Policy control options for comparative air pollution study in urban areas

THE WORLD BANK FILE COPY POLICY PLANNING AND RESEARCH STAFF Environment Deparnment Policy Control Options for Comparative Air Pollution Study in Urban Areas Robert D. Hamrin April 1990 Environment Working Paper No. 28 FILECOPY 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 actg on their behalf. Robert D. Hamrin is a Principal Consultant with Robert D. Nathan Associates based in Washington, D.C. The present paper was written in conjunction with the diagnostic case study of the air pollution situation in Mexico city which was carried out by the World Bank and the World Health Organization team led by lona Sebastian, Economist, in the Environment Department of the World Bank. 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 affiliased 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 delay, 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 This paper reviews alternative policy controls to internalize the costs associated with environmental externalities and of the methods used to evaluate policy control options. The approach of the paper is strictly non-academic, operations-oriented. Specifically, command-and-control regulations, and market-based approaches: marketable permits, emission charges, and demand side management are considered. There is an analysis of the properties of each form of control, the necessary institutional and market conditions for its implementation, and the results from places where it has been implemented. The paper shows that market-based approaches have proven both the inherent features and the practical results to indicate that they will be increasingly employed as a supplement to, or in some cases, as a substitution for command and control regulations. The principal advantage of market-t-ased approaches is that through greater flexibility and by using direct or indirect price or cost signals on pollution, as well as demand management, individual decision makers can determine what is best for their own circumstances. The individual responses to the incentives, taken together, result in reducing pollution at the lowest possible control cost. A number of weaknesses in marketable permit systems are highlighted, including the potential for monopoly elements in the permit market and a number of critical unanswered problems regarding the establishment of markets in pollution rights. The paper concludes by examining emission charges. These are fees levied by the government on the absolute or relative amount of pollution emitted by an industrial facility. There is an examination of the three major strengths as well as the major weaknesses of this approach which center on both practical and political problems. POLICY CONTROL OPTIONS Coparative Air Pollution Study in Urban Areas TABLE OF CONTENTS LINKAGES BETWEEN THE ENVIRONMENT AND THE ECONONT......................1 The Environment as the Foundation for Economic Growth........... Integration of Environmental Management and Economic Policy.....2 ALTERNATIVE POLICY CONTROLS TO INTERNALIZE COSTS ASSOCIATED WITH ENVIRONMENTAL EXTERNALITIES...........................5 Command and Control.............................................5 Major Features..............................................5 Major Strengths and Weaknesses..............................6 Implementation..............................................7 Requirements.............................................7 Results7............ .................................... 7 Market-Based Approaches.........................................8 Marketable Permits..........................................9 Basic Features...........................................9 Major Strengths and Weaknesses..........................11 Implementation..........................................12 Requirements.........................................12 Results..............................................15 Emission Charges...........................................17 Basic Features..........................................17 Major Strengths and Weaknesses..........................19 Implementation..........................................21 Requirements.........................................21 Results..............................................24 A Comparative Evaluation of Marketable Permits and Charges.25 Demand-Side Management.....................................27 Basic Features..........................................27 Major Strengths and Weaknesses..........................27 Implementation..........................................28 Requirements.........................................29 Results..............................................29 BIBLIOGRAPHY ........................................................31 LINKAGES BETWEN.THE ENVIRONMENT AND THE ECONOMY The Environment as the Foundation for Economic Growth Environmentalists have helped to make the expression "everything is connected to everything else" both commonplace and universally accepted. Yet it is only in recent years that economists and environmentalists have begun to recognize the profound influence that the economy and the environment have on each other. Even then, the influence has often been viewed as a negative one, with a prevailing belief that the pursuit of environmental quality (through regulation) of necessity adversely affects economic performance. In short, a zero-sum, trade-off mentality has prevailed. This mutual neglect and suspicion are ironic in light of the fact that the words "economics" and "ecology" share the same root, "eco" derived from the Greek word for a house or home. Ecology is interpreted as the study of the natural mechanisms of resource dynamics (or nature's housekeeping), and economics as the human process of managing resources (or human housekeeping) which constraints economic growth and environmental quality are not only mutually compatible and complementary in many ways. Environmental objectives can be attained by applying sound economic principles of cost internalization, provision of public goods, and correction for market failures. Environmental protection based on such principles can improve economic efficiency and increase economic welfare while preserving the resource base on which growth in productive capacity inevitably depends. Also, the environmentalism implicit in finding new and better ways to use and conserve energy is necessary for future economic growth. The economic growth achievable through the production and installation of new non-polluting technologies is key to environmentalism. Economic growth's ultimate objective is to attain an acceptable quality-of-life which has environmental quality as one component. In short, environmental values are economic values. The respected economist Alfred Kahn has noted: "Environmental values are economic values; it is in principle just as important, in the interest of economic efficiency and therefore economic welfare, to conserve our limited natural resources, to make wise and sparing use of our limited clean air, water, and living space, as it is to economize in the use of labor and capital; and using some of our limited economic resources to preserve or restore an acceptable environment is just as much a contribution to economic welfare as devoting them to travel, shelter, or national defense." As a result of aggressive policies to achieve high levels of economic development, we have learned the lesson that it is less the fact of growth than the manner of growth and the use made of it that have caused environmental problems over the years. - 2 - A serious shortcoming of economic policy making is the consistent failure to recognize the importance to economic performance of natural resources and the quality of the environment. For instance, a U.S. National Academy of Sciences report found that "emissions of sulfur and nitrogen oxides at current or accelerated rates will be extremely risky from a long-term economic standpoint as well as from the standpoint of biosphere protection". Yet government leaders often dismiss environmental and natural resource issues as irrelevant to the "hard" issues of economic growth, inflation, employment, and productivity. This perspective is faulty. A major challenge for policy makers and economists in the 1990s is to recognize, understand, and act on three fundamental principles: - Natural resources and the environment constitute the ultimate foundation upon which all future economic activity must be constructed. - Future economic progress will depend on the sustained integrity of the resource and environmental base. - Biological capital is as important as conventional capital for achieving long-run economic growth. Integration of EnvironMental Management and Economic Policy Putting these principles into practice means that economic policy and environmental management must be integrated. The traditional approach to environmental integration management is to invest in projects which have primarily environmental objectives, such as reforestation or sewerage schemes, or to ensure that components of other projects contain elements designed to mitigate adverse environmental impacts. This essentially project-by-project approach is important, and it must be continued. Alone, however, it is clearly inadequate, and needs to be supplemented by more powerful, wide-ranging policies. By concentrating on curative, piecemeal solutions rather than on the underlying causes, the traditional approach fails to confront the real issues, which have to do much more with the way society works, and less with the technical aspects of natural resource degradation. Environmentally related behavior and our policy toward it is in fact at the very heart of social, macroeconomic and sector policies relating to agriculture, energy, and industry; domestic and foreign investment; fiscal, monetary, and trade policy; income distribution; and regional planning. It appears therefore that the project-by-project approach should be supplemented by one which integrates environmental and natural resource management directly into economic and social policy. This can be done in two ways: - Through the design of investment programs supporting environmental and natural resource objectives - 3 - - Through promotion of economic, social, and institutional policies and incentives that influence the environmentally related behavior of government agencies, major resource users, and the countless small-scale resource-using activities which occur throughout a nation's economy. If project and policy measures are to be viable, they should also be based upon a sound understanding not only of the physical linkages between events, but also of the equally complex economic, financial, social, and institutional linkages which parallel them. Much work needs to be done in this area: in spite of the massive literature on the physical linkages relating to natural resource degradation, relatively little attention has been given specifically to those points in the series of interrelated physical events where institutional or individual behavior plays a key role, and therefore where policy interventions might be feasible. To be quite specific, integrating environmental management and economic policy could draw on elements from the following agenda: - Assisted by new technologies, assess the existing natural resource base, trends and patterns in resource utilization, and prospects for the future under various scenarios of economic growth, by major sector. - Identify investment programs and areas where policy interventions that will have a wide impact need to be introduced. - Design and introduce more complex interventions, calling for incentives price, tax, and subsidy policies which have an important, but often indirect impact on resource use, and which address externalities and the "commons" problem. - Continue efforts to address major underlying causes, not only of natural resource degradation, but of development problems generally, including income and land distribution, population growth, education and the status of women, and institutional reform. - Eliminate those government policies (subsidies and competition restricting practices) which are clearly perverse, not only in narrow economic terms, but also in terms of their direct environmental impact. The last item is of particular importance. All too often, economic policies encourage activities with high, externalized environmental costs, causing them to expand beyond the scale that would otherwise be optimal were costs made explicit. Consider the example of energy price controls or consumer subsidies. Such policies cause increased energy use. This leads to both great environmental damage and substantial economic losses, because the marginal benefits of additional energy use are less than the incremental costs of supply. Removing these policy distortions would lead private decision makers to reduce energy use, with large savings in conventionally measured economic costs and even larger savings in external environmental costs. While still not sufficient, policy changes of this kind are obviously important first steps toward better resource and environmental management. Correcting policies that generate perverse incentives has high priority for three rtasons. First, because such policy changes typically promote resource conservation and other economic objectives as well, their net social gains are large. They do not require trade-offs between development and "environmental" goals, and they do not create the impression of a false dichotomy between the two. For this reason, they command broad support. Indeed, many of these changes have been on the conventional agenda of policy reform all along and have been adopted to varying degrees in many countries. Second, unless underlying perverse incentives are removed, project investments that seek to enhance and protect natural resources will have little chance of overall success. Their remedial effect will be overwhelmed by the general pattern of unsustainable resource exploitation. Third, by improving market performance, complementary policy measures meant to internalize the costs of environmental degradation have a better chance to yield predictable results. Many of the opportunities for complementary policy reforms also imply reduced fiscal burdens on government and reduced inequities within the economy. The reason is that persistent distortions in economic policies usually involve the use of the machinery of government to transfer resources from the unorganized majority to an influential minority, through manipulation of market forces and capture of the fiscal powers of the state. Consequently, reform often threatens entrenched political interests, but offers large potential rewards in improved economic productivity, equity, fiscal stability, and resource conservation. In addition to substance, as contained in the agenda for environmental action, the critical issue of procedure must be addressed. In most government structures, the effective decision-making power is concentrated at the top, in a few central agencies of government. Beyond central agencies, the government is also organized along the lines of "productive sectors." Like the central agencies, the sectoral agencies develop advice and make decisions daily that are critical to environment. It is vital that these decision makers become aware of and sensitive to the environmental consequences of their decisions. The proposed Comparative Air Pollution Study is designed to demonstrate that environment and development are the responsibility of a range of agencies, public and private, from the central economic and financial institutions of government to their agriculture, energy, - 5 - transport, trade, and other agencies, whose policies and investments have a significant impact on the issues raised, are the source of many of the critical problems. ALTERNATIVE POLICY CONTROLS TO INTERNALIZE COSTS ASSOCIATED WITH ENVIRONMENTAL EXTERNALITIES Environmental control strategies per force or by design, internalize the costs associated with environmental externalities. There are two principal ways of achieving this: command and control and market- based policies. The most widely employed approach is command-and-control, which involves direct regulation of the quantity of pollution allowed by individual sources of the control technology that sources must use. A second major approach, gaining popularity in recent years, goes under the general rubric of market-based approaches. There are two major forms of market-based approaches (also reAerred to as economic incentives): tradeable discharge permits and emission charges. A third approach to pollution control is demand-side management. Demand-reducing strategies could reduce power plant and industrial emissions, lower the overall cost of air pollution controls, and mitigate the adverse effects of controls on electricity ratepayers. Electricity demand-side management consists mostly of utility-initiated strategies designed to alter a utility's daily, seasonal, and annual "load" to increase the efficiency of energy use and reduce the amount of coal needed to generate electricity. Efficiency standards for appliances, building insulation, and new construction codes are enforced in the industrialized countries and some developing ones in order to lower demand for household energy consumption. Transport demand-side management consists of strategies designed by the Public Transport Authorities and the municipalities to reduce the daily average vehicle miles travelled and reduce the amount of transport fuel burned to generate motor power. This section will examine, for each of these major approaches, their properties, the necessary institutional and market conditions for their implementation, and the results where they have been implemented. Command and Control Major Features The command-and-control approach to controlling air pollution involves a complex mix of requirements. Typical components of this approach include (1) health- or ecology-based ambient quality standards that are not to be violated, (2) technology-based or performance-based standards for discharges to the environment, and (3) permit approval and enforcement procedures for facilities. Liability assignment and fines are also features for obtaining measurable results. The targets of this approach are ambient standards, which establish the highest allowable concentration of the pollutant in the ambient air for each conventional pollutant. To reach these targets, emission standards (legal emission ceilings) are imposed on a large number of specific emission points such as stacks, vents, or storage tanks. Following a survey of the technological options of control, the control authority selects a favored technology and calculates the amount of emission reduction achievable by that technology as the basis for setting the emission standard. Technologies yielding larger amounts of control (and, hence, supporting more stringent emission standards) are selected for emitters in areas where it is very difficult to meet the ambicnt standard and for new emitters. In the United States, the Clean Air Act establishes technology- based or performance-based standards for emissions for new sources, requires Federal ambient air quality standards which states develop plans to meet, and creates elaborate permitting processes that can result in differential requirements depending on whether the source is new or old and whether it is located (or would be built) in an area complying with or violating ambient standards. MaJor Strengths and Weaknesses Perhaps the major advantage of the command-and-control approach is that it provides regulators with a reasonable degree of predictability in how much pollution levels will be reduced. Also, the approach is relatively equitable in the sense that everyone, i.e., all firms employing a particular production process, must use the same approach to managing emissions. This advantage begins to dissolve, however, when the mandated central technology is subject to economies of scale. In this case, relatively larger manufacturers will benefit from reduced pollution control costs per unit. The equity advantage can also be reduced if the regulations are implemented differently across industry segments or by age of plant. A weakness of the approach is its relatively high cost, due in turn to its inherent inefficiencies. The grossest inefficiency is that uniform regulations or across-the-board standards fail to allocate the assimilative capacity resource to those who could make best use of it. Because everyone is expected to reduce their discharges equally, no account is taken of those producers stressing activities which could reduce their pollution at a low cost, or those producers which have already implemented some type of abatement system. The allocation of waste disposal capacity is accomplished by general administration rather than by specific process. Another serious weakness is a set of implementation problems, particularly the need for massive amounts of information (see following section on implementation). Also, there are many emerging problems for which traditional command-and-control approaches may not be practical solutions at an acceptable cost. In particular, combustion byproducts and other industrial chemicals affecting global climate change may be the - 7 - ultimate pervasive environmental challenge for which application of treatment technologies would inevitably be insufficient, regardless of the scale of investment and deployment. Overall, the weaknesses can be summarized as follows: (1) high information costs for implementation, (2) encourages strategies of avoidance, (3) generally high fixed-cost outlays, (4) no account for economies of scale, (5) low flexibility once investments made, and (6) no incentive for further innovation. Implementation Requirements The command-and-control approach requires massive amounts of information. The regulatory agency must possess detailed information concerning production processes and the suitability of pollution control devices. In the case of a diverse set of industries, it will be extremely expensive and time consuming for the agency to generate the necessary expertise. A further problem develops if the regulatory agency must rely upon firms to provide the detailed engineering data for its regulatory determinations. Firms, recognizing the use for such information, would have incentives both to withhold such data and to inflate the costs involved. There is also a greater tendency to litigate under such policies. Further, once firms have made the necessary investments in the required pollution control technology, it is very difficult for the regulating agency to adjust its policy. Thus, flexibility is low for both the firms and the regulatory agency. As a consequence, there is little incentive to innovate and reduce cost. The result is that this class of policy options tends to be the most expensive, relying heavily upon capital expenditures, but, at the same time, the most politically acceptable. Results The command-and-control approach as practiced in the United States since the early 1970s has scored notable successes, both in emissions levels and in quality. Fro- 1975 to 1985, emissions of both lead and particulates decreased subt antially, by 96 percent and 36 percent respectively. Only aggregate nitrogen oxide emissions rose slightly. Emissions of volatile organic compounds, carbon monoxide, and sulfur oxides decreased by 29, 25, and 15 percent respectively. As far as quality, the measured average levels of the six criteria air pollutants decreased over the 10-year period from 1977 to 1986, but by different amounts. Concentrations of lead decreased by 87 percent, concentrations of nitrogen oxides by 14 percent, and concentrations of sulfur dioxide, carbon monoxide, particulate matter, and ozone by 37 percent, 32 percent, 23 percent, and 21 percent, respectively. For all the pollutants except lead, the percentage decrease over the second half of the 10-year period was markedly less than over the first half, even though the level of control stringency was increasing overall. - 8 - Another good result in the U.S. context is that the system has generally protected the competitive positions of competing facilities by requiring uniform abatement technology across the country for new facilities. For existing plants emitting air pollution, requirements are related to emission control capabilities of similar facilities, depending on imperatives imposed by ambient conditions. One disadvantage of placing strictest controls on new plants is that it has increased the incentive to keep old facilities in operation longer, i.e., a new source bias. The major negative result is that command and control as practiced in the United States has been unable to achieve several mandatory standards and deadlines established in the legislation. Nonattainment of ozone and carbon monoxide ambient air quality standards in dozens of urban areas is an example. The other major negative result is the cost of pollution control. Total expenditures on air pollution control in 1987 were $18 billion. This is roughly double, perhaps more, what the costs would be if controls could be tailored to individual plant costs. Market-Based Approaches Market-based approaches directly address this weakness. By using direct or indirect price or cost signals on pollution, individual decision makers can determine what is best for their own circumstances. While one may prefer to pay the governmentally imposed "market costs" and continue to pollute, others may find it cheaper to modify their activities in ways that eliminate or reduce pollution, thus avoiding or minimizing the external costs associated with the pollution borne by society. The total of individual responses to the incentives results in a reduction in pollution which is achieved at the lowest possible control cost, as polluters undertake abatement as long as the marginal cost of abatement continues to be lower than the additional cost of polluting as determined by the system of government fees and charges. The higher the governmentally set price of pollution, the greater the reduction in pollution, assuming adequate institutional support, monitoring, and enforcement. Whatever the market- based approach is utilized, the polluter is internalizing the external cost of pollution. The term "market-based approaches" does not imply markets in the form of daily price quotes and frequent exchanges of products or property. Rather, proposed approaches would act to increase the cost of polluting products and services to a sufficient level to cause businesses or consumers to reduce or eliminate pollution without direct regulatory involvement. In essence, the capacity of environmental media to absorb some types of wastes would be given an explicit price, thus rationing this absorptive capacity in a way similar to the way markets ration any scarce good or service. In general, the decentralized flexibility of economic incentive systems gives them several important advantages over command-and-control approaches: 1. Such systems achieve large cost savings by giving firms with relatively low control costs an incentive to control above the level mandated by uniform regulation, while allowing firms with high costs to control less. 2. Government regulators do not need to acquire the detailed information needed to determine the feasible and appropriate level of control for each plant or product. 3. Flexibility in control technologies is encouraged. 4. No penalty is imposed on new products and plants. 5. Enterprises are given an ongoing incentive to devise new products or production technologies to reduce still further the amount of pollution they produce. 6. Democratic accountability for environmental policy decisions is enhanced. 7. Such systems can provide government with an appropriate and important new source of revenue. There are two basic forms of market-based approaches: marketable permits and emission charges. In the following sections, the basic features, major strengths and weaknesses, and the implementation requirements and results of each form will be examined. Marketable Permits Basic Features An emissions-based pollution rigbs market requires that the regulatory body determine the desired level of total regional emissions. Marketable emission permits, equalling in number the desired total level of emissions, would be created and distributed. Firms are allowed to trade permits. Polluters with high marginal abatement costs would buy the right to pollute from those with lower marginal costs. Regulators need know nothing about the polluters' marginal cost of abatement functions. In equilibrium, the marginal cost of pollution removal would equilibrate across firms and would equal the price of a pollution permit. A competitive equilibrium in this market minimizes the total cost of abatement for the given level of emissions. How tradeable permits are initially allocated among firms has a major impact on the financial costs to firms and also the probable political acceptability. Many proponents call for auctioning discharge permits to industry, since that also generates revenue for the government. A counter proposal is to distribute the permits at no ecst, so that companies' existing but implicit property rights to dispose of wastes are recognized. - 10 - One method for establishing a tradeable permit system is for the regional authority to auction the quota of allowable permits. Firms bidding for the pollution rights will establish a price for the permits and determine how much abatement is undertaken by each firm. Revenues from the auction go to the government, but some proponents argue for rebates and subsidies to offset the financial impact on vulnerable parts of the industry. The other method for initial allocation of permits is to distribute them at no cost in proportion to each firm's current pollution rate. Since firms have already undertaken abatement activities, distribution in proportion to some other defendable indicator such as output or pretreatment pollution levels is probably necessary. If the total pollution allowed under the permits is less than firms generate, permits will command a price and trading between firms will allocate pollution abatement through market forces. The counter argument to this approach is that it makes entry difficult for new firms into the market. Marketable permit systems in the United States have taken one of two basic forms in the air pollution arena. The most all-encompassing, and the one with the longest track record, is the emission trading program. A short-lived, specific application was lead trading. The basic features of emissions trading follow. The. results of each program, and the basic features of lead trading, a-e highlighted at the end of this section. Emissions tradinf, is a systei. composed of a basic unit of currency and a set of policies governing how this currency can be stored and spent. The basic unit of currency is the emission reduction credit (ERC). Credits are created when polluticn sources reduce their emissions below the levels allowed by their permits. These reductions can be achieved in a variety of ways burning cleaner fuel, installing new control equipment, or even shutting down a source altogether. Once created, credits can be used elsewhere in a firm or sold to another firm. Firms that buy credits effectively buy the right to produce more emissions than their permits would otherwise allow. If one source can create credits for less than it would cost another source to reduce its emissions, then the two sources can both benefit by arranging a trade. Four policies govern how the ERC can be stored and spent: emissions netting, offsets, bubbles, and emissions banking. Emissions netting, which began in 1974, allows a firm creating a new emissions source within a plant to reduce emissions from another source in the plant so that net emissions do not increase significantly. The firm can thus avoid the stringent emissions limits which would otherwise apply to the new source. A firm using netting is only allowed to obtain the necessary emissions reduction credits from its own sources. This is called internal trading because the transaction involves just one firm. Offsets have been used for new emissions sources since 1976 in areas where air quality goals have not been met. The offset rule specifies that new sources may be located in these "nonattainment areas," but only if - 11 - they "offset" their new emissions by reducing emissions from eAisting sources by even larger amounts. The offsets can be obtained through internal trading, as can netting. However, they can also be obtained from another firm's sources, which is called external trading. The offset rule was developed by the EPA as an alternative to banning construction of new sources altogether in nonattainment areas. Bubbles were first allowed in 1979. A bubble allows a firm to sum the emissions limits for individual sources of a pollutant within a plant, and to adjust the levels of control applied to different sources as long as this aggregate limit is not exceeded. Bubbles enable firms to use their knowledge of pollution control costs to achieve efficiency gains without exceeding the overall emissions limits imposed by regulators. While the trading concept for bubbles is similar to that for netting and offsets, bubbles apply only to existing sources. In theory, a bubble can be used for more than one plant or firm, so that external trading could be involved. In practice, bubbles are nearly always used for single plants, so only internal trading occurs. Emissions banking was developed in 1979 in conjunction with the bubble policy. Banking allows firms to save emissions reduction credits for future use in emissions trading. Major Strengths and Weaknesses Marketable permits have four major strengths: (1) they hold the promise of achieving objectives which cannot be met otherwise; (2) whatever level of reductions are achieved, they are likely to come at lower cost; (3) the system has great flexibility over time; and (4) polluters would have increased incentives to invest in emission-reducing technologies during periods of regional growth, precisely when pollution is of most concern. Marketable permits also have a number of weaknesses. One is the potential for monopoly elements in the permit market. Large firms could exercise undue control over markets, hurting the position of financially less powerful competitors. Hoarding could be an outcome in some circumstances, particularly where there is major uncertainty about future conditions governing the permits or where local market power could affect competition and entry. The problem of the social versus the private cost of abatement is also implicit in this system, depending on how permits aro Allocated initially. If permits are obtained initially through bidding, then polluters pay both for abatement and for discharge permits. If permits are distributed to existing polluters at no cost, the additional financial burden on companies is avoided. How permits would be equitably distributed, given that firms have already undertaken varying degrees of abatement, is an open and difficult question. As with most regulatory systems, the outcomes may be affected by political maneuvering, which diminishes the role of market incentives. - 12 - The dynamic nature of environmental problems and regulations could also create trouble for a tradeable permit system. If companies must make long-term investments in pollution reduction, fear of later tightening of standards or technology requirements could be an impediment to development of permit trades to begin with. There are also a number of critical unanswered problems regarding markets in pollution rights. They include the specific form of the property right itself, the method of property rights distribution, the determination of the optimal size of trading areas, the number of rights to be issued, and the likelihood that competitive markets would be sustained. Finally, the application of marketable permits in specific situations raises important concerns: - Compliance costs could be greater for industry in some cases than under the current regulatory system, even though abatement expenditures may be reduced through incentive approaches. This is especially true where auctions of pollution permits would be involved. Where tradeable permits are grandfathered, lowering pollution abatement costs of marginal suppliers through trading could also affect industry-wide market prices and thus profits. Consequently, some members of concentrated industries may oppose systems that lower costs, even if the consumer and general public are better off. - Incentive systems because they allow greater latitude in how and where reductions occur could sharply increase the need for detailed monitoring and enforcement, at either government or industry expense. - Where environmental deterioration is severe, the relative advantages of incentive approaches may be small. In these circumstances, every feasible means of technical control may be needed to achieve environmental standards, if major lifestyle changes are to be avoided. - Use of market-based approaches assumes that markets will not have major imperfections. Too few buyers and sellers, hoarding, lack of adequate information, and uncertainty about security of rights are all concerns that surround trading proposals in the environmental area. Implementaion Reauirements Three major steps will be involved in the implementation of a marketable permit system. First, an accurate emissions baseline must be established to form the basis for initial permits. The simplest approach will be to use existing emissions for firms which are complying with all current requirements, and to use current legal limits for firms which are - 13 - not in compliance. States should be given the flexibility, however, to deal with local situations in ways that avoid punishing firms that have aggressively reduced emissions, and any credits established under existing programs should be recognized. Second, once the baseline permit levels are established, states should be required to institute systems for evaluating, carrying out, recording, and monitoring exchanges and sales of permits among firms. Again, local discretion will be necessary. Third, the emissions level permitted will not constitute a "right" to pollute. State implementation plans for nonattainment areas should be based on systematic scheduled reductions of aggregate emissions levels permitted. That is, once the inventory is complete, a schedule should be issued whereby levels permitted will be reduced by a fixed percentage every year until attainment is achieved. A plant which currently (legally) emits 10 tons of hydrocarbons will begin with 10 permits. But it will know that those 10 permits will become, for example, 8 next year, 6 in 1991, and 5 in 1992. Since the issues of establishing the appropriate baseline and monitoring are so critical, it is useful to examine them in further detail. The baseline definition issue gets right to the heart of the politics involved in marketable permits, for it deals with defining and distributing property rights. Indeed, the politics of emissions trading can best be understood in terms of a struggle over the nature and distribution of property rights. The struggle concerns not only measurable outputs, such as costs and environment quality, but also underlying values. There has been a continuing struggle over the quantity of rights to which firms are entitled. The core of the problem hinges on determining the appropriate baseline to use for allocating emissions rights. Measuring actual emissions is technically difficult, legally contentious, and politically sensitive. Ambiguities about baselines have contributed to uncertainty about the allocation of property rights and the likelihood that regulators will approve proposed trades. Moreover, the absence of well-defined emissions rights helps to explain why trading has been predominantly internal. Obtaining emissions credits from another firm involves significant uncertainty and search costs. These uncertainties are greatly exacerbated if the amount of emissions rights each firm is entitled to, and the potential uses of these rights, are not well specified. If emission rights were better defined, more external trading would take place. Also, external trading would become more attractive to firms if there were fewer restrictions on the use of banked rights. Monitoring of compliance is a major challenge in the implementation of marketable permits. Also, the focus on performance rather than on which hardware a company has installed has increased the difficulties of enforcement. Systems of monitoring are both necessary and likely to be more complex and expensive for a marketable permit system than for a - 14 - performance- or technology-based approach. With the latter, regulators know approximately what abatement levels are being achieved and where. To actually measure pollution levels from specific sources in order to impose fees requires regular monitoring or self-reporting with periodic verification. For air emissions, for example, most sources do not have continuous emissions monitoring, nor are all potentially controlled pollutants monitored. Relatedly, information and exchange mechanisms (public or private) would need to evolve quickly to help bring interested parties together. In many situations, responsible agencies would have to be able to monitor market development and to deal with possible problems of market dominance by very large organizations or other serious noncompetitive actions. It is clear that implementing a marketable permit system will require that regulators change the way they think about their jobs. No longer will regulators be in the business of evaluating different pollution control technologies and strategies. Firms will do that for themselves, driven by the price of continued pollution. What regulators will have to do is manage the permit system. They will need to keep track of each source's current permit level, which will require monitoring all emissions trading, and they will have to review proposed sales of permits to insure that emissions to be increased are environmentally comparable to those being reduced. Regulators may at first feel that they have less control over the system, because actual pollution control decisions will be made by polluters, not by the government. This, of course, is the whole point of the marketable permit approach, and the system will be effective only if this decentralization of decision making is allowed to work. Regulators must resist temptations to restrict the ways in which emissions are reduced to produce tradeable permits. For example, a utility which reduces loads through a conservation program must be permitted to sell the resulting pollution reductions, as would a utility that switched to cleaner fuel or installed a new control technology. One particularly knotty implementation issue for regulators is the task of defining permits in different ways for different pollutants. Permits designed to control pollutants which are mixed uniformly in the atmosphere (such as volatile organic compounds, one type of precursor for ozone formation) can be defined simply in terms of a rate of emissions flow per unit time. Permits sharing this design characteristic are called emission permits. Permits to control stock pollutants (such as chlorofluorocarbons) which build up in the environment over time are defined in terms of cumulative emissions, rather than an emissions flow. They are typically called cumulative emissions permits. The former is a rather standard flow permit, while the latter is a rather standard stock permit. Permit design is somewhat more difficult when the pollution target being pursued is defined in terms of concentrations measured at a number of specific receptor locations (such as sulfur oxides or particulates). In this case it is necessary for the cost-effective permit design to take into account the location of the emissions (including injection height) as well - 15 - as the amount of emissions. As long as the control authorities can define for each emitter a vector of transfer coefficients, which translate the effect of a unit increase of emissions by that emitter into an increase in concentration at each of the receptors, receptor-specific permits can be defined which will allocate the responsibility cost-effectively. The permit design which is consistent with cost-effectiveness in this context is called an ambient permit. Unfortunately, although the design of the ambient permit is not very complicated, implementing the markets within which these permits would be traded is rather complicated. In particular, for each unit of planned emissions, an emitter would have to acquire separate permits for each receptor affected. When the number of receptors is large, the result is a rather complicated market. Results Uplike netting and bubbles, offsets result in no direct emission control cost savings because the use of offsets does not allow a firm to avoid any em ssion limits. However, since a firm using offsets is allowed to locate major new emission sources in nonattainment areas, presumably there is some economic advantage to the firm; if there were not, it would locate in an attainment area where offsets are not required. The willingness of firms to go to the expense of obtaining offsets indicates that they derive some net gain from doing so, but the extent of this gain cannot be estimated. According to EPA data, approximately 1,500 sources used offsets between 1977 and 1980; between 1980 and 1986, offsets were used by approximately 500 sources. Cost savings from banking also cannot be estimated, but must be small given the number of transactions that have occurred. The second major result to look at is the impact on environmental quality. For each of the four elements of emission trading, the effects on environmental quality have been, on the whole, insignificbnt. While there have been some small emissions increases from individual sources involved in netting tiansactions, the overall effect has been inconsequential. Offsets, which require trading ratios greater than 1:1, will naturally lead to reduced emissions. For bubbles, the lack of systematic data collection leaves the question of effects on environmental quality unresolved, but early reports indicated that aggregate effects may be slightly positive. Emission credits for a few of these transactions have been created by lowering permitted emission levels, but not making any actual reduction in emissions. Such transactions ha%re an adverse environmental effect in the sense that emission reductions that would otherwise have been required were foregone. However, their aggregate effect on air quality in local areas is thought to be inconsequential. Banking has probably had a very slight positive effect, since banked credits represent emission reductions that have not been used to offset emission increases. However, because there has been little banking activity, this effect is also very small. The performance evaluation of emissions trading activities reveals a mixed bag of accomplishments and disappointments. The program has . 16 - clearly afforded many firms flexibility in meeting emission limits. This flexibility has resulted in significant aggregate cost savings to the industry - in the billion of dollars - without significantly affecting environmental quality. However, these cost savings have been realized almost entirely from internal trading. They fall far short of the savings that could be realized if there were more external trading. Finally, the emissions trading program has had a mixed effect on enforcement. By lowering compliance costs it has made it possible for a larger number of previously noncomplying sources to attain compliance. It has also reduced the firms' incentives to seek relaxation of the emission standards and to view noncompliance as a less expensive alternative. On the other hand, by encouraging the use of new sources of control with which the control authorities are less familiar, it opens the possibility for introducing reductions which are ultimately more difficult to monitor, and for allowing trades which degrade the environment. Both have happened in the U.S. program. Shifting control to discharge points which are cheaper to control is not clearly more cost-effective in the long run if the difficulty of enforcing compliance is so high as to invite noncompliance. Lead trading, the other major marketable permit program implemented in the United States, stands in stark contrast to emissions trading. It comes by far the closest to an economist's ideal of a freely functioning market. The purpose of the lead trading program was to allow gasoline refiners greater flexibility during a period when the amount of lead in gasoline was being significantly reduced. The EPA began its regulatory program to reduce lead in gasoline in 1973. In 1982, the EPA instituted lead trading at the same time that it imposed new, lower limits on gasoline lead content. The trading program was developed, in part, in response to concerns that some refiners, especially relatively small ones, would experience difficulty in meeting the new standards and would benefit from a program providing a degree of flexibility for a period of time. In 1985, the EPA further reduced the amount of lead allowed in gasoline and specified that the lead trading program would be terminated at the beginning of 1986. Later in 1985, the EPA instituted lead banking, with banking allowed retroactively to the beginning of that year. Refiners were then allowed to trade and use banked credits until the end of 1987. Though the EPA has not collected data on the actual cost savings realized by firms as a result of lead trading, the agency estimated that banking could lead to a savings of as much as $226 million to refiners. Without other evidence, it is reasonable to estimate that the costs savings to refiners from lead rights trading and banking have amounted to hundreds of millions of dollars. The picture of the lead rights trading and banking program that emerges from a brief performance evaluation is that of a well-designed program that gave rise to a vigorous market in lead rights. Trading in this market enabled refiners to achieve large cost savings. Moreover, the introduction of trading probably enabled some refiners to make the - 17 - transition to lower lead content standards, when they would otherwise have been unable to do so and remain in business. The environmental effect of the program as planned was neutral, since it did not allow any increase in the total amount of lead added to gasoline. In short, trading and banking of lead rights is an example of a successful marketable permits program. It is important to understand several key differences between emissions trading and lead trading. In the case of lead standards, there appears to be agreement about the distribution of property rights, and the standard that defined them. Refiners had the right to put lead in gasoline at specified levels during specified time periods. Lead in gasoline was reduced to a very low level at the end of 1987. In contrast, there is great disagreement about the underlying distribution of property rights regarding emissions trading. Environmentalists continue to adhere to the symbolic goal of zero pollution. Industry believes and acts as if its current claim on the environment, without any emission reductions, represents a property right. In the case of lead trading, output could be monitored relatively easily using the existing regulatory apparatus. This was not the case for emissions trading. A new system was set up for evaluating proposed trades. This was due in part to existing weaknesses in the current system of monitoring and enforcement. It was also a result of concerns that environmentalists had expressed about the validity of such trades. This comparison suggeats that it is possible to gain important insights into the likely performance and choice of instruments by understanding the forces that led to their creation. Analyzing the underlying beliefs about property rights to pollution may be vital for the political success of the measure and for its success in terms of pure economic efficiency. Finally, the success of this market in promoting cost savings over a period in which lead was being reduced makes it important to understand why the market was successful. The lead market had two features which distinguished it from other markets in environmental credits. The first was that the amount of lead in gasoline could be monitored easily with the existing regulatory apparatus. The second was that the program was implemented after agreement had been reached about basic environmental goals. In particular, there was already widespread agreement that lead was to be phased out of gasoline. This suggests that the success in lead trading may not be transferred easily to other applications for which monitoring is a problem, or environmental goals are poorly defined. Nonetheless, the fact that this market worked well demonstrates the value of employing economic incentives for environmental regulation in certain situations. Emission Charges Basic Features An environmental charge is defined as a fee levied by the government on the absolute or relative amount of pollution emitted by an - 18 - industrial facility. An environmental charge is thus distinguished from the concept of an administrative fee, the level of which is set to reflect the costs to the government of performing an activity that directly benefits a private entity. The costs to government of an environmental program may be an appropriate target for revenues from an environmental charge, but the amount paid by any polluter is a function of emissions, not cost to the government. An environmental charge can range from a set fee on any entity that generates pollution emissions independent of the type, level, or hazards of the emissions, to a charge based on the actual environmental damages associated with the specific emissions of concern. Other potential charges that lie within this range are characterized by varying data and information requirements and are associated with different economic effects. We have defined three environmental charges schemes to provide a basis for illustrating the range of possible charges and their application to specific environmental problems: - Effluent fees based on environmental damage - General taxes on pollution - Taxes on polluting inputs The basic features of each charge are highlighted below. Under an effluent fee system, a regional authority would impose an emissions or effluent fee on discharges of each discrete pollution source in proportion to an estimate of the environmental damage caused by the discharges. The regional focus is a critical component of this approach, for this limited scope is necessary to determine damages which will vary depending on climate, ecology, population density, and so on. In this way, the social costs of the polluting behavior become part of the business costs of the source; in the parlance of economists, the external costs of pollution are internalized. General taxes on pollution are a variant of the regionally based effluent fee system. Here, pollution could be taxed as both a revenue source for general government funds and a disincentive to po7luting activities. Because the tax is on volume of emissions, there is a direct incentive to find ways to reduce or eliminate the cost. This would differ from an effluent fee system in that it would be independent of location and thus could be less complex to implement. Taxing inputs that cause pollution could be much less complicated than taxing emissions, while still providing environmentally beneficial economic incentives for substitution and technical change. Taxing polluting inputs is based on the same general logic as effluent fees. Taxes allow users to determine their own best solutions, leading to more cost-effective reduction in pollutants than across-the-board controls. - 19 - Input taxes are more likely to be proposed for pollution problems which are pervasive and hard to monitor than for situations where pollution sources are concentrated and control technologies are commercially available. Taxing of chemicals used to manufacture toxic products or of certain fuels used in energy generation would be an example. Malor Strengths and Weaknesses Emissions charges have three overall major strengths. The first is that they would provide an efficient alternative to the command-control approach, inducing firms to reduce pollution at lower cost than a technology-based approach. Relatedly they could lead firms to choose the most efficient statement methods, this minimizing the resource cost of achieving environmental goals. A second strength is that they can be good sources of revenue, irrespective of the current (and perhaps future) abatement efforts taken by firms. Several recent analyses have compared pollution charges with the economic effects of other revenue-raising methods traditionally used by government, most notably the corporate and personal income taxes. In some cases, environment charges appear to perform well, as judged by the usual public finance criteria of minimal economic distortion, low administrative cost, revenue stability and predictability, and an equitably distributed burden. Their final strength is that they achieve compensation for the unpaid costs of industrial activity typically borne by society at large. The first of these costs is fairly easy to quantify the governmental outlays devoted to the development and enforcement of environmental regulations. The second major cost arises from the emissions that are legally emitted but that still cause damage to natural resources, even if their abatement is not cost-effective. The critical issue regarding the strengths of emission charges is the potential efficiency of charges compared with command-and-control standards. Here, two concepts are most useful: external costs and uncertainty. External cost is the difference between the private cost of producing a commodity the market value of the resources devoted to its production and the total social cost of prodi-cing that good which includes the cost of any environmental damage. A charge levied on emissions, if properly specified, can allow the competitive market to "regulate" pollution in a theoretically straightforward fashion. If producers are forced to pay the full social cost associated with their output, then the incentive to minimize production costs will include the incentive to reduce pollution. Producers will then employ the most economically efficient methods of pollution abatement (for example, investing in additional control equipment, changing the input mix, or reducing output) until the costs incurred by abating another unit of pollution outweighs the charge that they must pay to emit that unit. If the charge truly captures the external cost of production, then the distortions created by external costs will be corrected. - 20 - The theoretical economic case for effluent charges requires that the regulators know the monetary value of pollution damage from each source, and that they set charges for individual plants at a level exactly equal to the marginal value of pollution damages avoided. Such a determination depends on the hazards posed by specific pollutants, the production and abatement technologies present or available to the emitting sources, and the value society places on environmental gains in the receiving media. Comprehensive information of this sort is not available in most cases. Instead, the level of environmental quality is often specified in regulations with extremely limited informat!-n about costs and benefits. In this solution, pollution charges may represent the most efficient method of attaining any level of emissions. Since the cost Lf abatement varies substantially between firms, a charge encourages those firms with the cheapest available techniques to reduce emissions more than those without such opportunities. Seen this way, an emission charge provides a method of distributing any amount of total abatement effort efficiently among different firms. It is most instructive to consider the weaknesses of emissions charges within the context of the three cases highlighted earlier. The weaknesses of effluent fees center on both practical and political problems. A basic objection is that industry will always prefer control through standards to a system of fees with the same abatement costs, because payment of the fee on remaining discharges will cost them more in total. Since the effluent fee is a transfer within society, it does promote the most efficient use of abatement resources, but it also lowers the profitability of the affected industry. In the real world of conglomerates, multiplant operations, and conflicting corporate objectives, a fee system could affect longer run goals such as competitiveness and plant location. Equally telling, there is as yet no scientific, let alone politically accepted, way to put monetary values on many of the damages caused by pollution. Thus, settiag a fee level for even the simplest pollutants, such as organic discharges from a paper mill, must be somewhat arbitrary, on a trial-and- error basis. This type of public financial experimentation is not done in other areas and would probably not be acceptable in the environmental area. Financial problems also arise in the disposition of funds. Other problems follow from the complexity of pollutior. sources, particularly toxic and hazardous substances and nonpoint sources. Developing and administering fees for numerous pollutants from single- or multi-point sources would probably not be a feasible undertaking with the resources available for monitoring and enforcement. Strong local authorities to handle the complex planning, analysis, monitoring, enforcement, litigation, and interjurisdictional negotiations may only exist in a few major metropolitan areas and a handful of states. - 21 - As with the experience of river basin authorities in France, the presence of large industries may result in "capture" by the regulated interests and modification of the effluent-fee system. Finally, drawing regional boundaries to apply the effluent f6e system poses major difficulties. While a watershed is relatively easy to define, how far up tributaries does the jurisdiction of the authority extend? For airsheds, the problem is considerably more complex, given the variability of pollution dispersion. As for general taxes on pollution, the parties affected are not likely to look with favor on a cost increase, particularly in light of past expenditures on abatement. Taxes could be economically disruptive and could affect competitiveness. Large inequities would exist since some industries and some regions have made greater progress than others in abating pollutants under the current command-and-control system. Experimenting with tax levels would probably not he easy to do, although a gradually rising rate with interim assessment on abatement progress offers a possible alternative to trial and error. The pressures, however, might be to lower such taxes, particularly in real or inflation- adjusted terms. Only conventional air and water pollutants would seem to fit the open nature of this approach. Taxes on polluting inputs could disrupt input markets and invite charges of confiscatory taxation. It is also internationally difficult for a country if it penalizes some segments of its industry without imposing similar costs on imported products in proportion to their input content. Predicting the reduction in pollution as a result of a tax on inputs might not be possible in ways precise enough to satisfy traditional policy and regulatory processes. Furthermore, not all pollutants can be tied explicitly to specific inputs or sources, which limits the scope for this approach. Implementation Requirements Two specific implementation issues are especially important: the cost of administering the programs and the burden of the charge on the pollution sources affected. Charges tend to be efficient to the extent that their administrative costs are lower per dollar collected. These costs include the costs of collecting and evaluating the information needed to assess the charge; the fixed costs of developing the fee schedule and data management system for processing and monitoring payments; and the costs of enforcement associated with ensuring collections. These costs will be a function of the specific program characteristics. Three basic alements of the design of specific charge schemes can affect administrative costs: - 22 - - The size of the charge base - The information requirements of the charge scheme - The utilization of existing transaction points Assessing the burden of the charge on the pollution sources affected is an integral part of the implementation process. Ideally, the regulators should address fully the potential industry economic impact. Such an analysis would require detailed industry-by-industry modeling that could assess how much of the change will be passed by various firms through to consumers in the form of higher prices. A rough gauge of the relative short-run burden on various industries can be constructed that compares the aggregate (air, water, and hazardous waste) charge impacts on specific industries in terms of their pre-tax profits. This measure of relative financial burden assumes that in the short run firms would pay the environmental charge out of profits. This assumption is admittedly crude and would be more true for firms that are competitively constrained than for others. Nevertheless, it would illustrate the rough impact of environmental charges. Again, it is most instructive to consider the actual implemeutation requirements that arise under the three general forms of charges. To actually implement an effluent-fee approach, several necessary institutional, political, and technical conditions would have to exist or be created. Briefly, they would include - Regional authorities encompassing natural boundaries such as water or air sheds, with the legal authority to impose and enforce taxes on polluters - Analytical methods and data to establish the monetary value of damages caused by various pollutants or a method for estimating the level of fee necessary to meet regional environmental quality standards - Adequate resources to monitor emissions from each source, or the authority to require self-monitoring with periodic verification and enforcement - Legal authority over disposition of revenues As for general taxes on pollution, the regulators would have to set the taxes higher than some polluters' incremented abatement costs. Setting the tax level could be an interactive process, since actual reductions in pollution would be hard to predict in advance. Revenues would not be returned in the form of subsidies to polluters, since that would defeat the general revenue objective. An existing local tax system could provide a ready vehicle for managing an input tax system. To have much effect, however, special conditions must hold. Unless the input represents an important cost to the - 23 - user, cost increases through an input tax must be large relative to the price of the input in order to produce much short-term reduction in use. Where input costs are a small fraction of total costs, doubling or tripling the price through an input tax is not likely to have much effect unless there are acceptably priced substitutes. If fewer polluting substitutes exist, then small increases in input price caused by a tax may be an effective incentive for substitution and innovation over the longer term. But if revenue from an input tax is rebated to those taxed, it reduces or eliminates any disincentive against pollution, as with an emissions tax. Finally, regulators would have to consider which pollutants would be the best candidates for emission charges, and also what form of charge would be best. Separate sets of such considerations apply for stationary sources and for mobile sources. Stationary sources of sulfur dioxide (SO2) and nitrogen oxides (NOx), several exhaust emissions from mobile sources, and the production of chlorofluorocarbons (CFCs) are likely candidates for environmental charges. Any charge applied to S02 or NO emissions would most reasonably be applied to permitted sources of these pollutants, and might be based on permitted or estimated emission levels. For both pollutants, the burden of the charge would fall most heavily on the electric utility industry. Since mobile sources are a primary contributor to air pollution in most major cities in developing countries, it is constructive to look at the types of environmental charges that could be best applied to mobile sources. Two types of charges stand out. The first would assess a charge based on the presumed emission characteristics when vehicles are first sold. This type of charge would have manufacturers decide how much additional emission reduction technology might be incorporated into vehicles, based on likely consumer response to price increases. The second type of system would impose a charge based on periodically measured emissions and would allow vehicle owners to decide whether or not the additional maintenance or repair expense is justified compared with paying the charges. This could promote additional abatement effort throughout the lifetime of the vehicle and encourage consumers to acquire information about new or used vehicle emission rates. However, the second system would cost more to implement, even accounting for inspection and maintenance programs that may be in place. The cities in this study could also examine and modify based on their particular circumstances the permit fee system which is at the center of the system used by California's South Coast Air Quality Management District (SCAQMD). The SCAQMD, which includes the Los Angeles Basin, has required companies to pay an annual permit fee based on their emissions of air contaminants. The SCAQMD is a fee-based system; it does not receive funds from state or federal sources. The SCAQMD is supported by several fee programs, including permitting, application fees, processing fees, annual fees, and the emissions fees which are required of each facility operating under SCAQMD's jurisdiction. While other air quality control districts throughout the United States use the permit fees to . 24 - finance their activities, the SCAQMD is worthy of some analysis because of its high fee level relative to other jurisdictions, and the manner in which fees are computed. Fees in the SCAQMD are quantity-based; most other jurisdictions do not set permit fees for different sources on the basis of the quantity of emissions. Companies conducting business in the Los Angeles Basin must pay a set fee per ton of each of the following air contaminants which they emit: total organic gases (other than those specifically mentioned on this list), methylene chloride (along with 1,1,1,-trichloromethane, trifluoromethane, and chlorinated-fluorinated hydrocarbons), carbon monoxide, oxides of nitrogen (expressed as nitrogen dioxide), gaseous sulfur compounds (expressed as sulfur dioxide), and particulate matter. These fees have greatly increased since the original fees were put into place. Results Emissions charges have been applied in the United States and in a few European countries. A brief survey shows that there are a wide array of fee-based systems in place, but that most are applied to water pollution. In the United States, 40 states have legislated authority to issue permits controlling releases of air pollutants, and 32 of these 40 levy permit fees. In addition, 40 states are authorized to assess fees on the generation or disposal of hazardous waste, and 35 states are entitled by state law to charge fees for water permits. There is a wide variation in the activities subject to fees and in the bases for the fees. For example, fees are charged for permits to construct or operate facilities which discharge pollutants into the air, for bubble or banking applications, or for asbestos removal permits. France is the only country in Western Europe which uses a charge system for air emissions. In 1988, it implemented a charge on sulfur oxide emissions. The charge applies to some 400 firms that generate 50 MW or more of electricity or emit more than 2,500 tons of sulfur oxides or nitrogen oxides per year. The charge rate is ECU 19 (approximately $21) per ton and yields total revenues of ECU 13 million. As currently structured, the French air pollution charge has limited incentive effect. First, the rate is low relative to the cost of scrubbing and other technological options. Second, over 90 percent of the revenues are returned to the affected firms as subsidies for installing engineering controls. Air pollution is attacked through fuel charges in countries such as the Netherlands, Norway, Sweden, and Japan. In 1988, the Netherlands replaced a group of environmental charges (on chemical waste, industrial noise, lubricating oil, and motor fuels) with a charge system on fuels. R1venues from the new fuel charges will be substantial: over ECU 129 million ($142 million) annually. With the new fuels charge system, unleaded gasoline is taxed at ECU 0.1 per 100 liters (about $0.004 per - 25 - gallon); leaded gasoline at ECU 1.74 per 100 liters (about $0.08) per gallon); and other taxes are levied on diesel and other fuel oils, LPG, natural gas, and petroleum coke. The fuel taxes are designed to have some incentive effects; witness the differential taxation of leaded and unleaded gasoline. Revenues from the fuels taxes are used to finance the environmental program of the Ministry of the Environment. Differential taxation of leaded and unleaded gasoline is common throughout Europe. Besides in the Netherlands, differential taxation of the two fuels is practiced as follows (per gallon dollar equivalents of the differentials in parentheses): Denmark ($0.20), Finland ($0.21), German ($0.04), Norway ($0.18), Sweden ($0.09), and the United Kingdom ($0.20). It should be noted that France, Germany, and the Netherlands make extensive use of water charge systems. Also, a number of countries in Europe use charges to control hazardous wastes, using either a tax on waste outputs or a tax on feedstacks, which are usually correlated with the level of waste produced. The United States also has a diverse range of taxes imposed on hazardous waste streams. An overview of the various charge systems in place shows that there are several repeating patterns. First, the major motivation for implementing emission fees is to raise revenues, which are then usually earmarked for activities which promote environmental quality. Second, most charges are not large enough to have a dramatic impact on the behavior of polluters. In fact, they are not designed to have such an effect. They are relatively low, and not directly related to the behavior of individual firms and consumers. Third, there is a tendency for charges to increase faster over time than inflation. Presumably, starting out with a relatively low charge is a way of testing the political waters as well as of determining whether the instrument will have the desired effects. Finally, in the overwhelming majority of cases, the direct economic effect of fees appears to have been small. A Comparative Evaluation of Marketable Permits and Charges An examination of the charge and marketable permits schemes reveals that they are rarely, if ever, introduced in their textbook form. Most of these approaches were not implemented from scratch; rather, they were grafted onto regulatory systems in which permits and standards play a dominant role. Perhaps as a result of these hybrid approaches, the level of cost savings resulting from implementing charges and marketable permits is generally far below their theoretical potential. Cost savings can be defined in terms of the savings which would result from meeting a prescribed environmental objective in a less costly manner. As noted, most of the charges to date have not had a major incentive effect. It can be inferred from this that polluters have not been induced to search for a lower cost mix of meeting environmental objectives as a result of the implementation of charge schemes. Thus, it seems unlikely that charges have performed particularly well on grounds of narrow efficiency. The - 26 - experience on marketable permits is similar. Hahn and Hester (1986) argue that cost savings for emissions trading fall far short of their theoretical potential. The only apparent exception to this observation is the lead trading program, which has enjoyed very high levels of trading activity. The example of lead trading leads to an important observation: in general, different charge and marketable permit systems exhibit wide variation in their effect on economic efficiency. On the whole, there is more evidence for cost savings with marketable permits than with charges. While the charge systems and marketable permit systems rarely perform well in terms of efficiency, it is important to recognize that their performance is broadly consistent with economic theory. This observation suggests that the performance of the markets and charge systen. can be understood in terms of basic economic theory. For example, where barriers to traders are low, more trading is likely to occur. Where charges are high and more directly related to individual actions, they are more likely to affect the behavior of firms or consumers. If these instruments are to be measured by their effect on environmental quality, the results are not very impressive. In general, the direct effect of both charges and marketable permits on environmental quality appears to be neutral or slightly positive. The aggregate effect of lead trading has been neutral. The aggregate effect of emissions trading on environmental quality has probably been neutral or slightly positive. The direct effect of charges on polluter incentives has been modest, although the indirect environmental effect of spending the revenue raised by charges has been significant. The evidence on charges and marketable permits points to an intriguing conclusion about the nature of these instruments. Charges and marketable permits have played fundamentally different roles in meeting environmental objectives. Charges are used primarily to improve environmental quality by redistributing revenues. Marketable permits are used primarily to promote cost savings. Finally, what can be expected in the future regarding the further adoption of market-based approaches? Because marketable permit approaches have been shown to have a demonstrable effect on cost savings without sacrificing environmental quality, this instrument can be expected to receive more widespread use. One factor which will stimulate the application of this mechanism is the higher marginal costs of abatement that will be faced as environmental standards are tightened. A second factor which will tend to stimulate the use of both charges and marketable permits is a "demonstration effect." Several countries have already implemented these mechanisms with some encouraging results. The experience gained in implementing these tools will stimulate their use in future applications. A third factor which will affect the use of both of these approaches is the technology of monitoring and enforcement. As monitoring costs go down, the use of mechanisms such as direct charges and marketable permits can be expected to increase. The combination of these factors leads to the prediction that greater use of these market-based environmental systems will be made in the future. - 27 - Demand-side Management Basic Features All of the approaches to pollution abatement described above deal with the interface between the production process and the environment. However, one alternative not yet discussed is an adjustment at the interface between consumption and environment. The economic incentive causes the consumer's cost to reflect the higher social cost as compared to the private cost above. For example, rate structures for natural gas, municipal water, and electricity are ways in which consumer pricing could reduce some of the user stress on ecosystems. This mechanism, then, has two complementary effects. First, raising the price of a service or product (in effect, a product is really a service since consumers only benefit from the use of a commodity) decreases demand, thereby reducing the user stress on the environment. And second, the increment that the price is raised reflects the damage or cost to the environment and should therefore be used to ameliorate this damage (either by treatment or by future abatement). Otherwise stated, the overall pattern of demand is rearranged rather than lowered. Demand-side management as employed to date has been principally electricity demand-side management. It aimed at increasing the efficiency of electricity use and reducing peaks in demand offers a promising way to control emissions from power plants that burn coal and to reduce the adverse electricity rate impacts of conventional controls like scrubbers. Demand-side management strategies center on the need and opportunity to directly influence the level and timing of customer electricity demand over a day, week, season, and year through deliberate strategies. These strategies fall into three broad categories: - Local management. Reducing the peaks of highest demand during times like hot summer afternoons, when utilities must bring expensive-to-operate generating units on-line to meet the increased load caused by air conditioning. - Conservation. Increasing the overall efficiency cf electricity use in the home, office, and factory, which reduces electricity demand in peak, as well as off-peak, periods. - Cogeneration. Producing electricity and thermal energy from the same primary fuel source, which increases efficiency of fuel use and leads to a reduction in electricity demand in peak, as well as off-peak, periods. Major Strengths and Weaknesses Load management, conservation, and cogeneration offer opportunities for reducing both emissions and the overall cost of emissions controls. These demand-side strategies cut emissions in two ways. They can reduce overall energy use or peak energy use, thereby reducing the - 28 - burning of fossil fuels and the emissions that come from these fuels. They can also encourage the use of new technologies or alternative fuels that generate electricity with less emissions. Demand-side strategies also can create substantial dollar savings for utilities and their customers. These savings occur in several ways: - Fuel savings. By reducing peak and off-peak demand, utilities purchase less fossil fuel burned by peaking units. * Reserve marin savings. Maintaining peak generating capacity year round is expensive, but it must be done to assure adequate power reserves it is called the "reserve margin" in the industry to meet peaks that are solely at the mercy of the weather. Demand- side strategies give utilities more control over customer demand in peak periods and enable them to achieve a closer match between demand and supply. - Capital savings. Demand-side strategies allow utilities to slow demand growth and thereby defer the need for new power plants. Since a new coal-fired plant may cost $1,500-1,800 per kilowatt (KW), a new 1,000- MW plant costs more than a billion dollars. Avoiding these costs can free up substantial internal capital, reduce dependence on external debt, and decrease the utility's revenue requirements and need for electricity rate hikes to pay for the new plants. Demand-side strategies are not free, however, and their costs have to be considered. - Customer electricity bills. Although utility expenditures to implement demand-side strategies may lead to a rise in electricity rates, most customers who participate in demand-side programs may very well enjoy more than offsetting savings, because they will use less electricity. Demand-side strategies are not a panacea however and they have some disadvantages as emissions controls. One disadvantage is that some utilities with excess generating capacity want load growth, not reduced demand. Another is that more costly demand-reducing measures are generally not as attractive where electricity rates are relatively low, since they will not be economic. Utilities also face uncertainty in planning, administering, and evaluating demand-side programs. Good marketing and financial analysis is scarce, and it is difficult to predict customer response. Data gathered by EPRI on 80 utility programs revealed customer participation rates ranging widely from 1 percent to 90 percent, with a 29 percent median. Another problem is that demand-side strategies, inasmuch as they reduce the need for burning coal, may adversely affect coal mining jobs. - 29 - Inplementation Requirements In order for utilities to seriously consider alternative control strategies, lawmakers will have to successfully maximize incentives to make such strategies attractive. Indeed, the regulatory framework itself will be key to encouraging or discouraging such alternative control strategies. For instance, the extent to which it permits utilities to take credit for actual tonnage emissions reductions (as opposed to only emission rate reductions), or to enjoy some flexibility in tailoring controls with varying degrees of effectiveness to different boilers are important aspects of the framework. The law should also contain incentives to encourage individual utilities to form new institutional arrangements that permit these cost reduction strategies to be more implemented effectively in power pools that cross utility service territories. Such arrangements will be particularly important for conservation strategies. The encouragement of demand-side management control options will require lowering the regulatory barriers to their wider use. The most important provisions would concern what reductions are required at each source (flexibility is best), what standard is used to measure and enforce reductions (a tonnage reduction standard is better than an emissions rate limit), and what compliance timetable is required (flexibility to allow time for commercialization of promising retrofit technologies would help). Results Demand-side management strategies have been employed most extensively in the United States. Since they are relatively new, evaluation of the strategies is in its infancy. As a result, information about the effectiveness and relative costs and benefits of these strategies is scarce and difficult to assemble. Thus, only a descriptive evaluation of the programs enacted to date is possible. Load management has been practiced for a number of years, with Detroit Edison being one of the first utilities to install devices on customers' waterh6aters to directly reduce their use during peak periods of the day. As a result, the utility has avoided the high cost of operating peaking generators to supply power during such periods. Similar kinds of load control programs, some relying on direct control devices that shut off other appliances and equipment during peak periods, some relying on "time- of-use" rates that encourage customers to reduce electricity consumption during such periods, are also widespread. Load management programs are particularly popular among utilities that operate close to capacity during winter or summer peak periods. These programs reduce electricity use during such peak periods and allow utilities to postpone or cancel the construction of new power plants designed to meet peak loads. - 30 - Energy efficiency or conservation programs have also gained in popularity over the past decade. They aim to reduce electricity consumption during all periods of the day and year, not just during peak periods, by improving the thermal performance of buildings and by encouraging customers to install and use more energy-efficient appliances and equipment. Utility-sponsored conservation programs have been a response, in part, to aggressive state public utility commissions (PUCs) seeking to reduce the cost of electricity services to the ratepayer, and, in part, to national conservation legislation. Utilities also sponsor such programs to improve relations with their customers, as well as to maintain adequate reserve margins of power and to postpone construction of new power plants. Cogeneration has come mostly from industries, institutions, and independent power producers seeking to take advantage of the Public Utilities Regulatory Policies Act of 1978 (PURPA). Some companies, for instance, that require significant electricity and that process steam for manufacturing have installed boilers that produce steam for both purposes. Any electricity generated in excess of a company's needs can, under PURPA, be sold back to the utility that services its territory. Thus, cogenerators displace utility generation not only with more efficient energy production, but with their own surplus power, which utilities are required to purchase. - 31 - BIBLIOGRAPHY "A Market Approach to Air Pollution Control Could Reduce Compliance Costs Without Jeopardizing Clean Air Goals", a Report by the U.S. General Accounting Office, Washington, DC, March 23, 1982. Anderson, Robert C., et. al., "The Use of Economic Incentives Ifechanisms in Environmental Management", American Petroleum Institute, Washington, DC, June 23, 1989. Dudek, Daniel J., "Chlorofluorocarbon Policy: Choices and Consequences", Environmental Defense Fund, New York, NY, April 1987. Dudek, Daniel J. and Palmisano, John, "Emissions Trading: Why Is This Thoroughbred Hobbled?", Columbia Journal of Environmental Law, Vol. 13, No. 2, 1988. "Emission Charges", unpublished paper, Congressional Budget Office, U.S. Congress, Washington, DC, 1989. Griffith, C.R., et. al., "The Theory and Practice of Alternative Mechanisms for Environmental Quality Management: A Bibliography", CPL Bibliography No. 57, CPL Bibliographies, Chicago, IL, July 1981. "Guidelines for Performing Regulatory Impact Analysis", U.S. Environmental Protection Agency, December 1983. Hahn, Robert W., "Economic Prescriptions for Environmental Problems", Journal of Economic Perspectives. Vol. 3, No. 2, Spring 1989, pp. 95-114. Hahn, Robert W. and Hester, Gordon, L., "The Market for Bads: EPA's Experience With Emissions Trading", Regulation. Nos. 3 and 4, 1987. Hahn, Robert W. and Gordon L. Hester, "Marketable Permits: Lessons for Theory and Practice", Ecology Law Quarterly, University of California, Berkeley, Vol. 16, No. 2, 1989. Hamrin, Robert D., "The Economy and The Environment: A Healthy Partnership", Report for the Taiwan 2000 Project, Taipei, October 1986. Hamrin, Robert D., "The States' Lead in Bringing The Economy and Environment Together", paper presented at the conference New State Roles: Environment, Resources, and the Economy; The Woodlands, Texas, October 1988. - 32 - Jessup, Philip, Strategies for Reducing the.Cost of Acid Rain Controls: Electricity Demand-Site Manaement and Clean Coal Technologies. Special Report, Environment and Energy Study Institute, Washington, DC, January 1988. Levin, Michael H., "New Directions in Air Pollution Control and Environmental Policy: Incentives", paper presented at ABA Section of Natural Resources Law, March 1988. Moore, John L., et al., "Using Incentives for Environmental Protection: An Overview", unpublished paper, Congressional Research Service, U.S. Congress, June 1989. Noll, Roger G., "Implementing Marketable Emissions Permits", AEA Papers and Proceedings. May 1982. Project 88: Harnessing Market Forces to Protect Our-Environment. a Public Policy Study sponsored by Senator Timothy Wirth and Senator John Heinz, Washington, DC, December 1988. Repetto, Robert, "Economic Policy Reform for Natural Resource Conservation", draft paper, World Resources Institute, Washington, DC, September 1986. Rusin, Michael, et. al., "Managing The Environment: A Review of Present Programs and Their Goals and Methods", Discussion Paper No. 057, American Petroleum Institute, Washington, DC, February 1989. Stewart, Richard B., "Controlling Environmental Risks Through Economic Incentives", paper presented at Colloqium on New Directions in Environmental Policy, Columbia University, October 1, 1987. Tietenberg, T.H., "Marketable Permits in the United States: A Decade of Experience", paper presented at International Institute of Pub'tc Finance, Paris, August 24-29, 1987.