之而进一步提高税收的国家应该有这 样做的空间。 对由更高税率引发的非法贸易的担忧,最好通过在行政和执法、信息共享和采用共 同的或可交互的信息系统的新技术方面进行有效合作,以更好地解决由更高税率引发 的对非法贸易的担忧。 24 世界卫生组织烟草税政策和管理技术手册 2.4　新型尼古丁和烟草制品 近年来,对烟草风险和危害的认识、烟草控制规定的实施(尤其是在世卫组织框架公 约下)以及法规收紧导致的卷烟销量下降,主要表现在高收入经济体,这改变了烟草市场 的动态。为了应对这些有效的烟草控制措施,烟草业通过推广新的产品组合来实现业务 多元化,他们声称这些产品是技术创新,可以减少传统的烟草制品,特别是与卷烟相关的 危害和风险。 所谓的新型烟草制品已被烟草业推广为“更清洁的替代品”、“更安全的替代品”和“无 烟无灰的危害/风险降低的产品”。在这些主张的基础上,他们就国家内部限制较少的监 管环境进行了谈判。一些新产品也被营销或推广用于戒烟,尽管这一结果的证据尚无定 论。在这些产品没有被禁止的情况下,全球卫生界的争论之一是关于它们的监管和税收 问题。 2.4.1　加热烟草制品（HTPs） HTPs是烟草制品,在加热烟草或激活装有烟草的装置时,会产生含有尼古丁和有毒 化学物质的气雾剂(或气溶胶)。这些气溶胶被用户直接吸入或借助吸烟设备吸入。它们 含有极易成瘾的物质尼古丁(存在于烟草中)以及非烟草添加剂,并且通常带有香味。 HTPs中的烟草可能是专门设计的卷烟(例如所谓的加热棒或 Neo 棒)烟弹或塞子。 这些产品包括 PMI 的 IQOS、日本烟草国际(JTI) 的Ploom TECH、英美烟草(BAT) 的 glo 和 PAX Labs 的 PAX。HTPs不仅不同于传统卷烟,还不同于 ENDS,其中的一些被 称为电子烟,因为 ENDS 不含烟草,而是一种尼古丁溶液(见下一小节)。然而,由于含有 尼古丁溶液和烟草的所谓混合烟草制品的出现,不同产品之间的界限变得越来越难以界定。 HTPs目前在 40 多个国家/地区可以使用,在不到 10 个国家/地区被禁止。即使在 它们受到监管的国家,监管方法也存在显著差异。多种因素影响一个国家控制和规范 HTPs 使用的效果,包括国家监管权力、执法能力监管框架、国家能力和烟草业干预[1]。 大多数国家在应用从量税或混合消费税时以低于卷烟的税率对 HTPs 征税,并以烟 草千克为基数(见表 2.4)。使用这样的基数对税收征收可能非常具有挑战性,特别是因 为检查每根烟棒中的烟草含量很困难。过去,一些国家按每千克烟草对卷烟征税,但今天 普遍的做法是不管烟草含量如何,对每支卷烟征税。 表2.4 HTPs的消费税于2018年7月首次征收(2020年7月更新) 与卷烟的总体比较 消费税类型 基本单位为千克,整 体费率低于卷烟 基本单位是支,费率 与卷烟相同 基本单位是支,费 率低于卷烟 其他 34第二章 烟草消费税政策 续表 与卷烟的总体比较 从量税 阿尔 巴 尼 亚、奥 地 利、白俄罗斯、波斯 尼亚-黑塞哥维那、 保加 利 亚、克 罗 地 亚、塞浦路斯、捷克、 丹麦、希腊、哈萨克 斯坦、吉 尔 吉 斯 斯 坦、拉脱维亚、立陶 宛、黑山、荷兰、新西 兰、北马其顿、罗马 尼亚、俄罗斯联邦、 斯洛伐克、斯洛文尼 亚、瑞典 阿塞拜疆a、日本、乌 克兰b 亚 美 尼 亚、匈 牙 利、约旦、意大利c、 菲律宾、韩国 摩尔多瓦、塞 尔维亚g 从价消费税(基 数为零售价,除 非另有说明) 西班牙、瑞士 沙特阿拉伯和阿拉 伯联合酋长国(基数 为零售价,不含消费 税和增值税) 印度尼西亚h 混合制(除非另 有说明,否则按 零售价计价) 法国、德国、波兰、葡 萄牙 哥伦 比 亚、格 鲁 吉 亚、以色列、约旦河 西岸和加沙地带(从 价消费税,税基为批 发价) a 适用的从量税率与进口卷烟相同,高于国产卷烟的税率。 b 该税率与每 1 000 支卷烟的最低消费税相同。费率和结构自2021 年 1 月 1 日起生效。 c 从量税被定义为基于卷烟和 HTPs 之间使用的等效性的卷烟消费税的 25%。2023 年将这一比例增加到 40%。 d 2020 年的从量税仅比卷烟低 11%。 e 从量税以烟草混合物的重量(千克)为基数,按每 1 000 支卷烟最低消费税的 40% 计算。 f 从量税高于卷烟,但与 HTPs 不同,卷烟也面临从价消费税。HTPs 的消费税总体略低。 g 从量税以烟草混合物的重量(千克)为基数,按每 1 000 支卷烟最低消费税的 40% 计算。计划分阶段增加这一 比例,目标是到 2025 年与卷烟持平。 h 虽然卷烟面临从量税率,但 HTPs 面临从价税率,即法律规定的最高税率,基于预先定义的最低价格。 资料来源:世界卫生组织2020 年卷烟和 HTPs 价格和税收数据收集,截至 2021 年 4 月未发布[1,100,101];儿童无 烟运动网站关于加热烟草制品的税收和价格,https://www.tobaccofreekids.org/what-we-do/global/taxation-price/ staging-tax-gap。 Liber[102]的一项研究比较了 34 个国家的 HTPs 和卷烟的价格,结果表明,虽然 HTPs 的税收普遍性低于卷烟,但在一半接受调查的国家中价格较高。 44 世界卫生组织烟草税政策和管理技术手册 关键要点22 HTPs 在征税时通常低于卷烟,尽管它们的价格通常高于卷烟。 需要明确的是,HTPs 是烟草制品,适用于烟草制品的规定同样也应适用于它们。世 卫组织关于 HTPs 的信息表[103]对此进行了阐述,该信息表提供了有关如何监管这类产 品的指导,以及关于新型烟草制品的 FCTC/COP8[22]决定。此外,帮助世界卫生组织成 员国实施世界卫生组织FCTC 减少需求条款的 MPOWER 措施适用于 HTPs,特别是税 收第 6 条。目前,没有证据表明 HTPs 比传统烟草制品的危害小。此外,HTPs 含有卷 烟烟雾中没有的化学物质,其对健康的影响尚不清楚。对行业数据的独立评估表明,在 HTPs排放中,超过 20 种有害和潜在有害化学物质明显高于卷烟烟雾[104]。因此,需要更 多地了解这些产品及其排放对健康的影响,以及接触这些排放造成的其他的影响。 关键要点23 目前,没有证据表明 HTPs 比传统烟草制品的危害小。 从公共卫生和税务管理的角度来看,HTPs 的征税水平和方式应与卷烟相同。一些 国家已经采用了这种方法,并按照与卷烟相同的标准对 HTPs 征税(如阿塞拜疆、哥伦比 亚、格鲁吉亚、以色列、日本、乌克兰以及约旦河西岸和加沙地带)。沙特阿拉伯和阿联酋 最近作为海湾合作委员会的成员国对烟草制品征收消费税,现在对卷烟和 HTPs 应用相 同的进口关税税率和消费税结构。 技术的持续发展和产品的变化要求政府对每单位 HTPs 征税。单位的定义可能因 HTPs 类别中的产品而异。例如,IQOS 的一个单位是一支热烟,Glo的单位是一个 Neo Stick,而Ploom TECH的单位是一个烟弹。政府需要为市场上允许的每种产品确定一 个单位的确切定义。市场潜在的复杂性强烈支持限制一个国家允许的 HTPs 类型,并制 定严格的法规来尽可能地标准化这类产品。 各国还可以考虑对用于消费 HTPs 的设备征税,即支架和充电器(见附件 3.1 中的 产品说明)。 关键要点24 HTPs 是烟草制品,这一点必须明确。在没有被禁止的情况下,HTPs应受到严格 的监管和征税。建议与卷烟相同水平,以每单位对它们征税。各国还可以考虑对用于 HTPs 消费的设备征税。 2.4.2　电子尼古丁和非尼古丁输送系统（ENDS/ENNDS） ENDS 和 ENNDS 等产品在过去十年中发展迅速。ENDS 对含有尼古丁(但不含烟 54第二章 烟草消费税政策 值得注意的是,ENDS 不是烟草制品,也不是新兴产品。该技术自20世纪80 年代后期就已经出现(例如 Premier、Eclipse 和 Accord)。然而,这些产品的最新一代或多或少地借鉴了电子烟的成功经验。 草)和其他可能对人体健康有害的液态化学物质(电子烟油)加热,形成气溶胶,供用户吸 入。ENDS 的例子包括 Juul Labs 的 Juul、BAT 的 Vype 和 Imperial Brands 的blu[1]。 电子非尼古丁递送系统(ENNDS)与 ENDS 基本相同,但使用的电子烟油一般不含 尼古丁。然而,经过测试,发现许多所谓的零尼古丁溶液含有尼古丁[105107]。 ENDS虽然通常被视为单一产品类别,但产品构成了一个多样化的群体,只是在有毒 物的产生和尼古丁的输送方面可能存在显著差异。市场上有几种共存的 ENDS/ENNDS 设备类型,包括第一代所谓的卷烟、第二代罐系统和更大的第三代个人蒸发器。总的来 说,它们通常也被称为电子烟、Vapes 或 Vape Pens。ENDS 的其他类别包括电子水烟、 电子烟斗和电子雪茄,因此,ENDS 是多个产品类别的统称。一些产品类似于传统的烟草 制品,比如卷烟、雪茄、小雪茄、烟斗或水烟袋,而其他产品的形状更像钢笔、USB 记忆棒 或基本圆筒。这些 ENDS 中还使用了不同形式的尼古丁,最近的一种是尼古丁盐,它能 提供高水平的尼古丁[1]。 ENDS/ENNDS 产品有两种类型:开放系统和封闭系统。开放系统是允许用户购买 电子烟油并在其设备中填充他们想要的混合物(不含尼古丁、不同的尼古丁浓度和/或口 味)的设备。封闭系统是带有预装容器(称为墨盒、吊舱或罐)的产品。 在过去的十年中,关于 ENDS 作为戒烟辅助手段的有效性(尤其是对于无法戒烟的 烟草使用者)以及 ENDS 在公共卫生中发挥作用的可能性,一直存在分歧,然而,证据仍 然不确定。尽管烟草业和其他相关行业推荐这些产品作为戒烟工具,但目前的证据并不 支持将其用作基于人群的戒烟策略的一部分[108]。因此,美国卫生总署于 2020 年 1 月得 出结论,电子烟是一种不断变化和多样化的产品,其使用方法多种多样。因此,很难根据 涉及特定电子烟的临床试验对戒烟功效进行概括,目前也没有足够的证据能证明电子烟 总体上会提高戒烟率[109]。 目前关于使用 ENDS 不利健康的证据越来越多,大多数 ENDS 用户的做法[110]是将 ENDS 与吸烟结合使用,这使两种或多种产品的不利健康的因素结合在一起。然而,没有 足够的数据来了解这些影响的全部范围,因为 ENDS 投放市场的时间还不够长,无法确 定其对健康的长期影响。 尽管如此,有证据表明大多数 ENDS 和 ENNDS中的气溶胶,其中一些成分是致癌 的化学物质。ENDS 还含有尼古丁,非常容易让人上瘾。此外,ENDS 可导致患心血管疾 病和肺部疾病的风险增加,并会对怀孕期间发育中的胎儿产生不利影响[108,110]。对于青 少年来说,使用尼古丁会导致上瘾,并可能损害其大脑发育。ENDS 的使用还可能导致新 一代尼古丁和烟草使用者更多,正如在一些国家看到的那样,这种现象是因为这些产品旨 在吸引年轻人。尽管与 ENDS 相关的具体风险水平尚未最终确定,但这些产品无疑是有 害的。因此,如果允许它们在国内市场销售,则应受到严格监管,并且必须远离儿童。税 收将成为监管手段的关键组成部分,因为它是影响消费者行为的有效工具。 一些国家已经做出了完全禁止这些产品的大胆决定。已采取的方法从部分禁止到全 面禁止,2018 年有 30 多个国家禁止使用 ENDS、ENNDS 产品 。在其他国家,它们作为 64 世界卫生组织烟草税政策和管理技术手册 为世界卫生组织RGTE 2019 收集的数据。 消费品、药品或烟草产品受到监管或完全不受监管。世界卫生组织建议,在 ENDS/ ENNDS 未被禁止的情况下,应对其进行监管以实现以下目标: 1.禁止非吸烟者和青少年使用 ENDS、ENNDS,要特别关注弱势群体; 2.尽可能减少 ENDS、ENNDS 用户的潜在健康风险,并保护非用户免受其排放物的影响; 3.禁止对 ENDS、ENNDS 做出未经证实的健康声明; 4.保护烟草控制活动免受与 ENDS、ENNDS 相关的所有商业和其他既得利益(包括 烟草业的利益)的影响。 ENDS、ENNDS 对年轻人构成风险,在一些国家(包括加拿大和美国)大量年轻人使 用电子烟[1,111]。例如,Juul 品牌在美国迅速获得了可观的电子烟市场份额[112,113]。它的 营销和普及导致美国食品和药物管理局(FDA) 提出严重关切并寻求解决方案,以有效防 止年轻人使用 ENDS、ENNDS[114]。旧金山市于 2019 年 6 月禁止销售电子烟[115]。 ENDS除了给年轻人带来开始吸烟的风险外,还可以吸引非烟草使用者或阻止当前 吸烟者戒烟。税收可以在控制这些产品的消费方面发挥作用,特别是对于非吸烟者、弱势 群体、儿童和青少年来说。 关键要点25 ENDS、ENNDS 产品对长期健康影响尚不清楚,但它们显然对健康有害。此外,关 于 ENDS 产品作为戒烟辅助手段的有效性的证据仍然没有定论。对这些产品征税可 能会阻止人们使用,特别是在非吸烟者、弱势群体、儿童和青少年中。 ENDS 产品需求的价格弹性 在税收方面,重要的是对 ENDS 的需求价格是否敏感提出疑问。初步证据虽然几乎 完全集中在美国的电子烟数据上,但事实却表明:对电子烟的需求价格可能比对传统卷烟 的需求价格更加敏感,因此可以使用税收来阻止从未吸烟的人开始吸烟[116123]。大多数 ENDS 产品需求价格弹性的研究也表明,电子烟和传统卷烟是部分替代品,即它们表现出 正的交叉价格弹性。弹性的大小表明产品之间的可替代程度:其价格越高,产品越接近替 代品;卷烟价格越高,电子烟销量越增加。一些研究还显示了另一个方面的可替代性效 应:电子烟价格上涨导致传统卷烟使用量增加[117,120]。所有研究都显示了可替代性的证 据,只有一项研究显示出相反结论[124],该研究区分了单一用户和双重用户,说明了没有 证据表明电子烟和传统卷烟之间存在替代性。同时(或双重)使用(即消费者同时使用传 统卷烟和 ENDS 产品)会使结果更加复杂化,也表明需要在该领域进行更多的研究以更 好地解决二者不同的影响。 Liber 等人[125]分析了 45 个国家样本的销售价格,并得出结论,传统卷烟的可比单位 的成本低于一次性电子烟。考虑为电子烟定价的单位包括电子烟油和可充电设备。单独 来看,电子烟油的平均价格远低于高收入国家和中低收入国家的卷烟价格。在大多数国 家/地区,回购可充电设备所需的时间估计不到两周。 有人可能会争辩说,通过进一步增加对普通卷烟的税收来增加价格差异可能会有效 地推动当前普通卷烟吸烟者转向电子烟[126],电子烟是一种潜在的低风险替代品[127]。然 74第二章 烟草消费税政策 而,ENDS 作为戒烟设备的有效性仍在争论中。Sweet 等人的一项研究[128]表明,双重使 用电子烟作为戒烟的潜在工具仅在短期内有效。此外,大量的吸烟者表示,如果卷烟价格 翻倍且没有电子烟,他们会戒烟[122],或者说如果电子烟不那么容易获得,他们永远不会 对尼古丁上瘾[129]。一旦电子烟使用者对尼古丁上瘾,就有开始使用传统烟草制品的风 险[130]。总的来说,政府可以通过更有力地实施其他已被证明可有效减少烟草使用的控 制政策来更好地促进戒烟。 关键要点26 关于 ENDS 产品价格弹性的研究很少,可用数据几乎完全来自美国。 早期研究表明,随着电子烟价格的上涨,人们对电子烟的需求将会下降。结果还表 明,一般来说,卷烟和电子烟是部分替代品,其中卷烟价格上涨会增加对电子烟的需求, 同时减少对卷烟的需求。 但是这些研究结果并没有区分那些只吸食卷烟(或电子烟)的人与那些两种产品都 吸食的人。 税收结构 不同国家对 ENDS、ENNDS 产品征收不同税收结构的税款 (详见表 2.5)。例如, 韩国基于每毫升 ENDS、ENNDS 电子烟油征收从量税[131],而印度尼西亚则对电子烟油 的零售价征收从价税;也有法律允许的烟草制品最高税率[132]。在美国,对电子烟征税的 州之间没有通用的方式来征税[133,134]。欧盟的情况类似,新型且流行的尼古丁和烟草制 品目前不在烟草税政策范围内,成员国可以根据自己的规则征收他们认为合适的烟草税。 所有对 ENDS 产品征税的欧盟国家都基于每毫升电子烟油征收从量税。 这些不同的税收待遇有可能扭曲内部市场的运作。2020 年 2 月,欧盟委员会得出结 论认为,协调政策的当前规定不再与 ENDS 和 HTPs 的税收相关,这从内部市场的角度 来看是一个令人担忧的问题[135]。2020 年 6 月,欧盟成员国重申,通过协调 ENDS、 ENNDS 和 HTPs 等新型产品的定义和税收政策来升级欧盟监管框架是紧迫和必要的[2]。 表2.5 截至 2019 年 7 月,适用于全球和美国各个州的 ENDS、ENNDS 产品电子烟油的消费税类型(截 至 2020 年 7 月更新,适用于除美国以外的所有国家/地区) 消费税类型 国 家 仅对含尼古丁的电子烟油(ENDS 产品)征税 对所有电子烟油征税 (ENDS和ENNDS产品) 从量税(基于每毫升 体积) 阿尔巴尼亚、哈萨克斯坦、吉尔吉 斯斯坦、巴基斯坦、葡萄牙、韩国、 罗马尼亚、俄罗斯联邦、斯洛文尼 亚、瑞典 阿塞拜疆、塞浦路斯、爱沙尼亚、芬兰、格 鲁吉亚、希腊、匈牙利、意大利、拉脱维 亚、立陶宛、黑山、摩洛哥、北马其顿、菲 律宾、塞尔维亚 84 世界卫生组织烟草税政策和管理技术手册 重点是用于 ENDS、ENNDS 产品的电子烟油。 续表 消费税类型 国 家 仅对含尼古丁的电子烟油(ENDS 产品)征税 对所有电子烟油征税 (ENDS和ENNDS产品) 从价税(占零售价格或 进口价值的百分比) 巴林a 印度尼西亚、约旦、阿拉伯联合酋长国、 也门 消费税类型 美国的个别州 仅对含尼古丁的电子烟油(ENDS 产品)征税 对所有电子烟油征税 (ENDS和ENNDS产品) 从量税(基于每毫升 体积) 特拉华州、伊利诺州(芝加哥)、库 克县、路易斯安那州、俄亥俄州、波 多黎各C、康涅狄格州C 堪萨斯州、北卡罗来纳州、华盛顿州、西 弗吉尼亚州、威斯康州 从价税(占批发或分 销商价格的百分比) 阿拉斯加(朱诺、马塔努斯卡 苏西 特纳区)、加利福尼亚州、伊利诺 州、缅因州、马里兰州(蒙哥马利 县)、明尼苏达州、内华达州、宾夕 法尼亚州、佛蒙特州、华盛顿特区、 维尔京群岛C 纽约 混合税 新墨西哥州、新泽西 a 对电子烟(或电子水烟)征收税,因为巴林禁止电子烟。 b 意大利对含有尼古丁和不含尼古丁的烟油实行不同的税率。 c 目前尚不清楚是否只对ENDS或同时征收ENDS和ENNDS产品税的州。 资料来源:来自FrankChaloupka和世卫组织2020年卷烟和 HTPs的价格和税收数据收集的补充数据[1,135],截 至2021年4月未发表)。 表2.6提供了关于确定 ENDS、ENNDS 产品电子烟油的税收结构和税基的不同考 虑的利弊的参考资料。 表2.6 ENDS、ENNDS 产品电子烟油的消费税选项 消费税 类型 税 基 优 势 劣 势 从量税 含尼古丁的电 子烟油的体积 (不考虑浓度) 1.降低同类产品之间的价格差距; 2.从税务管理的角度来看很简单, 因为只需要确定数量 1.如果寻求的税收与卷烟相当,就很 难进行比较; 2.需要实验室有能力检测电子烟油中 尼古丁的存在,因为目前还没有简单 的方法来确定电子烟油是否含有尼古 丁(行业陈述是不够的); 3.无法涵盖不含尼古丁的电子烟油, 这些烟油在通过电子烟吸入人体时也 是有害的,并且在独立测试中经常被 发现含有尼古丁,这与产品标签或行 业陈述相反; 94第二章 烟草消费税政策 续表 消费税 类型 税 基 优 势 劣 势 从量税 4.可能对每毫升尼古丁浓度较高的产 品有利,这些产品往往也是在被积极 营销的国家中最容易被青少年快速接 受的产品; 5.可能会鼓励更多自己动手(DIY)的 产品,这其中电子烟油由用户自己混 合,这增加了事故、疾病和死亡的风险 不考虑尼古丁 含量的电子烟 油体积 1.降低同类产品之间的价格差距; 2.从税务管理的角度来看很简单, 因为只需要确定数量; 3.涵盖不含尼古丁的电子烟油,这 些烟油在通过电子烟吸入时也是 有害的,并且在独立测试中经常被 发现含有尼古丁,这与产品标签或 行业陈述相反; 4.不需要有能力检测液体中尼古 丁存在的实验室 1.如果寻求的税收与卷烟相当,就很 难进行比较; 2.检测和区分在ENDS、ENNDS中使 用的液体(如丙二醇或植物甘油)是否 在进口和制造水平上被虚假申报为用 于其他目的(如食品、化妆品或药品) 方面面临挑战; 3.可能对每毫升尼古丁浓度较高的产 品有利,这些产品往往也是被积极营 销的国家中最容易被青少年迅速接受 的产品 所有电子烟油 的体积,每单位 尼古丁浓度额 外征税 1.更有可能通过影响最低价格类 别和高浓度尼古丁产品,让青少年 更难开始吸烟;同时降低了不同产 品之间的价格差距。 2.涵盖不含尼古丁的电子烟油,这 些烟油在通过电子烟吸入时也是 有害的,并且在独立测试中经常被 发现含有尼古丁,这与产品标签或 行业陈述相反。 3.相同产品类别的当前市场观察 (2019 年)表明尼古丁浓度不是价 格的主要决定因素;随着尼古丁浓 度升高而增加税收可能会改变这 一点,并会强化尼古丁令人上瘾而 非无害的健康理由。 4.减少制造商增加尼古丁浓度以 减轻税收负担的动机。 5.减少个人DIY动机,因为增加个 人消费的尼古丁浓度会增加用户 的成本 1.如果寻求的税收与卷烟相当,就很 难进行比较; 2.检测和区分在ENDS、ENNDS中使 用的液体(如丙二醇或植物甘油)是否 在进口和制造水平上被虚假申报为用 于其他目的(如食品、化妆品或药品) 方面面临挑战; 3.需要有能力检测液体中尼古丁存在 的实验室; 4.从税务管理的角度来看更复杂,因 为体积和尼古丁含量都需要评估 05 世界卫生组织烟草税政策和管理技术手册 续表 消费税 类型 税 基 优 势 劣 势 从量税 含尼古丁的电 子烟油量,每单 位尼古丁浓度 额外征税 1.更有可能通过影响最低价格类 别和尼古丁含量较高的产品,让青 少年更难开始吸烟;同时降低了不 同产品之间的价格差距。 2.相同产品类别的当前市场观察 (2019年)表明尼古丁浓度不是价 格的主要决定因素;随着尼古丁浓 度升高而增加税收可能会改变这 一点,并会强化尼古丁令人上瘾而 非无害的健康理由。 3.减少制造商增加尼古丁浓度以 减轻税收负担的动机。 4.减少个人 DIY 动机,因为增加个 人消费的尼古丁浓度会增加用户 的成本 1.如果寻求的税收与卷烟相当,就很 难进行比较; 2.需要实验室有能力检测电子烟油中 尼古丁的存在,因为目前还没有简单 的方法来确定电子烟油是否含有尼古 丁(行业陈述是不够的); 3.需要有能力检测液体中尼古丁浓度 的实验室; 4.从税务管理的角度来看更复杂,因 为体积和尼古丁含量都需要评估 从价税 含尼古丁的电 子烟油的生产 者价格/到岸价 格(不考虑尼古 丁浓度) 在产品存在较大异质性的情况下, 更容易计算和调节 ENDS产品和 卷烟之间的税收等价性(特别是如 果卷烟税是从价税,尽管这并没有 改变长期以来关于从量税对传统 卷烟更好的观点) 1.税基的真实价值难以确定,容易低 估; 2.需要较强的税务管理能力来有效实 施,特别是评估申报税基价值有效性 的能力; 3.需要实验室有能力检测电子烟油中 尼古丁的存在,因为目前还没有简单 的方法来确定电子烟油是否含有尼古 丁(行业陈述是不够的); 4.可能会鼓励更多自己动手(DIY)的 产品,这其中电子烟油由用户自己混 合,这也增加了事故、疾病和死亡的 风险 15第二章 烟草消费税政策 续表 消费税 类型 税 基 优 势 劣 势 从价税 所有电子烟的 生产者价格/到 岸价格(不考虑 尼古丁浓度) 1.在产品存在较大异质性的情况 下,更容易计算和调节ENDS产品 和卷烟之间的税收等价性(特别是 如果卷烟税是从价税,尽管这并没 有改变长期以来关于从量税对传 统卷烟更好的建议); 2.不需要有能力检测液体中尼古 丁存在的实验室; 3.涵盖不含尼古丁的电子烟油,这 些烟油在通过电子烟吸入时也是 有害的,并且在独立测试中经常被 发现含有尼古丁,这与产品标签或 行业陈述相反 1.税基的真实价值难以确定,容易 低估; 2.需要较强的税务管理能力来有效实 施,特别是评估申报税基价值有效性 的能力; 3.检测和区分在ENDS、ENNDS中使 用的液体(如丙二醇或植物甘油)是否 在进口和制造水平上被虚假申报为用 于其他目的(如食品、化妆品或药品) 方面面临挑战 含尼古丁的电 子烟油体的零 售价(不考虑尼 古丁浓度) 1.在产品存在很大异质性的情况 下,更容易计算和调节ENDS产品 和卷烟之间的税收等价性(特别是 如果卷烟税是从价税,尽管这并没 有改变长期以来关于从量税对传 统卷烟更好的建议); 2.税基不太容易被低估,因为税基 更容易评估(与到岸价格/生产者 价格税基相比) 1.需要实验室有能力检测电子烟油中 尼古丁的存在,因为目前还没有简单 的方法来确定电子烟油是否含有尼古 丁(行业陈述是不够的); 2.需要监控市场以评估市场价格的 能力; 3.不涵盖不含尼古丁的电子烟油,这 些烟油在通过电子烟吸入人体时也是 有害的,并且在独立测试中经常被发 现含有尼古丁,这与产品标签或行业 陈述相反; 4.可能会鼓励更多自己动手(DIY)的 产品,这其中电子烟油由用户自己混 合,这也增加了事故、疾病和死亡的 风险 25 世界卫生组织烟草税政策和管理技术手册 续表 消费税 类型 税 基 优 势 劣 势 从价税 不考虑尼古丁 含量的所有电 子烟油的零售 价格 1.在产品存在较大异质性的情况 下,更 容 易 计 算 和 调 节 ENDS、 ENNDS产品和卷烟之间的税收等 价性(特别是如果卷烟税是从价 税,尽管这并没有改变长期以来关 于 从 量 税 对 传 统 卷 烟 更 好 的 建议); 2.不需要有能力检测液体中尼古 丁存在的实验室; 3.不太容易被低估,因为税基更容 易评估(与到岸价格/生产者价格 税基相比) 1.需要监控市场以评估零售价格的 能力; 2.检测和区分在ENDS、ENNDS中使 用的液体(如丙二醇或植物甘油)是否 在进口和制造水平上被虚假申报为用 于其他目的(如食品、化妆品或药品) 方面面临挑战 设 置 有 低 从 量 税 的 从 价 税 或 混 合 税 从价税与最低从 量税: 1.含尼古丁的电 子烟液的量将是 最低规格的从量 税基数; 2. 零售价将作为 从 价 消 费 税 的 基数b。 或混合税: 1.含尼古丁的电 子烟液的量将是 最低规格的从量 税基数; 2. 零售价将作为 从价消费税的基 数b 1.本制度旨在发挥从量税制和从 价税制的优点: a.特定组成部分保证最低税收并 推高所有产品的价格; b.在产品存在大量异质性的情况 下,按从价成分更容易计算和监管 ENDS 产品和卷烟之间的税收等 效性 1.需要监控市场以评估零售价格的 能力; 2.需要实验室有能力检测电子烟油中 尼古丁的存在,因为目前还没有简单 的方法来确定电子烟油是否含有尼古 丁(行业陈述是不够的); 3.不涵盖不含尼古丁的电子烟油,这 些烟油在通过电子烟吸入人体时也是 有害的,并且在独立测试中经常被发 现含有尼古丁,这与产品标签或行业 陈述相反; 4.很难设定最低的从量税数额/从价 税数额,特别是在寻求税收等同于卷 烟的情况下; 5.可能会鼓励更多自己动手(DIY)的 产品,这其中电子烟油由用户自己混 合,这也增加了事故、疾病和死亡的 风险 35第二章 烟草消费税政策 续表 消费税 类型 税 基 优 势 劣 势 设 置 有 低 从 量 税 的 从 价 税 或 混 合 税 从价税与最低 限额: 1.含尼古丁的 电子烟液的量 将是最低规格 的基数; 2.零售价将作 为从价消费税 的基数b。 或混合税: 1.含尼古丁的 电子烟液的量 将是最低规格 的基数; 2.零售价将作 为从价消费税 的基数b 1.本制度旨在发挥从量税制和从 价税制的优点: a.特定组成部分保证最低税收并 推高所有产品的价格; b.在产品存在大量异质性的情况 下,按从价成分更容易计算和监管 ENDS、ENNDS 产品和卷烟之间的 税收等效性; 2.不需要有能力检测液体中尼古 丁存在的实验室 1.需要监控市场以评估零售价格的 能力; 2.很难设定最低的从量税数额/从量 税数额,特别是在寻求税收等同于卷 烟的情况下; 3.检测和区分在ENDS、ENNDS中使 用的液体(如丙二醇和植物甘油)是否 在进口和制造水平上被虚假申报为用 于其他目的(如食品、化妆品或药品) 方面面临挑战 a 税收等价性可以用不同的方式来衡量:①税收负担(作为零售价格的%);②作为等价数量的确切税额(假设一 定数量的电子液体和一包卷烟之间的等价性)。 b 也可以选择使用生产者价格/到岸价格作为从价税部分的基数,但这是一个较弱的选择,因为基数很难确定,因 此容易被低估。 注:表格是在 2019 年 9 月 2 日至 4 日于瑞士日内瓦举行的世卫组织电子尼古丁和非尼古丁传递系统征税专家 会议(ENDS/ENNDS) 之后编制的。 目前缺乏证据表明更支持上述其中一种方法的国家所面临的实际挑战。此外,由于 市场的性质在不断变化,这类数据难以获得。 然而,可以对哪些电子液体征税提出明确的建议。如表 2.5 所示,一些国家/地区对 所有电子烟油征税,无论它们是否含有尼古丁(ENDS 和 ENNDS 产品);而有些国家仅对 含尼古丁的电子烟油(ENDS 产品)征税。如表 2.6 所示,有证据表明,在许多情况下, ENNDS 产品确实含有一些尼古丁。此外,ENNDS 产品并非无害[136,137]。因此,建议对 所有电子烟油(ENDS 和 ENNDS产品)征税。 从健康角度来看,是否基于尼古丁含量采用差别征税的问题似乎是合理的,因为尼古 丁是一种有毒物质。然而,这可能会给税务管理员带来额外的工作,因为他们需要确定市 场上电子烟油的尼古丁浓度。此外,浓度与实际输送的尼古丁含量可能不再相关,因为技 术的进步表明产品的其他特征可能会导致超出电子液体实际浓度的尼古丁输送。现在可 以通过增加电池功率(通过降低电阻或增加电压)来增加低尼古丁浓度下的尼古丁输 45 世界卫生组织烟草税政策和管理技术手册 送[138]。 在实施方面,虽然大多数国家似乎基于每毫升 ENDS、ENNDS 电子烟油征收了从量 税,但实施从价税的一个好处是在产品种类繁多的情况下似乎相对更容易监管。然而,对 产品的零售价格价值征税是至关重要的,因为这个基数比制造商申报的任何其他价值都 更容易确定。 还需要补充的是,对 ENDS、ENNDS 产品的特性进行监管是必不可少的,并且应与 所采用的其他税收政策一起实施。这些政策应包括: 1.设定每毫升尼古丁的最高浓度,以保障公众健康,特别是减少青少年依赖的风险; 2.设置药筒的最大体积以减少毒物暴露并尽可能限制使用; 3.设置再填充容器的最大容量,以减少有毒物质暴露并尽可能限制使用; 4.设置最大的电池功率,以减少影响尼古丁和有毒物质输送的可能性; 5.对尼古丁征税,无论其来源如何(如烟草、雪茄以及其他合成物)。 各国也可以选择对ENDS和ENNDS设备 征收消费税。最简单的税收类型是根据 公布的零售价征收从价税。如果各国选择不对这些产品征收消费税,它们至少应征收常 规增值税或销售税。从管理的角度来看,对设备征收消费税可能是具有挑战性的,因为所 有组件都需要明确定义并分类为ENDS、ENNDS消费的设备。例如,如果设备是在进口 后组装的,并且某些部件可以用于ENDS、ENNDS消费以外的其他目的,当局可能在检 测和区分哪些部件应缴纳消费税、哪些不应缴纳消费税方面面临挑战。 关键要点27 目前没有足够的证据为 ENDS和ENNDS 产品推荐一种科学合理的税收结构。 然而,很明显,对用于消费的电子烟油征税是必要的。消费税应适用于所有电子烟油, 无论它们是否含有尼古丁。如果首选的消费税类型是从价税,则应适用于零售价。各 国也可以考虑对设备征税,但他们需要充分评估这样做的行政能力。 政策制定者需要注意 ENDS、ENNDS 产品的多样性和快速演变性,并相应地进行政 策调整。监管政策必须注意这一现实,以免政策漏洞被行业利用。例如,ENDS、ENNDS 产品不仅包括电子烟、Vape和Vape笔,还包括电子水烟、电子烟斗和电子雪茄等其他类 别。立法者需要明确 ENDS、ENNDS 产品是如何定义的,以便在法规生效时子类别不会 受到关注。在税收方面,定义也很重要。一个不明确的定义可能会导致矛盾的结果,例如 在巴林,电子烟被禁止,但电子水烟却没有被禁止 。 虽然政策制定者需要注意新产品的出现并必须采取适当的行动来保护其公民的健 康,但重要的是要记住,消费尼古丁的绝大部分来源仍然是烟草制品,尤其是卷烟。2018 年 ENDS、ENNDS 和 HTPs 销售额的总市值不到烟草销售额总市值的 2.2%,而仅卷烟 销售额就占了相同总市值的 91%[139,140]。 55第二章 烟草消费税政策 有关 ENDS、ENNDS 产品中使用的设备元素的概述,请参见附件2.3。 在巴林,生产和贸易部 2013 年第 38 号决定禁止电子烟,而财政部的官方应税产品清单包括电子水烟(或电子烟 袋),使其被正常征税,因此被视为合法的。 2.5　总结 对全球消费税应用的概述揭示了广泛的价格水平和税收水平,以及用于对烟草制品, 特别是卷烟征税的结构。然而,一些趋势表明,高收入国家的税收和价格水平更高。税率 也很重要:更高的税率与更高的价格相关,较高的价格会改变消费者行为,从而导致消费 的减少。越来越多的国家正在从从价税转向混合税或从量税制度,而且很少有国家不对 卷烟征收任何消费税。 显著提高烟草制品的税收和价格是控制烟草使用的最有效和最具成本效益的政策。 随着烟草制品价格的升高,增加的税收会转嫁给吸烟者,从而减少烟草的消费。 在设计烟草税政策或改革烟草税制度时,决策者面临几个挑战,从技术方面如何确定 合适的税收结构和税率,到政治经济方面,以及烟草部门对经济发展的公认贡献。 在设计税收政策时,所采用的税收结构不仅影响整体消费,还会塑造市场结构。 从价税鼓励行业设定的价格低于实行从量税时所设定的价格。 有证据表明,在从量税下,优质产品和低价产品之间的价格差距更小,因此减少了税 收增加后替代低价产品的动机。然而,随着行业整合,生产商扩大了其产品组合,新的证 据表明,该行业正在增加更便宜的品牌,同时提高其昂贵品牌的价格。因此,矛盾的是,这 导致了产品之间的价格差距反而有所扩大。 证据还表明,在从量税结构下价格更高。此外,从税务管理的角度来看,从量税更容 易实施,因为只需要确定生产的数量,而不是产品的价值。 税收结构的另一个方面是使用分级税收,即根据不同产品特性实行变化的税率。有 证据表明,使用分级消费税结构的国家的平均卷烟价格和一包卷烟的平均消费税水平,往 往比使用统一消费税的国家低得多。分级税收会刺激消费者从优质品牌向更廉价品牌的 转变,从而维持了吸烟率并降低税率增加对健康的影响。除了导致价格下降之外,分级征 税难以管理,并为烟草业创造了避税和逃税的机会。 税收结构的设计还必须考虑征税的基数。基数的选择应该对价格和收入产生尽可能 大的影响。对于从量税,税基是数量。当税收为从价税时,税基的选择不仅对健康方面的 考虑(通过对消费的影响)很重要,对税收收入和行业利润也很重要。基于生产者价格或 CIF 值的从价税为烟草制造商提供了减少纳税义务的机会,特别是当他们通过关联方控 制分销系统时。从价(或混合)消费税结构中的最佳做法是使用零售价作为税基并引入每 包最低消费税。 其他税收设计考虑因素包括对从量税使用自动调整和指数化通货膨胀和收入增长的 重要性,以避免税收随着时间的推移而受到侵蚀。 新出现的证据表明,烟草税并不总能达到预期的结果,因为烟草业会想方设法规避 它。定价监管等非税收政策(特别是最低加价或底价/最低价格)可被视为确保高价格水 平和抑制烟草制品消费的补充方法。迄今为止,这些政策尚未被证明会提高烟草制品的 平均价格。价格下限可能会导致行业利润增加,为行业营销策略(例如推出新产品)提供 更多资金,并降低政府的税收收入。通过减少价格竞争,价格下限允许公司在其他方面 65 世界卫生组织烟草税政策和管理技术手册 (例如产品规格)积极竞争市场份额。 然而,在一些特定情况下,强大的跨国公司在几乎所有市场板块都有所参与,他们有 能力对某些品牌超额转税收,而对另外一些品牌则少转税收,或者当价格促销不能被禁止 时,最低价格政策可能有助于提高增税的有效性。 其他影响价格水平的非税收政策与烟草制品的促销折扣和销售单支卷烟有关,这两 者都应该被完全禁止。禁止促销折扣通常在烟草广告、促销和赞助条款下的烟草控制法 中处理。 提高税收是劝阻消费的最有效方式,它还有为政府筹集资金的额外好处,因为这些资 金可以专门用于健康和教育项目,而不是作为利润流向烟草业。此外,为了使烟草制品的 消费税更有效地减少总体烟草使用并符合 WHO FCTC 第 6 条实施指南的建议(减少烟 草需求的价格和税收措施),所有烟草制品都需要以类似的方式征税,重点不应只放在卷 烟上。 税收选择和改革会对市场产生各种不同的、有时是相互矛盾的结果。例如,在数量、 种类或感知到的质量影响之间可能存在失衡。政府必须认识到,企业会对税收政策的变 化做出战略性的反应。因此,密切监测市场是必要的,以对行业反应形成正确的预期,从 而能够估计税收增加对消费和税收收入的影响。 为了估计税收增加对烟草制品需求和税收的总体影响,对需求的自身价格弹性、交叉 价格弹性和需求的收入弹性进行正确估计是很重要的。随着消费者做出决定的环境不断 变化,使用更新的需求弹性估计也很重要。例如,金融危机或成功的烟草控制干预措施, 都会改变需求和弹性。 衡量税收政策影响的另一个关键指标是税基弹性。决策者需要注意税基弹性的3个 关键组成部分:①烟草需求的价格弹性;②税收在消费价格中的份额;③消费税税率提高 到消费者价格的传递程度。这些因素受到增税影响的程度将影响需求和收入。目前,这 3个组成部分的总和在任何国家都不够高,不足以通过增加税收导致消费税收入减少。 必须承认的是,如果税收增加导致价格上涨低于收入水平的同步增加,它们将无法有 效减少消费,因为烟草在大多数国家仍然是正常商品。政策制定者在考虑增税时需要考 虑民众的负担能力。他们应该确保税收增加幅度足以使物价上涨超过收入的增长,从而 有效降低烟草消费。 在设计税收政策和决定正确的实施水平时,决策者需要评估和监测其政策决策的影 响。监测和评估很重要,测量影响的工具将是非常有用的,这样的工具有多种。例如,世 卫组织ISPT不仅关注税收政策的影响,还着眼于一系列烟草控制政策,这使国家政策制 定者和其他烟草控制专家能够探索未来的烟草控制政策的潜在影响。该工具从已发表的 文献、吸烟率趋势、国家人口信息和烟草相关死亡风险中选定的世卫组织FCTC减少需 求措施的影响因素。更具体的是,对于烟草税政策,世卫组织TaXSiM 评估了任意消费 税的增加和消费税结构的变化对价格水平、合法销售、烟草制品的消费税和其他税收收入 的影响。 建立和监测税收和烟草控制政策的指标有助于政策制定者评估其政策的有效性,以 及确定这些政策是否会随着时间的推移对烟草使用产生影响。MPOWER 一揽子计划的 75第二章 烟草消费税政策 实施是全面评估烟草控制的一项有用指标。如果将烟草税作为全面的 MPOWER 一揽 子计划的一部分加以实施,则其效果最佳。 选定烟草制品零售价格中的税收份额是税收政策有效性的指标之一,但更重要的是 考虑可负担性,即增税是否确实导致高于收入的价格上涨和总体价格上涨。评估税收政 策整体绩效的一个有用指标是烟草经济学-卷烟税收积分制,它通过烟草税收政策的4 个关键组成部分(价格水平、可负担性随时间的变化、零售价格和烟草税结构中的总税收 和消费税份额)来综合评估烟草税收的最佳实践方案。 农业、工业、贸易、金融和劳工方面的国内政策都有可能在烟草生产、制造和分销的不 同阶段创造机会或激励支持,这些措施可能对烟草控制和税收目标产生反作用。因此,应 促进政府不同部门之间的政策达到最大限度上的一致性,以确保这些部门的公共政策和 干预措施不会抵消烟草控制和税收调整对公众健康的预期影响。 不同的税收结构和利率有可能扭曲内部市场的运作。烟草税的协调确保了单一市场 的建立和正常运作,防止税收侵蚀、避税和逃税,并保护人民的健康。在这种情况下,税收 竞争,即各国简单地降低本国的税率,可能会阻碍各国政府实现其烟草控制目标和筹集足 够的资金来推行公共卫生政策。为了避免这种逐底竞争,各国可以对所有烟草制品制定 最低税率。常见的方法即设定较高的最低从量税,是确保税收和价格高于最低水平的最 佳方法。 近年来,世界上出现了很多新型烟草和尼古丁产品,包括 ENDS、ENNDS 和 HTPs, 业界声称这些产品比传统烟草制品更安全。而迄今为止的证据表明,这些产品可能对公 共健康构成威胁,特别是如果它们吸引新用户或年轻的用户,或阻碍当前的吸烟者戒烟的 话。 这些产品的市场和需求动态,以及不同社会经济群体烟草使用行为的开始、停止和转 变尚不清楚。在获得更多证据证明这些烟草制品好处之前,应谨慎制定税收政策。因此, 目前的建议是,无论烟草含量如何,对 HTPs 都应按每单位与卷烟相同的水平征税。在 未禁止使用 ENDS、ENNDS 产品的国家/地区,必须以不鼓励年轻人和非用户使用的方 式对 ENDS、ENNDS 产品进行监管和征税。对电子烟油征税是对ENDS、ENNDS 产品 征税的一个关键组成部分,含尼古丁和不含尼古丁的电子烟油应同等征税。如果选择从 价消费税作为结构,则应以零售价格为基数。各国还可以考虑对用于 ENDS、ENND 和 HTPs 消费的设备征税,但他们需要对自己的管理能力进行充分评估。 虽然新出现的烟草和尼古丁输送系统的演变值得管理人员和监管机构关注,但值得 记住的是,世界上大部分烟草消费仍然是传统卷烟。 85 世界卫生组织烟草税政策和管理技术手册 本章参考文献 1.WHO report on the global tobacco epidemic 2019:offer help to quit tobacco use. Geneva: World Health Organization;2019 (https://www.who.int/teams/health-promotion/tobacco- control/who-report-on-the-global-tobacco-epidemic-2019,accessed 24 January 2021). 2.Outcome of proceedings:Council conclusions concerning the structure and rates of excise duty applied to manufactured tobacco. Brussels:Council of the European Union;2 June 2020 (8483/20;https://www.consilium.europa.eu/media/44235/st08483-en20.pdf,accessed 28 September 2020). 3.Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco:Article 7,paragraph 2. Brussels:OJEU. 2011;176:24-36 (https://eur-lex.europa. eu/LexUriServ/LexUriServ.do? uri=OJ:L:2011:176:0024: 0036:EN:PDF,accessed 28 September 2020). 4.U.S. National Cancer Institute and World Health Organization.The economics of tobacco and tobacco control. NIH Publication No. 16-CA-8029A. Bethesda,MD:Department of Health and Human Services,National Institutes of Health,National Cancer Institute;2016 (National Cancer Institute Tobacco Control Monograph 21;https://cancercontrol.cancer. gov/brp/tcrb/monographs/monograph-21,accessed 28 September 2020). 5.Drope J,Schluger N,Cahn Z,Drope J,Hamil S,Islami F,et al. The tobacco atlas. Atlanta: American Cancer Society and Vital Strategies;2018 (https://tobaccoatlas.org/wp-content/ uploads/2018/03/ TobaccoAtlas_6thEdition_LoRes_Rev0318.pdf,accessed 24 January 2021). 6.Sector reports,cigarettes. In:GlobalData [website]. London:John Carpenter House;2019 (restricted access). 7.Velios N,Ross H,Perucic AM. 2018 trends in cigarette demand and supply in Africa. PLOS ONE. 2018;13(8):e0202467 (https://doi.org/10.1371/journal.pone.0202467,accessed 28 September 2020). 8.Keen MJ. The balance between specific and ad valorem taxation. Fiscal Studies. 1998; 19(1):1-37 (https://www.ifs.org.uk/fs/articles/fsmick1.pdf,accessed 24 January 2021). 9.Kay JA,Keen MJ. The structure of tobacco taxes in the European community. IFS report series. Institute for Fiscal Studies;1982. 10.Kay JA,Keen MJ. How should commodities be taxed?:Market structure,product heterogeneity and the optimal structure of commodity taxes. Euro Econ Rev. 1983;23(3): 339-358. 11.Kay JA,Keen MJ. Commodity taxation for maximum revenue. Public Finan Q. 1987; 15(4):371-385. 12.Kay JA,Keen MJ. Product quality under specific and ad valorem taxation. Public Finan Q. 1991;19(2):238-247. 95第二章 烟草消费税政策 13.Delipala S,Keen MJ. The comparison between ad valorem and specific taxation under imperfect competition. J Public Econ. 1992;49(3):351-367. 14.Suits DB,Musgrave RA. Ad valorem and unit taxes compared. Q J Econ. 1953;67(4): 598-604. 15.Skeath SE,Trandel GA. A pareto comparison of ad valorem and unit taxes in noncompetitive environments. J Public Econ. 1994;53(1):53-71. 16.Myles G. Imperfect competition and the optimal combination of ad valorem and specific taxation. Int Tax Public Finance. 1994;3:29-44. 17.Effectiveness of tax and price policies for tobacco control. Geneva:WHO Press,World Health Organization;2011. (IARC handbooks of cancer prevention:tobacco control:Vol. 14;https://publications. iarc.fr/Book-And-Report-Series/Iarc-Handbooks-Of-Cancer- Prevention/Effectiveness-Of-Tax-And-Price-Policies-For-Tobacco-Control-2011, accessed 28 January 2021) 18.Bevan J. GCC taxation. Aylesbury (UK):Rothmans Exports Limited;1992 (https:// www.industrydocuments. ucsf.edu/docs/yxpy0198,accessed 29 September 2020). 19.Morton IT. Note from IT Morton to I Birks regarding agreement on specific taxation;8 February 1992 (https://www.industrydocuments.ucsf.edu/docs/#id=nxpy0198, accessed 29 September 2020). 20.Galgut R. Oman. BAT (UK and Export) Limited;16 April 1992 (https://www. industrydocuments.ucsf. edu/docs/xxpy0198,accessed 29 September 2020). 21.Common excise tax agreement of the states of the Gulf Cooperation Council [website]. Abu Dhabi:UAE Ministry of Finance;2016 (https://www.mof.gov.ae/En/lawsAndPolitics/ govLaws/Documents/ Common%20Excise%20Tax%20Agreement.pdf,accessed 29 September 2020). 22.Brock B,Choi K,Boyle RG,Moilanen M,Schilo B. Tobacco product prices before and after a statewide tobacco tax increase. Tob Control. 2016;25(2):166-173. 23.Hanson A,Sulivan R. The incidence of tobacco taxation:evidence from geographic micro- level data. Natl Tax J. 2009;62(4):677-98. 24.Delipala S,O􀆳Donnel O. Estimating tax incidence,market power and market conduct:the European cigarette industry. Int J Ind Organ. 2001;19(6):885-908. 25.Keeler TE,Hu TW,Barnett PG,Manning WG,Hai-Yen S. Do cigarette producers price- discriminate by state? An empirical analysis of local cigarette pricing and taxation. J Health Econ. 1996;15:499-512. 26.Linegar DJ,van Walbeek C. The effect of excise tax increases on cigarette prices in South Africa. Tob Control. 2017;0:1-7 (https://tobaccocontrol.bmj.com/content/27/1/65, accessed17 February 2021). 27.Vilanti AC,Colins LK,Niaura RS,Gagosian SY,Abrams DB. Menthol cigarettes and the public health standard:a systematic review. Public Health. 2017;17:983 (https://www. ncbi.nlm.nih.gov/pmc/articles/ PMC5747135/,accessed 24 January 2021). 06 世界卫生组织烟草税政策和管理技术手册 28.Directive 2014/40/EU of the European Parliament and of the Council of 3 April 2014 on the approximation of the laws,regulations and administrative provisions of the member states concerning the manufacture,presentation and sale of tobacco and related products and repealing Directive 2001/37/EC:Article 7. Brussels:OJEU. 2014;57:1-38 (https://eur- lex.europa.eu/legal-content/EN/ TXT/PDF/? uri=OJ:L:2014:127:FULL&from= EN,accessed 24 January 2021). 29.Delipala S,Keen MJ. Product quality and the optimal structure of commodity taxes. J Public Econ Theory. 2006;8:547-554. 30.Obara T,Tsugawa S. The welfare comparison of ad-valorem tax and specific tax with both quality and quantity choice of a consumer. Theor Econ Lett. 2017;7:1 797-1 813 (https://www.scirp.org/journal/ paperinformation.aspx? paperid=79862,accessed 24 January 2021). 31.Anderson S,de Palma A,Kreider B. Tax incidence in differentiated product oligopoly. Journal of Public Econ. 2001;81(2):173-192. 32.Razzolini L,Shughart WF,Tolison RD. On the third law of demand. Econ Inq. 2003;41: 2. 33.Sobel RS,Garrett TA. Taxation and product quality:new evidence from generic cigarettes. J Polit Econ. 1997;105:880-887 (http://faculty.citadel.edu/sobel/Al%20Pubs% 20PDF/Generic%20Cigarettes. pdf,accessed 24 January 2021). 34.Nargis N,Fong GT,Chaloupka FJ,Li Q. The choice of discount brand cigarettes:a comparative analysis of International Tobacco Control (ITC) surveys in Canada and the United States (2002-2005). Tob Control. 2014;23:i86-i96 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/23/suppl_1/i86. ful.pdf,accessed 24 January 2021). 35.Chaloupka FJ,Peck R,Tauras JA,Xu X,Yurekli A. Cigarette excise taxation:the impact of tax structure on prices,revenues and cigarette smoking. Cambridge:National Bureau of Economic Research;2010 (NBER Working Paper 16287;https://www.nber.org/system/ files/working_papers/w16287/w16287. pdf,accessed 24 January 2021). 36.Cigarettes in India,2019. In:GlobalData [website]. London:John Carpenter House;2020 (restricted access). 37.Gilmore AB,Tavakoly B,Taylor G,Reed H. Understanding tobacco industry pricing strategy and whether it undermines tobacco tax policy:the example of the UK cigarette market. Addiction. 2013;108(7):1 317-1 326 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3746125/,accessed 24 January 2021). 38.Marsh L,Cameron C,Quigg R,Hoek J,Doscher C,McGree R,et al. The impact of an increase in excise tax on the retail price of tobacco in New Zealand. Tob Control. 2016; 25(4):458-634. 39.Brock B,Schilo BA,Moilanen M. Tobacco industry marketing:an analysis of direct mail coupons and giveaways. Tob Control. 2015;24:505-508. 40.Chaloupka FJ,Warner KE. The economics of smoking. Handbook Health Econ. 2000;1: 16第二章 烟草消费税政策 1 539-1 627. 41.Shang C,Chaloupka FJ,Fong G,Thompson M,O􀆳Connor RJ. The association between tax structure and cigarette price variability:findings from the ITC Project. Tob Control. 2015; 24:ii88-ii93 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/24/Suppl_3/ ii88.ful.pdf,accessed 24 January 2021) 42.Li Q,Hyland A,Fong GT,Jiang Y,Elton-Marshal T. Use of less expensive cigarettes in six cities in China:findings from the International Tobacco Control (ITC) China Survey. Tob Control. 2010;19(suppl 2):i63-i68 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/19/Suppl_2/i63.ful. pdf,accessed 24 January 2021). 43.Hyland A,Laux FL,Higbee C,Hastings G,Ross H,Chaloupka FJ,et al. Cigarette purchase patterns in four countries and the relationship with cessation:findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control. 2006;15(suppl III):ii59-ii64 (https://tobaccocontrol.bmj.com/ content/tobaccocontrol/15/suppl_3/ ii59.ful.pdf,accessed 24 January 2021). 44.Tsai YW,Yang CL,Chen CS,Liu,TC,Chen PF. The effect of Taiwan􀆳s tax induced increases in cigarette prices and brand switching and the consumption of cigarettes. Health Econ. 2005;14:627-641. 45.Cummings KM,Hyland A,Lewit E,Shopland D. Use of discount cigarettes by smokers in 20 communities in the United States,1988-1993. Tob Control. 1997;6(suppl 2):S25-S30 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/6/suppl_2/S25.ful.pdf, accessed 24 January 2021). 46.Ross H,Tesche J. Undermining government tax policies:common strategies employed by the tobacco industry in response to increases in tobacco taxes. Economics of Tobacco Control Project,School of Economics,University of Cape Town and Tobacconomics,Health Policy Center,Institute for Health Research and Policy. Chicago:University of Ilinois at Chicago;2016 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC5747349/,accessed 26 January 2021). 47.Technical manual on tobacco tax administration. Geneva:World Health Organization;2010 (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng.pdf? sequence=1,accessed 26 January 2021). 48.Hiscock R,Branston JR,McNeil A,Hitchman SC,Partos TR,Gilmore AB. Tobacco industry strategies undermine government tax policy:evidence from commercial data. Tob Control. 2017;0:1-10 (https:// apps.who.int/iris/bitstream/handle/10665/44316/ 9789241563994_eng.pdf? sequence=1,accessed 26 January 2021). 49.Filippidis F,Laverty A,Hone T,Been JV,Milett C. Association of cigarette price differentials with infant mortality in 23 European Union countries. JAMA Pediatr. 2017; 171(11):1100-1106 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5710372/#: ~:text=Cigarette%20price%20increases%20across%20 23,priced%20cigarettes% 26 世界卫生组织烟草税政策和管理技术手册 20during%20this%20period,accessed 26 January 2021). 50.López-Nicolás A,Belén Cobacho M,Fernández E. The Spanish tobacco tax loopholes and their consequences. Tob Control. 2013;22(E1):e21. 51.Balester L,Auchincloss A,Robinson L,Mayne S. Exploring impacts of taxes and hospitality bans on cigarette prices and smoking prevalence using a large dataset of cigarette prices at stores 2001-2011,USA. Int J Environ Res Public Health. 2017;14(3):e318 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC5369154/,accessed 24 January 2021). 52.Liber A,Ross H,Omar M,Chaloupka FJ. The impact of the Malaysian minimum cigarette price law:findings from the ITC Malaysia survey. Tob Control. 2015;24:ii83-ii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/ii83.ful.pdf, accessed on 26 January 2021). 53.Feighery E,Ribisl K,Schleicher N,Zelers L,Welington N. How do minimum cigarette price laws affect cigarette prices at the retail level? Tob Control. 2005;14:80-85 (https:// tobaccocontrol.bmj. com/content/tobaccocontrol/14/2/80.ful.pdf,accessed 26 January 2021). 54.Tynan MA,Ribisi KM,Loomis BR. Impact of cigarette minimum price laws on the retail price of cigarettes in the USA. Tob Control. 2012;22:e1. 55.Huang J,Chriqui JF,DeLong H,Mirza M,Diaz MC,Chaloupka FJ. Do state minimum markup/price laws work? Evidence from retail scanner data and TUS-CPS. Tob Control. 2016;25:i52-i59 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/ i52.ful.pdf,accessed 26 Jan 2021). 56.Whitehead R,Brown L,Riches E,Rennick L,Armour G,McAteer J,et al. Strengths and limitations of tobacco taxation and pricing strategies. Edinburgh:NHS Health Scotland; 2018 (http://www. healthscotland.scot/media/1829/rapid-evidence-review-strengths- and-limitations-of-tobacco-taxation-and-pricing-strategies.pdf,accessed 27 January 2021). 57.Golden S,Farrely M,Luke D,Ribisl K. Comparing projected impacts of cigarette floor price and excise tax policies on socioeconomic disparities in smoking. Tob Control. 2016; 25:i60-i66 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/Suppl_1/i60. ful.pdf,accessed 27 January 2021). 58.Doogan NJ,Wewers ME,Berman M. The impact of a federal cigarette minimum pack price policy on cigarette use in the USA. Tob Control. 2018;27(2):203-208. 59.Marynak K,Xu X,Wang X,Holmes CB,Tynan MA,Pechacek T. Estimating the impact of raising prices and eliminating discounts on cigarette smoking prevalence in the United States. Public Health Rep. 2016;131:536-543 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4937114/#:~:text=We%20found%20 that%20each%20price,smokers% 20aged%20%E2%89%A526%20years.,accessed 27 January 2021). 60.McLaughlin I,Pearson A,Laird-Metke E,Ribisl K. Reducing tobacco use and access through strengthened minimum price laws. Am J Public Health. 2014;104(10):1844-1850 36第二章 烟草消费税政策 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4937114/#:~:text= We% 20found%20that%20each%20price,smokers%20aged%20 %E2%89%A526%20years.,accessed 27 January 2021). 61.European Commission v. French Republic,C-197/08,European Court of Justice 4 March 2019 (https:// eur-lex.europa.eu/legal-content/EN/ALL/? uri = CELEX% 3A62008CC0197,accessed on 27 January 2021). 62.Tobacco price promotion:policy responses to industry price manipulation. Boston:Public Health and Tobacco Policy Center;October 2011 (http://tobaccopolicycenter.org/ documents/Price%20 Promotion%20Overview%20FINAL.pdf,accessed 28 September 2020). 63.Golden SD,Holt Smith M,Feighery EC,Roeseler A,Rogers T,Ribisl KM. Beyond excise taxes:a systematic review of literature on non-tax policy approaches to raising tobacco product prices. Tob Control. 2016;25:377-385. (https://tobaccocontrol.bmj.com/ content/tobaccocontrol/25/4/377.ful.pdf,accessed 27 January 2021). 64.Branston J,Gilmore S. The case for Ofsmoke:the potential for price cap regulation of tobacco to raise 500 milion per year in the UK. Tob Control. 2014;23(1):45-50 (https:// www.ncbi.nlm.nih.gov/pmc/ articles/PMC3812939/,accesed 27 January 2021). 65.von Lampe K,Kurti M,Johnson J. “I􀆳m gonna get me a loosie” understanding single cigarette purchases by adult smokers in a disadvantaged section of New York City. Prev Med Rep. 2018;12:182-185. (https:// www.sciencedirect.com/science/article/pi/ S2211335518302195,accessed 27 January 2021). 66. de Ojeda A,Barnoya J,Thrasher JF. Availability and costs of single cigarettes in Guatemala. Nicotine Tob Res. 2013;15(1):83-87. 67.Lal P,Kumar R,Ray S,Sharma N,Bhattarcharya B,Mishra D,et al. The single cigarette economy in India -a back of the envelope survey to estimate its magnitude. Asian Pac J Cancer Prev. 2015;16(13):5 579-5 582 (http://journal.waocp.org/article_31295_ 9ebd3981652934bbf5d78ec066b6a39d.pdf,accessed 27 January 2021). 68.Coady MH,Chan CA,Sacks R,Mbamalu IG,Kansagra SM. The impact of tax on purchasing behaviors among New York City smokers. Am J Public Health. 2013;103(6): e54-e60 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC3698745/,accessed 27 January 2021). 69.Cigarettes in Thailand,2019. In:GlobalData [website]. London:John Carpenter House; 2020 (restricted access). 70.Cigars in Germany,2019. In:GlobalData [website]. London:John Carpenter House;2020 (restricted access). 71.Report from the Commission to the Council on Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels:European Commission;2018 (COM(2018) 17 final;https://ec.europa.eu/taxation_customs/sites/taxation/files/report _excise_duty_manufactured_ tobacco_12012018_en.pdf,accessed 28 September 2020). 46 世界卫生组织烟草税政策和管理技术手册 72.Galet CA,List JA. Cigarette demand:a meta-analysis of elasticities. Health Econ. 2003; 12(10):821-835. 73.Guidelines for implementation of article 6 of the WHO FCTC. Geneva:World Health Organization;2016 (https://www.who.int/fctc/guidelines/adopted/Guidelines_article_6. pdf,accessed 29 September 2020). 74.DeCicca P,Kenkel D,Liu F. Excise tax avoidance:the case of state cigarette taxes. J Health Econ. 2013;32(6):1 130-1 141 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3855906/,accessed 27 January 2021). 75.Jawad M,Lee JT,Glantz S,Milett C. Price elasticity of demand of non-cigarette tobacco products:a systematic review and meta-analysis. Tob Control;2018;27(6):689-695 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6056338/,accessed 27 January 2021). 76.John RM. Price elasticity estimates for tobacco products in India. Health Policy Plan. 2008;23(3):200-209 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2800993/, accessed on 27 January 2021). 77.Wasserman J,Manning WG,Newhouse JP,Winkler JD. The effects of excise taxes and regulations on cigarette smoking. Journal of Health Econ,Elsevier. 1991;10(1):43-64. 78.Cheng K,Kenkel D. U.S. cigarette demand:1 944-2 004. B.E. Journal of Econ Analysis & Policy. 2010;10(1):10.2 202/1 935-1 682.2 438 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC3820165/,accessed 28 January 2021). 79.Bobak M,Jha P,Nguyen S,Jarvis M. Poverty and smoking. In:Jha P,Chaloupka F, editors. Tobacco control in developing countries. Oxford:Oxford University Press;2000. 80.Peck,RM. Equity issues,tobacco,and the poor. Yurekli A,de Beyer J,editors. Washington (DC):World Bank Group;2013. (Economics of Tobacco Toolkit,Tool 6;https:// openknowledge.worldbank.org/ handle/10986/16265,accessed 28 January 2021). 81.Kenkel DS,Schmeiser MD,Urban C. Is smoking inferior?:evidence from variation in the earned income tax credit. J Hum Res. 2014;49:1 094-1 120. 82.Ruhm,CJ. Healthy living in hard times. Journal of Health Econ. 2005;24(2):341-363. 83.Ruhm,CJ. Are recessions good for health? The Quarterly Journal of Econ. 2000;115(2): 617-650. 84.Tarantilis F,Athanasakis K,Zavras D,Vozikis A,Kyriopoulos I. Estimates of price and income elasticity in Greece. Greek debt crisis transforming cigarettes into a luxury good:an econometric approach. BMJ Open. 2015;5(1):e004748. (https://bmjopen.bmj.com/ content/bmjopen/5/1/e004748.ful.pdf,accessed 28 January 2021). 85.Kaiser M,Reutter M,Sousa-Poza A,Strohmayer K. Smoking and the business cycle: evidence from Germany. Discussion Paper Series. Bonn,Germany:IZA Institute of Labour Economics;2017 (http:// ftp.iza.org/dp10953.pdf,accessed 28 September 2020). 86.Tingum EN,Mukong AK,Mdege N. The effects of price and non-price policies on cigarette consumption in South Africa. Tob Induc Dis. 2020;18(July):62 (https://www.ncbi. 56第二章 烟草消费税政策 nlm.nih.gov/pmc/articles/ PMC7398599/,accessed 28 January 2021). 87.Vardavas CI,Filippidis F,Agak,I,Mytaras V,Bertic M,Connoly GN,et al. Tobacco taxation:the importance of earmarking revenue to health care and tobacco control. Tob Induc Dis. 2012;10(1):21 (http://www. tobaccoinduceddiseases.org/Tobacco-taxation- the-importance-of-earmarking-the-revenue-to-health-care-and-tobacco,66110,0,2.html, accessed 28 January 2021). 88.Marquez P,Catinkaya V. Tobacco taxation in Turkey:an overview of policy measures and results. Washington (DC):World Bank Group;2017 (https://openknowledge.worldbank. org/handle/10986/26387,accessed 10 November 2020). 89.Blecher EH,van Walbeek CP. An international analysis of cigarette affordability. Tob Control. 2004;13:339-346 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/13/ 4/339.ful.pdf,accessed 28 January 2021). 90.Goodchild M,Sinha P,Munish VG,Tulu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018:a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73-81 (https://apps.who.int/iris/ handle/10665/331836,accessed 28 January 2021). 91.John RM,Dauchy E. Trends in affordability of tobacco products before and after the transition to GST in India. Tob Control Online First. 20 March 2020. 92.Chaloupka FJ,Drope J,Siu E,Vulovic V,Stoklosa M,Mirza M,et al. Tobacconomics cigarette tax scorecard. Chicago:Health Policy Center,Institute for Health Research and Policy,University of Ilinois at Chicago,2020 (https://tobacconomics.org/files/research/ 636/uic-tobacco-scorecard-report-final.pdf,accessed 28 January 2021). 93.Kanbur R,Keen MJ. Jeux sans frontieres:tax competition and tax coordination when countries differ in size. Am Econ Rev. 1993;83(4):877-892 (https://www.researchgate. net/publication/4901220_Jeux_ Sans_Frontieres_Tax_Competition_and_Tax_Coordination _When_Countries_Differ_in_Size,accessed 28 January 2021). 94.Bouw A. Tobacco taxation in the European Union:an overview (English). Washington (DC):World Bank Group;2017 (http://documents.worldbank.org/curated/en/ 493581492415549898/Tobacco-taxation-in-the-European-Union-an-overview,accessed on 28 September 2020). 95.Blecher E,Drope J. The rewards,risks and chalenges of regional tobacco tax harmonization. Tob Control. 2014;0:1-5. 96.Directive No.n° 01/2017/CM/UEMOA of 22 December 2017. Paris:Droit-Afrique;2017 (https://www. droit-afrique. com/uploads/UEMOA-Directive-2017-01-droits-accise- tabacs.pdf,accessed 29 September 2020). 97.Mirzoev T,Zhu L,YangY,Pescatori A,Matsumoto A,Calen T. The future of oil and fiscal sustainability in the GCC region. Middle East and Central Asia Research Department. Washington (DC):International Monetary Fund;2020 (https://www.imf.org/-/media/ Files/Publications/DP/2020/English/FOFSGCCEA. ashx,accessed 28 January 2021). 66 世界卫生组织烟草税政策和管理技术手册 98.State cigarette excise tax rates & rankings. Washington (DC):Campaign for Tobacco-Free Kids;2020 (https://www.tobaccofreekids.org/assets/factsheets/0097.pdf,accessed 28 September 2020). 99.A map and table comparing cigarette prices in Canada (March 2018). Toronto:Non- Smokers􀆳 Rights Association/Smoking and Health Action Foundation;2018 (https://nsra- adnf.ca/wp-content/ uploads/2018/03/180302_map_and_table.pdf,accessed 4 February 2021). 100.European Commission excise duty rates on manufactured tobacco. Brussels:European Commission;31 December 2017 (https://ec.europa.eu/taxation_customs/business/ excise-duties-alcohol-tobacco-energy/excise-duties-tobacco_en,accessed 1 July 2018). 101.Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels:European Commission;2019 (https://ec.europa.eu/ taxation_customs/sites/taxation/ files/study-on-the-tobacco-taxation-directive-2019_en. pdf,accessed 28 September 2020). 102.Liber AC. Heated tobacco products and combusted cigarettes:comparing global prices and taxes. Tob Control;2018;28(6). 103.Heated tobacco products:information sheet -2nd edition. Geneva:World Health Organization;2020 (https://www.who.int/publications/i/item/WHO-HEP-HPR-2020. 2,accessed 28 January 2021). 104.Kass Lempert L,Glantz S. Analysis of FDA􀆳s IQOS marketing authorisation and its policy impacts. Tob Control 2020;0:1-9. 105.Chivers E,Janka M,Franklin P,Mulins M,Larcombe A. Nicotine and other potentialy harmful compounds in “nicotine-free” e-cigarette liquids in Australia. Med J Aust. 2019; 210(3):118-119 (https://onlinelibrary. wiley.com/doi/epdf/10.5694/mja2.12059, accessed 28 September 2020). 106.Wenzl T,Zelinkova Z,Regueiro J,Giri A. Scientific-technical support activities to DG TAXUD-C-2 on the option to include e-cigarettes within the scope of excisable goods for the Impact Assessment on a possible revision of Directive 2011/64/EU. Administrative Arrangement TAXUD/2015/DE/329. Luxembourg:Publications Office of the European Union;2017 (https://publications.jrc.ec.europa.eu/ repository/bitstream/JRC105828/ kjna29062enn.pdf,accessed 28 September 2020). 107.Lisko JG,Tran H,Stanfil SB,Blount BC,Watson CH. Chemical composition and evaluation of nicotine,tobacco alkaloids,pH,and selected flavors in e-cigarette cartridges and refil solutions. Nicotine Tob Res. 2015;17(10):1 270-1 278. (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC4573955/,accessed 28 January 2021). 108.Ghebreyesus TA. Progress in beating the tobacco epidemic. Lancet. 17 August 2019; 394:548-49 (https:// www.thelancet.com/pb-assets/Lancet/pdfs/S0140673619317301. pdf? hss_channel=tw-27013292,accessed 29 September 2020). 109.Smoking cessation,a report of the Surgeon General. Atlanta:U.S. Department of Health 76第二章 烟草消费税政策 and Human Services,Centers for Disease Control and Prevention,National Center for Chronic Disease Prevention and Health Promotion,Office on Smoking and Health;2020. (https://www.hhs.gov/sites/default/ files/2020-cessation-sgr-executive-summary.pdf, accessed 8 October 2020). 110.E-cigarettes are harmful to health. Geneva:World Health Organization. 5 February 2020 (https://www. who. int/news-room/detail/05-02-2020-e-cigarettes-are-harmful-to- health,accessed 29 September 2020). 111.Shan L,Manzione LC,Azagba S. Psychological wel-being and dual-use of cigarettes and e- cigarettes among high school students in Canada. J Affect Disord. 2020;265:357-363. 112.King BA,Gammon DG,Marynak KL,Rogers T. Electronic cigarettes sales in the United States,2013-2017. JAMA. 2018;320(13):1 379-1 380 (https://jamanetwork.com/ journals/jama/fularticle/2705175,accessed 28 January 2021). 113.LaVito A. Popular e-cigarette Juul􀆳s sales have surged almost 800 percent over the past year [website]. CNBC. 2 July 2018 (https://www.cnbc.com/2018/07/02/juul-e- cigarette-sales-have-surged-over-the-past-year.html,accessed 28 September 2020). 114.Statement from FDA Commissioner Scott Gottlieb,M.D.,on new enforcement actions and a youth tobacco prevention plan to stop youth use of,and access to,JUUL and other e- cigarettes. Washington (DC):US Food and Drug Administration;23 April 2018 (www. fda.gov/NewsEvents/Newsroom/ PressAnnouncements/ucm605432.htm,accessed 29 September 2020). 115.Paul K. San Francisco becomes first US city to ban sale of e-cigarettes. The Guardian. 25 June 2019 (https://www.theguardian.com/us-news/2019/jun/25/san-francisco-e- cigarettes-vaping-juul-ban,accessed 29 September 2020). 116.Pesko MF,Courtemanche CJ,Maclean JC. The effects of traditional cigarette and e- cigarette tax rates on adult tobacco product use. J Risk Uncertain. 2020;60:229-258. 117.Cotti CD,Courtemanche CJ,Maclean JC,Nesson ET,Pesko MJ,Tefft N. The effects of e- cigarettes taxes on e-cigarettes prices and tobacco product sales:evidence from retail panel data. Cambridge:National Bureau of Economic Research;2020 (NBER Working Paper 26724;https://www.nber.org/ system/files/working_papers/w26724/w26724.pdf, accessed 28 January 2021). 118.Saffer H,Dench DL,Grossman M,Dave DM. E-cigarettes and adult smoking:evidence from Minnesota. Cambridge:National Bureau of Economic Research;2019 (NBER Working Paper 26589;https://www. nber.org/system/files/working_papers/w26589/ w26589.pdf,accessed 28 January 2021). 119.Huang J,Gwarnicki C,Xu X,Carabalo RS,Wada R,Chaloupka FJ. A comprehensive examination of own-and cross-price elasticities of tobacco and nicotine replacement products in the U.S. Prev Med. 2018;117:107-114 (https://www.ncbi.nlm.nih.gov/ pmc/articles/PMC6195827/,accessed 28 September 2020). 120.Zheng Y,Zhen C,Dench D,Nonnemaker JM. U.S. demand for tobacco products in a 86 世界卫生组织烟草税政策和管理技术手册 system framework. Health Econ. 2017;26:1 067-1 086. 121.Stoklosa M,Drope JM,Chaloupka FJ. Prices and e-cigarette demand:evidence from the European Union. Nicotine Tob Res. 2016;18(10):1 973-1 980. 122.Grace RC,Kivel BM,Laugesen M. Estimating cross-price elasticity of e-cigarettes using a simulated demand procedure. Nicotine Tob Res. 2015;17(5):592-598. (https://www. researchgate.net/ publication/270291421_Estimating_Cross-Price_Elasticity_of_E- Cigarettes_Using_a_Simulated_ Demand_Procedure,accessed 28 January 2021). 123.Huang J,Tauras J,Chaloupka FJ. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob Control. 2014;23:ii41-ii47 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/23/suppl_3/ii41.ful.pdf, accessed 17 February 2021). 124.Cheng KW,Chaloupka FJ,Shang C,Ngo A,Fong GF,Borland R,et al. Prices,use restrictions and electronic cigarette use—evidence from wave 1 (2016) US data of the ITC Four Country Smoking and Vaping Survey. Addiction. 2019;114(suppl 1):115-122 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC6661221/,accessed 28 January 2021). 125.Liber AC,Drope JM,Stoklosa M. Combustible cigarettes cost less to use than e- cigarettes:global evidence and tax policy implications. Tob Control. 2016;0:1-6 (https:// www.researchgate.net/ publication/299473181_Combustible_cigarettes_cost_less_to_use _than_e-cigarettes_Global_ evidence_and_tax_policy_implications,accessed 28 January 2021). 126.Chaloupka FJ,Sweanor D,Warner KE. Differential taxes for differential risks — toward reduced harm from nicotine yielding products. N Engl Journal Med. 2016;373(7):594- 597. 127.Goniewicz ML,Knysak J,Gawron M,Kosmider L,Sobczak A,Kurek J,et al. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob Control. 2014;23:133-139. 128.Sweet L,Brasky TM,Cooper S,Doogan N,Hinton A,Klein EG,et al. Quitting behaviors among dual cigarette and e-cigarette users and cigarette smokers enroled in the tobacco user adult cohort. Nicotine Tob Res. 2019;21(3):278-284 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6379027/,accessed 28 September 2020). 129.Wils TA,Knight R,Wiliams RJ,Pagano I,Sargent JD. Risk factors for exclusive e- cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics. 2014; 135(1):e43-e51 (https://www. ncbi.nlm.nih.gov/pmc/articles/PMC4279062/, accessed 28 January 2021). 130.Electronic Nicotine Delivery Systems:report by WHO. Geneva:World Health Organization;2014. (FCTC/ COP/6/10 Rev.1;https://apps.who.int/gb/fctc/PDF/ cop6/FCTC_COP6_10Rev1-en.pdf,accessed 28 January 2021) 131.WHO report on the global tobacco epidemic 2017:monitoring tobacco use and prevention 96第二章 烟草消费税政策 policies. Geneva:World Health Organization;2017 (https://www.who.int/tobacco/ global_report/2017/en/,accessed 28 January 2021). 132.Regulation of the Minister of Finance of the Republic of Indonesia,number 146,PMK. 010/2017 on excise tax rates of tobacco products. Jakarta:Ministry of Finance Indonesia; 2017. 133.Gourdet CK,Chriqui JF,Chaloupka FJ. A baseline understanding of state laws governing e-cigarettes. Tob Control. 2014;23:ii37-ii40 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC4078672/,accessed 28 January 2021). 134.An act to ensure that nicotine products are taxed equaly. 127th Maine Legislature:first regular session-2015. H.P. 670,Legislative Document No. 973. 2015 (https://www. mainelegislature.org/legis/bils/getPDF.asp? paper= HP0670&item =1&snum = 127&PID=,accessed 28 January 2021). 135.Evaluation of the Council Directive 2011/64/EU of 21 June 2011 on the structure and rates of excise duty applied to manufactured tobacco. Commission Staff Working Document. Brussels:European Commission;2020 (SWD(2020) 32 final;https://ec.europa.eu/ taxation_customs/sites/taxation/files/10-02-2020-tobacco-taxation-report.pdf,accessed 28 September 2020). 136.Electronic nicotine delivery systems and electronic non-nicotine delivery systems (ENDS/ ENNDS):report by WHO. Geneva:World Health Organization;2016. (FCTC/COP/7/ 11;https://www.who.int/ fctc/cop/cop7/FCTC_COP_7_11_EN.pdf,accessed 28 September 2020). 137.E-cigarette use among youth and young adults:a report of the Surgeon General. Atlanta: US Department of Health and Human Services,Centers for Disease Control and Prevention,National Center for Chronic Disease Prevention and Health Promotion,Office on Smoking and Health;2016 (https://e-cigarettes. surgeongeneral.gov/documents/2016 _SGR_Ful_Report_non-508.pdf,accessed 28 September 2020). 138.Wagener TL,Floyd EL,Stepanov I,Driskil LM,Frank SG,Meier E,et al. Have combustible cigarettes met their match? The nicotine delivery profiles and harmful constituent exposures of second-generation and third-generation electronic cigarette user. Tob Control 2017;26:e23-e28 (https://tobaccocontrol. bmj.com/content/ tobaccocontrol/26/e1/e23.ful.pdf,accessed 28 September 2020). 139.Vapour products data. In:Euromonitor International [website]. London:Euromonitor International;2019 (restricted access). 140.Global market data:segment insights. In:GlobalData [website]. London:John Carpenter House;2020 (restricted access). 07 世界卫生组织烟草税政策和管理技术手册 附录2.1 表A2.1 2018年采用不同类型卷烟消费税结构的国家 从量税 (65 个国家/地区) 从价消费税 (42个国家) 混合消费税 (63个国家) 无消费税 (15个国家) 阿尔巴尼亚、安道尔、澳大 利亚、阿塞拜疆、巴巴多斯、 白俄罗斯、伯利兹、玻利维 亚(别呤民族)、布隆迪、加 拿大、库克群岛、多米尼加、 厄瓜多尔、埃斯瓦蒂尼、斐 济、冈比亚、洪都拉斯、冰 岛、印度、印度尼西亚、牙买 加、日本、约旦、哈萨克斯 坦、肯尼亚、基里巴斯、吉尔 吉斯斯坦、莱索托、马来西 亚、毛来西亚、蒙古、莫桑比 克、缅甸、纳米比亚、尼泊 尔、新西兰、尼加拉瓜、挪 威、巴洛、巴布亚新几内亚、 秘鲁、菲律宾、韩国、圣卢西 亚、圣文森特和格林纳丁 斯、萨摩亚、塞舌尔、新加 坡、所罗门群岛、南非、斯里 兰卡、苏里南、塔吉克斯坦、 东帝汶、汤加、特立尼达和 多巴哥、乌干达、坦桑尼亚、 美国、乌拉圭、乌兹别克斯 坦、瓦努阿图、也门、津巴布 韦 阿根廷、亚美尼亚、 巴林、孟加拉国、贝 宁、布基纳法索、卡 波佛得角、柬埔寨、 喀麦隆、乍得、科摩 罗、科特迪瓦、古巴、 刚果民主共和国、赤 道几内亚、厄立特里 亚、埃塞俄比亚、加 蓬、加纳、格林纳达、 危地马拉、几内亚比 绍、利比比亚、马达 加斯加、马里、毛里 塔尼亚、尼日尔、巴 拿尔、巴巴拉马、圣 基茨和尼维斯、沙特 阿拉伯、塞内加尔、 塞拉利昂、苏丹、叙 利亚、多哥、多库、土 库曼斯坦、图瓦卢、 阿拉伯联合酋长国、 委内瑞拉、越南、赞 比亚 阿尔及利亚、奥地利、比利 时、波斯尼亚 黑塞哥维那、 哥茨瓦纳、巴西、保加利亚、 中非共和国、智利、中国、哥 伦比亚、刚果、哥斯达黎加、 克罗地亚、塞浦路斯、丹麦、 多米尼加共和国、埃及、萨 尔瓦多、德国、爱沙尼亚、芬 兰、法国、希腊、匈牙利、伊 朗、爱尔兰、以色列、意大 利、老挝、拉巴嫩、立陶宛、 卢森堡、黎巴嫩、卢森堡、马 耳他、墨西哥、黑山、摩洛 哥、荷兰、尼日利亚、北马其 顿、波兰、葡萄牙、摩尔多瓦 共和国、罗马尼亚、俄罗斯 联邦、卢旺达、圣多美和普 林西比、塞尔维亚、斯洛伐 克、斯洛文尼亚、西班牙、瑞 典、瑞士、泰国、突尼斯、土 耳其、乌克兰、英国、约旦河 西岸和加沙地带 阿富汗、安哥拉、 安 提 瓜 和 巴 布 达、朝鲜民主主 义人民共和国、 伊拉克、科威特、 利比亚、马尔代 夫、马绍尔群岛、 密 克 罗 尼 西 亚 (联 邦)、瑙 鲁、 Niue、阿 曼a、卡 塔尔a、索马里 a 本表显示了截至2018年7月的相关国家/地区卷烟消费税结构的状况,并没有考虑到该日期之后发生的变 化,特别是卡塔尔和阿曼,它们分别在2019年1月和2019年6月引入了烟草消费税。 来源:世卫组织RGTE。 17第二章 烟草消费税政策 附录2.2　税基弹性的分析 假设税收R=tsQ 或R=tvPQ,其中ts 为从量税,Q 为消费量,tv 为从价税,P 为消 费价格。 以下方程可以帮助说明税基弹性的不同组成部分。 在从量税制度下,收入的变化主要取决于消费的变化: ∂R ∂ts =Q(1+ηs) 其中R 为烟草税收入,为具体消费税,反映烟草税基弹性。 税基弹性包括: ηs=ε ∂Pts ∂tsP 其中,ε为价格弹性, ∂P ∂ts 是提高的从量税传导转稼到消费者价格的程度,ts P 是税收价 格比率。 在从价消费税制度下,收入的变化基本上取决于烟草支出的变化: ∂R ∂tav =PQ(1+ηav) 其中R 为烟草税收入,tav 为从价消费税,ηav 为烟草税基弹性。 这里的税基弹性包括: ηav= ∂Ptav ∂tavP (1+ε) 其中∂P ∂tav 是从价消费税对消费者价格提高的通过程度,tav P 是税价比,ε是价格弹性, ε= ΔQP ΔPQ . 27 世界卫生组织烟草税政策和管理技术手册 附录2.3　ENDS/ENNDS产品设备的构成 任何ENDS/ENNDS套件的组成主要包括: • USB充电器(非汽车充电器)。 • 内置电池。 此外,对于开放系统: • 带有可拆卸雾化器的罐(可移动容器,通常与雾化器捆绑销售)。 • 清除雾化器/可再填充的吊舱(不可拆卸的雾化器)。 • 电子烟油。 对于封闭系统: • 一次性电子烟:不可充电,抽完电子液体后扔掉。 • 非一次性电子烟: • 预填充雾化器(随雪茄套件设计的药盒)。 • 预装油箱补充/吊舱(与预装油箱/吊舱组件一起设计的吊舱或药盒)。 一些定义: • 喷雾器:使用一个加热元件来蒸发电子液体。 • 雾化器:结合药盒和雾化器。 • 清除器:与清除雾化器相同,使用不同的技术。 • 药盒:承装电子烟油的容器。 综上所述,ENDS/ENNDS产品设备包括: • USB充电器(非汽车充电器)。 • 内置电池。 • 一次性电子烟。 • 调制解调器/清除器。 • 带或不带雾化器的药盒。 • 预装盒(用于封闭系统,承装电子烟油)。 • 电子烟油(在开放式系统的药盒中添加)。 资料来源:ECigInteligence,2020年。信息也从相关网站获得,包括https://www. misthub.com/blogs/vape-tutorials/76788357-tutorial-atomizer-vs-cartomizer-vs-clearomizer, http://www.bestclearomizer.com/clearomizer-vs-cartomizer-vs-atomizer/,https://wayofleaf. com/accessories/vapes/atomizer-vs-clearomizer-vs-cartomizer,htps://wayofleaf.com/accessories/ vapes/atomizer-vs-clearomizer-vs-cartomizer,2020年7月15日访问。 37第二章 烟草消费税政策 第三章 烟草税的管理 3.1　简介 对烟草制品征收消费税通常不止一个目的。政府经常发现自己需要在财政目标和公 共卫生目标之间平衡利益。这两个目标最好可以通过一个高效、有效、具有强大执行力和 征税专业能力的主管机关来实现。主管机关是为完成某一特定活动依法设立的机构、组 织或部门。在实行烟草税的情况下,主管机关通常是税务管理部门、税务当局、海关部门 或财政部。 世卫组织《烟草控制框架公约》(FCTC)[1]第6条及其准则[2]为健全的税收管理提供 了坚实的基础。如指南第1.5节所述,烟草税制度应是高效和有效的。它们的结构应尽 量减少合规性和管理成本,同时确保增加所需的税收水平并实现健康目标。高效和有效 地管理烟草税制度提高了税收合规和税收征收,同时减少了逃税和非法贸易的风险。 税收管理的效率是指使每单位税收收入的成本最小化,它是通过比较所使用的资源 与所产生的收入来衡量的。税收管理的有效性是指高水平的合规性,也称为纳税人履行其 义务。因此,一个高效和有效的主管机关以最低的成本征收税款,同时确保政策符合规则。 烟草税是减少烟草使用的唯一且最有效的烟草控制措施,最好作为综合烟草控制计 划的一部分来实施[3]。非法贸易(包括走私、非法制造和避税)破坏了税收政策及其目标 的有效性[4]。反对增税的人经常提到增税对非法贸易的影响,他们认为增税会增加非法 贸易。他们认为,非法贸易会导致政府收入减少和价格下降。主管机关面临的挑战是确 保对在辖区生产和/或进口的所有烟草制品申报和征收应缴税款,同时检测在辖区非法生 产和/或进口的烟草制品,停止此类活动并起诉责任方。 本章描述了良好的税务管理部门的共同特征,包括基于国家经验的最佳做法。它经 常提到世卫组织的FCTC,并且鉴于税收管理和打击非法贸易造成的逃税的努力之间的 密切联系,广泛借鉴了《消除烟草制品非法贸易议定书》[5](以下简称《议定书》)。任何法 律禁止的与烟草制品的生产、运输、接收、占有、分销、销售或购买有关的做法或行为,包括 旨在促进该类活动的任何做法或行为,均被视为非法贸易(第1条)。《议定书》的目标是 消除和防止所有形式的烟草制品非法贸易。与此同时,《议定书》包括基于国际最佳做法 的烟草税管理措施,这使其涉及所有国家,甚至是非缔约方的国家。 该《议定书》于2012年缔约方第五届会议上通过,并于2018年9月25日生效。如序 言所述,它是为了应对日益增长的烟草制品国际非法贸易而制定的[5]。该议定书涵盖3 个主要领域:①控制供应链的措施(第三部分);②处理犯罪的措施,包括制裁(第四部分); ③国际合作(第五部分)。本章详细讨论了《议定书》的不同条款,第 3.4 节专门讨论控制 和执行。 3.2　制度安排 能够有效地征税的主管机关具有许多共同特征。这些机关的组织机构包含了明确界 定的角色、职责和相关机构之间协调的规则。此外,主管机关会定期收集数据并管理评估 风险所需的信息。成功风险管理的关键是在一国内部和国家之间的相关机关共享此信 息。有效和高效的主管机关还将根据关键绩效指标定期评估其绩效和实行问责制,以确 定需要改进的领域。以下各节将更详细地讨论这些特性。 3.2.1　明确界定主管部门的角色和职责 为执行税法指定主管机关(包括明确定义一个国家内许多机构之间的权力边界)对于 有效征税至关重要。还必须定义不同机构需要合作和共享数据的领域。不同机关的活动 重叠导致资源使用效率低下,且为欺诈创造了机会,导致税法无效。明确界定的角色和责 任的重要性不仅适用于税务机关和海关,也适用于执法机构,包括警察和边境管制部队。 不同国家的税收执行方式和执行组织都有所不同。最常见的结构是将海关和税收管 理分开。然而,自20世纪90年代以来,将这些职能合并为一个机构成为一种趋势,如英 国的税务海关总署(HMRC)、秘鲁的SUNAT 和阿根廷的AFIP 。一些国家通过建立 税收秘书处并实施将税收记录作为单一纳税人账户共享的系统,来加强税收和海关之间 的协调。税收政策与税务管理部门之间的协调也有所加强。人们可以考虑将两者合并成 财政部内的一个部门,或确保在税收政策过程中咨询税务管理机关。一些业务,如许可 证,可由卫生、农业或贸易部等其他部门处理。例如,文莱卫生部和新加坡卫生科学管理 局负责颁发烟草制品进口商的许可证[6]。在一些联邦国家,如哥伦比亚和美国,消费税 57第三章 烟草税的管理 最高法令061-2002-PCM-规定国家税务管理局-SUNAT与国家海关监管局 海关合并(2002年关于税收和海关 管理之间合并的最高法令)。利马:秘鲁,2002年7月12日(西班牙文,http://www.sunat.gob.pe/legislacion/ sunat/ds061-2002-PCM.pdf,访问日期为2020年11月13日)。 联邦英格雷索公共行政,第618/1997(联邦公共收入管理局,第618号法令)(西班牙语,http://servicios.infoleg. gob.ar/infolegInternet/an-exos/40000 44999/44432/norma.htm,2020年11月13日访问)。 (包括烟草税)由地方或州税务机关征收和执行。其他国家通过建立一个统一的税收机构 来组织国家税收的管理。特别是在较大的经济体中,该机构通常负责直接税和间接税,包 括消费税,并向财政部报告。有效和高效的税务管理所需的所有职能都在这些机构内建 立[7]。无论体制安排如何,各机构之间的合作和信息交流以及它们的权力在法律上找到 依据都是至关重要的。第 3.2.2 节提供了有关此主题的更多信息。 然而,许多国家都有独立的税收和关税征收机构。2015年,一项对全球135家税务 管理部门的调查发现,其中只有36%的税务机关同时负责税务管理和海关管理[8]。在大 多数国家,海关部门更有可能对进口产品征收消费税,而在许多国家,增值税或销售税与 烟草税共同征收,特别是对进口产品。通过统一通用的流程和程序来创造协同作用、简化 控制措施,从而为税务管理部门和纳税人节省成本。多个机构参与税收征收尤其需要良 好的合作和信息共享,以确保高效而有效地征收税收和关税。 关键要点1 具有明确定义的角色和职责的制度安排旨在防止出现重叠和漏洞,有助于有效和 高效的税务管理。 3.2.2　相关机构之间的有效协调 国家层面的协调 有关机构之间的协调是有效的烟草税管理的关键。这不仅意味着如上一节所述的明 确界定的角色和职责,而且还意味着主管机关、海关和负责制定、分析与实施税收政策的 部门之间的协调。无论制度安排如何,无论责任方都在财政部内还是在不同的政府机构, 各方都需要良好的合作和交换信息,以优化税收征收和税收政策的执行。在实践中,这意 味着信息应该在海关、发放许可证的地方政府单位和卫生部门之间共享,尤其是那些监管 烟草制品销售的部门。 对于税务机关来说,与消费税最相关的信息包括纳税人和参与烟草贸易的人的身份 (进出口数据、许可证、犯罪记录、纳税申报单、银行对账单等);制成品的类别、数量、价值 和位置;以及这些货物的流动,直到缴清所有税款。应牢记获取这些信息的法律障碍,如 银行保密或隐私法规,并在需要时,应将财政程序的例外情况纳入法律。缉获数据也是一 个有价值的信息来源,第3.4节提供了此方面的更多细节。 税务当局应定期与执法机构(如警察和边境控制部队,具体取决于国家法律)进行协 调,以适当监测与烟草有关的活动并执行税法。通常,主管税务机关和海关机关与反欺诈 团队会密切合作 。在立法或法规中可能需要协调和共享信息,以确保简化流程,避免混 淆。这可以根据需要临时进行,也可以通过正式计划的信息交流和定期会议进行。建议 至少建立和完善在政府机构之间交换或获取信息的法律依据,以防止在法律程序中提出 67 世界卫生组织烟草税政策和管理技术手册 例如,请参见“关注税务欺诈”。荷兰的税务和海关管理,2017 年 (https://customsnl-insight.nl/article/ 309563676,2020年10月3日访问)。 证据是非法的(没有提到索赔)。 一些国家不仅仅是交换信息和合作。例如,在荷兰,海关当局不仅为财政部门开展工 作,还为其他7个部门执行非财政任务,包括农业、自然和食品质量部,司法和安全部,以 及外交部[9]。这些活动往往是基于财政部和其他部门之间的双边协定。在其他国家,如 美国和加拿大,海关和边境保护部门不属于财政部的一部分,它们是美国国土安全部和加 拿大公共安全和应急准备部的一部分。这些机构还执行许多非财政任务。 随着实施新的烟草控制和税法,一些国家还设立了高级委员会,以确保良好的协调和 法律的执行。由卫生和财政部领导,委员会确保协调和微调,以达到预期的结果。博茨瓦 纳、智利、哥伦比亚、印度尼西亚和塞内加尔等国家已通过定期委员会会议成功地开始协 调、规划和监督烟草法的实施。这些委员会通常包括来自卫生部、财政部、税务部和海关 部、警察部、运输部以及在某些情况下的教育部的代表。 跨国界协调 控制烟草制品走私的有效方法需要在司法管辖区的边界进行干预,因此必须让边境 机构也参与进来。然而,随着贸易的全球化,不仅税务和边境管制机关之间需要密切合 作,而且在不同的司法管辖区之间也需要密切的合作。最近的案例表明,缺乏正式的合作 框架可能会使市场面临金融犯罪危险,包括洗钱和恐怖主义融资等[10]。 加入《议定书》、经济合作与发展组织(经合组织)、《税务事项行政互助多边公约》和其 他区域安排等国际合作协定,将大大有助于执法机构之间的有效信息交流和合作。来自 参与辖区的市场数据和信息的有效交换可以防止潜在的跨境犯罪和国内收入损失。国际 合作加强了制止非法贸易和增加急需收入的国内措施。《议定书》缔约方承诺相互合作并 共享信息以履行其在议定书下的义务(第 20 条)。《议定书》本身是允许缔约方跨境合作 和共享信息的法律文书。不是《议定书》缔约方或其他代表交换信息的法律文书的合作协 议缔约方的政府当局,可以缔结互助协议或信息交换协议,以指导制定有效进行信息交换 的程序。 世界海关组织(WCO) 颁布的 2010 年修订的京都公约建议签订双边协议的司法管 辖区要求其他司法管辖区提供运往其关税区的货物的抵达前的信息。2013年,一项对87 名 WCO成员的调查发现,绝大多数海关行政部门有与其他行政部门共享烟草制品供应 链相关信息的法律权力[11]。一些经济集团还制定了适用于其所有成员国的统一立法,以 提供行政支持,以便在税务问题上进行有效合作[12]。 协调可以包括建立一个专门机构以确保外部边界的安全和正常运作,例如欧洲边境 和海岸警卫队机构,也称为 Frontex(来自法语 frontières extérieures,外部边界)。在一 些由 Frontex 牵头的行动中,欧盟和非欧盟国家与国际组织合作打击跨境犯罪,包括走私 卷烟和生烟[13]。 从事烟草制品非法贸易的嫌疑人通常也在从事相关的违法活动,如贿赂、洗钱、腐败、 妨碍司法公正等,甚至为恐怖组织提供资金[14]。一些国际条约提供了通过从刑事司法角 度处理非法贸易的机制来打击这种行为的法律框架,例如《联合国打击跨国有组织犯罪公 约》、《联合国反腐败公约》和《禁止向恐怖主义提供资助的国际公约》。表3.1总结了这种 协调的结构类型。 77第三章 烟草税的管理 表3.1 关于协调机制的结构 协调类型 基 准 参与的角色 国家协调 各国机构之间有法律依据的协议 海关部门、财政部和负责制定、分析与实 施税收政策的部门;执法机构,如警察和 边境管制部队及反欺诈团队 各部门之间的协议或法律法规中关于设立 高级委员会的基础 卫生、财政、税务、司法、运输部,有时还 有教育和执法实体,如海关和警察 双边协调 双边合作协议 联合政府 区域协调 区域安排,如: · 适用于经济集团所有成员国的协调立法, 以提供税收行政合作,以在税务问题上进行 有效合作 · 共同设立一个特别机构,以确保外部边界 的安全和运作的规范 欧盟成员国、欧洲边境和海岸警卫队局 (Frontex)、海关、法律和边境执法机构 国际协调 具有跨国性的条约或公约,如: · 协议 · 经合组织《税务事项行政互助多边公约》 · 《联合国打击跨国有组织犯罪公约》 · 《联合国反腐败公约》 · 《禁止资助恐怖主义的国际公约》 国际条约和公约的缔约国、法律和边境 执法机构 关键要点2 无论制度安排如何不同,一个国家内部和跨司法管辖区的协调与合作对于优化税 收征管和税收政策的执行都是必不可少的。 3.2.3　绩效和问责制的评估 关键的战略指标有助于评估主管机关的绩效。绩效指标可包括净收入、与预算金额 相比的总支出、成本与收款的比率、申报和支付合规措施以及纳税人满意度等指标[15]。 包括国际货币基金组织、世界银行、美洲开发银行和经合组织在内的几个国际组织已经开 发了以关键绩效指标评估税收和海关的工具。本节提供了一些对衡量与烟草税有关的绩 效特别有用的指标信息,包括征收成本率、税收差距分析和税收目标。 征收成本比 各国的征收成本各不相同。征收成本比是预征收总额占征收净税收总额的百分比。 这一比率经常被用作衡量主管机关的效率和有效性的标准。在表3.2中,根据国际货币 基金组织的年度调查,按收入水平计算国家群体的征收成本。表中的数字显示了使用的 87 世界卫生组织烟草税政策和管理技术手册 资源和税收的收入。所有国家都采用了相同的征税成本定义。税收不包括进口产品的增 值税和消费税,因此它只反映了国内税:个人和企业所得税、国内生产的增值税和消费税。 关税也不包括在内。结果显示,不同收入水平的国家之间存在差异。其他影响因素包括 税收制度、经济状况和合规水平之间的差异。 表3.2 2015年每100单位的征收成本比率(平均经常性预算与征收收入的比率a) 分组(样本量) 2015 低收入国家(6) 1.3 中低收入国家(15) 1.6 中高收入国家(18) 0.9 高收入国家(36) 0.9 所有(75) 1.1 a不包括进口商品的增值税或消费税。 资料来源:参考文献8,附录表12。 正如人们所预期的那样,由于自动化和资源水平较低,低收入和中低收入国家的这一 比例更高,大于1.0(收集 100 个货币单位需要超过 1.0 个货币单位)。中高收入和高收 入国家的比率低于1.0,表明其拥有有效和/或更有效的收集系统。 征收的成本可能与所引入的税收无关,而不仅仅是财务目标,例如影响行为的变化。 特别是在对烟草制品征收消费税时,征税的成本并不能揭示全貌。如果消费税税率大幅 提高,或者至少高于通货膨胀率和收入增长,消费将会减少。因此,由于与烟草有关的死 亡率和发病率降低以及生产力提高,保健费用将会降低。这些节省并没有计入征收成本 与收入的比率中,但总体而言政府确实从这些较低的费用中受益。当然,征税成本可以作 为主管当局效率的指标。 税收差距分析 税收差距分析是确定烟草制品税收征收效率的另一种方法。税收差距是指应缴税款 和所征收的税款之间的差额。例如,对卷烟零售价格征收从价税的理论是一包卷烟的平 均价格乘以销售的包数(例如根据家庭支出调查估算)乘以税率。这个结果可以与实际征 收的税收进行比较[16]。 税收征收的有效性也可以通过使用宏观经济投入 产出矩阵来确定,以衡量经济部门 的附加值(这种情况下的烟草)和理论增值税,然后将结果与实际的增值税额进行比较。 该方法适用于衡量国内逃税情况(第4章第4.1节提供了关于使用该方法来估算非法贸 易的更多信息)。 税收收入目标 主管机关的表现也可以通过确定在一个特定的税收期间是否达到了税收收入目标来 评估。尽管收入预测经常被当作目标,但仍建议谨慎行事。预测收入可能包括经济增长、 97第三章 烟草税的管理 通货膨胀和税收等假设。然而,预测是一种良好的做法,主管机关应向政府提供预测收入 的投入,以提高估算的质量。主管机关应对照预测收入来监控实际收入情况,但由于各种 原因,理论收入可能无法实现。此外,税收目标可以激励一些海关和主管机关仅仅以达到 目标金额为努力方向,而不是努力利用现有资源收集尽可能多的税收。 3.3　税收征管流程 对于任何税种,都有相关的合规、控制和执行流程。税收征管流程通常包括注册和许 可、税务申报、记录保存、仓库存储、暂停征税、征税和退税。图3.1说明了税收征管流程 的主要节点。 注册和许可（续期） 许可活动，例如进口、生产、运输、存储、出口等 纳税申报 退还 纳税和征收 对税务申报单中提供 的信息的审计和控制 机构： 审计和控制 纳税人： 记录 1 2 3 456 图3.1 税收征管流程 3.3.1　注册和许可 除了规范和确保那些处理受管制货物或货主的人的诚信外,发放许可证的主要目标 是规范供应链。许可证是获取更多信息和确保烟草制品供应链安全的有力工具。 《议定书》缔约方承诺授权烟草制品和制造设备的制造、进出口(第6条)。此外,缔约 方承诺努力在适当的情况下授权参与烟草种植和烟草制品或制造的零售、运输、批发、中 介、仓储和分销的人员颁发许可证(第 6 条)。为了确保许可制度有效,缔约方应在适当 的情况下监测和收取可能征收的任何许可费用,并考虑将其用于许可制度的有效管理和 执行、公共卫生或任何其他相关活动。在可能的情况下,每一缔约方应要求零售商和烟草 种植者(除了在非商业基础上工作的传统种植者)根据其国家法律(第9.4条)保持他们所 参与的所有相关交易有完整和准确的记录。 《议定书》第6.3(b)条提供了该许可证申请人被要求提供的信息清单,包括: • 申请人的相关身份信息; • 生产单位或仓库的生产地点及生产能力; • 烟草制品和设备的详细清单; 08 世界卫生组织烟草税政策和管理技术手册 • 制造设备的安装和使用地点的说明; • 无犯罪记录的文件或声明; • 用于交易和付款的银行账户的信息; • 烟草制品的预期用途和预期销售市场的描述。 为了让当局更容易收集他们需要的所有信息,在许可过程中可以免除保密规则。 许可证可以是通用的(包括所有需要颁发许可证的活动),也可以为每个活动单独颁 发。例如不同的制造、进口和零售许可证。通用许可证对许可证颁发机构的责任较小,而 每种类型活动的许可证提供更大的控制权,但以更多的管理为代价[17]。实施许可制度的 成本应与该制度的潜在影响成正比。不仅应考虑许可证的类型,还应仔细考虑获得许可 证所需的过程和信息,以确保两者是否相称。过程所需的信息和系统对被许可人的义务 越严格,该制度对企业和必须管理和执行它的当局的责任就越大。收集的信息越多,合规 性和行政责任就越高。建议将信息的附加价值与额外的合规、行政和/或执行责任相平衡。 在确定严格程度时,应考虑到诸如活动的风险水平和执行能力的可用性等因素。对 于在税收收入潜在损失方面对政府构成更高风险的活动,例如进口、生产和处理尚未缴纳 消费税的应税产品,更严格的制度可能是必要的。政府可以考虑将许可费用设定在足够 高的水平,以支付管理和执行系统的成本。有关主要依赖许可和许可的系统的示例,请参 见方框 3.1 中的澳大利亚案例研究。 方框3.1 澳大利亚许可证的案例研究 澳大利亚采取了与许多国家不同的控制烟草税的方法。它没有使用财政标记或 跟踪和追查 。相反,它通过许可和基于许可的系统管理烟草税,旨在促进低风险实 体的运营,同时防止或严格控制涉及高风险实体的商业活动。国内税务机构澳大利 亚税务局(ATO)负责大部分控制。这些控制措施涵盖在澳大利亚种植或制造并作 为成品或烟叶进口的烟草。事实上,澳大利亚的合法烟草市场仅包括进口烟草成品。 2006年,ATO 取消了所有烟草种植许可证,因为制造商转而使用外部供应商提供的 更便宜的烟叶。最后一家国内卷烟制造商于2016年关闭,此后便没有合法的国内烟 草种植和卷烟制造。 ATO 承担与烟草和烟草制品进口相关的所有其他职能,包括用于储存进口产 品的保税仓库的许可,以及在许可的保税仓库之间或出口地点之间进行保税烟草制 品运输的许可。 进口商必须申请保税仓库的许可证来储存进口烟草。申请人必须满足一般标 准,例如适用性、记录保存和安全性 。这些标准旨在确保只有低风险实体才能进入 消费税系统。通过允许暂停或取消许可证的规定,风险水平也保持在可接受的水平, 但可上诉。许可证的有效期为三年,对于证明合规的许可证持有者,会自动续期。 18第三章 烟草税的管理 有关财政标记和跟踪和追查的详细讨论见第3.4.4和3.4.5节。 该实体不得根据《消费税法》或任何联邦、州或地区法被起诉,在过去12个月内罚款超过105000美元(或前10年 被定罪),在过去四年有遵守税法的历史,以前没有取消执照,有足够的财政资源,没有接受破产管理。 许可系统依赖于对商业记录的交易后审计。用于评估风险的标准包括义务责任 的大小、双方的合规记录以及转移到市场的可能性。当存在收入损失的感知风险时, 可能会拒绝申请或要求提供财务保证金。 烟草制品的出口也需要通过澳大利亚海关的出口申报程序。产品必须有经批准 的出口申报才能离开本国。 根据 2017年政府工作组的建议,自2019年7月1日起取消了免税保税烟草的 地位。此外,还引入了进口许可制度,禁止进口无证商业烟草。进口商必须在进口时 确定他们的关税责任并立即付款;没有可用的信用条款。在将烟草制品运到澳大利 亚之前,需要向澳大利亚海关全额支付关税和税款。 资料来源:参考文献20和21。 烟草制品的批发商、分销商和零售商也可能被要求在从事这些产品的贸易之前获得 许可证,这将使主管机关能够要求被告知与烟草制品的购买和销售有关的交易。此外,这 将使主管机关能够完成对整个供应链的审计跟踪,并获得将有助于税收和卫生决策者适 当和有效地监测烟草制品的数据。 各国政府还可以要求让包括农民在内的处理原材料或种植烟草的实体获得许可证。 如果烟农的许可证被认为是适当的并随后在一个国家被要求,那么烟农必须确定并登记 他们的农场区域和位置以获得许可证。要求农民获得许可证的好处是可以将合法供应链 的控制扩展到烟草制品原材料来源的识别。这也使得将未加工的烟草从合法供应链转移 到非法供应链变得更加困难。 许可证由世界各地的不同机构颁发。例如,在巴西由卫生监测机构负责颁发许可证。 经营者需要获得制造和仓储设施布局的批准才能经营。此外,他们必须证明他们将如何 遵守其他法律和法规。例如通过展示产品包装的设计,包括包装、纸箱和主箱;在制造商 能够获得许可证之前,必须确定工厂的位置;最后,进口生产烟草制品的机器需要许可 证[18]。 如果主管机关确定了申请人为获得许可证而必须提供的信息,则许可证可成为有用 信息的来源。这些信息可以包括数量、价格和烟草收获的处置方式,以及买家的身份。我 们建议采用一个有效的许可证制度收集信息,通过要求相关罪行的犯罪记录(如以前不遵 守烟草许可证或欺诈的犯罪记录),以确定申请人的身份等信息。 为避免生烟监测漏洞,烟叶进口商也可以获得许可证或至少被要求登记和报告进口 烟叶的数量、来源和销售信息。在一些国家,这些信息已经由税务机关以外的政府机构收 集,例如由农业部来收集。应通过立法和机构之间的协调来避免要求和报告的重复。如 果(小规模)农民的许可难以实施,那么供应链中的后续购买者(第一加工者)就可以获得 许可和监管。对于主管机关来说,许可第一加工者的执行负担通常较小,因为一般来说, 第一加工者的数量远少于种植者。例如,在欧盟,已经确定了 50 到 100 个第一加工者, 而农民却有 55000 名[19]。 各国还可考虑要求从事制造和进口或销售用于制造烟草制品的材料的个人或实体进 行登记,例如卷烟纸、烟叶、添加剂、黏合剂、醋酸盐或任何其他类型的用于卷烟的过滤嘴、 28 世界卫生组织烟草税政策和管理技术手册 水松纸和玻璃纸或塑料包装纸,以及用于卷烟包装和主盒的材料。此外,烟草制造商可能 被要求在购买这些材料之前获得许可证。《议定书》缔约方应根据关于对烟草制品制造必 不可少的关键投入物是否存在、是否可以确定并是否受到有效控制的研究来决定采取适 当的措施。 关键要点3 许可的目的是规范和保护供应链。它是获取信息以进行验证、进一步调查和审计 的强大工具。理想情况下,所有参与烟草种植和烟草制品零售、运输、批发、中介、仓储 和分销烟草制品或制造设备的人员都应获得许可。 许可请求 根据案例研究和最佳实践(包括在商品价值或暂停关税或税收很高的情况下管理保 税仓库的经验)可能需要以下类型的信息来获得许可证,特别是对于生产商、仓库和烟草 分销商: • 安装安全、生产和存储周边安全认证(可包括闭路电视 ); • 财务偿付能力证明; • 详细的在线、实时烟草制品和主要原材料库存记录,由税务部门管理; • 电子会计系统; • 业主和经理的详细名单; • 银行业务及其他财务记录; • 烟草制品交易的定期电子报告; • 随时征税入库权; • 强制性的电子纳税申报单和付款单; • 对烟草制品抵达前的强制性报关单; • 符合税务印花税制度的声明(如适用); • 对于涉及进出口的企业,授权经济经营者(AEO)认证; • 符合保证金或担保制度的证明; • 为合理的检查、跟踪和溯源成本提供资金许可。 如上所述,可以申请 AEO 认证作为许可过程的一部分。大多数海关部门都熟悉 AEO 的概念。AEO 原则由 WCO 创建,最初侧重于安全问题[22]。由于产品的特殊性, 建议为烟草供应链的经营者制定特殊的许可制度。对于尚未建立系统的国家,AEO 认证 可能是建立此类制度的起点。 AEO 被 WCO SAFE 标准框架[22]定义为参与国际货物运输的一方,无论其职能如 何,都已被国家海关部门或代表国家海关部门批准为符合 WCO 或同等供应链安全标准。 AEO 尤其包括制造商、进口商、出口商、经纪人、承运商、集运商、中间商、港口、机场、码头 运营商、综合运营商、仓储商和分销商。 38第三章 烟草税的管理 “闭路电视”一词通常用于描述监控摄像机系统。 多年来,在某些情况下,甚至自1970年以来,海关部门越来越多地参与国际贸易供应 链的安全。最近,海关部门在全球范围内制定了安全计划。AEO 是这些计划的一部分, 2005年,WCO 通过了 SAFE 标准框架。从那时起,许多交易者被要求进行大量投资才 能获得 AEO 身份,并且必须继续投资以维持该身份。 AEO方案也得到了贸易便利化协定的承认,这是一个由174个国家签署的多边协 定[23]。 一些区域集团进一步明确了 AEO 的标准,并在其网站上提供了清晰、结构合理的信 息,以指导和鼓励运营商申请 AEO 身份。这种做法的一个很好的例子是爱尔兰共和国 税务网站,其中包含方框 3.2 中显示的信息。 方框3.2 AEO:爱尔兰共和国的税务和海关 什么是AEO? AEO状态是由欧盟(EU)海关管理部门颁发的认证标准授权。它证明了经济经 营者符合以下方面的某些标准: • 安全和保障; • 管理商业记录的系统; • 遵守海关规定; • 财务偿付能力; • 能力或专业资格的实用标准。 这主要是一项贸易便利化措施,旨在认证可靠的运营商并鼓励国际供应链中的 最佳实践。作为 AEO,运营商可以受益于: • 全球公认的安全、可靠和合规的国际贸易商业伙伴; • 风险分析系统中的风险评分较低; • 如果进行物理控制,则优先处理; • 在联合海关合作协议下相互承认 AEO 计划,这可能会导致货物更快地通过 第三国边界; • 减少进出境摘要申报的数据集(这仅适用于 AEO 安全和安保); • 更容易获得简化的程序; • 减少或放弃综合担保。 AEO身份的条件适用于所有企业,无论规模大小。制造商、出口商、货运代理、 仓库管理员、清关代理、承运人和进口商都可以申请AEO身份。 来源:参考文献24。 3.3.2　数据收集、申报和会计核算 风险分析的有效性取决于可用数据的质量和可靠性,与烟草税相关的风险分析也是 如此。在许多国家,获取可靠数据可能是一项挑战,但大多数主管机关越来越多地使用电 子系统来收集和管理数据。世界上许多国家引入增值税极大地提高了可用于烟草税分析 48 世界卫生组织烟草税政策和管理技术手册 的数据的可用性,因为报告是在供应链上完成的,例如货物的价值、商品数量和交易日期。 大多数征收消费税的国家也有增值税制度。此外,越来越多的国家正在成为该《议定书》 的缔约方。随着《议定书》的实施,将有更多数据可用,因为除其他措施外,各国将有义务 实施具有报告要求的许可制度以及跟踪和溯源系统。第3.4节提供了有关跟踪和溯源系 统的更多信息。《议定书》的义务也将有助于监测烟草制品的库存。 理想情况下,参与烟草制品供应和分销链的所有实体都应获得许可,并被要求记录发 生的每一笔交易。由于这可能会给税务机关和纳税人带来负担,因此建议使用自动化和 电子系统以降低合规成本。这可能需要所有原材料、机械、制品和成品都有准确的库存系 统,也使对所需数据进行良好的记录保存更为重要。随着报告数据量的增加,将需要一个 良好的信息技术(IT) 系统。将 IT 用于定期纳税申报、会计、库存和财务数据对于获取准 确信息和降低整个报告系统的成本至关重要。 大多数国家现在都有一定程度的自动化促进数据分析。一个新兴趋势是使用由贸易 商开具的电子发票,作为税务管理的在线实时信息的一部分。各国通常首先在公用事业 公司使用电子发票,然后将使用扩展到大公司。电子发票最大限度地减少了纸张的使用, 有助于自动记录保存,并为税务管理提供准确、及时的交易信息。一些国家开始不要求公 司必须使用电子发票,后来强制其使用电子发票,特别是对于交易量大的大公司(包括烟 草业)。电子发票已在欧盟国家和几乎所有拉丁美洲国家以及几个亚洲国家成功实施。 为了验证信息是否准确,主管机关可以系统地与第三方信息(如来自银行、金融机构、 雇主)进行交叉核对,或将数据与其他政府机构登记册中的信息相匹配。交叉检查和数据 匹配的过程也可以实现自动化,以尽量减少管理工作量[25]。 关键要点4 可靠的数据对于有效的风险分析至关重要。虽然获取这些数据可能具有挑战性, 但电子系统可以通过自动化数据收集程序与不同来源的信息交叉检查来帮助工作人员 减轻负担。 3.3.3　保持记录 《议定书》缔约方承诺酌情要求所有从事烟草、烟草制品职业和制造设备供应链上的 个人或实体保持生产烟草制品所用材料的完整、详细和准确的记录(第9条)。相关信息 包括市场数量、趋势,对烟草制品的预测,以及在过境、转运和暂停关税的税务和海关仓库 中库存的烟草制品和制造设备的数量。这些信息应由参与供应链的个人和实体提供,并 按照法律的规定定期提交给主管机关。主管机关可以利用提交的信息来监测烟草法规的 遵守情况和纳税情况。具有这种详细级别的注册表实际上只能以电子形式保存。 记录必须对用于生产烟草制品的材料提供全面的概括。其目的是使税务当局和制造 58第三章 烟草税的管理 拉丁美洲的电子发票:西班牙文件的英文摘要;美洲开发银行,美洲税务管理中心,2018年(https://publications. iadb.org/publications/english/document/Electronic-Invoicing-in-Latin-America.pdf)。 商应能够将生产数量与生产中使用的投入进行协调,从而确保没有漏记录或非法生产。 还应对关键投入的供应商施加义务,以表明供应与需求[17]相称。 关键要点5 监控烟草、烟草制品供应链中的所有个人或实体的合规性和纳税情况,并要求相关 方对制造设备保存所有相关交易的完整、详细和准确的记录,以及用于生产烟草制品的 材料的详细信息。 3.3.4　仓储、储存和配送 根据《议定书》第6.2条,所有缔约方应努力向参与烟草和烟草制品或制造设备的批 发、中介、仓储或分销的人员发放许可证。维护一个授权系统,允许主管机关对生产和存 储设施进行控制,以确保纳税人纳税[2]。获得授权的审批过程可以包括对工厂或仓库的 布局、将要使用的机械以及生产、仓储和装运流程的评估,包括原材料和成品的入口和出 口点。监控生产和确保只有已纳税的产品从该场所流通到市场的基本方法是识别生产设 施和控制进出点。 主管机关应不时对其中所包含货物进行实物清点,以检查所有文件是否已适当准备 和批准,并确定所保存记录的准确性和完整性。如果司法管辖区要求在烟草制品上贴税 券,则只有贴有适当税票的产品才能销售。 一般来说,未缴纳必要税、未加盖财政标志的烟草制品不得进入仓库。出于实际原 因,许多国家允许暂停消费税,这意味着事先被授权的人可以生产、发送、接收和储存尚未 缴纳消费税的烟草制品。有关主管机关还可以要求已纳税的产品不应与暂停纳税的产品 储存在同一区域。显然,暂停缴纳消费税的产品面临高风险,这可能证明对生产、贸易、储 存和处理有更严格的要求。 澳大利亚拥有严格的许可系统和烟草制品运输许可要求。截至2019年7月1日,它 已经启用了一个新的系统,将保税仓库从供应链中删除。进口商必须拥有进口许可证,并 在进口后立即为卷烟支付消费税(见第3.3.1节中的方框3.1)。 3.3.5　关税暂停 许多国家/地区要求自然人或法人(作为授权仓库管理员)在其业务期间生产、加工、 持有、接收和发送需缴纳消费税的产品。需缴纳消费税的货物的生产、加工、持有、接收和 发送通常是在纳税前进行的,可以要求授权人员提供担保以确保其缴纳税款。这种系统 的特点可能包括给予授权的严格标准、仓库预授权访问、适当的库存控制措施、检查消费 税产品的来源和整个生产过程以及对产品进行编码和标记。使用计算机系统来监控消费 税暂停期间消费货物的流通也是一种措施。 也可以考虑针对暂停关税的产品颁发不同的许可证,这将使主管机关和运营商的执 法更容易,负担更轻。一般而言,建议仅在满足严格标准的情况下才允许在暂停关税的情 况下处理消费税产品。此类标准可包括预授权访问、充分的库存控制措施、检查消费产品 68 世界卫生组织烟草税政策和管理技术手册 的来源和整个生产过程以及对产品进行编码和标记。原则上,跟踪和追溯系统也应涵盖 烟草制品的流动。考虑到与这些产品相关的高风险,可以考虑进行额外的监控,例如在暂 停消费税的情况下监控记录消费货物流通情况的计算机系统。在设计此类系统时,建议 密切关注进出口海关程序,以确保一致并避免监管真空。 计算机系统的一个例子是欧盟的消费税移动和控制系统,该系统跟踪所有消费税产 品(包括烟草制品)的流通,这些产品尚未缴纳消费税。该系统实时记录流通情况,因此是 打击欺诈的重要工具。此外,该系统对于欧盟成员国相关主管机关之间的信息交流和合 作必不可少[26]。最后,在暂停消费税的情况下生产、进口或储存烟草制品之前需要获得 批准[27]。 关键要点6 暂停征收消费税的产品有更高的逃税风险,这可以证明采取严格措施是合理的,例 如要求保证支付税款、额外的许可要求、遵守计算机系统以监控暂停消费税产品的流通 以及现场授权和审核。 3.3.6　征收税款 为降低征税系统的复杂性,建议在制造、进口或从存储、生产仓库放行产品时记录数 量,以供消费时征收消费税,这是大多数征收消费税的国家的常用做法。在供应链的这一 层征税极大地限制了纳税人的数量,从而限制了控制他们所需的资源。鼓励纳税人使用 电子支付方式也可以增加征收所有税款的机会。 这同样适用于要求对某些高风险活动提供担保的情况,例如处理暂停关税的货物。 许多国家根据申请者的情况和常规业务活动在给定时间范围内所代表的风险水平(数量 或价值以及潜在的应缴消费税)逐案决定担保水平 。一些国家允许减少对具有良好合规 记录的运营商的担保。需要注意的是,担保不是对责任的限制,纳税人仍可能被要求并有 责任支付远高于担保水平的金额。 法律应要求在销售后的固定时间间隔或每月的固定日期缴纳税款[2]。许多国家都设 有专门的征收执法部门,专职负责征收税款。通过向发现有纳税人发送延迟申报或延迟 付款的消息和电话,凡有停止申报者或付款控制机构不合规行为时立即采取行动,这一点 很重要。这增加了保持纳税人合规的可能性。如果不申报或不付款持续存在,则可以执 行债券或担保政策。 在生产或进口前后征收消费税的另一个原因是在这些节点可以更有效地监控数量。 监测烟草制品供应链有多种选择,使用哪种监测系统取决于国家的财政、技术和人力资 源,最薄弱的监控形式是行业自我声明。验证合规性和确保收取全额应缴税款的活动可 以包括实物检查、审计,与第三方数据的交叉检查申报信息以及对管理和记录信息的检 查。 一般而言,在行政系统不完善的国家,强制合规是通过对生产或制造过程实施物理控 制来实现的。当消费税征收的官员有可能进行欺诈时,物理控制的成本就会增加。然而, 78第三章 烟草税的管理 当消费税官员在不同地点频繁轮换并且主管进行突然访问时,欺诈可以有显著减少。历 史上,一些国家(例如印度)已在生产厂家派驻税务管理人员以监控生产。在印度,主管当 局的一名工作人员全天候被派往卷烟和大型比迪烟制造工厂工作。每位官员记录每天的 产量和出厂的卷烟/比迪烟数量,并向下一位官员报告 。 更好的选择是远程监控生产。主管机关可以要求在整个制造和仓储设施的重要位置 安装闭路电视系统。有了这些,主管机关可以建立一个中央指挥所,从中可以持续监控和 记录设施和活动情况。此外,主管机关可以不时进行实物库存控制,如果实施电子发票, 则可以在发票和申报库存之间进行交叉检查。这也是防止主管机关工作人员与制造商或 进口商勾结的有效方法。例如,2015 年,菲律宾国税局要求所有烟草公司在其生产线和 仓库中安装闭路电视。做出这一决定是为了应对大量没收的未征税卷烟,目的是监控生 产以确保缴纳所有税款。 收集过程还必须得到 IT 系统的支持。这些系统必须提供透明度和准确性,以确保 从纳税人到税务库的付款流程安全。大多数国家都实施了自动电子系统,用于与国内和 进口的与每个申报相关的税收支付。如果适用,税务机关必须与银行系统达成全面协议, 以降低交易成本。一些国家实施了国家支付门户网站,允许公民在线支付税款和其他费 用,如汽车许可证和执照费以及农业、健康和环境费用。 关键要点7 应在制造、进口或从存储、生产仓库放行产品时记录数量,以供消费时征收消费税, 以确保有效监控数量。这也通过限制纳税人的数量以及控制他们所需的资源来降低税 收征收系统的复杂性。 3.3.7　退税 在消费税不出口的原则下,增值税、消费税和关税的退税在大多数国家是一个常见的 流程。退款频率和方式因国家/地区而异。每月退款是很常见的(如果期间有出口),退款 可能会直接发送给出口商或保留作为其他税款的抵免。一些出口量大的国家使用的替代 方案是所谓的零税率或暂停缴税,这意味着整个链条即从原料的进口、生产和包装直至出 口环节暂停间接税(增值税、消费税和关税)。该制度需要特殊的许可程序。由于烟草业 有出口业务,因此税务部门的退税过程需要特别注意税务管理。 3.4　控制和执行 控制和执行是税收管理的主要职能。事实上,大多数税法都包括“控制和执行税法” 的目标,对海关来说是“控制和执行边境纳税和税款”或类似的短语。《议定书》为烟草税 的控制和执行提供了指导。高效和有效的主管机关往往有一个战略计划来确保合规、一 88 世界卫生组织烟草税政策和管理技术手册 印度财政部,基于私人沟通获取的信息,2009年。 个基于风险的方法来确定链条中的问题点和将资源相应地导向高风险或高价值领域的能 力。 可以在控制和执行中发挥作用的任务包括控制注册和许可程序、尽职调查、核实申报 和征收税务工作。生产和分销控制,包括跟踪和追溯、财政标记、审计和进出口控制,都在 控制和执行中发挥了作用。本节介绍了改善控制和执法的主要活动,重点是烟草供应链。 还讨论了一旦发现非法烟草贸易就可以实施的程序和处罚政策。 3.4.1　控制和执行计划 战略计划 在现代税务管理中,通常会有一个以控制和执行为支柱的战略计划。适当控制合规 周期是保持纳税人遵守和防止非法贸易和避税的关键。因此,大多数税务管理部门将其 大部分资源集中在预防性政策上。其中的一些例子可以在英国和美国国税局(IRS)的战 略计划中找到。 在英国,HMRC多年来一直在制定完善的战略计划。该计划的一个关键支柱是保持 纳税人的合规性。这就是预防的概念:控制大多数纳税人最初的轻微不合规行为,同时对 不合规一方的少数人使用强有力的法律[28]。 美国国税局的战略计划也有类似的方法,重点是控制。如果检测到不合规问题,数据 分析和行为洞察力将被用来确定不合规问题的最佳方法。早期干预或自我纠正是解决检 测到的不遵守规定行为的方法。美国国税局还强调了解决不合规问题的重要性,以确保 纳税人对税收体系的信心和保护制度体系的完整性[29]。 基于风险的方法 在制定一个战略计划后,必须起草一个执行和控制计划。该计划应包括被执行的活 动的定义、被执行的纳税人和被执行的情况,以及为人员配备审计、基础设施和IT系统 资源。目标还必须确定,包括干预措施的数量和额外收取的收入或减少逃税的数额。一 些税务机关制订年度计划,定期审查,以改善结果,并根据需要纠正分配和目标,力争取得 更好的效果、更有效地利用资源、更低的纳税人成本和更有效地征税。换句话说,重点是 将干预措施放在那些有更高可能性不遵守规定的人身上。使用基于风险的方法可能特别 有益。 税收风险管理是现代税务管理中控制策略的一个关键要素。风险指对主管机关实现 有效和高效地征税等目标构成威胁。风险评估是分析风险和应对风险的最佳方法。从接 受到缓解再到规避,响应可能会有所不同。适当的风险评估使主管机关能够最有效地利 用其可用资源,并更有效地处理风险。它可以通过识别具有高风险的纳税人或活动类型 来提高合规性。具有相同特征的纳税人群体往往也有类似的风险。违规风险高的群体可 能会接受更大的审查。 烟草供应链中可能存在更大违规风险的领域包括进口、出口和进出仓库的转移,尤其 是在暂停关税的情况下。 从内部和外部来源收集风险相关信息是合规风险管理的最佳做法。这些来源可能包 括第三方信息(例如来自银行、信用卡公司、运输公司)、纳税人行为研究和合规问题研究、 98第三章 烟草税的管理 税收差距分析、税务审计和申报[30]。 风险管理使用这些不同的数据来源和算法来寻找高度不合规性的模式。风险分析也 可以表明风险降低。较低风险的领域可能需要更少的治理来确保合规性,这允许资源被 定向到其他地方。因此,风险评估可以帮助将有限的资源战略性地分配到风险最大的地 区,同时减轻低风险纳税人的负担。 风险管理一直由主管机关完成,但数据可用性和识别模式的统计方法已经改变了评 估风险的方式。尽管许多风险评估系统仍然是手动完成的或包含手动元素,但密集数据 技术的使用允许系统化、更深入和更有针对性的分析[31]。 现代风险评估利用纳税人、纳税额、其他税种申报(如增值税)和第三方信息的电子数 据。有了这些数据,税务机关可以进一步采取措施制定合规性指标。例如,增值税发票可 用于将报告的烟叶采购与烟叶批发商的销售发票相匹配。对于纳税人(即获得许可并提 供所需报告的纳税人),主管机关可以创建一个业务分析程序,以确定报告的数据在交易 的每一方是否一致。此外,在实行增值税制度的国家,主管机关可以将纳税人在增值税制 度下报告的数据与烟草消费税制度下报告的数据进行比较,以发现任何不一致之处。增 值税发票还可用于验证购入和销售数据。如果在供应链的各个层面征收增值税,政府就 更容易监控供应链以保证其缴纳消费税义务。 使用相同方法对烟草消费进行定期调查也可以获得有关消费税政策遵守程度的证 据。未反映在消费数据中的收入突然下降可能表明存在非法制造、非法进口、跨境购物或 买断情况。此外,缉获数据可以获得有关违规风险高的地区和活动的宝贵信息。 在进行风险分析时,还应考虑烟草税收政策的结构。如果消费税税率提高,则可能会 出现更大的买断或提前装载风险(请参阅本节后面关于反买断的讨论)。基于产品或包装 特征的不同消费税税率(例如软包装和硬包装或过滤嘴和非过滤嘴卷烟之间的区别)也容 易被经营者操纵,这可能会影响税收。减轻这些风险的一种方式是修改消费税政策并采 用统一的税率。 对于海关交易,风险管理是确定要检查的商品和支持申报单的关键要素。在1990年 之前,大多数海关机构使用随机标准来选择检查对象。从那时起,许多国家都采用了基于 风险的方法来选择检查。进口商和贸易社区的历史数据,辅以人工智能技术,表明风险管 理工具极大地提高了实物检查的效率。大多数现代海关机构都采用了此类技术,从而可 以实现更有效的控制流程,同时促进合规交易的流程更顺畅。方框 3.3 详细介绍了风险 管理流程的一些近期变化。 方框3.3 风险管理的变化 OECD 制定了图 3.2,以显示 2004 年理解合规风险的流程和关键步骤[32]。相 同的方法仍然用于识别、评估和确定风险的优先级。然而,许多主管部门现在使用新 技术和先进的数据分析,以及更多来源的信息,例如来自银行、雇主和增值税销售发 票的外部数据。 09 世界卫生组织烟草税政策和管理技术手册 经营环境 识别风险 评估和优先考虑风险 监控计划 的性能 分析依从性行为 （原因、治理方案） 计划和实施战略 确定解决策略 评估合规结果： 注册 归档 报告 付款 图3.2 合规性风险管理流程 来源:参考文献32。 识别风险和分析合规行为的方法也发生了变化。传统上,主管机关通过审计来确定 高风险案件。有了更多样化和更合理的数据,主管机关现在可以使用更多基于证据的方 法来检查风险模式。合规激活关系的成功现在通常是根据它们对整体合规环境的影响来 衡量的,而不仅仅是增加收入。 在印度尼西亚,合规风险管理流程的使用将非法贸易在卷烟总消费量中的份额从 12% 降低到 3%。在本章后面的“方框 3.13”中可以找到有关这方面的更多信息。 了解产品,以及供应链和分销链,可以让主管机关确定链条上的哪些区域风险最大, 因此需要更多的资源。附件 3.1 提供了有关选定烟草制品成分的详细信息。 关键要点8 风险分析有助于识别违规概率较高的干预点。具有针对性干预的基于风险的方法 可以获得更好的结果和更有效的资源利用,以确保税收征管的有效性。 3.4.2　控制烟草供应链 根据《议定书》第1条的定义,供应链包括烟草制品和制造设备的制造以及进出口,并 在一方决定时可扩展到以下一项或多项活动: 19第三章 烟草税的管理 有关具有多个指标和问题清单的有效风险管理的更多信息,请参见税务管理诊断评估工具。 1.烟草制品零售; 2.烟草种植,传统小规模种植者、农民和生产者除外; 3.运输商业数量的烟草制品或制造设备; 4.烟草和烟草制品或制造设备的批发、中介、仓储或分销。 《议定书》第4.1条要求各方“采取和实施有效的措施,以控制或规范烟草制品的供应 链,以防止、阻止、发现、调查和起诉此类商品的非法贸易,并为此目的相互合作”。下面将 进一步讨论规范供应链的具体措施,以及在这方面的最佳实践。 图 3.3 显示了供应链上报告和监控的主要场所:进口、出厂和从仓库中移除。与成 品进口商一样,制造商可能需要在边境报告进口投入。如果组件需要获得许可,则可能需 要在许可过程中提供相关信息。图3.3中的箭头代表运输,它也是供应链的一部分。 初级制造： 当地种植或进口的烟叶 被切碎、腌制和混合 进口卷烟： 来自卷烟进口商 卷烟生产： 将烟草卷成烟丝 卷烟包装： ·卷烟棒被放入包装中 ·财政标记，例如 作为税票贴在包装上 ·包装用玻璃纸包裹 ·包装被放入纸箱，然后放入主箱 分销商、批发商、仓库 零售商 1 2 图3.3 卷烟供应链从制造或进口到零售销售 3.4.3　许可证 和尽职调查 许可系统只有在得到适当控制的情况下才有效。大多数税务部门都有酒精饮料和能 源产品消费税许可程序的经验。强烈建议在实施和执行与烟草有关的许可时,应参照从 这些产品许可过程中汲取的经验。 许可提供了及时和准确的数据,可以作为审计的基础,因为它可以识别和控制合法的 运营商。对于新的运营商来说,获得许可证的过程可能包括参观和核实生产工厂、储存设 29 世界卫生组织烟草税政策和管理技术手册 这里将在尽职调查和执行的背景下讨论许可。关于如何设置许可以及可以要求提供哪些信息的详细信息见第 3.3.1节。 施和分销场所。没有许可证制度并希望开始申请许可证的国家,可以允许现有运营商有 一个过渡时期,以便其遵守新的许可证要求。许可证控制的过程必须定期进行和更新,特 别是通过控制债券或担保的有效性、所需系统的正常功能和记录保存。 在需要许可证的情况下,法律应强制规定,不得从无证供应商购买或向无证购买者销 售产品。这意味着供应商和买家都需要核实那些与他们有业务来往的人。这一要求为执 法机构提供了一个切入点,以在两端执行许可制度。此外,对制造设备的许可证要求有助 于主管机关查明和起诉非法制造烟草制品,从而大大减轻了举证责任。在许多国家,制造 设备的存在并不足以证明正在发生非法制造;只有当主管机关检查该地点时,机器必须正 在运行和生产非法烟草制品才能证明。然而,有了许可证要求,没有许可证的机器就足以 使主管机关采取行动。 许可证的有效性应有时间限制,到期后需要续期或重新申请,以保持高水平的控制。 遵守许可证所需的条件应由主管机关控制,对不遵守规定的处罚(例如暂停或吊销许可 证)应严重到足以起到威慑作用[33]。许可证条例应规定主管机关有权检查被许可方的产 品和场所,并对其违规行为进行处罚,其中可能包括对严重或重复犯罪的刑事和民事 起诉。 如《议定书》第6.3(a)条所述,缔约方需要建立或指定一个单一的权力机构或多个权 力机构来颁发、续期、暂停、撤销和取消许可证。 根据《议定书》第 7 条,法律要求参与供应链的人员在建立业务关系之前和期间进行 尽职调查。他们还必须向主管机关报告任何证据表明客户从事违反《议定书》规定的义务 的活动。该要求包括客户识别、销售监控以确保数量与预期市场内此类产品的需求相称, 并采取措施确保合规。了解市场需求对于确定是否存在供过于求的情况是必不可少的。 如果向低税国外市场供应的烟草制品超过需求,就会增加这些产品被走私回高税国家的 风险,从而损害高税管辖区的政策目标和有效性。过去,如果供应量明显高于需求量且被 走私回其管辖范围的风险被认为很高,一些政府就会决定对烟草公司处以罚款[34,35]。 关键要点9 许可有助于识别和控制合法运营商。从许可中获得的数据可以作为审计的基础。 应定期控制许可证并定期更新以确保其有效性。 3.4.4　财政标记（例如税票） 财政标记是控制和监测烟草制品生产和进口的另一个重要工具。它的作用通常被认 为适用于增加遵守税法的情况。财政标记也可以帮助区分真假烟草制品。贴在卷烟或烟 草制品包装上的税票或其他财政标记有助于征收消费税、审计和执法行动。财政标记的 存在使主管机关和公众都能够监测市售烟草制品是否依法纳税。因此,它有助于主管机 39第三章 烟草税的管理 所有欧盟国家适用的消费税税率可以在欧盟委员会的网页上找到:https://ec.europa.eu/taxation_customs/ business/excise-duties-alcohol-tobacco-energy/excise-duties-tobacco_en。 关调查非法贸易和起诉违规行为。 财政标记包括印花税票、增强印花税票(banderols)和数字印花税票。财政标记的例 子有烟草印花、税印花、消费税印花、税贴和印花税票。方框3.4介绍了印花税票的不同 类型和特征的详细信息。出口的烟草制品通常需要标记为出口。方框3.5提供了关于国 际标准化组织(ISO)消费税印花标准的有用信息。 术语“财政标记”并不表示该标记的特征。每包烟草制品上都贴有财政标记。要求标 准包装尺寸有助于标记[2]的应用。在用玻璃纸包裹每包卷烟(或其他烟草制品)之前,通 过将标记贴在每包卷烟(或其他烟草制品)上,可以阻止欺诈者试图重复使用财政标记(特 别是税票)的做法[36]。 在大多数情况下,税票是由生产商或进口商购买的,并应用于销售的每个产品,作为 缴纳消费税的证明[33]。只有在已全额缴纳烟草制品的消费税或建立担保时,才应向烟草 制品的制造商或进口商发放财政标记。 方框3.4 印花税票的类型和特点 随着时间的推移,印花税票和标记变得更加防伪。过去,印花税票通常是纸质 的,很容易伪造。新的印花税票使用额外的安全功能使其更难以伪造。防伪认证解 决方案可以利用各种安全功能,包括: • 显性特征———肉眼可以验证的特征; • 隐蔽特征———只能通过使用专用和专业电子阅读器来验证的特征; • 半隐蔽特征———需要简单工具且操作人员无须大量培训即可验证; • 化学特征———可以通过实验室分析识别的特征。税收标记可以是物理的或 数字的; • 物理标记———信息包含在文件或附在包装上的设备中; • 数字标记———通过与数据库的链接以及使用用于创建数据的工具和密钥解 密来获取信息。“数字印花税票”一词有时会引起混淆,因为一些带有数字组件的纸 质印花也被称为数字印花税票。完全数字化的税务标记不包含附在包裹上的文件或 设备中的信息。 说数字印花税票比纸质印花税票更安全可能过于武断了。这两种类型都有优点 和缺点。例如,纸质和数字印花税票在安全功能上都可能弱或强。然而,纸质税票的 管理、生产、销售、运输和监控需要更多的安全措施。因为纸质印花税票与钞票具有 相同的价值,并且可能是盗窃、丢失和欺诈的对象。 例如,在比利时,审计法院在 2015 年严厉批评了对印花税票的制作和库存管理 缺乏控制。审计法院认为,印花税票的发行应按照公认的安全实践和程序进行,避开 与各种生产、分销和发行过程相关的安全风险。此外,有人指出,在包装上使用数字 印花税票的新印刷技术可以促进库存管理并减少欺诈。2016 年 3 月,比利时更换了 税票,印刷成为联邦政府财政部门的内部流程。新税票仍印有水印,但也有数字组 件。这一变化通过标准化尺寸和优化生产过程降低了成本。 49 世界卫生组织烟草税政策和管理技术手册 一些更先进的财政标记技术包括嵌入式线程、水印、特殊油墨与涂料(例如所谓 的隐形油墨、全息图和箔)和计算的或可变的内容。由于其增强的安全功能,这些税 票可能比传统税票更贵。在美国加利福尼亚州,传统税票每 1 000 张售价 0.42 美 元。第一代税票的成本高出 10 倍,为每 1 000 张 4.77 美元。第二代加密税票的价 格几乎翻了一番,达到每 1 000 张 8.20 美元。尽管如此,在实施加密印花税票后的 第一个10年中,加利福尼亚州增加了约 4.5 亿美元的额外税收。这种额外的收入远 远大于实施和执行的成本。 其他司法管辖区也修改了它们的税票以纳入新技术。例如,密歇根州在 2015 年用数字加压印花税票取代了热贴卷烟印花税票。密歇根州部署了一种印花税票, 该印花税票具有多种公开和隐蔽的安全功能以及独特的快速响应(QR) 代码和序列 号。QR 码(由一系列黑白方块组成的机器可读代码,通常用于存储 URL 或其他信 息)可用于跟踪和溯源之外的用途。消费者可以使用智能手机或平板电脑应用程序 读取二维码,可以获取有关戒烟计划的信息、举报违反该州青少年准入政策的行为、 连接到举报热线以举报不合规的包装并了解非法烟草销售的危害和购买。执法机构 可以使用基于智能手机的 eTRACS(电子税务报告和审计合规系统)验证印章。作 为该系统实施的一部分,密歇根州警察局在该州七个地区的每个地区都建立了执法 人员团队,而州财政部则建立了自己的执法团队。 资料来源:参考文献37~42。 方框3.5 ISO关于消费税印花税的标准 ISO于 2018年10月发布了其消费税印花税标准(ISO/TC 292/SC)。建立ISO 标准的目的是协助税务和财务部门加强对消费税法规的监管。 印花税票被定义为贴在某些类型的消费品上的可见印花、标签或标记,以表明产 品已缴纳适用的消费税。ISO 标准适用于物理性质的印花税票,而不适用于直接印 刷在包装上而没有物理组件的数字标记。本标准中的“认证”是指对印花税票的认 证,而不是对贴有印花税票的产品进行认证。换句话说,卷烟包装上的税票认证意味 着税票是真实的,但并不能保证包装是真实的。此外,还需要在印花时采取控制措 施,以验证印花税票与相应产品的符合性。 该标准对用于表明已缴纳所需税款和印花税票真实性的实物印花税票的内容、 安全性、发行和检查提供了指导。没有介绍使用印章来促进供应链内的跟踪和追溯。 具体而言,ISO 标准处理以下问题: • 定义印花税票的功能; • 识别和咨询利益相关者; • 规划采购流程和选择供应商; • 印花税票的设计与制作; • 提供税票保护的公开和隐蔽的安全特征; • 印花税的整理和申请流程; 59第三章 烟草税的管理 • 印花税供应链的安全性; • 用于印花税票的序列化和唯一标识符代码; • 检查税票; • 监测和评估印花税表现。 一张税票可以完成许多功能,但印花税的核心业务是确保和促进税收的征收。 印花税票必须使用安全特征的组合。税务机关应确保印花税票可以被认证,并且能 够检测出伪造、变造、篡改或其他欺诈性印花税票。 该标准提供了有关印花税票不同组成部分的详细信息,例如基材、油墨、黏合剂、 承压材料、认证或安全特征,以及用于检查所需税款支付的唯一标识符。 采购过程在标准中有详细讨论。税务机关应确保采购过程公开、透明并符合可 持续发展目标。税务机关应制定目标和要求,让投标组织在提出可能与主管机关指 定的解决方案不同的最佳解决方案时有更多余地。 该标准不是规定性的;相反,它提供了一个选项目录。例如,它没有推荐特定的 安全功能,但它确实描述了确保印花税安全所需的不同类型的功能。税务官员仍然 需要做出决定并选择最适合他们的选项,但对于那些想在其管辖范围内引入印花税 计划的人来说,该标准仍然推荐阅读。 ISO 标准不是免费提供的,但可以在 ISO 商店(www.ISO.org) 或从 ISO 国家 成员机构购买。 资料来源:参考文献43。 根据《议定书》第8条,每一缔约方应要求在卷烟的所有单位包装、包装和任何外包装 上粘贴或构成其唯一、安全和不可移除的识别标记,例如代码或税票,以实施10年内跟踪 和追溯系统,以及在议定书对该缔约方生效后10年内跟踪和追溯其他烟草制品。 该协议规定至少以下信息应构成唯一标记的一部分: • 生产日期和地点; • 生产设施; • 产品描述; • 尽可能标记目标市场。 在一些国家,二维码被用作烟草和酒精税收控制的财政标记。每个税票都有一个唯 一的识别码和一个二维码。二维码中存储的数据提供以下产品信息: • 制造商; • 生产地点; • 税票订购日期; • 税务状况和类别; • 品牌; • 目标市场; • 唯一标识符(序列号)。 69 世界卫生组织烟草税政策和管理技术手册 关键要点10 使用财政标记通常被认为是提高遵守税法的合适工具。 财政标记也有助于区分真假烟草制品。 3.4.5　跟踪和溯源 跟踪和溯源系统有助于主管机关确定烟草制品的来源和在适用时的转移点,并监测 和控制烟草制品的流通及其法律地位。建立跟踪和溯源系统的目标是使当局能够通过整 个烟草制品供应链获得所有交易的信息,直到关税被支付或其他义务被履行。 可追溯性并不只应用于烟草制品。它还可应用于改善供应链功能,如包裹服务,以及 出于产品安全原因,以管理潜在的产品召回和监管。 跟踪是一个过程,它始终监控产品的位置,同时还为所有移动创建时间和位置记录。 追踪是识别产品过去位置的能力,以便可以追踪产品的路线及其原产地[44]。换句话说, 可追溯性是“追踪对象的历史、应用或位置的能力”[45]。 跟踪和追溯系统必须能够唯一地识别单个产品。通过使用唯一代码或标识符标记产 品,可以明确记录该产品的移动。 其他必要的特征包括共享注册的流通信息和验证产品的能力。这使得可以通过供应 链捕获产品的状态,并对其历史进行追溯识别和验证。 根据《议定书》第 8.4.1 条,缔约方应要求提供以下信息: • 生产日期和地点; • 制造设施; • 使用的机器; • 生产班次或制造时间; • 第一个不属于制造商的客户的姓名、发票、订单号和付款记录; • 零售产品的描述和目标市场; • 全部仓储和运输; • 全部已知的后续购买者的身份; • 预定的运输路线、日期、目的地、出发地和收货人。 一个良好的跟踪和追溯系统使政府能够正确监控供应链,提高其政策执行能力确保 征收适当的关税和税收,提供认证能力识别标记是否真实,匹配产品和提高执行法律的能 力并提供足够的证据来证明存在违法者。 有效的跟踪和追溯系统需要以下要素[46]: • 每个产品包装的序列化唯一识别标记。这些标识符是数字、字母或两者的独特组 合。它们不能被预测或被多次使用。包装上标识符的表示可以是人工可读的(字母或数 字)或机器可读的(条形码)。应排除作为烟草业专利一部分的代码和加密的生成。 • 带有序列化唯一标识符和其他信息的数据载体,例如制造日期和地点、制造设施、 产品描述以及(如果有)预期零售市场。该信息应可供《议定书》任何缔约方的授权机构阅 读。数据载体应符合质量标准,适用于高速生产线。例如,二维条码符合这些标准,并且 79第三章 烟草税的管理 可以用廉价的设备读取。 • 不同包装单元之间的链接和父子关系(称为聚合),提供跟踪托盘的选项,而无须 扫描该托盘的所有包装和主箱。 • 记录整个供应链中的所有运输和接收事件。例如,这包括出发地点和到达地点, 以及涉及的运营商。建议使用 ISO 的国际标准来捕获和交换数据和事件。 • 对数据载体中编码的关键信息使用国际标准[5]。国际公认的产品唯一标识符的 一个例子是全球贸易项目编号。 以下有关信息存储和共享的详细信息来自协议的各个部分。供应链中的数据和事件 必须存储在由政府主管部门控制的独立数据库中。在全球层面,国家和/或区域数据库可 以相互链接,以促进主管机关的国际调查。《议定书》缔约方同意在世卫组织框架公约秘 书处设立一个全球信息共享联络点,所有缔约方均可访问,使他们能够进行查询并接收相 关信息。每一缔约方应确保根据第9款的要求,全球信息共享联络点可通过与其国家和/ 或区域中心点的标准电子安全接口访问根据《议定书》第8条第5款记录的信息。全球信 息共享联络点应编制一份缔约方主管当局名单,并将该名单提供给所有缔约方。 跟踪和溯源系统的成本是许多国家关注的问题,但如《议定书》第 8 条第 14 段所述, 司法管辖区可能要求烟草业承担在一个国家建立跟踪和溯源系统相关的所有成本[46]。 在巴西,卷烟制造商的成本为每包 0.018 5 美元[42]。在肯尼亚,制造商的成本为每包 0.024美元[42]。 除了在选择特定系统时考虑跟踪和追溯系统的特点外,还必须避免利益冲突,确保与 供应商的交易公平透明,对腐败或反竞争行为实施零容忍政策并确保合规 。 任何跟踪和追溯系统都应符合 FCTC 第 5.3 条(涉及行业干扰)和《议定书》第 8 条。 第 8.13 条规定,“每一缔约方应确保其主管机关在参与跟踪和溯源制度时,仅在实施本 条例所必需的范围内与烟草业和代表烟草业利益的人进行互动”。方框 3.8 提供了一个 不符合世界卫生组织 FCTC 第 5.3 条的跟踪和溯源系统的警示示例。 虽然《议定书》包含大量关于跟踪和溯源系统应满足的要求的信息,但出现了与此类 系统的实施相关的问题。为实现《议定书》的目标,议定书缔约方会议(MOP)作为条约的 理事机构,有权设立附属机构,如专家组和工作组。在 FCTC/MOP1[6]号决定中,MOP 设立了一个工作组,负责根据《议定书》第 8 条开发和实施跟踪和溯源系统,包括全球信 息共享联络点(第 8.1 条)和唯一识别标记烟包和包装(第 8.3 条),以进一步详细说明接 下来的步骤。该工作组将编写一份综合报告,汇编关于实施跟踪和追溯系统以及在国家 或地区层面为卷烟包装提供独特识别标记的良好做法和经验。工作组还被授权对如何建 立全球信息共享联络点进行概念分析。 实施带有财政标记的完整跟踪和追溯系统需要时间。在大多数已经实施跟踪和追溯 的国家中,从制定法律框架到最终实施需要数年时间。还需要克服几个障碍: • 法律框架的批准通常会被烟草行业推迟。 • 税务管理部门缺乏与跟踪和溯源相关的技术和知识。 • 招标和投标过程都比较复杂。 • 国内税务机关与海关之间的协调十分薄弱。 89 世界卫生组织烟草税政策和管理技术手册 尽管这个过程可能很长,但对跟踪和溯源系统的投资将以不因逃避而损失的烟草税 的数额来偿还。 在实施一个新的跟踪和追溯系统时,税务行政当局应要求已成功实施这些系统的政 府间组织和国家提供合作和技术援助,以加快和确保该进程的成功。智利、肯尼亚和欧盟 的跟踪和追溯系统实施实例详细见方框3.8、方框3.9和方框3.10。对于已经有财政标 记的国家,应考虑到这些标记与跟踪和追溯系统实施之间的潜在相互作用。关于这种交 互作用的进一步信息见方框3.6。 方框3.6 跟踪和溯源以及财务标记 税票包含一些跟踪和溯源功能变得越来越普遍,例如独特的识别标记和描述公 司、税务状况或产品本身的产品基本信息。目的是用唯一的识别标记标记每个包装, 以便从生产厂到零售商对其进行监控,包括中间的每个步骤,从而创建完整的时间和 位置历史记录。 虽然印花税票可以满足《议定书》第8条的要求,并具有跟踪和溯源功能,但总的 来说,印花税票系统的侧重点与跟踪追溯系统不同。跟踪和溯源不仅仅是烟草制品 包装上独特、安全和不可移动的识别标记。它意味着阅读或扫描代码,链接包装、纸 箱、主箱和托盘之间的代码,将信息上传到数据库,记录供应链中的任何运输和接收 事件并互联不同的数据库。 虽然新的印花税计划包含追踪和追查功能,但它们的主要目的是促进国内市场 的征税,而不是追踪免税的跨境贸易或产品出口。印花税系统的重点是提供真实性 和纳税证明。跟踪和追溯系统的重点是通过监控和调查产品的过去和未来位置来进 行唯一标识和控制供应链中的流通。印花税计划侧重于库存管理、验证(印花与产品 对应)和认证(印花是真品),而跟踪和溯源系统的重点是原产地、预定路线、第一位客 户和最终目的地。印花税的重点主要是针对已完税的国内市场的单个包装,而跟踪 和溯源系统的重点是所有包装(包装、纸箱、主箱、托盘),当然,也不仅限于免税出口 市场。 然而,有时可以产生协同作用。例如,出于财政目的需要印花税票或国家识别标 志的欧盟国家可以选择将其用作跟踪和溯源目的的安全功能,前提是满足要求。总 而言之,当转换后的系统提供包装、纸箱和主箱之间的聚合并记录供应链中的所有移 动时,印花税票可以转换为跟踪和溯源系统的一部分。对于出口市场,应增加唯一识 别标记。 资料来源:参考文献39 42。 方框3.7 不应该做什么:使用行业解决方案进行跟踪和追溯 Codentify 是一种序列化系统,用于生产卷烟包装标记,用于验证卷烟包装是否 合法。它由 PMI 授予专利,但随后免费授权给其他主要卷烟制造商。2016 年, Codentify被转移到Inexto,后者是法国 Impala 集团的子公司。 99第三章 烟草税的管理 Codentify/Inexto有什么问题? 主要问题是 Codentify/Inexto 与烟草业的联系同2018 年 9 月生效的《议定书》 不兼容。《议定书》规定分配给缔约方的义务不得由烟草业履行或委托给烟草业 。 此外,许多元素表明它是一种无效的身份验证手段。例如,由 Codentify 生成的 12 个字符的数字代码可以轻松复制或克隆并用作假冒或正品包装的原件,然后可以 通过系统的基本验证测试。这些代码也是由相对不安全的商用设备生成的,并且不 包括能够保护标识符号码真实性的高安全性特征。使用多层高级安全解决方案让分 销商、零售商、客户和主管机关能够识别不合规产品的系统更加安全。 另一个问题是 Codentify/Inexto 无法像其他可用系统那样有效地跟踪产品。它 需要更强的执法能力才能达到与烟草业无关的其他系统相同的检测率。主管机关将 不得不检查在 Codentify 系统下标记的包装比在其他系统下所需的数量要多得多, 以实现不遗漏欺诈包装的相同确定性。此外,并非所有利益相关者都能够验证在 Codentify 系统下标记的包装是否为真品,而其他可用系统确实提供了这种可能性。 资料来源:参考文献47。 方框3.8 成功的烟草追溯系统:SITRAF,智利 智利国税局(SII,Servicio de ImpuestosInternos)成功地协调和实施了烟草制品 的跟踪和溯源系统。 合规管理模式的应用 在合规管理模式的框架内,烟草制品的可追溯系统利用结构性措施来减少逃税。 据估计,智利卷烟市场每年逃税(增值税和消费税)额占该国总市场的 16.6%———约 3 亿美元。 智利实施的可追溯系统[SITRAF(TAB2)]使主管机关能够以某种方式及时了 解生产或进口到该国的卷烟数量。此外,它有助于主管机关区分假冒产品和不符合 纳税要求的原装产品。2018年,通过公开招标的方式,将追溯系统的实施授予了一 家公司。根据与SII签订的合同,该公司负责该系统的实施和5年运行,并且必须为 该项目配置一个20人的团队。 直接标记适用于智利生产的供国内消费的物品,而税票适用于进口产品。对于 这两种类型的产品,标记均基于数据矩阵代码,该代码使用安全墨水印刷,该安全墨 水与获奖公司提供的具有特定设备的任何其他类型的墨水不同。出口产品虽然不受 标识限制,但由追溯系统控制和核算。 对于国内产品,每条生产线上都安装了识别正在生产的包装类型的设备,在每个 包装上打印一个唯一的代码,然后读取(激活)以将所有信息保存在位于生产工厂的 服务器上。该信息被传送到系统的中央服务器,然后传送到 SII。它也可用于现场 检查。 对于进口产品,税票必须由每个进口商在智利购买,然后寄到国外的生产商,生 产商负责在用玻璃纸包裹每包香烟之前,使用生产线上的涂抹器将它们贴在每包香 001 世界卫生组织烟草税政策和管理技术手册 烟上。一旦卷烟进入智利,税收确定过程已在服务(临时免费运输指南)中完成,并已 缴纳海关税款(进口申报),进口商必须在可追溯系统的平台上输入进口商使用的税 票上的数据,经过验证后,这些税票可以在系统中被激活。也就是说,它们被认为有 效,可用于商业化。 除了公司提供的设备外,SII 还开发了一款智能手机应用程序,供市民验证。虽 然应用程序无法验证所用墨水的真伪,但它可以验证代码是否正确生成并显示代码 可追溯系统中包含的信息(品牌、品种、卷烟数量、产品/进口商)以便纳税人可以验证 其一致性。 合规性管理模式的进展情况 可追溯性系统的实施需要在服务部门内生成指令、程序和计算机开发,以及与国 家海关和卫生部等机构进行协调。在协调全国不同烟草公司和供应商公司的生产线 实施系统方面,这是一个史无前例的项目。该项目的一些主要里程碑包括: 2014年9月:《税收改革》第20780号法律规定,在确定义务纳税人的决议公布 后的6个月内实施该制度。 2015年5月:第47号决议确定了义务纳税人。 2015年6月:第47号通告描述了将税票或独特标记作为可追溯机制的义务。 2016年2月:关于税收改革的第20899号法规简化了系统的定义,允许系统外 包或由SII提供,同时使可追溯性类型更灵活。 2016年8月:发行了可追溯性系统法规D.S.1027(2016年12月28日实施)。 2017年3月:财政部第49号豁免决议授权SII外包全部或部分可追溯性系统。 2017年6月:公开市场公布招标基准。 2018年2月:使选定的公司中标。 2018年6月:由共和国总审计长做出合同决定。 2018年8月:第61号决议确定了纳税人有义务应用可追踪能力的系统。 2018年8月:在海关、卫生部、供应商和SII的参与下,首次举办了有关项目详细 定义的研讨会。 2018年9月至10月:参观工厂,与生产商协调,并确定调整生产线,以进行系统 实施。 2019 年 1 月:第 6、第 7 和第 8 号决议通过,响应生产商延期的请求;开始在全 国所有生产线实施系统。 2019 年 2 月:第 16 号决议确立了将剩余库存商业化的期限,无须标记。 2019 年:第 24 号决议要求延长进口商(包括海关人员)的培训。 2019年3月:可追溯性系统已启动。 2019年6月:所有进口卷烟开始印花。 目前,该系统在该国的所有生产线上安装和运行,几乎国产和进口的卷烟100% 都有可追溯性标记。2019年,可追溯系统能够控制约11.75亿包卷烟:全国消费 7.44亿包,出口4.09亿包,年进口0.22亿包。 101第三章 烟草税的管理 来源:参考文献48。 方框3.9 肯尼亚实施跟踪和追溯系统的案例研究 肯尼亚使用目前的跟踪和溯源系统之前,在税收结构和消费税管理方面进行了 一系列改革。改革包括对出口进行电子货物监测,从而允许自动监测和报告。该系 统非常高效,因为它需要的容量更少,而且比早期的系统更不容易被操作。肯尼亚的 经验表明,一个中低收入国家可以成功地实施一个能够减少非法贸易的复杂制度。 它还表明了其他措施的重要性,如加强执法,加强不同机构之间的合作和沟通,以及 增加对违法行为的处罚力度。根据肯尼亚税务局(KRA)的衡量,非法贸易在改革初 期估计约占市场总消费的15%。2015年引入新系统后,降至5%。 重大改革的时间表: 2003年:在此期间,使用的纸质税票有唯一的标识符,并用颜色编码表示产品的 类型。定期进行合规性检查。2007年,一张税票的成本为2.124肯尼亚先令,相当 于每包0.023美元。然而,这些税票很容易被伪造,而且不能与特定的品牌联系起来。 2008年:KRA提出了一个跟踪和追溯系统,并提高税率。新的系统被逐步 引入。 2010年:在纸质税票上增加了更强的安全功能,包括紫外线标记。当税票被出 售和使用时,税票将清晰可见,这样打开一个包装就会破坏税票。这些税票在供应链 上的4个不同点进行了验证。成本仅略高于之前的税票,为每包0.024美元。 为国内制造商引入了许可证制,但每年更新一次。进口商被要求在KRA注册。 许可证需要提交关于公司董事、库存和设备、会计系统、投入生产比和生产品牌的详 细资料。对违规行为的处罚增加,最高可判处三年监禁。 电子货物追踪系统。在集装箱或卡车上安装了电子密封件,并使用GPS技术进 行跟踪。出口应支付保证金以支付消费税和增值税。只有在货物到达最终目的地并 缴纳税款时,保证金才被返还。 涉及两国的商业交易的验证在边境进行。电子系统提供有关货物的离开和到达 以及解除封条的信息。货物离开国内生产设施前通知进口国的相关机构。该系统减 少了所需的检查点和工作人员的数量,并生成可通过增值税退税请求进行验证的到 达报告。 由于这些变化,3家工厂和10家进口商中有7家因不合规而关闭。向科特迪 瓦、厄立特里亚、马里和苏丹的出口停止了,因为公司无法提供证据证明货物到达最 终目的地、并缴纳了税款。2011年,收回了超过 1100 万美元的消费税损失。KRA 估计,非法贸易下降至8%。 2013年:4月签署了一份合同,引入烟草和酒精跟踪和追溯系统,即应税商品管 理系统。该系统为现有系统增加了生产盘点、跟踪和溯源、库存控制、处理等数据收 集。基础设施要求包括在生产设施、仓库、KRA和端口的高速宽带互联网,以及在这 些地方的可靠的电力或备用发电机。计划分3个阶段进行实施: 201 世界卫生组织烟草税政策和管理技术手册 • 第一阶段,引入了具有唯一标识符的新电子数字税票。它包括一个数据矩阵 代码和明显标记(全息图、荧光纤维、用于 KRA 认证的安全链接和用于验证和激活 的可见二维码)、半隐蔽标记(紫外线特征、可通过专用设备检测的荧光打印、用于零 售商的迷你文本打印)和用于起诉的法医标签。税票还包括人工可读的代码,用于使 用 KRA 网络通过短信服务进行验证。 • 第二阶段,控制和监控系统于2014年2月实现了自动化。制造商必须在生 产线上安装光敏读取器,数据会自动实时发送到KRA。每个税票都被激活,并与生 产线上的一个品牌和包装大小相关联。KRA数据库每15分钟自动更新一次。 • 第三阶段,市场监测开始了,83名官员被授权没收非法卷烟并进行逮捕。这 些人员配备了手持设备,将数据传输到KRA进行认证。经销商和零售商对销售没 有缴纳消费税的产品负有责任,并因不遵守规定而被处以罚款和最多三年的监禁。 2016年,一款智能手机应用程序问世,公众可以据此对卷烟包进行认证。进口商现 在必须购买数字税票,并送到其他国家的出口设施进行粘贴。从工厂运出或进口时 应纳税。电子货物监控系统仍然有效。 2016年:《消费税和税收程序法案》澄清了新的义务和处罚。 2017年:推出新的海关综合管理体系。 KRA估计,非法贸易水平现在为5%左右。目前,更全面的数字系统比以前的 纸质印花税制度更便宜。制造商为生产监控系统支付费用,但它被计入企业纳税申 报单上的商业费用。2018年,两家制造商和10家持牌进口商在肯尼亚运营。 2018年:包装、纸箱和主箱之间的标记聚合尚未实施,但预计即将实现。 来源:参考文献36。 方框3.10 欧盟新跟踪和溯源系统案例研究,2019年5月 据统计,欧盟的卷烟走私和其他形式的非法贸易每年会造成100亿欧元的收入 损失。2018年,欧盟海关查获了420万包(每包20支)非法卷烟。非法烟草产量也 在增加:爱尔兰的一家非法工厂(于2018年拆除)每小时能够生产25万支卷烟。 欧盟的烟草控制政策在烟草制品指令中都有描述,并受到《议定书》的影响。该 指令第15条要求在2019年5月前实现对卷烟和RYO烟草制品以及在2024年5月 前实现对其他烟草制品的可追溯性。欧盟的跟踪和追溯系统足够灵活,可以在区域 和单一国家层面实施。只要满足基本要求,各国就可以在提供者中进行选择。该策 略提供了高水平的保护,以防止任何操纵数据的企图。报告义务涵盖了参与烟草制 品生产和分销的所有经济经营者。 欧盟系统要求所有烟草制品包装标明唯一标识符,这些产品的移动信息由第三 方数据存储提供商存储。唯一标识符和数据存储的供应商应在财务和法律上独立于 烟草行业。出于执法目的,欧盟成员国将完全获取这些数据。 唯一标识符的生成以及经济运营商、设施和机器预注册所需的所有其他代码将 由指定的标识符发布者在成员国层面完成。制造商和进口商在要求发行商提供唯一 301第三章 烟草税的管理 标识符时,必须提供与产品和生产线相关的信息。然后,发行人将生成并交付成批的 唯一标识符。在生产线上,烟草制品制造商将完成每个唯一标识符,并带有指示日期 和时间的标记。唯一标识符将是机器可读的光学的一维或二维条码。必须事先安装 了能够创建不可更改的独立验证过程记录的防篡改设备。公共管理机构可以访问此 附加记录以进行潜在的调查和检查。 可以在整个供应链中跟踪和溯源单位包裹以及聚合包裹(例如纸箱、主箱或托 盘)。只要单位包裹保持可追溯性,也允许在聚合包装级别进行跟踪。 在运输过程中,必须记录和报告每次发货和到达信息,直到最终发货到第一个零 售店。所有记录的信息必须提交给独立的第三方数据存储设备,一般在3小时内,或 在发货和转运前 24 小时内。 根据《议定书》第8条,包括运营成本在内的成本将转移到烟草行业。欧盟的烟 草可追溯性和安全功能系统于2019年5月20日开始运行。 来源:参考文献36,49 50。 关键要点11 跟踪和追溯系统可协助主管机关确定烟草制品的来源和转移点(如适用),以及监 测和控制烟草制品的流动及合法性。 3.4.6　反买断 “买断”是一个术语,它描述了由于预期税收增加而增加的产品产量或库存[2]。涉及 这种做法的其他术语包括“储存”和“前端装载”。当制造商、进口商增加其已纳税库存或 通过增加产量或进口以争取到先前较低的税率而使市场供过于求时,就会发生买断。它 减少和延长了税收措施的有效期。新税率的生效日期将推迟,收入将减少,对价格和消费 者行为可能产生的影响也将推迟。为了说明反买断措施的作用,附件 3.2 中提供了一个 示例。 反买断措施必须有法律依据,否则,政府无法阻止该行业的买断行为。处理买断的法 律措施包括[51,52]: 1.限制原税率下可放行的烟草制品数量,对超过该限制的产品征收新税。 2.对所有仍库存且尚未供应给最终消费者的商品按新税率征税。 3.限制按照上调前的税率签发的印花税票数量或限制旧税率印花税票产品的销售 时间。 4.要求生产商和进口商每年或在增税后购买新的印花税票。 前3项措施中,主管机关按以前(较低)税率确定征税限额。允许的数量可能基于烟 草制品的保质期———卷烟约为6个月———或正常库存水平,例如前3年的平均水平。 第1项措施是限制可放行的烟草制品数量,需要主管机关提供资源来执行。主管机 关可能会决定在每个生产厂家中派驻检查员,但即使没有派驻检查员,也需要有程序来确 401 世界卫生组织烟草税政策和管理技术手册 定何时超出了允许的数量以及采取何种后续行动。 第2项措施,对尚未供应给最终消费者的库存商品征收新税,可能难以实施。主管机 关需要监控制造过程,至少在很短的时间内对供应链中的所有参与者进行盘点,包括制造 商、进口商、批发商和零售商。如果监控仅涵盖制造商和进口商的库存,则可以通过确保 将库存出售给供应链中的其他人或通过设立单独的分销公司来购买库存来轻松规避此措 施。在零售层面控制库存是一项繁重的工作,鉴于卷烟零售商数量众多,在管理上可能不 可行。如果对零售商没有许可要求,这会变得更加繁重,因为他们首先必须被识别。 如果一个国家使用印花税,则第3项措施与第1项措施的作用相同。第4项措施也 需要印花税票,对主管机关来说更简单,但对烟草公司来说负担更重,因为必须每年购买 印花税。方框3.11提供了欧盟国家反买断措施的例子。 方框3.11 欧盟国家反买断措施案例 一些欧盟成员国已采取措施限制买断。似乎需要采取谨慎的方法来设计此类措 施,以确保它们符合欧盟立法和欧盟法律的一般原则———特别是比例原则。欧盟成 员国实施反买断措施的主动权未发生争议。尽管如此,一些欧盟成员国不得不在欧 盟法院面前为它们的措施辩护,该机构确保所有国家立法符合欧盟法律并一致适用 该法律[53]。法院承认,反买断措施适用于打击逃税和避税行为。此外,法院强调财 政立法是阻止烟草制品消费并因此保护公众健康的重要而有效的工具[54,55]。但是, 所采取的措施应该与目标相称。相称性原则是指只采取实现目标所需的行动,不应 超出必要的范围。该原则规范了欧盟内部采取的措施,并包含在欧盟条约中。 法院要求葡萄牙修改其立法以确保遵守相称原则。比利时、爱沙尼亚和匈牙利 也被敦促改变其反买断措施,使其符合欧盟立法[56,57]。 大多数欧盟成员国都制定了反买断措施,但这些措施没有统一。下面是一些 例子: • 葡萄牙将一年中最后4个月可释放的卷烟数量限制为前 12 个月的平均值加 10%。此外,制造商和进口商必须在3个月内销售带有上一公历年税收标记的卷烟 包装。对于其他烟草制品,适用更长的限制。 • 丹麦将按旧税率增税前发行的印花税票数量限制,通常比在年底前两个月内 购买的印花税票多 20%。 • 在波兰,印花税票仅在当前公历年有效,带有旧印花税票的卷烟只能销售到 次年二月。 • 在罗马尼亚,公司必须向海关申请放行许可。 来源:参考文献58。 501第三章 烟草税的管理 考虑到有关反买断事件的法院案件的频率,这些措施可能已经被取代或修正。 关键要点12 买断会降低和延迟税收措施的有效性。实施反买断措施可以限制加税的延迟及其 对税收和消费者行为的预期影响。 3.4.7　额外的国家审计和控制 除了上述措施外,还可以实施若干定期审计和控制,以提高对税法的遵守程度。最常 见的审计和控制方法如下: • 成本审计:成本审计方法通过模拟卷烟的中间成本和最终成本,提供了预期的增 值税和烟草税的征收。它从原材料库存开始估计附加值和最终成本,然后将结果与从烟 草供应链中实现的实际收入相匹配。附件3.1提供了关于组成一些选定(烟草)产品成分 的更多信息。 • 转让定价审计:为确保企业缴纳公平份额的税款,应对关联公司之间的交易价格 进行评估,当价格不符合市场条件时,应予以纠正。在国际层面经营的公司(跨国公司), 包括许多烟草公司,可以操纵商品或原材料的进出口价格,目的是降低税率,并且可以将 这些利润转移到税率较低的国家。 • 价格和市场监测:零售价格调查可以提供有关特定地点市场价格差异的信息,突 出潜在避税或非法贸易的领域。对这些地点的实际控制需要快速反应小组,正如在菲律 宾实施的那样。为了监督纳税人的税收合规性,菲律宾主管部门需要了解烟草市场;它必 须有关于品牌、细分市场和产品价格的信息。这些信息使主管机关能够估计税收和价格 变化对消费者行为和收入的影响。市场数据可以作为风险管理和反欺诈分析的一部分进 行分析,以确定向谁调查违规行为以及何时调查。可以对销售数据进行三角测量以验证 其他数据源,例如关于流行率的家庭调查。市场数据和趋势也是确定是否存在供过于求 的有用指标。另见第 3.4.3 节。 • 消费者控制:通过宣传活动让公众参与也被证明是有效的。消费者有权确信市场 上出售的产品是真实的并且来自合法渠道。因此,了解并能够验证他们购买的是正品符 合消费者的利益。烟草制品财政标记的特征应有助于消费者区分正品和非法产品。一些 国家,例如肯尼亚(见方框 3.9 中的案例研究),使用智能手机应用程序,让任何人都可以 检查隐蔽和公开的特征,并报告任何标记不正确的卷烟。其他国家,例如荷兰,已经开发 了一款智能手机应用程序,允许任何人举报消费税欺诈的可疑案例。 • 交叉检查控制:主管机关应考虑使用多种来源获取市场数据,并确定这些数据是 否与纳税申报一致。增值税申报可用于验证原材料和最终产品的供应商和采购商报告的 金额是否相同。银行信息可用于验证供应链上交易双方的交易行为。任何差异都可以提 醒主管机关对可能的非法烟草贸易或逃税行为进行进一步调查。 关键要点13 可以执行几种不同类型的定期审计和控制来提高合规性,包括成本审计、转让定价 审计、价格和市场监控、消费者控制和交叉检查控制。 601 世界卫生组织烟草税政策和管理技术手册 3.4.8　进出口管制 《议定书》缔约方应仅允许由经正式许可的自然人或法人实体进出口烟草制品和制造 设备(第6.1条)。欺诈者使用的一个众所周知的策略是申报出口产品,以便出口国不征 收任何关税。这些产品通过其他国家运输,利用在途制度,允许临时暂停关税,直到货物 到达最终目的地。在到达最终目的地(应缴纳消费税的地方)之前,货物再被转移到非法 供应链时消失或丢失。货物可能从未离开该国,或者可能在未申报或缴纳关税的情况下 被走私回出口国。这种收入损失的风险可以通过要求提供担保或保证金来减轻,只有在 另一个国家缴纳关税的情况下,这些担保或保证金才会被解除。 根据国家法律,除非烟草制品上贴有必要的财政标记(如税票或出口标签),否则烟草 制品不得进入司法管辖区。供出口的烟草制品应当标明该产品将出口。 如果对烟草生产和制造设备的制造商和出口商有管辖权的各主管机关在批准装运和 已经装运时向目的地的主管机关事先提供信息,则非法烟草贸易可以大幅减少。这些信 息可以包括收货人的姓名、发运货物的描述和数量。此外,目的地的主管机关应通知对托 运人有管辖权的主管机关已收到货物以及与该货物有关的相关信息。 进出口还需要一个良好的IT系统。建议对到达前的清单和进口申报进行电子化处 理。大多数国家已经实施了在线海关系统来处理进出口申报,包括所有需要的数据,如原 产国或出口国、商品描述、价值、重量、货物保险、承运人、进口商或出口商和中间人身份、 支付的详细税费和最终目的地。 世界贸易组织(WTO)的《贸易便利化协定》为更好地控制贸易提供了若干种工具,包 括海关行政机关之间的合作,信息交换、使用非侵入性设备和到达前的进口申报[23]。 海关的非侵入式检测设备检测违禁品的能力很强。最常见的工具是用于小包裹、集 装箱、卡车和火车的 X 射线扫描仪。大多数现代港口也实施了 X 射线扫描仪的使用,这 种技术正在将控制速度提高到每个集装箱两分钟。尽管扫描仪的成本正在下降,但对于 资源有限的国家来说,它仍然难以获得。幸运的是,扫描仪通常可以出租,因此这些国家 的税务部门可以租用它们。 较不复杂和更便宜的检测设备包括内镜、镜子、夜视设备、相机和自动车牌阅读 器[33]。一个更便宜的替代选择是使用狗,它可以通过训练来检测卷烟和其他有机产品。 许多国家同时使用扫描仪和狗来检测违禁的烟草制品。 还可以采用特殊的物理控制措施,以减少违禁品。这些操作包括将加工操作与已征 税和免税产品的密封存储分开。在操作的一部分或整个过程中,可由消费税主管机关的 官员进行物理和直接控制(例如,由个别卡车或车队护送货物从边境到边境的运输,或将 无线电或卫星跟踪系统,如GPS设备应用于货物、运输工具、车辆或集装箱)。 对边界的控制至关重要,应注重运用综合技术并加强与边境站各机构的合作。前线 官员应得到管理层适当的情报、指导和监督,以及对执法的技术协助。 在一个国家,移动消费税控制单元有助于验证在国内运输时验证应税商品。这些单 位应被派往重要的交通走廊、通信中心和拥堵地区,如桥梁、渡船和通道。这些行动需要 701第三章 烟草税的管理 警察、边防警卫和其他公共服务部门之间的密切协调。 出口还需要特别注意,特别是如果对烟草制品的出口给予增值税和退税时。确认申 报金额的真实出口对于避免非法重返领土和不当退税至关重要。对于任何退税,都必须 进行包括税收抵免信息在内的审计。审计可能包括涉及出口的整个链条的发票,包括烟 农、第一加工商、制造商、批发商、储存和运输商。 关键要点14 为确保进出口控制,建议只允许获得正式许可的个人和实体进出口烟草制品和制 造设备。 3.4.9　自由贸易区和转运点 “自由贸易区”一词内涵非常广泛,可以指许多不同类型的区域。金融行动特别工作 组在其 2010 年的报告[59]中列出了以下类型:自由贸易、出口加工、企业、自由港、外贸、经 济特区和保税仓库。其中许多领域可包括烟草制造和贸易。根据定义,自由贸易区内的 监管等控制措施不如其他地区严格。这可以使它们吸引参与非法卷烟制造或贸易的 人[17]。事实上,与自由贸易区有关的非法活动(不限于烟草)由承认这种联系的组织定期 记录。这些活动包括洗钱、逃税和假冒商品或其他非法商品贸易[60]。欧洲议会的一份报 告[61]特别提到了自由港,提到使用它们的动机包括“高度保密和推迟进口关税和间接税 的缴纳”。该报告甚至提出“紧急淘汰自由港”。在欧洲议会的报告中,自由港是作为高价 值商品(半)永久存储区的自由贸易区。 该《议定书》包括对自由贸易区烟草制品的所有制造和交易进行有效控制的有时限规 定(第 12 条)。自由贸易区被定义为缔约方领土的一部分,在那里货物被视为在关税领 土之外征收进口关税和税收(第 1.5 条)。这与《简化和协调海关程序国际公约》(经修订 的京都公约)[62]中使用的定义相同。《议定书》缔约方必须在《议定书》生效后三年内对自 由贸易区实施有效控制。对于尚未成为《议定书》缔约方的国家而言,在自由贸易区对涉 及烟草制品的制造和交易进行严格控制是有效和高效税收管理的重要组成部分。 在《议定书》内处理自由贸易区的措施之一包括“通过使用本《议定书》规定的所有相 关措施,对在自由贸易区内烟草和烟草制品的所有制造和交易实施有效控制”。正如国际 刑警组织的一份报告[17]所指出的,自由贸易区的一个重大弱点是不同的经济活动(例如 制造、组装、重新包装和仓储)发生在主管机关的控制之外。因此,海关部门必须在自由贸 易区行使其权力,以有效识别和打击烟草制品的非法贸易。应采取《议定书》中所列的相 关措施,其中包括为自由贸易区内的所有运营商颁发许可、尽职调查和记录保存,以及实 施跟踪和追溯制度。取消消费税豁免是加强控制和取消使用自由贸易区作为逃税手段的 激励措施的另一种方式。例如,印度尼西亚对其自由贸易区的卷烟制造征收消费税 。 《议定书》缔约方还应禁止在从自由贸易区运出时将烟草制品与非烟草制品混合在单 801 世界卫生组织烟草税政策和管理技术手册 印尼财政部,基于私人沟通获取的信息,2020年1月。 个容器或任何其他类似运输装置中。最后,各缔约方应“按照《议定书》的规定,对烟草制 品和生产设备的国际过境或转运采取控制和核查措施”。 《议定书》第 13 条涵盖烟草制品的所有免税销售,要求《世卫组织框架公约》缔约方 考虑禁止或限制向国际旅行者销售进口退税或免税烟草制品,如世界卫生组织 FCTC 第 6 条。这些销售削弱了旨在减少对烟草制品需求的税收和价格措施的影响,并通过在税 收结构中制造漏洞对政府收入产生不利影响[2]。 关键要点15 海关部门应在自由贸易区行使权力,以防止不同的经济活动发生在它的控制之外。 相关措施包括为自由贸易区内的所有运营商发放许可证、尽职调查和记录保存,以及实 施跟踪和追溯制度。 3.4.10　检测非法烟草贸易的后续程序 前面章节中描述的程序旨在提高合规性并防止非法贸易。当发现走私或非法贸易时 (通过多种途径,比如审计、跟踪和追溯系统、申报核查或边境管制途径)必须立即采取行 动,例如扣押或销毁走私的非法烟草以及征收应缴税款。为了阻止进一步的非法行为,还 必须对参与违法行为的每个人和所有事物进行全面审计。可以扣押涉及非法活动的资产 和船只,并冻结金融账户。包括英国、加拿大和智利在内的一些国家也采取了一种被称为 “追随金钱”的策略,以获取更多有关资助非法贸易的人的信息。该战略旨在通过针对为 非法产品的运输、生产和储存提供资金的人来产生更深远的影响。英国、加拿大和智利都 成立了专门小组,负责识别和锁定嫌疑人。 此外,《议定书》第 18 条规定没收和销毁烟草、烟草制品和制造设备。主管机关在行 使权力扣押和没收用于制造或分销烟草制品的产品和/或设备时面临的困难之一是在销 毁前保留或储存货物和机器的成本。因此,法律还应规定处理和销毁被扣押和没收的货 物或机器的机制和时间表,同时规定这些财产仍可作为司法程序中可采信证据的机制。 方框 3.12 和方框 3.13 提供了英国和印度尼西亚成功打击非法贸易的例子。 关键要点16 一旦发现烟草制品的走私或非法贸易,就必须采取征税、扣押和销毁走私的非法烟 草等行动。 方框3.12 英国贸易经验 2000年,非法卷烟占英国卷烟市场的22%。为解决这个问题,英国关税与消费 税局 开展了一项重大的反走私工作。该战略在2011年通过额外的资源和措施进行 了更新,并在2015年进行了审查。结果是,到2014年,非法卷烟市场稳步下降至10%。 901第三章 烟草税的管理 在进行这项战略更新时,相关机构是英国关税与消费税局以及英国的边境机构。 所采取的措施是全面的,包括雇用1 000名新的海关官员和调查员。此外,还引 入了烟草供应链立法,旨在阻止烟草制造商为走私提供便利。 更严厉的制裁包括对制造商处以最高500万英镑的罚款、最高7年刑期的刑事 起诉、没收资产作为犯罪所得的一部分、对没收的货物缴纳关税以及最高100%的处 罚关税,禁止销售烟草制品长达6个月,在2015年3月13日之后出售没有英国完税 财政标志的烟草,将被处以罚款,并吊销运输商的执照。使用移民制裁,拒绝烟草走 私者入籍英国并提起民事诉讼(包括破产)。 这些措施的成本在该计划的前3年为 2.09 亿英镑,到2009年每年约为 1 亿英 镑。该数字仅涵盖 HMRC,不包括英国边境管理局的任何成本。2014 年,烟草税收 为 95 亿英镑。 在 2015 年对该战略进行审查后,2017 年引入了审批制度,加强了对生烟的控 制。任何制造、购买、获取、拥有烟草制品制造机器的人都必须在2018 年 8 月 1 日 前已获得海关许可[63]。 英国于 2018 年 6 月 27 日批准了《议定书》,是第 40 个批准国。这是该《议定 书》生效的触发点。 英国纳税消费 非法市场 20 00 -0 1 20 01 -0 2 20 02 -0 3 20 03 -0 4 20 04 -0 5 20 05 -0 6 20 06 -0 7 20 09 -1 0 20 07 -0 8 20 08 -0 9 20 10 -1 1 20 11 -1 2 20 12 -1 3 20 13 -1 4 20 14 -1 5 20 15 -1 6 20 16 -1 7 20 17 -1 8 60 50 40 30 20 10 0 十 亿 支 卷 烟 图3.4 对英国非法卷烟市场和纳税卷烟消费的估计 来源:参考文献16和64。 011 世界卫生组织烟草税政策和管理技术手册 方框3.13 成功案例:印度尼西亚将非法卷烟的市场份额从12%减少到3% 在一些国家,大部分非法卷烟是进口的,但在印度尼西亚,大多数非法卷烟是由 未注册的制造商在国内生产的,规模相对较小。在几个地区,特别是在爪哇岛,非法 卷烟已经传承了几代人。这种做法得到了烟草原料和丁香的供应以及廉价劳动力成 本的支持,尤其是对女性工人而言。印度尼西亚还生产大多数其他国家没有的卷烟 产品。这些被称为手工丁香卷烟(sigaret kretek tangan)的产品含有丁香,生产过程 包括混合、卷制和包装。部分过程是手工完成的,99% 的工人是女性。 印度尼西亚打击非法卷烟贸易的战略分为两大部分:预防行动和应对行动。印 度尼西亚在有效处理非法贸易方面的成功可归因于以下关键因素。 监测和监视 预防措施包括通过优化消费税服务信息系统(ExSis) 进行风险管理的行政措 施,例如发放许可证和消费税印花税票购买机制。有了这个 IT 系统,海关总署 (DGCE) 可以监督工厂的日常交易和日常生产。当获得可疑活动的信息时,DGCE 可以暂停购买消费税印花税票。打击非法卷烟贸易的尝试还包括邀请利益相关者参 与,以获取有关高风险地区和地区政府的信息。 战略沟通和社区参与 DGCE不断传播信息并开展公众教育,以打击非法卷烟。这些尝试每年都在进 行,并会使用特别的竞选口号。2019年的口号是“GempurRokokIlegal”(“打击非法 卷烟”)。 DGCE单位和办事处的关键性能指标 在采取上述预防措施的同时,DGCE 还不断开展应对行动:对卷烟厂的执法、调 查和审计活动。在规划和衡量执法活动时,要考虑到有限的人力资源和受监测地区 的规模。为了证明行政和执法措施在遏制非法卷烟贸易方面的有效性,这两种活动 都被转化为所有 DGCE 工作单位和办事处(包括区域办事处和人员)的关键绩效 指标。 技术和智能系统的使用 负责监察卷烟生产和销售的不同海关办事处之间的协作,由先进的IT应用程序维 持,以便有效地分发信息和开展调查活动。这些应用包括海关情报和战术中心实现数据 分析,以及确保指挥和控制中心(Pusat Komando dan Pengendalian/Puskodal)海上巡逻 工作的有效和高效。 独立评价 为了评估印度尼西亚减少非法卷烟流通的努力和活动(例如无印花税票、假印花税票 或二手印花税票的卷烟),政府于2016年委托日惹木田大学采用分层随机抽样方法进行 了一项调查。为了保持客观性和独立性,由这所知名大学的一个独立机构来进行该调查。 调查结果显示,印度尼西亚的非法卷烟流通水平占总消费量的12.1%。 2018年,DGCE委托Gadjah Mada大学进行另一项调查。结果显示,非法卷烟的流 111第三章 烟草税的管理 通量已减少到7.0%。2019年,DGCE采用与该大学相同的方法进行了一项调查,结果显 示下降至3.0%。 图3.5显示了调查的结果。 12.1% 10.9% 70% 30% 2016 2017 2018 2019 图3.5 2016—2019年印度尼西亚非法卷烟贸易占卷烟总消费的份额 资料来源:印度尼西亚财政部海关和消费税部门,基于私人沟通获取的信息,2020年。 所采取行动的结果可作为DGCE制定服务和执行消费税政策的反馈。DGCE将为 控制消费、维护劳工保护、优化收入以及(最重要的是)在全国范围内不断减少非法卷烟消 费的最佳政策模式提供建议。 3.4.11　处罚 处罚和制裁必须足以阻止非法活动。否则,支付的罚款只被当作经商成本,而非法活 动仍然会持续。《议定书》规定了缔约方的承诺,并为非缔约方提供了有关最佳做法的信 息。《议定书》第 IV 部分第 14.1 条要求每一缔约方了解非法活动包括制造、批发、中介、 销售、运输、分销、储存、运输和进出口烟草制品或制造设备,以及未支付适用的关税或税 收,或未使用财政印章或其他要求的标记或标签。 《议定书》第14.2条和第15条规定缔约方确定第14.1条规定的哪种违法行为构成 刑事犯罪。缔约方必须采取立法和其他措施来实施这些决定,并确定从事非法烟草贸易 的责任是否属于刑事、民事或行政罪行。第17条还规定,双方应考虑采取必要措施,授权 主管机关征收与实施非法贸易造成的税收和关税损失成比例的罚款。方框3.14提供了 一个关于哥伦比亚如何利用处罚来打击非法贸易的案例研究。 方框3.14 哥伦比亚使用处罚措施打击非法贸易 2017年,哥伦比亚的卷烟从量税翻了一番,从2016年的每包COL$700增加到 2017年的COL$1 400。2018年税率比2016年增加了两倍,达到每包 COL$2 100。在2018年后增加了一项规定,即每年增加通货膨胀4%比率的税收。 2015年,在增税之前,第1 762号法出台了多项措施,以更有效地打击非法贸易。 贩卖走私卷烟的刑期从3~5年增加到4~12年。此外,为非法贸易提供便利的政府 211 世界卫生组织烟草税政策和管理技术手册 官员(或任何参与非法烟草运输或零售的人)都将面临类似的监禁。该法律允许没收 用于走私的车辆,并增加了对在经济特区等地区进行非法贸易的处罚。根据该法律, 非法贸易被认为是一种洗钱的源泉,这意味着金融情报部门在调查其他非法金融活 动时也可以使用同样的方法。这种做法并不常见。 该法律特别规定了与逃避酒精和烟草消费税有关的新制裁,包括扣押商品、罚 款、关闭零售店以及暂停或吊销许可证、授权或登记。 根据新法律,逮捕和扣押的人数有所增加。事实上,自其颁布以来,法律主管机 关报告称,在2016年至2018年,共捣毁了5个犯罪组织,53人被逮捕,72项资产被 没收。此外,2 236人被逮捕,503辆运输走私货物的车辆被没收。根据法律,运输走 私货物现在也被视为犯罪。 更重要的是,由于大幅增税,2017年和2018年香烟消费下降,政府财政收入大 幅增长(图3.6)。据估计,2016年哥伦比亚五个城市的非法卷烟贸易占总消费的 3.5%,远低于行业数据显示的估计值。2017年,在实施了9个月的增税后,一项类 似的研究发现,非法卷烟的消费量仍然很低,占总消费量的6.4%。 2016 2017 2018 800 700 600 500 400 300 200 100 0 673.7 521 446.4 195.1 301.2 386.2 33.7%减少 97.9%增加 卷烟消费税收入/ 百万美元（以当时汇率计算） 卷烟销售量/ 百万包 图3.6 2016—2018年哥伦比亚增税前后的卷烟销售和烟草税收入 资料来源:参考文献65~67和哥伦比亚财政部基于私人沟通获取的信息,2020。 对于持有非法烟草的消费者,最低处罚应是没收和销毁他们持有的非法烟草制品,并 要求支付这些产品的未缴税款和关税。 在美国加利福尼亚州,拥有应纳税但尚未缴纳的烟草制品是非法的。证明已缴纳税 款的责任在于拥有产品的人。该规定由加利福尼亚州税费管理部和当地执法机构执行。 违规是一种轻罪,最高可处以5 000美元的罚款和最多一年的监禁。非法卷烟会被扣押 和没收。 大多数国家已通过立法打击有组织犯罪和洗钱活动。一些国家正在使用此类立法来 解决烟草非法贸易问题。对参与非法贸易的资产没收和加重处罚也变得越来越普遍。扣 311第三章 烟草税的管理 留甚至没收参与走私的卡车在一些国家也很常见。 关键要点17 施加的处罚和制裁应足以阻止非法烟草贸易活动。罚款的数额应与非法贸易造成 的税收和关税损失成比例。 3.5　其他烟草制品的税收管理 原则上,对卷烟以外的烟草制品征收类似于对卷烟征收的烟草税。然而,其他产品缺 乏标准化,有时还存在大型非正规市场。例如,据估计,欧盟2/3的水烟烟草是免税 的[19]。其他烟草制品(例如东南亚的比迪烟、东地中海地区的水烟烟草和瑞典的鼻烟)被 认为是一个国家传统的一部分。这有时会导致政府不愿对这些产品进行严格监管和征 税。 一些产品,如印度尼西亚的丁香卷烟与印度和孟加拉国的比迪烟,主要在一个市场销 售。其他烟草制品更可能是小规模手工生产,因此难以被发现和征税。这同样适用于 RYO 烟草,它可以手工或使用小型机械小规模生产。烟草原料贸易和小规模家庭烟草 生产通常在监测和控制系统之外进行[19]。如第3.3.1节所述,各国已找到解决此问题的 多种解决方案,包括采购或销售原材料的事先批准,以及对所有处理烟草原料的经营者和 种植者进行登记、授权或许可。 3.5.1　新型的尼古丁和烟草制品 原则上,将产品添加到现有税收框架中不太可能产生大量成本。可以合理预期,在对 新产品征税方面将面临与处理传统烟草制品时所面临的类似挑战,因为市场参与者将试 图利用税收监管中的漏洞尽可能避免或逃税。然而,当这些新产品涉及快速变化的技术 并且烟草市场动态广为人知时,预计会出现新的挑战。 此外,鉴于一个国家的应税项目和税基的不同特点,对新的烟草制品征税可能需要额 外的能力,一个国家可能需要确定新的代理与机构。 3.5.2　加热烟草制品（HTPs） 许多国家根据烟草重量对烟草制品征收特定的消费税(见第2章表2.4)。对于 HTPs,这可能会带来挑战,因为评估加热的烟草棒中的烟草含量将是额外的负担。从税 务管理的角度来看,主管机关将更容易对每支或每单位征税,就像对卷烟所做的那样。 3.5.3　电子尼古丁和非尼古丁传送系统（ENDS/ENNDS）产品 一些国家只对含尼古丁的电子烟油征税,而其他国家则对含尼古丁和不含尼古丁的 电子烟油征税。仅对含尼古丁的电子烟油征税需要实验室有能力检测尼古丁的存在(见 第 2 章表 2.5)。行业的自我声明是不够的,因为一些标有无尼古丁的电子烟油已被发现 411 世界卫生组织烟草税政策和管理技术手册 含有尼古丁(见第 2.4.2 节)。因此,对含尼古丁和不含尼古丁的电子烟油征税更简单。 对所有电子烟油征税的一个可能挑战是检测和区分 ENDS/ENNDS 中使用的电子烟油 是否在进口和制造层面被错误申报为用于其他目的的能力。第 2 章表 2.6 中提供了有 关不同消费税政策优缺点的更多信息。 对 ENDS/ENNDS 产品的其他成分征税的挑战在于它们的多样性(参见第 2 章第 2.4.2 节)以及某些成分用于其他目的(例如用于电池)的可能性。如前所述,快速变化的 技术以及缺乏对市场的控制和了解使得对 ENDS/ENNDS 设备征税具有挑战性。可能 正是因为这个原因,大多数对这些产品征税的国家只针对电子烟油。 在对这些较新的产品征税时,各国应该意识到许多客户在线购买他们所需的产品。 因此,建议决定对这些产品征税的国家制订适当的实施计划,包括如何对进口产品和在线 销售征税。并非所有国家都允许在线跨境销售烟草制品。一些欧盟国家已禁止此类销 售,其中包括从零售店向另一个国家的消费者进行在线销售。当然,这样的禁令只有在有 能力执行的情况下才有意义。 与对烟草制品征税一样,以下行动可以更有效地对这些产品征税: 1.实施强有力的执法机制,例如许可、记录保存和供应链控制,包括但不限于: a)对零售商、进口商和制造商实施严格的许可;理想情况下,为所有参与供应链的人 员颁发许可证,并为 ENDS/ENNDS 产品和 HTPs 制定跟踪和追溯制度(为了分摊成本, 这可以与为卷烟开发的系统一起完成)。 b)行使设定审计或控制的频率及类型的权利。 c)行使没收货物的权利。 d)若不遵守法律则实施处罚、罚款和/或吊销执照(如适用)等处罚。 2.对含尼古丁电子烟油宣称为“不含尼古丁”的生产商实施严厉的处罚。 3.当新产品投放市场或对现有产品进行重大修改时,需要支付实验室测试费用(分摊 成本)。 有关对 ENDS/ENNDS 征收消费税的政策选项的更多信息,请参见第 2 章第 2.4.2 节表 2.6。 关键要点18 原则上,新出现的尼古丁产品和烟草制品的税收管理应与卷烟相似。 然而,由于这些产品缺乏标准化,需要对它们及其供应链进行快速和不断发展的了 解,以实现有效和高效的税收管理。 3.6　良好税收制度的更广泛要素 3.6.1　主管当局的适当资源配置 除了采取行动和执行的法律工具和法律依据之外,还应为负责执行消费税法的主管 511第三章 烟草税的管理 机关提供足够的资源,雇用必要的人员来正确执行这些法律。必要的人员配备可能包括 多个机构,并且通常需要机构之间的合作,因为某些方面(例如监管、许可和边境控制)可 能由主管机关以外的机构执行。 主管机关的工作人员需要必要的工具、设备、培训和用品来履行其职责。该要求包括 构建、购买或维护允许纳税人以电子方式提交所需信息的软件系统的方法。电子申报对 纳税人和机关都有好处。它最大限度地降低了纳税人的合规成本,因此可以支持自愿合 规[68]。为识别违规风险,软件系统应让主管机关能够分析纳税人提交的数据,并将其与 其他税收(如增值税)和第三方来源(如银行和住户调查)的数据进行交叉核对。 主管机关的另一个选择是更有效地利用现有资源。例如,主管机关可以通过改用基 于风险的方法来优化风险管理系统:可以通过审计基于风险分析更有可能不合规的纳税 人而不是审计所有纳税人来节省资源。 其他挑战主管机关有效运作的问题是缺乏连贯的战略以及与缺乏培训或腐败相关的 专业问题[69]。制定战略可避免将资源用于不太重要的领域。该战略应始终与目标保持 一致,以便主管机关可以确定它们应该采取哪些步骤以及它们应该采取什么顺序来实现 这些目标。策略对于确定和组织资源的优先顺序是必不可少的,以便可以有效地解决已 识别的问题或风险。 3.6.2　腐败 主管部门应廉洁奉公地执行税法,并制定严格的腐败查处规章制度。还应制定严格 的规章制度,以惩处从事腐败行为的机构人员和纳税人。主管机关内部的腐败导致对税 收合规性的不当监测,是烟草制品非法贸易泛滥的原因之一。它还削弱了人们对主管机 关的信心,并最终削弱了对整个政府的信心。除了有效的法律法规外,还应实施包括预 防、调查和制裁在内的强有力的内部审计。为改进预防措施,应创建风险地图,突出显示 不当行为和可能有漏洞的领域。应制定更新控制措施的行动计划,以改进程序和系统中 检测到的薄弱环节。还应定期安排审核。配备有调查权力的内部审计师是必要的。对腐 败的制裁,包括行政制裁和刑事起诉,必须是强有力的。 3.6.3　强大的司法系统 司法系统在事实和观念上都应该是廉洁和独立的。争端应该迅速解决,而不是像一 些国家那样需要几年时间。上诉程序应该有限制,以使上诉不能持续数年。还应考虑使 用刑事而非民事指控,尤其是在非法贸易的情况下。 关键要点19 一个良好的税收系统的更广泛的要素包括: (1)适当的资源,如雇用员工和配置必要的设备和系统; (2)严格的规则和制度,以检测和惩罚机构人员和纳税人之间的腐败; (3)确保司法系统廉洁、独立,尽快解决纠纷。 611 世界卫生组织烟草税政策和管理技术手册 3.7　总结 如果政策得到适当的实施和执行,就会更有效。主管机关在实现消费税的财政和公 共卫生目标方面发挥着关键作用。鉴于税收管理与打击非法贸易导致的逃税行为之间的 密切联系,本章广泛借鉴了《议定书》相关内容。该《议定书》提供了解决非法贸易问题的 措施蓝图,即使是非缔约方的国家也可以将其当作样板。 有效且高效的税务管理的品质包括制度安排,其中明确界定主管机关的角色和职责 以避免重叠和空白。此外,必须促进相关机构之间的有效合作。在国家层面,在任何组织 安排中,机构合作和交流信息以及它们的权限在法律上找到依据是至关重要的。应确保 政府机构之间交换或获取信息的法律基础。在国际层面,特别是在边境管制方面,海关的 作用是关键,获得《议定书》等国际合作协议支持非常有用。一个组织的税务管理结构必 须包括通过预先定义的关键指标进行绩效评估和问责的系统。 为确保合规,税务合规周期信息的准确性是关键,包括清晰明了的纳税人登记和许 可、申报、记录保存、仓储、分配、征收和退税流程。 • 许可证是获取信息和确保烟草制品供应链安全的有力工具。理想情况下,所有参 与烟草种植以及烟草制品或制造设备的零售、运输、批发、代理、仓储和分销的人员都应获 得许可。 • 收集尽可能多的烟草业务信息并记录所有交易是减少逃税的关键,但这对主管机 关来说可能是负担。将信息技术用于定期纳税申报、会计、库存管理和财务,对于获取准 确信息至关重要,有助于降低整个报告系统的成本。 • 应确保记录保存。从事烟草、烟草制品和制造设备供应链的所有个人或实体都应 完整、详细和准确地记录所有相关交易以及生产这些产品所用材料的详细信息。 • 维护仓储授权系统使主管机关能够对生产和存储设施进行控制,以确保缴纳税 款。理想情况下,应从供应链中剔除保税仓库。 • 关税暂停———通常在需缴消费税货物的生产、加工、持有、接收和发送过程中应 用———只有在满足严格的标准时才应准予。例如,给予授权、仓库预授权访问、充分的库 存控制措施,检查需缴消费税产品的来源和整个生产过程以及产品的编码和标记。 • 为了限制主管机关必须管理的纳税人数量,征税应在生产和进口地点附近进行。 • 根据不出口税款的原则,增值税、消费税和关税的退税在大多数国家都很常见。 必须密切监控退税过程,以避免逃税的机会。 控制和执法(税收管理的关键组成部分)包括一系列保护供应链的措施:许可和尽职 调查、财政标记、跟踪和追溯、反买断措施、审计和控制、进出口控制以及对自由贸易区和 转运点的关注。控制和执法需要作为税收管理战略计划的支柱。必须设计执法和控制计 划来明确受执法约束的活动和纳税人,并分配人员、审计、基础设施和 IT 资源。必须确 定目标,包括干预的数量以及任何额外的征税或减少逃税。这包括为违规概率较高的人 选择干预措施(基于风险的方法)。在烟草供应链中,进口、出口和进出仓库的转运可能是 违规风险更大的领域。 711第三章 烟草税的管理 • 许可证提供了及时和准确的数据,可以作为审计的基础,因为它识别和控制合法 的运营商。必须定期执行和更新许可控制过程,特别是通过控制债券或担保的有效性以 及所需系统和记录保存的正常运行。在需要许可证的情况下,法律应包括禁止从无证供 应商处购买或向无证购买者销售的条款。这意味着供应商和采购商都需要核实与他们有 业务往来的人。这将大大有助于减轻当局的举证责任。此外,为了保持高水平的控制,应 及时限制许可证的有效期,从而需要更新或重新申请。 • 控制和监测烟草制品生产和进口的另一项重要措施是使用财政标记(例如税票)。 除了增加对税法的遵守外,财政标记还可以帮助区分合法的和非法的烟草制品。财政标 记的使用使主管机关和公众能够监测烟草制品税是否已正确缴纳。除了本地生产和进口 的产品外,还应要求出口的烟草制品有标记,但要注明它们是用于出口的。要求标准的包 装尺寸可以促进财政标记的应用。为了降低欺诈者试图重复使用财政标记(特别是印花 税票)的机会,在用玻璃纸包裹每包卷烟(或其他烟草制品)之前,应该在每包卷烟(或其他 烟草制品)上进行标记。只有在产品的消费税已全额支付或已建立担保后,才应向烟草制 品的制造商或进口商发放财政标记。财政标记应包括若干安全特征,以使其更难以伪造。 这些可以包括公开的、隐蔽的、半隐蔽的和取证的特征。 • 跟踪和追溯系统有助于主管机关确定烟草制品的来源和转移点(如适用),以及监 测和控制烟草制品的流动及其法律地位。跟踪和追溯系统的目标是向主管机关提供关于 整个烟草制品供应链的所有交易的信息,直到缴纳关税或其他义务被履行为止。任何跟 踪和追溯系统都必须能够唯一地识别单个产品。通过用唯一的代码或标识符标记一个产 品,就可以明确地记录该产品的流动。 良好的跟踪和追溯系统使政府能够正确监控供应链,提高其能力来确保征收适当的 关税和税收,验证识别标记是否真实和匹配产品,提高执法能力和提供足够的证据查处违 法者。为了减轻实施这一制度的财政负担,司法管辖区可要求烟草业承担成本。任何跟 踪和追溯系统都应符合FCTC第5.3条,政府应确保该系统独立于烟草行业。虽然财政 标记与跟踪和追溯系统的目标不同,但印花税票越来越多地包含跟踪和追溯功能。 • 实施法律措施以防止买断可以限制加税的延迟及其对收入和消费者行为的预期 影响。当制造商或进口商通过在增税前增加产量或进口以按先前较低的税率纳税,带来 库存增加或供过于求,此时就会发生买断、囤积或提前装载。 • 可以实施定期审计和控制以提高合规性。其中包括成本审计、转让价格审计、价 格和市场监控、消费者控制和交叉检查控制。 • 应仅允许获得正式许可的自然人或法人实体进出口烟草制品和制造设备。可以 通过要求提供担保或保证金来降低收入损失的风险,只有证明在另一个国家支付关税的 情况下才会解除控制。根据法律规定,除非在包装上贴有所需的财政标记(例如税票或出 口标签),否则不得允许任何烟草制品进入司法管辖区。出口的烟草制品应带有表明该产 品运往出口市场的标记。司法管辖区之间关于货物流动的信息交换也可以降低逃税的风 险。 非侵入性检测设备(例如 X 射线扫描仪)可用于海关检查,以检测违禁品。一种更便 宜的选择是使用经过训练可以检测卷烟和其他有机产品的狗。许多国家同时使用扫描仪 811 世界卫生组织烟草税政策和管理技术手册 和狗来检测走私烟草制品。还可以采用特殊的物理控制措施来减少违禁品。这些措施包 括将加工操作与已征税和未征税产品的密封储存分开。在一个国家内,转移消费税控制 单元有助于验证在国内运输的应税货物。这些单位应该被派往重要的交通走廊、通信中 心和交通拥堵的地区,如桥梁、渡轮和通道等地。实体控制行动需要警察、边防警卫和其 他公共服务部门之间的密切协作。 • 在自由贸易区和转运点,监管和监督等控制通常不那么严格。这可以使自由贸易 区吸引参与非法卷烟制造或贸易的人。海关管理部门应在自由贸易区行使其权力,有效 地识别和打击烟草制品的非法贸易。相关措施包括对自由贸易区内的所有运营商发放许 可证、尽职调查和记录保存,以及实施跟踪和追溯制度,并取消消费税的豁免。其他行动 包括禁止在将烟草制品从自由贸易区转移时,将烟草制品与非烟草制品混合在一个集装 箱或其他类似的运输单位中。应禁止向国际旅行者出售退税或免税的烟草制品,因为这 些销售会削弱旨在减少对烟草制品需求的税收和价格措施的影响,并通过在税收结构中 制造一个漏洞,对政府收入产生不利影响。 应明确界定发现烟草制品非法贸易后的程序。如果通过审计、跟踪和追溯系统、申报 核查或边境管制发现走私或非法贸易,必须立即采取查获、销毁走私和非法烟草以及征收 应缴税款等行动。同样重要的是,惩罚和制裁必须足以阻止非法活动。在非法活动继续 时,可以简单地支付较低的经济处罚作为开展业务的成本。对持有非法烟草制品的消费 者的最低处罚应是没收和销毁他们所拥有的产品,并要求支付这些产品的未缴税款和关 税。大多数国家已通过立法来打击有组织犯罪和洗钱活动。一些国家正在利用这类立 法,并利用它来解决烟草非法贸易问题。 原则上,对卷烟以外的烟草制品征收烟草税与对卷烟征收烟草税类似。对其他产品 征税的挑战包括这些产品缺乏标准化,有时是大型非正规市场。对产品和供应链的了解 大大有助于促进有效的税收管理。生烟草贸易和小规模家庭生产RYO和比迪烟等其他 产品通常发生在监测和控制系统之外。解决这一挑战的最佳方法是对原材料的购买或销 售实施事先批准,并要求所有生烟草的经营者和种植者进行登记、获得授权或许可。 原则上,在现有的税收框架中增加新的尼古丁和烟草制品预计不会带来重大成本。 我们有理由预期,在对这些新产品征收税收方面也将面临类似的挑战,因为市场参与者将 试图利用当前税收监管的漏洞,尽可能避免或逃避对这些产品的税收。然而,随着新产品 涉及快速变化的技术,而且它们的市场动态被普遍未知,预计将会出现挑战。此外,由于 应税项目和税基的不同特点,一个国家可能需要确立新的机构,因此可能需要额外的能 力。由于较新的尼古丁和烟草制品在网上被广泛购买,决定对这些产品征税的国家应该 制订一项适当的实施计划,其中包括关于如何对进口产品和网上销售征税的规则。一些 国家不允许在线跨境销售。 良好的税收体系的要素包括:(1)为主管机关提供足够的资源,以雇用必要的工作人 员来适当实施和执行消费税法;(2)廉洁奉公和严格的规章制度,以发现腐败,并惩处从事 腐败行为的机构人员和纳税人;(3)确保司法系统在事实上和观念上是廉洁和独立的。争 端应该迅速解决,而不是像一些国家那样需要几年时间。 911第三章 烟草税的管理 本章参考文献 1.WHO Framework Convention on Tobacco Control. Geneva:World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf; jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B? sequence=1,accessed 5 October 2020). 2.WHO FCTC guidelines for implementation of Article 6 Price and tax measures to reduce the demand for tobacco. Geneva:World Health Organization;2014 (https://www.who.int/ fctc/treaty_instruments/ Guidelines_Article_6_English.pdf? ua=1,accessed 5 October 2020). 3.The economics of tobacco and tobacco control. Bethesda:US Department of Health and Human Services,National Institutes of Health,National Cancer Institute;and Geneva:World Health Organization;2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf,accessed 28 September 2020). 4.Joossens L,Raw M. From cigarette smuggling to ilicit tobacco trade. Tob Control. 2012; 21:230-234 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.ful.pdf, accessed 5 October 2020). 5.Protocol to Eliminate Ilicit Trade in Tobacco Products. Geneva:World Health Organization; 2013 (https://apps.who.int/iris/bitstream/handle/10665/80873/9789241505246_eng.pdf? sequence=1,accessed 5 October 2020). 6.Reyes IP. Measures to control the tobacco supply chain in the ASEAN:a review of excise tax stamps,track and trace mechanisms,and licensing systems in the ASEAN region. Bangkok: Southeast Asia Tobacco Control Aliance;2014 (https://seatca.org/dmdocuments/ SEATCA%20MEASURE%20BOOK%20. pdf,accessed 5 October 2020). 7.Italy􀆳s tax administration:a review of institutional and governance aspects. Paris: Organisation for Economic Co-operation and Development;2016 (https://www.oecd.org/ tax/administration/italy-tax-administration-a-review-of-institutional-and-governance-aspects. pdf,accessed 5 October 2020). 8.Crandal W,Gavin E,Masters A. ISORA 2016:understanding revenue administration. Washington (DC):International Monetary Fund;2019 (Departmental Paper Series No. 19/ 05;(https://www.imf.org/en/ Publications/Departmental-Papers-Policy-Papers/Issues/ 2019/03/07/ISORA-2016-Understanding-Revenue-Administration-46337,accessed 5 October 2020). 9.Dutch customs in 2018. Heerlen:Customs Administration of the Netherlands;2019 (https://over-ons. belastingdienst.nl/wp-content/uploads/2019/04/Dutch-Customs-in- 2018.pdf,accessed 8 October 2020). 10.ESAs transform the way competent authorities cooperate with each other on AML/CFT 021 世界卫生组织烟草税政策和管理技术手册 matters [press statement]. European Banking Authority;16 December 2019 (https://eba. europa.eu/esas-transform-way-competent-authorities-cooperate-each-other-amlcft-matters, accessed 5 October 2020). 11.Ilicit trade report 2013. Brussels:World Customs Organization;2014 (http://www. wcoomd.org/-/media/wco/public/global/pdf/topics/enforcement-and-compliance/activities-and- programmes/ilicit-trade-report/itr_2013_en.pdf? la=en,accessed 5 October 2020). 12.Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation and repealing Directive 77/799/EEC. Brussels:OJEU;2011. (L 64/1; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/? uri=CELEX:32011L0016&from =en,accessed 29 January 2021). 13.439 stolen cars seized in 12 days as part of Frontex-led operation. Frontex News Release;14 October 2019 (https://frontex.europa.eu/media-centre/news-release/439-stolen-cars- seized-in-12-days-as-part-of-frontex-led-operation-vvvirI,accessed 5 October 2020). 14.Focus on:the ilicit trafficking of counterfeit goods and transnational organized crime. Vienna:United Nations Office on Drugs and Crime;2014 (https://www.unodc.org/ documents/counterfeit/FocusSheet/ Counterfeit_focussheet_EN_HIRES.pdf,accessed 5 October 2020). 15.Crandal W. Revenue administration:performance measurement in tax administration. Washington (DC):International Monetary Fund;2010 (Technical Notes and Manuals 10/ 11;https://www.imf.org/ external/pubs/ft/tnm/2010/tnm1011.pdf,accessed 7 October 2020). 16.Measuring tax gaps 2019 edition -tax gap estimates for 2017-18. London:HM Revenue and Customs;2019 (https://webarchive.nationalarchives.gov.uk/20200701215139/https:// www.gov.uk/government/ statistics/measuring-tax-gaps,accessed 29 January 2021). 17.Countering ilicit trade in tobacco products:a guide for policy-makers. Lyon:INTERPOL; 2014 (Legal handbook series;https://www.aph.gov.au/DocumentStore.ashx? id= 704d9168-2aa7-4df4-bb81-0fe1f357dfa6&subId=408832,accessed 29 January 2021). 18.Resoluçāo-RDC No. 226 de 30 de Abril de 2018 [Resolution of April 30,2018]. Ministério da Saúde/ Agência Nacional de Vigilância Sanitária/Diretoria Colegiada;2018 (in Portuguese) (https://www.in.gov. br/materia/-/asset_publisher/Kujrw0TZC2Mb/ content/id/12527411/do1-2018-05-02-resolucao-rdc-n-226-de-30-de-abril-de-2018- 12527407,accessed 17 February 2021). 19.Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels:European Commission;2017 (https://ec.europa.eu/ taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 20.Preece R. Addressing the ilicit flow of tobacco products in Australia. In:Dutta S (editor). Confronting ilicit tobacco trade:a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1. worldbank. org/curated/en/ 121第三章 烟草税的管理 677451548260528135/pdf/133959-REPL-PUBLIC-6-2-2019-19-59-24- WBGTobaccoIlicitTradeFINALvweb.pdf,accessed 29 January 2021). 21.Excise on tobacco [website]. Canberra:Australian taxation office;2019 (https://www. ato.gov.au/ business/excise-on-tobacco/,accessed 5 February 2021). 22.Safe Framework of Standards. Brussels:World Customs Organization;2018 (http://www. wcoomd. org/-/media/wco/public/global/pdf/topics/facilitation/instruments-and-tools/ tools/safe-package/ safe-framework-of-standards.pdf?la=en,accessed 29 January 2021). 23.The Trade Facilitation Agreement. Geneva:World Trade Organization;2017 (https:// www.wto.org/ english/docs_e/legal_e/tfa-nov14_e.htm,accessed 29 January 2021). 24.Authorised economic operator (AEO) [website]. Kildare:Irish Tax and Customs;2020 (https://www. revenue.ie/en/customs-traders-and-agents/authorised-economic-operators/ index.aspx, accessed 7 October 2020). 25.Performance outcome area 1:integrity of the registered taxpayer base. In:Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC): TADAT Secretariat;April 2019 (https://www. tadat.org/assets/files/IMF_TADAT- FieldGuide_web2.pdf,accessed 29 January 2021). 26.Commission implementing regulation (EU) 2016/323 of 24 February 2016 laying down detailed rules on cooperation and exchange of information between member states regarding goods under excise duty suspension pursuant to Council Regulation (EU) No 389/2012. Brussels:OJEU;2018 (http://data. europa.eu/eli/reg_impl/2016/323/2018-02-15, accessed 7 October 2020). 27.Council Directive 2008/118/EC concerning the general arrangement for excise duty repealing Directive 92/12/EEC. Brussels:OJEU;2009 (https://eur-lex.europa.eu/legal- content/EN/TXT/PDF/? uri=CELEX:3 2008L0118&from=en,accessed 29 January 2021). 28.Our strategy. London:HM Revenue & Customs,2017 (https://assets.publishing.service. gov.uk/ government/uploads/system/uploads/attachment _ data/file/629941/HMRC- Strategy.pdf,accessed 31 January 2021). 29.Strategic plan FY2018-2019. Washington (DC):Internal Revenue Service,2017 (https:// www.irs.gov/ pub/irs-pdf/p3744.pdf,accessed 31 January 2021). 30.Performance outcome area 2:effective risk management. In:Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC):TADAT Secretariat;April 2019 (https://www.tadat.org/ assets/files/IMF_TADAT-FieldGuide_web2.pdf,accessed 29 January 2021). 31.Tax administration 2017:comparative information on OECD and other advanced and emerging economies. Paris:Organisation for Economic Co-Operation and Development; 2017. 32.Compliance risk management:managing and improving tax compliance. Paris:Organisation for Economic Co-Operation and Development;2017 (https://www.oecd.org/tax/ 221 世界卫生组织烟草税政策和管理技术手册 administration/33818656. pdf,accessed 7 October 2020). 33.Petit P,Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund;2016 (International Monetary Fund How To Notes,No.3/ 2016;https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 34.Evans P. U.K. Fines British American Tobacco for Oversupplying Cigarettes in Belgium. The Wal Street Journal. 13 November 2014 (https://www.wsj.com/articles/u-k-fines- british-american-tobacco-for-oversupplying-cigarettes-in-belgium-1415873470,accessed 31 January 2021). 35.Doward J. BAT fined for oversupplying tobacco in low-tax European jurisdictions. The Guardian. 16 November 2014 (https://www.theguardian.com/business/2014/nov/16/ bat-fined-for-oversupplying-tobacco-in-low-tax-European-jurisdictions,accessed 31 January 2021). 36.Confronting ilicit tobacco trade:a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1. worldbank. org/curated/en/ 677451548260528135/pdf/133959-REPL-PUBLIC-6-2-2019-19-59-24- WBGTobaccoIlicitTradeFINALvweb.pdf,accessed 29 January 2021). 37.Cour des comptes. Accises sur les produits du tabac,Bruxeles [Court of Auditors,Excise duties on tobacco products,Brussels]. Bruxeles:Cour des comptes;2015 (in French) (https://www.ccrek.be/ docs/2015_27_AccisesProduitsDuTabac.pdf,accessed 7 October 2020). 38.Chaloupka FJ,Edwards SM,Ross H,Diaz M,Kurti M,Xu X,et al. Preventing and reducing ilicit tobacco trade in the United States. Atlanta:Centers for Disease Control and Prevention;2015 (https://www. cdc.gov/tobacco/stateandcommunity/pdfs/ilicit-trade- report-121815-508tagged.pdf,accessed 29 January 2021). 39.The Tax Stamp Forum 2015. Hexham:Reconnaissance International (https://10times. com/tax-stamp-forum-miami,accessed 10 September 2020). 40.Tax stamps:a technical study and market report. London:Reconnaissance international; 2012. 41.ЕАНС/2013/Health/l concerning the provision of an analysis and feasibility assessment regarding EU systems for tracking and tracing of tobacco products and for security features. Brussels:European Commission,2015 (https://ec.europa.eu/health/sites/health/files/ tobacco/docs/2015_tpd_tracking_ tracing_frep_en.pdf,accessed 29 January 2021). 42.Ross H. Controling ilicit tobacco trade:international experience. Economic research informing tobacco control policy. Cape Town:University of Cape Town;2015 (https:// tobacconomics.org/ uploads/misc/2015/05/Ross_International_experience_05.28.15.pdf, accessed 18 December 2020). 43.Security and resilience — authenticity,integrity and trust for products and documents — guidelines for the content,security,issuance and examination of excise tax stamps. Geneva: 321第三章 烟草税的管理 International Organization for Standardization;2018. 44.Borkowski F,Twomey C. European Union:an update on EU policies. In:Dutta S (editor). Confronting ilicit tobacco trade:a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1. worldbank. org/curated/en/ 677451548260528135/pdf/133959-REPL-PUBLIC-6-2-2019-19-59-24- WBGTobaccoIlicitTradeFINALvweb.pdf,accessed 29 January 2021). 45.Quality management systems — Fundamentals and vocabulary. ISO 9001:2015(en). Geneva:International Organization for Standardization;2015 (https://www.iso.org/obp/ ui/#iso:std:iso:9001:ed-5:v1:en,accessed 16 February 2021). 46.Protocol to Eliminate Ilicit Trade in Tobacco Products:questions & answers. Geneva: World Health Organization;2019 (https://www.who.int/fctc/protocol/faq/en/,accessed 7 October 2020). 47.Ross H,Eads M,Yates M. Why governments cannot afford Codentify to support their tracking and tracing solutions. Tob Control. 2018;27(6):706-708. 48.Servicio de Impuestos Internos,Chile,Noticias,28 May 2019. (http://www.si.cl/ noticias/2019/280519noti01er.htm,accessed 20 October 2020). 49.Directive 2014/40/EU of the European Parliament and of the Council. OJEU. 2014;127:1- 38 (https:// ec.europa.eu/health/sites/health/files/tobacco/docs/dir_201440_en.pdf, accessed 7 October 2020). 50.Report on the EU customs enforcement of intelectual property rights:results at the EU border. Luxembourg:Publications Office of the European Union,2019 (https://ec.europa. eu/taxation_customs/ sites/taxation/files/2019-ipr-report.pdf,accessed 7 October 2020). 51.Matthews P. Forestaling ahead of property tax changes. London:Office for Budget Responsibility;2016 (OBR Working Paper No. 10;https://obr.uk/docs/dlm_uploads/ Working-paper-No.10-1.pdf,accessed 29 January 2021). 52. Ross H,Tesche J,Velios N. Undermining government tax policies:common strategies employed by the tobacco industry in response to tobacco tax increases. Prev Med. 2017; 105S:S19-S22 (https://www. sciencedirect.com/science/article/pi/S0091743517302165? via%3Dihub,accessed 17 February 2021). 53. European Commission v Portugal supported by Belgium,Estonia and Poland,Case C-126/ 15. Judgement of the Court (Fifth Chamber);29 June 2017 (https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/? u ri=CELEX:62015CJ0126&from=EN,accessed 31 January 2021). 54. Amalia Valesko v Zolamt Klagenfurt,Case C-140/05. Judgement of the Court (Second Chamber);5 October 2006 (http://curia.europa.eu/juris/showPdf.jsf;jsessionid= 02DEE17CC11845BAE0F42FC7 CFD24C85? text = &docid = 65599&pageIndex = 0&doclang=en&mode=lst&dir=&occ=first&part=1&c id=2591459,accessed 29 January 2021). 55. European Commission v French Republic,Case C-197/08. Judgement of the Court (Third 421 世界卫生组织烟草税政策和管理技术手册 Chamber);4 March 2010 (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/? uri =CELEX:62008CJ0197&qid=1612223322830&from=EN,accessed 31 January 2021). 56. Commission requests that Belgium amend its legislation on tobacco:letter of formal notice. Brussels:European Commission;8 November 2018. (Memo 18/6247;https://ec.europa. eu/commission/ presscorner/detail/EN/MEMO_18_6247,accessed 29 January 2021). 57. Commission requests that Belgium amend its legislation on tobacco,MEMO/18/6247 of 8 November 2018. 58. Pederson H,Floristean A,Iseppi L,Dawkins R,Smith C,M? rup C,et al. Study on the measuring and reducing of administrative cost for economic operators and tax authorities and obtaining in paralel a higher level of compliance and security in imposing excise duties on tobacco products. Brussels:European Commission;2014 (https://ec.europa.eu/ taxation_customs/sites/taxation/files/docs/body/ rambol-tobacco-study.pdf,accessed 3 October 2020). 59. Money laundering vulnerabilities of free trade zones. Paris:Financial Action Task Force; 2010. https:// www. fatf-gafi. org/media/fatf/documents/reports/ML% 20vulnerabilities%20of%20Free%20Trade%20 Zones.pdf,accessed 7 October 2020). 60. Omi K.‘Extraterritoriality􀆳of free zones:the necessity for enhanced customs involvement. World Customs Organization;2019 (WCO research paper no. 47;http://www.wcoomd. org/-/media/wco/public/global/ pdf/topics/facilitation/ressources/permanent-technical- committee/225-226/itemixb_fzs_wco_e. pdf?la=en,accessed 29 January 2021). 61.European Parliament resolution of 26 March 2019 on financial crimes,tax evasion and tax avoidance. European Parliament;2019 (2018/2121(INI));http://www.europarl.europa. eu/doceo/document/TA-8-2019-0240_EN.html,accessed 7 October 2020). 62. International Convention on the Simplification and Harmonization of Customs Procedures. Brussels:World Customs Organization;2008 (http://www.wcoomd.org/Topics/ Facilitation/Instrument%20 and%20Tools/Conventions/pf_revised_kyoto_conv/Kyoto_ New,accessed 7 October 2020). 63. Evidence Notice 2004:tobacco duty -tobacco products manufacturing machine licensing scheme. London:HM Revenue & Customs,2018 (https://www.gov.uk/government/ publications/excise-notice-2004-tobacco-duty-tobacco-products-manufacturing-machine- licensing-scheme/excise-notice-2004-tobacco-duty-tobacco-products-manufacturing-machine- licensing-scheme,accessed 31 January 2021). 64. Tackling ilicit tobacco:from leaf to light:the HM Revenue & Customs and Border Force strategy to tackle tobacco smuggling. London:HM Revenue and Customs;2015 (https:// assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/ file/418732/Tackling_ilicit_ tobacco_-_From_leaf_to_light 2015_.pdf,accessed 7 October 2020). 65. Maldonado N,Llorente BA,Iglesias RM,Escobar D. Measuring ilicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/ 521第三章 烟草税的管理 tobaccocontrol/29/Suppl_4/ s260.ful.pdf,accessed 29 January 2021). 66. Cardenas M. Case studies in ilicit tobacco trade:Colombia. Chicago:University of Ilinois at Chicago;2020 (Tobacconomics Fact Sheet;https://tobacconomics.org/files/research/ 606/UIC_Colombia-Ilicit-Trade-Fact-Sheet_v1.4.pdf,accessed 29 January 2021). 67. Maldonado N,Llorente BA,Escobar D,Iglesias RM. Smoke signals:monitoring ilicit cigarettes and smoking behaviour in Colombia to support tobacco taxes. Tob Control. 2019;29:s243-s248 ( https:// tobaccocontrol.bmj.com/content/tobaccocontrol/29/Suppl_ 4/s243.ful.pdf,accessed 29 January 2021). 68. Tax Administration Diagnostic Assessment Tool (TADAT) field guide. Washington (DC):TADAT Secretariat;April 2019 (https://www.tadat.org/assets/files/IMF_ TADAT-FieldGuide_web2.pdf,accessed 29 January 2021). 69. Pelechio A,Tanzi V. The reform of tax administration. Washington (DC):International Monetary Fund;1995 ( Working Paper no. 95/22;https://www.elibrary.imf.org/view/ IMF001/07242-9781451843941/07242-9781451843941/07242-9781451843941_A001.xml? language=en&redirect=true,accessed 29 January 2021). 621 世界卫生组织烟草税政策和管理技术手册 附录3.1　烟草制品组成 为了以最有效的方式实施和执行烟草税政策,主管机关应熟悉烟草制品的所有组成 部分,包括在其制造中使用的每种原材料、生产投入和烟草制造机械。了解应税产品和机 械的组成部分可提供有价值的信息,以识别违规风险高的活动,从而采取措施确保收缴所 有税款并防止非法贸易。 尼古丁、非尼古丁和烟草制品及其组成部分 在大多数国家,财政部决定税收政策,包括对哪些烟草制品征税,而卫生部负责产品 和使用监管。这可能会导致对同一产品的不同定义,具体取决于负责特定法律或法规的 部门。在可能的情况下,应制定明确和通用的定义,以简化程序并避免混淆。烟草制品有 多种形式,并非所有产品都在特定司法管辖区受到监管或征收消费税。除卷烟外,其他传 统烟草制品包括无烟烟草———如嚼烟、鼻烟———以及可以手工卷制或制造的比迪烟和丁 香卷烟、烟斗、水烟和雪茄。 卷烟 了解特定国家最常用的烟草制品的材料和成分是很重要的。在大多数司法管辖区, 就数量和税收收入而言,卷烟是最常见和最重要的烟草制品。 一根卷烟棒由以下部分组成: • 各种烟草植物(叶、茎及其他植物部分)和添加剂(包括香精)的烟草混合物; • 用于包裹烟草混合物以构成烟条的卷烟纸; • 醋酸纤维滤嘴,在过滤嘴卷烟尖端形成白色部分,与吸烟者的嘴直接接触; • 接装纸或包裹过滤器; • 将卷烟纸固定在烟草混合物和烟头周围的黏合剂[1]。 每个制造商都遵循特定的流程来生产卷烟。除了烟草混合物,制造商还会改变卷烟 纸和水松纸的尺寸以及每支烟嘴使用的醋酸纤维过滤嘴的长度[1]。在一些国家,这些元 素是标准化的。在理想的监管框架中,作为许可要求的一部分,制造商需要向主管机关提 交有关其生产的每个品牌和烟草制品衍生品的具体流程的信息(见第 3.4.3 节)。例如, 主管机关可以要求制造商提交此信息以获得许可证。行政部门的最低要求和应包含的信 息可以在法律或较低一级的法规中规定,以确保主管机关掌握所有许可制造商的信息。 通过将用于生产的材料数量和每支卷烟使用的数量与制造的卷烟总数进行比较,该有助 于验证公司是否报告了制造和销售的卷烟的实际数量。最终,这些信息还有助于验证制 造商是否正确缴纳税款。 图A3.1显示了一种典型的机械制造的传统卷烟的组成部分。 并非烟草制品的所有部分都受到相同水平的控制。根据《议定书》第 6.5 条,在本 《议定书》生效五年后,《议定书》缔约方会议(MOP)应确保在其下届会议上进行循证研 究,以确定是否存在任何对烟草制品制造至关重要的关键投入,是可识别的并可以受到有 效的控制机制约束。以此类研究为基础,MOP 应考虑采取适当的行动。 721第三章 烟草税的管理 F 过滤区 烟杆 过滤器 水松纸 卷烟纸 烟草 图A3.1 一种机械制造的卷烟的零部件 资料来源:作者的汇编。WalterKlerx的照片。 除了烟草制品的组成部分外,还可以监测将特定数量的烟支包装成一包卷烟(通常每 包 20 根)所需的材料。这些材料包括箔纸、包装纸(可以带有品牌名称、设计和健康警 告)、财政标记(如果需要)和塑料或玻璃纸包装。固定数量的卷烟,通常为 10 包,被装入 纸盒,也称为“条”。这些纸盒通常由软纸板或硬纸板制成,可能带有商标,并用塑料包装 或玻璃纸包裹。五十个纸盒包装在主箱中,主箱由更坚固、更厚的纸板制成,并堆放在托 盘上(通常一个托盘有 50 个主箱)。有效的监管框架将要求制造商和进口商向主管机关 提供有关包装和设计的信息,以及每个包装、纸箱和主箱数量。 购买或销售卷烟生产过程中使用的材料也可能需要事先批准。在菲律宾,此类原材 料的供应商,包括那些提供烟草纸和过滤嘴组件的供应商,都需要获得许可证[2]。在欧盟 的一些成员国,生烟草也受到财政和法律要求的约束。例如,在斯洛伐克和波兰,生烟只 能由授权经营者处理。虽然经授权的经营者不必为生烟草支付消费税,但如果未经授权 的经营者被检测到经营生烟草,则应缴纳消费税。匈牙利、意大利和英国要求所有生烟草 的经营者和种植者进行登记或授权[3]。 除了了解生产特定数量的受管制产品(例如卷烟)所需的投入外,主管机关还需要了 解供应链、制造链和分销链,以便能够正确地监控、监管和确定各环节是否已缴纳税款(另 请参见图 3.3)。 新型和新兴的尼古丁、非尼古丁产品和烟草制品 近年来,新产品已被引入多个市场,即ENDS、ENNDS和HTPs。 ENDS 通常包含含有尼古丁的电子烟油,但不包含烟草。ENNDS 本质上是相同的, 但(表面上)不含尼古丁。世界卫生组织 COP 要求公约秘书处邀请缔约方监测和报告科 学、监管和市场发展,例如 ENDS 和 ENNDS 的启动、停止、广告和推广。此外,COP 要 求世界卫生组织报告区域和国际标准制定组织为测试和测量这些产品的成分和排放制定 方法[4]。 电子烟有不同类型,最常见的 ENDS 和 ENNDS 类型目前有四代产品。但是,它们 可以分为两大类:开放系统和封闭系统。两种类型的电子烟都使用加热元件和热源来产 生气溶胶。加热元件中充满了电子液体,并使用微处理器来控制操作(并非全部包括此)。 821 世界卫生组织烟草税政策和管理技术手册 一些电子烟还带有 LED 灯来模仿传统卷烟的燃烧端[5]。图 A3.2 展示了开放和封闭系 统的例子。 开放 ENDS/ENNDS 系统（电子烟） 可充电电池 电源按钮（开始使用电子烟） 护套 加热器/加热元件 （热溶液，空气雾化尼古丁） 电子液罐 （可再填充的电子液体尼古丁罐） 烟油舱盖 封闭 ENDS 系统（电子烟） 电子液舱 可充电电池 吹口 加热元件 （热溶液，雾化尼古丁） 图A3.2 ENDS/ENNDS产品的开放和封闭系统的示例 来源:参考文献6。 与 ENDS/ENNDS 不同,HTPs 确实含有烟草。当加热烟草或激活含有烟草的装置 时,HTPs 会产生含有尼古丁和有毒化学物质的气溶胶[7]。HTPs 由两个元件组成:包含 烟草的棒或烟弹和用于加热烟草的装置。图 A3.3 显示了加热烟草制品的示例。HTPs 是烟草制品,因此受世卫组织FCTC 中包含的监管措施的约束。 921第三章 烟草税的管理 充电器手持器 烟叶梗 手持器 壳 电池 控制电子元件 加热片 烟草棒 MPF （7 mm） 空心醋酸纤维管 （8 mm） 外层纸 直径最大 （7.42 mm） 烟草插头 （12 mm） PLA （18 mm） 水松纸 加热棒总长：45 mm 图A3.3 HTP的组件 注:PLA:聚乳酸,MPF:吹口过滤器。 来源:参考文献8和9。 关于其他烟草制品的税收征管的更多信息见第3.5节。 031 世界卫生组织烟草税政策和管理技术手册 附录3.1参考文献 1.Memorandum by Galaher Group Plc:the tobacco industry and the health risks of smoking. Select Committee on Health Minutes of Evidence. London:UK Parliament,2000 (https:// publications. parliament.uk/pa/cm199900/cmselect/cmhealth/27/0011323.htm,accessed 7 October 2020). 2. Petit P,Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund;2016 (International Monetary Fund How To Notes,No.3/ 2016;https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 3. Study on Council Directive 2011/64/EU on the structure and rates of excise duty applied to manufactured tobacco. Brussels:European Commission;2017 (https://ec.europa.eu/ taxation_customs/sites/taxation/ files/study_on_directive-2011_64_main_text_en.pdf, accessed 28 September 2020). 4. Electronic nicotine delivery systems and electronic non-nicotine delivery systems. Geneva: World Health Organization;2016 (Decision FCTC/COP7(9) of the Conference of the Parties to the WHO Framework Convention on Tobacco Control;https://www.who.int/fctc/cop/ cop7/FCTC_COP7_9_EN.pdf? ua=1,accessed 7 October 2020). 5. Brown CJ,Cheng JM. Electronic cigarettes:product characterization and design considerations. Tob Control. 2014;23:i4-i10 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC3995271/pdf/ tobaccocontrol-2013-051476.pdf,accessed 31 January 2021). 6. E-cigarettes and vapor products [webpage]. King County,Washington (USA);2019 (https://www. kingcounty.gov/depts/health/tobacco/data/e-cigarettes.aspx,accessed 9 October 2020). 7. Heated tobacco products. Geneva:World Health Organization;2020 (WHO/HEP/HPR/ 2020.2 Information sheet;https://apps.who.int/iris/bitstream/handle/10665/331297/ WHO-HEP-HPR-2020.2-eng.pdf? sequence=1&isAlowed=y,accessed 7 October 2020). 8. Tobacco heating system (IQOS) briefing document. Silver Spring:US Federal Drug Administration;2018 (https://www.fda.gov/media/110377/download,accessed 31 January 2021). 9. Premarket tobacco product application:technical project lead review. Silver Spring:US Federal Drug Administration;2017 (https://www.fda.gov/media/124247/download, accessed 7 October 2020). 131第三章 烟草税的管理 附录3.2　买断和对策示例 这个例子是假设的,是受到菲律宾情况的启发。然而,数量和价格已经改变,并且假 设烟草制造商的正常库存为2个月。 本年度对一包卷烟征收的消费税为3.00美元;在新财年开始时(本例为1月)将增加 到3.30美元。X公司宣布的Y品牌卷烟的每月产量如下: 本年度的一个月份 卷烟包数 一月 10 000 000 二月 10 500 000 三月 9 900 000 四月 11 000 000 五月 10 200 000 六月 10 600 000 七月 9 700 000 八月 10 100 000 九月 9 900 000 十月 10 100 000 十一月 20 000 000 十二月 25 000 000 由于正常库存为两个月,因此不考虑新消费税实施前2个月的数量。烟草制品的保 质期约为6个月。如果没有增税,则计算 11 月之前6个月的平均值以获得推定生产或进 口的数量。5月至10月(含)的数量除以6为 10 100 000 包。因此,11月和12月(新税率 实施前的几个月)生产的超过 10 100 000 包的任何数量都将使用新税率进行评估。 在此示例中,11月生产的 10 100 000 个包装将按 3.00 美元的旧税率征税,9 900 000 个包装将按 3.30 美元的新税率征税。12月,10 100 000 包的税率为 3.00 美元,而 14 900 000 包的税率为 3.30 美元。如果不采取这些措施,政府将无法提高 2 480 万包的消 费税。此外,涨价对价格和消费者的影响将延迟大约2个月。 在使用印花税票的国家,扣留印花税票是一种众所周知的反买断方法。另一个实用 的解决方案是允许主管机关要求行业提供预付款以弥补收入不足,前提是此类请求有法 律依据。 231 世界卫生组织烟草税政策和管理技术手册 第四章 烟草税收的政治经济学 与任何拟定的政府行动一样,决策者需要在政策制定、实施和管理的每一个阶段控制 烟草税的政治环境。虽然每个国家独特的历史、文化、制度和结构塑造了其独特的政治环 境,但在烟草控制,特别是烟草税收方面,都有一些普遍的特征。这些特征可以归结为金 钱、权力和资源的分配。烟草业作为政治和经济的参与者,应该非常了解这些特征。 该行业一直有效地利用烟草税收的政治经济学原理来阻止烟草控制方面的重要进 展。尽管如此,精明的决策者可以通过考虑谁从行业青睐的政策措施和干预中受益来看 穿行业争论。该行业对烟草税政策的挑战可以分为五类 SCARE 威胁策略。本章提供了 一个路线图,以帮助决策者通过这些主题中的每一个来了解烟草税的政治经济学。 前五个部分剖析了烟草业每个问题的行业框架,精确地指出了每个论点的缺陷,确定 了每个问题的价值程度,并向负责任的政府针对如何解决每个问题提出了建议。这些讨 论得到了来自独立同行评审研究的公正证据以及来自国家经验的具体案例的支持。关于 SCARE策略的第 4.1 至第 4.5 节将为政策制定者提供他们所需的工具,以确保他们的 烟草税政策(假设按照本技术手册中阐明的指导方针制定和实施)将为他们的人民带来最 大的健康和经济利益,而不管行业会试图阻挠他们。第4.6节进一步支持决策者努力确 保其政策的有益影响,因为它描述了如何通过资助促进和支持人口健康和福祉的方案和 倡议来改善烟草税收的政治经济环境。 4.1　威胁策略S：走私和非法贸易 4.1.1　介绍 烟草业及其盟友认为,提高烟草税将不可避免地导致烟草制品非法贸易的增加[1,2]。 他们声称,更高的税率和更高的价格加强了犯罪企业从税收较低的司法管辖区供应卷烟、 促进国内逃税和鼓励吸烟者寻求更便宜的非法卷烟的经济动机。该行业还挑战烟草税增 加政府收入的论点,声称非法烟草市场的存在实际上会减少税收。这一论点的最新版本 (适用于解决对烟草使用的公共卫生问题)声称非法市场增长也抵消了吸烟率的下降,否 则烟草税增加会导致吸烟率下降。总之,烟草业及其盟友声称提高烟草税是无效的,甚至 适得其反,因为它们被非法市场规避,这会阻止政府实现其公共卫生目标并减少税收。 当一个国家考虑增加烟草消费税的提案时,烟草业及其盟友经常夸大该国非法烟草 贸易的规模和范围。反对增加烟草税的人认为,价格差异是非法贸易的唯一原因,或者至 少是主要原因。受这种令人恐惧的错误判断的影响,税务机关经常发现难以就烟草税做 出决定。然而,行业判断总是包含着同样的错误因素。首先,一个国家的非法贸易往往低 于该行业所描述的水平,而且该国对烟草制品的税收执法政策很少是独特的或与国家的 规范有任何不同[3]。其次,烟草非法贸易的规模并不完全一致,甚至不是主要由税收或价 格差异决定的。通常,它是由一系列以政府腐败、监管框架薄弱、烟草税收管理不善、刑事 司法系统无效、缺乏劝诫性制裁和/或参与非法和非正式市场规范薄弱为特征的治理问题 造成的[4~7]。 本节为税务部门和其他相关部门提供指导,指导他们如何应对烟草业的 SCARE 策 略,即提高烟草税将导致其国家的走私和非法贸易。税务机关需要了解非法贸易的性质、 原因和范围,以便正确界定问题属性并制定适当的应对措施。本章讨论了可用于更好地 定义和理解特定非法贸易问题的可用工具,特别是有助于对该贸易规模进行独立评估的 工具。第 3 章讨论了税务机关领域治理的改进,例如烟草税收管理的最佳实践以及提高 财政法规或规范对遏制非正规和非法市场的有效性的政策。 本节首先描述非法烟草贸易的性质,以突出其复杂性并确定解决问题的补充政策。 接下来,我们将讨论对非法贸易与高价格或税率变化之间的联系提出质疑的证据。最后, 为了帮助税务机关评估自己的情况,我们提出了几种不同的方法来估计非法烟草贸易的 范围,并评估某一特定国家或税收管辖区的贸易估计值。 4.1.2　非法烟草贸易的性质和范围 世卫组织FCTC将非法贸易定义为: 法律禁止的与生产、运输、接收、占有、分销、销售或购买有关的任何实践及行为,包括 旨在促进此类活动的任何实践及行为[8]。 在一个司法管辖区发现的未纳税烟草制品(即通过垃圾包装调查)可能是两种相关但 截然不同的活动的结果:逃税和避税。逃税是一组寻求不支付烟草税和关税的违法行为, 而避税的目的是避免支付部分或全部税收的合法行为,如把卷烟的法律津贴从一个较低 的水平调整到一个更高的税收补贴。避税并不违法,因此不被认为是烟草制品非法贸易 的一部分 。 本节的重点是逃税活动,这可能发生在跨境流通或在国内生产和分销中。当跨境发 生逃税时,它被称为走私[9],可以大规模或小规模进行。当获得许可和授权的生产商或经 431 世界卫生组织烟草税政策和管理技术手册 本节没有对美国各州和欧盟国家之间常见的避税做法进行分析。 销商只履行部分税收义务时,国内市场的逃税可能是部分的,当整个生产和分销系统非 法,并且在税务管理者的视线之外时国内市场的逃税也可能是全部的[5,1014]。 大规模的 逃税计划可以由不同类型的生产商及相关分销商实施,例如跨国烟草公司(TTC) 及其全 国子公司、其他生产自有品牌的当地烟草公司和非法假冒其他品牌或生产非法品牌的工 厂 。 例如,大规模走私涉及利用免税区和在运输前或运输过程中贴错标签[11]或使用复杂 的秘密网络。这种形式的逃税是系统性的,可以由 TTC[12]或在税收执法较低的国家生 产的当地公司进行,例如,巴拉圭[15]有人通过秘密分销商[16]网络供应邻国和地区非法中 心[15]、位于阿联酋的杰贝尔阿里和迪拜等免税地区的公司[16]、俄罗斯和塞浦路斯[17]等 地。随着受影响国家或市场通过加强执法和寻求与原产地司法管辖区的双边合作做出反 应,大规模走私的来源、路线和数量经常发生变化。例如,欧盟2013年的全面战略[10]加 强了与包括俄罗斯、白俄罗斯和乌克兰在内的国家打击进入欧洲的主要非法卷烟来源和 过境国的双边合作。这些协议改善了这三个国家的日常跨境合作,减少了烟草制品非法 流动,并逐步调整烟草消费税,使其提高到欧洲水平。 小规模走私(也称为蚂蚁走私)是数量大于允许限度(例如两箱)但小于大货量(例如 卡车、货柜)的卷烟跨境走私,通常以出售获利为目的[10]。这种类型的非法贸易可能存在 于邻近税收管辖区内有机会的地方。例如,小规模走私通常由居住在靠近低税收国家(例 如比利时、卢森堡、瑞士、西班牙、波兰和捷克)附近的法国和德国的个人进行[18]。 假冒是一种非法制造形式,涉及未经商标持有人批准生产烟草制品(包括包装和烟草 填料)[13]。 非法制造的另一种产品是所谓的廉价或非法白烟。廉价的白烟是合法或非法生产的 品牌卷烟(例如JinLing)或非品牌卷烟 ,并有意在非法市场上销售[17]。廉价的白烟通常 不是由 TTC 生产的[17,19] 。它们由一个国家的小型烟草生产公司生产,并经常在邻国 的非法市场上销售。例如,在巴拉圭,少数公司打着在国内销售的幌子大规模生产廉价白 葡萄酒,但很大一部分被走私到乌拉圭和巴西[4]。Iglesias等人[20]展示了 20世纪90 年 代 TTC 的廉价品牌如何通过巴拉圭非法运输到巴西和阿根廷市场进行销售。这有助于 巴拉圭公司增加廉价白烟的生产。在巴西,即使在立法阻止 TTC 的非法活动之后,这些 公司仍继续从事非法业务。 国内逃税是一种普遍现象,尤其是在中低收入国家。烟草制品的部分逃税可以在任 何税率或价格水平上找到,通常是立法有缺陷或税收执法不力的结果。 当生产者和分 531第四章 烟草税收的政治经济学 在文献14中,逃税通常被认为是非法制造。 Ross 等人[17]分析了这个问题,发现出售给第一个购买者通常是合法的。他们的分析涵盖了自由地区(即阿联 酋、俄罗斯和塞浦路斯)的生产,以及从越南、印度尼西亚和中国出口的产品。在这种情况下,没有必要非法进行 第一次销售。然而,巴拉圭廉价的白色产品是卖给国内分销商的,其中大多数销售是完全非法的。 Ross等人[17]和Gilmore等人[19]确定了一些由TTC销售的廉价白色品牌,如在乌克兰生产的总统(PMI)牌和在 印度尼西亚生产的Esse(韩国烟草和人参公司,KT&G)。 当许可和授权生产商低估实际数量并通过非法渠道销售未纳税的产品时,就会发生这种情况。在许多中低收入 国家的从价计算系统中,还会包括生产者根本不报告数量的例子。 销商是秘密的,或者当两个税务管辖区之间存在严重的政策冲突时,就会有完全或全部逃 税。如美国和美洲土著保护区之间的税收执法。近年来,非法制造的证据在世界多个地 方有所增加,包括欧盟[10]和巴西[21]。 直到 20 世纪末,TTC 在非法贸易活动中占主导地位,即使有新的参与者进入非法贸 易,TTC 也没有完全退出。Gilmore等人[22]分析了行业资助的数据和缉获数据,并得出 结论认为,尽管四个 TTC 与欧盟签署了反走私和反假冒协议,但 TTC 仍参与欧洲的非 法贸易[23] 。使用行业资助的数据,Gilmore 等人表明 58% 的非法欧盟卷烟可归因于四 种主要的 TTC。当使用缉获数据时,69%~73% 的非法欧盟卷烟可归因于这些公司[22]。 由于其非法性、全球性和不断变化的性质以及数据收集方面的问题,评估全球非法烟 草贸易的程度是很困难的[24]。在 21 世纪之前,当 TTC 在知名卷烟品牌的大规模走私中 几乎是唯一的参与者时,根据全球卷烟出口和进口之间的差异可以很好地估计这个问题 在全球范围内的规模[7]。然而,随着过去二十年非法制造、廉价白烟的制造以及这些产品 的非法流通的普遍增长,贸易统计数据不再像过去那样有用。Joossens 等人[25]试图通过 添加 2007 年左右准备的不同类型的国家的规模来估计全球非法卷烟市场的规模。他们 发现估计的规模为每年 6 570 亿支卷烟,占全球卷烟市场的 11.6%。Joossens 等人称, 非法贸易使各国政府损失 405 亿美元的税收,而消除非法烟草贸易将在高收入国家增加 130 亿美元的直接收入,在中低收入国家增收 183 亿美元 。 对非法贸易类型的描述有助于制定打击非法贸易战略的第一个组成部分:评估问题 的性质和规模。表 4.1 列出了打击非法贸易战略的所有主要组成部分。为了在第一部 分取得进展并了解问题,管理机关可以使用和调整现有的健康监测工具、寻求与学术界和 独立专家的伙伴关系,使用不同的方法来调查所涉及的问题(见第 4.1.4 小节和附件 4.1 关于评估问题性质和规模的方法)。了解问题的性质需要不同政府行为者的合作,例如, 调查非法贸易背后有组织的金融犯罪活动。收集关于非法贸易性质的定性信息应与衡量 问题严重程度的统计工作同时开始。 631 世界卫生组织烟草税政策和管理技术手册 从 2000 年起,欧盟委员会和 10 个成员国针对几个 TTC 发起了关于走私和洗钱的法庭案件。为了结束法庭案 件,签署了反走私和反假冒协议,要求 TTC 对其供应链实施严格控制(通过跟踪和追踪、尽职调查和反洗钱和报 告义务),分享与成员国和欧盟的运营情报,支付扣押罚款,以及 12 年期间的年度付款。与 PMI 的协议已经结 束,与 JTI 的协议将于 2022 年结束,而与该国烟草和 BAT 的协议将持续到 2030 年(23)。 世界海关组织每年发布一份非法贸易报告,利用基于海关查获的数据,说明包括烟草在内的关键产品中非法流通 的主要特征和趋势。 表4.1 打击非法烟草贸易战略的组成部分 1 2 3 1.评估问题的性质和规模 利用和调整现有的健康监测和其他现 有的国家调查来评估这个问题 寻求与学术界和独立专家的合作关系, 以寻找严格研究非法贸易的方法 利用金融和警察的调查来查明和打击 在非法贸易中经营的有组织犯罪 2.开始确定和执行适当的具 体国家的政策和战略,以杜绝 非法贸易 完善税务、海关管理,堵住助 长非法贸易的法律和行政漏 洞 实施其他适当的政策,以处理 具体国家的问题 3.成为缔约方和执行《议 定书》 调整协议中的供应链控制 义务 调整对非法贸易犯罪的国 家处罚 寻求和建立国际合作 可用于估计非法贸易的性质和规模的方法在第 4.1.4 小节中讨论,这是应对 SCARE 策略中的第一步。第 3 章详细讨论了相关的税收管理措施和最佳做法,以尽量 减少烟草制品非法贸易的机会。除了第 3 章所述的最佳做法外,表 4.2 列出了旨在解决 特定类型非法贸易的适当政策和战略示例。 在完成该战略的第一步后,税务、卫生和司法机关应讨论如何面对特定国家的问题, 不仅要考虑税收和海关管理措施,还要考虑社会、执法和国际合作政策和战略。 表4.2 解决针对特定国家的非法贸易问题的建议政策和战略 辖区内非法贸易的主 要类型 问 题 要使用的政策和策略 非法制造 邻近的低税收司法管辖权 协调烟草税收制度的双边谈判 在土地边界广泛的国家,控制人 员流动困难 为边境地区的目标人群确定和建立适当 的社会保护或就业政策 广泛的陆地边界和多重通道 与执法部门和边境警察进行双边合作, 监测通往主要消费市场的通道 在邻近司法管辖区进 行的大规模走私 邻近低税收管辖权和在边界控 制方面的困难 双边谈判,以协调烟草税收制度和双边 执法合作 低税收辖区的生产者和分销商, 旨在为高税收辖区提供产品 双边合作,协调税收制度,控制原产国的 生产者和分销商,为合法出口和进口征 税创造条件 731第四章 烟草税收的政治经济学 续表 辖区内非法贸易的主 要类型 问 题 要使用的政策和策略 从第三国或免税区进 行大规模走私 旨在尽可能提供免税烟草制品 的生产商和分销商 海关和其他形式的国际合作,以控制和 监测来自确定地区的出口 国内逃税 存在许多小型的非正式或半正 式的生产者 通过生产者协会和合作社鼓励企业集 中,创造激励措施正式建立原材料和生 产的许可规则和基本电子信息系统 来自正规生产商的漏报 通过诸如投入品和生产的基本电子信息 系统等政策来改善税收管理,建立中立 程序以验证生产,改进审计系统,增加关 于投入品以及烟草制品生产的第三方 信息 秘密工厂 与原材料和机器生产商及分销商的商业 协会执法调查 政府应努力查明鼓励和允许国内非法贸易流通的激励措施和治理问题。如表 4.2 所示,处理这些问题的政策的设计和实施并不完全取决于税务和海关机关;它们还取决于 警察和执法部门、司法机构的努力。换言之,需要在不同类型的政府机构和专业部门之间 进行大量的协调和协商,才能做出适当的反应。从表 4.2 中也可以看出,正规生产者的 国内逃税可以由税务机关处理,主要与议定书的供应链控制条款有关。 《议定书》[26]以世界卫生组织 FCTC 第 15 条为基础并对其进行了补充,该条将打击 烟草制品非法贸易的手段作为全面烟草控制战略的一个关键方面。《议定书》是解决这一 问题的措施蓝图,其条款应成为打击非法市场战略的一部分。它本身就是一项具有法律 约束力的条约,于 2018 年 9 月 25 日生效。正如第 3 章详细描述的那样,该《议定书》具 有三个主要行动路线:供应链控制、关于如何处理与非法烟草贸易有关的违法行为以及建 立机制以寻求和建立国际合作以打击这种贸易。各国甚至可以在谛约之前就开始实施世 卫组织框架公约第 15 条和《议定书》建议的适当政策或战略,选择最适合解决其特定问 题的政策或战略。这种过渡性工作将促进《议定书》的最终实施,因为任何纠正税收和海 关做法漏洞的计划都将使政府当局更接近《议定书》中建议的最佳做法。 4.1.3　逃税的决定因素：价格水平的作用 关于价格和税率是非法烟草贸易的主要决定因素的观点已经说服了一些政府(如乌 拉圭和格鲁吉亚)避免实施可能导致卷烟价格上涨的政策(如消费税税率上涨)[4]。 一些政府(如1994年的加拿大、1999年的巴西和2017年的巴基斯坦)[20,27]甚至降低 了税率,试图减少非法贸易。 更广泛的学术文献表明,非法贸易不是单一因果现象[7],而是多种因素的结果,其中 大多数与治理问题有关。政府腐败、监管框架薄弱、烟草税收管理不力、刑事司法系统效 831 世界卫生组织烟草税政策和管理技术手册 率低下、非法和非正规市场的存在以及邻国之间的冲突[5]都促成了非法烟草贸易的存在 和增长。 很难将价格的作用与其他每个因素分开,因为①获取非法贸易的价格和数量指标本 身就具有挑战性;②在大多数国家,卷烟品牌众多,各品牌价格各不相同;③缺乏应对影响 非法贸易的非价格因素的良好措施,例如政府腐败和刑事司法系统不力。这些限制使得 获取价格和其他因素如何影响非法贸易严格的经验证据具有挑战性。尽管存在这些根本 性挑战,但经济文献已经提供了可靠的证据,表明价格始终只是决定非法贸易程度的一个 因素———而且往往不是最重要的因素。 许多关于价格和其他因素影响的计量经济学研究都集中在跨境购物(或从低税收到 高税收管辖区的小规模走私)。因为在美国和欧洲 可以获得低税收和高税收司法管辖 区的销售数据,并按地理区域以方便的方式分类 (即靠近或远离边界)。在控制了影响跨 境销售的其他重要因素后,这些研究试图将低税收管辖区的非法贸易流动或相对较高的 销售额解释为低税收管辖区与周围高税收管辖区之间的价格和税收差异的函数。例如靠 近边界和腐败程度[6,7,11,28] 。研究的主要结论是,非法贸易流动不仅仅与价格有关[29]。 一些研究显示出价格差异与其他因素的显著影响,但其他研究没有发现显著的价格差异 效应。这些分析的重要政策含义是,在高税收管辖区降低烟草税率和实际价格可能对非 法市场份额的影响很小或没有影响 。 观察和案例研究提供的信息可以改进公共政策,即使它们无法提供令人信服的因果 关系证据。一些观察性研究使用大量国家样本将价格水平与非法市场份额相关联。 Joossens 等人[25]发现,与税收份额和价格较低的国家相比,税收和价格较高的国家通常 具有较低的非法贸易份额。在他们的样本中,高收入国家通常具有相对较高的卷烟价格 和税收份额,但它们的有利结果(即较低的非法贸易水平)与有效的税收管理和较低的腐 败水平有关。相比之下,中低收入国家的价格和税收份额普遍较低,非法市场份额也很 大。Joossens 等人将打击非法贸易的困难归咎于税收和海关管理薄弱,以及在大多数情 况下面临制度和法律方面的挑战[25]。 图4.1说明了价格和非法贸易之间的关系,使用2018年94个国家的最畅销品牌卷 烟的单包价格(以美元计)和非法贸易估计水平 。这两个变量之间没有明显的独特关 联。以零售价格为解释变量,以非法交易份额为因变量进行线性回归,显示价格与非法市 931第四章 烟草税收的政治经济学 这是美国关于各州间贸易的传统文献链,在欧洲国家的程度较小,其中大多数国家使用传统但不准确的非法贸易 测量。 最近,PMI-Altria资助了一些影响跨境销售的因素的研究。其中一项研究,Prieger 和 Kulic[28]批评了 Merriman 等人[9],得出的结论是,在跨境购物中,价格差异对于确定非法贸易的规模很重要。 巴西在21世纪初降低了税率和实际价格,以打击来自低税收司法管辖区的非法贸易。然而,根据业内人士[20]的 说法,在这一行动之后,政府损失了收入,非法贸易的规模和范围仍然没有改变。 各国对非法贸易规模的估计存在争议。烟草业的数字高估了这个问题,并且基于有问题的方法。无法获得同一 年大量国家样本的严格和透明的估计数。为了将价格水平与非法市场份额进行比较,选择了 Euromonitor 对非 法市场份额的估计,原因有二:它们是给定年份中大量国家样本的可比估计值,并且没有人会争辩说它们偏向于 烟草控制的观点。使用 Euromonitor 数据并不意味着世界卫生组织完全同意所有细节和用于获取这些数据的方 法。 场份额之间的关系呈反比,但并不显著。 图4.1说明了一些特殊情况。 （以美元计） 50 45 40 35 30 25 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 SGP 16 NZL BLR UKR 零售价/（美元/每包） UZB BGD TUN MKD EGY KHM LAO CIV BOL TUR SRB DZAIDN TZA RUS JORTHA CHN KAZ SVK KWT HRV CZE KOR JPN LKA DNK USA GEO CHE SAU ITA NLD FIN DEU BEL PER PRT SVN HUN ESP SWEBGR MAR ARGKEN PHLAZE ESTMEX BIHMMR COL CHL POL LTU ROM LBNAGO VNM NGA IRQ CMR HND GTM SLV IND LVA ZAF ARE GRCURY CRI GHA ETH DOM PAN 非 法 贸 易 /% AUT FRA ISR CAN GBR AUS NOR IRL PAK ECU BRA MYS OMN 图4.1 2018年按国家/地区分列的非法贸易份额与 销量最高的卷烟品牌零售价的比例(以美元计) 注:卷烟非法贸易的程度由欧洲监测机构衡量为一个国家非法卷烟的估计数量除以该国家估计的卷烟总消费量。 来源:参考文献27和30。 041 世界卫生组织烟草税政策和管理技术手册 必须考虑其他因素,以将这种观察性分析转化为严格的因果分析。 • 许多价格低廉(即每包低于 2 美元)的国家在样本中的非法贸易水平最高,例如巴 西(BRA)(1.33 美元和 46.3% 的非法份额)、巴基斯坦(PAK)(0.39 美元和 40%)、埃塞 俄比亚(ETH)(0.55 美元和 32.9%)、加纳(GHA)(1.06 美元和 29%)和喀麦隆(CMR) (0.89 美元和 25%)。 • 相比之下,许多价格在 4 美元到 8 美元的国家———这可以被认为是足够高的经济 激励措施———其非法贸易份额不到总消费量的 10%。这些国家包括韩国(KOR)(4.02 美元和 0.8%)、捷克(CZE)(4.31 美元和 2.9%)和斯里兰卡(LKA)(6.89 美元和 1.6%)。 • 除爱尔兰外,所有价格非常高(高于 8 美元)的国家的非法贸易份额均低于 20%。 这些国家包括法国(FRA)(9.39 美元和 17.8%)、瑞士(CHE)(8.71 美元和 5.5%)、新加 坡(SGP)(10.35 美元和 3.7%)和挪威(NOR)(14.51 美元和 9.6%)。 正如对烟草制品非法贸易的众多研究和分析所表明的[4,5],解决问题的最有效方法 不是放弃增税,而是加强打击非法贸易的能力。因此,重要的是要考虑善治与非法贸易之 间的关系。一个国家总体上打击非法贸易的能力越强,非法贸易的水平就越低。 由经济学人智库(EIU) 编制的全球非法贸易环境指数衡量各国整体打击非法贸易 的结构能力。EIU 指标是4个指标或类别的组合,旨在评估各国在这些领域的表现;总 体指标越接近 100,该国打击非法贸易的能力就越好。这四类 是: 1.政府政策:衡量政府积极监测和防止非法贸易的承诺; 2.供求关系:衡量国内环境阻碍或鼓励对非法商品的供求关系的程度; 3.透明度和贸易:衡量适用于自由贸易区和运输的透明度和治理程度; 4.海关环境:衡量海关服务如何有效地促进合法贸易,同时防止非法贸易。 图4.2说明了EIU指标与一组国家卷烟非法贸易的估计水平之间的关系。该指标 与卷烟非法贸易的估计水平之间存在反比和统计上显著的关系。这表明,随着打击非法 贸易的能力普遍增加,卷烟非法贸易就会下降 。 严格的独立研究表明,尽管存在非法贸易的挑战,但对烟草制品征税是一种有效的公 共卫生干预措施,可显著减少烟草使用并产生政府收入[5]。此外,当卷烟税增加时,政府 会产生更高的收入并使卷烟的消费量减少[32,33]。然而,无效的税收管理会导致非法贸易 增长,并可能通过提供更便宜的卷烟来破坏烟草税的一些好处。例如,马来西亚走私卷烟 的平均街头价格比其法定纳税等价物低 55%[34]。非法烟草贸易还会减少政府税收,并 可能增加与吸烟相关的健康成本和与警务相关的成本。 141第四章 烟草税收的政治经济学 有关此指标的更多详细信息,请访问http://ilicittradeindex.eiu.com/。 使用两个变量之间的线性回归,这种关联在 90% 的置信水平上显著不等于零。 非 法 贸 易 香 烟 /% 60 55 50 45 40 35 30 25 20 15 10 5 0 10 20 30 40 50 60 70 80 90 EIU 指标 NZL USA DNKSGP CZE SVKCHN THA HRV ITA SVN NLD FIN DEUBELPRT HUNTUR SAU DZA RUS IDN MAR SRB ARG ESP AUS GBR SWECHL POLROM ISRTW EST AUT FRA LTU COL MEX BGR VNM BIH ZAF CAN LVA IRLGRC ARE URY IND GTM CRI KOR JPN KAZ TUN BLR UKR KHM LAO PHL PER MMR IRQ PANDOM PAK ECU BRA MYS 图4.2 2018 年 70 个国家/地区的非法贸易份额与 EIU 指标 来源:参考文献30和31。 4.1.4　测量烟草制品中的非法贸易 非法贸易的规模是税收政策讨论中的一个有力论据,因此烟草业资助对其特别感兴 趣的国家或地区的非法贸易估计(即欧盟的 Sun 项目和 Star 项目以及东亚的牛津经济 241 世界卫生组织烟草税政策和管理技术手册 研究院)。然而,最近对非法贸易行业数据的系统审查发现,行业委托报告存在大量方法 论缺陷[24]。此外,Blecher 等人[35]认为行业资助的研究倾向于系统地高估非法贸易的规 模,以说服当局放弃烟草税改革。独立研究人员还发现了烟草业资助估计值的不一 致[36]。 表 4.3 给出了一些夸大与行业相关的非法贸易估计值的例子,该表将同行评审和独 立研究与烟草业资助的估计值进行了比较。由于某些国家/地区有来自不同来源或年份 的多个行业估计值,因此表 4.3 列出了发表独立研究的文章中包含的估计值,因为它被 认为具有代表性且足以说明高估。在所有情况下,行业估计都超过了独立研究的估计。 衡量非法贸易的规模对政府来说可能是一项艰巨的任务,因为独立研究人员、政府和 烟草业采用了不同的方法。尽管如此,这些研究还是值得的,因为它们会推动政策讨论, 并可用于评估政策的影响(例如增税、平装和健康警告)。 表4.3 独立研究中估计的非法市场份额与烟草业资助的研究中估计的非法市场份额相比 国家 非法市场规模和估 计 年 份———独 立 研究 独立研究的来源 非法市场的规模和 估计 年 份———行 业资助的研究 负责行业资助研究的 机构 哥伦比亚 2016年占5个城市 总市场的3.5% Maldonado et al., 2018[37] 2014年占总市场份 额的13% FND and INVAMER, 2015 智利 占大都市总市场的 16.3% 圣地亚哥地区,2017 Parajeetal., 2020[38] 2017年占总市场份 额的24.3% ObservatoriodelComercio IlícitoBATC,2017 巴西 2014年占总市场份 额的28.8% Iglesiasetal, 2017[39] 2014年占总市场的 34% BAT publicstatement, 2015 墨西哥 2017年,占8个主要 城市总市场份额的 8.8% Saenz de MieraJuarez et al.,2020[40] 2012年占总市场份 额的16.6% Confederaciónde CámarasIndustriales, 2012 如表 4.4 所示,衡量非法贸易的方法可分为三类:①直接衡量;②残差法;③专家意 见[12]。直接测量依赖于与实际违法行为和包装观察直接相关的证据;剩余方法基于有关 消费和合法销售的理论和证据推断逃避;专家意见提炼从与最直接了解烟草市场的个人 交谈中获得的信息。每种方法都有优点和缺点。没有任何一种方法能明显优于其他方 法,但直接测量法和残差法更有利于确定非法市场的规模,而专家意见可以提供对市场运 作细节的洞察 。 表 4.4 列出了实施每种主要测量方法所需的资源相对数量和专业程度,以及主要目 的、数据收集特征、抽样特征和分析单位。附件 4.1 中提供了每种方法的简要说明。 Merriman[11]和 Ross[9]提供了更广泛的细节。 341第四章 烟草税收的政治经济学 在与来自烟草行业的专家进行访谈时,需要遵守FCTC第5.3条及其准则的规定。 表4.4 所需的资源和专业知识的概述和测量方法的主要目的 方法 资源 水平 专业 测量方法的主要目的 数据整理 样本 分析 单位 设置(D) $ 低 确定产品类型、运输方 法、入口点和品牌名称的 趋势 二 次 数 据 使用 非概率 船运 使用现有的健 康 监 测 调 查———自 我 报 告的消费量(D) $ 低,只有额 外的问题 非法贸易的规模,增加或 改进关于上次购买的品 牌、价值和数量的问题 其 他 主 要 数据集合 概率 个人 差距分析(R) $ 中级 提供一种衡量非法贸易 变化的措施 二 次 数 据 使用 普遍 国家 计量经济学模 型(R) $ 高级 估计从纳税到非法产品 的替代价格弹性 二 次 数 据 使用 普遍 地理 专家访谈(E) $ 低级 非法贸易的特点 主 要 数 据 收集 非概率 个人 吸烟者拦截和 包装观测调查 (D) $$$ 中级 非法贸易的规模和特征, 以概率为基础的样本来 代表人口 主 要 数 据 收集 概率 个人 包装退货和换 货调查(D) $$$ 中级 非法贸易的规模和特征, 以概率为基础的样本来 代表人口 主 要 数 据 收集 概率 个人 小包装调查(D) $$$ 中级 非法贸易的规模和特点, 与使用空包调查的行业 估计的比较 主 要 数 据 收集 概率 个人 秘密购买(D) $$$ 中级 非法贸易的产品类型和 贸易渠道 主 要 数 据 收集 概率 地理区 域 自我报告消费 者调查(D) $$$ 高级 非法贸易的规模和特征, 以概率为基础的样本来 代表人口 主 要 数 据 收集 概率 个人 注:普遍包括总人口;D = 直接测量,R = 残差法,E = 专家意见。假设一个中等规模的研究(例如一个数百万 地区的代表性研究)的资源成本规模:$(最便宜)———熟练工时的周数;$$(中等昂贵)———1 到 2 个月的熟练工时; $$$(最贵)———6 到 12 个月的熟练和非熟练工时。 没有简单的选择规则来决定使用哪种测量方法。选择一种或多种方法时需要考虑的 主要因素包括:①非法贸易问题的性质和特征(问题在哪里以及如何表现,是国内逃税还 是外国品牌的非法流入或两者兼而有之);②以前收集的数据;③可用的预算;④可用分析 441 世界卫生组织烟草税政策和管理技术手册 师的专业知识。 可用的预算和员工技能往往是政府面临的主要限制。因此,表 4.4 根据所需资源和 可用专业知识对测量方法进行排序。例如,残差方法和专家意见可以以低成本提供粗略 但有用的估计,并且需要最低水平的技术。对于采用人口健康监测调查的国家来说,另一 个低成本选择是添加衡量非法贸易的问题,例如品牌名称、价值和上次购买的数量。相比 之下,直接测量方法通常需要复杂的研究设计和昂贵(且耗时)的实地研究。 附件 4.1 中给出了对不同测量方法的更详细说明。为进一步协助主管机关决定选 择哪种方法,表 4.5 列出了每种测量方法的主要特征,以及每种方法的主要优点和缺点。 各国可以从需要较少资源和较少技能的方法开始,以获得对问题的概览。缉获(执法工作 的副产品)提供了第一步,各国可以分析获得的信息(产品来源、品牌、位置等)并报告结 果,以提高公众对该问题的认识 。或者,各国可以将与非法贸易有关的问题添加到由卫 生监督机构和统计机构定期进行的现有和资助的卫生监督调查中。这样,卫生部门(烟草 控制办公室和卫生监督单位)、税务和海关部门以及国家统计局之间在使用现有测量方法 方面的合作可以成为确定非法贸易问题的性质和规模的起点。 表4.5 非法贸易计量方法的主要特点和优缺点 方法 主要特征 主要优势 主要劣势 缉获量(D) 被地方和国家当局没 收的烟草制品的统计 数据 随 时 可 从 执 法 机 构 获得 可能不能提供关于非法贸 易的规模和/或性质的代 表性图片 利用现有的健康监 测调查来获得自我 报告的消费量(D) 添加或改进有关品牌 名称、数量、价格、购买 地点等因素的问题 在国家层面产生良好 的估计 由于消费非法烟草制品的 社会违法性,自我报告的 数据可能会有偏见 差距分析(R) 比较自我报告的消费 数据与观察到的(通常 是管理性的)纳税销售 数据 当质量数据可用时,是 简单和容易复制(提供 随时间推移的测量)且 可解释 关于纳税销售和/或消费 的数据往往不准确,在许 多情况下并不提供关于非 法市场规模的信息,而只 能提供随时间变化的信息 计量经济学模型(R) 根据模型给出的纳税 销售与预测消费之间 的差额进行估算 因为它符合经济理论 的悠久传统,可以评估 经验估计 需要一段时间内各种重要 变量的高质量数据和先进 的计量经济学建模专业 知识 541第四章 烟草税收的政治经济学 缉获对于获得有关非法活动的定性信息很有用,但在预测问题的规模时必须非常谨慎地对待它们。各国可能认 为他们有很大的问题,因为他们有主管机关在寻找非法商品方面做得非常出色。另一方面,国家的主管机关效率 较低,缉获量很少,而在这种环境中,缉获量并不能说明问题的大小和性质。 续表 方法 主要特征 主要优势 主要劣势 专家访谈(E) 专家包括研究人员(如 经济学、刑事司法和公 共卫生领域)、记者、税 务和执法专家、产品制 造商和批发商 有助于确定市场的性 质和趋势(例如,出售 非法卷烟的场所、进入 方式),访谈可用于定 义评估非法贸易规模 的方法 获得的信息可能无法概 括,专家知识可能过时或 受专家经验限制;此外,专 家往往有很强的偏见 吸烟者/零售商拦截 和包装观察调查(D) 检查吸烟者和卷烟零 售商的包装,便利性或 基于概率的样本 是直接和客观的,吸烟 者不受任何价值判断 难以确定代表烟草使用人 群的地区并对重要的亚人 群(如老年人和行动不便 的吸烟者)进行抽样,但住 户调查可以克服抽样问题 包装退货和换货调 查(D) 还有一个包装观察研 究,使用调查抽样技术 来检查吸烟者的特征 可能会减少与传统吸 烟调查相关的耻辱感 在中低收入国家,由于邮 件传递系统问题,调查分 发可能不可靠 小包装调查(D) 也称为空废弃包装调 查;公开丢弃的包装具 有表明它们是否符合 税收规定的特征(例如 印花 税、公 共 卫 生 警 告) 收益率估计不太可能 因社会期望、回忆错误 和保密性等问题而产 生偏差的估计值 聘请实地研究人员来收 集、编码和分析数据可能 需要大量预算;调查不提 供有关吸烟者和支付价格 的信息 秘密购买(D) 买包装和单支卷烟来 衡 量 非 法 卷 烟 的 可 用性 直接确定非法卷烟的 来源 很难为非法来源的零售商 创建一个抽样框架,或者 很难知道吸烟者实际购买 了什么以及购买了多少 自我报告消费者调 查(D) 调查可以使用各种分 配方式分配给个人或 家庭 在 国 家 一 级 的 良 好 估计 由于消费非法烟草制品的 社会违法性,自我报告的 数据可能会有偏差 注:D=直接测量,R=残差法,E=专家意见。 许多中低收入国家越来越多地通过不同类型的调查(例如拦截吸烟者或零售商、退回 包装、乱扔包装检查和秘密购买卷烟)对包装进行直接观察。这些活动扩展了独立研究人 员和学术界的技能,并增加了对这些方法的了解。此外,越来越便宜的数字技术允许采访 者在直接观察调查或更大的国家自我报告消费者调查中拍照和记录包装特征。 最终,在能力允许的情况下,需要在给定时间点使用多种方法进行更可靠的评估。理 641 世界卫生组织烟草税政策和管理技术手册 想情况下,应定期进行评估,以弄清非法贸易随时间的演变及其与政策变化的可能联系。 4.1.5　CRI衡量非法市场规模的批判性研究 衡量非法贸易对研究人员、行业和政府来说都是一项挑战,因为根据定义,这种贸易 是隐藏在视线之外的。买家和经销商有时会竭尽全力确保隐瞒他们参与的违法活动。多 年来,研究人员和政府机构对估计非法市场的规模和确定有效的干预措施越来越感兴趣。 作为研究的消费者,政府应该严格审查现有的研究并评估它们的科学严谨性和方法透明 度。与非法烟草贸易有关的衡量问题尤其重要,因为这是一个政治化的话题。过高估计 可能会引发有关烟草业控制供应链的能力、其参与非法转移、税收政策的影响和执法策略 的有效性的问题。虽然该行业将自己描绘成在衡量和打击非法贸易方面采取积极立场 (例如,由毕马威有限责任公司实施但由 PMI 支付费用,后来由 Sun 项目进入封闭市 场),但在过去,它曾将打击走私作为一种策略来打击非法贸易。例如,在中国和俄罗 斯[19,41]。政府仔细审查有关工业或准工业来源产生的非法烟草贸易的证据,并建议寻找 替代证据。准行业报告是由行业委托但由私人研究公司(例如 Ernst and Young、Oxford Economics)发布的研究[11]。 良好分析的特点 良好分析的主要特点之一是科学严谨,包括使用相关理论框架、健全的统计方法和对 结果稳健性的检查(例如敏感性分析)。高质量的研究报告对研究方法和所采取的统计分 析步骤,以及为确定研究结果稳健性的补充分析提供了透明的解释。例如,Joossens等 人[25]清楚地描述所使用的数据来源(限制和优势以及可以找到它们的地方)以及对如果 消除非法卷烟的全球市场份额将挽救的生命数量进行的计算。解释应该足够详细,以便 后来的研究人员能够仔细检查分析并复制结果。 可复制性是优秀科学的另一个标志。例如,垃圾包装的研究应该详细说明数据收集 的地点和时间、每个地理单位收集了多少包装、识别非法包装的协议(例如警告标签、品 牌、印花税等特征)和统计分析的细节。还应说明所选地理区域的代表性。未能提供这种 深度的信息可能会被质疑研究的普遍性以及所选方法是否存在缺陷。 在有关非法贸易的政策决定的背景下,最有用的数据提供了关于具有代表性的个人 和地理样本的信息。例如,仅限于基于特定特征或特定地理区域的一组个体的研究可能 会产生有偏见的信息。研究报告还应该清楚地说明研究的局限性。例如,衡量非法贸易 的研究通常不衡量产品假冒,也不将非卷烟烟草制品纳入其评估值[42]。当将其纳入统计 估计时,应提供置信区间和点估计,以说明由简单随机机会引起的不确定性[11]。 有缺陷的分析的特征 有缺陷的分析可能会混淆和扭曲有关非法贸易的科学知识。缺陷通常表现在数据、 方法、统计分析和结果解释中[11]。研究可能故意设计的、具有方法上的缺陷,以产生对贸 易的高或低评估。例如,研究表明非法贸易占整个市场的很大一部分,可能被用来支持税 收导致非法贸易急剧增加的论点,而较低的估计可能被用来支持某些政府干预(例如增加 零售检查)的论点是有效的。扭曲数据的压力也可能与经济有关。例如,高估可以促使政 府为执法活动提供更多资源。通过使用有偏估计的数据源或数据收集方法,可以有目的 741第四章 烟草税收的政治经济学 地歪曲分析。 有缺陷的研究有时对其方法的描述不完整或不准确。他们可能缺乏有关数据质量的 详细信息或有关如何收集和分析数据的信息。例如,行业资助的废弃包装研究的一个共 同弱点是没有报告收集和医学分析的方法,表面上是因为它们是“专有”信息[43]。然而, 这些方法细节是评估研究结果是否因抽样误差、模型错误指定、测量误差、无响应或其他 缺陷而产生偏差的关键。 如果研究人员未能披露调查工具中包含的问题,则可能无法评估测量误差,导致用于 衡量非法烟草贸易的调查项目可能不准确。例如,询问受访者购买“廉价”卷烟的频率可 能会产生有偏差的估计,特别是如果消费者可以使用优惠券以折扣价购买卷烟。为了更 准确地衡量逃税情况,调查必须包括有关上次购买地点、购买价格、公共卫生警告标签和 品牌名称等问题。 有缺陷的研究还可能通过其他方式无意或故意扭曲对非法贸易的估计。例如,数据 收集者可以故意对已知为非法销售热点的区域或低税收管辖区居民访问的站点进行过度 抽样。研究人员可以在税收较低的国家/地区附近收集丢弃的烟盒,以(有意或无意地)证 明税收政策的副作用。在未经同行评审或轻度同行评审的媒体上发表的研究,例如编辑 过的书卷或政策简报,应该比在备受推崇的同行评审的媒体上发表的研究更加可疑。 4.1.6　总结 在全球范围内,非法烟草贸易仍然是税务管理者关注的主要问题,因为它对收取更高 的收入以及准确和独立衡量带来了挑战。行业数据就问题的严重程度提供了扭曲的结 论:通常对非法贸易和烟草税之间的联系做出单一因果解释。非法贸易包括多系统问题, 需要多种策略应对。在世界范围内,不同经济发展水平的国家已经实施了各种有效措施 来打击烟草制品的非法贸易。例如,菲律宾和英国已将非法贸易作为其整体烟草税改革 的一部分[4]。 价格(和税收)水平不是非法贸易的关键决定因素,缺乏税收管理能力加剧了非法贸 易的存在。不加税不是解决办法;相反,各国应采取至少包括以下3个主要组成部分的综 合战略来应对: 1.应该独立于行业识别问题的性质和维度。有必要以最佳统计方法科学地评估非法 贸易的规模,以了解问题的特征和范围。 2.应确定并实施旨在解决该国正在经历的特定类型非法贸易的适当政策和战略。应 该直接解决特定国家的体制和治理挑战———以及缺乏会加剧非法贸易的多边协调———并 改进第 3 章所述的税收和海关管理实践。 3.如果某国还不是缔约方,则应实施《世卫组织框架公约消除烟草制品非法贸易议定 书》中包含的最佳做法并加入该议定书。 有解决非法烟草贸易的适当方法和政策。如果各国开始实施适当的政策,即使存在 非法贸易,它们也可以提高烟草税并获得健康和税收收益。 841 世界卫生组织烟草税政策和管理技术手册 4.2　威胁策略C：法庭和法律挑战 4.2.1　介绍 烟草业认为,精心设计和大幅增税是对其业务利润、增长和长期可持续性的威胁。正 如1985年PMI所指出的: 在所有担忧中,税收最让我们警觉。虽然(其他限制)……确实会降低交易量,但根据 我们的经验,税收会更严重地降低交易量[44]。 然而,与其他烟草控制措施相比,该行业不太可能对消费税发起直接的法律挑战(详 见方框 4.1),因为税收(尤其是消费税)是一种相对完善的监管措施;在许多司法管辖区, 对烟草制品征税已有一个多世纪。烟草业参与者反对税收的共识也较少,因为不同烟草 公司的市场地位差异会影响他们在税收政策中的利益。这反过来又降低了他们就该问题 采取集体行动的可能性[45]。BAT在20世纪90年代初期的既定战略是在烟草税的水平 和结构方面影响政府,以促进市场增长并确保竞争优势[46]。 尽管如此,当重大税收措施的设计、采用或实施中的漏洞显而易见时,烟草业参与者 仍将在法律上挑战或至少在法律上威胁这些措施。 方框4.1 对烟草税措施的法庭和法律挑战 有证据表明,烟草业及其盟友对税收措施采取的法律行动少于对其他烟草控制 措施: 1.无烟儿童运动的烟草控制法数据库只包含少数有关烟草税措施的案例,但有 数百个涉及其他烟草控制主题的案例。这种模式也可以在2018年对烟草控制法律 挑战的审查中看到,该审查审查了这个数据库和另外两个数据库,从而选择了与世卫 组织FCTC在诉讼中的有效性问题相关的96个案例[47]。这96起案件中只有6起 是与税收措施有关的。 2.2013年,一项关于烟草业干预烟草税政策的实证研究的系统回顾发现,36篇 相关文章中只有9篇报道了将诉讼作为烟草业策略的具体用途[1]。所有9个都涉及 对美国烟草税倡议指定条款的宪法挑战。 3.2015年一项关于中低收入国家行业干预的研究指出,15个国家对烟草控制措 施的法律挑战是行业干预的例子,但这些挑战与烟草税措施无关[48]。 4.2016年,一项对行业税收和营销措施的系统审查中发现,65篇关于烟草税的 论文中,只有5篇与使用诉讼或诉讼威胁干预烟草税措施有关[49]。 烟草业广泛使用法律专家[1,50,52],他们仔细研究所有相关法律和法规,以确定其可能 的和有争议的界限,以便操纵法规和监管机构[1,50,52]。根据这些专家建议,烟草公司知道 法规何时仍处于国际和国内义务的范围内,但仍然可以争辩,如果通过法律允许的烟草控 941第四章 烟草税收的政治经济学 制措施将在诉讼中失败[48,51,52]。由于仅靠法律挑战的威胁就可以为行业带来优势,因此 很少需要或不希望诉诸诉讼[1,45,48,51,5356]。即使提起诉讼,其目的也可能是推迟或削弱某 项措施,而不是根据案情取胜[1,45,48,53]。为了应对实际和可能面临的法律挑战,政策制定 者在准备和实施烟草控制措施时需要了解相关的法律义务。幸运的是,烟草行业的剧本 是相对可预测的。因此,可以设计税收和其他烟草控制措施来加强监管机构的法律地位 以应对真正威胁,并使他们能够消除毫无根据的行业威胁。 4.2.2　烟草税面临法律挑战的国家经验 与烟草税收相关的法律义务包括国内法和国际文书(如国际贸易协定和国际投资协 定(IIAs))规定的义务。 税收措施可能遇到的一些法律问题列于表4.6。来自不同国家 的案例研究说明了这些法律问题在烟草税的通过、设计和实施中是如何避免的。然而,这 些问题不是常态,不应引起过度担忧。根据法律决定支持相关税收措施的程度,案例研究 被评为正面、混合或负面。 表4.6 烟草税措施的潜在法律问题 脆 弱 点 法 律 义 务 案例研究 协商不足和其他程序漏洞 国内程序法 1,2 国际投资协定下对投资者的正当程序保护 None WTO协议和海关工会下的程序要求 3 对进口商品或投资者的歧视 世贸贸易协定和海关工会下的非歧视义务 8,9,10 国际投资协定下的非歧视义务 11 投资激励或诱因 投资者 国家合同下的仲裁机制 12 国际投资协定的公平和公正待遇条款 None 其他实质性违约 宪法权利和税收限制 4 对征税的法定限制 6 IIAs的征收条款 5 超权限(法律权限范围) 7 避免税法中的程序漏洞 通过在推进和实施监管或立法规定时非常小心,可以避免程序上的缺陷。程序问题 使烟草控制监管机构陷入困境。世界卫生组织FCTC 第 5.3 条和 COP 的实施指南指 出,政策制定者和监管者应仅在绝对必要的情况下并在绝对必要的范围内与烟草业互 动[57]。对于税收措施,互动可能是必要的,因为协商和协商程序可以在国内宪法条款和 051 世界卫生组织烟草税政策和管理技术手册 相关国际贸易协定包括世贸组织协定和欧盟、东非关税同盟及南方共同市场等关税同盟。相关的国际投资协定 包括双边投资条约以及自由贸易协定和关税同盟内的投资章节。 善治程序、国际投资协定的正当程序要求和一些国际贸易协定中规定。烟草业可以利用 这些要求作为杠杆来延迟、扭曲或控制规则制定过程,但这违反FCTC第 5.3 条。因此, 与烟草业的互动应仅限于以透明或公开的方式进行的严格必要的磋商,但要注意这不能 以牺牲措施的可辩护性为代价。适当的平衡将取决于相关管辖权,因为宪法、成文法和适 用的国际法律义务各不相同。 案例研究1(混合):对立法程序的行业操纵 2012年,哥斯达黎加立法议会通过了一项法案,规定征收新的卷烟消费税。然而,该 法案的通过是在“紧急”的情况下进行的,尽管有待于进行宪法调查(宪法调查旨在阻止一 项法案的通过) 。 事 件 多数决定法 少数人决定 教 训 法院是否可以 在法案通过后 考虑调查,以及 法案的通过可 能产生的影响 最高法院宪法庭的 多数票[58]接受了这 项调查。签署和公 布行政部门的法案 被宪法部门暂停,等 待他们对案件的是 非 曲 直 做 出 决 定———最终,该法案 的任何合宪性问题 都是毫无根据的[58] 由于法案提交太晚 并且立法机关在阅 读法案之前没有收 到提交通知,因此调 查不可受理[58]。与 多数意见不同,少数 人认为法院不能考 虑调查或暂停行政 部门签署法案——— 成为法律的最后一 步 这一挑战表明烟草业的捍卫者可能 会试图挫败和阻碍税收措施的通 过。在这种情况下,挑战似乎是一 种拖延策略,因为它是在法案最终 阅读的同一天发布的。其作者可能 希望其悬而未决的性质导致立法机 关推迟,或者在没有延迟的情况下 为程序和宪法挑战创造条件。挑战 本身的所有理由都被发现毫无根 据。尽管无法避免此类琐碎的挑 战,但可以而且应该预期它们不会 导致税收措施的失败 案例研究2(积极):遵守国内的程序要求 肯尼亚的烟草控制法规要求烟草业缴纳税款,以补偿国家的医疗保健和其他吸烟的 负面外部效应。在 2016 年针对这些规定提出的挑战中,原告 BAT 在各项指控上均未成 功[59,60]。尽管法院并未将征税视为一项税收措施,但该案例研究对监管机构如何保护税 收措施免受程序挑战具有指导意义。 151第四章 烟草税收的政治经济学 “紧急”是通过立法机关快速推进法案的程序。 事 件 法律和论据 决 定 教 训 政府就该措施进行 的磋商是否充分 断言宪法和法定文书 法一起意味着需要“与 可能受影响的人进行 适当的协商”,因为该 措施可能对企业产生 重大影响。BAT 声称 未达到此标准。肯尼 亚政府声称它没有义 务与烟草业进行特别 或广泛的磋商 法官裁定肯尼亚政府 胜诉,并指出:(1) 协商 的要求并不意味着任 何特定观点都需要占 上风;(2)对协商程度 的不满不是决定性的; (3) 事实上,该行业被 允许并经常被邀请派 代表参加所有相关的 公众协商会议和议会 委员会听证会;(4) 对 法规的咨询充分(59) 烟草业仔细审查立法和监 管程序是否存在缺陷。在 这种情况下,肯尼亚政府 官员适当地与烟草业保持 距离,不允许其代表进行 特殊考虑,但允许他们参 加公开会议并按照常规程 序提交意见。通过这种方 式,世界卫生组织 FCTC 第 5.3 条背后的原则和 肯尼亚法律下的协商要求 都得到了遵守 避免税务征管中的程序性问题 案例研究3(消极):违反国际义务中的程序要求 2010 年,世贸组织的一个专家组裁定,泰国违反了《海关估价协议》(CVA),因为该程 序被用于对 Philip Morris(PM) Thailand 从关联方菲律宾进口到该国的卷烟进行估价。 海关价值很重要,因为它们是关税的计税基础,并且可以作为根据商品价值征收的其他税 费的基础,例如从价消费税和增值税。PM 泰国申报的交易价值因受双方关系的影响而 被泰国税务机关拒绝,并以扣除额确定的海关价值取代[61]。 事 件 法律义务 (如合同契约等) 决 定 教 训 泰国在拒绝其宣布 的交易价值[61]之前, 是否与菲律宾进行 了充分的协商 CVA需要真诚的交换 理由和信息,并有机会 回应[61] 泰国未能正确解释其 拒绝交易价值的原因, 以及其认为价格受双 方关系影响的观点[61]。 这违反了 CVA。泰国 没有对这些调查结果 提出上诉 泰国当局在与烟草业打交 道时需要更加谨慎,以确 保他们履行相关的程序义 务。在这种情况下,CVA 规定了具体和高标准的咨 询(提供详细的理由和回 应的机会)而泰国未能达 到 确保符合实质性要求 国内法和国际法中的规则也规定了实质性义务。本小节侧重于国内法中的实质性义 务。下一小节将讨论有关歧视和投资激励措施的国际义务。 251 世界卫生组织烟草税政策和管理技术手册 案例研究4(积极):发现税收措施符合宪法 智利政府大幅增加了烟草和燃油消费税,1995年,一个纳税人联盟对该措施提出了 宪法挑战[62]。 事 件 决 定 教 训 根据智利宪法第19条,该税 收是否“明显不成比例或不公 正”[62] 增加消费税并没有违反宪法, 因为它既不是征用,也不是明 显的非理性 一般适用的消费税不容易因过度、 不公平或不成比例而受到挑战 案例研究5(积极):未能给予退税不是 IIA 下的征用 本案例研究是对违反投资条约征收条款的索赔示例。此类条款保护外国投资者免受 可被解释为直接或间接扣押投资或剥夺其价值的措施[63]。在本案例中,一名投资者被墨 西哥政府拒绝退税十多年。这影响了在国外购买和转售墨西哥卷烟业务的盈利能力,投 资者于 2002 年向投资协议仲裁庭提起诉讼[64]。 事 件 决 定 教 训 墨西哥未能向投资者证明回 扣是否超出了有效监管的范 围,构成了对投资者投资的间 接征收[65] 没有征用。仲裁庭指出,并非 所有商业问题都是违规行为: 投资者无权参与卷烟“灰色市 场”出口,并且有充分的理由 限制该市场[65]。此外,投资者 能够参与其他商业活动并实 际上继续取得商业成功[65] 根据国际投资协定提出的间接征用 索赔不太可能成功,因为普遍适用 的税收措施是一种合法的监管形 式。仅仅损失利润是不够的。除非 对投资结果造成实质性或显著剥 夺,否则征用索赔不会成功 案例研究6(消极):违反上级国内立法的规定 2011 年,印度尼西亚烟草行业协会组织 FORMASI 对新的消费税法规提出质疑。 自2009 年以来,政府一直在实施基于一系列特征(生产规模、卷烟类型和价格水平)的分 层从量消费税制度。2011 年,几乎所有 19 个等级的消费税都提高了,但参考价格没有相 应调整,这就产生了一个法律问题。(表4.12) 351第四章 烟草税收的政治经济学 事 件 决 定 教 训 新的消费税法规是否违反了 上级消费税法规定的 57% 的 烟草 制 品 零 售 价 消 费 税 上 限[6670]。挑战规定,国产手卷 丁香卷烟(丁香卷烟)的消费 税超过了这一上限[68,71] 法院裁定烟草业协会胜诉,并 要求政府立即撤销 2011 年的 规定[6970] 建议遵守规则并了解法律等级 (包 括上级国内立法)。烟草业仔细审 查所有增加的烟草税。在这种情况 下,违反单一类别烟草制品的立法 要求导致印度尼西亚收入损失,并 在减少烟草消费的努力中受挫 确保税收措施在当局的法律权力范围内 当一项税收措施超出制定机构的法律权力时,它就是越权。与案例研究 6 一样,这 是一个涉及法律等级的法律问题。然而,在越权案件中,问题不是集中在下级法律和上级 法律之间的冲突,而是制定税收措施的主管机关是否有权这样做。当税收措施由地方司 法管辖区或根据法定授权行事的行政部门颁布时,可能会出现此类问题。 案例研究7(消极):违反澳大利亚宪法的烟草税 事 件 决 定 教 训 新南威尔士州的许可费和罚 款制度是否通过其他方式构 成消费税,违反了澳大利亚宪 法对联邦政府的独家授权[72] 法院认定,州许可费是消费 税,这违反了澳大利亚宪法[72] 制定烟草税措施的主管机关必须在 其法定权力范围内行事 避免对进口和外国投资的歧视 虽然关税本身具有歧视性,但它的使用可能受制于一国贸易协定中商定的上限。旨 在提高烟草价格以减少需求和促进人类健康的消费税和其他税收应与原产地无关:它们 不应寻求对当地产品征收比外国产品更低的税,或旨在对外国产品区别对待。当烟草税 措施旨在将进口价格提高到高于当地产品的价格,或寻求比其他人更轻地负担受青睐的 市场参与者时,它们追求的目标不是健康。仅对健康有保护作用的烟草税也不会直接违 反国际投资协定(IIAs)的国民待遇(NT)、最惠国待遇(MFN)、征收和公平与公正待遇 (FET)条款中对投资者的反歧视保护[63,7374]。有可能以歧视以外的理由对违反国际义务 提出索赔,但这种索赔一般不大可能成功。 451 世界卫生组织烟草税政策和管理技术手册 案例研究8(消极):BAT诉乌干达案(2017年东非法院) 歧 视 教 训 乌干达对进口卷烟,包括来自东非关税同盟 伙伴国的卷烟,征收高于当地卷烟的消费 税。它的实施是违反关税同盟议定书第 15 条的歧视规定[75] 明确基于原产地的差别征税可以被解释为违反国际义 务的保护主义歧视。烟草业也可以求助于世贸组织之 外的国际贸易协定,特别是关税同盟机制 当表面上与原产地无关且保护健康的烟草税导致对烟草制品征税不同时,就会出现 困难[73]。歧视并不仅仅因为税收不同而存在,税收对进口商品的不利影响必须大于对本 地产品的不利影响,对一个国家的进口产品的不利影响大于另一个国家的进口产品,或对 特定投资者的产品的不利影响大于同类产品。如果产品类别之间的不同税收导致歧视, 而不同税收仅基于相关产品类别之间的合法监管区别,则该税收通常仍然是合法的 。 如果歧视是无意的,缺乏歧视意图不足以作为违反国际贸易协定或关贸总协定[73,76] 下的辩护理由。决策者应仔细审查以下措施,以确定: 1.税收措施的设计或实施的某个方面是否对进口产品或外国投资者损害大于当地产 品或国内投资者; 2.税收措施的潜在歧视性方面是否有助于支持税收措施(即实现健康目标所必需的); 3.是否有任何合理的替代方案可以在不存在歧视的可能性的情况下达到同样的效果 (即这是必不可少的); 4.当它被需要和不可或缺的时候,很有可能它是站得住脚的。 下面的案例研究提供了与烟草税措施有关的歧视的例子。 案例研究9(消极):泰国———对菲律宾卷烟的海关和财政措施(2010 WTO小组) 本案例的事实已在案例研究3中提出。本案例研究调查的是歧视的主张,而不是程 序问题。 551第四章 烟草税收的政治经济学 具体的适用规则因有关法律义务的性质而异。根据《关贸总协定》,可以根据第三条第2款规定的产品之间的科 学区别对同类或直接竞争的产品进行不同征税,另可以根据明确制定健康保护措施XX(b)条[73]所规定的是必要 的歧视。对于国际投资协定的最惠国和新台币条款,差异税收可以认为是非歧视的基础上,差异伤害意味着产品 “不一样”,或者,歧视是合理基于差异危害的科学证据和差别税收保护健康作用的合理原因[63]。 歧 视 解 释 教 训 泰国实施了确定卷烟 增值税税基的政策不 一致[61] 泰国采用了一种方法来确定进口卷烟 的税基,特别是营销成本部分,这与当 地产品的税基不同[61]。这导致进口卷 烟的营销成本部分高于一般方法下的 成本。这种待遇上的差异没有充分的 理由,因此被认为是歧视性的 由于从价税的税基固定时可能会 出现无意的歧视,因此税基的确 定必须一致且合理[61]。本案例 研究展示了政策制定者在设计和 实施从价税时需要如何小心,以 确保它们是非歧视性的并且在法 律上是可辩护的 泰国的增值税退税政 策可能带来更高的税 收负担,同时也对进口 卷烟提出了更繁重的 管理要求[61] 政府实体生产的卷烟经销商可免征增值 税[61]。尽管进口卷烟的经销商有资格 获得增值税的税收抵免,但这不是一个 自动过程[61]。进口卷烟经销商和本地 卷烟经销商的区别对待导致前者面临更 高的增值税负担的风险[61]。不同的待 遇还给进口卷烟经销商带来了额外的行 政负担并改变了竞争条件[61] 征收和执行税收义务的规则应在 形式和效果上与国内和进口烟草 制品相同或尽可能相似 案例研究10(消极):多米尼加共和国———影响卷烟进口和内部销售的措施(2004 WTO 专家组;2005 WTO 上诉机构) 根据 GATT 第 XX(d) 条,确保遵守合法税收措施所必需的歧视将是正当的,前提是 没有减少歧视的替代方案。在这种情况下,这种理由没有成功使用。 歧 视 解 释 教 训 多米尼加共和国的印花税规定 对进口商品具有歧视性[77]。根 据规定,所有卷烟包装都必须加 盖印花税票,但进口卷烟进口后 在当地税务机关的监督下加盖 印花税,而本地生产的卷烟可以 在生产加工过程中加盖印花税 本地产品和进口产品之间的这 种事实上的区别改变了竞争条 件,从而损害了进口卷烟,因为: (1)增加了进口商的成本;(2)损 害了进口产品的美观[77]。专家 组认为这种歧视是不合理的:没 有必要强制执行税收措施,因为 有限制较少的替代方案,例如允 许进口商在生产过程中贴上印 花税票(多米尼加共和国———影 响措施)[77]。该小组的调查结果 在上诉中得到维持[78] 为确保遵守税收措施而制定的 政策也必须是非歧视性的。当 进口产品的合规成本高于本地 产品并且这种事实上的区别是 可以避免时,就会产生歧视索 赔。重要的是要考虑负担较轻 的替代方案是否可以实现相同 的目标 651 世界卫生组织烟草税政策和管理技术手册 案例研究11(消极):费尔德曼·卡尔帕诉墨西哥案[2002年ICSID(国际投资争端解 决中心)仲裁庭] 当所得出的区别与公共福利目标之间存在合法联系时,仲裁庭已接受给予受国际投 资协定保护的投资者的不同待遇[76]。本案的事实在案例研究 5 中呈现。本案例研究审 查了涉及投资者主张歧视的案件的各个方面,而不是征用的实质性问题。歧视索赔的理 由与征收索赔的理由不同,这就是对这一索赔作出不同判决的原因。 歧 视 教 训 拒绝外国投资者的退税要求。向类似的当 地投资者授予退税申请[65],这违反了 IIA 的 国民待遇条款[65] 必须同等对待外国和本地投资者,并且必须使用一致 且文件齐全的政策来指导行政决策。拒绝退税可能是 合理的,但由于缺乏文件,政府无法证实这一点 避免投资激励陷阱 应避免烟草业与政府之间的投资者与国家之间的合同。它们不仅是国内法意义上的 “合同”,因为即使在没有适用的 IIA 的情况下,它们也可以国际化,为投资者提供以下权 利:①将争端解决从州法院撤出,支持独立仲裁;②从国家法律框架中移除争议,支持一般 法律原则[63,76]。因此,这些条款下的承诺不能在立法上得到缓和或消除,也不能在国内 法院内限制责任,因为国内法院更有可能支持国家为公共健康进行监管的权利[76]。 投资者与国家之间的合同和其他非合同诱因可以通过 IIA 中的总括条款进一步国 际化。这些条款使违背对投资者的承诺违反了 IIA[76]。此外,即使在没有总括条款的情 况下,合同和利诱也可以支持合法期望和违反公平和公正待遇的索赔,也可以加强投资者 对间接征收的索赔[63]。仲裁裁决明确指出,尽管税收可能会有所不同,烟草也会受到监 管,但投资者可以合理期望各国遵守正式的诱因和书面合同承诺。 投资者与国家合同中的一个共同条款,即稳定条款,对以证据为基础的烟草控制最有 效的措施是毁灭性的:增加消费税。稳定条款旨在从投资之时起冻结特定的国内法[63]。 看似不那么繁重的经济平衡条款要求缔约国补偿对投资价值产生负面影响的监管变 化[63]。这两种类型的条款在效果上几乎没有区别:违反均衡经济条款的成本责任可能非 常繁重,使其在财政上具有挑战性并在政治上令人不快。 案例研究12:投资者 国家合同 某州于 2001 年与 TTC 就其国有烟草企业私有化和建立合资企业达成了一项投资 协议。这项投资是为了根据协议提供经济利益:合资企业将利用 TTC 的现金和专业知 识增加出口和利润,同时确保优先考虑当地就业、制造业和资源。最终的投资者与国家合 同包括一种经济平衡条款,根据该条款,在规定日期之前适用于公司烟草制品的消费税税 率的任何增加都将得到补偿。虽然该协议并未从州法律中删除,但它规定了在对其赔偿 发生争议的情况下进行独立仲裁。此外,东道国与 TTC 子公司所在的另一个国家之间 有双边投资条约,其中包括 FET 条款(如果需要,这可以加强投资国独立安排提供的保 护合同)。对单独的 TTC 也有类似的不太正式的诱因。 751第四章 烟草税收的政治经济学 向烟草业提供激励的程度尚不清楚,但在国有烟草利益私有化以及投资者与国有烟 草企业之间的交易中,可能会提供合同和诱因[63]。尽管各国已被其对工业的激励所困, 但投资者与国家的合同提供了最清楚的例子,说明烟草业的承诺和诱因如何破坏烟草控 制[56,79,80]。各国应避免提供行业激励措施,尤其是与行业签订合同承诺。更系统的,政府 应考虑避免将激励和诱因置于合理和合理监管之上的国际投资协定。 4.2.3　结论 健康保护和原产地中性的烟草消费税在法律上是合理的,而行业威胁通常是毫无根 据的。但是,政府可能需要考虑某些关于程序、设计和协商的规则: 1.各国政府应了解国内法所要求的协商标准和任何适用的国际义务(案例研究1、2 和3)。重要的是,在允许的范围内,将烟草业与决策过程保持距离。不要给予行业特殊 考虑,但要确保在意识到潜在的程序操纵的同时,按照要求进行咨询———例如,通过提供 公开会议、及时的信息和提交行业意见的能力(案例研究1、2和3)。 2.消费税通常不会受到国内或国际法规定的没收或征用的挑战(案例研究4和5), 但是可以在其他法律或国家宪法或授予当局的税收权力限制中找到明确的税收限制(案 例研究6和7)。 3.在设计、实施或执行税收措施时,必须避免对外国烟草制品或投资者的明显和事实 上的歧视(案例研究8、9、10和11)。法律问题可能不是来自税收措施本身,而是来自支 持其实施的辅助措施(案例研究9和10)。 4.如果这种区分对进口产品的影响最大,并且必须根据对健康的影响的证据和缺乏 替代品来证明其合理性,基于产品对健康的影响而明确区分产品,就可能被视为歧视。 5.不应以奖励或合同承诺的形式提供投资奖励,因为这些可能具有约束力(案例研究 12)或可能根据国际投资协定提出挑战;它们也违反了《世界卫生组织烟草控制框架公约》 第5.3条准则。 4.3　威胁策略A：反贫困言论（累退性） 4.3.1　介绍 烟草业及其附属机构在努力游说反对增税时,经常声称烟草税的增加将伤害穷 人[81,82]。这一观点是基于有关税收的累退性的概念。从概念上讲,如果这意味着低收入 人群比富人支付更大比例的家庭收入来满足税收负担,那么税收可能是递减的。换句话 说,低收入家庭的税收负担往往相对高于中高收入家庭。然而,该行业的观点有两个局限 性。首先,仅基于税收负担的累退性概念没有考虑到烟草使用造成的更广泛的健康和经 济危害。第二,更高的烟草税和价格会导致人口之间的行为变化,这反映在需求的价格弹 性上[83,84]。综合起来看,这些更广泛的考虑有效地使烟草税成为一种渐进而不是倒退的 公共卫生干预措施。 851 世界卫生组织烟草税政策和管理技术手册 4.3.2　累退性和更广泛的视角 从狭义上说,烟草税可以被视为递减的,因为当烟草产品在税收增加后变得更加昂贵 时,低收入人群必须拿出比富人相对更多的家庭收入来支付烟草制品。在许多国家,来自 低收入群体的人比其他人更多地使用烟草[85]。世卫组织的一项系统文献综述发现,在成 年人中(包括男性和女性),低收入和高吸烟率之间存在密切的关联[86]。这一发现在30 年的研究中都是一致的,涵盖大多数地理区域和不同收入分类的国家中。例如,在印度, 高烟草使用率(即由30%以上的成年人使用),只在阿萨姆邦和奥里萨邦等低收入邦发 现,这些邦的人均净国内生产总值仍低于人均10万卢比(图4.3)[87]。 印度不同的邦 50 40 30 20 10 0 0 50 100 150 200 250 300 350 400 成 年 人 （大 于 15 岁 ）吸 烟 率 （ % ） 人均净国内生产总值（1 000 印度卢比） 图4.3 2016 2017年印度各邦和联邦地区成人烟草使用与人均净国内生产总值的关系 来源:参考文献87。 然而,这一发现并未考虑决定对家庭的全面影响的更广泛的健康和经济因素。当这 些更广泛的因素得到适当的考虑和解释时,烟草税实际上可以被视为一种累进的,或有利 于穷人的政策。在健康问题方面,低收入人群相对较高的烟草使用量转化为这些人群因 烟草引起的疾病负担更大,包括更高的发病率和死亡率。 低收入群体也更无力负担治疗烟草引起的疾病的医疗费用,而高额的自付医疗费用 会使许多家庭进一步陷入贫困。因此,许多穷人得不到甚至不寻求他们需要的医疗服务。 一项研究发现,在孟加拉国,55% 被诊断患有与烟草相关疾病的患者没有寻求进一步的 医疗护理。这种缺乏医疗保健的部分原因是自付费用过高[88]。高烟草使用率和无法获 得负担得起的医疗服务的结合意味着烟草使用显著增加了包括中国和印度在内的一些高 烟草负担国家的贫困率[89,90]。 951第四章 烟草税收的政治经济学 众所周知,穷人比富人对价格更敏感。低收入吸烟者比高收入吸烟者表现出更高的 价格弹性。世界银行最近的研究证明了这一点,其结果如图 4.4 所示[91]。与其他人相 比,穷人通过减少对烟草制品的使用而对更高的烟草税和价格做出更强烈的反应,因此他 们在避免与烟草相关的死亡、疾病和相关医疗费用方面受益更大。一项关于烟草控制政 策对烟草流行后期高收入国家社会经济不平等的人口影响的系统审查得出了类似的结 论[92]。审查发现了 16 项与税收有关的相关研究,其中只有一项发现税收与社会经济梯 度之间存在递减关联(7项发现存在累进影响,而其他研究结果喜忧参半)。 波斯尼亚和黑塞哥维那 乌克兰 摩尔多瓦共和国 南非 孟加拉国 印度尼西亚 俄罗斯联邦 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 -1.2 1 2 3 4 5 6 7 8 9 10 十分位 弹 性 图4.4 烟草消费的价格弹性,中等估计,按十分位数计算 来源:参考文献91。 世界银行的扩展成本效益分析(ECBA) 框架解释了这一更广泛的经济观点,该框架 评估了烟草税增加对健康以及其他因素的分配影响[82,83]。也就是说,ECBA 框架超越了 累退性的简单或部分定义(即收入水平对家庭支出的影响),以涵盖福利的全部分配,包括 健康和收入的改善。 ECBA 框架已应用于多个国家的研究,包括孟加拉国、波斯尼亚和黑塞哥维那、智利、 印度尼西亚、摩尔多瓦共和国、南非、俄罗斯联邦、乌克兰和越南。这些研究的证据支持这 样一种观点,即有效的烟草税政策可以产生有利于穷人和改善福利的结果。当考虑到因 与吸烟相关死亡率降低而导致的医疗支出减少和工作年限延长时,增加烟草税的总体政 策将变得累进而不是累退(图 4.5)[84]。对高收入国家的研究也得出了类似的结论,例如 美国于2009 年提高了烟草税[93]。 烟草税的增加通常也会导致富裕的吸烟者比贫穷的吸烟者对税收总收入的贡献相对 更多。这是因为较贫穷的吸烟者减少消费最多,因为他们对价格更敏感,而较富裕的吸烟 061 世界卫生组织烟草税政策和管理技术手册 者也倾向于购买优质(价格和征税更高)的烟草制品[94]。因此,就戒烟带来的额外收入和 健康收益以及从一开始就防止年轻人开始吸烟而言,提高烟草税可以被视为渐进式的。 中国的一项研究表明,50% 的税收增加将在 50 年内筹集 7 030 亿美元,其中只有 14% 的增长由收入最低的1/5的吸烟者承担[95]。此外,税收增加将使与烟草相关疾病的支出 节省 240 亿美元,其中约 28% 的节省由收入最低的1/5的吸烟者享用。 ECBA 框架实际上对提高烟草税的净收益进行了相当保守的描述,因为它不包括其 他收益来源,例如减少接触二手烟造成的危害、提高生产力以及贫困家庭受益的潜力来自 通过增加税收资助的社会计划[96]。 波斯尼亚和黑塞哥维那 乌克兰 摩尔多瓦共和国 南非 孟加拉国 印度尼西亚 俄罗斯联邦 5 4 3 2 1 0 1 -1 2 3 4 5 6 7 8 9 10 收 入 （ % ） 图4.5 价格上涨100%的影响,具有中等弹性,按十分位数计算 来源:参考文献91。 对美国 2009 年烟草税增加的分配影响的评估发现,税收收入用于扩大中低收入家 庭子女的健康保险覆盖范围,提高了烟草税增加的总体累进性[97]。考虑到这种扩大的覆 盖范围增加了整体立法方案的累进性,底线是对低收入1/5的影响是积极的,平均而言, 对低收入家庭的影响最大[93]。 同样,菲律宾所谓的罪恶税改革的很大一部分烟草税收入用于补贴贫困和接近贫困 的家庭的全民健康保险覆盖(UHC)。在全球范围内,已知有 37 个国家将部分烟草税收 入指定用于卫生计划,其中许多计划间接使穷人和弱势群体受益的比例超过其他群体[27] (有关指定用途的详细信息,请参见第 4.6 节)。 161第四章 烟草税收的政治经济学 4.3.3　总结 与烟草税是累退税的看法相反,当考虑到更广泛的经济影响时,这是一项强有力的扶 贫政策。税收负担不是累退性的完整指标,因为它不包括烟草引起的疾病对健康和经济 的负面影响,也不包括因税收和价格上涨而改变行为的积极影响。 烟草引起的疾病的健康和经济负担不成比例地落在穷人身上,他们往往对烟草的使 用率更高,而且最无力负担必要的医疗保健。由于穷人往往对价格更敏感,他们比富裕的 吸烟者更大幅度地减少他们对烟草制品的使用和消费以应对增税,这反过来又降低了他 们下游的健康和经济成本。 烟草税可通过再分配政策专款专用于某些特定的社会产品和服务来惠及低收入阶 层,从而提高其累进性(见第4.6节)。 4.4　威胁策略R：收入减少 4.4.1　介绍 烟草业及其盟友认为,增加烟草税会导致政府的税收减少。他们认为,收入的减少要 么是由于替代了更便宜、税收较低的或走私的烟草制品,要么是由于总体消费的减 少[98,99]。 烟草业经常引用拉弗曲线来证明这个论点。根据这条曲线,税收会随着税率增加到 一定程度,之后进一步提高税率会导致税收的下降。在考虑烟草税时,烟草行业假定各国 已经接近或甚至超过了临界税率水平[98]。 然而,这一论点建立在一个狭隘的理论和经验不足的基础上[98100]。许多国家对烟草 的价格无弹性需求和价格相对较低的税收份额解释了公共卫生和财政的双赢,即消费下 降和收入增加可以同时发生[98,101]。此外,许多国家的例子(见下文的案例研究)表明,设 计良好和实施良好的烟草税增加会导致收入的增加,至少在中短期是这样[98,100]。虽然消 费将随着烟草税的增加而减少,但单位消费税的百分比大于烟草消费的百分比,至少消除 了消费减少对收入的一些影响[98,99]。 在所有其他影响消费的因素保持不变的情况下,税率的变化相当于税收的变化,并由 沿拉弗曲线的移动来表示。随着税率的变化,税基的弹性也会发生变化;拉弗曲线上的每 个点都对应着不同的税基弹性。当一个或多个其他因素发生变化时,将影响曲线的位置, 而税基弹性将在给定的税率下发生变化。例如,一项成功的无烟政策或广告禁令可以减 少对烟草的需求,使曲线下降,降低每个税率的税收潜力。 为了证明很少有国家(如果有的话)超过了拉弗曲线上的收入最大化点,表4.7显示 了在不同情景下增加消费税对收入的影响,根据国家的收入水平,使用不同的需求价格弹 性,不同水平的增税和不同的起征点。这是衍生拉弗曲线的税基弹性方法(更多细节请参 见第2.2.3节和附件2.2)。显示的总税收和消费税份额是每个国家的收入组根据 RGTE数据集计算出的加权平均值。收入增长通过消费税的累进税率(25%、50%、75% 261 世界卫生组织烟草税政策和管理技术手册 和100%)和不同的价格需求弹性(-0.4~-1.2)进行模拟。 表4.7 在不同的税收水平、税收增加和价格弹性情况下的消费税收入增长百分比 收入阶层 总税 占 零 售 价格的比例 消 费 税 占 零 售 价 格 的比例 消 费 税 的 增加比例 当需求的价格弹性变化时,消费税收入增加: -0.4 -0.6 -0.8 -1.0 -1.2 低收入 38% 22% 25% 22% 20% 19% 17% 16% 50% 43% 39% 36% 33% 29% 75% 63% 57% 52% 46% 41% 100% 82% 74% 66% 59% 51% 中等收入 58% 41% 25% 19% 17% 14% 11% 9% 50% 37% 31% 26% 20% 15% 75% 54% 45% 36% 27% 19% 100% 71% 57% 45% 34% 23% 高收入 68% 55% 25% 18% 15% 11% 8% 5% 50% 35% 27% 21% 14% 8% 75% 50% 39% 29% 19% 10% 100% 65% 50% 36% 23% 11% 数据来源:作者根据RGTE收集的数据计算得出[27] 。 在模拟中考虑的所有场景中,收入都出现了大幅增长。这些结果重申了大部分已知 的事实,即更高的税收增加会产生更高的收入收益,并且这些收益随着需求缺乏弹性的增 加而增加。即使当需求具有相对的价格弹性(-1.2) 时,模拟预测收入也会增加。价格 中的税收份额也会影响收入潜力。税收在价格中的份额越低,收入潜力就越大。这表明, 只有在极端情况下(非常有弹性的需求加上非常高的当前税收份额),才会发生消费税增 加导致的收入减少。需要注意的是,大量的实证文献表明烟草价格普遍缺乏弹性;因此, 决策者不应相信极端情况。此外,如表 4.7 中的数据所示,大多数国家的税收份额相对 较低,增强了烟草税增加带来的收入潜力。 361第四章 烟草税收的政治经济学 这些预测使用来自185 个国家的 2018 年数据。这些国家根据世界银行收入组进行分类,每个国家的平均总税收 份额、消费税份额和增值税/销售税份额根据当前成年吸烟者的数量加权。为了计算每个规定弹性的预计收入, 假设消费税增加将完全转嫁,同时非消费税(增值税/销售税)占零售价格的百分比不变。随之而来的价格变化与 各自的弹性相乘,得出预期的消费变化。通过将新的消费数据与增加的消费税相乘,可以很容易地计算出预计的 收入。 这些计算没有考虑品牌替代(交叉价格弹性)、收入效应或非法贸易。消费税被假定为从量税,而非消费税(增值 税和其他税)被捆绑并作为以零售价为计税基础的从价税处理。零售价减去所有税款之间的差额也被假定为常 数,税费增加完全转嫁给消费者。 烟草税的收入潜力确实相当可观。据估计,2018 年,卷烟消费税在全球共产生了 3 610 亿美元的收入,其中中低收入国家的收入为 1 620 亿美元。如果所有国家都将每包 卷烟的消费税提高1 美元,消费税收入将增加 1 780 亿美元至 2 190 亿美元,或在 2018 年的水平上增加 49%~61%。中低收入国家将从这些增税中获益最大,这些国家的消费 税收入将增加 1 330 亿美元至 1 670 亿美元,即增加 82%~103%。 受调查国家的收入减少是由于其他原因,而不是税收增加本身。例如,汤加在 2016 年大幅提高了卷烟消费税,消费量大幅下降(下降 40%),随后收入下降。这是因为 20% 的吸烟者转而使用一种名为 Tapaka Tonga 的免税、廉价的当地散装烟草制品[102]。吸取 的教训是汤加需要对其所有烟草制品征收相同的税,以避免低价、免税烟草制品的替代。 另一个与税收增加无关而与税收管理不善有关的收入减少的例子是南非的情况(见本节 后面案例研究中的解释)。最后,由于烟草使用长期下降趋势导致的收入下降不应与超出 拉弗曲线的收入最大化点混淆。例如,在英国,烟草使用长期下降,尽管消费税保持不变, 但 2017 年至 2018 年收入出现名义下降。相反,即使烟草消费税非常高的国家也会因烟 草税增加而增加收入[98,100](见下文澳大利亚的案例研究)。这表明,很少有国家(如果有 的话)超出拉弗曲线上的收入最大化点。 烟草消费预计将缺乏税收弹性,即使由于成功的烟草控制干预措施,需求实际上变得 具有价格弹性。税收是实现财政和公共卫生目标的工具。如果在成功的烟草控制干预之 后,价格达到需求有弹性的水平,由于税收转移不足,税基仍然最有可能缺乏弹性,因为当 需求有弹性时过度转移不是一个好的定价政策(关于税收转移更详细的讨论见第 2.2.2 节)。换句话说,提高税率与非烟草价格控制措施相结合,使消费者对价格(税收)增长更 加敏感,导致边际收入下降但仍为正。 从长远来看,包括价格和税收措施在内的烟草控制政策在减少消费方面可能非常成 功,以至于收入将趋于平稳或下降。这最终是长期的政策目标。减少政策的影响并结束 全球烟草流行是烟草控制的目标,而不是可以避免的事情。然而,不幸的是,在中短期内 无法预见结束全球烟草流行;因此,政府目前可以依靠烟草税作为可靠的收入来源[103]。 4.4.2　消费税增加的收入影响：案例研究 以下案例研究说明了4个关键点:①大规模和定期的税收增加导致收入的大规模和 持续增长;②高税收和烟草使用流行率下降的国家仍然可以通过增加税收来增加收入; ③减税国家收入下降;④面对非法贸易而增税的国家仍然增加收入。 大规模和定期的增税通常意味着大规模且持续的收入增长 南非的经验表明,即使在税收大幅增加之后,连续不断的税收增加,远高于通货膨胀 率,而且年复一年地产生了额外的收入。在20世纪70 年代和 20世纪80 年代,随着每包 卷烟实际消费税的下降,实际收入连续下降了 20 年,南非从 1994 年到 2011 年连续实施 了消费税上调(图4.6)[98,104]。在调整通货膨胀后,这导致实际消费税收入增加了 245%[98]。随着税收增长停滞,收入从 2012 年开始趋于平稳。它们在 2015 年之后开始 461 世界卫生组织烟草税政策和管理技术手册 Goodchild M、Perucic AM、Paul J 将烟草税作为实现全球吸烟率目标的战略。未发表的手稿。2020 年 10 月。 下降,但不是因为税收增加,而是因为行政能力急剧下降和执法措施不力,包括税务管理 机构在内的政府大规模腐败加剧了这种情况[105]。税收管理和执法的迅速和灾难性的下 降一直是备受关注的主题[106]。 每包消费税 消费税收入 20 15 10 5 0 每 包 兰 特 （以 20 20 年 兰 特 币 值 为 基 数 ） 18 14 10 6 0 消 费 税 收 入 （以 2020 年 兰 特 币 值 为 基 数 ） 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11 20 13 20 15 20 17 20 19 图4.6 1961—2020年,南非每包卷烟的实际消费税和实际的消费税收入 数据来源:数据由南非开普敦大学提供,2020年。 类似地,菲律宾提供了一个令人信服的例子,说明在税收结构改革的同时,大规模和 定期的税收增加能够导致大量和持续的收入增长,在这种情况下,也伴随税收结构的改革 (图4.7)[98]。2012 年的罪恶税法在 2013 年将国家的4个税级合并为2个,并在 2017 年 建立了统一的结构。同一法律规定了全面的大额累进增加,特别是针对最低税种[98]。不 仅收入增长显著,而且超过了该法律通过之前对 2013—2017 年的所有预测[98]。 乌克兰是另一个在过去10年里经常增加税收、收入增加、消费和吸烟人数减少的国 家的例子。图4.8显示了2008—2017年消费税、收入、卷烟销售和吸烟人数的变化趋势。 消费税的增加一直伴随着收入的增加。2014—2015年,消费税的增幅并没有超过通货膨 胀(尤其是2015年的通货膨胀非常高,达到48.7%),因此消费税和收入的实际价值下 降。但从数据中可以明显看出,即使销售下降,收入也密切遵循消费税水平的路径。 561第四章 烟草税收的政治经济学 实际烟草消费税收和 不同价位卷烟每包实际消费税 图4.7 2012—2018年菲律宾每包香烟(多层、双层、单一)的实际消费税和烟草实际消费税总收入 注:数据根据通货膨胀进行了调整,使用了2020年4月国际货币基金组织《界经济展望》中的平均消费者价格的 年百分比变化,并以2012年为基准年。 资料来源:参考文献107、108和菲律宾财政部共享的数据,2020年9月。 每天吸烟的人数，以百万为计数单位 实际烟草消费税收入，单位为十亿乌克兰格里夫纳 每包 20 支烟的实际平均消费税，乌克兰格里夫纳 征税卷烟数量（销售额），以十亿支为单位 125 112 95 11.2 12.1 13 11.8 12.7 13.4 10.1 9.2 10.3 88 82 75 76 73 74 9.7 67 7.3 8.18.4 8.68.7 7.8 3.6 6.2 6.5 6.3 4 3.33.2 2.7 3.5 2.92.52.2 1.4 0.6 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 图4.8 2008—2017年乌克兰平均实际卷烟消费税税率、 实际卷烟消费税收入(2008年为基准年)、卷烟销量和吸烟人数 注:根据2020年4月国际货币基金组织《世界经济展望》中平均消费者价格的年百分比变化,并以2008年为基 准年,针通货膨胀对数据根据进行了调整。 资料来源:康斯坦丁·克拉索夫斯基提供的数据,2020年7月。 税收高和烟草使用流行率下降的国家仍然可以通过增加税收而增加收入 烟草税已经很高且烟草使用迅速减少的国家仍然可以通过增加税收来增加收 入[98,109]。澳大利亚实施了全面的烟草控制政策,并在已经是世界最高税率的基础上持续 增加烟草税(见图 4.9)。2001 年至 2010 年,收入随着税率的提高而增加,但按实际计算 (经通胀调整后)保持不变[109]。然后,在 2010 年,引入了 25% 的消费税上调,并计划从 2013 年开始每年大幅上调[98,109]。近十年来,这项税收政策的结果是收入年复一年持续 地大幅增加,即使是在税率已经很高的情况下也是如此。 661 世界卫生组织烟草税政策和管理技术手册 2012年和2013年收入的明显减少是由于2001—2011年和2012—2016年的数据来源发生了变化。2012年和 2013年的数据不包括关税,而其他所有年份都包括关税。 总 关 税 收 入 （以 20 19 年 为 基 准 年 进 行 了 通 货 膨 胀 调 整 ， 澳 大 利 亚 元 ，百 万 ） 12 000 9 000 6 000 3 000 0 200 1 200 2 200 3 200 4 200 5 200 6 200 7 200 8 200 9 201 1 201 0 201 2 201 3 201 4 201 5 201 6 201 7 201 8 201 9 总 消 费 税 收 入 （每 包 关 税 , 以 2019 年 为 基 准 年 进 行 了 通 货 膨 胀 调 整 ， 澳 大 利 亚 元 ） $0.8 $0.6 $0.4 $0.2 $0 从量税/每包进口关税（低于 0.8 g 的卷烟或雪茄） 总税收收入（所有烟草制品） 图4.9 2001—2019年澳大利亚每支烟草的实际消费税和关税以及实际总收入(所有烟草制品) 注:由澳大利亚税务局和澳大利亚移民和边境保护部公布的税率,根据澳大利亚统计局消费者价格指数税率进 行调整。2011—2012年和2012—2013年的数据不包括关税,这是收入明显下降的原因。根据官方披露,斯科 洛和贝利估计,这些年的关税分别为7 397.2美元和7 687.2美元的[110]。 资料来源:参考文献109和110。 减税和收入下降的国家 1982 年之前,加拿大降低了卷烟税,经历了收入下降和吸烟增加的情况———尤其是 在年轻人中。随后在 1982 年至 1992 年将卷烟税增加了五倍,导致收入增加、零售价格 上涨和消费量大幅减少,青少年吸烟率下降了近2/3[5]。在20世纪90年代初期,卷烟非 法贸易日益增长,其中出口到美国的加拿大卷烟被走私回加拿大[5]。烟草业后来被发现 参与了这种非法贸易并从中获利,试图将加拿大的高税率定为走私的原因[111,112],并成功 说服了联邦政府以及六个省政府,大幅降低烟草税[111,112]。结果,联邦税收显著下降(是 政府预测的两倍多),成人和青年的吸烟率开始上升[5,112]。加拿大政府后来改变了策略, 恢复了联邦消费税,从而增加了收入并减少了吸烟人数[5]。随后,加拿大的重点转向使用 海关执法而不是税率,作为打击非法贸易的最佳手段[5]。 面对非法贸易增加税收,但仍增加收入的国家 如第 4.1 节所述,烟草业利用非法贸易作为破坏烟草税收政策的战略,其目标是阻 止政府提高烟草税。已经形成的说法是,较高的烟草税率导致非法贸易增加,并通过导致 烟草使用减少(或不减少)或收入减少(或不增加)而破坏政策目标。然而,如第 4.1 节所 述,经验证据并不支持行业论点。此外,证据表明,该行业及其盟友一直吹嘘和夸大非法 贸易的规模和程度(见第 4.1 节)。 761第四章 烟草税收的政治经济学 与加拿大的情况一样,巴西的烟草税收政策担心非法市场会扩大,除非合法市场的价 格竞争削弱它,人们认为最好通过减税来鼓励这种竞争[20]。从 1999 年到 2000 年中期, 实际消费税税率下降,因为名义增幅低于通货膨胀率。这导致实际税收收入下降[20]。这 些年来,烟草业利用减税来增加利润率而不是降低价格并战胜非法市场,同时也夸大了非 法贸易问题的规模和范围[20]。这导致该行业关于非法贸易和收入的论点失去可信度,并 导致从 2007 年开始提高税率,并于 2011 年通过了一项重大改革[20]。法律规定烟草消费 税和最低价格在 2011 年至 2015 年提高到高于预期通胀的水平[20]。这导致每包烟草的 消费税和总收入大幅增加,到 2015 年,从 2013 年的低点到 2015 年实际增长 50% 以上 (图 4.10)。这项改革的成功表明,尽管存在相当大的非法市场,但可以通过提高税率来 增加收入[113]。最近的数据显示,巴西的收入在 2015 年和 2016 年有所下降,这与非法贸 易增加相吻合,但同时也发生了异常严重的经济衰退,这些年巴西的 GDP 下降了 3% 以 上。 每 包 卷 烟 实 际 消 费 税 （2013 巴 西 雷 亚 尔 ） 1999—2014 年巴西每包卷烟平均实际消费税 国内卷烟销量 实际消费税收入 6 5 4 3 2 1 0 3 2 1 0 国 内 销 售 量 （十 亿 包 ） 实 际 消 费 税 收 入 （十 亿 ， 2 01 3 巴 西 雷 亚 尔 ） 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 图4.10 1999—2014年巴西每包卷烟平均实际消费税、实际消费税收入和国内卷烟销量 注:联邦收入秘书处根据国内销售和烟草消费税收入的数据,使用消费者价格指数,以2013年巴西雷亚尔为索 引。使用消费者价格指数计算2013雷亚尔。 来源:参考文献20。 税收管理和执法的改进也可以增加收入。在肯尼亚,包括采取财政标记和后来的先 进跟踪和追溯系统在内的多项措施改善了税款征收,从而增加了合法销售和税收收入并 减少了非法销售[114]。此外,这些治理不善的例子表明,应将注意力集中在行政和执法能 力严重丧失而损害税款征收的国家。 861 世界卫生组织烟草税政策和管理技术手册 4.4.3　总结 烟草业将收入问题作为一种 SCARE 策略来避免、稀释和延迟烟草税的增加。增加 税收会减少收入的论点在理论上是合理的,但现实世界的例子表明这并不现实。此外,模 拟表明,即使当前平均税收份额大幅增加,也会产生可观的收入收益。 烟草业对拉弗曲线的使用应该受到挑战和反驳。卷烟需求的价格相对缺乏弹性,加 上税收份额较低且没有过度转移,这意味着大多数(如果不是全部)国家仍远未达到收入 最大化点,这表明增加税收将导致收入增加。 本节中的案例研究驳斥了烟草业关于因增税而导致潜在收入损失的每一个论点。南 非、菲律宾和乌克兰的经验表明,大规模和定期的税收增加会导致大量和持续的收入增 长。事实证明,精心设计的税收结构在创造收入方面也发挥着重要作用。澳大利亚的经 验表明,即使是税率已经很高且烟草使用率下降的国家,也可以通过定期大幅增税来增加 收入。加拿大的经验告诫人们不要听从烟草业的建议,将减税作为打击非法贸易的一种 方式。它清楚地表明,降低烟草税将减少收入并鼓励消费,而不是打击非法贸易。巴西的 经验表明,存在大量非法贸易问题的国家仍然可以通过增加税收来增加收入。 最后,在收入减少的少数情况下,减少的原因与税收增加并不严格相关。汤加就是这 种情况,那里的税收增加仅适用于卷烟,而不是它们的近似替代品(散装烟草),导致吸烟 者更换产品。在南非,收入减少是政府机构能力削弱的结果。而在乌克兰,实际收入仅在 税收没有超过通货膨胀率的两年内下降。 4.5　威胁策略E：影响就业 4.5.1　介绍 在反对增加税收的过程中,烟草业往往试图将烟草税视为经济问题而非公共卫生问 题[5,48,53,115]。特别强调税收增加对烟草种植和制造以及其他相关行业的就业构成的所谓 威胁[5]。然而,这种所谓的健康和工作之间的选择是基于几个错误的前提,包括[5]: 1.在更广泛的劳动力市场背景下,烟草是重要的就业来源,国内烟草税的增加将对国 内就业产生巨大影响[48,53,116,117]; 2.烟草消费是创造就业不可或缺的引擎[5,48,54]; 3.烟草提供高度繁荣、可持续和不可替代的生计[5,53,118]。 实际上,烟草税与就业之间的关系远比该行业想象的要复杂得多。事实上,有充分的 证据表明,烟草税对公共卫生和财政空间是双赢的,对就业没有可衡量的风险。 4.5.2　烟草行业就业情况与烟草税率之间的联系 烟草种植、生产和制造业(包括一些国家的手工卷制,其中大部分在东南亚)占劳动力 的一小部分,即使在该行业最集中的国家也是如此[5,48,103,116]。由于技术进步、贸易自由 化、市场整合以及国有烟草公司的私有化,全球烟草种植和制造业的就业率一直在下 961第四章 烟草税收的政治经济学 降[5,103,119]。这些相同的趋势导致烟草种植和制造高度集中在少数几个国家———而且在 这些国家内部,通常只有少数地区[103,119121]。即使在烟草种植和制造领域处于领先地位 的国家,烟草在农业和制造业就业总人数中所占的总体份额也相对较小,而且随着生产效 率提高,烟草业从业人员所占的份额通常会下降[5,117,118,122]。 同样,烟草业声称烟草税会减少就业的说法被夸大了,而且通常忽视了推动烟草业就 业的更广泛趋势。事实上,与烟草控制政策相比,烟草业的发展和创新在减少烟草业就业 方面发挥了更大的作用[103]。尽管烟草业声称烟草税会影响就业,但生产地点的特征,例 如市场规模、劳动力成本、生长条件和烟叶偏好,与烟草业利益的关系比与烟草税率的关 系要大得多[103,123]。此外,以出口为主要目的的烟草生产国的就业机会不会受到税收增 加导致当地消费减少的重大影响[5,103,116,123]。最后,事实证明,提高烟草税对零售业的就 业情况没有显著影响,因为大多数零售企业还销售其他商品[103]。 对烟草税上调对就业总影响的估计表明,确实发生的失业可以通过增加收入来弥补。 世界银行 2018 年的一项研究表明,在印度尼西亚,一项雄心勃勃的税收改革将简化层级 并将价格提高近 50%,将使烟草制造业的总就业减少不到 0.5%(失去 2 914 个工作岗 位)。政府可以向失业工人提供收入支持(例如,通过培训、转移或收入支持),而税收增加 不到 2%[117]。同样,孟加拉国国家税务局 2019 年的一项研究表明,烟草税的大幅增加将 导致 7 012 人失业,但比迪烟行业与这些失业相关的总收入仅占所获得收入的 3. 5%[120]。因此,增加的收入可以补偿那些失去工作并需要在过渡到新工作之前获得新技 能的人的支出[120]。 框4.2 就业担忧阻碍了印度尼西亚的烟草税改革 2017 年,印度尼西亚财政部决定于 2019 年实施烟草税上调,到 2021 年实施分 级简化[124]。然而,这项决议在一年内被放弃,因为烟草业参与者及其盟友开展了一 项协调一致的运动,将增加的费用重新定义为一个经济问题,重点关注增加烟草税对 就业的影响[124]。尽管估计与失业相关的收入损失与税收带来的额外收入相形见 绌[117],但烟草税倡议的失败还是发生了。早先的分析估计烟草税的增加对就业产 生了巨大的净正面影响[125],这说明了对烟草控制的负面社会经济影响的假设提出 挑战的证据往往被低估[126]。在这种情况下,烟草业的论点似乎引起了西爪哇、东爪 哇、中爪哇和西努沙登加拉选区的政治家的强烈共鸣,这些地区的烟草种植和制造业 就业集中[124]。尽管烟草制造业仅占制造业总就业人数的 5.13%,但工作和经济活 动的集中意味着关于就业的争论尤为突出[117,119]。因此,尽管有强有力的反对证据, 烟草业认为烟草税会对就业和农民生计产生负面影响的论点盛行[119]。烟草行业活 动在国家内部的集中对克服行业关于就业的争论构成了特别的障碍,需要给予谨慎 关注。 4.5.3　烟草税增加对就业的净影响的证据 对增加烟草税对就业的影响的正确分析必须检查它们对净就业或经济范围就业的影 071 世界卫生组织烟草税政策和管理技术手册 响。与烟草控制相关的烟草支出减少并不意味着支出就会消失;相反,它们被重新分配给 其他商品和服务的消费,从而在经济的其他地方创造就业机会[5,103,123]。同样,尽管提高 烟草税对净消费的影响可以说更加模糊,但这种干预的收入确实产生了卫生和教育等公 共服务的支出、投资和就业[5]。烟草控制政策通常对净就业产生边际中性或积极影响,特 别是在烟草原料或制成品净进口的国家,因为这些进口商品的支出往往流出该国[5]。以 出口为导向的烟草生产商对当地需求不太敏感,并且不受国内烟草税措施的显著影响,国 内烟草税措施可能具有近乎中性的净影响[5]。在某些情况下,净就业影响是一个非常小 的负数,通常不到 1%[127,128]。 最近的一项研究估计,在坦桑尼亚联合共和国(一个烟草生产和出口大国)吸烟率降 低 30% 将导致整个经济体的净就业仅下降 0.5%[129]。巴基斯坦的一项类似研究发现, 根据烟草消费支出的再分配情况存在一些差异,显著减少卷烟就业支出(10 亿卢比)对就 业的总体净影响将增加 6 651 至 5 803 个工作岗位[122]。之所以会出现这种增长,是因为 卷烟支出在更广泛的经济中产生的就业机会远少于食品和教育支出[122]。在坦桑尼亚联 合共和国,与其他地方一样,增加的收入可用于帮助失业者过渡到新的生计。 4.5.4　改善生计的可行性 在反对增加烟草税的争论中,烟草业提出了一个神话,即从事烟草生产的人们(尤其 是烟草种植业及制造业)没有任何其他相对有吸引力的生计前景。然而,基于印度尼西 亚、肯尼亚、马拉维、菲律宾和赞比亚的广泛调查数据的研究表明,尽管烟农需要为他们的 作物投入大量劳动力,但他们往往遭受损失而不是获利[119,121,123,130—132]。此外,印度尼西 亚的研究表明,与其他放弃烟草种植的农户的经验相比,烟草种植对家庭收入和机会有负 面影响[119]。 烟草税增加导致的消费下降是渐进的,并且容易受到几十年来发生的相同渐进式适 应的影响[5,103]。虽然在一些国家,政府需要帮助农民从长远来看过渡到其他作物种植或 行业,但这一过程不会对就业或更广泛的经济造成重大的短期冲击[123]。 由于烟草种植和制造可以集中在一个国家的少数几个地方,烟草业的失业可能在一 个地方产生不成比例的影响,而减少消费带来的就业收益可能会分布在整个国家[120,123]。 一项关于孟加拉国烟草税增加对就业影响的研究估计,由于行业高度集中,多达 60% 的 失业将仅发生在该国最贫困的两个地区[120]。在这些情况下,研究对支持的需求以及提 供支持的方式和资金来源尤为必要。除了需要确保公平和支持就业之外,未能提供有针 对性的救济会加剧人们对失业的恐惧,并可能对烟草税提案造成致命影响[120]。 方框4.3 支持菲律宾的替代生计 菲律宾将2012年烟草税增加收入的15%专项用于支持烟农和工人的经济上可 行的替代生计[5]。菲律宾的烟草种植是区域性的,烟草业以前成功地调动了对小烟 农的关注,从而破坏了烟草控制措施[136]。经济支持是遏制烟草业“威胁”策略的政 治有效对策,减轻了增税法案的通过阻力。 171第四章 烟草税收的政治经济学 鉴于菲律宾与全球烟草市场和需求的一体化,烟农没有受到烟草税增加和内需 减少的严重影响[108,132]。尽管如此,从专用税转移到烟草种植地区的资金是大量 的[108]。这些资金支持逐渐过渡到替代生计,鼓励农民种植替代作物,并建立基础设 施,如从市场到农场的道路,使这些替代作物在经济上更可行[108,133,134]。 支持烟农和其他烟草工人的替代生计很重要,因为如此可以抵消行业争论的政治影 响,即使国内税收增加通常对就业只有温和和渐进的影响。当烟草部门的就业因全球或 国家需求下降而逐渐减少时,有多种支持替代生计的模式。 菲律宾堪称典范,但许多其他国家已经实施或尝试支持作物转型。土耳其的替代作 物计划是在预期该国卷烟垄断企业私有化的情况下实施的,已证明有效地支持了许多烟 农转向其他种植作物[135]。肯尼亚和中国云南省的小规模作物替代项目展示了政府的财 政、监管和基础设施支持如何促进作物转型[5,53]。阿根廷、孟加拉国、墨西哥和美国的马 里兰州提供了关于政府如何支持这些转型的额外案例研究[5,136]。 4.5.5　总结 烟草业夸大了烟草就业的重要性,并夸大了地方税收导致的国内需求减少对服务于 全球市场的烟农的影响。该行业还通过只关注烟草的总就业人数来简化就业与税收的关 系,这忽略了这样一个现实,即烟草支出不会消失,而是重新分配给可以产生类似或更多 就业机会的其他消费。 许多详细研究发现,烟草种植的利润和可持续性比烟草业声称的要低得多。世界各 地的烟农已成功过渡到种植其他作物,尽管这种过渡通常需要政府或其他利益相关者的 临时或额外支持。这种支持的程度受到现实的限制,从种植烟草到种植其他作物的过渡 是一个长期考虑的现实。 4.6　将烟草税收入用于资助卫生事业 4.6.1　介绍 指定税收收入的用途涉及将全部或部分收入从一种税款或一组税款中分离出来,以 用于特定目的[137]。在全球范围内,有 80 多个国家将烟草税收入用于健康[138],37 个国家 将烟草税收入用于健康[27]。有两种主要类型的专用标记:硬标记(也称为实质性的)和软 标记或象征性标记[139]。硬性专项拨款将支出与立法中的收入来源联系起来。如果专项 收入是资金的主要来源,这可能会限制资金,或者当筹集到的收入超过用于指定用途的支 出时,会导致盈余浪费累积。软性专项拨款包括专项资金或将资金用于特定目的的承诺。 它们不一定具有法律约束力。例如,在法国,大部分烟草税收用于资助社会保障(包括健 康保险和医疗保健),但没有硬性、正式的专项拨款[140]。标记也可以是软硬结合。在菲 律宾,烟草税专项资金具有法律约束力,但专项收入归普通基金所有,作为预算申请的一 271 世界卫生组织烟草税政策和管理技术手册 部分,卫生部必须提交涵盖项目的年度预算。 专款专用是一个广泛而有争议的话题,它超出了指定烟草税用途的具体细节。关于 这个话题的讨论属于公共财政管理的范围,一般不鼓励指定用途。然而,从烟草控制的角 度来看,最好将指定烟草税用途理解为向公众、政治家和官员宣传大幅增加烟草税好处的 一种方式。它是改善烟草税收政治经济的工具,是在减少烟草需求的主要目标之后的次 要问题。使用专项拨款来改善烟草税收的政治经济学的一种方法,是将烟草使用者的纳 税与他们通过资助戒烟,支持实现烟草控制计划,或通过增加对他们依赖的健康计划的资 助,使烟草税政策与他们获得的利益联系起来,这被称为利益原则。 为烟草控制划拨专项资金是有道理的,因为其财务成本相对较小,而且当在一揽子补 充性烟草控制措施中实施烟草税时,可以更有效地减少需求。指定用途改善烟草税收政 治经济的另一种方式,是防止税收本身产生任何可察觉的或潜在的负面影响。这对于消 除错误的但往往令人信服的烟草业反对有效烟草税收政策的论点很重要。例如,菲律宾 将罪恶税的大部分额外收入用于穷人的健康保险费。此外,该国的一部分烟草税收专门 用于为烟草种植者和烟草种植区提供经济福利,其总体目标是促进经济上可行的烟草种 植和制造替代品,以防止潜在的烟草种植和生产减少对从业者带来的冲击[141]。 然而,烟草税专项资金是复杂的,在确定一项特定的专项拨款是否需要基于政治经济 理由以及基于公平和经济效率的理由时需要谨慎。在评估烟草税专项拨款时,许多已用 于评估一般专项拨款适当性的相同标准也适用。在本节的其余部分,将列出这些标准,并 与精心设计的烟草税专项资金能够满足这些标准的原因相匹配。探讨了烟草税专项拨款 的类型和结构以及国家经验的描述,以就如何使用烟草税专项拨款、何时合理使用以及设 计它们的最佳方法提供指导。 4.6.2　可能适合使用专款专用的情况 对专款专用的怀疑由来已久,而且是有道理的,但大部分争论都集中在一般的专款专 用上,并没有特别关注烟草税专款专用的优点[138]。表 4.8 列出了关于专款专用的主要 关注点,并附有关于如何构建指定用途以解决这些问题的建议。 表4.8 对专款专用的担忧以及避免这些担忧的建议保障措施 专款专用的主要顾虑 如何设计专款专用 保障措施可以解决问题 民主问责和监督:专项拨款阻碍立法和行政 对支出的监督,从而破坏民主进程 建立适当的监督和问责程序对于确保资金不会管理不 善很重要[138]。此外,如果采用软专项结构,将收入转 移到普通基金,然后从中分配,这将不是问题 371第四章 烟草税收的政治经济学 续表 专款专用的主要顾虑 如何设计专款专用 保障措施可以解决问题 预算刚性: 指定用途可能会造成预算僵化,从而导致资 源分配效率低下[138] 特定用途的特殊设计决定了引入多少刚性[138]。柔性 软耳标比硬耳标更不容易引入刚性。加入日落条款可 以减少对僵化的担忧,该条款确保指定的时间段过去 后会自动中止或审查[138]。进一步的保护措施是将专 项资金设立为瀑布账户,任何超过设定金额的超额收 入都将分配给普通基金 碎片化: 指定用途可能导致支出分散和不协调。这 意味着与指定用途互补但超出其范围的政 策可能没有资金[138,142] 这是一个合理的担忧。碎片化的负面影响无法完全消 除,但它们可能会被烟草税专款专用的其他优点抵消。 也就是说,应仔细审查烟草税专项提案,以确保资助项 目至少具有成本效益 权益减少: 如果根据所支付的款项狭义地界定个人获 得福利的机会,公平就会减少 这个问题不太可能出现在烟草税专款上,但可以想象, 并且应该在专款设计中加以防范 按特殊利益获取: 因为专项拨款通常是政治权宜之计的结果, 指定用途可能由促进税收通过的强大特殊 利益集团决定,而不是对资源进行仔细优先 排序[138] 精心设计的专项专用将保证为资源不足的项目和高优 先级项目提供资金 虽然上述担忧可能是有效的,设计也很重要,但烟草税和其他促进健康的税收在收入 专款专用的合理性方面不受同样的关注[138,143]。表4.9[138]列出了区分烟草税专项拨款与 专项拨款的更普遍评论的一些因素。 表4.9 对烟草税专款专用和区分因素的担忧 专项拨款引起的普遍担忧 烟草税专款的区别因素 顺周期性: 专项收入通常是顺周期性的,容易受 到繁荣和萧条的影响[138,139,142] 烟草税收入通常不是周期性的(它们可以抵御经济衰退),并 且相对于大多数其他间接和直接税收而言,税收是可以预测 的[103] 预算刚性 烟草税专项资金只占预算的一小部分;因此,任何刚性的影响 都将相对微不足道。部分由于涉及的金额相对较小,只有有 限的现实证据表明烟草税专款专用引入了有害的刚性[143] 471 世界卫生组织烟草税政策和管理技术手册 另请参阅第 4.6.3 节中有关烟草税专项拨款金额的内容。 续表 专项拨款引起的普遍担忧 烟草税专款的区别因素 被特殊利益集团获取 应仔细审查所有专项拨款,以确保其资金用途具有成本效益。 然而,就烟草税专项拨款而言,政治经济方面的考虑可能意味 着,如果资金能够激发有效提高烟草税所需的政治意愿,则为 低优先级项目提供资金就足够了。在这些情况下,特殊利益 集团被有目的地迎合,以确保增加烟草税。当然,反对烟草税 和烟草税专款专用的争论是由特殊利益集团主导的,例如烟 草业[144147] 收入不足: 指定用途的收入来源可能不足以为 其目的提供资金[139,142] 尽管从长远来看,随着更多烟草使用者戒烟,税收可能会下 降,但预计这种下降将是渐进的。请参阅第 4.4 节详细了解 如何通过有效的设计,做到即使消费下降,税收也会普遍增加 除了这些关于专款专用的普遍担忧不能完全适用于精心设计的烟草税专项资金的原 因之外,还有许多令人信服的理由支持实施烟草税收专项拨款用于资助烟草控制或公共 卫生: • 大幅提高消费税是减少消费的最有效、最具成本效益的机制,但最好作为一揽子 补充烟草控制措施的一部分来实施,例如世卫组织 MPOWER 一揽子计划。将烟草税收 入指定用于可能无法以其他方式获得资金的干预措施,可以促进总体烟草需求减少[148]。 • 烟草税增加的政治经济学也使专项拨款具有吸引力: 事实证明,当人们知道收入将用于有针对性的社会项目时,他们会更加支持提高烟草 税[143,149,150]。将专项税收收入用于健康或烟草控制,将烟草税视为公众心目中的一种公 共卫生干预措施,否则公众可能仅仅将其视为一种收入来源[138]。研究表明,使用专项拨 款将烟草税与健康联系起来也有助于提高人们对吸烟危害的认识[143]。 当烟草税收专门用于惠及弱势群体的计划时,税收变得更加有利于公平。尽管较低 的社会经济群体和年轻人在中期从烟草税的增加中获得了不成比例的健康和经济利益, 但由于烟草税,这些群体将在短期内花费更多的收入。将烟草税收入指定用于为这些群 体提供直接利益的UHC或戒烟服务等方案,抵消了对烟草税的一些批评(例如,2009 年 美国联邦消费税增加和 2012 年菲律宾罪恶税改革,说明了提高公平性的指定用途如何 促进大幅增税方案的通过)[93,138,143,150]。 4.6.3　专款专用的做法和国家示例 将税收专用于卫生事业是80个国家的普遍做法。2018年,来自世界各地的37个国 家将烟草税收专用于健康目的。 571第四章 烟草税收的政治经济学 有关按国家/地区指定的专用税的详细信息,请访问 https://www.who.int/tobacco/global_report/Table 9 4 Use-of-earmarked-tobacco-taxes.xls? ua=1。 烟草税专款专用政治经济学案例研究 2012年,菲律宾全面改革烟酒消费税。烟草税大幅增加,到 2017 年,许多税级减少 到只有一级。尽管增加收入是一些官员的首要动机,但改革明确围绕通过减少烟酒消费 来增加全民健康覆盖资金和促进公共健康。将税收指定用于全民健康覆盖对于政治妥协 至关重要,这种妥协使这种开创性的增税成为现实。它确保了原本可能被视为倒退的增 长在公众的想象中被框定为进步的公共卫生措施,同时也安抚了烟草种植者及其政治代 表。指定用途也很重要,因为指定用途通过实现关键的政治优先事项确保了对税收的高 度支持[151]。它的软专项拨款结构意味着它不是卫生部的空白支票,这解决了财政部内 部的担忧。 同样,在澳大利亚,指定收入的用途有助于克服社区对烟草税和更普遍的烟草控制的 反对,这是由于 20世纪80年代烟草业赞助体育和艺术而引起的。由于来自体育、艺术和 赛车游说团体的强大压力,他们声称禁令会损害这些活动,因此完全禁止烟草广告和赞助 的尝试没有成功。从维多利亚州开始,各州的应对措施是为健康促进基金会提供专项资 金,这些基金会接管了烟草业的赞助活动,并为反吸烟运动提供资金。1997 年,在高等法 院裁定宪法不允许各州征收消费税后,这些专项拨款被终止。然而,由于对基金会的成功 工作的认可,联邦政府开始直接从联邦预算中为其提供资金[152]。虽然这个例子可能难 以精确复制,但它表明,一旦证明具有成本效益的试点干预措施有效,带有内置日落条款 的专项拨款就有可能逐步转变为来自一般预算的资金。它还更广泛地展示了专项资金如 何消除社区反对意见、重新调整烟草税的增加以及释放推进有效烟草控制措施所需的政 治意愿。 管理专项烟草税的结构 2016 年对9个国家烟草税收专款专用经历的审查确定了治理和分配收入的3种安 排[151]。表 4.10 列出了分配安排的一些例子。在一些国家,指定用途的烟草税收与酒精 税收相结合。 表4.10 分配专用烟草税收的说明性安排 确定烟草税专项收入的可能性 预 算 分 配形式 收入归普通基金,然后 分配给指定用途中指 定的官方参与者 收入不通过一般预算,而是 支付给属于专款中指定的 官方参与者的单独账户 专项税收直接支付给管理自主或 半自主基金的实体账户 案例 在菲律宾,收入先进入 普通基金,然后再提交 使用预算后分配给卫 生部[140] 在罗马尼亚,收入直接归入 不同于普通基金的卫生部 账户。在巴拿马,收入支付 给3个接收机构(卫生部、 国家癌症研究所和海关部 门)的子账户中 在泰 国,泰 健 康 基 金 会 (Thai Health) 直接从自己的账户中收 到专项收入 在越南,越南烟草控制基金将收 入直接存入其管理但属于卫生部 的子账户 来源:(151). 671 世界卫生组织烟草税政策和管理技术手册 钱花在哪里? 专项烟草税收入用于多种健康目的,包括烟草控制、健康促进和全民健康覆盖。许多 其他项目也得到了烟草税专项资金的资助,包括救灾(例如,印度用于治疗 COVID 19 的医院医疗用品和设备),也门的青年项目、体育和手工艺工作,摩洛哥的社会融合,阿根 廷依赖烟草种植地区的健康和社会计划,泰国的健康促进和烟草控制、针对烟农的替代生 计计划以及菲律宾烟草种植省份的经济项目。 表4.11显示了专项烟草税收分配到的3个主要类别的健康计划,以及第四个杂项类 别,每个类别都有特定国家的示例。 表4.11 指定烟草税收入分配的项目 烟草控制 NCD 预防和控制计 划(括号内另有说明) 医疗保险扩大(例 如通过医疗保险) 其他更普遍或未具体说明的保健方案 哥 斯 达 黎 加、科特迪 瓦、伊 朗、 马 达 加 斯 加、巴拿马 (戒烟和打 击 非 法 贸 易)、瑞士、 越南 库克群岛、哥斯达黎 加、毛里塔尼亚(抗癌 研究)、帕劳(仅预防 非传染性疾病)、巴拿 马 (国 家 肿 瘤 研 究 所)、巴拉圭 哥伦比亚、刚果、 埃及、帕劳、菲律 宾 阿尔及利亚、阿根廷、孟加拉国、博茨瓦 纳、佛得角、乍得(提供抗逆转录病毒药物 的方案)、哥伦比亚(体育)、科摩罗(体育、 医院紧急情况)、刚果(体育)、科特迪瓦 (艾滋病计划)、萨尔瓦多、爱沙尼亚 (体 育)、危地马拉、印度尼西亚、伊朗伊斯兰 共和国(体育)、爱尔兰、牙买加、立陶宛 (体育)、马达加斯加(体育)、摩洛哥、尼泊 尔、巴拉圭(体育)、(促进健康)、罗马尼 亚、泰国(促进健康)、美国、也门(体育) 注:当国家指定的税收用于多个特定的卫生项目时,它们会出现在不止一列中。 来源:参考文献27。 与烟草税专款相关的金额 9个国家在烟草税专款专用方面的经验案例研究表明,与政府卫生支出相比,专款专 用资金相对较少(见表 4.12),因此,就 GDP 而言甚至更小。因此,烟草税专项拨款会给 公共财政管理带来僵化的论点可能不适用。 表4.12 烟草税专项资金在政府支出中的比例 国家 来自专项税的估计年度总资金 来自烟草税专项资金的年度资金占 2013 年政府卫生总支出的百分比 博茨瓦纳 2014—2015 年:400 万普拉(48 万美元) NA 埃及 2013—2014年:3.92亿埃及镑(5 206万美元), 专项税仅占卷烟总税的 1.8% 1.086% 冰岛 2014 年:1.083 亿克朗(89 万美元) 0.083% 771第四章 烟草税收的政治经济学 附件 4.2 提供了有关如何指定烟草税收入的详细信息。 续表 国家 来自专项税的估计年度总资金 来自烟草税专项资金的年度资金占 2013 年政府卫生总支出的百分比 巴拿马 2014年:2 780万美元 1.322% 菲律宾 2014年:501.8亿菲律宾比索(11.8亿美元) NA 波兰 2013 年:来自一般预算的 100 万兹罗提(31.6 万 美元) 0.001% 罗马尼亚 2014 年:110 万列伊(33 万美元),占卫生总预算 的 14.4% 0.004% 泰国 2014年:40.647 4亿泰铢(1.251 5亿美元),占卫 生部预算的 1.78% 和国家健康保障基金的 1.84% 0.932% 越南 2014年:2 991.71亿越南盾(1 391万美元),占国 家卫生预算的 0.5% 0.335% 来源:(151). 4.6.4　总结 尽管一些财政部门最初原则上反对专款专用,但经验表明,将烟草税和其他税收收入 用于消费具有负外部性的产品可以确保政治和公众支持。成功的专款专用需要一个完善 的结构来将资金用于健康目的。即使是反对专款专用的政府间组织(例如国际货币基金 组织)也承认,当收入用于特定的具有成本效益的计划时,精心设计的烟草税专款是合理 的[153,154]。 有效指定用于健康的烟草税收数额相对较小,很难在政府预算中引入令人担忧的僵 化。此外,在一些国家,这些资金帮助实施了急需的卫生计划(例如澳大利亚、菲律宾、泰 国)。越来越多的政府正在考虑将此选项作为稳定的中期可靠资金来源,用于烟草控制等 项目。随着患病人数减少和对烟草相关疾病的医疗保健需求减少,未来将看到回报。在 澳大利亚,一项专项税用于资助一项被证明是成功、有效和有影响力的需要且资源不足的 计划,该计划现在得到了可持续的资助,并纳入了联邦预算。 当特定的政治经济能够实施有效的烟草税以提高价格并减少消费时,专项拨款是可 取的。然而,当考虑到以下问题时,它在公共财政管理、经济效率和民主治理方面也将是 合理的。尽管并非每个问题都需要肯定的答案,但对以下许多问题的回答都是回答“是” 的政策制定者,可能会被认为是有效和合理的烟草税专项拨款: • 烟草税专款的目的是否与受援国方案的目的合理联系起来? 在利益原则下,资助 烟草控制或其他健康计划的专项资金比资助诸如儿童教育等无关项目的资金在经济上更 合理,即使无关项目的受欢迎程度可能使增税在政治上变得合情合理。 871 世界卫生组织烟草税政策和管理技术手册 改编自参考文献 138 139,143,150。 • 烟草税专项资金的数额是否与受援计划的需求合理联系起来? 不能被受助计划 吸收的专用资金实际上是从其他需求中扣除的资金。 • 当烟草税专项资金用于资助健康计划时,是否向公众明确传达了这一点,以确保 将增加烟草税作为一项健康措施来加强减少烟草需求的效果? • 由烟草税专款资助的计划是否是一项在政治上被忽视但极具成本效益或急需的 计划,一旦确定为概念证明,就有机会从总预算中获得资助? • 烟草税专项资金的目的是否与税收本身的效果有合理的联系? 为更多地造福于 社会经济地位较低的群体或为前烟草工人和农民的替代生计提供专项拨款将产生促进公 平的效果,这将加强烟草税已经具有的累进性质。 • 专项资金的用途范围是否足够窄,可以主要由烟草税专项资金提供资金,以确保 收入是可增加的,而不仅仅是替代原本来自普通基金的支出? • 烟草税专款的设计是否提供了灵活性,以确保意外收入的征收不会浪费在已经充 斥着过度资金的目的上? • 烟草税专款的设计是否包括触发其自动终止或审查的日落条款? 这样一个条款 的存在将防止由于政治惰性而导致分配效率低下和得不到纠正的情况。 971第四章 烟草税收的政治经济学 本章参考文献 1.Smith KE,Savel E,Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tob Control. 2013;22(2):144-153 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 31 January 2021). 2.LaFaive MD,Drenkard S,Nesbit T. Strong link between cigarette tax and ilegal smuggling rates [website]. Midland:Mackinak Center for Public Policy;2018 (https://www. mackinac.org/strong-link-between-cigarette-tax-and-ilegal-smuggling-rates#:~:text= An%20increase%20of%2011%20 percentage,packs%20to%2030.6%20milion%20packs. &text=Tax%20evasion%20and%20avoidance%20 is,high%20excise%20taxes%20on% 20cigarettes.,accessed 31 January 2021). 3.Ross H,Blecher E. Ilicit trade in tobacco products need not hinder tobacco tax policy reforms and increases. Chicago:University of Ilinois at Chicago;2019 (Tobacconomics White Paper;https:// tobacconomics.org/research/ilicit-trade-in-tobacco-products-need- not-hinder-tobacco-tax-policy-reforms-and-increases-white-paper/#:~:text=additional% 20related%20research-,Ilicit%20 Trade%20in%20Tobacco%20Products%20Need% 20Not%20Hinder%20Tobacco,Reforms%20 and%20Increases%20( White%20Paper) &text= The%20paper%20cites%20research%20 that,consumption%2C%20and% 20improves%20overal%20health,accessed 31 January 2021). 4.Confronting ilicit tobacco trade:a global review of country experiences. Washington (DC): World Bank Group; 2019 (http://documents1. worldbank. org/curated/en/ 677451548260528135/pdf/133959-REPL-PUBLIC-6-2-2019-19-59-24- WBGTobaccoIlicitTradeFINALvweb.pdf,accessed 12 October 2020). 5.The economics of tobacco and tobacco control. Bethesda:US Department of Health and Human Services,National Institutes of Health,National Cancer Institute;and Geneva:World Health Organization;2017 (National Cancer Institute Tobacco Control Monograph 21; https://cancercontrol.cancer.gov/sites/ default/files/2020-06/m21_complete.pdf,accessed 28 September 2020). 6.Joossens L,Chaloupka FJ.,Merriman D,Yürekli A. Issues in smuggling of tobacco products. In:Jha P,Chaloupka F,editors. Tobacco control in developing countries. Oxford: Oxford University Press;2000 (http://documents1.worldbank.org/curated/en/ 602821468330954036/pdf/709670WP0tobac00Box 370064B00PUBLIC0.pdf,accessed 18 February 2021). 7.Merriman D,Yürekli A.,Chaloupka,FJ. How big is the worldwide cigarette smuggling problem. In:Jha P,Chaloupka F,editors. Tobacco control in developing countries. Oxford: Oxford University Press;2000 (http://documents1.worldbank.org/curated/en/ 602821468330954036/pdf/709670WP0tobac 00Box370064B00PUBLIC0.pdf,accessed 18 081 世界卫生组织烟草税政策和管理技术手册 February 2021). 8.WHO Framework Convention on Tobacco Control. Geneva:World Health Organization; 2003 (https:// apps.who.int/iris/bitstream/handle/10665/42811/9241591013.pdf; jsessionid=B6558076A38E8DEB 6D3A9E6CDA96151B? sequence=1,accessed 5 October 2020). 9.Merriman D. Understand,measure,and combat tobacco smuggling. Washington (DC): World Bank Group;2003 (Economics of tobacco toolkit 7;https://openknowledge. worldbank.org/bitstream/ handle/10986/16267/805800NWP070Sm0Box0379807B00PUBLIC0. pdf? sequence=1&isAlowed=y,accessed 31 January 2021). 10.Progress report on the implementation of the Commission Communication. Stepping up the fight against cigarette smuggling and other forms of ilicit trade in tobacco products -a comprehensive EU strategy. Brussels:European Commission;2017 (COM(2017) 235 final; https://eur-lex.europa.eu/ legal-content/EN/TXT/PDF/? uri=CELEX:52017DC0235&from =en,accessed 31 January 2021). 11.Ross H. Understanding and measuring cigarette tax avoidance and evasion:a methodological guide. Tobacconomics;2015 (https://tobacconomics.org/wp-content/ uploads/2015/03/Ross_Methods_to_ Measure_Ilicit-Trade_03-17-15.pdf,accessed 12 October 2020). 12.Understanding the US ilicit tobacco market:characteristics,policy context,and lessons from international experiences. Reuter P,Majmundar M. (editors). Washington (DC): National Academies Press;2015. 13.von Lampe K. The ilegal cigarette trade. In:Natarajan M,editor. International crime and justice. New York:Cambridge University Press. 2010:148-154. 14.Joossens L,Raw M,From cigarette smuggling to ilicit tobacco trade. Tob Control. 2012; 21:230-34 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/2/230.ful.pdf, accessed 31 January 2021). 15.Gomis B,Lee K,Carrilo Botero N,Shepherd P,Iglesias RM. “We think globaly”:the rise of Paraguay􀆳s Tabacalera del Este as a threat to global tobacco control. Global Health. 2018;14(1):110 (https:// globalizationandhealth.biomedcentral.com/articles/10.1186/ s12992-018-0412-3,accessed 31 January 2021). 16.Communication from the commission to the Council and the European Parliament:Stepping up the fight against cigarette smuggling and other forms of ilicit trade in tobacco products - A comprehensive EU Strategy. Brussels:European Commission;2013 (https://eur-lex. europa.eu/legal-content/EN/TXT/ PDF/? uri=CELEX:52013DC0324&from=EN, accessed 31 January 2021). 17.Ross H,Velios N,Smith KC,Ferguson J,Cohen JE. A closer look at ‘cheap white􀆳 cigarettes. Tob Control. 2016;25(5):527-31 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/25/5/527.ful.pdf,accessed 31 January 2021). 18.Nagelhout GE,van den Putte B,Alwright S,Mons U,McNeil A,Guignard R,et al. 181第四章 烟草税收的政治经济学 Socioeconomic and country variations in cross-border cigarette purchasing as tobacco tax avoidance strategy. Findings from the ITC Europe Surveys. Tob Control. 2014;23:i30-i38 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/23/suppl_1/i30.ful.pdf, accessed 1 February 2021). 19.Gilmore AB,Rowel A,Galus S,Lugo A,Joossens L,Sims M. Towards a greater understanding of the ilicit tobacco trade in Europe:a review of the PMI funded ‘Project Star􀆳 report. Tob Control. 2014;23:e51-e61 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/23/e1/e51.ful.pdf,accessed 1 February 2021). 20.Iglesias RM. Increasing excise taxes in the presence of an ilegal cigarette market:the 2011 Brazil tobacco tax reform. Revista Panam Salud Publica. 2016;40(4):243-9 (https:// www.scielosp.org/pdf/ rpsp/2016.v40n4/243-249,accessed 1 February 2021). 21.Operaçāo Huno combate mercado clandestino de cigarros em sete Estados [Operaçāo Huno fights clandestine cigar market in seven states]. Brasilia:Federal Police of Brazil;2015 (in Portuguese) (http:// www.pf.gov.br/agencia/noticias/2015/11/operacao-huno-combate- mercado-clandestino-de-cigarros-em-seis-estados,accessed 12 October 2020). 22.Gilmore AB,Galagher AWA,Rowel A. Tobacco industry􀆳s elaborate attempts to control a global track and trace system and fundamentaly undermine the Ilicit Trade Protocol. Tob Control. 2019;28:127-40 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/28/ 2/127.ful.pdf,accessed 1 February 2021). 23.Technical assessment of the experience made with the Anti-Contraband and Anti- Counterfeit Agreement and general release of 9 July 2014 among Philip Morris International and affiliates,the Union and its Member States. Brussels:European Commission;2016 (SWD (2016) 44 final;https://ec.europa.eu/anti-fraud/sites/antifraud/files/technical_ assessment_pmi_24022016_en.pdf,accessed 1 February 2021). 24.Galagher AWA,Evans-Reeves KA,Hatchard JL,Gilmore AB. Tobacco industry data on ilicit tobacco trade:a systematic review of existing assessments. Tob Control. 2019;28 (3):334-45 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/28/3/334.ful. pdf,accessed 1 February 2021). 25.Joossens L,Merriman D,Ross H,Raw M. How eliminating the global ilicit cigarette trade would increase tax revenue and save lives. Paris:International Union Against Tuberculosis and Lung Disease;2009 (https://theunion.org/sites/default/files/2020-08/How% 20Eliminating% 20the% 20Global% 20Ilicit% 20 Cigarette% 20Trade% 20Would% 20Increase%20Tax%20Revenue%20and%20Save%20Lives%20-%20 EN.pdf,accessed 1 February 2021). 26.Protocol to Eliminate Ilicit Trade in Tobacco Products. Geneva:World Health Organization; 2013 (https://apps. who.int/iris/bitstream/handle/10665/80873/ 9789241505246_eng.pdf? sequence=1,accessed 5 October 2020). 27.WHO report on the global tobacco epidemic 2019:offer help to quit tobacco use. Geneva: World Health Organization;2019 (https://www.who.int/teams/health-promotion/ 281 世界卫生组织烟草税政策和管理技术手册 tobacco-control/who-report-on-the-global-tobacco-epidemic-2019,accessed 24 January 2021). 28.Prieger JE,Kulic J. Cigarette taxes and ilicit trade in Europe. Econ Inquiry. 2018;56(3): 1706-23 (https:// papers.ssrn.com/sol3/papers.cfm? abstract_id=2718519,accessed 18 February 2021). 29.Stoklosa M. Prices and cross-border cigarette purchases in the EU:evidence from demand modeling. Tob Control. 2020;29:55-60. 30.Passport:ilicit trade in tobacco products. In:Euromonitor International [website]. London:Euromonitor International;2018 (restricted access). 31.The Global Ilicit Trade Environment Index. London:The Economist Inteligence Unit; 2018 (http:// ilicittradeindex.eiu.com/documents/EIU%20Global%20Ilicit%20Trade% 20Environment%20Index%20 2018%20-%20Overal%20Results%20White%20Paper% 20June%206%20FINAL.pdf,accessed 1 February 2021). 32.Chaloupka FJ,Yürekli A,Fong GT. Tobacco taxes as a tobacco control strategy. Tob Control. 2012;21:172-180 (https://tobaccocontrol.bmj.com/content/tobaccocontrol/21/ 2/172.ful.pdf,accessed 1 February 2021). 33.Chernick H,Merriman D. Using littered pack data to estimate cigarette tax avoidance in NYC. Natl Tax J. 2013;66(2):635-668 (https://www.ntanet.org/NTJ/66/3/ntj- v66n03p635-668-littered-pack-data-cigarette-tax.pdf,accessed 1 February 2021). 34.Liber AC,Ross H,Omar M,Chaloupka FJ. The impact of the Malaysian minimum cigarette price law:findings from the ITC Malaysia Survey. Tob Control. 2015;24:ii83- ii87 (https://tobaccocontrol.bmj. com/content/tobaccocontrol/24/Suppl_3/ii83.ful. pdf,accessed 1 February 2021). 35.Blecher E,Liber A,Ross H,Birckmayer J. Euromonitor data on the ilicit trade in cigarettes. Tob Control. 2015;24:100-1 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/24/1/100.ful.pdf,accessed 1 February 2021). 36.Stoklosa M,Ross H. Contrasting academic and tobacco industry estimates of ilicit cigarette trade:evidence from Warsaw,Poland. Tob Control. 2014;23(e1),e30-e34 (https://www. researchgate.net/ publication/255954649_Contrasting_academic_and_tobacco_industry_ estimates_of_ilicit_cigarette_ trade_Evidence_from_Warsaw_Poland,accessed 1 February 2021). 37.Maldonado N,Llorente BA,Iglesias RM,Escobar D. Measuring ilicit cigarette trade in Colombia. Tob Control. 2020;29:s260-s266 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/29/Suppl_4/ s260.ful.pdf,accessed 29 January 2021). 38.Paraje G,Araya D,Drope J. Ilicit cigarette trade in metropolitan Santiago de Chile. Tob Control 2020;29:68-73. 39.Iglesias RM,Szklo AS,de Souza MC,de Almeida LM. Estimating the size of ilicit tobacco consumption in Brazil:findings from the global adult tobacco survey. Tob Control. 2017; 26(1):53-59. 381第四章 烟草税收的政治经济学 40.Saenz de Miera Juarez BS,Reynales-Shigematsu LM,Stoklosa M,Welding K,Drope J. Measuring the ilicit cigarette market in Mexico:a cross validation of two methodologies. Tob Control;2020(0):1-7 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/ early/2020/03/31/tobaccocontrol-2019-055449. ful.pdf,accessed 1 February 2021). 41.Lee K,Colin J. ‘Key to the future􀆳:British American Tobacco and cigarette smuggling in China. PLoS Medicine. 2006;3(7):e228 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC1502159/pdf/pmed.0030228. pdf,accessed 2 February 2021). 42.Barker DC,Wang S,Merriman D,Crosby A.,Resnick EA,Chaloupka FJ. Estimating cigarette tax avoidance and evasion:evidence from a national sample of littered packs. Tob Control. 2016;25 (Suppl 1):i38-i43 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/25/Suppl_1/i38.ful.pdf,accessed 2 February 2021). 43.Aziani A,Kulick J,Norman N,Prieger JE. Empty discarded pack data and the prevalence of ilicit trade in cigarettes;2017. (https://ssrn.com/abstract=2906015,accessed 12 October 2020). 44.Philip Morris International. The Perspective of PM International on Smoking and Health Issues. In:Legacy Tobacco Documents Library [website]. San Francisco:University of California;1985. Bates no. 2023268328-2023268349 (https://www.industrydocuments. ucsf.edu/tobacco/docs/#id=pxpb0040,accessed 5 February 2021). 45.Mandal S,Gilmore A,Colin J,Weishaar H,Smith K,McKee M. Block,amend,delay: tobacco industry efforts to influence the European Union􀆳s Tobacco Products Directive (2001/37/EC). Smokefree Partnership/Cancer Research UK;2009 (http://www. smokefreepartnership.eu/images/Reports-and-Position-Papers/EU_TI_TPD_report_May_ 2012.pdf,accessed 15 October 2020). 46.Bishop DJ. Key area paper:indirect taxation -British American Tobacco. In:Legacy Tobacco Documents Library [website]. San Francisco:University of California;1994. Bates no. 503911771-503911780 (http:// bat.library.ucsf.edu/data/h/f/k/hfk11a99/hfk11a99. pdf,accessed 15 October 2020). 47.Zhou S,Liberman JD,Ricafort E. The impact of the WHO Framework Convention on Tobacco Control in defending legal chalenges to tobacco control measures. Tob Control. 2009:28:s113-s118 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/28/Suppl_ 2/s113.ful.pdf,accessed 2 February 2021). 48.Gilmore AB,Fooks G,Drope J,Bialous SA,Jackson RR. Exposing and addressing tobacco industry conduct in low-income and middle-income countries. Lancet. 2015;385:1029- 1043. 49.Ulucanlar S,Fooks G,Gilmore A. The policy dystopia model:an interpretive analysis of tobacco industry political activity. PLoS Med. 2016:13(9):e1002125 (https://journals. plos.org/plosmedicine/ article?id=10.1371/journal.pmed.1002125,accessed 2 February 2021). 50.Apolonio D,Glantz S. Tobacco manufacturer lobbying to undercut minimum price laws:an 481 世界卫生组织烟草税政策和管理技术手册 analysis of internal industry documents. Tob Control. 2020;29:e10-e17 (https://www. ncbi.nlm.nih.gov/pmc/ articles/PMC7374022/pdf/nihms-1590806.pdf,accessed 2 February 2021). 51.Crosbie E,Glantz S. Tobacco industry argues domestic trademark laws and international treaties preclude cigarette health warning labels,despite consistent legal advice that the argument is invalid. Tob Control. 2012;23:e7 (https://www.researchgate.net/ publication/233769722_Tobacco_industry_ argues_domestic_trademark_laws_and_ international_treaties_preclude_cigarette_health_warning_ labels_despite_consistent_legal_ advice_that_the_argument_is_invalid,accessed 2 February 2021). 52.Puig S. Tobacco litigation in international courts. Harv Int Law J. 2016;57(2):1-68. 53.Crooked nine:nine ways the tobacco industry undermines health policy. New York:Vital Strategies;2019 (https://exposetobacco.org/wp-content/uploads/2019/09/Crooked-9- STOP.pdf,accessed 2 February 2021). 54.Balwicki L,Stokłosa M,Balwicka-Szczyrba M,Tomczak W. Tobacco industry interference with tobacco control policies in Poland:legal aspects and industry practices. Tob Control. 2016;25(5):521-526 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/25/5/ 521.ful.pdf,accessed 2 February 2021). 55.Sud SR,Brenner JE,Shaffer ER. Trading away health:the influence of trade policy on youth tobacco control. J Pediatrics. 2015;166(5):1303-07 (https://www.jpeds.com/action/ showPdf? pi=S0022-3476%2815%2900017-7,accessed 2 February 2021). 56.Lee S,Ling PM,Glantz SA. The vector of the tobacco epidemic:tobacco industry practices in low and middle-income countries. Cancer Cause Control. 2012;23:117-29 (https:// www.ncbi.nlm.nih.gov/ pmc/articles/PMC3332051/pdf/nihms-367996.pdf,accessed 2 February 2021). 57.Zhou S,Liberman J. The global tobacco epidemic and the WHO Framework Convention on Tobacco Control—the contributions of the WHO􀆳s first convention to global health law and governance. In:Burci GL,Toebes B,editors. Research handbook on global health law. Cheltenham:Edward Elgar Publishing;2018 (https://papers.ssrn.com/sol3/papers.cfm? abstract_id=3339718,accessed 18 February 2021). 58.Legislative Consultation with Constitutional Division of the Supreme Court,No. 2012- 003918. Constitutional Division of the Supreme Court of Costa Rica;20 March 2012 (https://www. tobaccocontrolaws.org/files/live/litigation/830/CR _ Legislative% 20Consultation%20with%20.pdf,accessed 2 February 2021). 59.British American Tobacco Kenya Ltd v Ministry of Health,Petition No. 143 of 2015. High Court of Kenya;2016 (https://www.tobaccocontrolaws.org/files/live/litigation/2643/KE _British%20American%20 Tobacco%20Kenya.pdf,accessed 2 February 2021). 60.British American Tobacco v Ministry of Health,et al,Civil Appeal No. 112 of 2016. Kenyan Court of Appeal;2017 (https://www.tobaccocontrolaws.org/files/live/litigation/ 2521/KE_British%20 American%20Tobacco%20Ltd%20v.pdf,accessed 2 February 581第四章 烟草税收的政治经济学 2021). 61.Thailand -customs and fiscal measures on cigarettes from the Philippines:Report of the Panel. Geneva:World Trade Organization;2010 (WT/DS371/R;https://docs.wto.org/ dol2fe/Pages/SS/directdoc. aspx? filename= Q:/WT/DS/371R.pdf&Open= True, accessed 2 February 2021). 62.Tribunal Constitucional de Chile [Constitutional Court of Chile] (1995 Rol No. 219-95; https://www. tribunalconstitucional.cl/descargar_sentencia2.php? id=3074,accessed 4 April 2020.) 63.McGrady B. International investment law. In:Voon T,Mitchel AD,Liberman J,editors. Regulating tobacco,alcohol and unhealthy foods:the legal issues. New York:Routledge; 2014. 64.Marvin Roy Feldman Karpa v United Mexican States,Case No. ARB(AF)/99/1 International Centre for Settlement of Investment Disputes;2002 (https://www.italaw. com/sites/default/files/case-documents/ ita0319.pdf,accessed 2 February 2021). 65.Feldman Karpa v Mexico (2002 ICSID Arbitral Tribunal) (https://www.italaw.com/ sites/default/files/ case-documents/ita0319.pdf,accessed 18 February 2021). 66.Undang-Undang Nomor 11 Tahun 1995 tentangCukai (Lembaran Negara Republik Indonesia Tahun1995 Nomor 76) (Excise Law,in Indonesian). Article 5. 67.Excise Regulation (2011) Reference No. PMK Number 167/PMK.011/2011 (in Indonesian) (https:// peraturan. bpk. go.id/Home/Details/145897/pmk-no- 167pmk0112011,accessed 3 February 2021). 68.Mahkamah Agung. Indonesia Supreme Court,Case No. 73/PDT/2018/PT SBY (https:// putusan3. mahkamahagung.go.id/direktori/putusan/74efff9cfe18d60990fcde19583ecd5f. html,accessed 6 February 2021). 69.Maya D. Analisis kebijakan kenaikam tarif cukai dan penyesuaian batasan harga jual eceran hasil tembakau [Policy analysis on excise tax increases and price limit adjustments retail tobacco products];2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-09/ S45937-Deacy%20Maya accessed 3 March 2020). 70.Indriawati EN,Susanti L,Qurnia MA. Kebijakan tarif cukai hasil tembakau terhadap perekonomian [Tobacco excise tariff policy on the economy];2014 (in Indonesian). (https://www.academia. edu/6611977/MAKALAH_CHT_kel_6,accessed 15 October 2020). 71.Arroyani DW. Evaluasi Kebijakan Penetapan Tarif Cukai Hasil Tembakau Berdasarkan PMK [Evaluation of Tobacco Excise Tariff Determination Policy Based on PMK].167/ PMK.011/2011. 2013 (in Indonesian) (http://lib.ui.ac.id/naskahringkas/2015-08/ S44664-Dwira%20Wanti%20Arroyan,accessed 15 October 2020). 72.Ngo Ngo Ha and Another v. State of New South Wales and Others (Matter No S 45 of 1996);Walter Hammond & Associates Pty Ltd v. State of New South Wales and Others (Matter No S 165 of 1996). Constitutional law;(1997) 146 ALR 355:5 August 1997 681 世界卫生组织烟草税政策和管理技术手册 (https://staging.hcourt.gov.au/assets/ publications/judgments/1997/029--NGO_NGO_ HA_AND_ANOR_v_STATE_OF_NEW_SOUTH_WALES_&_ ORS_(Matter_No_S_45_ of_1996);_WALTER_HAMMOND--(1997)_146_ALR_355.html,accessed 2 February 2021). 73.McGrady B. Trade and public health:the WTO,tobacco,alcohol,and diet. New York: Cambridge University Press;2011. 74.Giorgetti C. Health and international investment law. In:Burci GL,Toebes B,editors. Research handbook on global health law. Cheltenham:Edward Elgar Publishing;2018. 75.British American Tobacco v Attorney General and Center for Human Rights and Development. Constitutional Court of Uganda Petition No.46 of 2016;2017 (https:// media.uli.org/files/judgments/ ugcc/2019/4/2019-ugcc-4_0.pdf,accessed 18 February 2021). 76.Lim CL,Ho J,Paparinskis M. International investment law and arbitration:commentary, awards and other materials. Cambridge:Cambridge University Press;2018. 77.Dominican Republic -measures affecting the importation and internal sale of cigarettes : Report of the Panel. Geneva:World Trade Organization;2004 (WT/DS302/R;https:// docs.wto.org/dol2fe/ Pages/SS/directdoc.aspx?filename=Q:/WT/DS/302R.pdf&Open =True,accessed 21 February 2021) 78.Dominican Republic — measures affecting the importation and internal sale of cigarettes: Report of the Appelate Body. Geneva:World Trade Organization;2005 (WT/DS302/AB/ R;https://docs.wto. org/dol2fe/Pages/SS/directdoc.aspx? filename= Q:/WT/DS/ 302ABR.pdf&Open=True,accessed 21 February 2021). 79.Gilmore A. Tobacco and transition:understanding the impact of transition on tobacco use and control in the former Soviet Union [Thesis]. London:London School of Hygiene and Tropical Medicine,June 2005 (https://escholarship.org/uc/item/3rw2c04c,accessed 2 February 2021). 80.MacKenzie R,Colin J,Sopharo C,Sopheap Y. ‘Almost a role model of what we would like to do everywhere􀆳:British American Tobacco in Cambodia. Tob Control. 2004;13(suppl II):i112-i117 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/13/suppl_2/ i112.ful.pdf,accessed 2 February 2021). 81.Smith KE,Savel E,Gilmore AB. What is known about tobacco industry efforts to influence tobacco tax? A systematic review of empirical studies. Tobacco Control 2013;22(2):144-53 (https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3701860/pdf/nihms481283.pdf, accessed 21 February 2021). 82.Price and tax. In:TobaccoTactics [website]. Bath:University of Bath;2020 (https:// tobaccotactics.org/ topics/price-and-tax/,accessed 15 December 2020). 83.Vulovic V,Chaloupka FJ. Questioning the regressivity of tobacco taxes:a distributional accounting impact model of increased tobacco taxation—commentary. Tob Control. 24 June 2020. 781第四章 烟草税收的政治经济学 84.Verguet S,Kearns PKA,Rees VW. Questioning the regressivity of tobacco taxes:a distributional accounting impact model of increased tobacco taxation. Tob Control. 23 June 2020 (https:// tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/08/11/ tobaccocontrol-2019-055315. ful.pdf,accessed 2 February 2021). 85.Palipudi KM,Gupta PC,Sinha DN,Andes LJ,Asma S,McAfee T. Social determinants of health and tobacco use in thirteen low and middle income countries:evidence from Global Adult Tobacco Survey. PLoS ONE. 2012;7(3):e33466 (https://www.ncbi.nlm.nih. gov/pmc/articles/PMC3306395/pdf/ pone.0033466.pdf,accessed 2 February 2021). 86.Systematic review of the link between tobacco and poverty. Geneva:World Health Organization; 2014 (http://www. who. int/tobacco/publications/economics/ 9789241507820/en,accessed 21 October 2020). 87.Goodchild M,Sinha P,Gil Munish V,Tulu FT. Changes in the affordability of tobacco products in India during 2007/2008 to 2017/2018:a price-relative-to-income analysis. WHO South-East Asia J Public Health. 2020;9(1):73-81 (https://apps.who.int/iris/bitstream/ handle/10665/331836/whoseajph2020v9i1p73. pdf? sequence=1&isAlowed=y,accessed 2 February 2021). 88.Faruque GM,Ahmed M,Huq I,Parven R,Wadood SN,Chowdhury SR,et al. The economic cost of tobacco use in Bangladesh:a health cost approach. Dhaka:Bangladesh Cancer Society;2020 (https://www. researchgate.net/publication/342445467_The_Economic_ Cost_of_Tobacco_Use_in_Bangladesh_A_ Health_Cost_Approach,accessed 2 February 2021). 89.John RM,Sung H,Max WB,Ross H. Counting 15 milion more poor in India,thanks to tobacco. Tob Control. 2011;20:349-352 (https://www.researchgate.net/publication/ 49809290_Counting_15_milion_ more_poor_in_India_thanks_to_tobacco,accessed 2 February 2021). 90.Liu Y,Rao K,Hu TW,Sun Q,Mao Z. Cigarette smoking and poverty in China. Soc Sci Med. 2006;Dec;63(11):2784-2790. 91.Fuchs A,González Icaza F,Paz D. Distributional effects of tobacco taxation -a comparative analysis. Washington (DC):World Bank Group;2019 (Policy research working paper 8805; https://openknowledge. worldbank.org/bitstream/handle/10986/31534/WPS8805.pdf? sequence=4&isAlowed=y, accessed 2 February 2021). 92.Smith CE,Hil SE,Amos A. Impact of population tobacco control interventions on socioeconomic inequalities in smoking:a systematic review and appraisal of future research directions. Tob Control. 29 September 2020 (https://tobaccocontrol.bmj.com/content/ tobaccocontrol/early/2020/09/28/ tobaccocontrol-2020-055874.ful.pdf,accessed 2 February 2021). 93.Furman J. Six lessons from the U.S. experience with tobacco taxes -World Bank conference “Winning the tax wars:global solutions for developing countries”. Washington (DC):Obama White House Archives;24 May 2016 (https://obamawhitehouse.archives. 881 世界卫生组织烟草税政策和管理技术手册 gov/sites/default/files/page/files/20160524_cea_ tobacco_tax_speech.pdf,accessed 21 October 2020). 94.Önder Z,Yürekli A. Who pays the most cigarette tax in Turkey. Tob Control. 2016; 25(1):39-45. 95.Verguet S,Gauvreau CL,Mishra S,MacLennan M,Murphy SM,Brouwer ED,et al. The consequences of tobacco tax on household health and finances in rich and poor smokers in China:an extended cost-effectiveness analysis. Lancet Glob Health. 2015;3:e206-16 (https://www.thelancet.com/action/ showPdf? pi=S2214-109X%2815%2970095-1, accessed 2 February 2021). 96.Fuchs Tarlovsky A,González Icaza F. The welfare and distributional effects of increasing taxes on tobacco in Vietnam. Washington (DC):World Bank Group;2019 (http:// documents1.worldbank.org/ curated/en/758391561477648302/pdf/The-Welfare-and- Distributional-Effects-of-Increasing-Taxes-on-Tobacco-in-Vietnam.pdf,accessed 2 February 2021). 97.Expanding the global tax base:taxing to promote public goods:tobacco taxes. Washington (DC): World Bank Group;2016 (http://pubdocs.worldbank.org/en/799981480947591462/ Taxing-to-Promote-Public-Goods-tobacco-taxation-chapter-conference-25-nov-2016-final- version-pmarquez-002.pdf,accessed 2 February 2021). 98.Blecher E. Increasing tobacco taxes significantly wil increase revenues. Chicago:University of Ilinois at Chicago;2018. (Tobacconomics policy brief;https://tobacconomics.org/files/ research/470/Laffer-Curve-Policy-Brief_v2.2.pdf,accessed 2 February 2021). 99.Savedoff W,Lopert R. Sugar,rum,and tobacco:domestic resource mobilization for low- income countries through excise taxes. Washington (DC):Centre for Global Development; 2020 (CGD Policy Paper 164,https://www.cgdev.org/sites/default/files/PP164-Savedoff- Health-Taxes-DRM.pdf,accessed 2 February 2021). 100.Lavado R,Postolovska I,Sotomayor RE. Beyond the perceived regressivity of higher tobacco taxes,turning short-term losers into long-term winners? In:Tobacco reform:at the crossroads of health and development:a multisectoral perspective. Washington (DC): World Bank Group;2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/ uploads/2020/01/WB-Report-CrossRoads.pdf,accessed 2 February 2021). 101.Petit P,Nagy J. How to design and enforce tobacco excises? Washington (DC): International Monetary Fund;2016 (International Monetary Fund How To Notes,No.3/ 2016;https://www.imf.org/external/ pubs/ft/howtonotes/2016/howtonote1603.pdf, accessed 7 October 2020). 102.Tonga NCD related taxation policy assessment preliminary findings. Presentation made at the innovations and actions against NCDs conference ADB headquarters,Manila,17 July 2018. 103.Moreno-Dodson B. Raising tobacco taxes:why,what,and how:where public health and economics converge. In:Tobacco tax reform:at the crossroads of health and development: 981第四章 烟草税收的政治经济学 a multisectoral perspective. Washington (DC):World Bank Group;2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/2020/01/WB-Report-CrossRoads. pdf,accessed 2 February 2021). 104.Fuchs A,Del Carmen G,Mukong AK. Long-run impacts of increasing tobacco taxes: evidence from South Africa. Policy Research Working Paper 8369. Washington [DC]: World Bank;2018 (https://openknowledge. worldbank.org/bitstream/handle/10986/ 29497/WPS8369.pdf? sequence=4&isAlowed=y, accessed 18 February 2021) . 105.van Walbeek C. Case studies in ilicit tobacco trade:South Africa. Tobacconomics fact sheet. Chicago:University of Chicago,Tobacconomics,Health Policy Center;2020 (https://tobacconomics.org/files/ research/607/UIC_South-Africa-Ilicit-Trade-Case- Study-v1.3.pdf,accessed 18 February 2021). 106.van Loggerenberg,J. Tobacco Wars. Cape Town:NB Publishers;2019. 107.Marquez P. Global tobacco control:a development priority. In:Tobacco tax reform at the crossroads of health and development:a multisectoral perspective. Washington (DC): World Bank Group;2017 (https://untobaccocontrol.org/taxation/e-library/wp-content/ uploads/2020/01/WB-Report-CrossRoads. pdf,accessed 2 February 2021). 108.Kaiser K,Bredenkamp C,Iglesias R. Sin tax reform in the Philippines:transforming public finance,health,and governance for more inclusive development. Washington [DC]:World Bank;2017 (https:// tobacconomics.org/files/research/607/UIC_South-Africa-Ilicit- Trade-Case-Study-v1.3.pdf,accessed 18 February 2021). 109.Scolo M,Bayly M. Revenue from tobacco taxes in Australia. In:Scolo M,Winstanley M, editors. Tobacco in Australia:facts and issues. Melbourne:Cancer Council Victoria;2016 (https://www.tobaccoinaustralia. org.au/chapter-13-taxation/13-6-revenue-from-tobacco- taxes-in-australia,accessed 21 October 2020). 110.Scolo M,Bayly M. Tobacco taxes in Australia. In:Scolo M,Winstanley M,editors. Tobacco in Australia:facts and issues. Melbourne:Cancer Council Victoria;2016 (https://www.tobaccoinaustralia.org.au/ chapter-13-taxation/13-2-tobacco-taxes-in- australia,accessed 21 October 2020). 111.Breton E,Richard L,Gagnon F,Jacques M,Bergeron P. Fighting a tobacco-tax rolback:a political analysis of the 1994 cigarette contraband crisis in Canada. Journal of Pub Health Policy. 2006;27:77-99. 112.Fanta E,Garcia M. Preventing and controling ilicit tobacco flows. In:Tobacco tax reform:at the crossroads of health and development:a multisectoral perspective. Washington (DC):World Bank Group;2017 (https://untobaccocontrol.org/taxation/e- library/wp-content/uploads/2020/01/WB-Report-CrossRoads.pdf,accessed 2 February 2021). 113.Goodchild M,Perucic AM,Zheng R,Blecher E,Paul J. The health impact of raising tobacco taxes in developing countries. In:Tobacco tax reform:at the crossroads of health and development:a multisectoral perspective. Washington (DC):World Bank Group;2017 091 世界卫生组织烟草税政策和管理技术手册 (https://untobaccocontrol.org/ taxation/e-library/wp-content/uploads/2020/01/WB- Report-CrossRoads.pdf,accessed 2 February 2021). 114.Ross H. Tracking and tracing tobacco products in Kenya. Prev Med. 2017;105:S15-S18 (https://reader. elsevier. com/reader/sd/pi/S0091743517301457? token = 424D9EC1D42A9C69BDB55B90090C3F7853AA 20A4970D0E379C91EE64C73A273A0FA949557F4F120A0B636701581EC055,accessed 2 February 2021). 115.Mamudu H,Hammond R,Glantz S. Tobacco industry attempts to counter the world bank report curbing the epidemic and obstruct the WHO framework convention on tobacco control. Soc Sci & Med. 2008;67(11):1690-9 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC2662513/pdf/nihms81164. pdf,accessed 2 February 2021). 116.Gilmore A. Understanding the vector in order to plan effective tobacco control policies:an analysis of contemporary tobacco industry materials. Tob Control. 2012;21:119-26 (https://tobaccocontrol. bmj.com/content/tobaccocontrol/21/2/119.ful.pdf,accessed 2 February 2021). 117.The economics of tobacco taxation and employment in Indonesia:policy implications technical brief. Washington (DC):World Bank Group;2018(https://openknowledge. worldbank.org/bitstream/handle/10986/29814/126158-WP-P154568-PUBLIC-add-series- WBGPNIndoEmploymentUPDATEFin alWeb.pdf? sequence=1&isAlowed=y,accessed 2 February 2021). 118.Sahadewo GA,Drope J,Li Q,Nargis N,Witoelar F. Tobacco or not tobacco:predicting farming households􀆳 income in Indonesia. Tob Control. 2020. 119.The economics of tobacco farming in Indonesia:health,population,and nutrition global practice. Washington (DC):World Bank Group,2017 (http://documents1.worldbank. org/curated/en/161981507529328872/pdf/120307-REVISED-PUBLIC-WBGIndoEconomics TobaccoFarming.pdf,accessed 2 February 2021). 120.The revenue and employment outcome of biri taxation in Bangladesh. Dhaka:National Board of Revenue,Government of the People􀆳s Republic of Bangladesh;2019 (https:// www. who.int/docs/ default-source/searo/bangladesh/pdf-reports/cat-2/biri-study- report-03-12-2019.pdf? sfvrsn=b8fea69_2,accessed 2 February 2021). 121.Magati P,Lencucha R,Li Q,Drope J,Labonte R,Appau A,et al. Costs,contracts and the narrative of prosperity:an economic analysis of smalholder tobacco farming livelihoods in Kenya. Tob Control. 2019;28(3):268-73 (https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC6512316/pdf/nihms-1012064. pdf,accessed 2 February 2021). 122.Ghaus K,Iqbal MA,Sabir M,Saleem W,Ali A,Aamir N. Macroeconomic impacts of tobacco use in Pakistan. Karachi:The Social Policy and Development Centre;2018 (https://tobacconomics.org/files/ research/483/Macroeconomic-impact-of-Tobacco-in- Pakistan.pdf,accessed 2 February 2021). 123.Ross H. The effects of tobacco taxes on employment and labor productivity. In:Tobacco 191第四章 烟草税收的政治经济学 tax reform:at the crossroads of health and development. Washington (DC):World Bank Group;2017 (https:// untobaccocontrol.org/taxation/e-library/wp-content/uploads/ 2020/01/WB-Report-CrossRoads.pdf,accessed 2 February 2021) 124.Bigawanto M. Tobacco industry interference undermined tobacco tax policy in Indonesia. Bangkok:Southeast Asia Tobacco Control Aliance (SEATCA);2019 (https://seatca. org/dmdocuments/ Indonesia%20TII%20in%20Tax.pdf,accessed 2 February 2021). 125.Ahsan A,Wyono N. Analysis of the impact of higher cigarettes prices on employment in Indonesia. Bangkok:Southeast Asia Tobacco Control Aliance;2007 (https://seatca.org/ dmdocuments/An% 20 Analysis% 20of% 20the% 20Impact% 20of% 20Higher% 20Cigarette%20Prices%20on%20Employment%20 in%20Indonesia.pdf,accessed 2 February 2021). 126.Higashi H,Khuong TA,Ngo AD,Hil PS. Evidence and decision making:tobacco control policy and legislation in Vietnam. Int J of Health Plann. 2013;2(1):e72-e94. 127.Buck D,Raw M,Godfrey CA,Sutton M. Tobacco and jobs:the impact of reducing consumption on employment in the UK. York:University of York;1995 (https://www. researchgate.net/ publication/4814523_Tobacco_and_jobs_the_impact_of_reducing_ consumption_on_employment_ in_the_UK,accessed 2 February 2021). 128.Petkova R,Ileva R and Rusenov A. Tobacco control policies and tobacco industry employment in Bulgaria. In:Economic,social and health issues in tobacco control. Report of a WHO international meeting,Kobe,Japan,3-4 December 2001. Geneva:World Health Organization;2003. 129.Jha V,Narayanan BG,Wadhwa D,Tesche J. Economic and environmental effects of reduction in smoking prevalence in Tanzania. Tob Control;2020;29:24-8 (https:// tobaccocontrol.bmj.com/content/ tobaccocontrol/29/1/24.ful.pdf,accessed 2 February 2021). 130.Lencucha R,Makoka D,Drope J,Appau A,Labonté R,Li Q,et al. Costs,revenues and profits:an economic analysis of smalholder tobacco farmer livelihoods in Malawi. Tob Control. 2017;26(6)634-40 (https:// www.ncbi.nlm.nih.gov/pmc/articles/ PMC5957074/pdf/nihms965842.pdf,accessed 2 February 2021). 131.Goma FM,Labonté R,Drope J,Li Q,Zulu R,Kangwa E. The economics of tobacco farming in Zambia:tobacco farmers survey report. Lusaka:University of Zambia and American Cancer Society;2019 (https://www.cancer.org/content/dam/cancer-org/ research/economic-and-healthy-policy/zambia-report-7-16-19.pdf,accessed 28 October 2020). 132.Chavez JJ,Drope J,Li Q,Aloria MJ. The economics of tobacco farming in the Philippines. Quezon City:Action for Economic Reforms and Atlanta:American Cancer Society,2016 (https://www.cancer. org/content/dam/cancer-org/research/economic-and-healthy- policy/the-economics-of-tobacco-farming-in-the-philippines.pdf,accessed 22 October 2020). 291 世界卫生组织烟草税政策和管理技术手册 133.Augstin CP,Cardenas RP,Cortez JB,Gimongala GC,Ramos AR. The effects of the sin tax reform law of 2012 to tobacco farmers of Amulung,Cagayan. IJARMSS. 2017;6(12):57- 75. 134.Mercado A. Tobacco tax:good for the tobacco farmers,too. Business World;2019 [website] (https:// www.bworldonline.com/tobacco-tax-good-for-the-tobacco-farmers- too/,accessed 25 October 2020). 135.Yürekli A,Önder Z,Elibol M,Erk N,Cabuk A,Fisunoglu M,et al. The economics of tobacco and tobacco taxation in Turkey. Paris:International Union Against Tuberculosis and Lung Disease;2010 (https:// www.who.int/tobacco/en_tfi_turkey_report_feb2011. pdf,accessed 2 February 2021). 136.Latif E,Warner V. Legislation and smoking prevention. In:Loddenkemper R,Kreuter M, editors. The Tobacco Epidemic:progress in Respiratory Research. 2015;42:136-148. 137.Tandon A,Cashin C. Assessing public expenditure on health from a fiscal space perspective. Washington (DC):World Bank Group;2010 (http://documents1. worldbank.org/curated/en/333671468330890417/pdf/560530WP0Box341penditureFiscal Space.pdf,accessed 2 February 2021). 138.Cashin C,Sparkes S,Bloom D. Earmarking for health:from theory to practice. Geneva: World Health Organization;2018 (https://apps.who.int/iris/bitstream/handle/10665/ 255004/9789241512206-eng. pdf? sequence=1,accessed 2 February 2021). 139.Wilkinson M. Paying for public spending:is there a role for earmarked taxes? Fiscal Studies. 1994;15(4):119-135 (https://onlinelibrary.wiley.com/doi/abs/10.1111/j. 1475-5890.1994.tb00213.x,accessed 2 February 2021). 140.ITAF:impôts et taxes affectées à la protection sociale. Les impôts et taxes affectés (ITAF)[taxes and duties alocated to social protection]. Vie publique:au coeur du débat public [website]. Paris:Directorate of Legal and Administrative Information;2018 (in French) (https://www.vie-publique.fr/fiches/22010-impots-et-taxes-affectees-itaf- protection-sociale,accessed 31 October 2020). 141.Seventeenth congress of the Philippines,Metro Manila,RA 11346,23 July 2018 (https:// www. officialgazette.gov.ph/downloads/2019/07jul/20190725-RA-11346-RRD.pdf, accessed 31 October 2020). 142.McCleary W. The earmarking of government revenue:a review of some World Bank experience. World Bank Res Obs. 1991;6(1):81-104 (http://documents1.worldbank. org/curated/en/109871468764672767/ pdf/multi-page.pdf,accessed 2 February 2021). 143.Strategic investment of tobacco tax revenue. Washington (DC):Campaign for Tobacco- Free Kids;2020 (https://www.tobaccofreekids.org/assets/global/pdfs/en/strategic_ investment_tobacco_tax_revenue. pdf,accessed 31 October 2020). 144.Bialous SA,Glantz SA. Arizona􀆳s tobacco control initiative ilustrates the need for continuing oversight by tobacco control advocates. Tob Control. 1999;8:141-51 (https:// tobaccocontrol.bmj.com/content/ tobaccocontrol/8/2/141.ful.pdf,accessed 2 February 391第四章 烟草税收的政治经济学 2021). 145.Goldman LK,Glantz SA. The passage and initial implementation of Oregon􀆳s measure 44. Tob Control. 1999;8:311-22 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC1763939/pdf/v008p00311.pdf,accessed 2 February 2021). 146.Heiser PF,Begay ME. The campaign to raise the tobacco tax in Massachusetts. Am J Public Health. 1997;87(6):968-73 (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC1380932/pdf/amjph00505-0082. pdf,accessed 2 February 2021). 147.Landman A,Bialick P. Tobacco industry involvement in Colorado. Chicago:American Lung Cancer Association;2004 (https://escholarship.org/uc/item/9d880448# main, accessed 18 February 2021). 148.Technical manual on tobacco tax administration. Geneva:World Health Organization; 2010. (https:// apps.who.int/iris/bitstream/handle/10665/44316/9789241563994_eng. pdf? sequence=1,accessed 26 January 2021). 149.Vardavas CI,Filippidis FT,Agaku IT,Mytaras V,Bertic M,Connoly GN,et al. Tobacco taxation:the importance of earmarking the revenue to health care and tobacco control. Tob Induc Dis. 2012;10:21 (http://www.tobaccoinduceddiseases.org/Tobacco-taxation-the- importance-of-earmarking-the-revenue-to-health-care-and-tobacco,66110,0,2.html, accessed 31 October 2020). 150.Bird RM. Tobacco and alcohol excise taxes for improving public health and revenue outcomes:marrying sin and virtue?. Washington (DC):World Bank Group;2015 (Policy Research Working Paper 7500;https://openknowledge.worldbank.org/bitstream/handle/ 10986/23464/ Tobacco0and0al0ing0sin0and0virtue00.pdf? sequence=1&isAlowed=y, accessed 2 February 2021). 151.Earmarked tobacco taxes:lessons learnt from nine countries. Geneva:World Health Organization;2016 (https://apps. who.int/iris/bitstream/handle/10665/206007/ 9789241510424_eng.pdf? sequence=1,accessed 2 February 2021). 152.The establishment and use of dedicated taxes for health. Manila:World Health Organization Western Pacific Regional Office;2004 (https://www.who.int/health_ financing/documents/dedicated_taxes. pdf? ua=1,accessed 2 February 2021). 153.Teja RS. The case for earmarked taxes:theory and example. Washington (DC): International Monetary Fund;1988 (IMF Working Paper;https://www.elibrary.imf.org/ doc/IMF001/14514-9781451922677/14514-9781451922677/Other_formats/Source_PDF/ 14514-9781455212002.pdf,accessed 2 February 2021). 154.Keen M. Taxation and development-again. Washington (DC):International Monetary Fund;2012(IMF Working Paper WP/12/220O;(https://www.imf.org/external/pubs/ ft/wp/2012/wp12220.pdf,accessed 2 February 2021). 491 世界卫生组织烟草税政策和管理技术手册 附录4.1　评估非法烟草贸易性质和规模的方法 A4.1　直接测量 A4.1.1　吸烟者拦截和烟包观察调查 非法贸易可以通过检查吸烟者的烟盒来直接衡量。吸烟者自己可以提供有关购买模 式、品牌偏好和支付价格的信息。研究人员可以根据便利样本(可能不具有代表性的样 本)或基于概率的样本(选择在统计上代表潜在人群的样本)来选择个人或零售商进行调 查。 从包装中收集的数据可以揭示包装是否符合当地税法。信息可以从客观标记中获 得,例如品牌、公共卫生警告标签、印花税票、外语标签或免税标签。在这些停留期间,研 究人员可以记录人口统计信息(例如吸烟者的年龄和性别)、与吸烟相关的历史(例如每天 吸烟的数量)和价格信息。这有助于了解能够并愿意避免卷烟税的吸烟者的概况。包装 观察可与基于人口的住户调查结合使用,以获得基于人口的非法烟草贸易估计值[1,2]。 例如,作为定期全国健康调查的一部分,Kaplan 等人对土耳其的吸烟者进行了一项横断 面研究,采用面对面访谈者管理的调查和包装观察[3]。作为研究方案的一部分,他们能够 收集社会人口统计、生活方式和医疗细节以及包装观察。 吸烟者拦截和包装观察调查的优缺点 进行包装观察的一个主要优点是它是直接和客观的,吸烟者不受任何价值判断的影 响[2]。与调查数据相结合,群体观察可以适当地解释不是他们所调查地区居民的受访 者[4]。缺点包括难以确定代表烟草使用人群的区域,以及难以对重要的亚群(例如老年人 和行动不便的吸烟者)进行抽样。此外,在白天进行的调查可能会低估在校青少年吸烟者 的数量。另一个缺点是相当多的吸烟者可能拒绝出示他们购买的烟包[2]。卡普兰等人发 现在土耳其抽样的吸烟者中有 24% 没有向研究采访者展示他们的烟盒[3]。可以通过要 求用户提供有关所购买品牌的信息、是否张贴任何公共卫生警告以及支付的价格来缓解 此问题[5]。尽管响应基于回忆,但它们仍可能产生有用的信息。例如,Joossens 等人允许 未出示烟盒的吸烟者提供自我报告的信息,并且发现这些受访者与确实出示烟盒的吸烟 者之间在非法烟盒方面没有统计差异[2]。基于吸烟者拦截准确测量的另一个障碍是无法 确定购买单支卷烟的吸烟者的纳税情况,因为这些吸烟者通常没有得到一包烟。然而,通 过要求吸烟者报告购买的品牌和支付的价格,仍然可以在自我报告调查中获取信息。 供读者参考以获得额外指导的主要研究: Kaplan B、Navas-AcienA、Cohen JE。土耳其非法卷烟消费的流行情况及相关因素。 烟草控制。2018,27(4):442 447。 A4.1.2　包装退货和包装更换调查 包装退货和包装更换调查属于更广泛的包装观察研究类别,该研究使用调查抽样技 591第四章 烟草税收的政治经济学 术来检查吸烟者的包装特征并确定他们是否符合税收规定。对于这些调查,分析单位是 个人。换包调查和包装退货调查之间的主要区别在于,换包调查为吸烟者提供替换包,而 包装退货则内置于邮件调查中,并允许受访者以未开封的包装邮寄。换包和包装退货调 查使用概率和非概率抽样程序。概率抽样允许研究人员推广到更广泛的人群。 政府可以使用这种方法快速评估特定地理区域内非法产品的供应情况或衡量非法市 场的份额。如果出现新型烟草制品(例如新的廉价白色品牌)或涉嫌伪造税票特征,则可 能会进行快速评估。使用便利抽样策略的快速评估可以将研究人员安置在繁忙的十字路 口附近,在那里他们可以要求吸烟者允许查看他们的烟盒或拍摄照片。一项基于人群的 研究需要一个与烟草使用人群密切相关的样本。 包装退货和包装更换调查的优缺点 包装退货和包装更换调查可能有助于克服与传统吸烟调查相关的污名。例如,当研 究人员要求吸烟者查看他们的烟盒时(或拍照时),并没有做出任何价值判断。这些调查 是很好的快速评估工具,可用于检查旨在阻止非法贸易的包装(例如具有跟踪和追踪技术 或高科技印花税票的包装)的物理特征的有效性。此外,它们还可以辅以基于人群的烟草 使用调查。当与此类调查数据结合使用时,这些方法允许研究人员获得有关非法购买背 景的相关信息,包括来源(例如街道、同行网络、零售店)和价格。邮寄调查是在受访者舒 适的家中填写的,没有家人或路人在场,这可以向他们保证答复将被保密。一个潜在的缺 点是,同时购买非法卷烟和纳税卷烟的吸烟者可能会不成比例地寄回符合要求的包装。 此外,在中低收入国家中,由于与邮件传递系统相关的问题,这种调查分发模式可能不可 靠。 供读者参考以获得额外指导的关键研究: Fix BV、HylandA、O􀆳ConnorRJ、CummingsKM、FongGT、Chaloupka FJ。一种估计 美国免税卷烟流行率的新方法:2009 年和 2010 年国际烟草控制调查的结果。烟草控制。 2014,23:61 66。 A4.1.3　丢弃烟包调查 丢弃烟包调查,也称为废弃空烟包调查,主要用于高收入和中等收入国家(例如美国、 法国、加拿大、新西兰、墨西哥和波兰)。这种不引人注目的方法依赖于吸烟者公开丢弃包 装的前提(例如在街道、人行道和公共垃圾桶中)。这些包装具有说明它们是否符合税收 规定的特征(例如税票、健康警告)。例如,在希腊丢弃的卷烟包装上的阿尔巴尼亚健康警 告标签提供了该包装运往阿尔巴尼亚市场的证据。该包裹可能是由犯罪企业家走私到希 腊的,也可能是由游客带来的。从具有代表性的地理样本中收集丢弃的包装并检查这些 特征可以提供税收合规性的估计。在操作上,这种数据收集方法使用生态方法,其中地理 区域是分析单位。地理单位旨在代表城市/国家的吸烟者,可以通过行政方式定义(例如, 由国家人口普查局或过境区)或可以反映研究人员定义的社区(例如,公共汽车站或活动 空间附近的半英里缓冲区)。 加拿大的研究人员已经扩展了收集乱丢包装的方法,包括对 25 个中学后校园的烟 蒂的收集和分析。烟蒂提供了所售品牌(或缺乏品牌)的信息,并允许研究人员区分合法 691 世界卫生组织烟草税政策和管理技术手册 和非法产品[6]。乱丢包装方法的最新创新扩展是从卷烟零售商处收集包装。约翰和罗斯 从印度注册和未注册零售商的样本中收集了空包的烟草制品[7]。鉴于单支卷烟在印度的 非法市场占主导地位,从零售商处收集包装很重要。购买单支卷烟的吸烟者无法在包交 换或街道拦截调查中提供一包卷烟,因此从地面收集乱扔的包装会低估单支卷烟的销售 量。这种方法的可行性取决于研究人员和零售商之间的关系(增强信任)以及为确保机密 性所做的努力。在某些国家/地区,零售商可能会因为对保密性的担忧而难以遵守这种研 究方法,因为零售商可能面临刑事和民事处罚,具体取决于研究结果。 丢弃烟包调查的优缺点 丢弃烟包调查通常对政府有利,因为它们有助于与行业估计进行比较。这是最受欢 迎的方法之一,因为它产生的估计不太可能因困扰调查研究的社会期望、回忆错误和保密 性问题而产生偏差,而且它们比在吸烟者拦截或家庭调查中使用的面对面访谈容宜得多。 然而,这些调查存在一些问题,特别是在高收入国家,包括无法区分避税和逃税[8]。例如, 柏林的一包带有越南印花税的包裹可能是大量走私或由临时访客带来的。考虑到旅游业 带来的潜在偏差,研究人员规避了这个问题,并扩大了衡量卷烟税收违规行为的范围。丢 弃烟包调查的另一个缺点是需要更大的预算来聘请现场研究人员来收集、编码和分析数 据。并非所有国家/地区都在其卷烟包装上使用印花税票,这可能会使衡量税收合规性变 得困难。这些调查也可能低估了印度等低收入国家的市场,这些国家的非法贸易主要项 目是单支卷烟[7]。如果乱丢包装行为与从事非法贸易的意愿相关,调查也可能高估非法 贸易。最后,丢弃烟包调查和烟蒂收集提供了非法烟蒂和包装比例的信息,而不是购买非 法卷烟的吸烟者比例[6]。 读者可阅读以下相关重要文献: Barker DC,Wang S,Merriman D,Crosby A,Resnick EA,Chaloupka FJ. Estimating cigarette tax avoidance and evasion:evidence from a national sample of littered packs. Tob Control. 2016;25(Suppl 1):i38-i43. Merriman D. The micro-geography of tax avoidance:evidence from littered cigarettepacks in Chicago. Am Econ J Econ Policy. 2010;2(2):61-84. Stoklosa M.,Paraje G.,Blecher E.,A Toolkit on Measuring Ilicit Trade in Tobacco Products. A Tobacconomics and American Cancer Society Toolkit. Chicago, IL:Tobacconomics,Health Policy Center,Institute for Health Research and Policy,the University of Ilinois at Chicago,2020 (https://tobacconomics.org/files/research/621/ uic-ilicit-trade-tool-kit-eng-v2.0-2.pdf,accessed 18 February 2021). A4.1.4　自我报告调查 将自我报告调查分发给具有代表性的人口样本时,可以提供有关税收不合规发生率 的有意义的数据。调查可以通过多种方式分发给个人或家庭,包括面对面、电话、邮件和 互联网。专门针对非法购买的问题可以作为补充问题添加到现有的健康或烟草调查中。 一些国家在其成人和青年烟草调查中包含此类问题,以估计逃税/避税情况。例如,加拿 大的年度青年吸烟调查询问吸烟者购买第一民族/本土品牌卷烟的频率[9,10]。戴维斯等 791第四章 烟草税收的政治经济学 人使用纽约成人烟草调查的数据来衡量最近购买的卷烟的购买来源(美洲原住民保留区、 低税收邻国或国家、免费电话号码、互联网、免税商店)和支付的价格[11]。作为 ITC 项目 的一部分,目前有 28 个国家使用调查来衡量税收违规情况[12]。可以使用全球成人烟草 使用调查中的问题进行类似的分析。例如,伊格莱西亚斯等人使用巴西全球成人烟草使 用调查将自我报告的价格与定义的阈值零售价格进行比较,以估计巴西吸烟者中非法卷 烟使用的比例[13]。鼓励各国使用现有的全球健康调查或将与非法贸易有关的类似问题 纳入其年度健康调查。向受访者询问每包支付的价格(包括税款)、品牌名称和购买卷烟 的地点(例如免税店、无牌供应商、互联网)有助于更好地了解非法烟草贸易。 自我报告调查的优缺点 自我报告调查可以随着时间的推移重复进行,以衡量与卷烟税收增加相关的购买趋 势和进展。精心设计的调查还可以提供国家层面的普遍估计。根据样本的大小,自我报 告调查可以提供跨地域的可比数据,帮助政府确定资源目标。例如,非法卷烟在城市地区 更为普遍的调查结果可能会导致更多的教育活动和有针对性的执法。 自我报告调查的局限性包括因参与非法贸易带来的社会污名而产生偏见的可能性, 这可能导致调查受访者少报参与情况。此外,调查可能无法衡量个人是否与避税与逃税 有关。最后,有证据表明,自我报告的家庭调查可能无法充分代表吸烟者。 读者可阅读以下相关重要文献: Calaghan RC,Veldhuizen S,Ip D. Contraband cigarette consumption among adolescent daily smokers in Ontario,Canada. Tob Control. 2011;20(2):173-174. Davis K,Farrely M,Li Q,Hyland A. Cigarette purchasing patterns among New York smokers:implications for health,price,and revenue. Albany (NY):New York State Department of Health,Tobacco Control Program;2006. A4.1.5　秘密采购调查 在高收入、中等收入和低收入国家进行的许多研究使用秘密购买包装和单支卷烟来 衡量公共和半私人空间中非法卷烟的供应情况[1417]。美国的烟草业也使用这种方法来 识别销售假冒卷烟的零售商[18]。 秘密购买调查不提供对非法贸易规模(即市场量)的估计。相反,它们作为一种监视 工具来确定非法卷烟的销售地点以及它们渗透到合法企业的程度。例如,秘密购买调查 可以检查非法卷烟是否通过合法零售商销售。它还可以用来衡量对新兴烟草控制政策的 遵守情况,这些政策侧重于产品标准化或新的香料法规(例如,平装或对调味烟草制品的 禁令)。 秘密购买调查使用训练有素的研究人员访问选定的零售商样本,并通过直接购买或 询问掌握非法烟草制品的供应情况。零售商不了解研究的目标,确定非法产品可用性的 方法各不相同。例如,在一些研究中,隐蔽购买者不会直接询问非法产品。相反,他们以 全价购买一包烟草制品,以确定零售商是否打着合法的幌子销售非法产品[14]。然后研究 小组检查包装以确定它们是否合法。在美国,研究人员观察到,一些消费者正在为从低税 收州走私的非法未征税包装支付全价[14]。其他研究协议直接向零售商索要非法产 891 世界卫生组织烟草税政策和管理技术手册 品[14,16,17]。在危地马拉,Arevalo 等人专门向零售商询问“进口卷烟”[17]。秘密买家索取 非法产品的方式也可能因地域而异。例如,在一些国家,隐蔽的买家可能会要求“更便宜” 的包装或非法的白装。 秘密购买调查的优缺点 秘密购买使研究人员能够直接识别非法卷烟的来源。它还允许他们衡量和测试买卖 双方之间的动态。例如,研究人员可以进行实验,看看反复尝试购买产品是否会增加购买 的可能性(称为熟悉协议)[16]。与秘密购买相关的方法学挑战之一是难以为非法来源创 建抽样框架,因为有些可能是未知的(例如酒吧或家庭)。传统方法是在合法网点购买,这 可能会影响估计值。这种方法的另一个问题是,如果买家(也称为评估者)不熟悉卖家或 不符合购买者的典型人口统计特征,他们将很难购买产品。因此,使用秘密购买调查的研 究人员必须对市场有详细的了解,包括个人特别要求了解非法烟草制品的方式,并且他们 必须知道他们是否反映了社区的人口统计数据。例如,在一项对南布朗克斯吸烟者的研 究中,von Lampe 等人发现吸烟者寻找某些线索来评估他们是否被出售非法卷烟[19]。总 体而言,这种方法可能非常昂贵,因为它需要培训研究人员、前往零售商处和购买产品。 秘密采购调查无法让研究人员估计非法贸易的水平,但他们可以提供有关供应情况的信 息。 读者可阅读以下相关重要文献: Silver D, Giorgio MM, Bae JY, JimenezG.,MacinkoJ.Over-the-countersalesofout- of-stateand counterfeit tax stamp cigarettes in New York City. Tob Control. 2016;25 (5):584 586.Arevalo R,Corral JE,Monzon D,Yoon M,Barnoya J. Characteristics of ilegal and legalcigarettepackssoldinGuatemala.GlobalHealth.2016,12(1):78. A4.1.6　缉获货物 缉获是地方、国家和国际组织没收非法制造、运输和销售的烟草制品的执法活动的结 果。扣押旨在通过没收收益(例如现金、汽车或房屋)和贸易工具(例如印刷和烟草机械) 来减少与非法贸易相关的利润。缉获可能发生在供应链的各个环节。 缉获数据通常由负责机构统计并用于衡量计划有效性或作为请求额外资源(例如人 员)的理由。一些数据可能会提供给国际海关组织,包括世界海关组织[20]。记录保存的 质量各不相同。例如,一些机构可能在数据库中保存犯罪档案,详细说明扣押日期、品牌 名称和实验室测试。 缉获提供了关于犯罪活动范围的初步数据,有助于确定关键趋势,以指导执法机构的 工作。例如,缉获可以确定走私者的作案手法趋势,包括运输方式(例如海运与卡车)、入 境点和品牌偏好。缉获的统计数据也可用作衡量干预措施效果的初步测试。例如, Stoklosa 和 Ross 使用加拿大新斯科舍省的缉获数据来测试 2015 年薄荷醇禁令的影响。 他发现禁令前后缉获的薄荷卷烟数量没有统计学上的显著变化[1]。 缉获货物的利弊 通常,通过向机构看门人(例如公共信息官员)提出正式请求,可以很容易地从执法机 构获得缉获的统计数据。然而,缉获数据通常不能代表非法活动的情况。例如,某些地区 991第四章 烟草税收的政治经济学 可能会产生更高的缉获量,因为这是进行大量业务开展的地方。警察机构可能会专注于 某些地区(例如靠近边界的位置)而不是随机检查,他们的调查结果可能仅限于这些特定 地区。缉获数据也可能因所使用的调查程序类型而产生偏差。大量缉获可能是长期调查 (窃听或剔除机密线人)的结果,而较小的缉获可能来自涉及跨境购买少量卷烟(少于 1 000支)的反走私案件[2]。与执法机构的行业合作也可能会影响缉获量。例如,烟草业 更有可能支持执法查获假冒产品而不是走私案件,因为假冒产品会影响其品牌完整性。 A4.2　剩余法 由于非法烟草贸易通常是分散的,因此很难直接观察。然而,研究人员有时无须直接 观察就能推断出其规模,方法是将观察到的烟草税收与假设对所有烟草消费征税时所征 收的税收数额进行比较。观察到的和假设的收入之间的差异称为剩余收入,可用作判断 非法贸易规模的指标。即使残差只是一个近似的衡量标准,其规模的变化也可能是交易 规模变化的可靠指标。当可靠地观察到实际烟草税收时,剩余法的主要挑战是对假设所 有烟草消费都征税的情况下本应征收的税收数额做出准确的估计。 A4.2.1　差距分析 差距分析是首选的残差方法,因为它直观、直接且相对容易向决策者和公众解释,并 且已广泛应用于政府研究[4]。使用差距分析的研究人员将基于调查的自我报告的消费数 据与观察到的(通常是行政)纳税销售数据进行比较。基本前提是,如果自我报告和观察 数据都准确,报告的消费量和纳税销售量之间的任何差异都可以解释为合法进口的免税 卷烟(如免税销售)、出口的征税卷烟、逃税或避税。 与大多数残差方法一样,实施差距分析的最大研究挑战是获得可靠和准确的烟草消 费估计值。在其最简单的实施方式中,差距分析计算的残差是调查中报告的烟草消费量 与纳税销售量(通常可从行政来源获得)减去出口量之间的差值(应至少为零)。然而,这 种简单的计算通常是有缺陷的,因为报告的烟草消费调查低估了真实消费。调查对象对 披露被视为不健康和潜在的社会不良行为保持沉默,可能会导致烟草消费的低报。根据 法律和文化背景,某些群体(例如妇女或青年)可能比其他群体更可能少报消费。其他群 体(例如叛逆的青年男女)可能准确报告甚至高估了消费量。Reuter 和 Majmundar 用美 国的国家税收总销售额来衡量实际消费,人们普遍认为美国合法免税进口和出口的征税 烟草都非常小,他们发现自我报告的消费量与实际消费量的比率仅为 65 %[4]。在将这 项调查结合他们的分析并考虑差距分析和文献中的证据后,Reuter 和 Majmundar 发现 美国的非法市场(主要由避税或逃税组成)介于消费量的 8.5% 和 21%[4]。较高的估计 范围与先前使用基于人群的群体观察研究进行的估计一致[21]。 对合法进口未征税烟草(例如免税产品)或(合法或非法)出口征税卷烟更为重要的国 家或地区使用差距分析的研究人员应尝试将有关这些因素的数据或估计纳入他们的计算 中。应从报告的消费量中减去合法的未征税进口烟草(在对少报进行调整后),应从征税 销售额中减去征税烟草的出口量。获取关于合法免税进口和出口征税烟草的数据可能具 002 世界卫生组织烟草税政策和管理技术手册 有挑战性,因为这些进口和出口可能是个人旅行者在跨越税收边界时分散决策的结果。 这些活动的数据不一定通过正常的行政活动收集。尽管存在这些挑战,差距分析可能仍 然有用。 例如,如果研究人员有理由相信烟草消费的误报以及合法免税进口和征税出口的数 量相对稳定或随着时间的推移遵循已知趋势(例如正在下降),则差距分析可用于提供非 法贸易的估计值或下限(或上限,当有数年的征税销售额和报告消费数据可用时)。数据 来源可能是有权访问销售数据的国家税务管理员和有权访问烟草消费的卫生部门。在这 种情况下,多年报告的消费和纳税销售数据可以让研究人员估计非法贸易规模的变化,即 使很难衡量绝对水平。Paraje 使用 2008 年全球成人烟草调查和 2013 年全国健康调查 来衡量巴西报告的烟草消费量[22,23]。 差距分析的优缺点 差距分析的一个主要优点是,当可以获得高质量的数据时,它很简单、易于复制并且 可以向决策者和公众解释。然而,可能无法获得关于报告消费量的高质量数据,尤其是在 低收入国家。在许多情况下,差距分析不能提供有关非法市场规模的可靠信息,而仅提供 有关规模随时间变化的信息[22]。此外,一些低收入国家可能没有对纳税卷烟销售的可靠 估计,卷烟销售的二级数据存储库可能对其方法不透明[24]。 差距分析的另一个缺点是它通常不能用于分别估计避税和逃税。如果烟草消费调查 不能代表人口,也可能导致有偏差的估计[25]。此外,通常不可能量化估计的精确度或与 估计相关的不确定性,因为使用样本不能完全代表人口(例如吸烟的人口比例)导致的统 计不确定性和关键事实的不确定性,例如调查受访者少报其烟草消费量等事实。 读者可阅读以下相关重要文献: SzkloA, IglesiasRM, CarvalhodeSouzaM, SzkloM, MariadeAlmeidaL. TrendsinilicitcigaretteuseinBrazilestimatedfromlegalsales, 2012 2016. AmJPublicHealth. 2018; 108 (2 ): 265 269. ParajeG. IlicitcigarettetradeinfiveSouthAmericancountries: agapanalysisforArgentina, Brazil, Chile,ColombiaandPeru.NicotineandTobRes.2019,21(8):1 079 1 086 A4.2.2　计量经济学模型 使用数据来估计需求函数参数的传统由来已久,这些需求函数将消费的商品数量与 消费者面临的价格、他们的收入和其他变量联系起来。由于烟草的成瘾性,以及与烟草使 用相关的重要公共卫生和公共政策问题,经济学家特别关注卷烟需求函数的估计值[26]。 随着有关该主题文献的发展,很明显,如果一些消费者从非法市场获得烟草,征税烟草销 售将成为烟草消费的一个有偏见的指标。同样,如果某些销售未纳税,合法市场上的卷烟 价格可能会高估消费者支付的价格。 虽然经济学家通常无法观察非法市场的销售情况,但他们已经能够开发模型来预测 消费者规避烟草税的条件。他们认为非法烟草市场的相对规模主要取决于两个变量:征 税和免税消费的相对价格以及获得低成本(免税)烟草的难易程度。其他变量,包括逃避 税法带来的社会耻辱和非法烟草的相对质量,也可能影响对它的需求。虽然无法直接观 102第四章 烟草税收的政治经济学 察到非法贸易,但可以通过纳税销售额与预测消费量之间的差异进行估计。当零售商或 消费者逃税时,纳税销售额可能低于预期消费额。如果在辖区内购买一些已纳税的卷烟, 然后在税后价格较高的地区消费,则税率可能会更高。 非法贸易的计量经济学模型估计必须适合所研究国家的情况,因此对数据要求可能 因情况而异。使用这种方法的研究人员应该熟悉文献,也应该了解他们研究领域的条件。 它们必须始终包括对某种烟草消费或销售量和本国烟草价格的某种量度,以及已知会影 响烟草需求的其他变量(例如收入)。通常还需要包括衡量非法烟草供应量和相对价格的 变量,这通常可以通过将本国的烟草税与非法烟草来源地区的烟草税进行比较来衡量。 计量经济学模型的优缺点 计量经济学模型的主要优点是它与经济理论和实践的悠久传统相一致,因此可以根 据广泛接受的标准评估建模技术和经验估计的质量。实证分析提供了对避税的价格弹 性、收入弹性和烟草价格弹性的估计。大量的文献基础使得可以严格量化估计的不确定 性,并测试它们对建模过程中所做的各种假设的稳健性。估计值可以与文献中可用的其 他估计值进行比较,这些分析的结果可用于模拟政策变化(包括税收和执法增加)对消费 和避税的影响。由于这种方法提供了对结果不确定性的直接估计,研究人员可以指定他 们对结果的置信度。 计量经济学建模的一个缺点是它需要一段时间内各种重要变量的高质量数据,以及 先进的计量经济学建模专业知识。此外,由于计量经济学模型的结果基于统计推断和经 济理论,而不是直接观察(例如没有印花税的包装比例),因此很难向政策制定者和公众 解释。 读者可阅读以下相关重要文献: BeckerGS, GrossmanM, MurphyKM, ( 1994 ). Anempiricalanalysisofcigaretteaddiction.AmerEconReview.1994,84(3):396 418. Schafferer C,Yeh CY,Chen SH,Lee JM,Hsieh CJ. A simulation impact evaluation of acigaretteexcisetaxincreaseonlicitandilicitcigaretteconsumptionandtaxrevenuein36Euro peancountries.PublicHealth.2018,162:48 57. A4.2.3　专家意见（关键知情人调查和访谈） 对非法贸易动态的洞察可以来自该领域的专家,包括研究人员(例如经济学、刑事司 法和公共卫生人员)、税务部门、执法机构、产品制造商、批发商和零售商,其他关键线人包 括获得机密线人的记者和学者。专家可以提供有关新兴趋势的新信息(例如新的走私路 线)。在某些情况下,研究人员可以获得对被监禁或现役罪犯的采访[2729]。例如,研究东 非卷烟走私的研究人员采访了 150 多名乌干达烟草走私者[29]。 可以通过调查或半结构化访谈来询问专家。当抽样框可用时(例如税务部门雇员名 录),调查比访谈更方便。然而,当专家很难找到抽样框时,建议采用非随机抽样策略与访 谈相结合的方法。确定专家可能需要招募一名负责帮助研究人员寻找更多专家的看门 人;或有目的抽样,根据既定标准识别个人(例如,他们是地方政府雇用的税务专家)[30]。 202 世界卫生组织烟草税政策和管理技术手册 关键知情人调查和访谈的优缺点 线人访谈可以成为识别市场趋势的有用起点(例如,销售非法卷烟的场所或进入方 式)。依赖线人的缺点之一是从他们那里获得的信息可能无法概括。专家知识可能会过 时或受到线人经验的限制。此外,专家的意见是主观的,可能会因专家的就业状况和所使 用的抽样方法而产生偏差。例如,执法人员可能会高估走私的程度,以便为未来的行动获 得额外资金。同样,寻求避税的制造商可能会高估非法贸易,以说明税收与非法行为之间 的联系。或者,烟草控制倡导者可能会低估非法市场措施,以支持税收不会增加非法贸易 的观点。 读者可阅读以下相关重要文献: JoossensL,RawM.CigarettesmugglinginEurope:whorealybenefits? TobControl. 1998;7:66 71.doi:10.1136/tc.7.1.66PMID:9706757. TitecaK,JoossensL,RawM.Bloodcigarettes:cigarettesmugglingandwareconomies incentralandeasternAfrica.TobControl.2011,20(3):226 232. A4.3　混合和多方法研究 鉴于上述评估非法烟草贸易的性质和规模的方法存在缺陷,政府可能希望通过使用 混合或多种方法来验证他们的调查结果。混合方法使用两种方法论范式———以定性和定 量作为探索和解释的工具。例如,混合方法研究可以使用丢弃烟包调查来衡量市场规模 以及吸烟者的自我报告调查,以了解购买模式,包括来源、频率和社会规范。例如, Stoklosa 和 Ross 使用基于人口的自我报告调查和乱扔包装调查估计了波兰非法市场的 份额[1]。同时使用两种类型的调查使政府能够评估其在估计非法市场规模方面的有效 性。或者,政府可以采用多方法研究,即使用与传统相似的多种方法(例如焦点小组和半 结构化访谈)[31]。Saenz de Miera 等人使用面对面访谈(家庭)、垃圾收集和观察单一卷烟 卖家,这使他们不仅能够交叉验证两种主要方法,还可以查看单一卷烟的品牌是否是一个 很好的合法与非法贸易衡量标准[33]。 混合和多方法研究的优缺点 混合和多方法研究使研究人员能够检查他们发现的有效性。在非法贸易估计被政治 化的情况下,首选多种方法。例如,烟草行业可能会质疑低估计值,而烟草控制研究人员 和支持者可能会质疑高估计值。混合和多种方法(例如乱扔包装调查和线人访谈)可以使 政府了解非法贸易运作的状态和背景,包括所涉及的行为者和销售场所。使用混合和多 种方法的一个缺点是成本。受到紧缩预算限制的政府可能会选择使用提供最准确信息的 单一方法。然而,鉴于每种方法面临的问题,这可能并不可行———每种方法都有局限性。 相反,政府可以将高成本方法与低成本方法相结合(例如,将访谈与空包装调查相结合,或 将执法扣押数据与面对面的消费者调查相结合)。 读者可阅读以下相关重要文献: Zaloshnja E,RossH,Levy DT.The impactoftobaccocontrolpoliciesinAlbania. TobControl.2010,19:463 468. 302第四章 烟草税收的政治经济学 附录4.1参考文献 1.Stoklosa M,Ross H. Contrasting academic and tobacco industry estimates of ilicit cigarette trade:evidence from Warsaw,Poland. Tob Control. 2014; 23(e1),e30 e34(https://www. researchgate.net/publication/255954649_Contrasting_academic_and_tobacco_industry_ estimates_ of _ ilicit _ cigarette _ trade _ Evidence _ from _ Warsaw _ Poland, accessed1February2021). 2.Joossens L,Lugo A,La Vecchia C,Gilmore AB,Clancy L,Galus S. Ilicit cigarettes and hand-roledtobacco in 18 European countries: a cross-sectional survey. Tob Control. 2014; 23:e17 e23 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3812425/pdf/ nihms491463.pdf,accessed2February2021). 3.KaplanB,Navas-AcienA,CohenJE.Theprevalenceofilicitcigaretteconsumptionandrelatedfact orsinTurkey.TobControl.2018;27(4):442 7. 4. ReuterP, MajmundarM. UnderstandingtheUSilicittobaccomarket: characteristics, policycontext,and lessons from international experiences.Washington(DC): National Academies Press; 2015. 5.Maldonado N,Llorente BA,Iglesias RM,Escobar D. Measuring ilicit cigarette trade in Colombia. TobControl. 2020;29:s260 s266(https://tobaccocontrol.bmj.com/content/ tobaccocontrol/29/Suppl_4/s260.ful.pdf,accessed29January2021). 6.Barkans M,Lawrance KA. Contraband tobacco on post-secondary campuses in Ontario, Canada:analysis of discarded cigarette butts. BMC Pub Health. 2013;13(1):335(https:// bmcpublichealth.biomedcentral.com/articles/10.1186/1471 - 2458 - 13 - 335, accessed2February2021). 7.John RM,Ross H. Ilicit cigarette sales in Indian cities: findings from a retail survey. Tob Control.2018;27(6),684 688. 8. DavisKC, GrimshawV, MerrimanD, FarrelyMC, ChernickH, CoadyMH, etal. CigarettetraffickinginfivenortheasternUScities.TobControl.2014;23(e1):e62 e68. 9.Calaghan RC,Veldhuizen S,Leatherdale S,Murnaghan D,Manske S. Use of contraband cigarettesamongadolescentdailysmokersinCanada.CMAJ.2009;181(6 7):384 6(https:// www.cmaj.ca/content/cmaj/181/6-7/384.ful.pdf,accessed2February2021). 10.GuindonGE,BurkhalterR,BrownKS.LevelsandtrendsincigarettecontrabandinCanada. TobControl.2017;26(5):518 25(https://www.researchgate.net/publication/ 307890937 _ Levels _ and _ trends _ in _ cigarette _ contraband _ in _ Canada, accessed2February2021). 11.DavisK,FarrelyM,LiQ,HylandA.CigarettepurchasingpatternsamongNewYorksmokers: implications for health,price,and revenue. Albany: New York State Department of Health; 2006(https://www.health.ny.gov/prevention/tobacco_control/docs/cigarette_ purchasing_patterns.pdf,accessed2February2021). 402 世界卫生组织烟草税政策和管理技术手册 12.International Tobacco Control Evaluation Project(2018). Surveys. Waterloo: University of Waterloo;2018(http://www.itcproject.org/surveys,accessed12October2020). 13.IglesiasRM,SzkloAS,deSouzaMC,deAlmeidaLM.Estimatingthesizeofilicittobaccoconsumpt ioninBrazil:findingsfromtheglobaladulttobaccosurvey.TobControl.2017;26(1):53 9. 14.Silver D,Giorgio MM,Bae JY,Jimenez G.,Macinko J. Over-the-counter sales of out-of- state andcounterfeittaxstampcigarettesinNewYorkCity.TobControl.2016;25(5):584 586. 15. BrownJ, WeldingK, CohenJE, CherukupaliR, WashingtonC, FergusonJ, etal. Ananalysisofpurchasepriceoflegalandilicitcigarettesinurbanretailenvironmentsin14low - andmiddle-incomecountries.Addiction. 2017;112:1854 60(https://www.ncbi.nlm. nih.gov/pmc/articles/PMC5600117/pdf/ADD-112-1854.pdf,accessed2February2021). 16.Scolo M,Bayly M,Wakefield M. Availability of ilicit tobacco in smal retail outlets before and aftertheimplementationofAustralianplainpackaginglegislation.TobControl.2015;24 (e1):e45 e51. 17.Arevalo R,Corral JE,Monzon D,Yoon M,Barnoya J. Characteristics of ilegal and legal cigarettepacks sold in Guatemala. Globalization and Health. 2016;12(1):78(https:// globalizationandhealth.biomedcentral.com/articles/10.1186/s12992-016-0219-z, accessed2February2021). 18.Philip Morris USA Inc. v. Shalabi. United States,District Court,C.D. California; 2004 ( https://www. casemine. com/judgement/us/5914b6cbadd7b0493477b3da, accessed2Febru ary2021). 19.von Lampe K,Kurti M,Johnson J,Rengifo AF. ‘I wouldn’t take my chances on the street’ navigatingilegalcigarette purchases in the SouthBronx. J Res Crime Delinq. 2016;53(5): 654 80(https://www.researchgate.net/publication/298515449_I_Wouldn’t_Take_My_ Chances_on_the_Street_Navigating_Ilegal_Cigarette_Purchases_in_the_South_Bronx, accessed2February2021). 20.CustomsandTobaccoReport2009.Brussels:WorldCustomsOrganization;2009.(http:// www. wcoomd. org/en/media/newsroom/2010/june/~/media/ 83967DFEB9F74D388924A 4C61F279DC4.ashx,accessed12October2020). 21.Fix BV,Hyland A,O’Connor RJ,Cummings KM,Fong GT,Chaloupka FJ,et al. A novel approach toestimatingtheprevalenceofuntaxedcigarettesintheUS: findingsfromthe2009and2010 InternationalTobaccoControlSurveys.TobControl.2014;23:i61 i66(https://www.ncbi. nlm. nih. gov/pmc/articles/PMC3984758/pdf/nihms567943. pdf, accessed2February2021). 22. ParajeG.IlicitcigarettetradeinfiveSouthAmericancountries: agapanalysisforArgentina, Brazil,Chile,ColombiaandPeru.NicotineTobRes.2019;21(8):1079 86. 502第四章 烟草税收的政治经济学 23.NationalSurveyofHealth.InstitutoBrasileirodeGeografiaeEstatística;2013(inPortuguese) (https://www.ibge.gov.br/en/statistics/social/health/16840-national-survey-of-health. html? =&t=downloads,accessed27November2020). 24.BlecherE,LiberA,RossH,BirckmayerJ.Euromonitordataontheilicittradeincigarettes. TobControl.2015;24:100 1.(https://tobaccocontrol.bmj.com/content/tobaccocontrol/ 24/1/100.ful.pdf,accessed1February2021). 25.RossH.Understandingandmeasuringcigarettetaxavoidanceandevasion:amethodologicalguide. Tobacconomics; 2015(https://tobacconomics.org/wp-content/uploads/2015/03/Ross_ Methods_to_Measure_Ilicit-Trade_03-17-15.pdf,accessed12October2020). 26.ChaloupkaFJ,WarnerKE.Theeconomicsofsmoking.In:ArrowKJandIntriligatorMD,editors. HandbookofHealthEconomics.Amsterdam:Elsiver;2000.pp1539 1627. 27.AntonopoulosGA.Cigarettesmugglers:anoteonfour‘unusualsuspects’.Glob.Crime.2007;8 (4):393 8(https://www.researchgate.net/publication/248955367 _ Cigarette_ Smugglers_A_Note_on_Four_’Unusual_Suspects’,accessed2February2021). 28. AntonopoulosGA. TheGreekconnection (s ): thesocialorganizationofthecigarette- smugglingbusinessin Greece. Eur J Criminol. 2008;5(3):263 88(https://www. researchgate.net/publication/249752218 _ The _ Greek _ ConnectionsThe _ Social_ Organization _ of _ the _ Cigarette - Smuggling _ Business _ in _ Greece, accessed2February2021). 29.Titeca K,Joossens L,Raw M. Blood cigarettes: cigarette smuggling and war economies in central andeastern Africa. Tob Control. 2011;20:226 32(https://www.researchgate. net/publication/49809289_Blood_cigarettes_Cigarette_smuggling_and_war_economies_in_ central_and_eastern_Africa,accessed2February2021). 30.Babbie,ER.Thebasicsofsocialresearch. Boston:CengageLearning;2013. 31.Tashakkori A,Teddlie C.,editors. Sage handbook of mixed methods in social & behavioral research.NewYork:Sage;2010. 32.Saenz de Miera Juarez B,Reynales-Shigematsu LM,Stoklosa M,Welding K,Drope J. Measuring theilicitcigarettemarketinMexico: acrossvalidationoftwomethodologies. TobControl;2020(https://tobaccocontrol.bmj.com/content/tobaccocontrol/early/2020/ 03/31/tobaccocontrol-2019-055449.ful.pdf,accessed2February2021). 602 世界卫生组织烟草税政策和管理技术手册 附录4.2　如何指定烟草税收入的用途 引入烟草税收入专项拨款几乎总是与消费税的增加(或新的附加税)相结合,而不是 对现有收入进行重新分配[1]。表 A4.1 提供了几个国家用于指定烟草税收入的不同方法 的例子。 表A4.1 用于指定烟草税收入的方法 资金来源/税种 税基和税率示例 取消费税收入的一 部分(烟草、酒精) 从量税 韩国:每包 841 韩元 (0.75 美元)或从量 消费税的 29% 哥斯 达 黎 加:每 包 467.8 科隆(0.83 美 元)或从量消费税的 100% 刚果:每包 20 非 洲金融共同体法郎 (0.036 美元)或从 量消费税的 50% 从价税 哥伦比亚:零售价的 10%(相当于从价税率的 100%) 新增税收入(对现有 的消费税征收附加 税或完全新征一种 税) 从量税 埃及:每包加收0.75 埃及镑(0.042 美元) 从价税 泰国:对消费税基数征收 2% 的附加税 博茨瓦纳:新烟草税按生产成本或 CIF 的 30% 征收 按消费税收入的百 分比 库克群岛:50% 的税收来自 烟草消费税 危地马拉:100% 的税收来自烟草 消费税 注:使用 2018 年 7 月 31 日(数据收集日期)国际货币基金组织的官方汇率换算的当地货币金额。 资料来源:附录4.2参考文献2。 702第四章 烟草税收的政治经济学 附录4.2参考文献 1.Earmarkedtobaccotaxes:lessonslearntfromninecountries.Geneva:WorldHealthOrganization; 2016(https://apps.who.int/iris/bitstream/handle/10665/206007/9789241510424_eng. pdf? sequence=1,accessed2February2021). 2.WHO report on the global tobacco epidemic 2019: offer help to quit tobacco use. Geneva: WorldHealthOrganization;2019(https://www.who.int/teams/health-promotion/tobacco- control/who-report-on-the-global-tobacco-epidemic 2019,accessed24January2021) 802 世界卫生组织烟草税政策和管理技术手册 第五章 烟草税收政策和管理的最佳实践 5.1　税收政策 使用消费税增加来实现减少因吸烟导致的死亡和疾病的公共卫生目标 广泛的研究清楚地证明了提高烟草制品税率和价格在减少烟草使用及有害后果方面 的有效性,特别是在穷人和年轻人中。事实上,提高烟草消费税率是减少烟草使用的最有 效和最具成本效益的政策。消费税是对烟草制品征收的最重要的税种,因为它们能够提 高绝对价格和相对价格。烟草消费税的增加还会产生可观的新收入,这些收入将在中短 期内持续下去。从长远来看,烟草税率的持续增加,再加上其他烟草控制政策和计划的实 施,将导致烟草使用及其后果更大幅度的减少。 将大幅提高烟草消费税率作为减少烟草使用的综合战略的一部分 政府应采取全面的烟草控制战略,其中包括减少成人烟草使用和预防青少年吸烟的 目标。大幅提高消费税率是减少消费的最有效也是最具成本效益的措施。当与其他减少 需求干预措施相结合时,增加税收对烟草使用的影响甚至更大。这些干预措施包括所有 公共场所的全面无烟政策,全面禁止烟草广告、烟草公司的促销和赞助,关于烟草使用后 果的大型图形健康警告,无装饰包装,帮助当前使用者戒烟的广泛努力和大众媒体公共教 育活动。实施减少烟草使用的综合战略可进一步减少烟草使用的有害后果,为提高税收 建立公众和政治支持,并最大限度地提高税收在实现公共卫生目标方面的有效性。 在考虑修改税收政策时,从一开始就让主管部门参与 税务机关和海关部门等主管部门是有效实施税收政策的关键合作伙伴。政策制定者 需要确保在税收政策修订过程中征求主管机关的意见并让其参与,以便从一开始就考虑 到它们对政策变化和执法影响的担忧。这也有助于在执法过程的早期识别和解决可能存 在的漏洞。相关机构之间的协调,包括密切合作和信息共享,将优化税收政策的执行和税 收征收。为了简化合作和信息交流的过程,需要建立法律基础。此外,税务管理机关在整 个税收政策制定过程中的参与对于确保政策的有效实施非常重要。 促进农业、工业、贸易、金融和劳工等部门之间更大的政策一致性 需要在国家层面加强多部门整合和政策一致性,以实现有效的健康改善。尤其重要 的是要确保非卫生部门(例如农业、工业、贸易、金融和劳工)的公共政策和干预措施不违 背烟草控制和税收的预期公共卫生影响(此类干预措施包括为烟草种植或制造提供 补贴)。 5.2　税收设计 税收结构很重要,越简单越好 复杂的税收结构难以管理,为避税和逃税创造机会,并且在实现公共卫生和税收目标 方面不如简单的结构有效。简化烟草消费税的结构将促进税收管理,减少避税和逃税,增 加收入,并有利于减少烟草制品或品牌之间因税收增加而替代,从而对烟草使用产生更大 影响。根据产品特征(例如价格水平、长度、重量、烟草类型)制定多级烟草税率的国家应 降低并最终取消这些差别税率。适当的过渡策略是减少税率随时间的变化,目的是对特 定烟草制品实施统一税率(即无论消费税是从价还是从量,都适用单一税率)。对特定烟 草制品的所有品牌征收统一税也传达了一个明确的信息,即它们同样有害。 更多地依赖特定的烟草消费税来推动价格上涨 通过缩小高价和低价替代品之间的价格差距并限制用户因税收增加而减少吸烟的机 会,更多地依赖从量消费税可以最大限度地提高烟草税对公共卫生的影响。对于目前依 赖从价税的国家,适当的第一步是通过增加相当大的特定组成部分或引入较高的最低特 定消费税(最低消费税)来转向混合税制。对于依赖从价税和从量税相结合的国家,从量 税部分应定期增加,使其在消费税总额中占更大份额。 大幅提高烟草税以降低烟草制品的可负担性 为了最大限度地提高烟草税对公共卫生的影响,同时创造更高的收入,政府应该大幅 提高税收以提高价格并降低烟草制品的可负担性。在许多中低收入国家,烟草使用随着 收入的增加而增加,而且由于收入的增长速度快于烟草制品价格,因此这些产品变得越来 越便宜。为了降低可负担性,税收增加需要导致实际价格上涨高于实际收入的增加。 如果以增加收入为目标,则依靠定期增加消费税 如果政府想要增加烟草收入,就必须定期增加消费税。从税收角度看,重要的决定因 素是税基弹性,它具有3个关键组成部分:烟草需求的价格弹性、税收在零售价格中的份 额和消费税税率对零售价的传递程度。至少在中短期内,增加税收将增加收入,因为需求 是无价格弹性的,税收水平占零售价格的比例通常较低,而且税收增加对零售价格的传导 不太可能高于税收增加本身(没有过度转移)。此外,在拥有强有力的烟草控制政策的衰 退市场中,提高税率是唯一可以扭转收入减少的政策措施。 根据通货膨胀和收入增长自动调整特定烟草税 除非定期调整特定烟草税,否则它们的实际价值将随着一般价格水平的上升而下降。 012 世界卫生组织烟草税政策和管理技术手册 当这种情况发生时,它们在减少烟草使用方面的效果将会减弱。各国政府应建立自动调 整从量税以跟上通货膨胀的机制。最近,一些政府开始扩大这种指数化,将收入增长也包 括在内,进一步确保烟草不会随着时间的推移变得更加负担得起。 定价法规不能被视为消费税的替代品。但是,在某些特定情况下,可以将定价法规与 消费税结合使用,以帮助确保有效实施增税措施 在增加税收具有挑战性或税收结构薄弱的某些情况下,定价监管(特别是最低加价和 底价/最低价格)等非税收政策可能被视为确保高税收和劝阻烟草制品消费的次佳选择。 然而,这些政策不一定会导致理想的价格水平,也不能保护消费者和政府免受行业操纵。 然而,强大的跨国公司在所有细分市场销售品牌并且很容易将增税转移到更便宜的品牌 的背景下,或者在不能禁止价格促销的情况下,最低价格政策可能有助于提高税收增加的 有效性,特别是如果最低价格定期上涨。 实施影响价格水平的非税收政策,例如禁止烟草制品的促销折扣和单支卷烟的销售 禁止促销折扣通常在烟草广告、促销和赞助条款下的烟草控制法中处理。 不要让对提高烟草税的通货膨胀影响的担忧来阻止税收增加 鉴于工资或一些政府支出可能与价格指数挂钩,政府可以通过使用不包括烟草制品 的价格指数来减少对增加烟草税的通货膨胀影响的担忧。 税收公平 以类似的方式对所有烟草制品征税。提高某些烟草制品的消费税而不增加其他产品 的消费税会导致其他产品的相对价格发生变化。这导致人们转向相对便宜的产品———例 如,从消费昂贵的卷烟换为消费其他更便宜的烟草制品,如 RYO 烟草、比迪烟、雪茄烟或 嚼烟。因此,烟草使用的总体减少幅度小于所有税收增加可比金额时的减少幅度。通过 尽量减少替代机会,对所有烟草制品增加税收,从而最大限度地提高烟草税收增加对公共 卫生的影响。此外,增加对所有烟草制品的税收将带来更大的收入增长。 严格监管未被禁止的新型烟草和尼古丁产品并征收消费税 近年来,世界一直在经历新的和新型的烟草和尼古丁产品的兴起,包括 ENDS、 ENNDS 和 HTPs。烟草业声称这些新产品比传统烟草制品更安全,但迄今为止的证据 表明,它们可能对公共健康构成威胁,特别是如果它们吸引新的或年轻的使用者或阻止现 有吸烟者戒烟的话。 新产品的市场和需求动态,以及不同社会经济群体之间的开始、戒烟和转换行为尚不 清楚。根据现有知识,对新型烟草和尼古丁产品征税的最佳做法是: 1.HTPs 应按与卷烟相同的水平征税,从结构上来说,不管烟草的含量如何,都应该 按单位征收特定的消费税。HTPs 含有烟草,应被视为烟草制品。 2.ENDS/ENNDS 产品的征税方式应不鼓励年轻人和非使用者使用。含有电子烟油 的尼古丁和非尼古丁释放系统应同等征税。 3.各国还可以考虑对用于 ENDS/ENNDs 和 HTPs 消费的设备征税,但它们需要充 分评估其这样做的管理能力。 虽然这些较新的产品给烟草控制带来了额外的挑战,但重要的是,到目前为止,卷烟 仍然是主要的烟草制品,提高卷烟的税收和价格,从而减少对它们的使用,应该仍然是重 112第五章 烟草税收政策和管理的最佳实践 中之重。 5.3　监控和评估 了解市场 很好地了解您的市场。您选择的税收结构类型及其对消费和税收的影响取决于您所 在市场的特定动态。了解您所在市场的竞争性质和程度对于选择适当类型的税收结构和 政策以实现您的公共卫生和收入目标至关重要。这些知识还将有助于更准确地估计增税 的影响,以及更好地预测行业反应。 评估您的政策的影响,以设计和实施最有效的烟草消费税政策 监测和评估对于有效的烟草税收至关重要,它们应纳入烟草税收政策的初始设计或 重新设计中。现有多种工具可帮助决策者预先评估拟议的烟草控制政策对消费、吸烟流 行率和挽救生命的影响。世界卫生组织 TaXSiM 使用目标模拟来帮助政府预测特定的 税收变化将如何影响市场中的消费价格、消费和税收。 采用可帮助您衡量税收政策改进及其影响的指标 建立和监测税收和烟草控制政策的指标有助于政策制定者评估其政策的改进情况, 并确定这些政策是否会随着时间的推移对烟草使用产生影响。特定烟草制品零售价格的 税收份额是一个关键指标,应与可负担性指标结合使用。建议各国追求的目标是征收至 少占烟草制品零售价 70% 的消费税。评估税收政策整体绩效的另一个有用指标是税收 记分卡的使用,它通过结合税收政策的4个关键组成部分(价格水平、可负担性随时间的 变化、零售价格中的总份额和消费税份额及烟草税收结构)来综合评估烟草税的最佳 做法。 5.4　税务管理 在一般税务管理中实施最佳实践方法,使烟草税务管理更加有效和高效 最佳实践方法包括:①明确界定主管当局的角色和职责;②确保国家和国际层面相关 机构之间的有效协调;③根据预先商定的指标对绩效和问责制进行评估,以找出改进之 处。 确保税务合规周期信息的合规性和准确性 为此,请执行以下操作: • 需要获得制造、进口、出口、零售、种植、运输、批发、经纪、仓储和分销烟草制品的 许可证。这将有助于保护供应链,同时获得有价值的信息,例如通过访问公司的会计和库 存系统获取相关信息。 • 确保所有参与烟草、烟草制品和制造设备供应链的人员和实体保存所有相关交易 的完整和准确记录以及用于生产烟草制品的材料的详细信息。 • 确保纳税申报表收集尽可能多的纳税人信息。 • 在生产和进口地点附近征税,以限制主管机关需要管理的纳税人数量。 212 世界卫生组织烟草税政策和管理技术手册 • 维护仓储授权系统,以对生产和存储设施进行控制,以确保缴纳税款。 • 通过最好的 IT 信息系统进行申报和征税。这允许对申报中提供的信息与来自其 他政府机构和第三方的信息进行交叉核对。 确保对供应链的控制和执行 为此,请执行以下操作: • 将控制和执法作为整体税务管理战略计划的基本支柱。 • 使用基于风险的方法选择明确的执法和控制目标,例如违规概率较高的目标。 • 在许可过程中,确保不允许从无牌供应商处购买或向未经许可的购买者销售。还 要确保许可证的有效性在时间上是有限的,并需要更新或重新申请以保持高水平的控制。 • 使用具有强大安全功能的印花税票来降低伪造印花税票的风险。这些标记有助 于征收消费税、审计和执法行动。 • 实施烟草制品跟踪和追溯系统。跟踪和追溯系统可协助当局确定烟草制品的来 源和转移点(如适用),并监测和控制烟草制品的流动及其法律地位。 • 实施反买断措施,使买断不会延迟增税及对收入和消费者行为的预期影响。 • 通过仅允许获得正式许可的自然人或法人实体开展此类活动来控制烟草制品和 制造设备的进出口。 • 加强边境管制,例如通过利用非侵入性工具,如X射线扫描仪和狗来检测烟草 制品。 • 限制或严格控制,最好是禁止在免税区内进行与烟草制品的生产和贸易有关的活 动,以避免逃税的机会。 • 从免税区运出时,禁止将烟草制品与非烟草制品混合在一个容器或任何其他类似 的运输装置中。 • 禁止向国际旅行者出售进口退税或免税烟草制品。这些销售削弱了旨在减少对 烟草制品需求的税收和价格措施的影响,并通过在税收结构中制造漏洞对政府收入产生 不利影响。 明确规定在发现烟草非法贸易后应遵循的程序 • 立即采取行动没收和销毁走私和非法烟草并征收应缴税款。 • 确保对参与非法烟草制品贸易的人实施迅速和严厉的制裁,例如处罚、罚款和吊 销许可证。根据法律将烟草制品的非法贸易视为洗钱的来源也可能是有效的。 加入和实施《世卫组织FCTC议定书》以消除烟草制品非法贸易。 《世卫组织FCTC议定书》提供了处理非法贸易的最佳做法和政策的蓝图,应成为任 何打击非法贸易战略的一部分。 尽可能对所有烟草制品以及新出现的尼古丁和烟草制品实施相同的税收管理、执法 规则、条例实施广泛的政策以确保良好的税收制度,这将通过以下方式逐步落实到烟草制 品良好的税收管理: • 确保为主管机关获得适当的资源; • 制定严格的规章制度来发现腐败并惩处从事腐败行为的人员和纳税人; • 确保一个强大在事实和认知上都是独立的司法系统,可以迅速解决争端。上诉程 312第五章 烟草税收政策和管理的最佳实践 序应该有限制,以便上诉不能持续数年。还应考虑使用刑事而非民事诉讼,特别是对于非 法贸易。 政治经济 除了税收政策和管理最佳实践的技术健全性之外,提高烟草税的一个关键因素是获 得政府最高层政治支持的能力。一项关键策略是解决对烟草税收政治经济问题,烟草业 经常利用这一问题来阻止重大改革。 恐吓战术 烟草业使用 SCARE 策略来劝阻政府实施增加烟草税。其中包括走私和非法贸易 (S)、法庭和法律挑战(C)、反贫困言论(A)、减少收入(R) 和就业影响(E)。下面描述了 应对这些策略的最佳实践。 S:走私和非法贸易 不要让增加消费税对烟草非法贸易的影响的担忧影响您增加消费税的决定。依靠您 自己对非法贸易的水平和性质的估计,而不是依靠行业的估计。 烟草制品的非法贸易仍然是税务管理者的主要关切,因为难以对其进行准确和独立 的衡量,也难以消除。行业数据扭曲了人们对问题严重程度的理解,以及对非法贸易与烟 草税之间联系的因果解释。因此,建议各国政府:①独立评估非法贸易的规模并采用最佳 统计做法,以评估问题的范围;②直接解决特定国家的体制和治理挑战,包括多边协调,并 改进税务和海关管理实践;③实施《世卫组织框架公约》中包含的打击非法贸易的最佳做 法,以消除烟草制品非法贸易。理想情况下,如果尚未成为缔约方,最好加入该公约。 C:法庭和法律挑战 不要让烟草业的法庭威胁和对增税或改革的法律挑战阻止您改进税收政策。严格遵 守设计、程序和咨询的法律要求,以加强您的法律地位,并最大限度地减少任何挑战的可 能性。 保护健康和非歧视性的烟草消费税在法律上是有道理的,行业威胁通常是毫无根据 的。但是,您可以通过谨慎对待税收措施的程序、设计和咨询来加强您的法律地位: (1)确定国内法和任何适用的国际义务要求的协商标准; (2) 在允许的范围内将烟草业与决策过程保持距离; (3) 避免在税收措施的设计、实施或执行过程中对外国烟草制品或投资者进行不必 要和不合理的歧视; (4) 不以诱因或合同承诺的形式提供投资激励,因为这些可能是本身具有约束力或 成为根据国际投资协议提出质疑的理由。 A:反贫困言论 不要让对提高烟草税的累退性的担忧阻止烟草税的增加。 近年来,关于烟草税和税收增加对分配的积极影响的证据显著增加。事实上,一旦这 些更广泛的考虑因素得到适当考虑,烟草税和税收增加实际上是一项累进或有利于穷人 的政策。在游说反对增税的过程中,烟草业经常声称烟草税会伤害穷人。这一论点基于 与税收有关的累退性概念。从概念上讲,如果税收意味着低收入人群必须支付相对较大 比例的家庭收入来满足纳税义务,那么税收是累退的。但是,该行业的论点有两个局限 412 世界卫生组织烟草税政策和管理技术手册 性。首先,仅基于税收负担的累退性概念没有考虑较低社会经济群体较多地经历吸烟造 成的更广泛的健康和经济危害。其次,较高的烟草税和价格会导致人口的行为改变,正如 需求的价格弹性所反映的那样,这意味着低收入的吸烟者将最大限度地减少吸烟,因此将 从减少烟草的消费和使用中获得较多的健康收益。事实上,这些更广泛的考虑使烟草税 成为一种累进而非累退的公共卫生干预措施。 R:收入减少 不要让对潜在收入减少的恐惧阻止您提高烟草制品的消费税。 即使在税收已经很高的国家,增加税收也会带来额外的收入。烟草控制反对者认为 增加税收不会导致收入增加的论点是没有根据的。卷烟需求的价格相对缺乏弹性,再加 上税收份额低和税收没有过度转移,这意味着对于大多数(如果不是全部)国家来说,收入 的增加将伴随着税收的增加。如果加税经过精心设计并且税收管理是有效的,那么加税 就极不可能导致收入减少。 E:就业影响 不要让对就业影响的担忧来阻止烟草税的增加。 烟草业通常试图将烟草税视为经济问题而不是公共健康问题。特别强调了所谓的增 税对烟草种植和制造以及相关行业的就业构成的威胁。然而,这种所谓的健康和工作之 间的选择很大程度上是基于夸大其词。烟草业夸大了烟草就业相对于全国就业总量的重 要性,并夸大了地方税收减少和国内需求对服务于全球市场的烟农的影响。烟草业使用 的论点也忽略了这样一个事实,即烟草支出并没有消失,而是被重新分配到可以产生类似 或更多工作岗位的其他消费中。案例研究证明了政府支持农民过渡到种植其他作物的可 能性和方法,这些作物提供类似且通常更好的回报和更大的可持续性。 专款专用 考虑将烟草税收入指定用于以健康为重点的计划,特别是如果它有助于推进烟草控 制工作,更具体地说,有助于实施大幅提高烟草税和税收改革。这可能会带来额外的好 处,即为资金不足或未被优先考虑的卫生计划提供资金。 从烟草控制的角度来看,最好将烟草税专款专用理解为向公众、政治家和官员推销大 幅增加的烟草税的一种方式。专项拨款是改善烟草税收政治经济学的一种工具,在减少 烟草需求的主要目标之后,这只是次要问题。有证据表明,当至少部分增加的收入明确用 于支持以健康为重点的计划时,公众对提高烟草税的支持会更大。目前的证据表明,有效 指定用于卫生计划的金额相对较小,因此不太可能导致政府预算僵化。与此同时,在一些 国家,这些资金帮助实施了急需的资源不足的卫生计划。随着越来越少的人生病并且需 求减少的与烟草相关疾病的医疗保健,未来将看到回报。将烟草税收入用于资助高负担/ 低优先级的健康计划可以为提高对此类计划的重要性及其有效性的认识铺平道路,从而 说服政府重新确定其优先事项并承诺将这些计划纳入正常预算。 512第五章 烟草税收政策和管理的最佳实践 5.5　烟草税改革清单（适用于税收政策制定者） 第1步 关注烟草税的目标。 烟草税政策的目标不仅应该是增加收入,还应该是减少消费和改善健康。为了增加 收入和减少消费,您需要:①简化烟草税结构;②显著提高税率以影响价格水平;③定期调 整税率以至少考虑通货膨胀和收入增长。 第2步 分析税收结构并确定其弱点。 您需要分析和识别当前税收结构的问题,以了解下一步要考虑的步骤。用哪种税收 结构:从量税、从价税、混合税或无消费税? 第3步 确定对现有税收政策/结构的改进。 您目前的税收结构和税收状况将为您提供下一步理想的实现步骤,以实现第一步的 目标。 从量税: 1.确保税收根据通货膨胀和收入增长的影响自动调整。 2.确保禁止所有价格促销活动。 从价税: 1.确保从价计税的税基为零售价。 2.引入高从量税部分(和最低从量税)。 3.确保特定消费税和最低特定消费税根据通货膨胀和收入增长效应自动调整。 4.确保禁止所有价格促销活动。 混合税: 1.确保从价部分的计税基础是零售价。 2.确保您使用高特定消费税成分和最低特定消费税成分。 3.确保特定消费税和最低特定消费税根据通货膨胀和收入增长效应自动调整。 4.确保禁止所有价格促销活动。 无消费税: 1.引入高从量消费税。 2.确保利率根据通货膨胀和收入增长的影响自动调整。 3.确保禁止所有价格促销活动。 第4步 评估烟草税的政治经济学 改革必须从评估烟草税的政治经济学开始:①从过去的成功和失败中吸取教训——— 哪里出了问题,哪里做得对,这次您可以采取哪些不同的做法;②评估改革的优势和劣势、 可能的机遇和风险;③根据过去的改革和现状,确定政府内外改革的主要支持者和反对 者;④预测将用于反对改革的论点(参见 SCARE 策略)。 第5步 制定实施改革的计划 关注第 1 步确定的总体目标、第 2 步和第 3 步确定的实现这些目标的步骤以及第 4 步确定的围绕本次改革的政治经济学,准备您的计划: 612 世界卫生组织烟草税政策和管理技术手册 1.明确改革的不可谈判目标以及您准备为实现这些目标而进行的权衡。 2.制订计划以接近潜在盟友并争取他们参与改革工作。 3.确定回应之前确定 SCARE 论点所需的反驳论点。 4.提前准备您需要的证据。为此,请获得学术界和相关政府机构间的支持。 第6步 动员改革联盟 1.制订战略沟通计划:旨在获得最高层和公众的政治支持(将烟草税定为健康问题有 助于在许多国家赢得政治支持)。 2.确定政府中的拥护者:确保财政和卫生官员步调一致,从一开始就让执行部门参与 进来,例如执法机构。 3.动员学术界、民间社会和私营部门的盟友来应对来自烟草业、其代理人及其盟友的 预期抵制。 第7步 监测和评估 为了做出最明智的政策决定,应监测改革结果以评估其总体影响及对关键指标的影 响,这将有助于确定需要解决的问题,同时也为进一步的改革努力创造强有力的证据 基础。 获取并分析相关数据,以更好地了解市场情况及其动态: 1.监控市场及其演变(例如零售价格、完税销售、市场份额)。 2.定期估计价格弹性(包括交叉价格弹性)、收入弹性和税基弹性,以评估烟草需求的 任何变化。 使用相关工具评估税收政策对消费和收入的影响: 1.使用特定工具来分析消费税对价格、消费和收入的影响(例如世界卫生组织 TaXSiM)。 2.使用全球工具评估增税对流行率的影响(例如世界卫生组织 ISPT)。 密切监控关键指标以评估随时间推移的改进情况: 1.税收占零售价格的百分比变动情况。 2.烟草制品的可负担性随时间变化情况。 3.烟草税计分卡的变化,它结合了税收政策中的最佳做法组合。 4.烟草制品销售、流行和非法贸易的变化情况。 5.MPOWER 各项措施的落实进展。烟草税改革清单 712第五章 烟草税收政策和管理的最佳实践