_S I sI bZ POLICY RESEARCH WORKING PAPER 1672 Controlling Industrial Conventional discussions of pollution regulation in Pollution developing countries have been too shallow - devoting inordinate attention to the choice of instrument while ignoring the preconditions for Shakeb Afsah applying any instrument BenoFt Laplante effectively. They have also David Wheeler been too narrow - focusing only on the interaction of state and factory. and ignoring the role of the market and the community. The World Bank Policy Research Department Environment, Infrastructure, and Agriculture Division October 1996 POLICY RESEARCH WORKING PAPER 1672 Summary findings Afsah, Laplante, and Wheeler call for a revised model for and communities need timely, accurate, public the regulation of industrial pollution. They think the information to assess factories' environmental traditional emphasis on "appropriate instruments," while performance. ultimately correct, is premature, because agencies in most * Orchestration, not dictation. Potentially high- developing countries have too many problems with leverage programs to add to the mix include community information and transaction costs to implement any environmental education, public disclosure of factory instruments comprehensively. performance ratings, and technical training programs for Once regulators have better information, more environmental personnel in polluting factories. integrated information systems, more capacity for setting * Community control. This should be a current priorities, and a stronger public mandate, it will not be reality, not a goal of future programs. Strengthening difficult for them to manage pollution more cost- central regulatory agencies should not empower them to effectively. Overhasty introduction of market-based impose uniform standards on heterogeneous instruments will not work and will probably discredit communities under the guise of "efficiency." Local those potentially powerful regulatory tools. variations in regulation are legitimate. The new model of regulation should relegate * Structured learning. Agencies should initiate pilot regulators to their proper place in the scheme of things. projects and build larger programs as lessons from the Factories' environmental performance is shaped by the pilot projects are absorbed. interaction of agents with different incentives. The state * Adaptive instruments. Newly industrializing should play a role in regulating pollution externalities, economies can experience rapid changes in ambient but the role of the community and market must also be quality across air- and watersheds. Regulation should recognized. In the authors' view, appropriate regulation focus on adaptation to these rapid changes. Regulators in developing countries should incorporate five key should be empowered to counter environmental features: degradation by tightening existing regulations, but the * Information intensity. Regulators need reliable data, system should also minimize disruption for investors. integrated information systems, and the ability to set Adjustment rules should be transparent and linked to priorities that reflect relative costs and benefits. Markets publicly available data on quality and emissions. This paper - a product of the Environment, Infrastructure, and Agriculture Division, Policy Research Department - is part of a larger effort in the department to develop more cost-effective approaches to regulation of externalities. The study was funded by the Bank's Research Support Budget under research project "The Economics of Industrial Pollution Control in Developing Countries" (RPO 680-20). Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact David Wheeler, room N10-023, telephone 202-473-3401, fax 202-522-3230, Internet address dwheelerl@worldbank.org. October 1996. (16 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An obiective of the series is to get the findings out quickly, even if tre presentations are less than fully polished. The papers carry the names of the authors and should be used and cited accordingly. The findirgs, interpretations, and conclusions are the authors' own and should not be attributed to the World Bank, its Executive Board of Directors, or any of its member countries. Produced by the Policy Research Dissemination Center CONTROLLING INDUSTRIAL POLLUTION: A NEW PARADIGM SHAKEB AFSAH BENOIT LAPLANTE DAVID WHEELER1 I Shakeb Afsah and Benoit Laplante are Consultants and David Wheeler is Principal Economist in the Environment, Infrastructure and Agriculture Division of the World Bank's Policy Research Department. This paper could not have been written without the contribution of David Witzel and numerous inputs from our colleagues in PRDEI. Our thanks to Michael Bruno, Lyn Squire and Zmarak Shalizi for useful comments. In addition, we would like to express our gratitude to our many collaborators in the pollution control agencies of Brazil, China, India, Indonesia, Mexico and Philippines. 1. Introduction Factories in developing countries exhibit great variety in environmental performance despite the widely acknowledged weaknesses of the regulatory framework. Even in the poorest countries, some plants would satisfy OECD emissions standards. Similarly, a great variety in environmental performance is observed in developed economies. These facts create a problem for conventional thinking about controlling industrial pollution. Given the weaknesses of the regulatory framework in developing countries, plants should treat the environment as a 'free' input and undertake no effort to control emissions. On the other hand, factories in societies with stronger regulatory agencies should generally be in compliance with the standards. Since neither conclusion is consistent with the facts, we must question the premises and develop a new paradigm for understanding the performance of industrial polluters. Our current research is addressing this problem in six large developing countries. By establishing partnerships with environmental agencies in those countries, we have been able to observe regulator-polluter relationships at first hand.2 Our experience as 'participant-analysts' has revealed the limitations of the conventional regulatory model, and suggested a number of significant directions for revision. This paper summarizes our findings to date, along with the potential implications for regulatory policy. First, it is clear that the basic assumptions which support the model of 'optimal regulation' -- full information and zero transactions costs -- are not met in practice. This undermines the implementation of both traditional command-and-control regulation and economic instruments. Secondly, we find that the regulator is not the sole source of pressure on plants to improve their environmental performance. Local communities and market agents also play important roles. As an alternative to the traditional view, we therefore propose a model of interactions linking four agents: plant, state, community and market. This model focuses on the process that leads to efficient levels of pollution, rather than on a priori identification of the optimum point by state regulators. To illustrate the main features of our model, we present findings from two recent studies. Our analysis of China's non-compliance fee (or levy) for water polluters highlights the effect of local conditions on the actual enforcement practices of regulatory agencies. When viewed through a non-traditional lens, practices commonly criticized as symptoms of 'inefficient administration' appear closer to optimal behavior. In a second study, we 2The six countries are: Indonesia, Mexico, Brazil, China, Philippines and India. Our agency partners are BAPEDAL (National Pollution Control Agency), Indonesia; INE (Instituto Nacional de Ecologia), Mexico; CETESB (Pollution Control Agency of Sao Paulo State), Brazil; FEEMA (Pollution Control Agency of Rio de Janeiro State), Brazil; NEPA (National Environmental Protection Agency), China, Tianjin Environmental Protection Bureau, China; and DENR (Department of Environment and Natural Resources), Philippines. We are also collaborating with the Pollution Control Boards of eight Indian states in a nationwide survey of the environmental performance of Indian factories. 2 analyze Indonesia's recently-introduced program for public disclosure of environmental performance ratings for factories. There is no room for such a program in the traditional model, but we explain why it may strongly affect polluters in a world of multiple agents and multiple incentives. Finally, we argue that the new paradigm has State important implications for regulatory policy. Regulators The Law Regulators lose their role as sole enforcers, but gain the potential for greater effectiveness through new policies which leverage the power of communities and markets. We suggest five key principles that can form the basis for *Regulatory Stasndards structuring environmental agencies and their * Market-based tnstru mets program design and implementation. Igai tiability Figure la 2. The Traditional View of Regulation Figure Ia presents the classic paradigm for analyzing pollution control issues. Here the State holds center stage, with two principal agents -- Regulators and The Law -- expected to set and enforce rules of environmental behavior. In keeping with this understanding of the problem, the policy analysis literature has focused on appropriate roles for 'ex ante' regulation (standards vs. market-based instruments) and 'ex post' liability claims by injured parties. Figure lb provides the conventional view from the perspective of environmental economics. Pollution (N) is measured on the horizontal axis and costs ($) are measured on the vertical axis. In this textbook view of the problem, the regulator can quantify the increase in Marginal Social Damage (MSD) as the pollution level rises. There is also sufficient information to quantify increases in Marginal Abatement Cost (MAC) as polluters reduce their emissions. The regulator determines 'optimum pollution' at point N*, where MSD = MAC. The regulatory problem in this world is straightforward: Having determined N* with full Traditional View information, the regulator seeks to attain it by Assumes: S Mai using command-and-control (mandating factories *Fult Infornation Abatem.et not to pollute above a determined level) or 4No Transaction market-based instruments (setting a pollution Costs charge P*, or allowing factories to trade pollution Focses on Damage permits within the limit N* ). Able to enforce at *InStruments will because transactions costs are zero, the OtinilitX regulator simply dictates the terms and the Nz N Nt N factories respond appropriately. By assumption, the central regulator is and should be the sole Figure lb 3 decision agent in such a world. As environmental economists, we support the view that optimum pollution is an appropriate concept for regulation. We also believe (and are working with our partner agencies to demonstrate) that pollution charges and tradable pollution permits can be effective regulatory instruments under the right conditions. However, our research and field experience have convinced us that the conventional regulatory approach does not pay sufficient attention to defining the right conditions. 3. Strengthening the Foundations Indeed, it would be impossible for us to defend some basic tenets of the conventional model to our agency partners. They would not know what to make of assumptions like 'full information' and 'zero transactions costs.' These are not just 'approximations' under developing-country conditions; they are dangerous chimeras which can divert attention and scarce resources from real agency problems to grandiose programs which have no chance of working. Let us be more specific: 3.1 Information and Transactions Costs Our partner agencies are plagued by problems with: * Information: Monitoring quality is frequently so poor that compliance with regulations is difficult to assess. Fragmentary data on factory emissions and ambient quality are often non-computerized, and closely held by separate agency units charged with different responsibilities. Information on abatement costs is almost never available. * Bureaucracy: The air and water quality monitoring units frequently don't talk to each other, nor do they share information with those monitoring air and water emissions. * Human and technical resources: Agencies generally have little capacity for assessing the net benefits of alternative programs and using the results to establish priorities for allocation of scarce resources. Few trained inspectors are available, and it is impossible to monitor more than a modest fraction of polluting factories. * Political support: Serious enforcement frequently encounters potent political resistance. To summarize, life in our partner agencies is one long encounter with limited information and high transactions costs. 4 3.2 First Things First Under such conditions it is extremely difficult to implement my pollution control program, including market-based instruments. Indeed it would be pointless, and ultimately counterproductive, to advocate large-scale implementation of pollution charges or tradable permits under conditions which practically guarantee their failure. Near-term policy problems are more pressing and should be addressed first: * Identification of the small group of serious polluters which the agency can regulate effectively with existing resources; * Mobilization of political and community support for meaningful action; . First-stage development of an integrated information system with good quality control; * Establishment of ambient quality targets for polluted air- and watersheds; linkage to pollution reduction measures applied to target polluters; * Use of simple cost-effectiveness principles in the reform of licensing and inspection procedures; * Development of internal capacity for priority-setting using integrated information systems; * Small-scale pilot experimentation with new regulatory instruments (charges, permits, public disclosure, etc.) If successfully implemented, these "simple" steps will lay the necessary foundations for more sophisticated pollution control strategies. Without them we are likely to witness a demoralizing series of failures, as fancy programs attempt to lift off with no launching pad. 4. Broadening the Vision We have argued above that a regulatory approach based on inappropriate assumptions about information and transactions costs has distracted policy analysts from the real implementation issues in developing-country agencies. At a more general level, we would also argue that the traditional view of regulation is misguided because its focus is too narrow. Conventional policy discussion has focused almost exclusively on interactions between the State and the Plant. However, our research has suggested powerful roles for two additional 'players': the Community and the Market. 5 4.1 The Community Recent evidence from Asia, Latin America and North America suggests that neighboring New Elements f/) communities can have a powerful influence on factories' environmental performance. Communities which are richer, better educated, + Power Plants and more organized find many ways of enforcing *Social Norms environmental norms. Where formal regulators are * Negotiations munity present, communities use the political process to influence the tightness of enforcement. Where - formal regulators are absent or ineffective, 'informal regulation' is implemented through .__ community groups or NGOs. Figure 2 The agents of informal regulation vary from country to country -- local religious institutions, social organizations, community leaders, citizens' movements or politicians - - but the pattern is similar (Figure 2): Factories negotiate directly with local communities, responding to social norms and/or explicit or implicit threats of social, political or physical sanctions if they fail to reduce the damages caused by their emissions. In countries as different as China, Brazil, Indonesia and the US, much of the variation in factories' environmental performance is explained by inter-community variation in income, education and bargaining power.3 4.2 The Market Factories operate in local, national and international markets, where many agents can New Elements (11) affect revenues and costs (Figure 3). Environmental considerations now affect the decisions of many of these agents. In both Plants * Reputation industrial and developing countries, *Profits environmentalism in the middle and upper onsu classes is a significant factor in consumer Markets decisions. With the worldwide advent of In environmental legislation, investors are also scrutinizing environmental performance. . _ Among other factors, they have to weigh the Figure 3 potential for financial losses from regulatory penalties and liability settlements. In recent years, the importance of investor interest has been increased by the growth of new stock markets and the internationalization of investment. For similar reasons, international and 3For evidence from Asia, see Pargal and Wheeler (1996), Hettige, Huq, Pargal and Wheeler (1996), Huq and Wheeler (1993), and Huq, Hartman and Wheeler (1996). Evidence from Brazil and Mexico can be found in Wheeler and Witzel (1995) and Hettige and Witzel (1996). 6 local suppliers of financing, industrial equipment, and engineering services are increasingly reluctant to do business with flagrant polluters. Recent evidence from both the OECD and developing countries suggests that environmental reputation matters for firms whose expected costs or revenues are affected by judgments of environmental performance by customers, suppliers, and stockholders.4 Many factors can affect firms' evaluation of their environmental reputation, including company size, export orientation, and multinational ownership. For reputationally- sensitive companies, public certification of good or bad performance may translate to large expected gains or losses over time. 4.3. Multiple Agents, Multiple Incentives: A New View of Regulation Once the Community and the Market are . . ~~~~The New Model: introduced, we have a much richer and Multiple Agents, Multiple Incentives more robust model for explaining the State observable variations in factories' environmental performance. Clean factories are perfectly plausible in poor countries, and the survival of dirty factories in rich countries is not hard to understand. In place of the paired State/Factory model Mar Communit of regulation, we therefore propose the 'Regulatory Triangle' which is depicted in Figure 4. . _ __ __ __ __ __ __ __ __ _._ Figure 4 Once we introduce a world of multiple agents and multiple incentives, we must also rethink the regulator's appropriate role in pollution management. No longer is this role confined to producing, monitoring and enforcing rules and standards. Instead, the regulator can gain leverage through non-traditional programs which harness the power of communities and markets. Within the 'triangular' regulatory framework, for example, there is ample room for information-oriented approaches such as voluntary participation/compliance programs5 and public disclosure of factories' environmental performance. A broader implication is that one size no longer 'fits all' for regulatory policy design: Optimal combinations of regulatory tools will depend on country-specific social, economic and institutional conditions. What does this expanded view of regulation mean in practice? To draw out some of the implications, we will summarize the results of recent collaborative projects with the National Environmental Protection Agency of China (NEPA) and the National Pollution Control Agency of Indonesia (BAPEDAL). 4See Arora and Cason (1994), Hamilton (1995), Hettige, et. al. (1995) and Laplante and Lanoie (1994). 5 See Afsah, Laplante, and Makarim (1996) for a discussion of PROKASIH, Indonesia's river management program. 7 Case 1: China's Pollution Levy China's pollution levy is one of the few economic instruments with a long, documented history of application in a developing country. Article 18 of China's Environmental Protection Law specifies that "in cases where the discharge of pollutants exceeds the limit set by the state, a compensation fee shall be charged according to the quantities and concentration of the pollutants released". At present, approximately 300,000 factories are monitored and potentially subject to levy collections by national, provincial and local regulators. Although the levy experience has not previously been analyzed systematically, a number of case studies have suggested that the system is poorly administered, that enforcement is largely arbitrary, and that the system is ineffective in controlling pollution. We recently tested this view of the levy system in a collaborative project with NEPA and the Bank's Country Department EA2, using a new database which records the experience of 29 Chinese provinces and urban regions during the period 1987-1993. 6 We studied the water pollution levy because its implementation and impact were well-documented in the information available to us. Our econometric analysis focused on explaining variations in two province-level measures: Industrial emissions intensity (provincial emissions/output) for chemical oxygen demand (COD -- a common measure of organic water pollution) and the effective water pollution levy rate (provincial levy collections for above-standard wastewater discharge/total above-standard wastewater discharge). Differences in factory- level monitoring and enforcement can cause the effective levy rate to vary widely across provinces. The official levy rate determined by the national government applies uniformly China's Effective Pollution Levy across China. However, Figure 5 shows that the effective levy rate varies significantly across provinces. More importantly, the pattern of variation is not random: Effective levies are much higher Dn 31M in urbanized/industrialized provinces of the country, particularly in the eastern coastal regions. Large increases in the official levy since 1987 and significant variations in Figure 5 enforcement have also created a strongly- differentiated pattern of pollution intensities across provinces and over time. We have estimated that from 1987 to 1993, provincial COD intensities fell at a median rate of 50% and total COD discharges declined at a median rate of 22%. 6Wang and Wheeler (1996) 8 Hence, contrary to the conventional wisdom, . our results suggest that the water pollution DeterminantsofLocalEnforcement levy has been neither arbitrarily administered nor ineffective in China. As shown in Figure 6, two sets of local factors make significant contributions to explaining variations in the _tam effective levy. The first, reflecting the principles of environmental economics, is X Da Cr Caa t local valuation of pollution damage. This P"on Load C informaion has three components: total pollution load; a Expiosed Population a Education size of exposed population; and local . Income BargainingPower income. The second is community capacity Figure 6 to understand and act on local environmental problems, indexed by measures of information, education and bargaining power.7 Our results are consistent with the multiple-agent model. Lacking the appropriate information for determining optimal pollution levels in each province, the national government sets the official pollution levy at a 'reference level' and lets officials in each province trade off the costs and benefits of effective implementation. The implications of this result are very clear: The uniform implementation of uniform standards and/or levy rates is not optimal; local conditions determine what these should be.8 Thus, while enabling national environmental authorities in developing countries is an important objective, institutional strengthening programs should also recognize that much of the action takes place (and rightly so) at local levels. Case 2: Indonesia's Public Disclosure Program Enforcement of formal regulation in Indonesia is currently weak, and the modest size of the regulatory budget assures that this weakness will persist in the near future. However, manufacturing is growing at over 10% annually, and the Indonesian Government recognizes the mounting risk of severe pollution damage. Faced with this dilemma, Indonesia's National Pollution Control Agency (BAPEDAL) has decided to initiate a program for rating and publicly disclosing the environmental performance of Indonesian factories. BAPEDAL hopes that pressure on factories from public disclosure will provide a low-cost substitute for formal enforcement of the regulations, and create incentives for the adoption of cleaner technologies. 7Results of a similar nature have also been obseved in Canada and the United States. For more details, see Deily and Gray (1991), and Dion, Lanoie and Laplante (1996). 8 Our results do not imply that current effective levies are optimal. Provincial regulators do not have all the requisite information, nor do they have the capacity for a full assessment of this information. In addition, constraints imposed by low levels of community education or organization may reduce the pressure on local regulators to enforce at optimal levels. However, our results do suggest that provincial effective levies reflect significant elements of self-interest, and are closer to optimum arrangements than has commonly been supposed. 9 In late 1994, BAPEDAL invited us to participate in the design, implementation and analysis of the public disclosure program. The Bank's Policy Research Department and Country Department EA3 agreed to support the project. After six months of intensive work by the BAPEDAL/PRD team, Indonesia's Vice President Tri Sutrisno introduced the program to the public in June, 1995. It is called PROPER -- Program for Pollution Control, Evaluation and Rating (or PROPER).9 In PROPER, a polluter is assigned a color rating based on BAPEDAL's evaluation of its Grading Factories: environmental performance (Figure 7). A Blue rating is given to factories which are in compliance with national regulatory standards; BAPEDAL 's \ Gold is reserved for world-class performers, A and Black for factories which have made no Five-Color B attempt to control pollution and are causing System D serious damage. Intermediate ratings are Red, for factories which have some pollution control but fall short of compliance; and ._. Green, for factories whose emissions control Figure 7 and housekeeping procedures significantly exceed those needed for compliance. Why might PROPER be expected to have a significant impact on pollution? We turn to the PROPER'S Potential Impact regulatory triangle model (Figure 8) for an BAPEDAL explanation. First, while we have noted a \ Infometon & pervasive pattern of 'informal regulation,' or Rtn community influence on polluters' behavior, t our findings also suggest that information problems may distort communities' perceptions ZC of their pollution problems. For example, it is Market often easy to see (and/or smell) the impact of Markets ommunity organic water pollution or sulphur oxide air pollution. However, emissions of Figure 8 bioaccumulative metals and toxins are likely to escape notice. Even where pollutants are clearly visible, local communities frequently cannot gauge the severity of their long-run impact. In addition, communities downstream from polluting industrial complexes often have difficulty identifying individual culprits. Public disclosure offers significant empowerment to local communities in this context. Armed with government-certified performance ratings, they are in a much stronger position to negotiate pollution control agreements with neighboring factories. 9 For more details, see Wheeler and Afsah (1996). 10 Secondly, from the market perspective, PROPER provides a novel application of 'incentive Carrots and Sticks in regulation' principles. Traditional regulation has PROPER been plagued by an important principal-agent GOLD problem: Regulators need good data about firms' * Beyond Compliance performance, but firms have clear incentives to withhold such information. Incentive regulation follows traditional practice by penalizing non- compliance with regulatory standards. However, * Compliance it also addresses the agency problem by rewarding superior performance. This improves the regulators' information by encouraging good Figure 9 performers to identify themselves. It also provides competitive incentives for superior performers to help the regulators identify poor performers, since the latter will be penalized by disclosure. PROPER is expected to work in a similar manner. For non-compliant firms, BAPEDAL expects that that the program will provide an enforcement 'stick' which costs less than conventional procedures. The program also offers important 'carrots' in the form of Green and Gold ratings. BAPEDAL hopes many firms will conclude that the reputational value of Green or Gold status will warrant the costs associated with cleaner production. Moreover, it is important to note that because of PROPER, the Agency subjects itself to scrutiny and creates incentives to improve its performance through transparency. Hence, while an information release program may create incentives for polluters to improve on their environmental performance, it also creates incentives for the Agency to improve on its ability and capacity to collect and process information. In the pilot phase of PROPER, 187 plants were rated. When the program was officially launched PROPER's Short-Term Impact in June 1995, only the names of the five Green ) ..I995 DI,1995 plants were publicly announced. The 121 plants GOLD o O rated as Red or Black were privately notified, and given until December 1995 to improve their l_ (3 1 performance. Full disclosure was implemented on 61 (3 3 ) + 72 December 29; the pilot-phase results are displayed 6% in Figure 10. They suggest that PROPER's short- 115 (61%lblOS term impact as a 'stick' has been substantial. 6 (3%) SO0% 3 Before full disclosure in December, half the Black 6. - 3 plants made successful efforts to upgrade their Figure 10 status, along with a substantial number of Red plants. No short-term impact is observable in the 'carrot' range, but this is not surprising. Attaining Green or Gold status will require longer-term investments, while rapid 11 installation of basic abatement equipment can be sufficient to promote escape from a Black rating. Though preliminary, these results from PROPER suggest that industrial polluters respond to the incentives created by multiple agents. Since the state is not the sole actor, it is important for regulators to recognize that their role is not strictly limited to that of enforcer. In fact, they have access to a much larger set of instruments. Further research will be needed to determine the conditions under which these instruments will reinforce or substitute for one another. 5. Toward a New Paradigm In this paper, we have emphasized two points of departure for a revised model of regulation. First, Broader Implications: we think that the traditional emphasis on State 'appropriate instruments,' while ultimately correct, is premature because most developing- Mk +Cmui country agencies have too many information and _ transactions cost problems to implement any instruments in a comprehensive manner. New Model for Pollution Manaaement First things first: Once regulators have higher- quality information, more integrated information ._. systems, more internal capacity for priority- F igure 11 setting, and a stronger public mandate, it will not be difficult for them to manage pollution more cost-effectively. Overly-hasty introduction of market-based instruments will not work, and is likely to discredit these potentially-powerful regulatory tools. Secondly, the new model of regulation should relegate the regulators to their proper place in the larger scheme of things. The environmental performance of factories is determined by the interactions of multiple agents, with multiple incentives. Although the State can and should have a continuing role in the regulation of pollution externalities, the importance of the Community and the Market must also be recognized. When these two sets of factors are taken into account, a different model of regulation emerges. In our view, appropriate regulation for developing countries should incorporate five key features: * Information Intensity: Effective pollution management by the State is impossible unless regulators have reliable data, integrated information systems and the capacity to set priorities which reflect comparative benefits and costs. Markets and Communities need timely, accurate, public information to make appropriate assessments of factories' environmental performance. An effective regulatory 12 agency will therefore allocate fewer resources at the margin to conventional enforcement and more to the generation and distribution of appropriate information products. Orchestration, not Dictation: A pollution control agency is only one player in the environmental performance game. Agency activities which influence polluters indirectly, through other agents, may be as important as direct enforcement. Potentially high-leverage programs include community environmental education; public disclosure of factory performance ratings; voluntary, public agreements for pollution reduction by industry groups in environmentally-degraded regions; and technical training programs for environmental personnel in polluting factories. Community Control: This should be accepted as a current reality, not as the goal of future programs. And in fact, a substantial role for local communities is appropriate from the perspective of environmental economics. Regardless of the state of formal regulation, local 'informal regulation' is stronger in areas with higher pollution loads, larger affected populations and higher incomes. We also find independent effects for local education and bargaining strength. Taken together, our findings have three implications. 1. Strengthening central regulatory agencies should not empower them to impose uniform standards on heterogeneous communities under the guise of 'administrative efficiency.' Much local variation in regulation is legitimate, and should be recognized as such. 2. Regulatory agencies can play a key role in facilitating negotiations between local communities and neighboring factories. This role includes provision of reliable information on emissions and local ambient quality; technical advice on abatement alternatives; and the transfer of experience from other locations. 3. Central regulators can use their authority to 'level the playing field' for communities which are excessively polluted because their lack of education, organization and bargaining power prevents them from negotiating effectively with local factories. 13 * Structured Learning: Environmental policy reform is a complex business, STRUCTURED LEARNING which will inevitably be subject to many uncertainties. Because it is difficult to know exactly what will work in advance, reforms should emphasize structured leamning. Rather than pre-committing to broad-based programs, agencies should initiate a variety of pilot projects and build -;, larger programs as the lessons are absorbed (Figure 12) Figure 12 . Adaptive Instruments: Newly-industrializing economies can experience rapid changes in ambient quality across air- and watersheds. Since regulation should primarily serve environmental quality objectives, it should be focused on adaptation to these rapid changes. Regulators should be empowered to counter environmental degradation by tightening existing regulations. On the other hand, the system should minimize disruption for investors. Meeting both objectives implies: 1. Transparent adjustment rules, linked to publicly-available data on ambient quality and emissions; 2. Adjustment which is, to the extent politically possible, automatically triggered by deterioration of ambient quality below mandated levels; 3. Adjustment lags sufficient for smooth adaptation by local economic agents. 6. Conclusion Our view of pollution control has been fundamentally changed by our collaboration with regulatory agencies in developing countries. To summarize, we think that the conventional policy discussion is both too shallow and too narrow: Too shallow, because it devotes inordinate attention to instrument choice while ignoring the preconditions for applying any instrument effectively; too narrow because it continues to focus on the State/Factory interaction as the sole determinant of environmental performance. In this paper, we have argued for a less heroic approach to new regulatory programs, and a broader model which includes the Community and the Market as major players in the determination of factories' environmental performance. Finally, we have summarized our current thinking in five key principles for agency reform: Information intensity, orchestration, community power, structured learning, and adaptive instruments. 14 We hope that these ideas will help promote a richer policy dialogue with our partner countries, better project opportunities and, ultimately, better pollution management. 15 References Afsah, S., B. Laplante and N. Makarim, 1996, "Program-Based Pollution Control Management: The Indonesian PROKASIH Program", Policy Research Working Paper 1602, World Bank. Afsah, S., and D. Wheeler, 1996, "Indonesia's New Pollution Control Program: Using Public Pressure to Get Compliance", East Asian Executive Reports, 18, 6, 9-12. Arora, S., and T. N. Cason, 1994, "A Voluntary Approach to Environmental Regulation: The 33/50 Program", Resources, Resources for the Future, Summer 1994. Deily, M.E., and W.B. Gray, 1991, "Enforcement of Pollution Regulations in a Declining Industry", Journal of Environmental Economics and Management, 21, 260-274. Dion, C., P. Lanoie and B. Laplante, 1996, "Monitoring of Pollution Regulation: Do Local Conditions Matter?" World Bank, Policy Research Department (mimeo). Hamilton J. T., 1995, "Pollution as News: Media and Stock Market Reactions to the Toxics Release Inventory Data", Journal of Environmental Economics and Management, 27, 1, 38-48. Hettige M., M. Huq, S. Pargal and D. Wheeler, 1996, "Determinants of Pollution Abatement in Developing Countries: Evidence from South and Southeast Asia", World Development (forthcoming). Hettige M., M. Singh, S. Pargal and D. Wheeler, 1995, "Patterns in Pollution Intensity: Formal and Informal Regulation of Industrial Pollution in the U.S. and Indonesia," presented at the Annual Meetings, American Economic Association, Washington, D.C., January 1995. Hettige, M. and D. Witzel, 1996, "Pollution Control Priorities for Mexico", World Bank, Policy Research Department (mimeo). Huq, M., and D. Wheeler, 1993, "Pollution Reduction Without Formal Regulation: Evidence from Bangladesh," World Bank, Environment Department Working Paper 1993. Huq, M., R. Hartman and D.Wheeler, 1996, "Why Paper Mills Clean Up: Survey Evidence From Four Asian Countries," World Bank, Policy Research Department Working Paper (forthcoming). Laplante, B. and P. Lanoie, 1994, "The Market Response to Environmental Incidents in Canada: A Theoretical and Empirical Analysis," Southern Economic Journal, 60, 3, 657-672. 16 Pargal, S., and D. Wheeler, 1996, "Informal Regulation in Developing Countries: Evidence from Indonesia," Journal of Political Economy (forthcoming). Wang, H., and D. Wheeler, 1996, "Pricing Industrial Pollution in China: An Econometric Analysis of the Levy System," World Bank, Policy Research Department Working Paper (forthcoming). Wheeler, D. and D. Witzel, 1996, "Development, Regulation and the Fate of Sepetiba Bay," World Bank, Policy Research Department (mimeo). 17 Policy Research Working Paper Series Contact Title Author Date for paper WPS1651 Bank-Led Restructuring in Poland: Cheryl W Gray September 1996 B. Moore Bankruptcy and Its Alternatives Arnold Holle 38526 WPS1652 Intra-Industry Trade, Foreign Direct Bernard Hoekman September 1996 F. Hatab Investment, and the Reorientation Simeon Djankov 35853 of Eastern European Exports WPS1653 Grants and Debt Forgiveness in Leonardo Hernandez September 1996 R. Vo Africa: A Descriptive Analysis 31047 WPS1654 Indonesia's Palm Oil Subsector Donald F. Larson September 1996 P. Kokila 33716 WPS1655 Uncertainty and the Price for Crude Timothy J. Considine September 1996 P. Kokila Oil Reserves Donald F. Larson 33716 WPS1656 The Investment Decision: A Cherian Samuel September 1996 C. Samuel Re-Examination of Competing Theories 30802 Using Panel Data WPS1657 Is There an Optimal Structure for Frannie Humplick September 1996 J. Williams Decentralized Provision of Roads? Azadeti Moini-Araghi 82557 WPS1658 Decentralizing Structures for Frannie Humplick September 1996 J. Williams Providing Roads: A Cross-Country Azadeh Moini-Araghi 82557 Comparison WPS1659 Unemployment Insurance in Algeria: Elizabeth Ruppert September 1996 H. Osselyn Implications for a Labor Market in 36039 Transition WPS1660 Mind Your P's and Q's: The Cost of Lant Pritchett October 1996 S Fallon Public Investment is Not the 38009 Value of Public Capital WPS 1661 Determinants of Public Expenditure Susan Randolph October 1996 Z. Bogetic on Infrastructure: Transportation and Zeljko Bogetic 623-7292 Communication Dennis Heffley WPS1662 From Learning to Partnership. Giorgio Barba Navaretti October 1996 M. Patena Multinational Research and Carlo Carraro 39515 Development Cooperation in Developing Countries WPS1663 Interinal Finance and Investment: Cherian Samuel October 1996 C. Samuel Another Look 30802 WPS1664 Pensions in Germany Monika Queisser October 1996 H. Arbi 34663 Policy Research Working Paper Series Contact Title Author Date for paper WPS1665 How Important Are Labor Markets Andrew D. Mason October 1996 D. Ballantyne to the Welfare of Indonesia's Poor? Jacqueline Baptist 87198 WPS1666 Is Growth in Bangladesh's Rice John Baffes October 1996 P. Kokila Production Sustainable? Madhur Gautam 33716 WPS1667 Dealing with Commodity Price Panos Varangis October 1996 J.Jacobson Uncertainty Don Larson 33710 WPS1668 Small is Beautiful: Preferential Trade Maurice Schiff October 1996 M. Patena Agreements and the Impact of 39515 Country Size, Market Share, Efficiency, and Trade Policy WPS1669 International Capital Flows: Do Punam Chuhan October 1996 T. Nadora Short-Term Investment and Direct Gabriel Perez-Quiros 33925 Investment Differ? Helen Popper WPS1670 Assessing the Welfare Impacts Dominique van de Walle October 1996 C. Bernardo of Public Spending 31148 WPS1671 Financial Constraints, Uses of Asli Demirgu,c-Kunt October 1996 P. Sintim-Aboagye Funds, and Firm Growth: An Vojislav Maksimovic 37644 International Comparison WPS1672 Controlling Industrial Pollution: Shakeb Afsah October 1996 D. Wheeler A New Paradigm Bemoit Laplante 33401 David Wheeler WPS1673 Indonesian Labor Legislation in a Reema Nayar October 1996 R. Nayar Comparative Perspective: A Study 33468 of Six APEC Countries WPS1674 How Can China Provide Income Barry Friedman October 1996 S. Khan Security for Its Rapidly Aging Estelle James 33651 Population? Cheikh Kane Monika Queisser WPS1675 Nations, Conglomerates, and Branko Milanovic October 1996 S. Khan Empires: The Tradeoff between 33651 Income and Sovereignty WPS1676 The Evolution of Payments in David B. Humphrey October 1996 T. Ishibe Europe, Japan, and the United Setsuya Sato 38968 States: Lessons for Emerging Masayoshi Tsurumi Market Economies Jukka M. Vesala WPS1677 Reforming Indonesia's Pension Chad Leechor October 1996 G. Telahun System 82407