Environmental policy and the public revenue in developing countries

Aj L) - 0c) 34 THE WORLD BANK SECTOR POLICY AND RESEARCH STAFF Environment Department Environmental Policy and the Public Revenue in Developing Countries Dennis Anderson July 1990 Snvironment Working Paper No. 36 This paper has been prepared for internal use. The views and interpretations herein are those of the author(s) and should not be attributed to the World Bank, to its affiliated organizations or to any individual acting on their behalf. This paper has been prepared by Mr. Dennis Anderson, Consultant to the Environment Department. The author wishes to acknowledge critical comments and insights in early drafts of this paper provided by Jeremy Warford, John English, Mohan Manasinghe, Emanuel Jimenez, and Gunnar Eskeland. The paper reports on findings that are particularly significant in the context of the overall work program of the Environmental Policy and Research Division. Departmental Working Papers are not formal publications of the World Bank. They present preliminary and unpolished results of country analysis or research that are circulated to encourage discussion and comment; citation and the use of such a paper should take account of its provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Because of the informality and to present the results of research with the least possible deloy, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. - ii - Abstract The paper analyzes approaches to environmental policy in developing countries. It reviews the policies of taxing pollution and the alternatives of regulating it, and the roles of public and private investment in environmental improvement. By --eference to a wide range of environmental issues, it shows that the 'tax approach' to environmental policy will generally be the better instrument in many circumstances, regulation (e.g., of toxic wastes) in others, public investment (e.g., in afforestation and soil conservation) in others, and sometimes a combination of the three. However, the developing countries would be better served by following the tax and investment approaches to environmental policy in most situations, rather than the regulatory route that the industrial countries have followed (often exclusively) over the past century. The paper also reviews issues to be addressed in the implementation of policies: their effects on efficiency and growth, on income distribution, on employment and earnings opportunities, and on trade and investment; the macro-economic context (stability being crucially important for the implementation of successful policies); their administrative requirements; and, not least, their financial requirements and 'affordability'. An underlying ieme is that policies can be designed to be both socially beneficial and environmentally beneficial. Indeed, this should be the primary ai. when setting environmental standards and deciding on the instruments of policy to be used to meet them. Preas for future research are also discussed, ENVIRONMENTAL POLICY AND THE PUBLIC REVENUE IN DEVELOPING COUNTRIES - CONTENTS - INTRODUCTION PART I ECONOMIC GROWTH AND THE ENVIRONMENT 1.1 Linkages 1.2 Parallels with Material Investment 1.3 Damage Accumulation and Irreversibilities 1.4. Value Added by Environmental Maintenance and Investment (EMI) PART II ECONOMIC EFFICIENCY, THE ENVIRONMENT AND THE PUBLIC REVENUE: SOME EVIDENCE 2.1 Microeconomic Evidence 2.2 The Macroeconomic Context PART III CHANGING PERSPECTIVES ON ENVIRONMENTAL TAXES 3.1 Introduction 3.2 Some Candidates for Environmental Taxes and Expenditures PART IV ENVIRONMENTAL TAX ANALYSIS 4.1 Key Issues 4.2 Economic Efficiency and Innovation 4.2.1 Costs 4.2.2 Innovation and Substitution 4.2.3 Environmental Improvements Over Time 4.3. Regulatory Alternatives 4.3.1 Direct Regulation 4.3.2 Pollution Permits 4.4 Environmental Taxes and Revenue Requirements 4.4.1 Tax and Revenue Goals 4.4.2 Estimating Suitable Tax Levels 4.4.3 Private and Public Components of Environmental Costs 4.4.4 Uncertainties 4.4.5 Financial Requirements of the Regulatory Route 4.4.6 Where Financial Self-Sufficiency is More Difficult to Achieve 4.5 Subsidies and Direct Public Investment 4.5.1 Areas where the Approach is Beneficial 4.5.2 Areas where the Approach has Drawbacks 4.6 Income Distribution Issues 4.6.1 Labour Demand and Productivity 4.6.2 Health and Living Conditions 4.6.3 Compensatory Policies 4.7 International Trade and Investment 4.8 Macroeconomic and Second-Best Issues PART V AREAS FOR RESEARCH PART VI CONCLUSIONS AND SUMMARY APPENDIX ENVIRONMENTAL TAX THEORY - Current Status and Areas for Research A.1.1 Standard Results A.1.2 Technical Change, Substitution and Simulation A.1.3 The Responses of the Polluted Parties A.1.4 Enlargement of the Model A.1.5 Uncertainty and Costs Annex Tables Al, A2, A3 REFERENCES ENVIRONMENTAL POLICY AND THE PUBLIC REVENUE IN DEVELOPING COUNTRIES' INTRODUCTION How might environmental!y desirable programmes and policies be financed in developing countries? The range of environmental concerns is now very wide, and few sectors of economic activity are untouched by them. At the same time, there is no shortage of good proposals, based on field experience and supported by ex post analysis, that would lead to environmental improvements if implemented: in forestry, in soil erosion control programmes, in the protection and management of wildlife and game reserves, in the long-familiar but still important areas of water supply, sewerage and solid waste disposal, in fuel and energy conservation, in congestion reduction, and (not lastly) in the treatment of industrial emissions and effluents. But claims on public and private financial resources are already large, so how "affordable" are environmental policies? Is there a role for taxes or user charges, to reduce environmental damage, raise resources for environmental maintenance and investment, and create profit opportunities for industry and agriculture in environmentally more benign processes and products? Or should the developing countries follow the regulatory route of the industrial countries? In either case, what might be the costs in terms of economic growth? The purposes of this paper are to discuss these questions and examine the role of environmental taxes in developing countries. The use of regulations is self-evidently appropriate in some cases, e.g. for hazardous wastes. In others, a combination of the regulatory and tax approaches may be merited, with the taxes serving revenue-raising aims. However, in the majority of cases, the tax approach has advantages that are well worth pursuing: it would be economically efficient and more wide-reaching in its impact; less demanding on information, and indeed would be especially suited to situations where information is poor; it would be administratively less burdensome, and could be grafted onto existing structures of tax administration; it would be less prone to problems of bribery and non-compliance; and, for a period, it would raise revenues. Economists have often held most such advantages to be applicable in the industrial counries. But they would apply with even greater force in the developing countries. In setting out the case, the paper reviews some practical issues that would be encountered in the implementation of the tax approach. 1 1 should like to acknowledge the most helpful and constructive comments I have received on this paper from Jerry Warford, John English, Mohan Munasinghc, Emanuel Jimenez and Gunnar Eskeland. 1 To provide a context, Part I of the paper reviews the linkages between environmental policy and growth and Part II summarises some evidence on the linkages. It is well known among environmental economists, though perhaps less so among those working on public revenue policies, (a) that environmental damage or neglect is widely being subsidised in developing countries,2 not taxed as it should be, and (b) that environmental programmes, by reducing external costs, often have good social rates of return. In both respects good environmental policies would be twice blessed, leading to environmental and economic improvements. Further, as discussed in Part IV, environmental policies may sometimes help in the alleviation of poverty. The linkages are not always clear-cut; for example, the coal industry may be seriously affected by sulphur and carbon taxes in some countries, industrial and developing alike, and the design of compensation and restructuring programmes to avoid adverse consequences for employment and earnings opportunities would self-evidently be important. In other areas the linkages may be wholly positive, most notably in soil conservation, afforestation and agricultural development programmes, and, in urban areas, the health benefits of emissions control, water supply and sewerage systems. Turning to the analysis of policy, Part III discusses changing perceptions on the role of environmental taxes in public policy, and raises the key questions to be addressed. Part IV presents the formal analysis of tax policies, and their comparison with the regulatory alternatives, Part V some areas for research and Part VI the conclusions. Perhaps ambitiously, the paper is addressed to two audiences (and thus may require the patience of both). One is economists and non-economists interested in the operational aspects of environmental policy; the other, economists interested in applied research. Some areas for research are raised in the course of the analysis and brought together in Part V. However, notwithstanding the recent upsurge of political interest in the subject, environmental economics is a long and firmly established subject.4 While there are, of course, areas for applied research and uncertainties to be addressed, there is much that can be recommended operationally on the basis of present knowledge. 2 See Repetto (1988) for an excellent discussion, and papers in Schramm and Warford (1989) for a representative view. These and other papers are reviewed further below. 3 This is also the theme of the literature on sustainable development, as it has come to be called. See the Brundtland Report (1987) and Pearce, Markandya and Barbier (1989). 4 See Dasgupta and Maler (1990) for a recent review, Pearce et al. (1989) and Beckerman (1974). Also the recent edition of the Oxford Review of Economic Policv, edited by Helm and Pearce (1990). PART I: ECONOMIC GROWTH AND THE ENVIRONMENT 1.1 Linkages In the analysis of economic growth, it is usual to divide income flows into two parts: (a) public and private consumption, and (b) gross investment. In the developing countries, gross investment rates have been in the range 20 to 25%, 30% in the newly industrialised countries, rather higher if (as it should be) investment in education were included. Investment rates fell dramatically in the heavily indebted countries in the 1980s on account of the crises of debt, fiscal deficits, inflation and the policies that brought them about. But in more stable periods and - an important condition - in a policy environment conducive to good private and social returns to investment, investment rates of 20 to 25% or more have been shown to be achievable, and are the major factor in economic growth. The ways in which the (socially efficient) accumulation and use of human and material assets combine to produce growth are represented schematically by the left-hand loop of Figure 1. FIGURE 1 LINKAGES BETWEEN GROWTH AND THE ENVIRONMENT + HUMAN AND ECONOMIC ' sRESIDUALS ENVIRONMENTAL ASSETS: MATERIAL OUTPUT aAtmosplim ASSETS o Forests e Seas - *Watar * siodiversity * Minerals+ + + + + NATIONAL INVESTMENT + (Health, EducatioL Industiy, Public services, ENVIRONMENTAL Infastructure R & 0) MAINTENANCE and INVESTMENT CONSUMPTION SOCIAL CHOICE 5 It is clearly a simple capital accumulation or investment model of growth, but is sufficient for the following analysis. The investment approach to growth has recently been revived by Scott (1989), in an important work. See also Sen (1983) and Anderson (1990b). Data on gross investment rates since 1965 are presented in the Statistical Annexes of the World Development Reports. 3 For the analysis of environmental issues it is necessary to show the environment as a separate input into the growth process. Land and mineral resources are traditionally treated in this way, but forests, water resources, the atmosphere and climate need to be included as well. This is the upper right-hand loop of the figure, which also indicates the discharge of residuals into the environment as a consequence of producer and consumer activities. Until recently the effects of changes in atmosphere and climate on output were usually regarded as random and exogenous variables about which little could be done. In actuality, however, much expense and effort are made to change these variables, or at least to counteract the effects of human activity on them. The reduction of urban smogs in industrial countries, the use of wind-break and agro-forestry programmes to improve micro-climatic conditions in agricultural regions. controls on acidifying gases and toxic emissions, and the recent CFC Accord, are all examples of where appreciable resources have been expended to bring about improvements in atmosphere or climate, while policies to reduce greenhouse gas accumulations may soon b- another. In the much discredited Club of Rome Report,6 it was assumed that the drawdown of environmental assets and the discharge of residuals would continue exponentially as growth proceeded, and would ultimately lead to environmental and social collapse. But (as economists were quick to point out) what the analysis neglected were the resources that may be diverted from consumption and investment into what are termed here environmental maintenance and in'estment activities (the lower right-hand loop). Indeed, as discussed below, they are an aspect of investment and maintenance as they are usually defined in economic analysis, and a significant part of value added in a country pursuing enlightened policies. Examples are numerous and include various forms of recycling; the use of scrubbers and catalytic convertors to reduce emissions; direct investment in soil erosion control programmes, forests and watersheds; investments in water supplies, sewerage systems and waste disposal; conservation measures of various kinds; and R & D and investment into environmentally more benign substitutes, products and processes. It is the potential of such activities to bring about environmental and economic improvements that has led to the idea of sustainable development, as it has come to be called. Environmental maintenance and investment is in other words a busy and innovative field, a rising source of value added in many economies and potentially so in others that have so far neglected the environment. From an economic perspective, therefore, the important questions are: (1) whether the resources being allocated to environmental maintenance and investment activities are sufficient; and, (2) what sorts of policies will bring a desirable level of maintenance and investment about? 6 The Limits to Growth (1972), by Meadows, Meadows, Randers and Behrens. 4 These issues are considered in Parts II, III and IV. But first consider some other conclusions that emerge from the above framework for thinking about growth and tle environment.7 1.2 Parallels with Material Investment The parallels between environmental maintenance and investment, and investment and maintenance as they are usually defined, are precke. As with what Scott (1989) aptly terms the required maintenance of material assets, environmental maintenance is a use of resources - labour, managerial, material and other - to maintain income and assets intact. Environmental investment, as with material investment, unavoidably involves environmental change; but as with soil conservation programmes, afforestation, investments in water and sewerage systems, waste disposal, and other areas, the aim is to change the environment sustainably and, often, with an environmental improvement. Indeed, the economic desirability of environmental investment and maintenance activities, as with material investment and maintenance, can in principle be assessed by cost - benefit methods, and there are now some examples available. The economic benefits of reducing or avoiding environmental damage are difficult to quantify in many situations; however, this does not undermine the principles involved. Rules for pricing or user charges also readily follow, with the appropriate prices or user charges being related to the marginal costs of avoiding or reducing environmental damage to desirable levels. These rules are discussed later, but the basic point is that the economist's "tools of the trade" apply, whether we are concerned with assessing the desirability of an environmentally-related investment, or an environmental tax or user charge policy. 1.3 Damage Accumulation and Irreversibilities Another parallel with material investment is that if maintenance is neglected, assets deteriorate. The primary aim of environmental policy should be to raise maintenance activities to required levels. However, there are several forces working against this. First, lags in environmental processes are often very long and uncertainties as to the causes of environmental damage appreciable. This is not only the case with issues like the greenhouse effect, where both the lags and the uncertainties have received much public discussion, but with the extent to which soils are eroding and desertification is happening in developing regions,9 or how micro- and regional climates may be changing. Over large areas, systems of measurement of ecological change, or even of local climatic conditions, are not in place or at 7 For a further discussion, see Anderson (1990a). More formal frameworks on similar issues are discussed in Pezzey (1989). 8 Schramm and Warford (1989) provide a number of examples from developing countries. For a recent review of methods see Pearce and Markandya (1989). 9 Stanford (1983), Nelson (1988), Southgate (1988). 5 best are very thin on the ground.'0 In the history of environmental policy, it is noticeable how often uncertainty has led to divisivc iess and paralysis in policy-making, and the long lags to a much-used excuse for procrastination. Second, a large number of present-day environmental problems arise from common property problems (atmosphere, common lands and forests, ground water resources) and other sources of external cost (downstream siltation and flooding from the clearance of trees in watersheds, for instance). While the aim of public policy is to address these problems by taxes, subsidies, regulation, institutional changes on the matter of property rights, or by direct expenditures, the starting-point is often one in which required maintenance has been neglected for many years. Hence what might have been a routine maintenance problem at one time later requires major expenditures on environmental defence, restoration and clean-up. Related to this are the political opportunities for exploiting a common resource when property rights are ill-defined, and for patronage - e.g. in the issuance of rights or licences to exploit the resource. Weak administrative and legal structures, and the lack of a satisfactory system of public accountability for those involved with decisions affecting the use of the resource, serve to compound the difficulties of arriving at a satisfactory policy. This issue is discussed later in the comparison of the tax or user charge approach with the regulatory alternative. Hence the initial conditions for policy-making are not ones where the aim is simply to bring maintenance up to desirable levels, but of making up for huge backlogs as well. This of course poses special problems for finance and adjustment, all the more so given the institutional difficulties just mentioned. 1.4 Value Added by Environmental Maintenance and Investment Another point that emerges from the above framework for analysis is that environmental policies ultimately require the redeployment of labour and capital into environmental maintenance and investment." The policies may thus become significant and new sources of employment and earnings opportunities - e.g. in soil erosion control programmes, forestry management, and in the construction of water, sewerage and waste disposal systems. The historical record shows that they are also potentially important sources of innovation both technologically (e.g. in emissions reductions and the treatment of water) and managerially (e.g. in the management of land and forestry resources). Under an appropriate policy 10 See e.g. Nicholson (1982) on the shortcomings in monitoring the climate in the Sahel. 11 The value added by such activities should appear in the National Accounts, and provides a good measure of a country's actual commitment to environmental policy. Although many economists have criticised national accounting concepts for omitting what is termed "environmental depreciation", this criticism is less valid once policies are in place because environmental concerns will ideally then be internalised in the pricing-cum-taxation system and reflected in value added as normally measured. For different views on accounting for the environment see Repetto (1989), Pearce et al. (1989, Ch. 4), Ahmed, Serafy and Lutz (1989) and Dasgupta and Maler (1990). 6 framework therefore they offer good social and private rates of return to capital - indeed, this is a requirement of good policies. These topics will be discussed in Part IV, in connection with the distributional and efficiency aspects of policy. 7 PART II: ECONOMIC EFFICIENCY, THE ENVIRONMENT AND THE PUBLIC REVENUE: SOME EVIDENCE From an economic viewpoint, the objective of environmental policies, whether in the form of a tax, user charge, regulation, direct investment or other, is to achieve an environmental and an economic improvement. A tax or regulation on a polluting product or process for example reduces pollution directly and provides an incentive for the introduction of less polluting products or processes that are otherwise similar. Economic efficiency is improved because the full costs of the product or process, including pollution costs, are then taken into account in consumers' and producers' decisions. Even where there is much uncertainty (as there often is about the costs of pollution or environmental damage). and environmental standards are based on value judgements or other criteria, the use of economic instruments can still lead to economic benefits by lowering the costs to producers and consumers of complying with the standards. Economic and environmental improvements are thus a standard expectation of good environmental policies, and there is now much evidence from developing countries and the historical experiences of industrial countries that both can be achieved. Indeed, as noted earlier, environmental damage is often being subsidised not taxed in developing countries, while financial constraints widely restrict environmentally desirable investments with potentially good social rates of return. Hence the possible gains from environmental policy, as Repetto and Warford have argued in a number of papers,'2 may be large because we are not starting from a position where external costs are merely neglected, but are greatly amplified by prevailing policies. These points have been particularly well illustrated by case studies in particular sectors - rainforests and forestry policy. for example, in agriculture and soil erosion, in urban congestion and other sectors. Some results are summarised below in the section on microeconomic evidence. But there are also reasons for thinking that macroeconomic stability and the broad structure of fiscal incentives are fundamentally important for environmental policies to be successful; they are discussed in the section on the macroeconomic context. 2.1 Microeconomic Evidence The following examples show that good environmental policies are (or would) very often be economically beneficial. Indeed, the converse also applies: good economic policies would very often be environmentally beneficial also. The examples are wide ranging. All are well reported in the references cited, and so the summaries are brief. 12 See Repetto (1988), Warford (1989a) and the World Bank's Report to its Development Committee, Warford (1989b). 8 (i) Rainforests. This has become the classical example of where good economic policies would lead to environmental improvements. It has been widely reported.! Table 1 summarises the subsidies that have gone into the unsustainable exploitation of the resource in Brazil. Simulation studies of land clearance and livestock ranching in the region indicate that the real private rates of return to investment may reach 20% or more if the land is grazed over-intensively, while the social returns - the returns that would obtain without the subsidies - would be zero or negative. A removal of the subsidies would probably eliminate most damage, improve resource allocation and reduce fiscal strains. As will be seen, this is not an isolated case of environmental damage being subsidised, though it is perhaps the most calamitous example. Table 1: Brazil's Rainforests: Social and Private Returns to Land Clearance and Ranchinp Private Returns Social Returns Low Cattle Price Assumptions: "Appropriate" Grazing Intensity 6 -6 Overgrazing 18 -7 High Cattle Price Assumptions: "Appropriate" Grazing Intensity 18 2 Overgrazing 24 0 Source: Quoted by Mahar (1989) from Hecht and others. (Assumes 15% increase in land values.) Main Forms of Subsidy: (1) Investment Tax. 50% credit against federal income tax liabilities if resulting savings invested in Amazonia. (2) Tax Shelters: 80 to 90% of agricultural income exempted; fixed investments, animals, machines and vehicles can be depreciated in first year, with rollover of unused depreciation allowances for four years. (3) Land Allocations: Claimants receive titles for up to 3 x area cleared. Titles used for collateral. (4) Credit: 20-year credit programme for ranchers at nominal rate of 12% (POLAMAZONIA). (Defaults on credit are another possible subsidy.) (5) Infrastructure Programmes: Not quantified above. (6) Stumpage: Government rent capture is very small (not quantified above). 13 See Mahar (1989) and Binswanger (1987). 9 (ii) Forestry Industry. However, the effects of such subsidies are not confined to the tropical forests alone, but must bear adversely on the development of the forestry industry in general. Cost curves for the industry are not readily available. But assuming, outside the tropical forests, it has to provide for its own renewal and maintenance, and operate without subsidy, the subsidies probably reduce the returns to and output of the plantation side of the industry - an industry in which developing countries have a comparative advantage. The actual picture is more complicated than suggested in Figure 2, since it is necessary to allow for different qualities and types of timber, for the build-up of stocks, and other factors. But the basic point applies nevertheless. FIGURE 2 OPEN FELLING, COSTS AND PRICES IN FORESTRY PRICE AND MARGINAL World MC Cure COSTIMC) Without open Felling Price Without Plantation Forestry - Open Felling 7 Id MC b' margin Word MC Cuve With Open Felling a Price With . di Open Felling c bs margin a a Worm Price- CDemand Curve Financial Coste of Open FellingI Open Felling ve Plantation *I I Forestry I ANNUAL VOLUME OF TIMBER (iii) Agro - Forestry. Agro-forestry practices, including the planting of wind-breaks, have multiple benefits to agro-pastoral activities. They stabilise and improve the nutrient context of soils; provide wood, fruit and fodder; and, by reducing surface wind velocities and surface run-off, raise the moisture content and fertility of soils. They also have low unit costs and, where farmer response is good, potentially high social rates of return.15 They have attracted much attention in recent years because of their potential for addressing desertification problems 14 Brandle, Hintz and Sturrock (1988), Gregersen, Draper and Elz (1989), Spears (1988), Wiersum (1984), Raintree (1984), Food and Agriculture Organisation (1985), Prinsley and Swift (1986), Anderson (1987) and the World Bank (1978). 15 Anderson (1987). 10 and improving microclimates; however, even without allowing for these factors, their own direct products (wood, mulch, fruit, fodder) and effects on soil moisture and nutrients can lead to good social returns. As with forestry management in general, the issue here is that they are generally given a low priority in expenditure policies when available analysis suggests they warrant a high priority - on economic as well as on environmental grounds. (iv) Agro - Chemicals.'6 Large subsidies for agro-chemicals in developing countries seem more the rule than the exception, and are certainly not environmentally benign. Table 2 on pesticides is taken from Repetto (1989). This is another example of pollution wrongly being subsidised instead of being taxed. Table 2: Pesticide Subsidies in Selected Countries Subsidies as % Annual of Retail Costs Value, $m China 19 285 Columbia 44 69 Ecuador 41 14 Egypt 83 207 Ghana 67 20 Honduras 29 12 Indonesia 82 128 Senegal 89 4 (v) Urban Congestion Pricing.17 Notwithstanding its well-researched potential, urban congestion pricing has been a much neglected policy. The Singapore Area Licensing Scheme introduced in 1976 is perhaps the best example of its effectiveness in reducing congestion.18 The Scheme worked with an administratively simple licensing system and would not be difficult to apply in developing countries. Its main features are summarised in Table 3. Aside from reducing congestion and improving urban transport efficiency, additional benefits would include (a) its potential for raising revenues for the finance of urban services and infrastructure improvements, and for earning a surplus, and (b) appreciable reductions in fuel consumption and vehicle emissions. 16 See Repetto (1989) for a review. 17 Walters (1968), Churchill (1972), Goodwin and Jones (1989) and Goodwin (1989). 18 See the recent evaluations of the Oxford University Transport Studies Unit, by Goodwin and Jones (1989). 11 Table 3: Singapore Area Licensing Scheme Features: Permit for entry to central business district at peak times. By-pass routes available. Complementary policies to encourage switching. High parking charges. 1976 1989 Hours 7.30 - 10.25 7.30 - 10.15 16.30 - 19.00 Charges ($/day) Cars 4 3 Company Cars 8 6 Taxis 3 3 Motorcycles - 1 Buses - 3 Car Pools (4 or more people) - 3 Other - 3 (vi) Water Utility Pricing Polices.'9 Water supply and sewerage systems are heavily regulated and subsidised in developing countries. Financial returns are frequently insufficient to cover operation and maintenance (0 & M) costs, let alone provide a good return to investment. Table 4 summarises recent tariff data for Africa, which should be compared with typical LRMCs (long-run marginal costs) of about $1.0 per cubic metre, and 0 & M costs of $0.3. The upshot is underinvestment in an environmentally - as well as socially and economically - important sector and a needless burden on the public revenue. The policy also increases losses and waste. This is a case where cost-reflecting pricing policies would lead to (a) an economic improvement, (b) an environmental improvement and (c) a saving in public revenues. Social or distributional concerns can be met via the tariff structure (low first blocks) and suitable investment policies (emphasis on yard taps and stand-pipes); the evidence also shows that policies which have subsidised supplies have generally favoured high-income consumers while restricting expansion in low-income areas. 19 See Warford (1966), Warford, Pellegrini, Kneese and Maler (1974), Julius and Warford (1979) and Arthur and Kirk (1985). 12 Table 4: Water Supply Tariffs for Some African Countries Country Year Tariff/cu. m. Exchange Tariff rate/$ $/cu. m. Cte d'Ivoire 1987 330 CFAF 300 1.1 Cameroon 1987 242 CFAF 300 0.8 Congo 1987 130 CFAF 300 0.4 Niger 1986 135 CFAF 314 0.4 Togo 1987 180 CFAF 300 0.6 Mali 1986 119 CFAF 314 0.4 Senegal 1986 88 CFAF 314 0.3 Benin 1986 86 CFAF 314 0.3 Ghana 1986 22 Cedis 150 0.15 Gambia 1986 1.6 Dalasis 7.5 0.2 Nigeria 1987 Niger State 1.1 Naira 4.0 0.28 Lagos State 0.35 Naira 4.0 0.09 Source: World Bank project files. The data are publicly available in each of the countries listed. See Anderson (1989). (vii) Electric Utilities and the Fuel Industry.20 Much the same can be said, though with less force, for the electric utilities. Real financial rates of return have been held below the opportunity cost of capital in many countries (Table 5),21 leading to (a) an enlarged dependence on the public revenue for capital financing (b) over-expansion of demand along with (c) financial constraints on supply expansion. Recent concerns about emissions from the electric industry also argue for cost-reflecting pricing policies, perhaps with emissions taxes. Fuel taxes in general are also favoured as an approach to improve conservation and raise revenues, while being administratively convenient. 20 Munasinghe and Warford (1982) and Turvey and Anderson (1977). Schramm (1985) on energy in Africa. 21 Munasinghe, Gilling and Mason (1988). 22 They are often discussed in relation to the greenhouse effect, and are a feasible option for developing countries. 13 Table 5: Distribution of Financial Rates of Return on the Assets of Electric Utility Companies in Developinp Countries Rate of Return For Projects For Projects at Project Completion, % Approved in 1966-78 Approved in 1979-83 No. of projects % No. of projects % Below 4.0 23 22 13 40 4.1 - 6.0 20 20 5 15 6.1 - 8.0 17 17 4 12 8.1 - 10.0 13 13 7 21 10.1 - 12.0 12 12 0 0 12.1 - 14.0 7 7 0 0 Over 14.0 10 10 4 12 102 100 33 100 Source: Munasinghe, Gilling and Mason (1988), based on analysis of electric utility data in World Bank project files. Note: Ideally the returns should be estimated on revalued assets, not book values. The authors comment: "For projects approved between 1966-78, 22% of utilities had rates of return equal to or less than 4% at project completion, and 17% had rates of return of over 12%. The situation has deteriorated for ongoing projects in 1984, when 40% of utilities had rates of return of 4% or less and only 12% had rates of return over 12%. Part of this deterioration was due to more rates of return being based on revalued assets." The above list does not of course exhaust the range of examples. All the sectors mentioned above, with the partial exception, perhaps, of agro-forestry and forestry management, are capable of being self-financing. But there are others which are not, and are more dependent on public revenue. The provision of funds for watershed management, soil erosion control programmes, and the protection and maintenance of wildlife reserves are further examples where good social rates of return are achievable from environmentally desirable activities. The above examples show that the relationships between socially efficient economic policies on the one hand, and environmental concerns on the other, are generally favourable. In several cases (energy and congestion pricing) there would be public revenue gains as well as economic and environmental improvements; in others (forestry, water and sewerage), revenue savings. Further, the gains in revenues would leave the authorities better placed to finance environmentally desirable investments with good social returns (in wildlife and game parks, for example, soil conservation, and the protection and management of forests) that are less capable of being self-financing. The economic case for bringing environmental concerns into the analysis of public revenue and expenditure policies is therefore a strong one on three grounds - economic efficiency, finance, and of course the environment. Indeed, as will soon be seen, there is often a good case on distributional grounds too. 14 2.2 The Macroeconomic Context Turning to macroeconomic policies and the environment, the evidence is less concrete, but nevertheless points in the same direction. Consider first the issue of macroeconomic stability in aggregate before turning to some more specific issues raised by the structure of macroeconomic policies. Aggregate Balances.- Most economists' writings on the environment focus on the microeconomics of market failure. There are, however, two other sources of failure in economic systems, quite separate from that of market failure. One is, of course, a failure of government administration, with Eastern Europe providing an example of high levels of intervention combined with chronic pollution damage. The other arises from "short-termism" brought about by financial instabilities, inflation, fiscal crises or foreign exchange crises, or sometimes (in regions of Latin America and Africa) all such factors together. As Beckerman has remarked, many of the environmental concerns prominent in Europe and the US in the early 1970s were soon left off the agenda in the wake of the oil price shocks, inflation and the severe recession of 1979-82, and it is perhaps not coincidental that with the restoration of growth and stability environmental concerns have resurfaced again.2 In the developing countries - and in the highly indebted developing countries in particular - macroeconomic conditions have often been highly unfavourable to investment and maintenance activities, whether in material or environmental assets. In the highly indebted countries, gross savings and investment rates declined by one-quarter to one-half in the 1980s, from 20 to 25% of GNP to 15 to 20%, sometimes less, and the pressures to disinvest in natural resources, as in material assets (for example ;n roads) by neglecting required maintenance, were appreciable. The restoration of investment and maintenance activities to desirable levels may thus depend crucially on the restoration of macroeconomic stability and fiscal balances. More generally, Killick has noted the following ways in which macroeconomic stability facilitates environmental policy-making: Confidence in the future. The pursuit and achievement of stability reduces uncertainties and risks, including the risks of political and social disruption; conditions become more favourable for policy-making at all levels; governments, business and people can look ahead; and environmental planning to address both immediate and longer term issues becomes feasible. 23 The next two paragraphs draw on an excellent commentary on the subject made by Tony Killick at a conference organised by the Overseas Development Institute in London, 27-28 March, on The Environment, Development and Economic Research. The proceedings are to be published by ODI in 1990. 24 In one important respect time was not lost, since there was an accumulation of scientific research in the period. 15 . Market stability. Price signals are clearer, and the use of the polluter pays principle becomes an operational possibility. Tax policies, along with the pricing policies of the infrastructure services, may thus focus more on questions of resource allocation rather than on alleviating budgetary crises, and the merits of environmental taxation and regulation more readily and objectively evaluated. Structural Issues. In addition to achieving satisfactory aggregate internal and external balances, the structure of macroeconomic incentives is similarly important for environmental policy, as it is for investment and growth in general. The example of the rainforests, discussed above, showed how the structure of fiscal and monetary policies (including interest rate and credit policies) may permeate and encourage environmental damage over large areas. Another example, on which less research has been completed, may be the effects of overvalued exchange rates on agriculture; as the 1987 Bank Staff Report to the Development Committee2 noted, by lowering agricultural prices, overvalued exchange rates reduce "the returns on investments in farmland development or conservation ... which reduces the farmers' ability and their incentive to invest in levelling, terracing, drainage, irrigation and other land improvements ..." A further example is the Carajas Ore Project in North-East Brazil. It seems broadly agreed that, within the confines of the project itself, environmental damage was contained. Indeed, there were environmentalhy beneficial expenditures incurred to control soil erosion along the railroad right-of-way, and to improve drainage and effluent control in the industrial area. These expenditures were necessary for the protection and maintenance of the project infrastructure, particularly the railway. They were described in the post-evaluation of the project as being sustainable, and good investments in economic as well as environmental terms.2 It was the opening up of the 890-kilometre corridor outside the project's confines, made possible by the project infrastructure, that ostensibly led to ecologically damaging intrusions into the forest areas. With hindsight, two lessons have emerged. The first, long advocated by Easter, Dixon, Hufschmidt and others in other contexts,27 is that it is often necessary to broaden the definition of project areas when analysing projects that have significant externalities in order to estimate costs and identify measures that may avoid the external damage. The second, rather less noticed, but perhaps more fundamental, is that ecological damage is sometimes less the result of externalities than of distortions in the structure of macroeconomic policies. However well designed and executed a project may be, and whatever provisions are made for the protection of its immediate environment, it is quite possible 25 See Warford (1987. p. 19). 26 "Environmental Aspects and Consequences of the Carajas Iron Ore Project", OECD, World Bank (draft, confidential report). 27 See e.g. Easter, Dixon and Hufschmidt (1986). 16 that ecological damage is unavoidable outside a project's confines, in ecologically sensitive areas, if macroeconomic policies encourage it. (The incentives for land acquisition and the clearance of the rainforests in the region, summarised earlier, suggest that this was probably the case in the Carajas project.) However, the converse also applies: given macroeconomic stability, and a structure of incentives more favourable to environmental investment and maintenance, ecological damage from industrial activity may often be reduced to negligible proportions. 17 PART III: CHANGING PERSPECTIVES ON ENVIRONMENTAL TAXES 3.1 Introduction Although the idea of taxing pollution is an old one in economics (going back to Pigou in the 1920s), it has found little application in practice. Historically, governments in all countries have favoured the instruments of law and regulation to address environmental problems, as with Clean Air Acts, for example, the setting of standards and controls on waste disposal systems, industrial effluents and vehicle emissions, and the recent International Accord on CFCs. Such laws, accords and regulations will always occupy a central place in environmental policy, as will "voluntary compliance" and privately negotiated arrangements for many categories of local pollution. However, the regulatory approach does have drawbacks; the costs that regulations impose on business and consumers may be hidden, but are no less real, and were the subject of much analysis by Kneese, Beckerman and others two decades ago.2 Increasingly, attention is being paid to areas where pollution taxes can be used to provide market-based incentives for reducing environmental problems:29 Lead - Free Petrol. Tax differentiation in favour of lead-free petrol is already in effect in Finland, Germany, the Netherlands, Norway, Sweden, Switzerland and the UK. Carbon Taxes. The use of carbon taxes to bring about reductions in greenhouse gas emissions - by encouraging conservation and development of new energy technologies - is now widely discussed in Europe and the UK. Germany is actively considering liquid fuel taxes, and taxes on packaging and beverages containers in a revenue-neutral context. The Italian Cabinet has approved taxes on fuels and plastics. Acid Deposition. Similarly, taxes on sulphur dioxide and nitrogen oxide emissions are being discussed as means of encouraging the use of cleaner fuels and/or scrubbers and - equally important - to provide revenues for environmental clean-up and restoration programmes. (The Swedish Government is actively considering a sulphur tax.) 28 See Kneese, Ayers and d'Arge (1970) and Kneese, Rolfe and Hamed (1971). More recent analyses can be found in his handbook with Sweeney (1985); in this, see especially the paper by Bohm and Russell, "Comparative Analysis of Alternative Policy Instruments". Also Beckerman (1974). 29 Opschoor and Vos (1989) have provided a recent review. See also Barde (1989) and Ch. 7 of Pearce et al. (1989). 30 E.g. the report by the UK Environment (now Trade and Industry) Minister, Nicholas Ridley, (1989), Pearce et al. (1989) and Grubb (1989). 18 Conestion Charges in Cities. The aims here are to use the price mechanism to reduce mounting traffic congestion and related environmental problems in cities (e.g. the recent proposals of the Dutch Government). Major advances in road-use tolling or metering technologies have further stimulated this debate.31 However, the Singapore Area Licensing Scheme, noted above, has shown that successful congestion pricing policies do not necessarily depend upon advanced technologies. Taxes on Fertilisers and Agro - Chemicals. The idea is to encourage environmentally more benign methods of agricultural production, and reduce surface and groundwater pollution. Such taxes exist or are proposed in Sweden and Finland. The use of Deposit Refund Systems and Differential Taxes on waste products to encourage recycling. These examples illustrate some of the areas in which the industrial countries are considering using taxes to provide market-based incentives for controlling environmental pollution.32 Let us now turn to the developing countries. 3.2 Some Candidates for Environmental Taxes and Expenditures It is useful to begin with a listing of candidates for environmental taxation and expenditures. The following list in Table 6 is not comprehensive, but is typical of the range of options budgetary authorities might eventually have to consider in the course of annual taxation and e.-penditure reviews. The table focuses on pollution taxes, user charges, conservation incentives and public investment. But of course there is a regulatory alternative to be weighed in most cases, including the possibility of marketable pollution permits, which will be discussed later. The distinction between the terms user charges and pollution taxes is not always clear-cut, but in practice can be useful. Here, the term user charge either relates to a direct charge for a service, such as water supply, waste disposal or a concession to log timber from a forest reserve or watershed, or to the use of taxes in lieu of charges, the best example of the latter being roac-user charges (vehicle licences, petrol taxes and so forth). 31 A recent report is Transport in the Nineties: The Shaping of Europe, by Bendixson (1989). 32 See also a World Bank Report (draft) by Hamrin (1989). 33 It is noticeable that standard texts on public finance in developing countries contain no or little discussion of environmental issues and externalities. E.g. Prest (1985) and Newbery and Stern (1987). 19 There is a widely used assumption that while pollution taxes are best left in general revenues, for allocation in accordance with government priorities, user charge revenues should accrue, in the first instance, to the bodies responsible for providing the service in question - roads, water and sewerage, waste disposal and so forth. It is still an open question, discussed further below, as to whether pollution taxes, like user charges, would not better be directed to environmental ends. A conservation incentive, as the term is used here, may typically be an indirect tax to reduce consumption of a good or service responsible for pollution, but which is not otherwise related to the costs of pollution. An example here is the use of indirect taxes to encourage fuel and energy savings; they are unrelated to emission or pollution characteristics of the fuel or energy used. Some problem areas can be addressed by more than one economic instrument. Thus emissions related to energy use (SO, NOx, particulates and greenhouse gases) would be influenced both by pollution taxes and conservation incentives. Others may place significant demands on the public expenditure side of the account, as well as requiring some form of tax or user charge: water supply and sewerage, waste disposal, congestion reduction, watersheds, rainforests, and wildlife reserves are all clear cases. Soil erosion control also requires significant investment. For these reasons it is good practice to analyse the revenue and expenditure sides of the accounts together, not separately, as is often traditional in budgetary analysis. Let us now turn to the issues involved in applying the instruments of environmental policy listed, and in comparing the tax or user-charge approach to environmental policy with the regulatory alternatives. 20 Table 6: Matrix of Problems, Instruments and Issues ISSUES The Regulatory Alternatives. International Trade and Investment. Administration and Compliance Costs. Revenues and Expenditures. The Use of Subsidies. Second-Best Issues. Income Distribution. Economic Efficiency and Innovation. The Macroeconomic Context of Environmental Policy. PROBLEM AREAS INSTRUMENTS Pollution User Conservation Public Investment Taxes Charges Incentives Urban - congestion x x - noise x - emissions x x Water & Sewerage x x x Greenhouse Gases - emissions x - carbon sinks x (afforestation) Waste Disposal x x x Acid Deposition x x Industry - effluents x - gases x Agriculture - chemicals x - fertilisers x - soil erosion x - soil quality' x Forestry - watersheds x x - rainforests x x - social forestry x CFCs x Wildlife Reserves x x 1 E.g. problems arising from dessication, losses of nutrients, salination or other, depending on the area. 21 PART IV: ENVIRONMENTAL TAX ANALYSIS 4.1 Key Issues In any study of taxes or user charges, there are a range of issues to be considered; they were listed in Table 6 above, and are discussed below under the following headings: . economic efficiency and innovation; . the regulatory alternatives; . taxes and revenue requirements; . subsidies and direct public investment for lowering pollution; . income distribution; . international trade and investment; . the macroeconomic context and second-best issues. That such issues arise in all problem areas is another good reason for analysing individual taxes or user charges within the context of an overall revenue and expenditure programme. As with most other issues of policy, the answers vary with the local situation, not least with the severity of the problems encountered and with such matters as budgetary constraints and the practicalities of administration. As discussed earlier, it is also difficult for any environmental policy to succeed when macroeconomic conditions are working against it. This issue is discussed last below, though arguably it is the first priority to be addressed when macroeconomic imbalances and distortions in the incentive structure for growth and development are severe. 4.2 Economic Efficiency and Innovation 4.2.1 Costs An initial aim of environmental taxes is to raise the prices of polluting or environmentally damaging products and practices such that the (external) costs to other parties, both present and future, are reflected in the price. The higher prices act to regulate the level of pollution or the damage to an asset. Less often noted, they also raise public and private financial resources for the further reduction of pollution or for the maintenance of an asset. If before the taxes are introduced prices reflected the marginal costs (MC) of production, and the taxes are chosen to reflect the marginal external costs (MEC) imposed on other parties, then prices rise to P = MC + Environmental Tax (= MEC) (1) and the use of the offending product or practice declines. The decline would of course be even greater in cases (like the tropical forests) if subsidies for environment damage were to be removed as well. 22 The same effect can be achieved by regulations or pollution permits, though at a higher cost (see Section 4.3 below). Ideally, the initial scarcity or rental element in prices when a regulation is applied, or the initial value of a permit, if a permit-to-pollute approach is followed, equals MEC. 4.2.2 Innovation and Substitution However, this is barely the beginning, and greatly understates what environmental policies may accomplish, in the developing no less than in the industrial countries. Once taxes or regulations are introduced, it is rare for products and practices to remain the same as before, and for their effects to be confined merely to quantitative reductions of demand and supply, with other things unchanged. If a tax or regulation were to add, say, 25% to marginal production costs (MC), environmental damage may typically fall by 10 to 15%, perhaps less, depending on demand and supply elasticities - a rather small effect for a large price increase, and one which would soon be eliminated in most instances by added demands arising from per capita income growth. it would rarely be sufficient to allay environmental concerns. A much more important consequence of environmental taxes, or indeed of regulation, is that rents and investment opportunities are created in products, methods of production and methods of asset management that may reduce pollution or environmental damage to low or sometimes negligible levels. There are three effects to consider, not one. Quantitative effects - arising from demand and supply reductions, as just noted. Substitution - towards less polluting or damaging products or practices. Innovation - again in the direction of less polluting or damaging products or methods of production. Historical experience shows that substitution, changes in practices and innovation become widespread once a policy is in place and is complied with. The technical and engineering developments in water supply and sewerage systems in the last century are a good example; this is also an area where there is significant project experience in developing countries to show that, in addition, it is unnecessary to adopt "best-practice" approaches to obtain major improvements. Another example is the substitution of clean fuels in urban areas in response to the Clean Air Acts introduced in the industrial countries in the 1950s and 1960s. More recently, the possibility of taxes or regulation on SO2 and NO. emissions from power stations has led to a range of innovations in emissions 34 Recall Turvey's original formulation of the external cost problem worked with two damage functions, the first being related to existing production methods, the second to less damaging ways of production arising from negotiations or tax incentives (Turvey, 1963). Beckerman (1974) also contains a good discussion of innovation and substitution. 23 control technologies and has also raised the rates of return to gas-fired power stations; an OECD study (1987b) found that the costs of reducing the offending SO, emissions by 90% or more would probably cost less than 10 - 15% of capital costs of new power stations, perhaps less than 5% if gas or low sulphur fuels were used. Solid waste disposal, the treatment of industrial emissions and effluents, and recycling are also highly innovative areas. In more traditional areas such as soil erosion control and forestry management (where, however, there is still appreciable R & D)35 knowledge as to sustainable approaches has evolved over several centuries; once again, the effects of environmental policies are usually to promote innovation, substitutions and changes in practices rather than to reduce output with technologies or practices remaining unchanged. Technical change and substitution will also be the most likely response to carbon taxes if they are eventually introduced on account of the greenhouse effect; a recent OECD compendium shows that a wide range of technical options is emerging.6 Further examples are provided in Annex Tables Al, A2 and A3, drawing on an OECD report on energy and the environment (1990). Technical options for pollution reduction in the energy industry and related activities are classified as follows: . Pollution control based on the use of add-on technologies such as electronic precipitators (an old and widely used invention to reduce particulate matter from power station exhausts) and scrubbers to remove acidifying gases (Table Al). . Fuel substitution. Increased use of natural gas and low sulphur oil and coal would be favoured by emission reduction incentives or regulations. . "Clean" energy technologies (Table A2). . Greater energy efficiency in production and consumption (Table A3). There is a very wide range of possibilities here, involving substitution, innovation and conservation. Under appropriate incentives, therefore, innovations and substitutions which reduce environmental damage appreciably are the rule rather than the exception, and lead to a far more optimistic conclusion than if the taxes or regulations simply restricted demand without changing the nature of the products or the practices by which they are produced. 4.2.3 Environmental Improvements Over Time The above analysis suggests that environmental improvements should be an evolving process under a successful policy. In the diagrammatic summary of the above discussion in Figure 3, the process is divided into three phases: 35 See e.g. the references in footnote 14. 36 OECD (1989). The costs of renewables are now approaching $40-60 per barrel or less, much lower than oil price projections for the 1990s made five years ago. 24 Phase 1: quantitative effects arising from the removal of subsidies and the introduction of environmental taxes. In the short run there would be a reduction of damage due to supply and demand contractions, or at least to supply and demand not growing as fast as they would at lower prices. Tax revenues rise rapidly in this period, and the improved budgetary situation leaves the public authorities better placed to put resources into such areas as soil conservation, afforestation, and the protection of forest and wildlife reserves, where financial self-sufficiency is more difficult to achieve. [The curve shown in Figure 3(b) assumes the tax is phased in gradually, which is why it rises as shown to P2.] Phase II: innovation and substitution. Environmentally more benign or sustainable products and practices are gradually brought on stream. In the private sector, this is the market response to the rental elements created by the policies. In the public sector improved pricing policies for utilities and waste disposal also raise investment opportunities. Environmental tax revenues decline as environmental damage declines. However, prices also decline (Figure 3). If MUC denotes the marginal cost of using the new product or practice relative to the old one, price declines towards P = MC + MUC (2) and what was originally an environmental tax (= MEC) has been transformed into a user charge to provide for the costs of achieving the environmental improvements. The extra value added by these activities is approximately MUC times output. Most are significant employers of labour and of capital, and the returns to capital should be at least equal to its opportunity cost (otherwise the environmental taxes will have been set too low). Interestingly, there have been instances where prices were lower after environmental improvements than before. A good example is the substitution of gas for coal as a domestic fuel after the Clean Air Acts in industrial countries. Another is the use of gas for power generation, which with combined cycle technologies is often the least-cost alternative, quite apart from its superior emission features (lower particulates, CO, and SO,). Yet another example is public water supply costs, which are frequently below the costs of alternatives - such as the costs of the family labour required to fetch and carry water from distant wells or stand-pipes.37 37 Several studies have found that the labour supply costs of the alternatives are often appreciably above the MC of public supplies. See Churchill (1987), and the World Bank's Lagos Water Supply Proiect: IBRD No. 6375-UNI (World Bank, 1988). 25 FIGURE 3 RESPONSES TO ENVIRONMENTAL POLICIES (a) Demand and Supply MC + MEC Prices P MC + MUC & Marginal Costs 10 lp-Initial MEC 10s0 MC MC-Subsidy P0 Demand Quantity (b) Price Response P3 Prices .................... ............... ...... MC +MUC Prices P r - - - - - - - - MC I.Quantitative II.Innovation III.Effects effects & substitution internalised I Time (c) Damage and Taxes 100 4 of initial I damage \a----Tax revenues Damage Time (d) Practices 100 % of Non-damaging total output Damaging Time 26 Phase III: In this phase, environmental concerns become fully incorporated or internalised in the pricing system. Perhaps surprisingly, they also become part of the National Accounts, being measured in the returns to labour and the returns to capital in the activities discussed. In the case of add-on technologies and substitutes that have been preferred on environmental grounds, it may be possible to identify these sources of value added. However, the more the improved products and practices become ingrained in the economic system, the more difficult it may be to identify them separately. Improved technologies or practices no longer become add-ons but an integral part of the process, like the health and safety features of buildings and machines, and P simply reflects the MC of the environmentally unoffending activity. How realistic is such a scenario? The main doubts relate less to its institutional, economic or technological realism, than to the willingness of public authorities to pursue it. In practically all cases cited - in forestry, agriculture, industry, energy, transport and the utilities - the options are well known and, as discussed in Part II, there is much historical experience to go on. The options can also be costed and an operational approach to the required policies defined, whether the tax approach or its regulatory alternative is chosen. While there are significant obstacles to be addressed, the desirable directions of policies, their effects on economic activities, and on the environment, can be clearly adumbrated. 4.3 Regulatory Alternatives There are two forms of regulation to be considered, one is direct, the other is through the use of pollution permits or licences. 4.3.1 Direct Regulation It was noted that much the same end-results as those described above for the tax approach can be achieved by regulation. Taking a long-term perspective, environmental damage - and thus taxes - would eventually decline to low or negligible levels once environmentally superior products or practices were introduced, and prices would decline to P = MC+ MUC. The regulatory approach goes to this result directly, "forcing" the technical and managerial innovations through, so to speak, by restrictions on emission levels or on the ways common resources are to be used. Hence the main differences between the tax approach and direct regulation lie in the short and medium terms, and the question to be addressed is, which is the best way of bringing the transition about? As with most questions of this kind, the answer rarely rests on economic efficiency alone, but on administrative and institutional practicalities as well. It also varies with the problem: regulations backed by legal liability and law are better suited (for instance) to hazardous wastes and environmental accidents, the tax approach to congestion, and a combination of the two, perhaps, to tropical forests. Further, neither approach can work properly without certain elements of the other. Regulation requires significant funds for administration, research and 27 the monitoring of compliance with policy, and thus requires tax revenues from one source or another. The tax approach, on the other hand, has many of those features normally associated with regulation, such as the setting of environmental standards, the monitoring of emissions (or of the practices responsible for them), and regulating the use of common resources; like regulation, it also relies heavily on the instruments of law and legal liability, if only to take defaulters to court. Nevertheless, there are clear distinctions in the approaches; and theory and experience suggest that far more weight should be given to the tax approach - particularly in the circumstances of the developing countries. It has three advantages.8 First, it lowers the costs to polluters of complying with policy.39 Polluters facing high costs of pollution abatement would prefer to pay a charge or tax while those facing low costs would prefer to avoid it by installing the necessary equipment or changing practices. Hence the least-cost forms of abatement are introduced first, and there is a better ranking of priorities. Second, it has institutional advantages in terms of cost and feasibility. Environmental taxes on commercial energy, fertilisers and agro-chemicals for instance could be grafted onto existing indirect taxes on such commodities. Indirect taxes have long been favoured forms of taxation in developing countries because of their administrative convenience. For many problem areas environmental taxes would work through existing administrative arrangements and would be operationally feasible. By being levied at wholesale or bulk supply points, problems of non-compliance with policy as well as administrative costs would be minimised. In addition, they would feed through to informal sector and small-scale activities, in so far as they used polluting products; such activities would easily escape the regulatory net (as they often do with income taxes). Chronic institutional weaknesses with respect to law and public administration in some countries must also raise doubts about any policy which rests heavily on regulation.4' Aside from reducing costs and the occurrences of non-compliance as just noted, taxes can also achieve useful effects even if the levels of pollution and the parties responsible for it are not precisely known: polluters still pay (as they would, for instance, with taxes on agro-chemicals or on polluting fuels) and there is the incentive for pollution reduction by demand changes, substitution and innovation. The regulatory approach, in contrast, is much 38 See also the discussion of Beckerman (1974, Ch. 6), and Pearce et al. (1989, Ch. 7). 39 This is a classical argument for the tax approach, and is due to Baumol and Oates (1971). See also the World Bank study on Finland's water pollution control programmes by Warford, Pellegrini, Kneese and Maler (1974). 40 Point recently made by David Turnham in an OECD research note. 41 Again the issue is one of emphasis and deciding on the best instrument in the circumstances. The legal-cum-regulatory route has been used with success in some countries; see e.g. Findley (1988) on Brazil. 28 more exacting on matters of damage measurement, monitoring compliance with standards, and so forth; its administrative burden is greater, and it would be more prone to problems of bribery and default when administration and laws are weak. Third, as variously emphasised above, financial resources still have to be raised for those environmentally and economically desirable programmes that are less capable of becoming financially self-sufficient - such as afforestation, soil conservation and the protection and maintenance of forest and game reserves. In addition, there are the requirements of administration and environmentally-related research to be met. While, for the reasons discussed, environmental tax revenues in a successful policy would eventually decline to low levels, the transition period would often not be short. The declines could in any case be offset by appropriate increases in indirect taxes on the activities in question, being justified as a conservation incentive, as well as for revenue-raising. These advantages notwithstanding, there are situations when the regulatory approach is preferable, toxins and hazardous wastes, as noted, being a clear case. The choice of the best policy must necessarily rest on local analysis of (a) costs, (b) what is institutionally feasible, and (c) revenue requirements. However, the advantages of the tax approach in all three respects have too often been ignored in environmental policy; they would generally outweigh those of the regulatory approach, though not in all cases. 4.3.2 Pollution Permits42 Marketable pollution permits have recently led to much excitement among economists because they promise to address the first of the above-mentioned disadvantages of regulation when it is compared with environmental taxes. This is that once a target level of pollution has been settled upon, and the total number of permits issued to buyers is chosen to be equal to or less than the target level, then buyers create a secondary market among themselves, with the permits going to those bidders who experience the highest costs in reducing pollution and who thus value them the most. As with the tax approach, a cost-effective means of pollution abatement would be achieved. The approach has been tried in the US, though with mixed results. 4 One application that was successful was trading among refiners for the right to put lead in gasoline at specified levels during a specified period; much of the success was due to the well-defined policies and targets of the unleaded fuels programme in the US, the technological sophistication of the refinery industry (not least its ability to exchange information), and the fact that the use of lead could be easily 42 See Pearce et al. (1989) for an introduction, and Pearce and Turner (1989). 43 Another reason for the current popularity of the idea (particularly in the US) is that it is seemingly an approach which avoids "new" taxes. 44 Hahn (1989). 29 monitored using the existing regulatory apparatus. In other instances, involving the trading of emissions permits, the programmes were unsuccessful because of unclear policies and targets, the difficulties of monitoring and arranging for information exchanges, and the fact that the institutional arrangements were ill defined and not in place. These findings are highly ielevant for developing countries because the problems of monitoring, of arranging for information exchanges and establishing new institutions would be far greater. Is the approach therefore suited to the circumstances of the developing countries? Once again, the answer will turn on institutional realities as well as economic efficiency, and it will also vary with the problem in hand. Thlis can be seen by reference to two instances where the approach, in all but name, has in fact been practised. In one case it has been a brilliant success, in the other, a calamity. The successful case is the Singapore Area Licensing Scheme for reducing the external costs - the "pollution of time" as Turvey once called it45 - of congestion. This scheme was noted in Part Il. It was introduced in 1976 and is well known and reported; its main features were summarised in Table 3. The issue of permits to enter the central business district during peak hours was administratively simple and compliance easy to monitor. There was a useful increase in revenues arising from the issue of licences, traffic flows were reduced by 50% relative to their free access levels, and congestion by more than this. The other case is the issue of licences or concessions for the "harvesting" of the tropical forests. It is not usual to think of this old system, much used in the forestry industry, as a permit to pollute, but the principles behind it are in all respects identical. Ideally, the number of licences issued should be such that the resource should be used in a sustainable manner (compliance with standards); there should be no bribery or corruption in their issuance and use; policies should be internally consistent such that financial incentives for the exploitation of the resource are not at variance with restrictions on its use; the licences should be tradable or returnable for reissue, as appropriate; and their value should rise over time as their supply becomes more restricted in relation to demand, thus providing further incentives for replanting and a sustainable use of the resource. The system has been a calamity not because it is wrong in principle, but because of weaknesses in administration and laws in the face of enormous political and commercial pressures. Pollution policies thus need to be related to institutional as well as economic and political realities, whether taxes, direct regulations or permits are used. Pollution permits have the merit of providing the authorities with another policy option, may reduce costs of pollution abatement to the polluters (though not of administration) and would also raise revenues. But they are not a panacea, and their relevance and applicability (as even the US has found) vary greatly. 45 Turvey (1972). 30 The tax approach will very often hold the advantage - it is theoretically equal to or better than the alternatives in terms of efficiency, administratively more convenient and feasible, less demanding on information and monitoring requirements, raises revenues, and is less prone to bribery and default. 4.4 Environmental Taxes and Revenue Requirements What should the ideal environmental tax be and how can it be estimated? In theory, it should be equal to the marginal external cost (MEC) of environmental damage. But in most situations MEC cannot be estimated reliably enough for tax purposes, and sometimes not at all - for acid deposition, for example, for the losses of biodiversity and other consequences of tropical deforestation, for the economic consequences of regional or global changes in climate, for the costs to health, land values and economic activity of poor sewage and solid waste disposal systems, and so forth. Cost - benefit studies can be done in many instances to form a judgement on the desirability of reducing environmental damage, and indeed they are very valuable for this purpose; but they are not precise enough for estimating desirable tax levels. 4.4.1 Tax and Revenue Goals However, such uncertainties (which also afflict the regulatory approach) do not preclude defining desirable objectives and working out the role of taxes in meeting them. For reasons discussed, not least their financial advantages, taxes will often be the more efficient and effective way of achieving the objectives. Experience with environmental policy in the industrial countries has shown that the objectives generally emerge as the outcome, not of precise estimates of a monetised damage function, but of (a) scientific, public and social inquiries into environmental damage, and its effects on public health and well-being, (b) the physical damage to public and private assets, and (c) the costs of meeting various objectives and targets. Once the objectives have been agreed upon, the marginal costs to public and private parties in meeting them can be estimated. Above, these costs were called the marginal costs (MUCs) of using improved products and practices; they can be estimated reliably in most cases, e.g. for the reduction of industrial emissions and effluents; recycling technologies; the use of environmentally more benign substitutes; 46 Pearce and Markandya (1989). 31 . water supply, sewerage and waste disposal projects - where LRMCs have been estimated for a variety of technologies and practices, often in the search for affordable solutions . urban congestion pricing policies - where a recursive and evolving approach can also be contemplated, with taxes or the prices of permits being adjusted periodically once the responses to previous taxes or prices have been assessed; . the protection and maintenance of forest and wildlife reserves; and . soil erosion control programmes on farmlands and watersheds. In each case, the MUCs of achieving different standards can also be estimated - and indeed this may be important for policy purposes, since major results can often be achieved at lower costs without necessarily turning to best-practice approaches. 4.4.2 Estimating Suitable Tax Levels An operational approach therefore would be to set the taxes at levels somewhat higher than the long-run MUCs of achieving the environmental goals. Provided non- or low-polluting products or practices are exempted, this will provide the required incentive for such products and practices to be introduced. How much higher taxes should be than MUC is bound to involve some judgement, and may be influenced (among other things) by short-run revenue requirements or the urgency of a situation. If there is a backlog of restoration and clean-up or of defensive expenditures to be met, or if there are significant fixed costs of establishing a programme, there will often be a good case for setting taxes suitably in excess of MUC to meet revenue requirements - quite apart from the point that the ideal tax, in terms of economic efficiency, would be much greater than MUC in any case. To translate this approach into practice, four further questions arise: (i) what are the public and private components of MUC and should they be treated differently? (ii) how do we deal with situations when the uncertainties as to a desirable policy and its costs are unusually large? (iii) how are revenue requirements to be met when the regulatory route is preferred? and (iv) how are the revenue requirements of those programmes less capable of financial self-sufficiency to be met? The answers will usually vary with local circumstances, though some practical points can be made on each. Some technical notes to complement the following discussion of them are provided in the Appendix. 47 See Saunders, Warford and Mann (1977). LRMC is routinely estimated in the World Bank's water supply and sewerage projects (Julius and Warford, 1979). For an earlier study, see Warford (1966). 32 4.4.3 Private and Public Components of Environmental Costs The public components are the costs of administering environmental policy in the area in question - energy, industry, transport, etc. They include . monitoring environmental conditions and change; . initiating and financing high-risk research into alternative technologies and practices, whether undertaken by public or contracted out to private bodies; . supporting educational and training programmes; . monitoring compliance with and the enforcement of policy. Such costs tend to rise with the severity of an environmental problem, and may properly be counted as part of MUC and financed out of environmental taxes. But they may also have significant fixed elements which persist after the environmental problems have been addressed; e.g. the costs of monitoring compliance and of enforcement are likely to continue long after a problem has been satisfactorily addressed, if only to prevent its recurrence. In the short run, tax rates (which need to be related to the sum of public and private MUC) are likely to be in excess of the requirements of the public revenue alone (in the sector in question). But in the longer run, they may not be, and there will often be a good case for meeting requirements out of an indirect tax on the sector once pollution or damage has been reduced. The private costs (including those of public enterprises, like water and electric utilities) similarly have several components: the R & D costs of developing new products and processes, and the subsequent costs of investment, operations and maintenance. Alternatively, they may be equal to the incremental costs of turning to environmentally more benign substitutes. If it is desired to stimulate substitution and innovation in these directions, then a minimum criterion is that the present value of the exemptions to environmental taxes - to which of course, as less damaging alternatives, they are entitled - should at least equal the present value of the costs to the producers of introducing them. In other words, the incentives should offer good rates of return to investments in the improved products and practices, or to investment in the substitutes. 4.4.4 Uncertainties When uncertainties are acute (as they are for instance with respect to the greenhouse effect), it is more appropriate to relate taxes to the shorter-term problems to be addressed (rather than to the long-term MUC, say, of non-greenhouse gas alternatives). Paradoxically, this is a more forward-looking and progressive approach than it might at first appear. Continuing with the greenhouse effect as an example, the relevant shorter-run costs include8 48 See e.g. the World Bank note on the greenhouse effect (1989). 33 (a) R & D expenditures into the causes of the problem; (b) the costs of conservation measures to be encouraged by the taxes; (c) addressing related issues, such as energy losses and emissions arising from congestion (again with the help of congestion taxes or permits); and (d) R & D on low emissions options and renewables, again stimulated by carbon taxes. Such policies would lead to short-run as well as to long-run gains; and they would leave an economy both better informed and better placed to address long-run issues. The revenues from taxes introduced in connection with (b), (c) and, possibly, (d) may often be appreciable; (b), (c) and (d) would also mean that precautionary policies were in place once findings of (a) became clearer. With the greenhouse effect, the issues, of course, go beyond national boundaries, and raise new questions about international aid and co-operation.49 It is beyond the scope of the present paper to discuss these issues. The point of the above example is to show that the resolution of long-term, highly uncertain problems, involving risky investments, requires significant expenditures - for R & D, and for precautionary purposes - in the short run. It is analysis of these types of expenditures to which estimates of MUC and tax policies, at least in the initial phases of a policy, might best be related. 4.4.5 Financial Requirements of the Regulatory Route While the regulatory route creates rents in improved products and practices, the public costs of administration noted above still have to be met. As with the tax approach, there is a good case for financing these out of an indirect or environmental tax on the activity or product in question. As far as the public components of environmental costs are concerned, the remarks in 4.4.3 apply equally to the regulatory route. As noted earlier, the two approaches may sometimes be complementary. 4.4.6 Where Financial Self - Sufficiency is More Difficult to Achieve In cases of soil erosion control, and the protection of forest and wildlife reserves, the main instrument at the authorities' disposal will be direct expenditures. It would generally be impractical, for instance, to tax farmers for having poor soil conservation practices, or not having sufficient tree stocks on their land for ecological purposes, or to use taxes as an incentive for planting wind-breaks. Some degree of cost-recovery is generally possible in other areas - e.g. from tourism for wildlife programmes, from stumpage fees and licences for forest cutting, or from the sales of timber from wind-breaks and afforestation 49 See Barbier and Pearce (1990) for a recent discussion. 34 programmes. But experience has shown that such revenues are often limited, and are not sufficient to cover the costs of sustainable practices in the circumstances of most developing countries. In practice the funding of such expenditures will have to be supported by general taxation. Another advantage of the tax approach to environmental policy, however, in areas where it can be applied, is that the surplus of revenues over the government requirements for applying policies in these areas either (a) can be used for supporting economically desirable, non-financially-self-sufficient activities elsewhere, or (b) reducing the "call" on general taxation. It thus greatly adds to flexibility while improving efficiency. 4.5 Subsidies and Direct Public Investment Subsidies and public expenditures may perhaps be more often used than taxation to complement the regulatory approach to environmental policy, notwithstanding some well-known and long-noted drawbacks summarised below. Typical devices are low-interest loans, grants and tax exemptions for pollution control equipment, or for relocation of polluting plants away from high-pollution areas.50 4.5.1 Areas where the Approach is Beneficial Before turning to their drawbacks, however, three economically beneficial uses of subsidies and public investments in support of private activities might be noted, of which the first two are uncontroversial. The first is the use of subsidies to support high-risk R & D activities in pollution control technology development. Examples might include subsidies for R & D into renewable energy technologies, research and demonstration projects in forestry, agro-forestry and erosion control, and pollution control and demonstration experiments in other areas where the technology is unproven and initial costs are high. As was concluded in a World Bank study on water pollution in Finland in 1974, "the benefits of subsidies of this kind are solidly supported by economic theory" while many other forms of subsidy are not.51 Second, there is the use of subsidies for socially desirable activities which by their nature are not capable of becoming self-financing. Environmentally desirable public goods come into this category - the protection and maintenance of wildlife and nature reserves, parks, forest reserves, historic buildings, and so forth. Education and extension services also come into it, e.g. in support of soil conservation and afforestation, as do some environmental works programmes such 50 For a recent analysis of their application in Brazil, see Findley (1988), who also provides an extensive discussion of the development and workings of pollution control policies in the country. See also Portney (1989). 51 Warford, Pellegrini, Kneese and Maler (1974). 35 as watershed management and soil erosion control. Again, the use of public revenues for such purposes, in cases where the social benefits are thought to exceed the costs, is well supported by economic principles and experience. Third, turning to a more controversial issue, there is the political-economy argument for "recycling" or earmarking revenues to environmentally desirable ends if taxes are used as the primary instrument of environmental policy. The literature has been long divided on the merits of earmarking, one school arguing that it imposes unnecessary and socially inefficient rigidities on government expenditures, the other stressing various advantages of political accountability.2 In the case of environmental policies the case for some form of earmarking may also be supported by the view that environmental taxes are taxes in lieu of charges for environmental services and can thus be legitimately viewed as user charges (just as vehicle and gasoline taxes are viewed as road-user charges). However, turning to the practical issues of policy, two points can be made. One is that, for reasons just noted, environmental policies will make significant claims on public (as on private) revenues. So environmental tax revenues may very well be needed for environmental purposes in any event. The other is that by using the public revenues for environmentally desirable programmes that are not capable of becoming self-financing, or that are inherently risky, the political-economy argument of people "seeing their environmental taxes at work", so to speak, can be turned into an undisputed economic gain. 4.5.2 Areas where the Approach has Drawbacks It is in the use of subsidies for investments where tax or regulatory policies (if they were pursued) would make environmentally improved technologies or practices financially self-sufficient that the use of subsidies is most questioned. Taxes, as noted, create rents in the improved technologies and practices, and bid up prices such that the costs of the improvements are met; regulations, provided there is reasonable compliance with policy, have a similar effect. In contrast, subsidies have long been known to have several drawbacks, of which only three need be highlighted:53 Technology Bias. In theory, the subsidy needs to be related to the level and benefits of pollution avoided; but the latter is even more difficult to measure than actual pollution produced, so in practice administrators have taken to subsidising pollution abatement equipment. The bias is generally in favour of capital-intensive technologies and often against substitutes or changes in practices that may accomplish the same end more cheaply. For instance, SO. scrubbers may be favoured by a subsidy for equipment to reduce acid deposition, whereas low sulphur 52 See e.g. Buchanan's classic article on earmarking (1963). 53 See e.g. the World Bank's study on water pollution in Finland by Warford et al. (1974) from which the following points are taken. The same conclusions would appear to apply to Brazil's experience, reported by Findley (1988), where the subsidy policies adopted are uncannily similar. 36 alternatives (gas, low sulphur coals and oil) would not. Further, as low-polluting options become integrated into the new production process, it becomes difficult to decipher what part of the costs is due to pollution control and what is not; investment decisions become biased "in favour of those expenditures that can be most easily claimed".M Administration. Related to this is the "scramble" for subsidies if firms have to comply with new standards. When funds are insufficient to support the policies (as they often are), standards are disputed, and lengthy delays may arise in the implementation of policy. Revenues - lastly, the subsidies are an avoidable claim on the public revenue. The principle to follow therefore is that when an environmental policy can be made to be financially self-sufficient by a tax, this is much to be preferred to one relying on subvention. 4.6 Income Distribution Issues Who would benefit from environmental policies? Little systematic research has been done on this subject.55 But as with most questions of this kind it can be anticipated that the answers will vary with the problem and the situation. Some general points can be made under three headings: the influence of environmental policies on labour demand and productivity; their influence on health and living conditions; and the design of compensatory policies. 4.6.1 Labour Demand and Productivity Environmental policies will often be labour-intensive, and should prove to be a means of raising real incomes and wages in low-income regions. The following are some examples: . the establishment of agro- or social-forestry practices on farmlands and watersheds; . the planting and maintenance of wind-breaks or shelter-belts, and the extraction of timber from them, in areas threatened by erosion and desertification; . soil erosion control programmes in general; . the protection and maintenance of wildlife and forest reserves; less chemically-intensive forms of tillage;56 54 Warford et al. (1974, p. 8). 55 It would be an important area for research. The work of Selowsky (1980) and Meerman (1979) on who benefits from public expenditures may provide a good model. 56 See Repetto (1989). 37 the construction and maintenance of water supply, sewerage, storm drainage and solid waste disposal systems. Besides being labour-demanding, such activities also fulfill another basic requirement of good expenditure policies, which is that they have prospectively good social rates of return to investment. In the cases of agro-forestry and wind-breaks, for instance, there is first of all the yield of "wood". But their favourable effects on soil moisture and nutrient retention as well as, self-evidently, the reduction of soil erosion, have all been shown to be beneficial for agriculture (see Part II). They would benefit smallholder agriculture in particular. It is not possible to assess without local analysis the implications of some other policies for distribution, at least from the employment side. Industrial emissions and effluent treatment facilities, for instance, or the manufacture of lean-burn engines and catalytic converters for vehicle exhausts, are all fairly capital-intensive activities. While their manufacture, installation, maintenance and repair may generate useful backward linkages with the local economy, it is unclear what the net effect on labour demands and real wages would be. Consider next employment in coal-mining. At first glance it might seem that such employment would be reduced by restrictions or taxes on acidifying gases and particulate matter emissions from power stations. However, it is important not to overlook the technological response: coal-cleaning, scrubber and particulate matter technologies for reducing such emissions are well proven and, even in situations where emission standards are very high, amount to less than 15% of power station capital costs.57 Other technologies available include fluidised bed combustion and changes of boiler designs to reduce combustion temperatures and NOx. It is thus possible that the coal industry would retain its electric power markets; employment and incomes overall might thus be increased once value added in the manufacture of emission control technologies is taken into account. As against this, environmental controls would undoubtedly provide another reason for favouring gas-fired stations in many countries. Still on coal, the greenhouse effect, if it turns out to be as serious an issue as many observers expect it to be, would of course pose more fundamental problems. However, a range of technological options are becoming available which may themselves become important sources of employment and earnings - solar thermal, photovoltaics, dentrothermal, afforestation, biomass and biomass-synthetic fuels in particular.ss 57 OECD (1988). 58 OECD (1989). 38 Urban congestion pricing policies are another area where the distributional impact is difficult to assess a priori. However, there seems no reason in principle why such policies could not be made to be progressive. In the first place, the revenues raised from congestion pricing would be appreciable, and could be used to finance urban infrastructure improvements - roads, drainage, traffic management and safety, pedestrian areas, and so forth. Second, congestion charges would be far higher per passenger for private cars than for buses, jeepneys and the like, and infinitely higher than for pedestrians and bicycles. Agrarian reform and reforms of agricultural prices are further policies likely to have efficiency and distributional benefits, which may at the same time impact positively on the environment - by raising the incentives for farmers to adopt more sustainable agricultural practices. Reliable evidence on this matter is still forthcoming. What can be said is that the long-discussed reforms in these areas should ideally leave farmers better placed to invest in soil erosion control programmes and social forestry. 4.6.2 Health and Living Conditions In the industrial countries, the worst environmental conditions historically were encountered by the most impoverished members of the population. The evidence - on deaths and diseases arising from industrial emissions and effluents, from smogs, from unsanitary water supplies, from the lack of sewerage and waste disposal facilities, from the lack of drainage systems and so forth - has been 39 well documented.59 No social group escaped such problems, though the poorest were the worst affected. The question is whether developing countries need to repeat, as many are 59 It is very well reviewed by Beckerman (1974), from which the following passages are taken (pp. 65/66): "In the middle of the last century it was reported [in Drainage and Sewerage (London, 1847, p. 607)]: The space bounded by Oxford Street, Portland Place, New Road and Tottenham Court Road, is one vast cesspool, the sewers being so imperfectly constructed that their contents are almost always stagnant ... Now when the reader reflects that thousands of working men are closely confined, for perhaps 14 to 15 hours out of the 24, in a room in which the offensive effluvium of some cesspool is mingling with the atmosphere ... he will cease to wonder at the amount of disease ... "More generally, the conditions in London in the mid-nineteenth century have been described in all their horrifying detail in the classic works of Dr Hector Gavin, notably in his Sanitary Ramblings. Almost any page of this work contains descriptions such as the following: Pleasant Row ... Immediately facing Pleasant Row is a ditch, filled with slimy mud and putrefying filth, which extends 100 feet. The space between Pleasant Row and the central square is beyond description, filthy; dung-heaps and putrefying garbage, refuse, and manure, fill up the horrid place, which is covered with slimy foetid mud. The east end has likewise its horrid filthy foetid gutter reeking with pestilential effluvia; the southern alley is likewise abominably filthy ... I entered one of these houses on the southern side, and found that every individual in a family of seven has been attacked with fever ... the privy of this house is close to it, and is full and overflowing, covering the yard with its putrescent filth ... "It is not surprising that the deaths from typhus alone in England in the mid-nineteenth century were nearly 20,000 a year, and that 60,000 deaths a year were attributed to tuberculosis, not to mention high death-rates from numerous other diseases associated with insanitary and unhealthy living conditions." 40 now doing, the same experiences in the course of their economic expansion.6 The industrial countries would have been economically as well as socially better off had their environmental policies been introduced earlier, instead of more than a century after urban smogs were known to be a serious hazard to life and health (a century and a half in the case of the UK),61 and it took the deaths of many thousands in the "great stinking fogs" to induce the authorities to introduce the Acts;62 safe water supply, sewerage, drainage and waste disposal systems were similarly introduced long after suitable practices and the resources to introduce them were available. It was, ultimately, the growth of social concerns, repeated environmental crises and epidemics and an ever-widening political accountability of governments for the prevailing conditions that eventually led to environmental 60 c.f. Findley's summary (1988, pp. 59 - 60) of the crisis that led to the introduction of a successful programme in Cubatao, Brazil: "Additional air quality monitoring by scientists at the University of Sao Paulo had disclosed, in the dust and smoke perpetually shrouding the city, "an infinity of ... chemicals, invariably toxic," as well as sulfates, phosphates, and nitric and phosphoric acids, in concentrations "at least a dozen times higher than the maximum acceptable levels". Although CETESB's preferred course of action was "evacuation [of residents] from the city, pure and simple", there seemed to be little chance of that happening. Meanwhile, acid raid continued to denude the slopes of the Serra do Mar, and the threat of massive landslides was described as a "geological time bomb". "On September 3, 1984, in the midst of an atmospheric inversion, officials of CETESB and SEMA visited Cubatao. A few hours later the governor of Sao Paulo, for the first time ever, decreed a "state of emergency" in Cubatao, citing particulate concentrations in excess of the ambient standard of 875 micrograms per cubic meter. CETESB promptly required nine industries in Vila Parisi to shut down and ordered residents to evacuate the district. Police were sent from the city of Sao Paulo to assist in the evacuation and to prevent looting of empty homes. Although the mayor of Cubatao protested that he had not been consulted before these measures were taken, he agreed to make the soccer stadium available for displaced residents and to provide food and blankets. Eventually, atmospheric conditions improved, the state of emergency was downgraded to a state of alert (defined as concentrations of particulates in excess of 625 micrograms per cubic meter), the eighth such alert of 1984, and people were allowed to return to their homes. A few months later, in the early morning of Saturday January 26, 1985, there was a massive, prolonged release of ammonia gas from a ruptured pipe at a fertilizer plant in Vila Parisi. Six thousand residents were evacuated; more than sixty persons were hospitalized. On Sunday, after the gas had dissipated, people were allowed to return to their homes. On Monday, CETESB assessed a penalty of twenty-four million cruzeiros against the owner of the plant. Later that day, after the Governor of Sao Paulo said the penalty was too small, the mayor of Cubatao signed a decree ... "Since 1985, in response to national media attention, public pressure, and the realisation that public health and safety no longer should be sacrificed completely, state and municipal officials finally have moved to deal with Cubatao's problems."/ 61 Brimblecombe (1987), Ashby and Anderson (1981) and Parker (1989). 62 Ibid. 41 policies being put in place. What is now known is that economic returns to such policies were sufficient to have warranted their introduction earlier - quite apart from their effects on poverty alleviation. 4.6.3 Compensatory Policies Where environmental policies have adverse effects on incomes or employment, there is still the option of compensatory policies on the expenditure side.63 This generally requires analysis of tax relief in various ways, expenditures, or tax restructuring to promote redeployment into new activities, or a combination of the three. The principles (and pitfalls) of such policies are well known. Once again, the revenues generated by the tax approach to environmental policy should give the authorities more flexibility in deciding on an appropriate response to the problem. 4.7 International Trade and Investment A question increasingly raised in the industrial countries concerns the impact of differing environmental policies on the competitiveness of trade and foreign direct investment (FDI). Are the countries with better environmental standards than others likely to find their competitive positions eroded in major industries or in agriculture? This is now an important policy issue in the EEC, which is seeking to harmonise policies on a range of issues (a recent example being the discharge of sewage into the North Sea). However, the potential impact of environmental policies on trade and competitiveness is perhaps of greater concern in the developing countries - and is a common rationalisation, for instance, of the subsidies for logging and ranching in the rainforest areas. The questions raised by environmental policy for trade and investment are relatively unresearched. The following are some general points that can be made at the present time; they may also be suggestive of future areas for analysis and research: (a) A policy of neglecting environmental damage in the interests of maintaining competitiveness does not have a good economic basis. Most environmental problems, by their nature, arise on account of externalities; thus while the polluters may escape the costs of pollution damage, the costs are borne by some other party in the economy. Similarly, when environmental damage is being subsidised, there is not only an increase in external costs to the polluted parties (who are generally uncompensated) but a loss of opportunities elsewhere through the loss of public revenues. The aims of environmental policy are to reduce social costs, not to increase them. (b) Where the main trade-offs seem to lie is in the setting of environmental standards. In principle, standards should ideally differ between countries because the costs of environmental policies and their benefits differ with (among other things) labour costs, land values and even the scarcity of foreign exchange. Thus 63 See the recent study by Smith and Pearson (1990). 42 it is quite legitimate to raise the issue of whether environmental standards may be too high or too low for the country in question, and to adjust standaids on the basis of objective assessments of costs and benefits - including their effects on the costs and benefits of trade and investment. A further point is that, as noted earlier, major environmental improvements can often be achieved without resorting to best-practice techniques, and there is indeed a wide range of technologies and practices available to address most problems. Thus a marginal lowering of standards will often make low-cost options available. (c) The revenues from environmental taxes could in principle be used to offset any competitive disadvantages that may arise from environmental policy. Offsetting tax exemptions on non-environmentally damaging inputs would be an obvious recourse that would not undermine the goals of environmental policy, and may even help them. (d) The regulatory approach to environmental policy would affect costs more, for the reasons give above (Section 4.3). It allows producers less flexibility and is administratively more burdensome; however, there are exceptions, of which the dumping of toxic wastes is again an example. (e) Creating rents on environmentally more benign products and methods of production would attract FDI into these activities. Thus the principles of taxation and, where relevant, the regulatory approaches, discussed in Sections 4.2 and 4.3 above, apply equally to foreign and domestic investment. It thus seems that issues of trade and competitiveness may be addressed without compromising environmental policies. The grey area is in the setting of standards. Even where conflicts arise, the tax approach to environmental policy provides some room for manoeuvre by leaving governments better placed to provide financial compensation. 4.8 Macroeconomic and Second - Best Issues It was noted in Part II that environmental policies would be difficult to implement in conditions of macroeconomic instability and when the structure of macroeconomic incentives is working against them. These are perhaps the main lessons to emerge from the much-discussed problems of deforestation and of the Carajas iron ore project in Brazil. Indeed, in some situations, economically desirable investments, containing all the appropriate provisions and safeguards for the environment, may have perverse effects unless macroeconomic conditions support them. Does this mean that little can be done until macroeconomic policies are put in place? The question can be answered again by reference to Brazil, which shows that there are indeed environmental policies (often of great urgency) which may be pursued despite the macroeconomic situation. The following draws on Findley's report noted earlier:6 64 P. 67. 43 "The case of Cubatao demonstrates, perhaps more dramatically than any other in the world, the most extreme results of unregulated occupation of land by mixed uses in a sensitive ecological setting: a large nation's most important and most polluting factories concentrated in a small area subject to atmospheric inversions and interlaced with thousands of homes; tens of thousands of people bathed in toxic effluent and sleeping directly over pipelines carrying flammable and explosive substances; acid rain defoliating adjacent mountainsides and causing landslides that threaten not only the factories and residences but the highways, railroad, pipelines, and harbor connecting the community to the outside world; and public officials at all levels seemingly in a state of paralysis, obsessed with the goal of economic development and striving to ignore the environmental catastrophe they have allowed.... "Since 1985 there has been real progress toward ameliorating some of the worst aspects of the environmental fiasco in Cubatao. A tremendous amount of pollution control equipment is being installed in industrial plants, which also have been switched to less-polluting fuel, thousands of residents are being assisted in relocating to more suitable living areas, CETESB has become more aggressive in using fines and temporary plant closures to deal with recalcitrant polluters, and the Ministerio Publico has initiated public civil actions seeking restoration of damaged wetlands, waterways, and hillsides. Moreover, through extensive national newspaper and television coverage of the agony and progress of Cubatao, all of Brazil has received an environmental education." Further, some environmental policies may help macroeconomic reform - and in so doing help themselves. The tax approach to environmental policy and the removal of subsidies for environmentally damaging activities in countries with significant fiscal deficits are good examples. Urban congestion pricing policies and the introduction of cost-reflecting tariffs for water, sewerage and waste disposal systems are others. Soil erosion control programmes may or may not be operational depending on the effects of agricultural and marketing board policies on investment in agricultural areas. Such problems can of course only be assessed locally. Some environmental policies may clearly be operational (and merit high priority) in difficult macroeconomic situations, others not; even the former, however, would be enhanced by stability and general improvements in the structure of macroeconomic incentives in many countries. 44 PART V: AREAS FOR RESEARCH What research might usefully be initiated in the area of environmental taxation and its regulatory alternatives in developing countries? The preceding analysis has sought to review a range of policies that might fruitfully be pursued on the basis of current knowledge and experience. However, particularly in new areas of policy-making, any programme of policy reform needs to be supported by economic research. At various pointa in the above analysis, a number of suggestions were made on topics requiring research; they can be grouped under the following headings: (i) environmental standards; (ii) costs, technologies and management practices; (iii) macroeconomic and second-best issues; (iv) the effects of environmental policy on income distribution and employment; (v) analytical issues. Consider each briefly in turn. (i) Standards. Any environmental policy, whether involving taxes or otherwise, is working towards some environmental standard. Economists have long argued for basing standards on comparisons of the costs and benefits of achieving them. There are of course situations in which value-judgements and intangibles are sometimes overriding, and others in which the benefits may be too difficult to quantify. Even here, however, the analysis of costs and what is known of benefits and intangibles has proved to be a good way of organising analysis for decision-making purposes. The cost - benefit (C - B) approach to deciding on standards can also be extended to the impact of environmental policies on trade and investment, an area of much concern in the developing as in the industrial countries. It is often said that developing countries cannot "afford" high environmental standards because of their impact on incomes and trade - though the point that good standards can often be achieved at a small cost, and sometimes with a net benefit (see (ii) below), is frequently overlooked. Nevertheless, there is a good reason why standards should, even in principle, differ between countries when their circumstances differ significantly; this is of course that costs and benefits of achieving environmental improvements also differ between countries. C-B analyses of proposed standards, allowing for their effects on trade and investment, would provide clarifications and a better documentation of issues in an important (and controversial) area. 65 See Pearce and Markandya (1989). 45 (ii) Costs, Technologies and Practices. A wide range of technologies and practices are available for effecting environmental improvements (Part IV). Further, it was noted that major improvements are possible without resort to "best-practice" techniques in practically all areas - erosion control, afforestation, the maintenance and protection of forest and wildlife reserves, waste disposal and recycling, congestion reduction, emissions and effluent reductions, and so forth. The marginal costs are often quite low or moderate for bringing about major improvements, but tend to rise steeply or standards are heightened. This feature of the marginal costs of environmental improvement or pollution abatement has obvious advantages for developing country policies and merits research in relation to local circumstances. Ideally, we need information for a range of emission and effluent reduction technologies which shows the percentage reductions possible in relation to the costs and technologies used.66 Similar information is needed in such other areas as soil erosion control and forestry management to show the percentage reductions of damage in relation to types of investment and management practice. Lastly, information of this kind is required for the estimation of environmental taxes where the tax approach is followed (Part IV) or of revenue requirements where the regulatory approach is followed. (iii) Macroeconomic and Second - Best Issues. Case studies of the linkages between macroeconomic conditions and the possibilities for environmental policy deserve some priority. There are three hypotheses here, raised in Parts II and IV. The first is that macroeconomic stability is not only beneficial for environmental policy but that there are instances when, without it, such policies may have little and sometimes perverse effects. The same hypothesis applies, second, to the structure of macroeconomic incentives: e.g. to the effects of exchange rate and agricultural pricing policies on the willingness of farmers to invest in rural afforestation and erosion control practices. However, third, there are instances when environmental policies may be pursued notwithstanding the macroeconomic situation, and (as with environmental taxes) may even help it; industrial pollution control is a good example. (iv) Income Distribution and Employment. On a first inspection (Part IV) there are some instances when environmental policies would have favourable effects on income distribution and employment (afforestation and soil erosion control programmes were cited along with the health benefits of treating urban effluents and wastes) and others when the effects may be unfavourable (congestion pricing, coal-mining). Much depends, however, on how environmental taxes are structured and used on the one hand, and the design of compensatory policies on the other. These are self-evidently important areas for research. It was proposed that they could be analysed under three headings: the effects of 66 For an excellent example of research of this kind, see Kneese and Bower (1979). The OECD has also provided some evidence for the industrial countries (OECD, 1988). 46 environmental policies on the deployment of labour; the incidence of benefits; and the design of compensatory policies when the effects on distribution and unemployment are unfavourable. (v) Analytics (see also the Appendix). There are several problems that cannot be examined reliably by simple (partial equilibrium) calculations, and when it is necessary to look at the impact of environmental taxes more broadly - to assess their impact on revenues, economic efficiency and income distribution, for instance, and to design compensatory policies when the effects of such taxes on distribution or on trade may be adverse. The same applies to many other aspects of tax and expenditure analysis. The development of modern models of taxation and expenditure systems has enabled the direct and indirect impacts of tax policies on efficiency and distribution to be assessed more rigorously, and has greatly facilitated the design of compensatory policies. Another use is to examine sensitivities when uncertainties are appreciable, that is, to test how robust a proposal may be. It has recently been proposed to adapt such models to research 67 on environmental policy in the industrial countries; their application to developing countries would help address the problems discussed. Other proposals on analysis are discussed in the Appendix. To sum up, the main theme of research should be perhaps to gather evidence on the economist's approach to environmental policy and how it compares with (and sometimes complements) the regulatory alternatives. Case studies of environmental tax analysis in a cross-section of countries would also help such research in four ways: they would add to the evidence; provide a means of developing and testing the tools of analysis; document the approach, and explore the issues more fully than is possible than in the course of operations; and they would be of operational interest to other countries seeking reforms. Research programmes of this kind, combining theory, evidence and case studies, have been very successful in the past, e.g. in industry and trade, in utility pricing and investment and in social infrastructure, to mention three examples. They would provide much-needed support to ongoing policy initiatives. 67 Institute for Fiscal Studies (1989). 47 PART VI: CONCLUSIONS AND SUMMARY The preceding analysis considered how environmental policies might best be implemented through the evolving budgetary and regulatory frameworks of developing countries, and what the gains might be. Until recently, the industrial countries have relied almost exclusively on regulation for their environmental policies. But there is no reason why the developing countries need follow this path. While there always is and must be a role for regulation, a policy of taxing activities or products responsible for pollution or environmental damage, when it is administratively feasible to do so (an important qualification), would have several merits. The policy would provide polluting parties more flexibility in adjusting plans and lower the cost of compliance; it would create profit opportunities, both to domestic and foreign investors, in environmentally better products and practices; it would be administratively feasible, less costly and less prone to default than the regulatory alternative; and it would raise revenues to provide for the costs of administration. Indeed, for a period, it could also earn surpluses, which might be used either for general expenditures or to help support environmental programmes - such as soil erosion, afforestation and wildlife programmes - less capable of becoming self-financing. However, the tax approach will not always be the best option. In some cases direct regulation may be preferable, in others regulation via permits or licences, and in others a combination of taxes and regulation.' In the light of the preceding analysis, Table 7 provides a rough classification of how the available instruments might optimally work out in practice, in relation to the problem areas discussed above (see Table 6 and the accompanying discussion). It is noticeable how many areas often ideally require more than one instrument, sometimes three or four. Thus it is quite consistent to seek cost-reflecting user charges for water supply, sewerage and waste disposal services, and yet to impose regulations on the standards and safety of supplies, on the discharge of effluents, on the treatment of wastes, and so forth. Similarly, in forestry, there may ideally be user charges or licence fees for the extraction of timber, regulations on access (often with physical protection of reserves), and public investment. 68 The same conclusion has been well stated by Bohm and Russell (1985) as follows: "All the alternatives are promising in some situations ... If the classic case for the absolute superiority of effluent charges is flawed by the simplicity of the necessary assumptions, the arguments for the superiority of rigid forms of regulation suffer equally from unstated assumptions and static views of the world. There is no substitute for careful analysis of the available alternatives in the specific policy context at issue ... That said, however, we are still tempted to stress the advantage of economic incentive systems in the long run context, at least as a complement to the regulatory approach." 48 Table 7: A Preliminary Classification of Environmental Policy Instruments in relation to Key Problem Areas INSTRUMENTS PROBLEM Tax -> User Direct Permit or Public AREAS Charge Regulation Licence Investment Urban - congestion x -> x or x x - noise x ->x x - emissions x -> x x Water & Sewerage x x X Waste Disposal x x X Industry - effluents x -> x or x - gases1 x -> x or x Agriculture - chemicals x - fertilisers x - soil erosion x - soil quality, X Forestr - watersheds x x x x - rainforests x x x X - social forestry x CFCs x -> x x Wildlife Reserves x x X Research X E.g. acid deposition, leaded fuels and greenhouse gases. 2 E.g. problems arising from dessication, losses of nutrients, salination or other, depending on the area. - > denotes "evolving into". The table shows that environmental taxes may often be expected to evolve into what was termed a user charge for an environmentally unoffensive product or practice; indeed, regulations may have a similar effect. The mechanism, discussed above, is this. When an environmental tax or regulation is introduced, the initial effect is to reduce the extent to which environmentally damaging products or practices are used. More important, however, they create rents and investment opportunities in non- or low-damaging products and practices, which of 49 course are exempted from the taxes or regulations; as the activities in question develop in these new, environmentally unoffensive ways, tax revenues decline to low levels, prices come to reflect the costs of adopting them, and environmental concerns become fully internalised in the pricing system. Indeed, the value added by labour and capital in these new activities is included in the National Accounts. It was suggested that since the marginal costs of using the improved products and practices can be estimated in most cases, they would form a good yardstick for the estimation of appropriate taxes. Such cost estimates are required in any case for the analysis of investment options: in most (if not all) areas of concern, major environmental improvements can be achieved without recourse to "best-practice" techniques, sometimes at a low cost. What would developing countries stand to gain from environmental policies - aside, of course, from environmental improvements? In the first place there would be economic gains because the external costs of environmental damage would be reduced. This is a standard expectation of policy, and is as relevant to developing as to the industrial countries. Developing countries would often gain more, if only because environmental damage is often subsidised, not taxed (rainforests, agro-chemicals), while in others there is underinvestment in activities with prospectively good social rates of return (soil erosion, social forestry, water supply, sewerage systems, waste disposal, and others). Such gains at the project or sector level must also depend on macroeconomic conditions. It will be difficult for environmental policies to succeed when macroeconomic conditions are working against them. Macroeconomic instabilities and defects in structure undermine the workings of the price mechanism, and thus of the polluter-pays principle; detract from long-term thinking and the planning ahead that good environmental policy-making requires; and, because they frequently are sources of fiscal crises, reduce the resources available for the maintenance of environmental (as of material) assets. Environmental policies may sometimes work despite the macroeconomic situation - and may sometimes help it, e.g. by raising resources if the tax approach is followed. But improved macroeconomic performance will generally be fundamental for a policy's success. In two ways, environmental policies should also mesh well with the goals of poverty alleviation, though again there are exceptions. First, they would be significant sources of employment and earnings opportunities, as might be inferred from some of the problem areas listed in Table 7. Secondly, the incidence of benefits - in social forestry, for example, in water supply, sewerage and waste disposal, and in the reduction of emissions and effluents - should generally be favourable to low-income areas. Where the policies have adverse effects on employment and incomes (as could happen in coal-mining) the design of compensatory policies would self-evidently be important, and would surely not present insurmountable problems to solve. 50 Compensatory policies may also be needed in those cases where environmental policies impact adversely on trade and investment. A policy of ignoring or subsidising environmental damage in order to maintain an economy's competitiveness does not rest on good economic principles. The problems of trade and the environment are perhaps better approached, it was suggested, in another way. This is to define environmental standards appropriately by comparing the costs and benefits of various standards (including their effects on trade) before settling on a policy. In trade and investment, as in other areas, the setting of standards is an important variable in policy analysis. Again, the cost - benefit approach provides an answer to the old question of whether standards should differ between countries whose circumstances differ greatly; it was suggested that in principle they should, because the costs and benefits of achieving environmental improvements also will very often differ greatly as well. At the same time, major improvements are often possible at a low marginal cost, as noted above; in most cases it is only when standards are tightened appreciably that the marginal cost curve begins to rise steeply. In other words, standards do not need to be sacrificially low. The introduction of a coherent set of environmental policies in developing countries using the available instruments of public finance and regulation would mark a new point of departure for development policy. Few countries have attempted to find an appropriate combination of the tax and regulatory approaches. While it is possible to define the broad directions of policy, several questions remain which by their nature can only be addressed in a local context: regarding the setting of standards; the estimation of the costs to which taxes should be related; how best to phase-in environmental taxes and regulations; the distributional impact of the policies; the assessment of administrative requirements, costs and problems of compliance with policy; the establishment of environmental monitoring procedures and research programmes; the issues posed for trade and foreign direct investment; and in the design of compensating or offsetting policies in those cases (which are not in the majority) when environmental policies may impact adversely on trade and distribution. Further research and case studies would obviously help to support policy reform and to provide clarifications. However, there is much evidence and experience to go on already, and little excuse for paralysis or procrastination. Several tax, regulatory and expenditure policies can safely be recommended on the basis of what is presently known, and there is every reason to be optimistic about what they might accomplish. 51 APPENDIX: ENVIRONMENTAL TAX THEORY Current Status and Areas for Research A.1.1 Standard Results Environmental taxes can be considered to be an extension of the system of indirect taxes. The aim is to find a tax structure that maximises consumers' plus producers' surplus subject to a public revenue constraint. Following Atkinson and Stiglitz (1980), the convention is followed that the n th producer inflicts damage d,(x.) on the i th activity, where i = 1...n, x, is the output of n, and d,(x.)= 0, the costs to the polluter of his own pollution being part of his cost function, c,(x,). With Pi(xi) denoting the price demand schedule, ci(xj) costs, the objective function is { P, (x,)dx, - C(. - d(.v) + B(R) - e(.(1) 0 where B(R) represents the net benefits of expending the public revenues, R, and e(x.) represents the public financial costs of environmental damage - of administration, monitoring, research, etc. (see Section 4.4.3). Of the revenues raised by indirect taxes, only R-e(x,) are available for general expenditures; the benefits in relation to revenues raised are thus B(R)= 4(R - e(.v,))- (R - e(x,n)); R = t1x1 (2);(3) tg being the indirect tax rate on i. Denoting derivatives by primes, the marginal conditions take the familiar form that the optimal indirect taxes vary inversely with price elasticities (E ), and polluters are taxed by an additional amount related to the costs of pollution damage: ti - P, i ...n- 1-$tn= 4 -1 Pn + d( where P = c, + t1 and d = Zd being the total marginal costs of environmental damage on private parties. Notice that since 0 ' represents the marginal gross benefits of public expenditures and (except for profligate governments) is greater than unity, the polluter gets a "discount" equal to 1/ ', on account of the "contribution" of their environmental taxes to the public revenue. This follows from the point that the taxes - so long as pollution continues - can be used for tax relief or expenditures elsewhere. The tax component e', however, is earmarked for the public costs of 52 administering environmental policy; it would be preposterous (not to say inefficient) to impose a tax to provide for such costs and then to divert the revenues elsewhere. With the exception of the extra term, e ', this is a standard result of tax theory.69 It rests on several simplifying assumptions that will need to be relaxed when it comes to the design of environmental policy. They are commented on below. The main areas for further work are technical change, substitution and simulation; the responses of the polluted party, and whether there is a case for compensation; enlargement of the model to include income taxes, and the effects of pollution on savings and investment; and uncertainty. A.1.2 Technical Chanize, Substitution and Simulation The above model is still much used in environmental tax theory. However, the assumption that pollution or damage is a simple function of the output of the polluting or damaging party needs to be amended to allow for the following: i) substitution - e.g. to alternative ways of producing the same product or service (e.g. gas or low sulphur coals) or to non-polluting versions of the previously polluting products; ii) innovation - changes in practices that may reduce pollution, e.g. add-on technologies (catalytic converters, scrubbers) or new forestry practices; iii) dynamic factors - the existing capital stock (power stations, vehicle fleets, etc.) takes time to change, as do management and labour practices. Both (i) and (ii) are a function of costs and taxes, as are the rates of substitution and innovation. Dynamic simulation models are available for most sectors - energy, transport, forestry and agriculture - and could be readily updated to allow for the effects of environmental policy on a sector's development. They are potentially a powerful tool.for policy analysis, especially for the analysis of costs, and the phasing of policies and investments. They are also, of course, adaptable to the analysis of the regulatory alternatives; the dual variables to the regulatory constraints give the imputed prices and rents, and could help in comparisons of the tax and regulatory approaches. The properties of the models also need to be explored, since results like (4) and (5), though instructive, are still too elementary for the formal analysis of policies. Lastly, the models are also valuable for the analysis of uncertainties and sensitivities (A. 1.5). 69 See e.g. Atkinson and Stiglitz (1980, Chs. 12 and 14). 53 A.1.3 The Responses of the Polluted Parties The polluted parties may often undertake defensive measures (such as conservation) or restructure their factor inputs on account of the higher maintenance expenditures arising from pollution. The above model ignores these complications and also assumes pollution is a lump-sum loss. Some variants, such as those of Baumol and Oates (1988) and Newbery (1980) have extended the approach such that the polluted parties can reduce damage by reducing their outputs (e.g. by moving away from a polluted area). However, this is still to over-simplify things - land cannot be moved, for example, while treating damage lump-sum ignores a variety of possible responses by polluted parties that would be part of an optimum policy. These too require analysis. A related and more controversial issue is whether the polluted parties should be compensated. The ambiguities in present theory have several sources, two of which have already been mentioned: over-simplified response functions, and the assumption of lump-sum losses. Another assumption is that the budgetary constraints on the polluted parties are not binding, so that lump-sum or other transfers (compensations) from the polluter have no effect on marginal conditions (and thus on efficiency). Yet another problem can be seen by reference to the Coase result. The Coase theorem shows that when negotiation is possible, compensation is efficient. When it is not, because transactions costs are too high, taxation is the appropriate alternative; ideally then the aim of government interventions should be to arrive at the situati -n that would obtain if the transactions costs were low; the government would then be acting as a financial intermediary, so to speak, between the suppliers (the polluted parties) and the demanders (the polluters) of environmental services. As Newbery has shown (1980, p. 125), there is then no real difference between the Coasean bargaining solution and the situation in which environmental tax revenues are used to compensate the polluted party. (The old objection that a tax and a negotiation for compensation together are inefficient, leading to too much damage reduction, does not apply if taxation is counted as part of the overall compensation package.) These are important issues, given the large costs of environmental defensive and maintenance expenditures, and further research, resting on more realistic analysis (a) of the nature and structure of these costs and (b) of the effects of compensation, is required in relation to actual situations. A.1.4 Enlargement of the Model Even if some forms of pollution or damage occur as lump sums, they still have effects (a) on profits and incomes of the polluted parties, (b), for this reason, on savings and investment plans, and (c) on income tax payments. As the analysis is enlarged, general equilibrium effects may become important. This would be another new point of departure for research on environmental taxes. 54 A.1.5. Uncertainty and Costs Uncertainty remains a central problem for analysis. Shortcomings in environmental monitoring and research in developing countries mean that physical - let alone monetised - damage functions are not known in most cases, for forestry (e.g. for biodiversity), agriculture for surface and ground water resources, or for emissions and effluents. In Part IV, it was proposed that a practical approach would be to initiate programmes of environmental monitoring, set objectives on this basis, and work not with monetised damage functions, di(xn), but with marginal costs (MUC) of damage avoidance or reduction. The objectives would appear as constraints in the models. They could be varied to assess the costs of meeting different objectives; the dynamic models, noted above (A.1.2), would therefore provide a good tool for making environmental strategy assessments, including assessments of the tax approach and its regulatory alternatives. A separate issue is that of risks and risk-sharing. It is apparent that environmental changes may require significant - often inherently risky - investments. This is another reason for assessing policies within the framework of dynamic, broadly-based computable general equilibrium models / optimisation models that properly reflect the uncertainties involved. Once again, the issues posed by environmental policy can be seen to pose novel problems for research, both theoretical and practical. 55 ANNEX 1 ENVIRONMENTAL TAX THEORY - Current Status and Areas for Research A1.1 Standard Results Environmental taxes can be considered to be an extension of the system of indirect taxes. The aim is to find a tax structure that maximises consumers' plus producers' surplus subject to a public revenue constraint. Following Atkinson and Stiglitz (1980), the convention is followed that the n th producer inflicts damage d (x) on the i th activity, where i = 1...n, Xn is the output of n, and d,(x,) = 0, the costs to the polluter of his own pollution being part of his cost function, c(x,). With Pj(xi) denoting the price demand schedule, ci(x) costs, the objective function is {fPi(x,)dx - c,(x,) - d( x,) + B(R)- e(xn) (1) where B(R) represents the net benefits of expending the public revenues, R, and e(x,) represents the public financial costs of environmental damage - of administration, monitoring, research, etc (see section 4.4.2). Of the revenues raised by indirect taxes, only R-e(xn) are available for general expenditures, the benefits in relation to revenues raised are thus B(R)= O(R-e(xn))-(R-e(x,n));R= tx (2);(3) ti being the indirect tax rate on i. Denoting derivatives by primes, the marginal conditions take the familiar form that the optimal indirect taxes vary inversely with price elasticities (E), and polluters are taxed by an additional amount related to the costs of pollution damage: -1P, 0'-1I Pn d' tI= .-i= 1 ... n- 1 ... (4); tn = 0, .c- +--7+ e (4); (S) where Pi = ci + t- and d = d d - being the total marginal costs of environmental damage on private parties. Notice that since 0 ' represents the marginal gross benefits of public expenditures and (except for profligate governments) is greater than unity, the polluters gets a "discount" equal to 1/0 ', on account of the "contribution" of their environmental taxes to the public revenue. This follows from the point that the taxes - so long as pollution continues - can be used for tax relief or expenditures 56 elsewhere. The tax component 0 ', however, is earmarked for the public costs of administering environmental policy; it would be preposterous (not to say inefficient) to impose a tax to provide for such costs and then to divert the revenues elsewhere. With the exception of the extra term, e ', this is a standard result of tax theory". It rests on several simplifying assumptions that will need to be relaxed when it comes to the design of environmental policy. They are commented on below. The main areas for further work are technical change, substitution and simulation; the responses of the polluted party, and whether there is a case for compensation; enlargement of the model to include income taxes, and the effects of pollution on savings and investment; and uncertainty. A.1.2 Technical Change, Substitution and Simulation The above model is still much used in environmental tax theory. However, the assumption that pollution or damage is a simple function of the output of the polluting or damaging party needs to be amended to allow for the following: i) Substitution - e.g. to alternative ways of producing the same product or service, (e.g. gas or low sulphur coals) or to non-polluting versions of the previously polluting products. ii) innovation - changes in practices that may reduce pollution - e.g. add on technologies (catalytic converters, scrubbers) or new forestry practices. iii) dynamic factors - the existing capital stock (power stations, vehicle fleets, etc) takes time to change, as do management and labour practices. Both (i) and (ii) are a function of costs and taxes, as are the rates of substitution and innovation. Dynamic simulation models are available for most sectors - energy, transport, forestry and agriculture - and could be readily updated to allow for the effects of environmental policy on a sector's development. They are potentially a powerful tool for policy analysis, especially for the analysis of costs, and the phasing of policies and investments. They are also, of course, adaptable to the analysis of the regulatory alternatives; the dual variables to the regulatory constraints give the imputed prices and rents, and could help in comparisons of the tax and regulatory approaches. The properties of the models also need to be explored, since results like (4) and (5), though instructive, are still too elementary for the formal analysis of policies. Lastly, the models are also valuable for the analysis of uncertainties and sensitivities (A.1.5). 70 See e.g. Atkinson and Stiglitz (1980), Chs 12 and 14. 57 A.1.3 The Responses of the Polluted Parties The polluted parties may often undertake defensive measures (such as conservation) or restructure their factor inputs on account of the higher maintenance expenditures arising from pollution. The above model ignores these complications and also assumes pollution is a lump-sum loss. Some variants, such as those of Baumol and Oates (1988) and Newbery (1980) have extended the approach such that the polluted parties can reduce damage by reducing their outputs (e.g. by moving away from a polluted area). However, this is still to oversimplify things - land cannot be moved, for example, while treating damage lump-sum ignores a variety of possible responses by polluted parties that would be part of an optimum policy. These too require analysis. A related and more controversial issue is whether the polluted parties should be compensated. The ambiguities in present theory have several sources, two of which have already been mentioned: over-simplified response functions, and the assumption of lump sum losses. Another assumption is that the budgetary constraints on the polluted parties are not binding, so that lump-sum or other transfers (compensations) from the polluter have no effect on marginal conditions (and thus on efficiency). Yet another problem can be seen by reference to the Coase result. The Coase theorem shows that when negotiation is possible, compensation is efficient. When it is not, because transactions costs are too high, taxation is the appropriate alternative; ideally then the aim of government interventions should be to arrive at the situation that would obtain if the transactions costs were low; the government would then be acting as a financial intermediary, so to speak, between the suppliers (the polluted parties) and the demanders (the polluters) of environmental services. As Newbery has shown (1980, p 125), there is then no real difference between the Coasean bargaining solution and the situation in which environmental tax revenues are use to compensate the polluted party. [The old objection that a tax and a negotiation for compensation together is inefficient, leading to too much damage reduction, does not apply if taxation is counted as part of the overall compensation package.] These are important issues given the large costs of environmental defensive and maintenance expenditures, and further research, resting on more realistic analysis (a) of the nature and structure of these costs and (b) the effects of compensation is required in relation to actual situations. A.1.4 Enlargement of the Model Even if some forms of pollution or damage occur as lump-sums, they still have effects (a) on profits and incomes of the polluted parties, (b), for this reason, on savings and investment plans, and (c) on income tax payments. As the analysis is enlarged, general equilibrium effects may become important. This would be another new point of departure for research on environmental taxes. 58 A. 1.5. Uncertainty and Costs Uncertainty remains a central problem for analysis. Shortcomings in environmental monitoring and research in developing countries mean that knowkdge of physical - let alone monetised - damage functions are not known in most cases, for forestry (e.g. of biodivesity), agriculture for surface and ground-water resources, or for emissions and effluents. In Part 4, it was proposed that a practical approach would be to initiate programmes of environmental monitoring, set objectives on this basis, and work not with monetised damage functions, di(xn), but with marginal costs (MUCi) of damage avoidance or reduction. The objectives would appear as constraints in the models. They could be varied to assess the costs of meeting different objectives; the dynamic models, noted above (A.1.2), would therefore provide a good tool for making environmental strategy assessments, including assessments of the tax approach and its regulatory alternatives. A separate issue is that of risks and risk-sharing. It is apparent that environmental changes may require significant - often inherently risky - investments. This is another reason for assessing policies within the framework of dynamic, broadly-based CGE/optimistation models that properly reflect the uncertainties involved. Once again, the issues posed by environmental policy can be seen to pose novel problems for research, both theoretical and practical. 59 REFERENCES Ahmed, Yusuf J, Salah El Serafy and Earnst Lutz (1989) Environmental Accounting for Sustainable Development. Report on a UNEP World Bank Symposium. The World Bank, Washington DC. Anderson, Dennis (1987) The Economics of Afforestation: A Case Study in Africa World Bank Occasional Paper No. 1/New Series Baltimore and London: Johns Hopkins University Press. Anderson, Dennis (1989) "Infrastructure Pricing Policies and the Public Revenue in African Countries". World Development, Vol. 17, No. 4, pp.525-542. Anderson, Dennis (1990a) "Economic Growth and the Environment". Shell Selected Papers. Shell International Petroleum Company Ltd., Shell Centre, London SE1 7NA, England. Anderson, Dennis (1990b) "Investment and Economic Growth". World Development (forthcoming). Arthur, James and John Kirk (1985) [ILO Paper on Water Supply in Developing Countries. Mimeo] Ashby, Eric and Mary Anderson (1981) The Politics of Clean Air. Oxford: Oxford University Press. Atkinson, Anthony B and Joseph E Stiglitz (1980) Lectures on Public Economics. New York: McGraw-Hill. Barbier, Edward B and David W Pearce (1990) 'Thinking Economically About Climate Change". Energy Policy, January/February, pp.11-19. Barde, Jean-Phillippe (1989) 'The Econom" Approach to the Environment". OECD Observer, June/July, No. 158 pp.12-15, Paris. Barrett, Scott A. (1986) 'The Use of Market Mechanism in the Regulation of Air Pollution" Paper for National Economic Research Associates (NERA), 18 Park Street, London WIY 3WD. Baumol, William J and Wallace E Oates (1971) 'The Use of Standards and Prices for Protection of the Environment". Swedish Journal of Economics. LXXIII (March). Baumol, William J and Wallace E Oates (1988) The Theory of Environmental Policy Cambridge: Cambridge University Press. Beckerman, Wilfred (1974) In Defense of Economic Growth. London: Jonathan Cape. Bendixson (1989) Transport in the Nineties: The Shaping of Europe, Royal Institute of Chartered Surveyors, October. 60 Binswanger, Hans (1987) "Fiscal and Legal Incentives with Environmental Effects on the Brazilian Amazon". Agriculture and Rural Development Department, World Bank. Processed. Bohm, Peter and Clifford S. Russell (1985) "Comparative Analysis of Alternative Policy Instruments" Ch.10 of Kneese and Sweeney (1985), Vol.1, pp.395-460. Brandle, James R, D L Hintz and J W Sturrock (1988) Windbreak Technology, Amsterdam: Elsevier Press. Brimblecombe, Peter (1987) The Big Smoke. London and New York: Reutledge. Buchanan, James M. (1963) 'The Economics of Environmental Taxes", vol.71, pp.457-69. Churchill, Anthony (1972) Road User Charges in Central America World Bank Staff Occasional Papers, No. 15 (Baltimore, MD: John Hopkins University Press). Churchill, Anthony (1987) "Rural Water Supply and Sanitation: Time for a Change". World Bank Discussion Paper, No. 18. Commonwealth Secretariat (1989) Climate Change: Meeting the Challenge. Report by a Commonwealth Group of Experts led by Martin W Holdgate. Marlborough House, Pall Mall, London SW1Y 5HX. Dasgupta, Partha and Karl-Goran Maler (1990) 'The Environment and Emerging Development Issues" Paper for the Annual World Bank Conference on Development (forthcoming proceedings). Easter, William K., John A. Dixon and Maynard M. Hufschmidt (Editors) (1986) Watershed Resources Management: An Integrated Framework with Studies from Asia and the Pacific. Studies in Water and Policy Management, No.10, Boulder and London: Westview Press. Falloux, Franqois, J. R. Mercier and N. Savine (1988) "Madagascar: Plan d'Action Environmental" Co-operative Study of the World Bank, US AID, Coop6ration Suisse, UNESCO, UNDP and the World Wide Fund for Nature. Available from the World Bank Technical Staff Department for Africa. Findley, Roger W. (1988) "Pollution Control in Brazil" Ecology Law Quarterly, Vol.15. No.1, pp.1-68. Food and Agriculture Organisation (1985) 'Tree Growing by Rural People". FAO Forestry Paper No. 64. Rome. Goodwin, P B and P M Jones (1989) "Road Pricing: The Political and Strategic Possibilities". A report of the Transport Studies Unit, Oxford University, Oxford OX2 6NB. TSU Ref: 40. 61 Goodwin Peter, B. (1989) 'The Rule of Three? A Possible Solution to the Political Problem of Competing Objectives for Road Pricing" Traffic Engineering and Control, October. Gregersen, Hans, Sydney Draper and Dieter Elz, editors, (1989) People and Trees: The Role of Social Forestry in Sustainable Development. Economic Development Institute Seminar Series, World Bank, Washington DC. Grubb, Michael (1989) 'The Greenhouse Effect: Negotiating Targets". Royal Institute for International Affairs, London. Hahn, Robert W (1989) "Economic Prescriptions for Environmental Problems: How the Patient Followed the Doctor's Orders". The Journal of Economic Perspectives, Vol. 3, No. 2. Spring, pp.95-114. Hamrin, Robert D (1989) "Policy Control Options for Comparative Air Pollution Study in Urban Areas". World Bank Energy Department, Mimeo, September 1989. Hecht, Suzanna B, R B Norgard and G Possio n.d. 'The Economics of Cattle Ranching in Eastern Amazonia". Graduate School of Architecture and Planning, University of California, Los Angeles. Processed. Helm, Dieter and David W. Pearce (eds) (1990) "Economics of the Environment" Special Edition of the Oxford Review of Economic Policy, Vol.6, No.1, Spring. Findley, Roger W. (1988) "Pollution Control in Brazil, Ecology Law Quarterly, Vol.15, No.1. Institute for Fiscal Studies (1989) 'Tax Reform and the Environment: A Research Proposal" November. London WIP 9LE. International Energy Agency (1989) Energy and the Environment: Policy Overview, OECD, Paris. Jones, Peter M (1989) 'The Restraint of Road Traffic in Urban Areas: Objectives, Options and Experiences". Transport Studies Unit, Oxford University. Discussion Paper 3. Julius, DeAnne S, and Jeremy J Warford (1979) "Water Rate Policy: Lessons from Less Developed Countries". Finance and Development, Vol. 16, pp.199-203. Kneese, Allen V., Sidney E. Rolfe and Joseph W Hamed, (eds) (1971) Managing the Environment: International Economic Cooperation for Pollution Control. New York: Praeger Publishers. Kneese, Allen V., Robert U. Ayers and Ralph C. d'Arge (1970) Economics and the Environment: A Material Balance Approach, Baltimore, Md: Johns Hopkins University Press. 62 Kneese, Allen V. and Blair T. Bower (1979) Environmental Quality and Residuals Management, Baltimore, Md: John Hopkins University Press. Kneese, Allen V. and James L. Sweeney (1985) Handbook of Natural Resource and Enery Economics, Volume 1 (of three volumes). Amsterdam: North Holland. Mahar, Dennis J (1989) "Deforestation in Brazil's Amazon Region: Magnitude Rate, and Causes". Chapter 7 in Schramm and Warford (1989). Meadows, Dennis, Donella Meadows, J Randers and W W Behrens (1972) The Limits to Growth, New York: Universe Books. Meerman, Jacob (1979) Public Expenditure in Malaysia: Who Benefits and Why. New York: Oxford University Press. Munasinghe, Mohan and Jeremy J Warford (1982) Electricity Pricing: Theory and Case Studies. Baltimore MD: John Hopkins University Press. Munasinghe, Mohan, Joseph Gilling and Melody Mason (1988) "A Review of World Bank Lending for Electric Power". World Bank Industry and Energy Department Paper, March. Muzando, Timothy R., Kenneth M. Miranda and A. Lans Bovenberg (1990) "Public Policy and the Environment" the Fiscal Affairs Department, IMF (processed). National Academy Press (1983) Agroforestry in the West African Sahel, Washington DC. Nelson, Ridley (1988) "Dryland management: The 'Desertification' Problem". World Bank Environment Department Working Paper No. 8. Newbery, David M G (1980) "Externalities: The Theory of Environmental Policy". Chapter 4 in G A Hughes and G M Heal (editors), Public Policy and the Tax System, London: George Allen and Unwin, 1980. Newbery, David M G and Nicholas Stern (1987) The Theory of Taxation for Developing Countries. Oxford: Oxford University Press. Nicholson, Sharon E (1982) 'The Sahel: A Climatic Perspective". Report for the Club du Sahel, Paris. OECD (1988) Emissions Controls in Electricity Generation and Industry, Paris: OECD Publications. OECD (1990) Energy and the Environment: Policy Overview, Paris: OECD Publications. OECD (1989) Enery Technology for Reducing Emissions of Greenhouse Gases. Proceedings of an Expert Seminar, Paris 12th-14th April, OECD Publications. OECD (1987) Renewable Sources of Energy, Paris: OECD Publications. 63 Opschoor, J B and H Vos (1989) Economic Instruments for Environmental Protection in OECD Member Countries, OECD, Paris. Parker, Andrew N (1989) "Environmental Policy: Case Studies in Water and Air Pollution". University College London. Processed. Pearce, David W, Anil Markandya and Edward Barbier (1989) Blueprint for a Green Economy, London: Earthscan. Pearce, David W and Anil Markandya (1989) Environmental Policy Benefits: Monetary Valuation, OECD, Paris. Pearce, David W and K Kerry Turner (1989) Economics of Natural Resources and the Environment. London and New York: Harvester Wheatsheaf. Pezzey, John (1989) "Economic Analysis of Sustainable Growth and Sustainable Development". World Bank Environment Department Working Paper No. 15. Portney, Paul R. (1989) "Regulating Industrial Pollution in Developing Countries: Some Possible Frameworks" World Bank Internal Discussion Paper, Latin America and the Caribbean Region, Report IDP-0056, December, Washington DC 20433. Prest, A R (1985) Public Finance in Developing Countries, London: Weidenfeld and Nicolson. Prinsley, R T and M J Swift (1986) Amelioration of Soil by Trees: A Review of Current Concepts and Practices. Commonwealth Science Council, Marlborough House, London SW1Y 5HX. Raintree, J B (1984) "A Diagnostic Approach to Agro-Forestry Design", in K F Wiersum (1984). Repetto, Robert (1988) "Economic Policy Reform for Natural Resource Conservation". World Bank Environment Department Working Paper No. 4. Repetto, Robert (1989) "Economic Incentives for Sustainable Production". Ch. 6 in Schramm and Warford (1989). Ridley, Nicholas (1989) "Policies Against Pollution". Policy Study 107, Centre for Policy Studies, London. Schramm, Gunter (1985) "Energy Situation in Africa". World Bank. Processed. Saunders, Robert J, Jeremy J. Warford and Patrick C. Mann (1977) "Alternative Concepts of Marginal Cost for Public Utility Pricing", World Bank Staf` Working Paper, No.259 (May). Schramm, Gunter and Jeremy J Warford, editors (1989) Environmental Management and Economic Development, Baltimore and London: The Johns Hopkins University Press. Scott, Maurice Fitzgerald (1989) A New View of Economic Growth. Oxford: Clarendon Press. 64 Selowsky, Marcelo (1980) Who Benefits from Public Expenditures? A case study in Columbia. New York: Oxford University Press. Sen, Amartya (1983) "Development: Which way Now?". Economic Journal, Vol 93 pp. 745-762. Smith, Stephen and Mark Pearson (1990) 'Taxation and Environmental Policy: Some Initial Evidence" Institute for Fiscal Studies, London W1P 9LE, IFS Commentary No.19. Smith, V. Kerry (ed) (1988) Environmental Resources and Applied Welfare Economics. Essays in honor of John V. Krutilla. Washington DC: Resources for the Future. Southgate, Douglvs (1988) 'The Economics of Land Degradation in the Third World". Worid Bank Environment Paper No. 2. Spears, John (1988) "Containing Tropical Deforestation: A Review of Priority Areas for Technological and Policy Research". Environment Department Working Paper No. 10, World Bank. Stanford, Stephen (1983) Management of Pastoral Development in the Third World. Overseas Development Institute London. Chichester and New York: John Wiley. Stern, Nicholas and David M G Newbery (1987) The Theory of Taxation for Developing Countries. Oxford: Oxford University Press. Turnham, David (1990) A note on environmental issues in developing countries (typescript) in preparation for an OECD conference. OECD. Turvey, Ralph (1963) "On Divergences between Social Cost and Private Cost". Economica, August 1963. Turvey, Ralph (1972) 'The Price of Time in a Technological Age", QUEST, No.22, Autumn 1972, pp.3,f-37 Walters, Alan (1968) The Economics of Road User Charges. Baltimore MD: Johns Hopkins University Press. Warford, Jeremy J (1966) "Water requirements: The investment decision in the water supply industry". Manchester School. vol. 34, pp.87-112. Reprinted (with revisions) in R Turvey, ed. Public Enterprise, Penguin Modem Economics (1968). Ch. 6. Warford, Jeremy J (1987) "Environment, Growth and Development" Paper No. 14 for the Development Committee of the World Bank and IMF, August 1987. World Bank Washington DC 20433. Warford, Jeremy J (1989) "Environmental Management and Economic Policy in Developing Countries". Chapter 2 in Schramm and Warford (1989). 65 Warford, Jeremy J Bank (1989) "World Bank Support for the Environment: A Progress Report". Staff Report No 22 to the Development Committee, World Bank. Washington DC. Wiersum, K F, ed (1984) Strategies and Designs for Afforestation, Reforestation and Tree Planting. Wageningen, Netherlands: Centre for Agricultural Publishing and Documentation (PUDOC). World Bank (1988) Staff Appraisal Report of the Lagos Water Supply Project. 6375-UNI Washington DC (Restricted). World Bank (1978) Forestry. A World Bank Sector Policy Paper. Washington DC. World Bank (1989) "Greenhouse Gases and the Potential for Global Warming: Energy Options for the Developing World and ESMAP's Role". A background paper for discussion at the Energy Strategy and Management Assistance Programm, November 27-29, Paris. World Bank Energy Department. Also: "Energy Efficiency Strategy for Developing Countries", (same source and meeting). World Commission on Environment and Development (1989) Our Common Future (the Brundtland Report), Oxford University Press. World Resources Institute, World Bank and United Nations Development Programme (1985). Tropical Forests: A Call for Action. International Task Force Report. Washington DC: World Resources Institute. 66 Al Examples of Add-on Air Pollution Emission Control Technologies Pollutant Impact on Controlled" Energy Status," Effciency' 1. STATIONARY SOURCES Fossil Fuel Combustion Facilities SLow-NO, bumiers NO% Variable Commercial SCatalytc combustion NO,. CO + R&D * Furnace sorbent injection SO, Variable Demonstrated . Off-stoichiometric control NO Variable Commercial . FGD (wet or spray-drver) SO, - Commercial . Combined de-SO J-NO, SO-, NO, - Commercial FGD C . SCR NOI Commercial . Electrostatic precipirators PM - Commercial . Baghouse fabric tilters PM - Commercial Other Stationary Facilities . Floating covers or roofs for VOC + Commercial storage tanks and waste water separator bays . Vapor recovery in petroleum VOC + Commercial distribution activities 2. MOBILE SOURCES Gasoline-Fuelled Vehicles Reduction catalyst NO - Commercial Oxidation catalyst HC' CO Commercial . Three-way catalyst NO,. HC. - Commercial CO . Thermal reactor HC. CO - Commercial . Eneine tuning (ignition timing. HC. NO,, - Commercial air/fuel ratio and mixture CO preparation) . Exhaust gas recirculation NO, - Commercial Diesel-Fuelled Vehicles . Trap systems PM/PAH Variable Commercial . Trap oxidizer systems PM/PAH Variable Demonstrated Notes: <in pollutant controiled: main target pollutants (2) impact on energy effictency: positive t+i or negative (-) (3) stats: technologies are classified here as still requirtng research and development (R&D). demonstrated in some locations (demons.), or avaiabe in the market in some locations (commeicial). Source. tEA Secretariat. 67 A I Examples of "Clean" Energy Technologies Pollutant Impact on Status Controlled Energy EMelency 1. STATIONARY SOURCES Fossil fuel combustion facilities . FBC SO,. NO, + Commercial . AFBC SO:. NO, Variable Demonstrated . PFBC SO;. NO, + Demonstrated * IGCC SO,. NO, + Demonstrated . High-efficiency gas turbines NO, + Commercial Co-firing SO,. NO, + Commercial Gas re-bum/with sorbent SO,-. NO,, + Commercial injection Plkf 2. MOBILE SOURCES Gasoline-fuelled vehicles . Changes in compression ratio. HC. NO, + Commercial combustion chamber design . High-compression lean-bum CO. NO, + Demonstrated entines Commercial . Stratified charge engines CO. NO, + R&D * Fast-bum systems NO. + Demonstrated Diesel-fuelled vehicles * Air/fuel ratio PM/PAH Variable Demonstrated . Controlled electronic fuel HC. NO,. Commercial injection CO . Four-stroke adiabatic engine PM/PAH 4 R&D . Turbo-charger/ HC. CO + Commercial turbo-intercooling Evaporative losses . Modifizations to the tank and VOC + Commercial carburetor . Carbon canisters VOC + Commercial Notes: Pollutant controlled main target pollutants. Impact on energy eficienev: positive (+) or negative *i. Status: technologies are classtfied here as still requinng research and development (R&D). demonstrated in some locations (demons.i. or available on the market in some locations (Commercialk. Source: 1EA Secretanat. 68 ~ E ent T«h~aninl ad ti E~a Po~esdaI fr Cons~vadoa Eness 8.d *0 Kf ohiI Saoe New Stoek 5~t Bes Awvøge Ui iMM t UrT g IMA Avrap AMr~e Aaie Avallelie Teeilg TPU Effdi ~ EMd~ary Tetlogy Te~.aiw <Units) t% Sasline Elidec 4% Svl RESIDENTIAL 20-25% . U.S. (All eC~re I 501 (Want per 328 -78% Over 30 yen •Sweden (AI I 242 (Was per 266 -78% eklne uses capital He and e 8-12% - 4ulding shen Oer 30 yemr - U.S.wintr 160 (KJ perm io <-37%> 50 -70% per den~ day) - Swedn (wintr> 13$ IKJ per m' 65 (-52%) 35 .74% per de day> Heatne 8-12% . Oia~ - System 020 ytan Et Eiciency US. 65.70% (% af 75-0% 84-94% .23-26% TPER convened t.13%1 to uneful heat) C4ådenq 1-2% - Cemal ak. 10.2< 7emn • U.S. 7 IEney 9 .22.) 14 -50% Etnciency Ratag Refrigerator 2% I0-Is yean us. I 5( I 300 t.13%) 750 .50% (kW>yearl • Gemny About J0 -20%) At le Ik~byr .20% • Japan 3$ W 28 (20%) Al east -20% W<ter Hu.e 3-5 I$ yeun -<U.S. 4 00 300 1710 -57% (kWillyearl COMMERCIAL 1$-20% Hewine and 10-1.2% 30+ yeam - US. 31 iGr per t.73 (444) 0.32. -75% - Sweden 1.0 0.76 t.27%> 02540.46 .5.75% Lare Office 8m/ldings - US. 270 (KBtuft 200 t-26%) Kø11 -63% 30+yea- yearl LJeknee 3-5% I-.10 years - U.5. .Bailast/Tub"e 64 :l men~Watil 73 9-2) 16 (2% .Control' <-20-30%) .Total .0-30% TRANSPORT 20-25% 4(tan.neIs 10-13% 10 yem - US. 190 tmilet per 26.1 <-344) 31.5 6% galion) • Japan II (knif) M>3 (15%) V«port 2.3% 25 vrger 30+ 40. -40% 15-30 yeas Air Trampi estgaMon (.20%) - Al cou~ues 2.3% R#Maruse,O,r INDUSTRY 3540% Che~vaks 6-8% - U.K. (Iorgamc -13% Ir«n at St~el % - U.SjlapanU.KJ1 22 24 (GIAonne~ 17.18 a.a. Atleau- 10-30yean Nether -225% 20-25% Mofe-rmu aMrtas 3% - OECD 15.17 13.3 .a. At leas 20-30 yean (Aufmflf "mWn)onne 1-10-20%) -10-20% 69 % of Euisting Stoch Ne.w Stock Sest Stn Ameage Useful Ut. et m En EA Awaga Amra Avaiab~ Aviem Tedm TPER Efei,ney E~ Tu~ Tecnu~gy (Unul t% Savng Eficieny t% S£Mi~ga> Paper 3% • U.K. (Pap~ and -30% Boond Making) S~one. Clky and 2% • U.SJFraneSwitz. 2.5 (Mkg) 1.5-2.0 a. At least 1030 yeas /U.K. (Buicks/ (-20^40%) -20.40% • Frni~J.KJ 3.6-3.8 (MiAg) 3.3 (4-13%) na. At las 10-30 yem SwitdOennany •13% (Camenu) Fond 1% Sp*.e Herng. Carding. Water. 2-3% H~eing. Lg~tin ALL SECTORS Elerme MatOrs 20% 75-90% (% 80-92% 85.93% .51-30% 10-20 yans convrted to (-2.7%) motive poerp Cemal ad On-Sie 35% Genraion~ • U.S. (Gas a. 30% (% 35% (-15%) 39-41% -25% Tuiones canvened to tee aSued Eastfee Canwu .a tra C.* . g7A. 07 70 REFERENCES Ahmed, Yusuf J., Salah El Serafy and Earnst Lutz (1989) Environmental Accounting for Sustainable Development. Report on a UNEP World Bank Symposium. Washington DC: World Bank. Anderson, Dennis (1987) The Economics of Afforestation: A Case Study in Afjica World Bank Occasional Paper No. 1/New Series. Baltimore and London: Johns Hopkins University Press. Anderson, Dennis (1989) "Infrastructure Pricing Policies and the Public Revenue in African Countries". World Development, Vol. 17, No. 4, pp. 525 - 542. Anderson, Dennis (1990a) "Economic Growth and the Environment". Shell Selected Papers. Shell International Petroleum Company Ltd., Shell Centre, London SEI 7NA, England. Anderson, Dennis (1990b) "Investment and Economic Growth". World Development (forthcoming). Arthur, James and John Kirk (1985) ILO Paper on Water Supply in Developing Countries. Mimeo. Ashby, Eric and Mary Anderson (1981) The Politics of Clean Air. Oxford: Oxford University Press. Atkinson, Anthony B. and Joseph E. Stiglitz (1980) Lectures on Public Economics. New York: McGraw-Hill. Barbier, Edward B. and David W. Pearce (1990) 'Thinking Economically about Climate Change". Energy Policy, January/February, pp. 11 - 19. Barde, Jean-Phillippe (1989) '"The Economic Approach to the Environment". OECD Observer, June/July, No. 158, pp. 12- 15. Paris. Barrett, Scott A. (1986) 'The Use of Market Mechanism in the Regulation of Air Pollution". Paper for National Economic Research Associates (NERA), 18 Park Street, London W1Y 3WD. Baumol, William J. and Wallace E. Oates (1971) "The Use of Standards and Prices for Protection of the Environment". Swedish Journal of Economics, LXXIII, March. Baumol, William J. and Wallace E. Oates (1988) The Theory of Environmental Policy. Cambridge: Cambridge University Press. Beckerman, Wilfred (1974) In Defense of Economic Growth. London: Jonathan Cape. Bendixson, T. (1989) Transport in the Nineties: The Shaping of Europe. Royal Institute of Chartered Surveyors. October. 71 Binswanger, Hans (1987) "Fiscal and Legal Incentives with Environmental Effects on the Brazilian Amazon". Agriculture and Rural Development Department, World Bank. Processed. Bohm, Peter and Clifford S. Russell (1985) "Comparative Analysis of Alternative Policy Instruments". Ch. 10 in Kneese and Sweeney (1985), pp. 395 - 460. Brandle, James R., D. L. Hintz and J. W. Sturrock (1988) Windbreak Technology. Amsterdam: Elsevier Press. Brimblecombe, Peter (1987) The Big Smoke. London and New York: Routledge. Brundtland Report: World Commission on Environment and Development (1989) Our Common Future. Oxford: Oxford University Press. Buchanan, James M. (1963) "The Economics of Earmarked Taxes". Journal of Political Economy, Vol. 71, pp. 457 - 469. Churchill, Anthony (1972) Road User Charges in Central America. World Bank Staff Occasional Papers, No. 15. Baltimore, MD: Jnhns Hopkins University Press. Churchill, Anthony (1987) "Rural Water Supply and Sanitation: Time for a Change". World Bank Discussion Paper, No. 18. Commonwealth Secretariat (1989) Climate Change: Meeting the Challenge. Report by a Commonwealth Group of Experts led by Martin W. Holdgate. Marlborough House, Pall Mall, London SW1Y 5HX. Dasgupta, Partha and Karl-Goran Maler (1990) "The Environment and Emerging Devt *)pment Issues". Paper for the Annual World Bank Conference on Development (forthcoming proceedings). Easter, William K., John A. Dixon and Maynard M. Hufschmidt (editors) (1986) Watershed Resources Management: An Integrated Framework with Studies from Asia and the Pacific. Studies in Water and Policy Management, No. 10. Boulder and London: Westview Press. Falloux, Franqois, J. R. Mercier and N. Savine (1988) "Madagascar: Plan d'Action Environmental". Co-Operative Study of the World Bank, US AID, Coop6ration Suisse, UNESCO, UNDP and the World Wide Fund for Nature. Available from the World Bank Technical Staff Department for Africa. Findley, Roger W. (1988) "Pollution Control in Brazil". Ecology Law Quarterly, Vol. 15, No. 1, pp. 1 - 68. Food and Agriculture Organisation (1985) 'Tree Growing by Rural People". FAO Forestry Paper No. 64. Rome. Goodwin, Peter B. (1989) "The Rule of Three? A Possible Solution to the Political Problem of Competing Objectives for Road Pricing". Traffic Engineering and Control, October. 72 Goodwin, P. B. and P. M. Jones (1989) "Road Pricing: The Political and Strategic Possibilities". A Report of the Transport Studies Unit, Oxford University, Oxford OX2 6NB. TSU Ref: 40. Gregersen, Hans, Sydney Draper and Dieter Elz (editors) (1989) People and Trees: The Role of Social Forestry in Sustainable Development. Economic Development Institute Seminar Series. Washington DC: World Bank. Grubb, Michael (1989) 'The Greenhouse Effect: Negotiating Targets". London: Royal Institute for International Affairs. Hahn, Robert W. (1989) "Economic Prescriptions for Environmental Problems: How the Patient Followed the Doctor's Orders". Journal of Economic Perspectives, Vol. 3, No. 2, Spring, pp. 95 - 114. Hamrin, Robert D. (1989) "Policy Control Options for Comparative Air Pollution Study in Urban Areas". World Bank Energy Department, Mimeo, September. Hecht, Suzanna B., R. B. Norgard and G. Possio (not dated) 'The Economics of Cattle Ranching in Eastern Amazonia". Graduate School of Architecture and Planning, University of California, Los Angeles. Processed. Helm, Dieter and David W. Pearce (editors) (1990) "Economics of the Environment". Special Edition of the Oxford Review of Economic Policy, Vol. 6, No. 1, Spring. Hughes, G. A. and G. M. Heal (editors) (1980) Public Policy and the Tax System. London: George Allen and Unwin. Institute for Fiscal Studies (1989) 'Tax Reform and the Environment: A Research Proposal". November. IFS, 180/182 Tottenham Court Road, London W1P 9LE. International Energy Agency (1989) Energy and the Environment: Policy Overview. Paris: OECD. Jones, Peter M. (1989) "The Restraint of Road Traffic in Urban Areas: Objectives, Options and Experiences". Discussion Paper No. 3, Transport Studies Unit, Oxford University. Julius, DeAnne S., and Jeremy J. Warford (1979) "Water Rate Policy: Lessons from Less Developed Countries". Finance and Development, Vol. 16, pp. 199-203. Kneese, Allen V., Robert U. Ayers and Ralph C. d'Arge (1970) Economics and the Environment: A Material Balance Approach. Baltimore, MD: Johns Hopkins University Press. Kneese, Allen V. and Blair T. Bower (1979) Environmental Quality and Residuals Management. Baltimore, MD: Johns Hopkins University Press. 73 Kneese, Allen V., Sidney E. Rolfe and Joseph W. Hamed (editors) (1971) Managing the Environment: International Economic Cooperation for Pollution Control. New York: Praeger Publishers. Kneese, Allen V. and James L. Sweeney (1985) Handbook of Natural Resource and Ener, Economics. Volume 1 (of three volumes). Amsterdam: North Holland. Mahar, Dennis J. (1989) "Deforestation in Brazil's Amazon Region: Magnitude, Rate and Causes". Ch. 7 in Schramm and Warford (1989). Meadows, Dennis, Donella Meadows, J. Randers and W. W. Behrens (1972) The Limits to Growth. New York: Universe Books. Meerman, Jacob (1979) Public Expenditure in Malaysia: Who Benefits and Why. New York: Oxford University Press. Munasinghe, Mohan, Joseph Gilling and Melody Mason (1988) "A Review of World Bank Lending for Electric Power". World Bank Industry and Energy Department Paper, March. Munasinghe, Mohan and Jeremy J. Warford (1982) Electricity Pricing: Theory and Case Studies. Baltimore, MD: Johns Hopkins University Press. Muzando, Timothy R., Kenneth M. Miranda and A. Lans Bovenberg (1990) "Public Policy and the Environment". Fiscal Affairs Department, IMF. Processed. National Academy Press (1983) Agroforestry in the West African Sahel. Washington DC. Nelson, Ridley (1988) "Dryland Management: The 'Desertification' Problem". World Bank Environment Department Working Paper No. 8. Newbery, David M. G. (1980) "Externalities: The Theory of Environmental Policy". Ch. 4 in Hughes and Heal (1980). Newbery, David M. G. and Nicholas Stern (1987) The Theory of Taxation for Developing Countries. Oxford: Oxford University Press. Nicholson, Sharon E. (1982) 'The Sahel: A Climatic Perspective". Report for the Club du Sahel, Paris. OECD (1987a) Renewable Sources of Energ. Paris: OECD Publications. OECD (1987b) Emissions Control. Paris: OECD. OECD (1988) Emission Controls in Electricity Generation and Industry. Paris: OECD Publications. OECD (1989) Energy Technolog for Reducing Emissions of Greenhouse Gases. Proceedings of an Expert Seminar, Paris 12 - 14 April. Paris: OECD Publications. 74 OECD (1990) Energy and the Environment: Policy Overview. Paris: OECD Publications. Opschoor, J. B. and H. Vos (1989) Economic Instruments for Environmental Protection in OECD Member Countries. Paris: OECD. Parker, Andrew N. (1989) "Environmental Policy: Case Studies in Water and Air Pollution". University College London. Processed. Pearce, David W. and Anil Markandya (1989) Environmental Policy Benefits: Monetary Valuation. Paris: OECD. Pearce, David W., Anil Markandya and Edward Barbier (1989) Blueprint for a Green Economy. London: Earthscan. Pearce, David W. and K. Kerry Turner (1989) Economics of Natural Resources and the Environment. London and New York: Harvester Wheatsheaf. Pezzey, John (1989) "Economic Analysis of Sustainable Growth and Sustainable Development". World Bank Environment Department Working Paper No. 15. Portney, Paul R. (1989) "Regulating Industrial Pollution in Developing Countries: Some Possible Frameworks". World Bank Internal Discussion Paper, Latin America and the Caribbean Region, Report IDP-0056, December. Washington DC 20433. Prest, A. R. (1985) Public Finance in Developing Countries. London: Weidenfeld and Nicolson. Prinsley, R. T. and M. J. Swift (1986) Amelioration of Soil by Trees: A Review of Current Concepts and Practices. Commonwealth Science Council, Marlborough House, London SWIY 5HX. Raintree, J. B. (1984) "A Diagnostic Approach to Agro-Forestry Design". In Wiersum (1984). Repetto, Robert (1988) "Economic Policy Reform for Natural Resource Conservation". World Bank Environment Department Working Paper No. 4. Repetto, Robert (1989) "Economic Incentives for Sustainable Production". Ch. 6 in Schramm and Warford (1989). Ridley, Nicholas (1989) "Policies Against Pollution". Policy Study No. 107. Centre for Policy Studies, London. Saunders, Robert J., Jeremy J. Warford and Patrick C. Mann (1977) "Alternative Concepts of Marginal Cost for Public Utility Pricing". World Bank Staff Working Paper No. 259, May. Schramm, Gunter (1985) "Energy Situation in Africa". World Bank. Processed. Schramm, Gunter and Jeremy J. Warford (editors) (1989) Environmental Management and Economic Development. Baltimore and London: Johns Hopkins University Press. 75 Scott, Maurice Fitzgerald (1989) A New View of Economic Growth. Oxford: Clarendon Press. Selowsky, Marcelo (1980) Who Benefits from Public Erpenditures? A Case Study in Columbia. New York: Oxford University Press. Sen, Amartya (1983) "Development: Which Way Now?". Economic Journal, Vol. 93, pp. 745 - 762. Smith, Stephen and Mark Pearson (1990) 'Taxation and Environmental Policy: Some Initial Evidence". IFS Commentary No. 19. Institute for Fiscal Studies, 180/182 Tottenham Court Road, London W1P 9LE. Smith, V. Kerry (editor) (1988) Environmental Resources and Applied Welfare Economics. Essays in honour of John V. Krutilla. Washington DC: Resources for the Future. Southgate, Douglas (1988) 'The Economics of Land Degradation in the Third World". World Bank Environment Paper No. 2. Spears, John (1988) "Containing Tropical Deforestation: A Review of Priority Areas for Technological and Policy Research". Environment Department Working Paper No. 10, World Bank. Stanford, Stephen (1983) Management of Pastoral Development in the Third World. Overseas Development Institute, London. Chichester and New York: John Wiley. Stern, Nicholas and David M. G. Newbery (1987) The Theory of Taxation for Developing Countries. Oxford: Oxford University Press. Turnham, David (1990) A note on environmental issues in developing countries (typescript) in preparation for an OECD conference. OECD. Turvey, Ralph (1963) "On Divergences between Social Cost and Private Cost". Economica, August. Turvey, Ralph (editor) (1968) Public Enterprise. Penguin Modern Economics. Turvey, Ralph (1972) 'The Price of Time in a Technological Age". QUEST, No. 22, Autumn, pp. 30-37. Turvey, Ralph and Dennis Anderson (1977) Electricity Economics: Essays and Case Studies. Baltimore, MD: Johns Hopkins University Press. Walters, Alan (1968) The Economics of Road User Charges. Baltimore, MD: Johns Hopkins University Press. Warford, Jeremy J. (1966) "Water Requirements: The Investment Decision in the Water Supply Industry". Manchester School, Vol. 34, pp. 87-112. Reprinted (with revisions) as Ch. 6 in Turvey (1968). 76 Warford, Jeremy J. (1987) "Environment, Growth and Development". Paper No. 14 for the Development Committee of the World Bank and IMF, August. World Bank, Washington DC 20433. Warford, Jeremy J. (1989a) "Environmental Management and Economic Policy in Developing Countries". Ch. 2 in Schramm and Warford (1989). Warford, Jeremy J. (1989b) "World Bank Support for the Environment: A Progress Report". Staff Report No. 22 to the Development Committee, World Bank. Washington DC 20433. Warford, Jeremy J., A. Pellegrini, Allen V. Kneese and K. G. Maler (1974) "Finland's Water Pollution Control Programs". World Bank Public Utility Department Research Paper. Wiersum, K. F. (editor) (1984) Strategies and Designs for Afforestation, Reforestation, and Tree Planting. Wageningen, Netherlands: Centre for Agricultural Publishing and Documentation (PUDOC). World Bank (1978) Forestry. A World Bank Sector Policy Paper. Washington DC. World Bank (1988) Staff Appraisal Report of the Lagos Water Supply Project. 6375-UNI Washington DC. Restricted. World Bank (1989) "Greenhouse Gases and the Potential for Global Warming: Energy Options for the Developing World and ESMAP's Role". A background paper for discussion at the Energy Strategy and Management Assistance Program, 27-29 November, Paris. World Bank Energy Department. Also: "Energy Efficiency Strategy for Developing Countries". Same source and meeting. World Resources Institute, World Bank and United Nations Development Programme (1985). Tropical Forests: A Call for Action. International Task Force Report. Washington DC: World Resources Institute. 77