FMC A WORLD BANK POLICY RESEARCH REPORT IR ~~~~~~~ctober 1999 a_ I f _ -I~~~ I b''--' !_ S i Greening Industry: New Roles for Communities, Markets, and Governments A World Bank Policy Research Report Greening Industry: New Roles for Communities, Markets, and Governments Published for the World Bank Oxford University Press Oxford University Press OXFORD NEW YORK TORONTO DELHI BOMBAY CALCUTTA MADRAS KARACHI KUALA LUMPUR SINGAPORE HONG KONG TOKYO NAIROBI DAR ES SALAAM CAPE TOWN MELBOURNE AUCKLAND and associated companies in BERLIN IBADAN © 2000 The Intemational Bank for Reconstruction and Development / The World Bank 1818 H Street, N\ W, Washington, D.C. 20433 Published by Oxford University Press, Inc. 198 Madison Avenue, New York, NMY 10016 Oxford is a registered trademark of Oxford University Press. All rights reserved. NVo part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopy- ing, recording, or otherwise, without the prior permission of Oxford University Press. Manufactured in the United States of America First printing October 1999 Cover credits: front cover, Tantyo Bangun-Indo Pix; back cover, from top to bottom, Curt Camemark/World Bank, Curt Carnemark/World Bank, Corbis The boundaries, colors, denominations, and other information shown on the maps in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsemettt or acceptance of such boundaries. LibraTy of Congress Cataloging-in-Publication (CIP) Data. Greening industry : new roles for communities, markets, and governments1' Development Research Group, World Bank. p. cm. Includes bibliographical references. ISBN 0-1 9-521127-3 1. Pollution-Developing countries. 2. Environmental policy-Developing countries. 3. Sustainable development. I. World Bank. Development Research Group. HC59. 72.PSSG74 1999 333. 7-dc2l 99-15989 CIP This book was printed using recycled paper and soy based ink. ® q * i The cover stock is 12pt. Comrwall, 15% post consumer waste. The cd sleeve is 1Opt. Cornwall, 15% post consumer waste. The text stock is 70 4 Mohawk 50% recycled, with 15% post consumer waste. Table of Contents Foreword ix The Report Team xi Acknowledgments xiii Executive Summary 1 1. Is Industrial Pollution the Price of Development? 7 1.1 Kuznets Revisited 8 1.2 Focusing on Pollution from Industry 10 1.3 How Economic Development Affects Pollution and Regulation 11 1.4 The Rise and Fall of Pollution Havens 18 1.5 Controlling Pollution: Benefits and Costs 21 1.6 The New Agenda 23 2. Regulating Pollution in the Real World 27 2.1 The Role of Economic Incentives 28 2.2 Pollution Charges: The Right Solution? 35 2.3 Targeting Enforcement 49 2.4 Options for Policy Reform 50 3. Communities, Markets, and Public Information 57 3.1 Communities as Informal Regulators 59 3.2 The Power of the Market 60 3.3 Getting PROPER in Indonesia 64 3.4 Evaluating PROPER 67 3.5 Regulating Pollution and Promoting Equity in the Information Age 74 4. Knowledge, Poverty, and Pollution 81 4.1 Helping Firms Adopt Environmental Management 86 4.2 Who's Complaining about Pollution? 94 4.3 Redefining Environmental Injustice 98 v GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS 5. National Economic Policies: Pollution's Hidden Half 105 5.1 How Trade Reform Influences Polluters 108 5.2 How Input Prices Affect Pollution 109 5.3 The Impact of Plant Ownership on Pollution 113 5.4 Accounting for Pollution's Hidden Half 117 6. Managing and Sustaining Reform 125 6.1 The Contribution of Information Systems 127 6.2 Creating Coalitions for Change 130 6.3 The Politics of Sustaining Reform 133 6.4 Living with Change 137 7. Greening Industry: The New Model 141 7.1 The Keys to Progress 144 7.2 The New Model for Controlling Pollution 147 7.3 The Role of the World Bank 148 Boxes 1.1 Four Fertilizer Plants in Bangladesh 12 1.2 Environmental Regulation and Economic Development 15 1.3 Controlling Air Pollution and Saving Lives in Beijing 22 2.1 Dutch Pollution Charges: An "Accidental" Success Story 38 2.2 Small Is ... Bad or Beautiful? 51 3.1 The U.S. Toxics Release Inventory 70 3.2 Changes in Compliance Patterns Under PROPER 72 4.1 Environmental Management and Regulatory Compliance in Mexico 89 4.2 In China the Poor Get More . .. Pollution 97 5.1 Beyond Anecdotes: Building a Database through Collaborative Research 111 5.2 Industrial Pollution in Indonesia's Financial Crisis 114 5.3 Economic Reform and Industrial Pollution in China 118 6.1 PROPER: Building Credibility 132 6.2 Sharing the Funds in Colombia 135 6.3 Sustaining Reforms in the Face of Political Change 138 Box Figures B1.1 Plants in Bangladesh 12 B2.1 The Impact of Dutch Pollution Charges 38 B2.2 Plant Size and Mortality in Brazil 51 B4.1a Mexican Polluters 89 B4.1b EMS and Compliance 89 vi CONTENTS B4.2 Income and Air Pollution 97 B5.1 Data for Comparative Research 111 B5.2 Financial Crisis and Pollution 114 B5.3a Chinese Provinces 118 B5.3b Plant Size and Ownership 118 B5.3c Pollution Intensities and Reform 119 B5.3d Dirty Sectors in China 119 B6.1 Steps in Developing PROPER 132 B6.2 Using Pollution Charge Revenues 135 B6.3 Elections in Developing Countries 138 Text Figures 1.1 Air Pollution in World Megacities 9 1.2 Air Pollution in Urban China 9 1.3 Polluting Factories in Philippines and Indonesia 11 1.4 Regulation vs. Income 14 1.5 Per Capita lncome and Industrial Pollution 16 1.6 Economic Development and Sectoral Change 16 1.7 Pollution Control Investment in Japan 19 1.8 Import/Export Ratio for Polluting Industries 20 1.9 The Cost of Air Pollution Control in China 23 2.1 Normal Variations in Emissions 29 2.2 Abatement Benefits and Costs 31 2.3 Penalties for Polluting 32 2.4 Abatement Cost 32 2.5 Plant-Level Pollution 33 2.6 Cost-Minimizing Pollution Choices 34 2.7 Optimal Pollution 36 2.8 CORNARE Region 40 2.9 BOD sources in Rio Negro 40 2.10 Results of Traditional Regulation 41 2.11 Malaysian Palm Oil Plantation and Processing Mill 43 2.12 Chinese Factories: Growing Pressure to Improve 45 2.13 Pollution Charges in China 47 2.14 Why Provincial Levies Differ 47 2.15 Polluters in Rio de Janeiro State, Brazil 50 3.1 Clean, Profitable Production 59 3.2 Communities and Polluters 60 3.3 Markets and Polluters 61 3.4 Environmental News and Stock Values in Philippines and Mexico 62 3.5 A Broader View of Regulation 63 3.6 Rating Polluters in Indonesia 65 vii GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS 3.7 Before PROPER 65 3.8 PROPER's First Impact 66 3.9 Results of Disclosure 67 3.10 Extended Impact 68 3.11 PROPER's Expansion: "2000 by 2000" 68 3.12 Public Disclosure in Philippines 73 3.13 PROPER's Legacy 74 4.1 Fuel Use and Pollution from Kilns 83 4.2 Mexican Brick Makers in the '90s: MAC vs. MEP 85 4.3 International Diffusion of ISO 14001 87 4.4 Plant Size and Monitoring Capacity 91 4.5 Plant Size and Compliance in Mexico 92 4.6 Results from Adoption of ISO 14001 93 4.7 Regional Distribution of Complaints 96 4.8 Literacy and Complaints 98 5.1 Air Pollution, 1984-1998 107 5.2 Ownership and Pollution 107 5.3 Trade Policy and Adoption of Clean Technology 109 5.4 Price Reform and Pollution Intensity 113 5.5 Plant Size and Pollution 120 5.6 Plant Size and Regional Development 120 6.1 Monitoring Pollution 128 6.2 Data Collection 130 6.3 Analysis 131 6.4 Reactions to Public Information 133 7.1 Policy Options for Pollution Control 146 7.2 New Dimensions for Policy 147 Tables 1.1 Sectoral Indices of Organic Water Pollution Intensity 14 1.2 Trends in Organic Water Pollution, 1977-1989 18 2.1 Pollution Charge Administration in Rio Negro 41 3.1 Environmental News and Stock Values in Canada and the United States 62 3.2 PROPER's First Impact, 1995 66 3.3 PROPER's Impact After 18 Months 67 4.1 Adoption Index for ISO 14001 Procedures by Mexican Factories 90 4.2 ISO 14001 Certification, 1999 by Country and Region 90 4.3 Mainstreaming Environmental Management in Mexican Factories 91 viii Foreword ver one hundred developing countries have joined the United Nations since Japan's Minamata disaster in 1956. Almost all have environmental agencies, in part because Japan's tragic encounter with heavy-metal poisoning helped spark an international effort to control industrial pollution. The first phase of this effort culminated in 1972, when the United Nations estab- ]ished its Environment Programme and the international commu- nity convened the Stockholm Conference on Sustainable Develop- ment. Between Stockholm and the Rio Earth Summit in 1992, most developing countries set up institutions to regulate pollution. They made steady progress, although it was usually eclipsed by media coverage of disasters such as the lethal landslides in Cubatao, Brazil, and the pesticide factory explosion in Bhopal, India, which killed and injured thousands. Still, pollution regulation arrived in the developing world as an import. Instead of creating new approaches from scratch, most agencies adopted traditional command-and-control regulation with technical assistance from the OECD countries. Unfortunately, this particular import didn't always adapt well to local conditions. By the early 1990s, regulators in many countries had concluded that conventional methods were too expensive and often ineffective. In- novators began experimenting with new approaches, and some yielded excellent results. At the same time, many national economic reforms were proving to be effective in fighting pollution. In this report, we show why these macro and regulatory policy reforms are defining a new model for pollution control in develop- ing countries. We write as participant-observers, because we have helped establish programs as well as studied their impact. Since 1993, we have been privileged to collaborate with pioneers of the new approach in Jakarta, Bogota, Beijing, Rio, Manila, Mexico City, ix GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS New Nations and Pollution Issues 1956-1998 ISO n^ .,T 180- 140 - = 120- -- R (D 100 - _ _ _ _ _ z 80-- l l l 60- 40 _ _ _ _ _ _ _ 1945 1955 1965 1975 1985 1995 Minamata Bhopa ubata |Japan ||India ||Brazil | and elsewhere. This report is really their story. It is also the story of our colleagues in the World Bank and other international agencies. Behind the scenes, they have worked tirelessly to provide new envi- ronmental agencies with financial support, technical assistance, and information about the progress of reform in other countries. The news we bring is hopeful. After six years of research, policy experimentation, and firsthand observation, we believe that en- vironmentally sustainable industrial development is within reach. Greening industry will take time, but even the poorest countries can accomplish it. In this report we show why, and suggest strategies for moving forward. x The Report Team T he principal author of Greening Industry: New Roles for Communities, Markets, and Governments is David Wheeler, Lead Economist for the Infrastructure/Environment Team of the World Bank's Development Research Group. Greening Industry summarizes six years of research and project work by a core team of economists, environmental engineers, and policy analysts: Shakeb Afsah, Susmita Dasgupta, David Gray, Raymond Hartman, Hema- mala Hettige, Mainul Huq, Benoit Laplante, Robert Lucas, Nlandu Mamingi, Muthukumara Mani, Paul Martin, Craig Meisner, Sheoli Pargal, David Shaman, Manjula Singh, Hua Wang, David Witzel and Ping Yun. The report was produced under the direction of Joseph Stiglitz, Lyn Squire, Paul Collier, and Zmarak Shalizi. To learn more about the World Bank's research in this area, please visit the New Ideas in Pollution Regulation website at http://www.worldbank.org/nipr. Material from this website is also included on the CD-ROM that accompanies Greening Industry. xi Acknowledgments T1 nhis report is the product of an extensive research effort by the World Bank's Development Economics Vice Presidency. The cornerstone of our research strategy has been a program of col- laboration with developing-country environmental agencies in the design, implementation and evaluation of new approaches to pollu- tion control. As participant-observers, we have learned a tremendous amount from the pioneers who are showing how innovative pro- grams can reduce pollution significantly, even in very poor countries. We are particularly indebted to the following colleagues: Brazil-from the Environmental Protection Agency for Rio de Janeiro State, FEEMA: Sergio Margulis, former President, Paulo de Gusmao, former Director for Environmental Planning, and Joao Batista; from the Brazilian Institute for Geography and Statistics, IBGE: Jos6 Enfilcio Rocha Collares, Chief of DERNA, Patrfcia Portella Ferreira Alves, Chief of DIEAM, Eloisa Domingues, ISTAM Project Manager, Rosane de Andrade Memoria Moreno and Lucy Teixeira Guimaraes, Technical Team, Industrial Pollution Project; from the Environmental Protection Agency for Sao Paulo State, CETESB: Luis Carlos da Costa. China-from the State Environmental Protection Agency, SEPA: Kunmin Zhang, Deputy Administrator, Xiaomin Guo, Fengzhong Cao and Qingfeng Zhang; from the Chinese Research Academy of Environmental Sciences, CRAES: Jinnan Wang and Dong Cao; from Nanjing University: Genfa Lu. Colombia-from the Ministry of the Environment, Office of Eco- nomic Analysis: Thomas Black Arbelaez, Director, Martha Patricia Castillo, Ana Maria Diaz-Ciceres and Maria Claudia Garcia; from the pollution control agency for Oriente Antioqueno, CORNARE: Leonardo Munoz Cardona, Director and Luis Fernando Castro, Pollution Control Director. xiii GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS India-from the Central Pollution Control Board: Dilip Biswas, Chairman; from the Environmental Protection Training and Research Institute: C. Uma Maheswari, Joint Director; from the Andrah Pradesh Pollution Control Board: Tishya Chatterjee. Indonesia-Sarwono Kusumaatmadja, former State Minister of Environment; from Indonesia's Environmental Impact Management Agency, BAPEDAL: Nabiel Makarim, former Deputy for Pollution Control, Made Setiawan and Dama Ratunanda. Mexico-from the Secretariat for Environment, Natural Resources and Fisheries, SEMARNAP; National Institute for Ecology, INE: Fran- cisco Giner de los Rfos, General Director for Environmental Regula- tion, Adrian Fernandez Bremauntz, Director General for Environmen- tal Management and Information, Luis R. Sanchez Catano, Director for Metropolitan Environmental Management, and Luis F. Guadarrama. Philippines-from the Department of Environment and Natural Resources, DENR: former Minister Victor Ramos and Bebet Gozun; from the University of the Philippines: Tonet Tanchuling and Alex Casilla. We are indebted to many colleagues in the World Bank who have directly supported our work and/or participated in our collab- orative program: Kulsum Ahmed, Adriana Bianchi, Dan Biller, Carter Brandon, Richard Calkins, Cecilia Guido-Spano, Ken Chomitz, Mau- reen Cropper, Shelton Davis, Adrian Demayo, Michelle De Nevers, Charles DiLeva, Yasmin D'Souza, Evelyn de Castro, Clara Else, Gun- nar Eskeland, Ben Fisher, Kristalina Georgieva, David Hanrahan, Patrice Harou, Nicholas Hope, Patchamuthu Illangovan, Gregory In- gram, Maritta Koch-Weser, Vijay Jagannathan, Emmanuel Jimenez, Todd Johnson, Andres Liebenthal, Lawrence MacDonald, Anna Maranon, Richard Newfarmer, Saed Ordoubadi, Mead Over, Jan Post, Violetta Rosenthal, Elizabeth Schaper, Teresa Serra, Katherine Sierra, Lyn Squire, Andrew Steer, Laura Tlaiye, Lee Travers, Walter Vergara, Joachim Von Amsberg, Konrad Von Ritter, Thomas Walton and Roula Yazigi. We would also like to thank the following World Bank colleagues for their assistance, advice, and insights: Richard Ackermann, Jean Aden, Nick Anderson, Bernard Baratz, Carl Bartone, Roger Batstone, Antonio Bento, Jan Bojo, Annice Brown, Shantayanan Devarajan, John Dixon, David Dollar, Alfred Duda, Jack Fritz, Richard Gains, Robert Goodland, Daniel Gross, Kirk Hamilton, Jeffrey Hammer, Na- garaja Rao Harshadeep, Gordon Hughes, Frannie Humplick, Ian John- xiv ACKNOWLEDGMENTS son, Bjorn Larsen, Stephen Lintner, Magda Lovei, Kseniya Lvovsky, Dennis Mahar, Desmond McCarthy, Jean-Roger Mercier, Ashoka Mody, Carl-Heinz Mumme, Lant Pritchett, Ramesh Ramankutty, Ge- offrey Read, John Redwood, Jitendra Shah, Sudhir Shetty, Karlis Smits, Sari Soderstrom, John Williamson, Jian Xie and C.H. Zhang. Book design, editing, and production were directed and man- aged by the Production Services Unit of the World Bank's Office of the Publisher. Sandra Hackman played a major role in editing the report. David Shaman coordinated production of the report for the Bank's Development Research Group. xv Greening Industry: New Roles for Communities, Markets, and Governments C onventional wisdom holds that developing countries cannot hope to clean up industrial pollution of their air and water until they reach a level of affluence seen today only in wealthy countries. According to this view, continued expansion of industrial output will inevitably worsen the already severe levels of pollution that are common today in urban areas of the developing world. Another prevalent belief is that growing global trade and open borders are encouraging dirty industries to move to developing countries, which cannot afford to curb environmental abuses. Six years of research, policy experiments, and firsthand observa- tion have shown this picture to be false. Factories in many poor countries run cleaner than a decade ago, and total emissions are ac- tually falling in some areas where industry is growing rapidly. What's more, "pollution havens"-developing countries that pro- vide a permanent home to dirty industries-have failed to material- ize. Instead, poorer nations and communities are acting to reduce pollution because they have decided that the benefits of abatement outweigh the costs. Environmental regulators in developing countries are trying fresh approaches and finding new allies in the battle to curb pollu- tion. These initiatives stem from widespread recognition that tradi- tional pollution regulation is inappropriate for many developing GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS countries. New regulatory institutions are often unable to enforce conventional discharge standards at the factory level. Many regula- tors also recognize that such standards are not cost-effective because they require all polluting factories to toe the same line, regardless of abatement costs and local environmental conditions. To break out of this one-size-fits-all approach, developing- country regulators are opting for more flexible and efficient systems that nevertheless provide strong incentives for polluters to clean up. Some of the pioneers have turned to financial incentives by charging polluters for every unit of their emissions. As results from programs in Colombia, China, and Philippines have shown, many managers opt for serious pollution control when they face steep, regular pay- ments for emissions. And pollution charges not only cut emissions but generate public revenue as well-which in turn can support local efforts to control pollution. Other environmental reformers are using simple rating systems to publicly recognize factories that adhere to local and national pol- lution standards-and to train the communal eye on those that do not. By classifying factories based on their reported emissions, and widely broadcasting the results, regulators are enabling communi- ties to identify serious polluters and pressure them to clean up. This channel for "informal" regulation has proven to be potent, even in cases where formal regulation is weak or absent. Such public disclo- sure programs also enlist the efforts of investors, lenders, and con- sumers, whose concern over liability from poor environmental prac- tices and desire to reward green manufacturers brings pressure to bear on polluters. Indonesia and Philippines, in particular, have shown that such public disclosure programs can curb pollution at modest cost. Public education regarding the sources and impacts of pollution also provides a powerful lever for improving the lives of poor people, who suffer greatly from emissions even as industry's pollution inten- sity declines. Armed with good information, poor citizens can work with environmental agencies and elect political leaders willing to pressure factories to curb emissions, as regions and countries make the transition to greener industry. To ensure the success of such programs, regulators are relying on low-cost computer technology that cuts the cost of gathering, processing, and distributing information. Selective, focused use of environmental databases and computer models, along with public 2 EXECUTIVE SUMMARY involvement, also helps communities and businesses negotiate envi- ronmental priorities and action plans based on a common under- standing of the impact of pollution and the cost of abating it. These initiatives are working because they have a solid eco- nomic foundation. Plant managers do not pollute because they enjoy fouling the air and water but because they are trying to minimize their costs, so they will tolerate emissions up to the point where the penalty for more pollution becomes greater than the cost of control- ling it. In fact, managers' sensitivity to costs gives regulators many opportunities to influence their decisions. At the factory level, for ex- ample, environmental agencies can lower pollution-control costs by supporting training in environmental management for small and medium-scale enterprises. Recent pilot projects in Mexico have shown that such programs can provide a cost-effective complement to conventional regulation. At the national level, economic reforms can also reduce pollution. Greater openness to trade can enhance managers' access to cleaner technology, while cutting subsidies for raw materials can encourage companies to reduce waste. State-owned enterprises are often heavy polluters, so privatization can contribute to cleaner production. Coun- tries as diverse as China, India, and Brazil have demonstrated the power of such measures to reduce pollution. But economic reforms are no panacea, because growth-promoting measures can make local pollution worse in some cases. To ensure sustainable development, economic reformers should anticipate such impacts and work closely with environmental agencies to offset them. Overall, the proliferation of innovative channels for reducing emissions has created a new model for pollution control in develop- ing countries. In this model, regulation is information intensive and transparent. As environmental agencies exert influence through for- mal and informal channels, they become more like mediators and less like dictators. Community representatives take their place at the negotiating table along with regulators and factory managers. Mar- ket agents make their presence felt through the decisions of con- sumers, bankers, and stockholders. The new model gives policymakers more options, but it also im- poses new responsibilities-for strategic thinking about the benefits and costs of pollution control; a strong commitment to public partic- ipation; clever, focused use of information technologies; and willing- ness to try new approaches such as pollution charges and public dis- 3 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS closure. Of course, regulators will always have important responsi- bilities for monitoring factories' environmental performance and en- forcing regulations. But in the new model, regulators use more re- sources to provide better public information, encourage informal regulation, furnish technical assistance to managers, and promote environmentally sound economic reforms. We write about this model as participant-observers, because we have helped establish many of the innovative programs that we dis- cuss, as well as studied their impact. Since 1993, we have collabo- rated with pioneers of the new model in Indonesia, Colombia, China, Brazil, Philippines, Mexico, and other countries. This report is really the story of those pioneers-their ideas, programs, and results. It is also the story of our colleagues in the World Bank and other interna- tional agencies who have worked tirelessly to provide the reformers with financial support, technical assistance, and information about environmental initiatives in other countries. Together, these experiences have persuaded us that the conven- tional wisdom is wrong: Economic development and industrial pol- lution are not immutably linked. We are convinced that developing countries can build on the new model to reduce industrial pollution significantly, even if they grow rapidly during the coming decade. 4 Chongqing, 1998 Source: Katnnka Ebbe ChaDter One Is Industrial Pollution the Price of Development? I n China, a generation of economic growth has given millions a lifestyle beyond the dreams of their grandparents. China's urban consumers celebrate their new prosperity by strolling through downtown malls in cities like Chongqing. But as China's cities have boomed, the simple pleasures of sunshine and clear air have been lost. Pollution from motor vehicles, smokestacks, and home hearths is so thick that Chongqing consumers can't see the tops of office towers a few blocks away. The most dangerous pollutants are partic- ulates-tiny airborne particles that lodge deep in the lungs, causing severe and sometimes fatal respiratory problems. In four Chinese cities alone-Chongqing, Beijing, Shanghai, and Shenyang-10,000 people will die prematurely this year from exposure to particulates. In the filthy clouds hanging over China's cities and the smog plaguing other poor countries, unspoken questions lurk: Is pollution simply the price of development? Does this generation have to en- dure an environmental tragedy for the sake of future generations? Many people in both developed and developing countries believe the answer is yes, as stories in the popular media often reinforce the idea that pollution control is limited to the industrial economies. After all, the proof is right before our eyes . . . or is it? In fact, recent evidence shows that many developing countries have already turned the corner in the fight against industrial pollu- tion. Factories are running cleaner than a decade ago, and total emis- sions are starting to fall even in areas where industry continues to GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS grow rapidly. The cleanup has begun because developing countries have decided that the benefits of pollution control outweigh the costs. This realization has prompted many countries to adopt innovative strategies that enlist local communities, consumers, investors, and economic policy reformers in the struggle against industrial pollution. Polluters, in turn, are discovering that they have no place to hide- and showing that they can reduce pollution quickly while producing profitably if regulators provide the right incentives. Industrial pollu- tion is still exacting a heavy price in developing countries, but there is no longer any reason to accept it as the price of development. 1,1 Kuznets Revisited A generation ago, the U.S. economist Simon Kuznets proposed that income inequality generally rises as development proceeds, falling only after the rewards of growth accumulate. Similarly, some researchers have claimed to identify an environmental Kuznets curve, in which pollution from industry, motor vehicles, and house- holds increases until development generates enough wealth to pro- mote significant pollution control. Whether the turning point occurs when countries reach per capita incomes of $5,000 or $15,000 per year has never been clear. But the implication is that for highly pol- luted cities in poor countries (Figure 1.1), another generation of growth will create nightmarish conditions. Fortunately, the evidence doesn't support such a bleak vision. Sao Paulo, for example, has lower particulate air pollution than Los Angeles (Figure 1.1), and Bombay's level is scarcely higher. Present- day Jakarta and Santiago have air quality comparable to that of many developed-country cities in the 1950s-yet the former have much lower incomes. China's growth experience casts further doubt on the environ- mental Kuznets curve, which would predict rapidly increasing pollu- tion in such a poor country. Recent data suggest that average urban air quality in that country has stabilized or improved since the mid- 1980s (Fig. 1.2). At best, environmental Kuznets curves provide snapshots of a dynamic relationship between pollution and development that is evolving in response to experience. To understand the forces under- lying this evolution, we need to pay closer attention to the complex factors driving environmental progress in developing countries. 8 IS INDUSTRIAL POLLUTION THE PRICE OF DEVELOPMENT? Figure 1.1 Air Pollution In World Megacities Particulate Air Pollution 500 400 Standard 0 100 0~~~~~ Source: UNEP/WHO, 1992 F igure 1.2 Air Pollution in Urban China, 1987-1995 Average Ambient TSP Concentration Weighted by Population in Chinese Cities soo - 400 300 20 -_ North 100 - _ ~~All Cities '-I-- Coastal 1987 1988 1989 1990 1991 1992 1993 1994 1995 Year Source: China Environmental Yearbooks (SEPA) 9 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS 112 Focusing on Pollution from Industry In many cities, a large share of air pollution comes from motor vehicles and home hearths, while household sewage is a major con- tributor to water pollution. Emissions from industry are also perva- sive, although they vary considerably in relative importance. At one extreme, China's State Environmental Protection Agency estimates that pollution from factories accounts for over 70 percent of the na- tional total, including 70 percent of organic water pollution, 72 per- cent of sulfur dioxide emissions, and 75 percent of flue dust, a major component of suspended particulates. Many polluting industries are located in China's densely populated metropolitan areas, where emissions exposure can cause particularly serious damage to human health and economic activity. In many Brazilian cities, by contrast, households and motor vehi- cles emit the lion's share of serious air and water pollution. These major pollution sources deserve serious attention. However, this re- port will focus on emissions from factories rather than attempting a comprehensive analysis of urban pollution. We have chosen to work on emissions from industry for two main reasons besides their signif- icant contribution to overall pollution. First, we are following the lead of our colleagues in developing-country environmental agencies. Dur- ing their first phase of development, they have focused their limited resources on major industrial polluters. Such polluters are feasible to regulate because they are stationary, relatively easy to identify, and more amenable to pollution control than smaller polluters such as households, informal-sector enterprises, and motor vehicles. Emissions from industry also provide an excellent domain for comparative analysis because they are more highly varied than those from other sources. Industry emits hundreds of air, water, and solid pollutants, contributing to smog, buildup of heavy metals, or- ganic water pollution, hazardous solid waste, and many other sources of damage to communities and ecosystems. Investigating these highly varied emissions has generated a wealth of new infor- mation for sound environmental policy making: on the sources of pollution, their relative contributions to environmental damage, and differences in the costs of controlling them. Rather than providing an exhaustive treatment of the issues sur- rounding control of industrial pollution, we highlight recent experi- ences with regulatory and economic policy reforms whose impacts have been documented. Rich sources of complementary social and economic data from standard national surveys facilitate this study. 10 IS INDUSTRIAL POLLUTION THE PRICE OF DEVELOPMENT? Equipped with such data, we have been able to investigate the role of many factors in promoting pollution reduction. 1.3 How Economic DevelopmentAffects Pollution and Regulation Because regulatory institutions in many poor countries are weak, we might expect factories to pollute with no restraint. However, con- sider the record of three Asian developing countries: Bangladesh, In- donesia, and Philippines. The poorest is Bangladesh: In a flood- and cyclone-prone area the size of an average U.S. state, 115 million Bangladeshis subsist on an average income of US$270 per year. The country is just beginning to regulate pollution, and industrial sectors such as paper, chemicals, and fertilizer nearly always discharge wastes into rivers that serve downstream populations. However, a study of fertilizer plants in Bangladesh finds wide variation in their environmental performance. Some are serious polluters, while others have made major efforts to control their emissions (Box 1.11). Traditionally, Indonesia and Philippines have also lacked a strong commitment to enforcing pollution control regulations. During the past few years, however, both countries have begun programs for rating and publicly disclosing factories' compliance with regulations (see Chapter 3). The programs have rated several hundred factories for over two years, and at least half now adhere to organic water pol- lution regulations in each country (Figure 1.3).1 Figure 1.3 Polluting Factories in Philippines and Indonesia FACTORIES' COMPLIANCE WITH WATER POLLUTION REGULATIONS PHILIPPINES INDONESIA Very Good Compliant Not Compliant .. - _i~~~~~~~~~~~~~~~~~~~~~~~...... Very Bad Sources: DENR (Philippines); BAPEDAL (Indonesia) 11 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 1.1 Four Fertilizer Plants in Bangladesh In 1992, a World Bank team surveyed four of Figure B1.1 Plants in Bangladesh the five urea fertilizer plants in Bangladesh (Huq and Wheeler, 1992). All of these plants were public enterprises managed by the Bangladesh Bangladesh Chemical Industries Corp. (BCIC), but their ages varied widely, and they were scattered in urban and rural locations through- out the country. All the plants were located on rivers into which they discharged their waste- water. All used natural gas as the basic feed- stock, included both ammonia and urea facili- aNDIA ties, and operated on self-generated electricity. Our survey investigated process technolo- gies, end-of-pipe treatment efforts, and the effi- Dumki ciency of general waste management. At that INDIA (LHCB) time, Bangladesh had no regulation-based in- centives for end-of-pipe (EOP) waste treat- ment, so we expected the enterprises to have eCUF invested little in such efforts. We were wrong. l . Despite their operational similarities, the factories' EOP treatment and pollution varied Bay of Bengal widely. NGFF (Natural Gas Fertilizer Factory, Syl- het) was Bangladesh's oldest urea fertilizer plant, built with Japanese assistance in 1961. Downstream villages had clearly identified the UFF and PUFF (Urea and Potash Urea Fer- plant's discharges as the cause of fish kills, tilizer Factories, Narsingdi) were built in differ- paddy-field damage, and health threats. Yet ent eras: UFF with Japanese assistance in 1968, community pressure for change was only mod- and PUFF with Chinese assistance in 1985. erate, as the area is primarily nonindustrial and Technologically, however, they were roughly at offers few other factory jobs. BCIC, too, regarded parity because the Chinese design closely re- the facility as obsolete and kept it open only to flected the two-decades-old Japanese design. preserve the local employment base. Everyone Both plants were clearly identifiable polluters recognized that the age and technology of the whose damage intensity fell in the mid range of plant precluded cleanup to a high standard, so our survey. Downstream fish kills and paddy nearby communities settled for some compensa- damage from polluted irrigation water were tion and a first-level cleanup effort. common, and the community exerted strong 12 IS INDUSTRIAL POLLUTION THE PRICE OF DEVELOPiMENT? Box 1.1 (continued) pressure for cleanup in the 1980s, emboldened CUF (Chittagong Urea Factory, Chittagong) by the relative abundance of other local jobs. was the country's newest, largest, and most ad- In response, UFF increased the number of em- vanced urea fertilizer factory. Because it was ployees working on pollution control, and both constructed in 1989 with Japanese assistance plants paid some compensation for damage and incorporated modern Japanese technology, claims. The two plants also shared a first-stage CUF was a very clean plant. A treatment la- treatment lagoon, constructed by UFF in 1980. goon had been excavated, but the effluent load Both factories used the lagoon to dilute the ef- was so low that the plant discharged waste- fluent with wastewater from their employee water directly into the Karnaphuli River. Al- housing complexes. UFF also used urea hydrol- though local employment alternatives were ysis, an ion exchange facility, and an oil/grease plentiful, neighboring communities had put no separation plant to clean up its effluent. PUFF, pressure on CUF, as they considered its envi- in turn, reduced the ammonia load in its efflu- ronmental controls acceptable. These controls ent with a steam stripping method, and spread surpassed any regulatory standards that the a simple cloth barrier over the outfall to cap- Government of Bangladesh was likely to en- ture some of the oil and grease. force in the coming decade. These findings suggest that an intriguing and hopeful story is un- folding in the developing world. Long before reaching middle-income status, countries like Indonesia, Philippines, and Bangladesh have begun an environmental transition in which some factories are demonstrating high levels of environmental performance. The role of economic development in this transition is clearly re- vealed by a study based on reports submitted to the 1992 U.N. Con- ference on Environment and Development held in Rio de Janeiro (Dasgupta, Mody, Roy, and Wheeler, 1995). The research shows a continuous relationship between national income per capita and the strictness of environmental regulation (Figure 1.4, Box 1.2). Accord- ing to a recent World Bank study, the result is a 1 percent decline in the intensity of organic water pollution-the amount per unit of in- dustrial output-for each 1 percent increase in income per capita. The study is based on extensive data from environmental agen- cies in Brazil, China, Finland, India, Indonesia, Korea, Mexico, the Netherlands, Philippines, Sri Lanka, Taiwan (China), Thailand, and the United States.2 Overall, the data reveal that pollution intensity falls by 90 percent as per capita income rises from $500 to $20,000 (Figure 1.5). Most important, the fastest decline occurs before coun- tries reach middle-income status. 13 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 1.4 Regulation vs. Income Development and Regulation in 31 Countries 2.6 - 2.4 - __ 2.2 - 2 . 2 C '1.8 1.2 2 2.5 3 3.5 4 4.5 Log (Income Per Capita) Source: Dasgupta, Mody, Roy, and Wheeler (1995) However, total pollution in developing countries could still rise if industrial output grows faster than pollution intensity declines. This is especially true because development affects the share of an econ- Table 1.1 Sectoral Indices of omy's polluting industries. An economy that depends largely on food Organic Water Pollution Intensity and paper production, for example, poses a much greater threat of organic water pollution than one based on metals and nonmetallic Sector Index minerals (Table 1.1). Yet an analysis of data from more than one hun- Food 100 dred countries reveals that development shifts production to sectors Pulp and Paper 87 that produce less organic water pollution per unit of output. This shift Chemicals 29 toward cleaner sectors reduces overall water pollution intensity by 30 Textiles 26 percent as income rises to around $5,000 per capita (Figure 1.6). Wood Products 18 We still need to estimate total organic water pollution in each Metals 3 sector of growing economies, to determine whether industrial ex- Nonmetallic Minerals 2 pansion yields more waste. Such estimates are difficult to find, so to produce them we have used another result from the 12-country Source: Hettige, Mani, and Wheeler study cited above. In those countries, we found that for each 1 per- (1998) cent rise in per capita income (and wages), labor intensity also 14 IS [NDUSTRIAL POLLUTION THE PRICE OF DEVELOPMENT? Box 1.2 Environmental Regulation and Economic Development We have analyzed international differences in tries. Our survey assessment uses a variety of environmental regulation using reports pre- questions to categorize the (i) scope of policies sented to the United Nations Conference on En- adopted; (ii) scope of legislation enacted; (iii) vironment and Development (UNCED, 1992) control mechanisms in place; and (iv) degree by 145 countries. The UNCED reports are simi- of success in implementation. The status in lar in form as well as coverage, and permit each category is graded "high, medium, or cross-country comparisons. To an impressive low," with assigned values of 2, 1, and 0 re- degree, they seem to reflect real environmental spectively. We have developed over 500 assess- conditions and issues. ment scores for each country and computed From the information in these reports, we separate composite indices of regulations for have developed a set of indicators that mea- air and water pollution. The resulting index sure the status of pollution control policy and values increase continuously with national in- performance in 31 randomly selected coun- come per capita. Income and Environmental Regulation GNP Air Water GNP Air Water Per Capita Regulation Regulation Per Capita Regulation Regulation Country ($1990) Index Index Country ($1990) Index Index Mozambique 80 56 98 Paraguay 1,110 84 117 Tanzania 110 50 90 Jordan 1,240 95 131 Ethiopia 120 20 56 Thailand 1,420 98 113 Bhutan 190 39 54 Tunisia 1,440 128 158 Malawi 200 93 116 Jamaica 1,500 114 168 Bangladesh 210 77 89 Bulgaria 2,250 168 198 Nigeria 290 75 106 South Africa 2,530 136 165 India 350 105 132 Brazil 2,680 113 127 China 370 98 127 Trinidad 3,610 118 149 Kenya 370 85 127 Korea 5,400 150 170 Pakistan 380 105 131 Ireland 9,550 203 223 Ghana 390 93 124 Netherlands 17,320 219 226 Zambia 420 87 115 Germany 22,320 236 242 Egypt 600 92 134 Finland 26,040 214 229 Philippines 730 93 113 Switzerland 32,680 231 240 Papua New Guinea 860 54 91 Source: Dasgupta, Mody, Roy, and Wheeler (1995) 15 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 1.5 Per Capita Income and Industrial Pollution Income and Industrial Water Pollution Intensity: International Comparison 100 _ a, 90 r -80- 70 70 60 50- 40- 30- 20- 10 0 I 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000 20,000 Per Capita Income (SUS) Source: Hettige, Mani, and Wheeler (199S) Figure 1.6 Economic Development and Sectoral Change Industrial Water Pollution Intensity: The Impact of Economic Development 5.5 5' 4a Y 3.5- 500 1,500 2,500 3,500 4,500 6,000 8,000 10,000 12,000 14,000 17,000 20.000 Income Per Capita ($US) BOD: Biochemical Oxygen Demand Source: World Bank BESD Database; Hettige, Mani, and Wheeler (1998) declines by approximately 1 percent. As countries grow richer, rising wages lead to lower demand for labor per unit of output. The study shows that economic development has a parallel im- pact on organic water pollution intensity: Stricter regulation and greater productive efficiency lead to lower pollution per unit of out- 16 IS INDUSTRIAL POLLUTION THE PRICE OF DEVELOPMENT? put. As a result, an average Indian paper mill employs far more workers and generates far more pollution than a U.S. mill with the same capacity. But because labor and water pollution intensities de- cline at about the same rate with development, the two mills have similar pollution/labor ratios. We can use these results to estimate total industrial water pollu- tion loads for a variety of countries, using a U.N. database that pro- vides annual employment figures for each industry sector and coun- try. For example, to estimate organic water pollution from paper production in each country, we multiply that country's paper-sector employment by our estimated (constant) pollution/labor ratio for the paper sector. We multiply employment in the metals sector by its pollution/labor ratio, and similarly for all other industry sectors. Then we add across sectors to obtain total estimated organic water pollution for each country. To determine the relationship between economic development and total organic water pollution during the 1970s and 1980s, we have selected 15 countries within four major economic groups: the Organisation for Economic Co-operation and Development (or OECD), represented by the United States, Japan, France, and Ger- many; the newly industrialized economies (or NIEs), represented by Mexico, Brazil, Taiwan (China), Korea, South Africa, and Turkey; the less-developed countries of Asia (or LDCs), represented by China, India, and Indonesia; and the ex-COMECON countries, represented by Poland and the former Soviet Union. In the OECD countries, despite continued economic growth, we estimate that total organic water pollution declined by 4 percent from 1977 to 1989 (Table 1.2), reflecting rising per capita income and more regulation. Organic water pollution in the NIEs increased by about 40 percent, while in the poorer Asian countries it grew at a slightly higher rate-49 percent. Since the latter three countries are very large, we estimate that they generated most of the pollution growth in our international sample (Table 1.2). The OECD and ex- COMECON countries dropped significantly in their share of pollu- tion, while the NIEs increased their share only marginally. Perhaps most impressive is our estimate that total organic water pollution from industry grew by only 16 percent in these 15 major industrial countries. Although economic growth sparked fears of skyrocketing pollution in the 1970s and 1980s, development was set- ting the stage for real improvements in environmental performance. 17 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Table 1.2 Trends in Organic Water Pollution: Selected Countries, 1977-1989 Emissions ('000 Kg/Day) % Ch. Region 1977 1980 1983 1986 1989 1977-89 OECD 5,776 5,847 5,501 5,403 5,523 -4 COMECON 4,127 4,218 4,302 4,228 4,039 -2 NIEs 1,565 1,917 1,848 2,197 2,188 40 ASIAN LDCs 4,617 5,030 5,566 6,183 6,883 49 TOTAL 16,085 17,012 17,217 18,011 18,633 16 % of Total Pollution 1977 1980 1983 1986 1989 OECD 36 34 32 30 30 COMECON 26 25 25 23 22 NIEs 10 11 11 12 12 ASIAN LDCs 29 30 32 34 37 Source: Hettige, Mani, and Wheeler (1998) 1.4 The Rise and Fall of Pollution Havens The pattern of international trade provides another measure of this record. Northern environmental groups have long expressed the concern that poor countries will become pollution havens, attracting industries that relocate from richer countries to avoid strict regula- tions, and siphoning away jobs in the process. Yet a look at overall trade statistics shows that permanent pollution havens have not emerged. Concern about pollution havens began in the early 1970s, when developed countries rapidly tightened pollution controls and most developing countries had not yet begun formal regulation. Business investment in pollution controls skyrocketed in Japan during that time (Figure 1.7), and companies in North America and Western Europe made similar investments. If such costs gave an edge to pol- luting industries in developing countries, the effect should have ap- peared in international trade patterns: Developing countries' exports of the products of dirty industries should have risen faster than their imports, lowering their import/export ratios for these products. The converse should have been true for developed countries. 18 IS INDUSTRIAL POLLUTION THE PRICE OF DEVELOPMENT? Figure 1.7 Pollution Control Investment in Japan Japan: Pollution Control Investment by Large Enterprises, 1973-1985 10,000 8,000 / 2 6,000- .2~~~~~~~~~N O4,000 2,000 0 l l 1973 1975 1977 1979 1981 1983 1985 Year Source: Mani and Wheeler (1998) Figure 1.8 shows that the shadow of pollution havens did emerge in five particularly polluting sectors: iron and steel, nonferrous metals, industrial chemicals, pulp and paper, and nonmetallic mineral prod- ucts.3 After the early 1970s, Japan's import/export ratio in these in- dustries rose rapidly, while the ratio declined steeply in the newly industrialized economies (NIEs) of the Republic of Korea, Taiwan (China), Singapore, and Hong Kong (China). And the same pattern oc- curred in mainland China and the other developing countries of East Asia a decade later. However, in each region the pollution haven story was markedly short. Both sets of Asian economies have stabilized their import/export ratios at levels greater than one, and remain net importers of pollution-intensive products from industrial countries. The story in the Western Hemisphere is similar. In North Amer- ica, the United States and Canada witnessed a steady climb in import/export ratios for polluting industries from the beginning of the environmental era to the late 1980s, while Latin America experi- enced the opposite after 1973. However, as in developing Asia, the Latin American ratio leveled off near one by the 1990s. Why didn't polluting industries continue to shift to developing countries? Economic growth-accompanied by more regulation- provides the best answer. Along with greater prosperity in the newly 19 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 1.8 Import/Export Ratio Trends for Polluting Industries 1.2 -- North America L2 Europe 1-~~~~~~~~~~~~~~~. 0 o6 02.> 0.4 -- 0.4 0.2 - . 1- 0-I11111111111 111 1 111111 111 1963 19661969 1972 1975 1978 1981 1984 1987 1990 1993 1963 19661969 1972 1975 1978 1981 1984 1987 1990 1993 1-2 Japan 4 5Latin America 0.9 -- ~~~~~~~~~~~3.5 0.6 ~~~~~~~~~~~ ~~2.5 0.4 ~~~~~~~~~~~~~1.5 0.3~~~~~~~~~~~~~~~. n n- 1963 1966 1969 1972 1975 1978 19811984 1987 1990 1993 1963 1969 1975 1981 1987 1993 9 --Asian NIEs 6 --Developing East Asia 5 7 6 4 3- 41 3 2- 2 HHHHH H HH H H 0- 1 i 1 1 i E ;sis||s||E| oll 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 Source: Mani and Wheeler (1998) 20 IS INDUSTRIAL POLLUTION THE PRICE OF DEVELOPMENT? industrialized countries came increased demands for environmental quality and better institutional capacity to regulate. The same process occurred in the Asian developing countries after a decade's delay. Faced with rising costs from environmental damage, they stabilized the terms of trade through measures to control their own pollution. 1.5 Controling Pollution: Benefits and Costs Poor countries are taking more steps to control pollution, but they must carefully justify such efforts because resources used to curb emissions could also be used to build schools or train doctors. Yet environmental policymakers in developing countries who look closely at the benefits and costs of controlling pollution are moving toward even stronger support for regulation. China provides an excellent case in point. Output from the coun- try's 10 million industrial enterprises grew by more than 15 percent annually during the 1990s, and industry is China's largest produc- tive sector, accounting for 47 percent of its gross domestic product and employing 17 percent of the country's labor force. Despite the country's progress in controlling pollution, serious environmental damage has undeniably accompanied this rapid growth. As we have noted, China's State Environmental Protection Agency (SEPA) esti- mates that industry accounts for over 70 percent of the nation's or- ganic water pollution, sulfur dioxide (SO2) emissions, and flue dust. Atmospheric concentrations of suspended particulates and S02 in urban areas routinely exceed World Health Organization safety stan- dards by large margins. China's pollution problem is clearly compelling, but how much more pollution control can the Chinese afford to undertake? To begin to weigh benefits and costs, a team of Chinese researchers has esti- mated the link between air pollution and mortality from respiratory disease in Beijing.4 Their analysis shows that a "statistical life" could be saved by removing 100 tons of SO2 annually from Beijing's atmosphere (Box 1.3). But how much would it cost to abate those 100 tons of SO2? To find out, we estimated abatement costs for large and small plants in China: Figure 1.9 shows the incremental cost per ton of pollutant re- moved as the degree of abatement rises. The scales on the vertical axes of the two graphs indicate that small plants have much higher marginal abatement costs than large plants, and that state-owned 21 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 1.3 Controlling Air Pollution and Saving Lives in Beijing Xu et al. (1994) have estimated dose-response damage from air pollution. In the case of relationships linking atmospheric pollution to China, there is reason to believe that 30 to 40 respiratory disease in Beijing. Their study percent of fine particulates are in the form of shows that atmospheric sulfur dioxide (SO2) sulfates from SO2 emissions. concentration is highly correlated with dam- In 1993, Beijing had a population of about age from respiratory disease. Recent scientific 11,120,000; the mortality rate was about 0.611 evidence provides some insight into the nature percent; total deaths were about 68,000; and of this relationship. Sulfur dioxide and other total S02 emissions were about 366 thousand oxides of sulfur combine with oxygen to form tons (of which 204 thousand were from indus- sulfates, and with water vapor to form aero- try). From this base, a decrease of 1,000 tons sols of sulfurous and sulfuric acid. These acid in S02 emissions decreases total emissions by mists can irritate the respiratory systems of hu- 1/366 x 100 percent. An independent econo- mans and animals. Therefore, a high concen- metric analysis of the relationship between tration of S02 can affect breathing, and may emissions and air pollution in China's cities aggravate existing respiratory and cardiovascu- predicts an associated decrease of 0.51 x 1/366 lar diseases. Sensitive populations include asth- x 100 percent in Beijing's ambient SO2 con- matics, individuals with bronchitis or emphy- centration. Applying the Beijing dose-response sema, children, and the elderly. result of Xu et al. to the new concentration, we The second, and probably more signifi- obtain an estimated saving of 10.4 lives per cant, effect of SO2 is traceable to the impact of year. Dividing both elements by 10 yields a fine particulates on mortality and morbidity. A useful round number for policy discussion: 1 review of recent evidence by the U.S. Environ- life saved per 100 tons abated annually. mental Protection Agency suggests that fine particulates are the source of the worst health Source: Dasgupta, Wang, and Wheeler (1997) enterprises (SOEs) have far higher marginal abatement costs than other plants. As large plants are a major source of air pollution in cities like Beijing, the numbers for those facilities are particularly interesting. Our results show that abating one ton of SO2, when 10 percent of emissions are controlled, would cost large plants about US$3. This is very low by international standards: U.S. environmental policymak- ers have been happy to discover that industry can abate S02 for less than $100 per ton. If abating 100 tons-at a cost of $300-will save a life, can anyone seriously argue that it shouldn't be done? In China, the numbers are clearly signaling that pollution control is far too lax. To consider how much further pollution control should be tight- ened, a simple exercise in valuation is worthwhile. In the West, en- vironmental agencies commonly use a value of at least $1,000,000 to 22 IS INDUSTRIAL POLLUTION THE PRICE OF DEVELOPMENT? Figure 1.9 The Cost of Air Pollution Control in China SO2 Marginal Abatement Costs: Large Plants S02 Marginal Abatement Costs: Small Plants 300 6,000 250 | Non-SOE* 5,000 -4- Non-SOE*| ,p 200 | _ SOE 4,000 | _- SOE a> 150- .u) 3,000 100 _2,000 50 1,000 0 O 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Rate of Abatement Rate of Abatement * State-Owned Enterprise Source: Dasgupta, Wang, and Wheeler (1997) assess the social benefit when pollution control saves one life. In Beijing, it costs only $300 to save a life by abating 100 tons of S02, Using the Western benefit standard, the implied benefit-cost ratio (1,000,000/300) is over 3,000 : 1. Some analysts have proposed much lower benefit estimates for China, but a figure as low as $8,000 would still yield a benefit-cost ratio of 24: 1. In either case, the implied social rate of return for air pollution abatement is ex- tremely high. China's regulators have reduced air pollution signifi- cantly by charging factories for their emissions (see Chapter 2). Yet even for a life-benefit value of only $8,000, we estimate that returns to further abatement are high enough to justify a 50-fold increase in the pollution charge rate. Of course, environmental, social, and economic conditions will yield different conclusions in different countries and regions. But ef- forts to apply the same methods to countries as varied as Brazil and Indonesia have yielded similar results: When the benefits of reduced pollution are weighed against the costs of control, today's regulation appears far too lax.5 Pollution control is a very attractive option for saving lives in the large cities of developing countries. 1.6 The New Agenda The record shows that developing countries are not destined to be the world's environmental dumping grounds: Even the most pol- luted areas are moving away from the nightmare landscapes antici- 23 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS pated only a few years ago. In China, air pollution has been stable or dropping during the past decade despite a rapid increase in income, and we see strong evidence that economic development boosts pol- lution control elsewhere in the developing world. Regulation has grown steadily with income, and its impact has quickly reduced the pollution intensity of industrial production. Yet benefit-cost studies in Asia and Latin America show that pol- lution damage remains unjustifiably heavy, given the low cost of abatement. More action is needed on three fronts: regulatory reform, economic policy reform, and better environmental management within factories. On the regulatory front, new, surprisingly low-cost strategies based on pollution charges and public information have reduced emissions from many factories. We take a detailed look at these strategies in Chapters 2, 3, and 4, drawing on new research and examples of imaginative and effective programs in developing countries. In addition, we explore the complex real-world decision making that these programs try to reflect. Chapter 4 also looks inside the factory gate for more clues to effective pollution fighting. Recent policy experiments suggest that pollution falls significantly when en- vironmental agencies broaden their mandate to include technical as- sistance to plant managers in the private sector. Chapter 5 explores the effects of economic reforms, such as privatization, market liber- alization, and curtailment of subsidies for materials and fuels, to de- termine which can best be used to prevent pollution. Regulatory and economic policy reforms do not take place in a vacuum: in Chapter 6, we identify political and institutional changes needed to support such efforts. In-depth research on these changes is lacking, because most relevant knowledge is in the heads of the people who are leading the process of policy innovation in develop- ing countries. We have been fortunate enough to work with many of them, and this chapter presents the lessons they have taught us. Finally, in Chapter 7, we summarize the main findings of this re- port and highlight the keys to progress. We see an urgent need for expanding the pilot projects we describe and disseminating their lessons internationally, and we hope this report will contribute to such efforts. We also suggest useful roles for our own institution in promoting the new agenda. On balance, given the recent record, we remain optimistic about the prospect for continued progress in con- trolling industrial pollution. 24 IS INDUSTRIAL POLLUTION THE PRICE OF DEVELOPMENT? References Calkins, R., et al., 1994, "Indonesia: Environment and Develop- ment" (Washington: World Bank). Dasgupta, S., A. Mody, S. Roy, and D. Wheeler, 1995, "Environmen- tal Regulation and Development: A Cross-Country Empirical Analysis," World Bank Policy Research Department Working Paper, No. 1448, March. Dasgupta, S., H. Wang, and D. Wheeler, 1997, "Surviving Success: Policy Reform and the Future of Industrial Pollution in China," World Bank Policy Research Department Working Paper, No. 1856, October. Hartman, R., M. Singh, and D. Wheeler, 1997, "The Cost of Air Pol- lution Abatement," Applied Economics, Vol. 29, No. 6. Hettige, H., M. Mani, and D. Wheeler, 1998, "Industrial Pollution in Economic Development: Kuznets Revisited," World Bank Devel- opment Research Group Working Paper, No. 1876, January. Huq, M., and D. Wheeler, 1992, "Pollution Reduction Without For- mal Regulation: Evidence from Bangladesh," World Bank Envi- ronment Department Working Paper, No. 1992-39. Mani, M., and D. Wheeler, 1998, "In Search of Pollution Havens? Dirty Industry in the World Economy, 1960-1995," Journal of Environment and Development, Vol. 7, No. 3. Von Amsberg, J., 1997, "Brazil: Managing Pollution Problems, The Brown Environmental Agenda," World Bank Report No. 16635- BR, June. Xu, X., J. Gao, D. Dockery, and Y. Chen, 1994, "Air Pollution and Daily Mortality in Residential Areas of Beijing, China," Archives of Environmental Health, Vol. 49, No. 4, 216-22. End Notes 1. Indonesia and Philippines have nearly identical color coding schemes, permitting direct comparison of results in Figure 1.3. 2. See Hettige, Mani, and Wheeler (1998). 3. See Mani and Wheeler (1998). 4. See Xu et al. (1994). 5. Similar case studies for Brazil and Indonesia, respectively, are available in Von Amsberg (1997) and Calkins (1994). 25 Colombia's Rio Negro Source: Dauid Shaman Regulating Pollution in the Real World R ising in the Andean highlands, the rivers of Colombia's Antio- quia region tumble wild and clean as they begin their descent to the Caribbean. The headwaters pass through upland ecosystems whose variety makes Colombia a world treasure of bio- diversity. As the highlands give way to broad valleys, the rivers of Antioquia flow more slowly past human settlements. Their sparkle fades as wastes pour in from farms, factories, and towns along their banks. Before the waters reach the sea, their life-sustaining oxygen is depleted and their beds are laced with toxic metals. Sustainers of richly varied life in the highlands, they become purveyors of death in their encounters with human society in the lowlands. Economic development has not been kind to Colombia's rivers. Regulations have stipulated limits on discharges for decades, and Colombians have supported corrective action, but polluters have flouted these regulations for just as long. Yet in the early 1990s com- munity support for cleaner rivers finally crystallized into demands for reform. The result is one of the world's most innovative pro- grams for controlling pollution. Its governing maxim is simple: All polluters-towns, factories, and farms-must pay for each unit of organic pollution they discharge into the waterways of the Antio- quia district. The result? Reported organic discharges have dropped by 18 per- cent during the program's first year. The most striking change has occurred along the Rio Negro, where factories have accounted for GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS over 40 percent of organic pollution: These factories have reduced their organic discharges by 52 percent. Colombia's recent experience reflects a movement toward regu- latory reform throughout the developing world. Decades of attempts to control pollution through traditional regulations, which make dis- charges above designated limits illegal, have often yielded disap- pointing results. Under traditional regulation, pollution above the legal limit is punishable by fines, plant shutdowns, or, in extreme cases, imprisonment of offending managers. But such an approach requires strong enforcement mechanisms: Regulators have to moni- tor and analyze pollution from each plant, determine whether it has violated the rules, and institute legal proceedings in cases where vi- olation is clear. These steps are not cheap, and many developing countries have not been able to implement them. What's more, such a system requires every commercial enterprise to toe the same regu- latory line regardless of cost. In an effort to break out of this one-size-fits-all approach, many countries are opting for more flexible and efficient regulation that nevertheless provides strong incentives for polluters to change their ways. Some countries have chosen strategies for traditional regula- tion that take benefits and costs into account. Some are using pollu- tion charges like those in Colombia-often combined with other strategies-to achieve impressive results. Still others, discussed in Chapter 3, are using public disclosure programs that pressure pol- luters to clean up their act. 2.1 The Role of Economic Incentives We begin with an obvious but important proposition: Plant man- agers respond mainly to economic incentives. Although public spirit moves a notable minority to control pollution, most managers are bound by pressures from markets and shareholders. They will reduce discharges only if they expect the additional cost to be less than the penalties that continued pollution will impose on them. Such penal- ties can include not only fines and plant closures but also pollution charges, credit refusals from bankers worried about liability, reduced sales to consumers who care about the environment, and even social ostracism within communities outraged about pollution. However, a manager's situation is uncertain because a plant's emissions vary daily, local regulators may be spread too thin to en- 28 REGULATING POLLUTION IN THE REAL WORLD force penalties, and reactions from markets and communities are un- predictable. Managers must find the right balance between the pos- sibility of heavy penalties from too much pollution and the certainty of high costs from too much abatement. Understanding this balanc- ing act is the key to more effective regulation. Figure 2.1 shows why this is not a simple problem. The figure portrays information on the concentration of organic pollution in emissions from a large Indonesian factory during 1994 and 1995. At the beginning of a public disclosure program in June 1995, Indone- sian regulators privately notified the plant's managers that they had received a poor rating because their average daily pollution ex- ceeded Indonesia's legal standard of 300 milligrams per liter for that industry sector. Faced with the threat of public disclosure, the man- agers quickly installed equipment designed to reduce concentration to around 100 milligrams per liter. By late November, the plant had Figure 2.1 Normal Variations in Emissions Pollution from an Indonesian Factory 100mg/I 900 mg/l 800 mg/I 700 mg/] 600 mg/I 500 mg/I yse 400 mg/I 300 mg/I 200 mg/I 100 mg/I 0 mg,/I L &= 2 1 2F 2F 2; 2 *ih N o N ,.I C4 ,5 4 O i H N O< N o X IH N X O Ni a N N Source: BAPEDAL 29 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS moved down the learning curve sufficiently to bring its typical emis- sions into the 100 milligrams per liter range. Yet Figure 2.1 shows that even before the treatment equipment was installed, effluent concentrations occasionally dipped below the legal standard. Suppose the standard had been 450 milligrams per liter-would the plant have been in compliance? The answer would have been no if regulators had insisted that all daily observations fall below the standard. Yet if regulators had averaged emissions over time, they might have judged the plant compliant. Faced with such variation, regulators and plant managers alike find themselves in a complex game.1 Inspectors need enough infor- mation to establish a characteristic pollution level. They would like the cooperation of plant management, since managers can easily delay or complicate the regulatory process. For their part, managers have no interest in antagonizing inspectors; because once a plant has become suspect, regulators will demand time-consuming and costly investigations and reports. However, managers will also tend to rationalize strings of bad observations as anomalies, and in some cases they will undoubtedly be justified. The result is that uncer- tainty reigns, and regulation involves continual negotiation. The Regulator's Dilemma Figure 2.2 illustrates the fundamental dilemma regulators must confront-as well as a way of resolving it. The red line shows that each additional (or marginal) unit of pollution creates more damage than the previous unit-progressively more respiratory disease from air pollution, fewer fish in contaminated water, etc. This is called the marginal social damage (MSD) schedule. Pollution abatement is subject to the opposite effect-a law of diminishing returns. As the blue line shows, each additional (or marginal) unit of pollution control costs more than the previous unit. This graph is the marginal abatement cost (MAC) schedule. It shows that pollution control can be cheap at low levels of abatement but expensive at high levels. If regulators target the brown level of pollution, the marginal cost of abatement will be much lower than the marginal social dam- age. This means that reductions in damage through pollution con- trol will more than compensate for increases in abatement costs. The opposite will be true for the yellow level of pollution, where MAC is much higher than MSD. The optimal choice for regulators is the green level of pollution, where MAC and MSD are equal. At this 30 REGULATING POLLUTION IN THE REAL WORLD Figure 2.2 Abatement Benefits and Costs How Much Pollution? Marginal Marginal Abatement Social Cost (MAC) Damage (MSD) b MAC > MSD MSD > MAC Society Gains Society Gains from Less ~ ~ ~ ~ frm or Pollution Control * Pollution Control Pollution point, neither increasing nor decreasing pollution will improve overall social welfare. MAC vs. MEP: The Manager's Dilemma Figure 2.3 illustrates the complex decisions factory managers face in weighing the penalties for polluting. The factory's cost is measured on the vertical axis, and pollution per unit of output (or pollution intensity) on the horizontal axis.2 The two upward-sloping lines show that marginal expected pollution penalties (MEP) in- crease as pollution intensity rises. That's because even weak regula- tors are bound to take notice if the plant's pollution intensity exceeds the legal limit by a wide margin. And even if regulators do not en- force legal standards, communities and markets will exact penalties from obvious, heavy polluters. The green and red MEP lines reflect differences in the strength of local regulation and the quality of in- formation on the factory's pollution available to banks, consumers, and local communities. Confronted with green or red MEP, a cost-minimizing manager needs information about abatement costs before deciding how much to pollute. Figure 2.4 illustrates the cost problem for two different 31 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 2.3 Penalties for Polluting Marginal Expected Penalty for Pollution (MEP) MEP I / | Strong Enforcement, I / | Good Public Information MEP Weak Enforcement, Poor Information Regulatory Pollution/Output Limit Figure 2.4 Abatement Cost Marginal Abatement Cost for Pollution (MAC) $ Change Factors Sector Size Ownership Market Conditions \Worker Skill Envnronmental Management Pollution/Output factories. The red factory incurs much more cost than the green fac- tory, although each incremental unit of abatement costs more for both. Recent research, discussed in later chapters, suggests that the lower MAC of the green plant correlates with factors such as larger 32 REGULATING POLLUTION IN THE REAL WORLD size, ownership by a private, multi-plant company, better-educated workers, and better environmental management. Figure 2.5 combines the red MAC with the green MEP to show how a manager can choose to react to penalties and abatement costs. At the brown level of pollution intensity, MEP is much higher than MAC, so the manager can lower costs by reducing pollution. At the green level of pollution intensity, MAC is much higher than MEP, so the manager can lower costs by reducing pollution control activity. The manager's cost-minimizing choice is the yellow level of pollution intensity, where MAC and MEP are equal. At this level, neither increasing nor decreasing pollution will lower a plant's over- all costs. Why Compliance Varies So Widely Our model of cost-minimizing pollution shows why plants in de- veloping countries vary widely in complying with regulation, even where it is weak. In Figure 2.6, the pairs of MAC and MEP schedules intersect at four points, colored green, blue, yellow, and red. The red (or "outlaw") case occurs when a plant with high MAC faces a weak Figure 2.5 Plant-Level Pollution Cost-Minimizing Pollution MAC Lower Cost by Increasing Lower Cost Pollution by Reducing Pollution/Output Cost-Minimizing Pollution/Output 33 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 2.6 Cost-Minimizing Pollution Choices Why Factories Differ: MAC vs. MEP $ Clean * Compliant * MEP Violator * Outlaw Pollution/Output Regulatory Limit regulatory and information environment in which MEP is low. Such a plant will pollute heavily. Strengthening regulation and public information will shift the MEP schedule from red to green and motivate the plant manager to lower overall costs by reducing pollution intensity to the yellow level. At this level, the plant still exceeds the legally permissible (blue) limit but by much less than in the red case. In contrast, even weak regulation can induce compliance if changes in a plant lower its MAC. In Figure 2.6, the legal (blue) pol- lution limit occurs at the point where green MAC equals red MEP. Recent research shows that plant managers sometimes reduce pollution below the blue point required by law, given other pres- sures from communities and markets (Chapter 3). In Figure 2.6, this occurs where green MEP equals green MAC. Chapter 1 has shown that more pollution control can yield a large social payoff in many developing-country cities. This entails shifting the industry mix from plants that are predominantly red and yellow to those that are predominantly blue and green. As Figure 2.6 shows, this can be achieved by changing MEP, MAC, or both from red to green status. Policies that promote these changes work be- cause they rely on plant managers' natural incentive to minimize their pollution-related costs. 34 REGULATING POLLUTION IN THE REAL WORLD 2.2 Pollution Charges: The Right Solution? Pollution charges, such as those imposed by Colombia, level the economic playing field by confronting all managers with the same price for each unit of pollution. Under such a system, managers are free to adjust their operations until they have minimized pollution- related costs-charges plus the cost of abatement. This system min- imizes overall abatement costs while providing the right incentives for managers to clean up. Yet at first glance a charge system looks unnecessarily complicated. Why not just require all factories to cut back pollution by the same uniform percentage until overall pollu- tion falls to the desired level? That system can also work, but it will heavily penalize factories with high marginal abatement costs. The challenge is to set pollution charges that promote the right level of cleanup from society's perspective. A recent study for Zhengzhou, the capital of Henan province in central China, shows how regulators can do so if they have good information. With a 1993 population of 1.8 million and an average industrial wage of 3,350 yuan per year, Zhengzhou is typical of China's large cities. Its indus- try pours approximately 45,000 tons of sulfur dioxide (SO2) into the atmosphere every year, contributing to an ambient SO2 concentra- tion of 90 micrograms per cubic meter. At this level, over 400 Zhengzhou residents die annually from S02-related pollution, and thousands suffer from serious respiratory illness. At current emissions (100 on the horizontal axis), Figure 2.7 suggests that the benefit from abating an additional ton of SO2-that is, reducing the marginal social damage-is $50, while the cost of abating it is $1.70. This illustrative case uses $8,000 as an extremely conservative estimate of the social benefit from saving a life through air pollution control. Figures in excess of $1,000,000 could be em- ployed, as we note in Chapter 1. However, even a rock-bottom value of $8,000 implies that about 70 percent of today's emissions should be eliminated to achieve the social optimum. The charge that will induce this reduction is about $90 per ton, at the intersection of MAC and MSD.3 This is the optimum charge for Zhengzhou, since a lower charge would leave socially profitable abatement opportunities unexploited, and a higher charge would impose an abatement cost higher than the social gain from further cuts in pollution. This analysis suggests that the air pollution charge should be in- creased more than fiftyfold in Zhengzhou and, by implication, the rest 35 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 2.7 Optimal Pollution Optimal Sulfur Dioxide Pollution in Zhengzhou 140 - 120' 100 j a 80- : 60 40 MAC 20 0 a I 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Sulfur Dioxide Emissions Index Current Pollution of urban China. As we saw in Chapter 1, the current pollution charge makes sense only if China's policymakers value the life of an average urban resident at less than $300. For the loss of a human life, this fig- ure seems ludicrously low compared with the pain, suffering, and elimination of a lifetime's contribution to China's economic output. A pollution charge not only cuts emissions but generates public revenue as well. If Zhengzhou's environmental regulators increased the SO2 levy to $90 per ton, the city's annual revenue from air pollu- tion charges would be approximately $1.1 million. For China as a whole, revenue from an SO2 charge of $90 per ton would be about $250 million-even then, only a small fraction of the charge's value as a lifesaving policy tool. Pollution Charges in Practice Why does pollution control in China and elsewhere fall short of the social optimum, as in the Zhengzhou case? Good studies of emissions and the damage they cause are still limited to a few air pollutants-principally particulates and sulfur dioxide-in a few cities. Guesstimates must be employed for water pollution and haz- ardous waste. As with traditional regulations, effective monitoring, and en- forcement of pollution charges can also be costly and time consum- ing. Claims from industry representatives about the excessive cost of regulation may be well received by high-level policymakers who are 36 REGULATING POLLUTION IN THE REAL WORLD not informed about the benefits of controlling pollution. And argu- ments against charging for illegal pollution are also common on the grounds that criminal acts should be punished, not merely subjected to fees. Thus, although the "golden rule" MAC = MSD provides a good framework for determining environmental goals and pollution charges, in the real world the actual levels are determined through the political process. Concrete information about lives lost, fisheries destroyed, and other damage can play some role, but it will never be the sole determining factor. Policymakers have to seek consensus on environmental goals and then use the available regulatory instru- ments to pursue them. In the 1970s, economists William Baumol and Wallace Oates wrote a classic book showing how pollution charges could be adapted to these political realities.4 They recommended a four-step approach: 1. Determine environmental quality goals; 2. Estimate the pollution reduction required by these goals; 3. Estimate the marginal cost of abatement at the desired level of pollution; 4. Set the pollu- tion charge equal to the estimated marginal cost. If the estimate is right, pollution should fall to the desired level. If it is wrong, the charge can be raised if there is too little abatement and reduced if there is too much. Baumol and Oates have joined other public-finance economists in arguing that all revenues from such a system should be rebated to the central treasury, where they can be allocated to the highest- priority spending categories. These categories might be environmen- tal, but they might also include health care, education, transporta- tion, and other public-sector responsibilities. Has any developing country-or, for that matter, any industrial country-actually instituted an ideal charge system? The answer is no, but some countries have come close. Box 2.1 describes a long- established pollution charge system in the Netherlands, which has applied this economic instrument more successfully than most other OECD countries. Several developing countries have also used charges to regulate pollution. Their experiences illustrate the problems and potential of this economic instrument as a regulatory tool for newly industrializing countries. (1) Colombia Colombia experienced a lamentable lack of success with tradi- tional regulation, and contamination of its air and water long went practically unchecked. In a strong attempt to break with the past, 37 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 2.1 Dutch Pollution Charges: An "Accidental" Success Story Among the OECD countries, the Netherlands has Figure B2.1 The Impact of Dutch Pollution had the most extensive and successful experi- Charges ence with charges for water pollution.5 By 1969, Organic Water Pollution From Dutch organic water pollution had mounted to the point Industry and Households where many Dutch waterways were biologically (Millions of PE) dead. Together industry and households were 50- 40 -E dumping 40 million population-equivalents, or 0 Actual 30 Pollution PE-the average organic pollution caused by 20- W Waste one person in a normal household-into Dutch 10- Tfroeated sewers and waterways every year. Heavy-metals o-a Sewers emissions from industry had also increased to 1969 1975 1980 1985 1990 dangerous levels. Discharges into Waterways or Sewers The Dutch responded with the Pollution of 50 Surface Waters Act (PSWA) in 1970, which pro- 40-_ hibited unlicensed discharges into surface wa- 30- F -Industry ters and imposed charges on polluting emis- 20- Households sions. Industry had to pay for emissions of 10- heavy metals, and all sectors of society were as- 0 sessed for estimated organic discharges: urban 1969 1975 1980 1985 1990 households, 3 PE; farm households, 6 PE; small enterprises, 3 PE; medium enterprises, PE esti- mated from engineering models; and large en- with charges equal to marginal abatement costs terprises, directly measured PE. Authorities at very high levels of cleanup. A careful statisti- granted rebates to small and medium enter- cal analysis by Bressers (1988) has shown that prises if they could prove that their actual emis- these high charges were much more important sions were lower than official estimates. than the permitting process in promoting reduc- The Dutch system began as a command- tions in emissions. By 1990, the system had and-control exercise, in which pollution charges halved both heavy-metals emissions and total were simply intended to finance construction organic discharges into waterways and sewers, of waste treatment facilities mandated by the and waste treatment facilities had expanded suf- PSWA. However, pollution-reduction efforts re- ficiently to reduce organic pollution of water- quired construction of high-cost facilities in ways to about 6 million PE. Industry displayed some areas, and charges escalated as construc- the strongest response to pollution charges from tion costs mounted. At some point, many Dutch 1969 to 1990, reducing its annual organic emis- factory managers found themselves confronted sions from 33.0 to 8.8 million PE (Jansen, 1991). 38 REGULATING POLLUTION IN THE REAL WORLD the country based its new pollution charge system on the Bau- mol/Oates principles. Analysis of abatement costs concluded that a charge of US$100 per ton would reduce industry's organic emissions to Colombia's waterways by 80 percent. However, the program began by charging only US$28 per ton for organic waste (biochemi- cal oxygen demand, or BOD), as well as $12 per ton for total sus- pended solids (TSS). These charges were considered high enough to bite, but not so costly as to provoke hostility from industry. The pro- gram will expand to include other pollutants based on the environ- mental and economic results of the first phase. Seven regions in Colombia with the greatest population, eco- nomic activity and pollution are the flagships for implementing the charge system, and most other regions will begin participation during the next few years. Each region starts by setting its own pollution- reduction goals, imposing the national base charges, and tracking total discharges for six months. If the targets are not met, regional authorities can raise charges for the ensuing six months, and this process continues until local targets are met. At that point the charges are frozen, although adjusted to reflect inflation. The pioneer in instituting this new program has been CORNARE, the pollution control authority in the Oriente Antioqueno region (Fig- ure 2.8). CORNARE's dynamic leaders have forged a good working relationship with local businesses and communities. Before begin- ning the program, for example, the agency worked closely with sev- eral large factories to develop plans for installing cleaner technolo- gies. CORNARE has also collected good information about local water pollution and thus can pinpoint the major sources of dis- charges into the Rio Negro and other rivers. Industry is clearly the kingpin of water pollution in the region, followed by sewage from towns (Figure 2.9). After consulting with factory managers and communities, CORNARE set a reduction target of 50 percent for organic discharges. Although industry leaders protested that such an ambitious target would prove too costly, in- dustry's recorded BOD discharges into the Rio Negro fell by 52 per- cent in the first six months under the plan, and TSS discharges fell by 16 percent. However, factories' responses varied widely: Of the 55 regulated plants on the Rio Negro, only 7 cut their recorded emis- sions of BOD, and only 8 cut TSS emissions. Obviously, the respon- sive plants reduced their pollution much more than average. Table 2.1 shows that CORNARE's administration of pollution charges has been quite efficient: Assessed charges have been signifi- 39 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 2.8 CORNARE Region 4. Santa Fe *de Bogota COLOMBIA cant, and collection rates have been high. Industries and municipali- ties have clearly gotten the message. So why have so few responded? One possibility is that the marginal cost of abatement remains above the charge for many factories, or managers may have simply not had enough time to adjust their pollution control practices. Indeed, CORNARE's director has noted that some plants that reduced pollu- Figure 2.9 BOD Sources in Rio Negro BOD Pollution: Rio Negro Basin Urban Industry * Agro-Industry _ F Urban Sewage Number of Volume of Sources Pollution Source: CORNARE 40 REGULATING POLLUTION IN THE REAL WORLD Table 2.1 Pollution Charge Administration in Rio Negro Total Total Charges Total Charge Pollution Sources Assessed Payments Sector Sources Charged (Mill. Pesos) (Mill. Pesos) Urban Sewage 8 8 57.3 57.3 Urban Industry 55 43 65.6 64.4 Agro-Industry 46 41 .2 .2 Source: CORNARE tion after the charges began had previously agreed to adopt cleaner technologies. Overall, although it is new, the Colombian experience provides support for the argument that a Baumol/Oates pollution charge system can work well in developing countries. (2) Philippines With a total surface area of about 90,000 hectares, Laguna Lake in Philippines is the second largest inland body of water in Southeast Asia. Twenty-one rivers flow into the lake, whose drainage region in- cludes Manila and many smaller cities. According to the Laguna Figure 2.10 Results of Lake Development Authority (LLDA), 1,481 factories occupied about Traditional Regulation 20 percent of the region's land area in 1994. While a few plants tap the lake's water for industrial cooling, most simply use the lake and Factory Ratings in Philippines its feeder streams as sinks for waste. Industry accounts for about 30 percent of the lake's pollution, while agriculture contributes about 40 percent and domestic sewage about 30 percent. Philippines has long maintained a traditional regulatory system, and over 60 percent of local factories have adopted at least nominal pollution control. However, polluters had very little incentive to take regulators seriously because the inspection rate was low, legal en- forcement was time consuming, and most ensuing fines were mini- mal. The results are evident in Figure 2.10, which summarizes a rig- orous audit of water polluters before recent regulatory reforms. Only 8 percent of polluters were found to be in compliance. Polluter Status To provide new incentives and restore Laguna Lake, the LLDA Compliant instituted an "environmental user fee" (EUF) for industrial pollu- tion. Initial studies identified five industries as the primary sources Poor of organic water pollution: food processing, hog farms, slaughter- houses, beverage firms, and textile makers. The agency first imple- Very Poor mented pollution charges-in this case EUFs-in 1997, for a pilot group of 21 plants. The system has two parts: a fixed charge deter- Source: DENR 41 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS mined by discharge volume, designed to cover administrative costs for LLDA, and a two-tier assessment for emissions. The latter in- cludes one charge per unit of emissions that meet the legally per- missible standard, and a higher unit charge for emissions above the standard. As in the Colombian case, abatement cost analyses pro- vided the basis for setting charges at levels that would induce plant managers to cut pollution significantly. After two years of implementation, LLDA reports that BOD dis- charges from the pilot plants have dropped 88 percent. Because pol- lution charges are remitted to LLDA, its resources for monitoring and enforcement have also increased significantly. In light of this experi- ence, the Philippine Government has announced its intention to im- plement the EUF system nationwide. Philippines' experience with pollution charges seems similar to that of Colombia in many respects. Faced with a continuous finan- cial drain rather than sporadic legal action, plant managers have moved quickly to reduce pollution to the point where the marginal cost of abatement is equal to the pollution charge. (3) Malaysia During the 1960s and 1970s, Malaysia grew rapidly while diver- sifying exports away from its two traditional products, natural rub- ber and tin. The country selected palm oil for promotion, and by 1975 private palm oil plantations covered two-thirds as much area as private rubber estates (Figure 2.11). This economic boom, unfortu- nately, was accompanied by an environmental tragedy. Malaysia's palm oil mills discharged their waste effluent directly into nearby waterways. Since this discharge was laden with organic pollutants, the effect on aquatic life was catastrophic. Freshwater fish could no longer survive in 42 of Malaysia's rivers, marine spawning beds near river mouths were dying, and the stench from decomposing anaero- bic waste was so bad that some riverside villages had to relocate. Faced with this crisis, in 1974 the government passed the En- vironmental Quality Act and established the Department of the Envi- ronment (DOE), which could withhold operating licenses from severe polluters. This gave a strong, credible signal to the Malaysian palm oil producers, who began working on waste-treatment technologies. By mid-1977, the DOE was satisfied that the available technologies would support rapid pollution reduction at feasible cost. The agency moved swiftly to enact a system that combined tra- ditional regulations with pollution charges. Within four years, palm 42 REGULATING POLLUTION IN THE REAL WORLD Figure 2.11 Malaysian Palm Oil Plantation and Processing Mill Source: Palm Oil Institute of Malaysia Source: Malaysian Palm Oil Promotion Source: Malaysian Palm Oil Promotion Council Council oil mills were required to reduce BOD in their effluent from 5,000 parts per million (ppm) to 500 parts per million, with the under- standing that the fourth-year standard would not be the final one. Operating licenses were issued for a flat M$100 fee, plus a charge of M$10 per ton of organic pollution discharged into water. Because the DOE had no way of valuing actual damages from pollution, it in- tended this charge to be high enough to provide some abatement in- centive without being burdensome. DOE added a surcharge of M$100 per ton for BOD discharges be- yond the allowable limits. The surcharge, intended to have real teeth, was based on mandatory quarterly discharge reports verified by in- dependent laboratories. Mills were required to apply for an operating license every year and include a description of their waste treatment system. DOE could reject license applications if it disapproved of the treatment approach-but it could also waive all fees for mills en- gaged in serious research and development on cost-effective pollu- tion control. In a single year, these combined measures produced a remark- able change: The mills' average daily discharges fell from about 220 tons to 125 tons. However, managers' decisions suggested that even the M$100 per ton surcharge was often below the marginal cost of abatement. Of 130 mills, 46 paid excess discharge fees of more than M$10,000, and 7 paid more than M$100,000. Compared with the compliance record in other countries, this was a good result; but full compliance with regulations should have lowered the average daily discharge to 25 tons, and the DOE professed disappointment. Now it faced a choice: It could retain the M$100 per ton surcharge while 43 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS continuing to tighten the standard, raise the surcharge to induce faster compliance, or abandon the polluter-pays approach for stricter enforcement along traditional lines. The Malaysian Government chose the third alternative. It aban- doned the surcharge, maintaining only the M$10 per ton discharge fee, and specified that the standards would henceforth be mandatory. The agency proved its intent by taking legal action against many non- compliant mills during the ensuing years. And its approach worked. During the second year, the average mill reduced its BOD discharge to 60 tons. In two years, total organic pollution from Malaysia's palm oil mills fell from 15.9 to 2.6 million person-equivalents.6 This oc- curred despite an increase in the number of mills from 131 to 147 and an increase in palm oil production from 1.8 to 2.6 million tons. By 1981, a sample survey suggested that 90 percent of the mills had cut BOD concentrations below 500 parts per million (ppm), and that 40 percent were below the sixth-year standard of 100 ppm. By 1991, 75 percent of the mills had dropped below 100 ppm, and organic pollu- tion was less than 1 percent of its level when regulation began, even though palm oil production was at an all-time high. To our knowledge, no study has attempted to separate the im- pacts of the fees, the legally imposed standards, and the waivers for R&D on abatement strategies. However, the regulatory package was clearly effective in reducing pollution and improving the quality of Malaysia's rivers. The estimated cost of compliance was also sub- stantial-M$ 100 million by 1984-and in a highly competitive world market, palm growers bore most of the cost. However, the Malaysian boom absorbed this cost with no apparent problem. Unemployment remained low, and palm oil production remained profitable for most producers. Malaysia might have reached the same goal more cheaply by relying almost exclusively on pollution charges, since they would have allowed plant managers the freedom to minimize pollution- related costs. However, efficient enforcement of emissions standards got the job done in a country where public institutions have tradi- tionally functioned very well.7 (4) China In response to its serious emissions problems, China instituted pollution charges in 1979 (Figure 2.12), and almost all of China's counties and cities have implemented this system. Some 300,000 factories have paid for their emissions and more than 19 billion yuan have been collected. About 80 percent of these funds have been 44 REGULATING POLLUTION IN THE REAL WORLD Figure 2.12 Chinese Industry: Growing Pressure to Improve Source: Curt Carnemark, World Bank Source: Corbis used to finance pollution prevention and control, accounting for about 15 percent of total investment in these activities. In sheer magnitude, the Chinese charge system may be without peer in the world; it is also one of the few documented long-term ap- plications of charges in a developing country. However, it differs greatly from an idealized charge system. Plants are charged only for pollution in excess of standards, and the charge is levied only on the single air or water pollutant that most seriously violates regulatory standards for each medium. The charges also provide insufficient economic incentives for compliance, since they are often too low to induce abatement to the legally required level. China's regulators do impose serious penalties, including shut- downs, for plants that persistently violate standards, and have man- dated that some large plants install abatement technologies. Charge revenues are earmarked to support regulators' budgets or pollution- control projects in the same region. Although it has weaknesses, this system has proven highly po- tent in fighting pollution and cutting pollution intensity. For exam- ple, each 1 percent increase in the water pollution levy has led to a 0.8 percent drop in the intensity of organic water pollution from Chi- nese industry.8 And each 1 percent rise in the air pollution levy has cut the pollution intensity of suspended particulates from industrial production by about 0.4 percent.9 The impact of these reductions during a period of rapid indus- trial growth has been remarkable. While industrial output has dou- bled, organic water pollution and air pollution have remained con- 45 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS stant, and even declined in some areas. China's industrial pollution problem is far less serious than it would have been without the levies and other regulatory instruments. China presents a paradox of success. According to estimates for cities like Zhengzhou and Beijing, the air pollution levy should be many times higher in China's major urban areas. But without the levy, pollution-related respiratory diseases would have seriously in- jured or killed hundreds of thousands more citizens. China can build on this demonstrated success. For SEPA, the State Environmental Protection Agency, adjustment of the pollution levy is an important task for the next round of policy reform. The record of responsiveness so far suggests that as the levy rises, Chi- nese industry could reduce pollution far faster than anticipated. 10 Lessons of Experience The experiences of China, Philippines, and Colombia suggest that charges can generate a rapid, large, and sustained decline in in- dustrial emissions. Charges appear to be an almost ideal tool be- cause they provide maximum flexibilitv for both industry and regu- lators, who can use them to pursue varying levels of environmental quality. Other significant lessons have also emerged. (1) Flexible Enforcement China's experience in using charges to control pollution shows that such charges generally flex with local circumstances. World Bank researchers recently investigated this experience in a collabora- tive project with SEPA. Using a new database on 29 Chinese provinces and urban regions from 1987 to 1993,11 they compared actual water pollution charges collected with the amount of wastewater discharged in each region. They found that actual charges per unit of emissions vary widely, although the official rate is supposed to apply uniformly across China (Figure 2.13). This variation is not random: Charges are much higher in urbanized and industrialized provinces of the coun- try, particularly the eastern coastal regions. Two factors explain these variations (Figure 2.14). The first is the price a community places on pollution damage, which varies with the total amount of pollution, the size of the exposed population, and local income per capita. The second is a community's capacity to understand and act on local en- vironmental problems, which is influenced by its level of informa- tion, education, and bargaining power. 46 REGULATING POLLUTION IN THE REAL WORLD Figure 2.13 Pollution Charges in China China's Effective Pollution Levy F- .00 to .05 F 05 to .10 3 ie 0 500 Soulrce: Wang and Wheeler (1996) Similarly determined variations in enforcement of emissions standards have appeared in other large countries such as Canada and India.12 Both within and across countries, the available evi- dence suggests that enforcement varies systematically with local cir- cumstances. Such community-level flexibility in administering na- Figure 2.14 Why Provincial tional regulations is probably critical to continued support for either Levies Differ charges or standards in countries with highly varied environmental, Local Enforcement in China social, and economic conditions. (2) Building Support Political realities indicate that industry has to support any _ charge system, and this support has proved contingent on four con- _ ditions. First, industry has to be convinced that the government is serious about environmental protection. Second, industrialists need . credible evidence that pollution control will not bankrupt them. In both Philippines and Colombia, industry support gathered steam after numerous meetings in which regulators and international ex- perts presented credible information about abatement costs. Third, plant managers tend to support charge systems once they under- 47 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS stand that these systems give them great flexibility. They can abate or pay, as their conditions warrant. The fourth condition relates to how the charge revenues are used. Pollution charges are effective regulatory instruments because they reduce pollution through economic incentives. But while this argument appeals to economists, it cuts little ice with factory cwn- ers. To them, the charge is simply a tax-a financial sacrifice they have incurred for the common good. With remarkable consistency, they refuse to support charges until they are guaranteed that the revenues will be used to finance public or private waste-treatment proiects in their own area. We will return to this issue in Chapter 6. (3) Technical Foundations To maintain a credible charge system, regulators must obtain re- liable data on plant-level emissions. This requires the ability to audit emissions records, enter and store data, and analyze variations in ef- fluent samples from each plant. Regulators also need good proce- dures for collecting and accounting for charge funds. These are stiff requirements, and many agencies are not capable of meeting all of them. Some analysts have argued that information problems can be circumvented through the use of presumptive charges based on en- gineering estimates of pollution from plants of different kinds. In this system, regulators charge a plant using assumptions about the pollution intensity of its operations. The plant can either pay or re- duce the charge by proving that its pollution is lower than the esti- mate. Presumptive charges have surface appeal because they seem to transfer monitoring costs to polluters, but regulators must still verify that emissions reports are correct, maintain consistent data- bases, and keep financial accounts. They are also saddled with the need to create and regularly update a large database of engineering parameters. And, of course, they inevitably have to deal with angry (and politically influential) factory owners who feel overcharged from the outset. In practice, regulators are solving their information and auditing problems by using subcontractors rather than presumptive charges. In Colombia, for example, regulators rely on reports from bonded au- ditors to analyze emissions. The regulatory agency has also subcon- tracted fee collection and financial accounting to Colombia's largest commercial bank, which receives a fixed percentage of the revenue flow. This solution has a triple advantage: The bank has the right ex- 48 REGULATING POLLUTION IN THE REAL WORLD perience to operate such a system, it knows how to collect debts, and failure to pay these debts can threaten a firm's credit rating. 2.3 Targeting Enforcement Despite the attractions of pollution charges, most countries still use traditional emissions standards to control air and water pollu- tion. Yet rigid standards can inflict much economic harm if they are enforced without regard to benefits and costs. Fortunately, regula- tory agencies can actually turn their inability to regulate all factories to their advantage by flexibly targeting plants for monitoring and enforcement. Targeting can crudely approximate the results of a charge system by raising the expected penalties for large pollution sources with low abatement costs. And these plants will often respond more vigor- ously than other factories because they tend to have more skills to draw on, resources to buy and run complex equipment, and ability to spread their administrative costs over many units of activity. Brazilian regulatory agencies have used such a targeting strategy to reduce pollution substantially while economizing on scarce ad- ministrative resources. The agencies assign factories to categories A, B, and C according to plant size, and target the largest (A) factories almost exclusively. How effective is the ABC approach? A good illustration is pro- vided by the case of FEEMA, the pollution control agency of Rio de Janeiro State. FEEMA program analysts have ranked several thou- sand factories according to their contribution to the overall volume and risks of local air and water pollution (Chapter 6). Figure 2.15 presents the results. Remarkably, the analysis suggests that 60 per- cent of the state's serious industrial pollution could be controlled by targeting only 50 factories in the A group. Controlling pollution from 150 plants in the B group would eliminate another 20 percent of the total. Targeting the first 300 plants in the C group, which numbers in the thousands, would cut 10 percent more pollution. Targeting larger plants seems to have impressive potential for reducing pollution, but will it also save lives? Larger factories have taller stacks, so their emissions are more dispersed and less danger- ous to nearby residents. But according to a recent study in Brazil (Box 2.2), large plants remain the source of most deaths because the sheer volume of their emissions simply overwhelms the higher per- unit hazard from smaller plants. 49 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 2.15 Polluters in Rio de Janeiro State, Brazil Rio's ABC Curve: Top 500 Polluters 100 90 so o-------------o-o-o- .2 70 0= 60- - - *j50 40 30 20 10- - - - 0 .-40 i 0 0 0 0 0 0 0 0 0 0 ,4 Ng LO 0 It) 0 It) 0 It) 0 It) 0 *I N CY N C') UV O It) Factory Rank Source: FEEMA The Brazilian pattern is not unique: Research in most countries and regions has revealed similarly skewed effects. The ABC ap- proach can substantially reduce pollution by raising the marginal penalties for a few large polluters with relatively low marginal abate- ment costs. With enough information, regulators can further focus on plants whose abatement costs are exceptionally low, or whose pollution damage is exceptionally high. 2.4 Options for Policy Retorm Our case studies suggest that many roads can lead to reduced pol- lution. They also highlight flexibility as an important key to effective reform. Pollution charges work well because they provide economic incentives for cleanup while affording maximum flexibility to factory managers. Tradable pollution permit systems can provide similar ad- vantages, although they have not been used as widely. Such systems fix overall pollution limits, and permit polluters to buy and sell rights to pollute within the overall limits. The United States has successfully used tradable permits to control national SO2 emissions, and Chile has instituted a tradable permit system to control air pollution in San- tiago. In the future, more developing countries may adopt tradable permit systems. At present, however, well-documented evidence on their implementation and impact remains scarce. s0 REGULATING POLLUTION IN THE REAL WORLD Box 2.2 Small Is ... Bad or Beautiful? Small enterprises have been controversial in Figure B2.2 Plant Size and Mortality in Brazil the environmental and development literature. Industrial Sources of Death from In Small Is Beautiful, E.F. Schumacher touted Air Pollution in Brazil small plants as the agents of choice for sustain- 200- able development. Wilfred Beckerman re- Plant Size sponded with Small Is Stupid, which attacked . 150- Large the idea that small factories are environmen- aMedium tally benign. Beckerman argued that small fac- X=6 Small tories are pollution-intensive, costly to regulate, _ __ and, in the aggregate, far more environmentally harmful than large enterprises. Recent reports _, from the World Bank and other international 1 2 3 4 5 6 7 8 9 10 institutions have tended to side with Becker- Municipalities by Income Decile man, but supporting data have been scarce. Recently, a team from the Brazilian Census * Deaths were added across municipali- Bureau (IBGE) and the World Bank addressed ties to obtain expected deaths by plant this issue by estimating pollution-related deaths size for each income decile. attributable to small, medium, and large facto- ries in Brazil. The team combined an IBGE database of 156,000 factories with economicEiueB2smarzsteeul,whc daaas f 5,00fctre wtBrazi clearly show that large plants account for most and demographic data from 3,500 Brazilian industry-related air pollution deaths in Brazil. municipalities. To provide another interesting Most of these deaths occur in large urban dimension, the team divided the municipalities areas such as S-o Paulo and Rio de Janeiro, into 10 groups according to per capita income. whose municipalities are in the two highest in- The study estimated the impact of emis- come groups. sions on mortality in four steps: The IBGE-World Bank study concluded * A standard World Bank model esti- that some truth lies on both sides of the argu- mated the impact of emissions from ment about small plants and pollution. Per small, medium, and large plants on the unit of output, small plants pollute more and atmospheric concentration of particu- inflict more health damage than large plants. lates in each municipality. However, large plants dominate the mortality * Ostro's "dose-response" function (Ostro, statistics because they produce far greater vol- 1994) converted the estimated concen- umes of output and emissions. Since they also trations to mortality rates. have much lower marginal abatement costs * Multiplication of mortality rates by mu- than small plants, they are a natural focus for nicipality populations yielded expected ABC targeting by regulators with tight budgets. numbers of deaths attributable to small, medium, and large plants. Source: Dasgupta, Lucas, and Wheeler (1998) 51 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Even under traditional standards-based regulation, ABC-style flexibility in targeting based on benefit-cost analysis can approxi- mate the workings of a pollution charge system. Regional flexibility in enforcing national charges or standards also appears important for maintaining the support of communities with different environ- mental, social, and economic conditions. In the next three chapters we will examine other effective roads to pollution control, including public disclosure of polluters' emis- sions and national economic reforms. Like pollution charges and tar- geted enforcement, these approaches reduce pollution by chang- ing the calculations of plant managers who are trying to minimize pollution-related costs. References Baumol, W., and W. Oates, 1988, The Theory of Environmental Policy (Cambridge: Cambridge University Press). Beckerman, W., 1995, Small Is Stupid: Blowing the Whistle on the Green (London: Duckworth Press). Bressers, H., 1988, "The Impact of Effluent Charges: A Dutch Suc- cess Story," Policy Studies Review, Vol. 7, No. 3, 500-18. Cohen, M., 1998, "Monitoring and Enforcement of Environmental Policy," Owen Graduate School of Management, Vanderbilt Uni- versity, August. Dasgupta, S., M. Huq, D. Wheeler, and C.H. Zhang, 1996, "WIater Pollution Abatement by Chinese Industry: Cost Estimates and Policy Implications," World Bank Policy Research Department Working Paper, No. 1630, August. Dasgupta, S., R. Lucas, and D. Wheeler, 1998, "Small Plants, Pollu- tion and Poverty: Evidence from Mexico and Brazil," World Bank Development Research Group Working Paper, No. 2029, November. Dasgupta, S., H. Wang, and D. Wheeler, 1997, "Surviving Success: Policy Reform and the Future of Industrial Pollution In China," World Bank Development Research Group Working Paper, No. 1856, November. Dion, C., P. Lanoie, and B. Laplante, 1998, "Monitoring of Pollution Regulation: Do Local Conditions Matter?" Journal of Regulatory Economics, Vol. 13, No. 1, 15-8. Jansen, H., 1991, "West European Experiences with Environmental Funds," Institute for Environmental Studies, The Hague, The Netherlands, January. 52 REGULATING POLLUTION IN THE REAL WORLD Ostro, B., 1994, "The Health Effects of Air Pollution: A Methodology with Applications to Jakarta," World Bank Policy Research De- partment Working Paper, No. 1301, May. Pargal, S., M. Mani, and M. Huq, 1997, "Inspections and Emissions in India: Puzzling Survey Evidence on Industrial Water Pollu- tion," World Bank Policy Research Department Working Paper, No. 1810, August. Schumacher, E. F., 1973, Small Is Beautiful: Economics As If People Mattered (Reprinted in 1989 by HarperCollins, New York). Vincent, J., 1993, "Reducing Effluent While Raising Affluence: Water Pollution Abatement in Malaysia," Harvard Institute for International Development, Spring. 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, No. 1644, September. - 1999, "China's Pollution Levy: An Analysis of Industry's Re- sponse," presented to the Association of Environmental and Resource Economists (AERE) Workshop, "Market-Based Instru- ments for Environmental Protection," John F. Kennedy School of Government, Harvard University, July 18-20. End Notes 1. For a recent survey of the environmental economics literature on monitoring and enforcement, see Cohen (1998). 2. We use pollution per unit of output to reflect traditional pollu- tion control laws. Environmental regulators do not expect a huge steel mill to produce the same pollution as a corner electroplating shop, but they do expect it to keep its pollution within feasible bounds. So tradi- tional regulations generally focus on discharge intensity-pollution per unit of output or effluent volume-rather than discharge volume. 3. The estimated MSD graph for Zhengzhou slopes downward to the right, while the theoretical graph in Figure 2.2 slopes upward. This difference is caused by the mathematical form of the health im- pact model that researchers have created for Chinese cities. See Das- gupta, Wang, and Wheeler (1997) for further discussion. 4. The most recent edition is Baumol and Oates (1988). 5. Our thanks to our World Bank colleague Carl Bartone for de- tailed documentation of the Dutch pollution charge experience. 6. A person-equivalent is the amount of organic pollutant in the waste produced by one person in one year. 53 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS 7. See Vincent (1993). 8. See Wang and Wheeler (1996). 9. See Wang and Wheeler (1999). 10. See Wang and Wheeler (1996); Dasgupta, Huq, Wheeler, and Zhang (1996). 11. See Wang and Wheeler (1996). 12. For evidence from Canada and India, respectively, see Dion, Lanoie, and Laplante (1998) and Pargal, Mani, and Huq (1997). 54 ; +::+ * :t::: A owNeS S: : e M;d: fft :; ; i - _l l 1S 1 - - E s :XH;d') SOUL0uT;S 00 :ff f; ::; I I I'ii - - l l l l | | | - I I 11 I I I 111 I I I I I I l l _l | l l l l l l l | * I I I * l I : : Development Dilemmas Source: Tantyo Bangun Indo Pix; Corbls Chan Three Communities, Markets, and Public Information S umatra, a huge island in the Indonesian archipelago, is home to the world's largest flower, Asia's largest volcanic lake, and indigenous peoples whose distinctive villages dot the volcanic highlands and forested lowlands. Sparsely populated and resource rich, Sumatra lies across narrow straits from Malaysia, Singapore, and the Indonesian island of Java. As their neighbors joined the East Asian miracle during the 1970s, the people of Sumatra found them- selves squarely in the path of onrushing development. They stood their ground in a succession of conflicts over land use, resource ex- ploitation, and environmental degradation. Some of these conflicts ended tragically, leaving social and environmental destruction in their wake. But some also ended happily, defining progressive new roles for government, business, and local communities. The success story of PT Indah Kiat Pulp and Paper (IKPP) offers some insights into these new roles. I The largest pulp producer in In- donesia, IKPP is also the cleanest. Its mill at Tangerang, West Java, has received several national and international environmental awards, and its Sumatran mill at Perawang is fully compliant with national pollution regulations. But IKPP wasn't always an environmental paragon. In 1984, its Sumatran operation began by importing an outdated factory from Tai- wan (China) that employed elemental chlorine and discharged its GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS wastes into the Siak River after minimal treatment. Round one of the mill's cleanup began in the early 1990s, with a backlash from local villagers. Allying themselves with local and national NGOs, the vil- lagers claimed severe health damage from the mill's emissions and demanded more pollution control and compensation for their losses. In 1992 Indonesia's national pollution control agency, BAPEDAL, me- diated an agreement in which IKPP acceded to the villagers' demands. As this settlement was concluded, Indonesia's export boom ush- ered in round two for the mill. To finance a huge expansion in capac- ity, IKPP needed access to Western bond markets on favorable terms. Faced with potential concern in these markets over the company's long-term liability for pollution damage, IKPP managers opted to make a high-profile investment in clean production. The new facility uses world-class technology that is largely chlorine free and could be converted to totally chlorine-free production. IKPP has absorbed this technology easily because its parent corporation has a large, sophisti- cated engineering staff. What's more, IKPP has shown that large-scale clean production can be profitable in a developing country. Its perfor- mance has been so good that the company's stock value has in- creased while the Jakarta composite stock index has plunged 60 per- cent during the country's current financial crisis (Figure 3.1). The PT Indah Kiat saga illustrates a new model for pollution control in developing countries. Formal regulation had little to do with pollution reduction at the Perawang mill. Defending their own interests, local communities applied pressure for cleanup and com- pensation. Abandoning the traditional agency role, BAPEDAL acted as a mediator rather than as a dictator of environmental standards. Later, pressure from international financial markets propelled IKPP to the next level of environmental performance. In our basic terminology, local and international forces con- fronted IKPP with growing marginal expected penalties (MEP), even though government regulation was weak. Because it was a large branch facility of a sophisticated multiplant firm, the Perawang fac- tory had relatively low marginal abatement costs (MAC). Faced with rapidly rising MEP and low MAC, IKPP's managers opted for a quick reduction in pollution intensity. In this chapter we will argue that the forces that influenced IKPP-links among local communities, market agents, and regula- tors-have sparked several of the world's most innovative experi- ments in environmental policy in countries where traditional regula- tion has failed. These creative programs harness the power of public information, enabling communities and markets to exert maximum 58 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION Figure 3.1 Clean, Profitable Production Indonesian Success Story 120 Indonesia Stock Pera g\ ~ 97 12/97 /6/98 12/98 t \ 9 1l / ~~~~~~BAPEDAL \Surnatera* / Excellent , 4 ~~~~~~~~~~Poor _ *~~~~J Very Poor *English: Sumatra **English: Java influence on polluters. The results suggest that such pioneering ef- forts can have a significant impact on industrial pollution in devel- oping countries. 3.1 Communities as Informal Regulators Abundant evidence from Asia, Latin America, and North Amer- ica shows that neighboring communities can strongly influence fac- tories' environmental performance.2 Where formal regulators are present, communities use the political process to influence the strict- ness of enforcement. Where regulators are absent or ineffective, NGOs and community groups-including religious institutions, so- cial organizations, citizens' movements, and politicians-pursue in- formal regulation by pressuring polluters to conform to social norms (Figure 3.2). Although these groups vary from region to region, the pattern is similar everywhere: Factories negotiate directly with local actors in response to threats of social, political, or physical sanctions if they fail to compensate the community or reduce emissions. Indeed, communities sometimes resort to extreme measures when sufficiently provoked. In the Asian Survey, Robert Cribb has 59 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 3.2 Communities and Polluters Informal Regulation Pants _ Elements * Power unity * Social Norms * Negotiations lNOs| recounted an Indonesian incident "reported from Banjaran near Jakarta in 1980 when local farmers burned a government-owned chemical factory that had been polluting their irrigation channels." In a similar vein, Mark Clifford has reported in the Far Eastern Eco- nomic Review that community action prevented the opening of a chemical complex in Korea until appropriate pollution control equip- ment was installed. When factories respond directly to communities, the results may bear little resemblance to the dictates of formal regulation. For exam- ple, Cribb also cites the case of a cement factory in Jakarta that- without admitting liability for the dust it generates-"compensates local people with an ex gratia payment of Rp. 5,000 and a tin of evaporated milk every month." In India, Anil Agarwal and colleagues (1982) describe a situation where, confronted by community com- plaints, a paper mill installed pollution abatement equipment-and, to compensate residents for remaining damage, the mill also con- structed a Hindu temple.3 If all else fails, community action can also trigger physical removal of the problem. In Rio de Janeiro, for exam- ple, a neighborhood association protest against a polluting tannery led managers to relocate it to the city's outskirts.4 3.2 The Power of the Market The environmental concerns of market agents create additional incentives for pollution control (Figure 3.3). Green consumers are al- 60 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION Figure 3.3 Markets and Polluters Markets as Regulators ME] s Elements * Reputation * Credit | Investors | * Profits ready well known, but investors have also become important actors. A high level of pollution intensity may signal to investors that a firm's production process is inefficient. Investors also weigh poten- tial financial losses from regulatory penalties and liability settle- ments. The importance of such scrutiny has grown with the rise of new stock markets and international financial instruments: Capital markets may revalue a firm in response to bad news about its envi- ronmental performance. News of good environmental performance or investment in cleaner technologies, on the other hand, can en- hance a firm's expected profitability and thus its stock value. Several studies have confirmed that U.S. and Canadian stock markets react significantly to environmental news. Table 3.1 sum- marizes the evidence from recent studies, which report gains from good news and losses from bad news in the range of 1 to 2 percent. Do such changes actually motivate polluters to clean up? A recent study of toxic polluters by Konar and Cohen (1997) suggests that the answer is yes: Firms that experienced the greatest negative impact on stock prices reduced emissions the most. To determine whether such forces affect firms in developing countries, World Bank researchers recently undertook a large-scale study of the impact of environmental news on stock prices in Ar- gentina, Chile, Mexico, and Philippines. None of the four countries has a strong record of enforcing environmental regulations. Never- theless, the study found that stock prices rise when authorities publi- cize good environmental performance and fall in response to public- 61 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Table 3.1 Environmental News and Stock Values in Canada and the United States Negative Performance Information Impact on Stock Value * Muoghalu et al. (1990) Average loss of 1.2% (33.3 M $) * Lanoie, Laplante (1994) Average loss of 1.6% to 2% * Klassen, McLaughlin (1996) Average loss of 1.5% (390 M S) * Hamilton (1995) Average loss of 0.3% (4.1 M $) * Lanoie, Laplante and Roy (1997) Average loss of 2 % Positive Performance Information * Klassen, McLaughlin (1996) Average increase of 0.82% (80 M $) ity surrounding citizens' complaints.5 In fact, the responses are much larger than those reported for U.S. and Canadian firms in Table 3.1: Gains average 20 percent in response to good news, and losses range from 4 to 15 percent in the wake of bad news. Figure 3.4 provides a striking illustration of such impacts for two firms operating in Philip- pines and Mexico. Overall, the message is clear: Capital markets Figure 3.4 Environmental News and Stock Values in Philippines and Mexico Good News Bad News 60.00 60,000 10 10 55.00 E.L 5000 & 50,000 .S 45.00 40,000 =40.00 1 0. 35.00 30,000 .630.00 . , 25.00 .0,000 se20.00 - | otooo 215.00 10.00 , :: : ,,0 -oo l l,l ,~ d ,~ - - ,l ,~ ,~ l o N N N N l N N N N N N NI i N N < N N N N N NI N N N N N N N N N N N 10 , .1 0 N N1 t N X0 10 N 10 N 10 - 10 1 I Date I Date Phtoijines: Regulatory authority highlights San Mexico: Gove'nment f ies Ki--berlv-Clark I Miguel Corporation's environmenta commitment fcr a vater pc lut on v oialtio. and the installation of pollut!on control equiprnent. Source: Dasgupta, Laplante, and Mamingi (1997) 62 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION Figure 3.5 A Broader View of Regulation The New Model everywhere are taking information about environmental performance into account, and firms are responding by cleaning up. Another powerful market influence has been exerted by the In- ternational Standards Organization through publication of ISO 14001, its most recent business performance standard. For the first time, this ISO standard includes explicit norms for environmental management. Hundreds of developing-country firms have already made the changes necessary to qualify for ISO 14001 certification. In Mexico, a recent study shows that even small enterprises seek ISO 14001 certification if they are interested in subcontracting relation- ships with large, ISO-certified enterprises (Chapter 4). Once the roles of communities and markets are introduced, we have a much more robust model for explaining variations in pol- luters' behavior. Even where formal regulation is weak or absent, pressure applied through these new channels can significantly in- crease a plant's expected penalties for polluting. Polluters will react by reducing emissions, just as if government inspectors were enforc- ing regulatory standards. This new story is captured by the regulatory triangle in Figure 3.5. Regulators still play an important part in controlling pollution, but their role is no longer confined to establishing and enforcing standards or charges. Instead, regulators gain leverage through pro- 63 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS grams designed to provide concrete information to communities and markets. 3.3 Getting PROPER in Indonesia The story of a pioneering Indonesian program illustrates the new model in action. Starting in the 1980s, the Indonesian Govern- ment charged BAPEDAL, the national pollution control agency, with enforcing standards on discharges from industrial plants. But en- forcement was weak because the regulatory budget was limited and the courts were plagued by corruption. Meanwhile industrial output was growing at over 10 percent annually. By the mid-1990s the gov- ernment was becoming concerned about the risk of severe damage from pollution. Faced with this predicament, BAPEDAL decided to initiate a program for rating and publicly disclosing the environmental perfor- mance of Indonesian factories. BAPEDAL hoped that the resulting pressure would provide a low-cost way to promote compliance with regulations, as well as create new incentives for managers to adopt cleaner technologies. The program that ensued is called PROPER-for Program for Pol- lution Control, Evaluation and Rating.6 Under PROPER, BAPEDAL rates each polluter on its environmental performance (Figure 3.6). Black denotes factories that have made no attempt to control pollution and are causing serious damage, while red denotes those that have in- stituted some pollution control but fall short of compliance. Factories that adhere to national standards receive a blue rating, and those whose emissions controls and production and waste-management procedures significantly exceed national standards receive a green label. World-class performers attain gold ranking. In the pilot phase of PROPER, which began in early 1995, BAPEDAL rated water pollution from 187 plants. (The agency chose to concentrate on water pollution first because it had data and expe- rience in that domain.) The pilot group included medium- and large- scale polluters from several river basins on the islands of Sumatra, Java, and Kalimantan. Initial ratings showed that two-thirds of the plants failed to comply with Indonesian regulations (Figure 3.7). Although this showing was dismal by Western standards, fully one-third of the rated factories were in compliance despite BAPEDAL's evident inability to enforce regulations. The PT Indah Kiat saga sug- gests why: Two-thirds of the regulatory triangle-local communities 64 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION Figure 3.6 Rating Polluters in Indonesia PERFORMANCE PROPER Rat LEVELS atings 4 GOLD ! . Clean technology, waste GOLD minimization, pollution prevention I Above standards & good maintenance, housekeeping * Efforts meet minimum standards * Efforts don't meet standards • No pollution control effort, • Serious environmental damage Figure 3.7 Before PROPER PROPER: FIRST RATINGS and markets-were already in place, albeit operating with poor infor- mation. These actors had already brought considerable pressure to bear. Public disclosure is a political act and a media event, so BAPEDAL's leaders thought carefully about strategy before releasing the results. In June 1995, Indonesia's Vice President Tri Sutrisno presided over a high-profile public ceremony to congratulate the "good guys"-the five green-ranked plants whose performance ex- ceeded formal requirements. After publicly rewarding these best ac- tors, BAPEDAL privately notified other plants of their ratings, and Very Good gave the non-compliant ones six months to clean up before full pub- lic disclosure. Compliant A scramble ensued as plants with red and black ratings consid- Non-Compliant ered their options, and by December striking changes had already occurred (Table 3.2, Figure 3.8). The most pronounced was a flight Very Poor from the black group, which contracted by 50 percent. Red plants, on the other hand, felt less immediate pressure-only 6 percent im- Source: BAPEDAL 65 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Table 3.2 PROPER's First Impact, 1995 June December Change % Change Gold 0 0 0 0 Source. BAPEDAL proved during the predisclosure period. One green plant also changed status, but not to gold: After the June announcement, the neighboring community had informed BAPEDAL that the plant was in fact polluting heavily under cover of darkness, and the facility was demoted to black. However, four of the original six black-rated plants improved their performance. This left three plants-the new- comer plus two laggards-in the black group by December. The net effect of these changes was an 18 percent expansion of the blue, or compliant, group. Even before public disclosure, PROPER had scored a considerable success. In December 1995, BAPEDAL delivered on its commitment to full disclosure, releasing ratings by industry group over several months to hold media attention. By December 1996-one year later-improvements had become much more pronounced (Table Figure 3.8 PROPER's First 3.3, Figure 3.9). Compliant plants, originally one-third of the sample, Impact now constituted over half. While the green group was unchanged, the blue group grew by 54 percent. Red-rated plants dropped by 24 PROPER RATINGS: percent, and the flight from black continued. Only one plant re- % CHANGE, JUNE-DEC. 1995 mained in the black category-a decline of 83 percent from the orig- 40 inal size of that group. 20 Evidence from mid-1997 suggests that the program's strong im- * pact has continued. For example, BAPEDAL's December 1995 ratings included 118 noncompliant factories-113 rated red and 5 rated -20 - black7-but by July 1997, 38 of those plants had achieved blue or -40o green ranking (Figure 3.10). Only 18 months after full disclosure, -60 PROPER had reduced pollution by more than 40 percent in the pilot group. Considerable turnover was evident in the lowest category: Source: BAPEDAL Four plants upgraded their ratings from black to red (3) or blue (1). 66 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION Table 3.3 PROPER's Impact After 18 Months June December 1995 1996 Change % Change Gold O 0 0 0 .. . . . I I I Blue 61 94 33 +54 Black 6 1 -5 -83 Source: BAPEDAL Four plants rated red in 1995 fell to black in mid-1997 as their con- dition changed or more information became available. With continued political support, the PROPER team hopes to rate 2,000 plants annually by the year 2000. BAPEDAL has also been pursuing its own version of Brazil's ABC targeting strategy, so the share of total water pollution under PROPER's purview is much greater than the proportion of Indonesia's 20,000 factories that it rates (Figure 3.11). If PROPER extends to 2,000 factories within the next two years, it will cover about 10 percent of Indonesia's medium and large industrial plants but about 90 percent of total water pollu- tion. As plant coverage expands, BAPEDAL intends to rate factories on air pollutants and toxic waste as well. 3.4 Evaluating PROPER Given Indonesia's previous regulatory history, this remarkable Figure 3.9 Results of Disclosure result suggests that performance ratings and public disclosure can be powerful tools for improving environmental conditions in developing PROPER RATINGS: countries. Several factors have contributed to PROPER's success. % CHANGE AFTER 18 MONTHS 80 Public Disclosure and Pollution Control 60 40 fl Armed with PROPER-type performance ratings, citizens are in a 20 much stronger position to negotiate pollution control agreements 20 with neighboring factories. This is especially true because lack of in- -40 formation can distort communities' perceptions. For example, resi- -60 l * dents can often see or smell organic water pollution and sulfur oxide --oo air pollution, but emissions of metals and toxins that accumulate in organisms' tissues are likely to escape notice. And even where pol- Source: BAPEDAL 67 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 3.10 Extended Impact lutants are clearly detectable, local communities may be unable to gauge the severity of their long-term impact or identify individual PROPER, PHASE II polluters. The PROPER system adds critical information to this pic- DEC. 1995 - JULY, 1997 ture and certifies the claims of local communities, which can use * cV PROPER's ratings to engage the most serious polluters. PROPER also / llil_ allows each community to more readily choose its own level of envi- / - _ ronmental quality. Better information can also influence the market side of the trian- * t, gle in Figure 3.5. Indonesia has a new stock market and, until the re- * \i S S cent crisis, its rapidly expanding industrial economy has had exten- sive credit needs. With BAPEDAL's ratings, the stock market can more accurately value companies' environmental performance, and banks -/ can factor pollution-related liability into their lending decisions. For consumers, nothing less than a green or gold ranking may Source: BAPEDAL suffice, and the availability of information through outlets such as the Internet-which PROPER has used-may greatly influence their decisions.8 All these factors should encourage polluters to clean up. BAPEDAL itself benefits from public disclosure. More widespread adherence to environmental standards has boosted BAPEDAL's credi- bility with industry, NGOs, and the public and enhanced its ability to do its job. All regulators need good data about firms' pollution, but noncompliant firms have a clear incentive to withhold such informa- tion. Under PROPER, clean firms have an incentive to identify them- Figure 3.11 PROPER's Expansion: "2000 by 2000" PROPER'S EXPANSION PLANS Plants Pollution m's 8 X000% t z 40,0!OOO 60% 400 40% 187 20% 68 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION selves, and the agency can then home in on serious polluters and keep them in the public spotlight. Rewarding good performers also insulates regulators from charges that they are anti-business. PROPER appealed to BAPEDAL because it had neither the re- sources nor the legal support to implement a traditional standards- based system. The agency's managers also decided that they lacked the capacity to enforce pollution charges. Viewing charge-based reg- ulation as an inside transaction between the agency and a plant, they feared that corruption of their inspectors would distort emis- sions information and undermine the market-based approach. Pub- lic disclosure, by contrast, allows communities to check an agency's claims against their own daily experience. PROPER bases its rankings on Indonesia's legal emissions stan- dards, but disclosure systems could also use other benchmarks, such as the average intensity of emissions in each industrial sector or in- ternational performance standards. In fact, public disclosure does not have to rely on benchmarks at all-regulators could simply re- port each plant's emissions. The OECD's Pollutant Release and Transfer Register and the U.S. Toxics Release Inventory (Box 3.1) are examples of such disclosure programs. However, in the developing world PROPER-type systems seem to be taking hold more rapidly. Their strength probably lies in two characteristics: They are compatible with standards-based regula- tions that are still on the books almost everywhere, and they rate en- vironmental performance in a clear, straightforward format that is easy for the media to report and citizens to understand. In principle, each locale could establish its own performance benchmarks to ensure maximum flexibility and efficiency. A sparsely populated area with few critical ecosystems, for example, could use laxer standards than a densely populated industrial area upstream from a marine sanctuary. A plant rated green in one area might well rate red in another. But neither the media nor political actors seem comfortable with such variations, and national and international market players and NGOs would find multiple-benchmark systems confusing. Still, as we have seen, uniform performance standards can raise pollution control costs. To accommodate regional differences and enhance the efficiency of PROPER-type systems, national perfor- mance benchmarks might include three plant sizes, three levels of local environmental quality (heavy, medium, and light pollution), and vary according to industry sector. 69 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 3.1 The U.S. Toxics Release Inventory The U.S. Toxics Release Inventory (TRI) has Community pressure is only one of several annually reported polluters' emissions of more channels through which TRI exerts its effects; than 350 toxic chemicals for a decade. Since the financial community has also responded Congress established the program in 1986, TRI strongly. Research by Hamilton (1995) and has published the names, locations, and toxic Konar and Cohen (1997) has shown significant emissions-by chemical and medium of re- negative market returns for publicly traded lease-of plants with 10 or more employees firms when TRI first reports their pollution. that use at least 10,000 pounds of any listed Firms' market valuation also responds to infor- chemical. The media and environmental mation about changes in the volume of toxic groups provide extensive coverage of the yearly pollution relative to toxic emissions from other announcements. As the accompanying table firms. These results, in turn, create significant shows, U.S. toxic emissions have declined sub- incentives to clean up: Firms with the largest stantially since TRI's beginning. Programs like TRI use informa- Total Releases of TRI Chemicals tion differently from programs like 1988-1994 ('000 Metric Tons) PROPER. In the Indonesian case, a % Change poor rating informs the public that a 1988 1992 1993 1994 1988-1994 firm is not in compliance with national Total Air environmental standards. Disclosure Emissions 1024 709 630 610 -40 programs such as TRI, by contrast, dis- Emissions to seminate "raw" information on toxic Surface emissions with no interpretation or risk Water 80 89 92 21 -73 assessment. Underground One problem is that some chemi- Injection 285 167 134 139 -51 cals covered by TRI are quite danger- On-Site Land ous, even in small doses, while others Releases 218 149 125 128 -41 are hazardous only after long exposure 981 899 -44 at very high levels. By treating all chemicals the same, raw disclosure programs stock market declines reduce emissions more may sometimes alarm the public unnecessarily than other firms. Numerous case studies have and pressure industry into adopting high-cost also shown that TRI induces firms to improve abatement programs that yield few social ben- their ability to manage materials and waste. efits. Academic researchers and NGOs have These successes have inspired similar ef- used media such as the Internet to inform the forts in other countries, including the Chemical public of the relative risks of different chemi- Release Inventory in the United Kingdom and cals, and to assist communities in identifying OECD sponsorship of pilot Pollutant Release large polluters and assessing their overall pol- and Transfer Registers (PRTRs) in Egypt, the lution problems. (The Environmental Defense Czech Republic, and Mexico. The PRTR pro- Fund maintains the most complete such Web grams use the same format as TRI but restrict site at http://www.scorecard.org.) listed chemicals to those with relatively high hazard ratings. 70 COMMUNITIES. MARKETS, AND PUBLIC INFORMATION The Costs of PROPER PROPER's direct costs should include only those entailed in developing the ratings from existing information on emissions and disseminating the results. However, during the program's first 18 months, BAPEDAL devoted most of the program's resources to up- grading the agency's ability to collect and analyze data-efforts nec- essary for any effective pollution control program. The pilot program also employed foreign consultants, although PROPER has since op- erated with much lower levels of foreign involvement. Despite these added expenses, PROPER's costs were only about $100,000 over the first 18 months. With 187 plants rated, the per- plant cost was $535, or $360 per year-just $1 per day. Given that this expenditure produced a 40 percent cut in organic water pollu- tion, PROPER must be judged spectacularly cost-effective. Of course, the acid test for the program comes after the interna- tional consultants have gone home and local industry realizes that life under PROPER will mean permanently increased pressure for pollution control. In Indonesia, the difficulties that normally accom- pany a program's adolescence have been compounded by the na- tion's severe economic crisis, which has produced dramatic cuts in BAPEDAL's budget. But paradoxically, the crisis seems to have strengthened PROPER's appeal. As resources for traditional monitor- ing and enforcement have diminished, Indonesia's leaders have found PROPER's low-cost leveraging of community and market ac- tion even more attractive. Overall, PROPER-type programs are efficient because they lever- age channels inaccessible to formal regulation. However, a recent study found that PROPER had a disproportionate impact on small factories, whose marginal abatement costs are typically high (Box 3.2). PROPER's effects also varied according to plant ownership: Reputation-sensitive multinationals responded most strongly, followed by private domestic firms, and then state enterprises. In short, PROPER induced factories of all types to cut pollution, but it shifted the relative burden to smaller plants and multinationals. The latter were relatively well positioned to bear this burden but the former probably were not. In the next chapter we will show how targeted efforts by government can help small factories overcome this disadvantage. Implementing PROPER-Type Programs Outside Indonesia When PROPER's impact first became apparent, other countries tempered their interest with a critical question: Since shame may be 71 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 3.2 Changes in Compliance Patterns Under PROPER Figure 3.7 summarizes initial compliance levels This was unusual by international standards, for the 187 factories in the pilot group for since research on other countries has shown PROPER. A preliminary BAPEDAL/World Bank that such plants are normally heavy polluters analysis of the program's impact has studied the (Chapter 5). However, after 18 months, the effects of plant size, ownership (public, local, compliance status of publicly owned plants multinational), degree of export orientation, was not significantly different from the status of province, and industrial sector on compliance. other domestically owned factories. The con- When PROPER began, plant size and pub- verse was true for multinational plants. Having lic ownership had significant, positive associa- started at parity in compliance with their do- tions with compliance; sectoral variations were mestic counterparts, they jumped to signifi- also important, but multinational ownership cantly higher compliance status. These results had no effect. After 18 months, compliance suggest highly varied sensitivity to environ- patterns were very different. Plant size, public mental reputation: Multinationals are the most ownership, and sectoral variation had become sensitive, followed by domestic private firms, insignificant as determinants of relative com- and then state enterprises. PROPER increased pliance. However, multinational ownership pollution control in all three types of plants, but jumped to a very high level of positive signifi- the strength of the response differed sharply. cance while export orientation moved the other This result is consistent with the idea that pub- way-toward negative significance. lic disclosure leverages pollution control through We interpret these results as follows. Plant the operation of markets in which environmen- size: Before PROPER, many big plants with low tal performance is valued. abatement costs had already reduced emissions Finally, and paradoxically, the response to because they faced significant expected pollu- PROPER seems to have been perverse among tion penalties from community and market ac- more export-oriented plants. These factories re- tion. Since smaller plants with higher abate- sponded more slowly than domestically ori- ment costs had not reduced pollution much in ented plants, so their relative (not absolute) the "old regime," they found themselves in the compliance was lower after 18 months. This re- spotlight once PROPER began. The resulting sult suggests that different market channels pressure forced them toward parity in compli- have very different sensitivity to environmental ance with big plants. Ownership: When information: International stockholders may be PROPER began, publicly owned plants in In- much more sensitive to environmental perfor- donesia were more compliant than average. mance than international importers. a strong motivator in a public disclosure program, does Indone- sian culture make such a program particularly effective? The answer has proven to be no as other countries have begun PROPER-type programs. 72 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION The most advanced such effort is in Philippines, where the De- partment of Environment and Natural Resources (DENR) has cre- ated a program similar to PROPER, called EcoWatch. In April 1997 EcoWatch published its initial accounting for 52 factories in the Manila area. The summary showed that 48 plants ranked red or black, representing a 92 percent rate of non-compliance. As in In- donesia, the most compelling argument for public disclosure was the evident failure of the traditional approach (Figure 3.12). To establish EcoWatch, the Philippine Government pursued a strategy similar to BAPEDAL's. President Fidel Ramos congratulated the blue plants in a public ceremony (there were no green or gold plants). Red and black plants were privately notified of their ratings and given a substantial period to reduce pollution. Full public dis- closure occurred in November 1998, with broad media coverage. As in the Indonesian case, the program dramatically increased plants' compliance with national regulations. Although no factories reached green or gold status, blue ratings jumped from 8 percent in April 1997 to 58 percent in November 1998. Red ratings fell sharply, while black ratings remained almost constant. Other countries are following closely in the wake of Indonesia and Philippines. Mexico is developing a program called Public Envi- ronmental Performance Indicators, or PEPI, and Colombia's public disclosure program will complement its pollution charge system. At least five other countries have also begun pilot implementation or active consideration of PROPER-type systems (Figure 3.13). What Figure 3.12 Public Disclosure in Philippines Philippines EcoWatch Results APRIL 1997 NOVEMBER 1998 ECOWATCH RATINGS _ World Class * Very Good _ Compliant Non-Compliant Very Poor Source: DENR 73 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 3.13 PROPER's Legacy Current/Planned Mexico China Venezuela Colombi lk , Philippines India t Bangadesh - Thailand Indonesia Papua New Guinea U Current * Planned began as a ripple in Indonesia is clearly building into an interna- tional wave. 3.5 Regulating Pollution and Promoting Equity in the Information Age Widespread acceptance of PROPER reflects a broader trend in public policy. Students of economic development are paying closer attention to the role of social capital-the informal relationships and institutions that strengthen developing communities. Similarly, legal scholars are focusing on the strong complementarity between social norms, which communities draw on to enforce public disclosure programs, and formal laws. The evidence shows that formal and in- formal regulatory mechanisms almost always coexist, but that the latter often dominate in developing countries where regulatory insti- tutions are weak.9 In environmental policy, new thinking about the role of local in- fluence reflects the insights of Nobel economist Ronald Coase, who called traditional regulation into question by noting that pollution vic- tims, as well as regulators, can take action if they perceive that the benefits outweigh the costs.10 As Coase noted, these costs stem from the need to acquire and analyze information, confront polluters, and 74 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION negotiate settlements. Without good information such settlements may be far from optimal. Polluters and regulators usually have the most concrete knowledge of emissions; but polluters are unlikely to share this information in the absence of outside pressure, and bureau- cratic inertia and/or legal constraints often prevent regulators from sharing information as well. Moreover, even if the public has infor- mation on emissions, it may not fully understand the risks it faces. Since polluters are also employers, good information on abatement costs is also important. In short, effective local negotiations require good environmental information, and regulators will often be best positioned to supply it. They can play a valuable new role by focusing more resources on information collection and dissemination, including public disclo- sure of polluters. But a new role for regulators does not mean that they should abandon the traditional one. Efficient enforcement of regulations will remain very important-for its own sake, and be- cause potential penalties provide an incentive for capital markets to react to public disclosure of non-compliance. In addition, as in the case of PT Indah Kiat in Sumatra, regulators can encourage local set- tlements by promoting negotiations, supplying the negotiators with objective information, and, as a last resort, posing the threat of offi- cial sanctions against non-compliant factories that refuse to negoti- ate with pollution victims. Regulators can also serve the special environmental protection needs of poor communities. In countries as different as the United States, China, Brazil, and Indonesia, much of the variation in facto- ries' environmental performance reflects the socioeconomic charac- teristics of the surrounding areas.11 Local residents pressure polluters more successfully if they are richer, more educated, and better able to bargain because they have more employment options. In developed countries, the so-called NIMBY (not in my back yard) phenomenon stems largely from wealthy communities' ability to exclude polluting activities completely. Employment concerns may lead poor commu- nities to welcome industrial activity, but such communities may lack enough political influence and environmental information to negoti- ate effective pollution control agreements. Economic development may be the best antidote to such problems in the long run, but in the meantime poor communities may suffer from excessive pollution.12 Here environmental agencies can help, by educating communities on the pollution risks they face and ensuring that polluters conform to basic national norms. We will return to this issue in Chapter 4. 75 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS References Afsah, S., B. Laplante, D. Shaman, and D. Wheeler, 1997, "Creating Incentives to Control Pollution," World Bank DEC Note, No. 31, July. Afsah, S., B. Laplante, and D. Wheeler, 1997, "Regulation in the In- formation Age: Indonesia's Public Information Program For En- vironmental Management," World Bank, March. Afsah, S., and J. Vincent, 1997, "Putting Pressure on Polluters: In- donesia's PROPER Program," A Case Study for the HIID 1997 Asia Environmental Economics Policy Seminar (Harvard Insti- tute for International Development), March. Agarwal, A., R. Chopra, and K. Sharma, 1982, "The State of India's Environment, 1982," New Delhi, India: Centre for Science and Environment. Arora, S., and T. Cason, 1994, "Why do Firms Volunteer to Exceed Environmental Regulations? Understanding Participation in EPA's 33/50 Program," Land Economics, Vol. 72, No. 4, 413-32. Clifford, M., 1990, "Kicking up a stink: South Korean Government reels from anti-pollution backlash," Far Eastern Economic Re- view, Oct. 18, 72-3. Coase, R., 1960, "The Problem of Social Cost," The Journal of Law and Economics, Vol. 3, October, 1-44. Cribb, R., 1990, "The Politics of Pollution Control in Indonesia," Asian Survey, Vol. 30, 1123-35. Dasgupta, S., B. Laplante, and N. Mamingi, 1997, "Capital Market Responses to Environmental Performance in Developing Coun- tries," World Bank Development Research Group Working Paper, No. 1909, October. Dasgupta, S., R. Lucas, and D. Wheeler, 1998, "Small Manufacturing Plants, Pollution and Poverty: New Evidence from Brazil and Mexico," World Bank Development Research Group Working Paper, No. 2029, December. Dasgupta, S., and D. Wheeler, 1996, "Citizen Complaints as Envi- ronmental Indicators: Evidence from China," World Bank Policy Research Department Working Paper, No. 1704, November. Hamilton, J., 1995, "Pollution as News: Media and Stock Market Re- actions to the Toxic Release Inventory Data," JouLrnal of Environ- mental Economics and Management, Vol. 28, 98-103. Hartman, R., M. Huq, and D. Wheeler, 1997, "Why Paper Mills Clean Up: Determinants of Pollution Abatement in Four Asian 76 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION Countries," World Bank Policy Research Department Working Paper, No. 1710, January. Hettige, H., M. Huq, S. Pargal, and D. Wheeler, 1996, "Determinants of Pollution Abatement in Developing Countries: Evidence from South and Southeast Asia," World Development, Vol. 24, No. 12, 1891-904. Hettige, H., S. Pargal, M. Singh, and D. Wheeler, 1997, "Formal and Informal Regulation of Industrial Pollution: Comparative Evi- dence from Indonesia and the US," World Bank Economic Re- view, Vol. 11, September. Huq, M., and D. Wheeler, 1992, "Pollution Reduction Without For- mal Regulation: Evidence from Bangladesh," World Bank Envi- ronment Department Working Paper, No. 1992-39. Klassen, R., and C. McLaughlin, 1996, "The Impact of Environmen- tal Management on Firm Performance," Management Science, Vol. 42, No. 8, 1199-214. Konar, S., and M. Cohen, 1997, "Information as Regulation: The Effect of Community Right to Know Laws on Toxic Emissions," Journal of Environmental Economics and Management, Vol. 32, 109-24. Lanoie, P., and B. Laplante, 1994, "The Market Response to Environ- mental Incidents in Canada: a Theoretical and Empirical Analy- sis," Soathern Economic Journal, Vol. 60, 657-72. Laplante, B., P. Lanoie, and M. Roy, 1997, "Can Capital Markets Cre- ate Incentives for Pollution Control?" World Bank Policy Re- search Department Working Paper, No. 1753, April. Muoghalu, M., D. Robison, and J. Glascock, 1990, "Hazardous waste lawsuits, stockholder returns, and deterrence," Southern Economic Journal, Vol. 57, 357-70. Pargal, S., and D. Wheeler, 1996, "Informal Regulation of Industrial Pollution in Developing Countries: Evidence From Indonesia," Journal of Political Economy, Vol. 104, No. 6, 1314 +. Pfeiffer, S., 1998, "Power Plants Spark Protest; Blackstone Valley Residents Seek Halt," Boston Globe, Sept. 26. Shane, P., and H. Spicer, 1983, "Market response to environmental information produced outside the firm," The Accounting Review, Vol. LVIII, 521-38. Sonnenfeld, D., 1996, "Greening the Tiger? Social Movements' Influ- ence on Adoption of Environmental Technologies in the Pulp and Paper Industries of Australia, Indonesia and Thailand," Ph. D. Thesis, University of California, Santa Cruz, September. 77 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Stotz, E., 1991, "Luta Pela Saude Ambiental: A AMAP Contra Cor- tume Carioca, S. A., Una Experiencia Vitoriosa," V. V. Valla and E. N. Stotz (eds.), Participacao Popular, Educacao e Saude, Rio de Janeiro, 133-60. Tietenberg, T., and D. Wheeler, 1998, "Empowering the Community: Information Strategies for Pollution Control," paper presented at the conference "Frontiers of Environmental Economics," Airlie House, Virginia, Oct. 23-25. 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, No. 1644, September. End Notes 1. For a detailed account, see Sonnenfeld (1996). 2. See Pargal and Wheeler (1996), Hettige, Huq, Pargal, and Wheeler (1996), Huq and Wheeler (1992), Hartman, Huq, and Wheeler (1997), and Dasgupta, Lucas, and Wheeler (1998). 3. Such arrangements are not confined to developing countries. Even in strictly regulated societies like the United States, communi- ties can make life difficult for plants that violate local norms, whether or not their activities meet formal regulatory requirements. A good example is the recent controversy over proposed construc- tion of four new electric power plants in Massachusetts' Blackstone Valley. Although the proposed plants easily exceed regulatorv re- quirements-and indeed, they have been lauded by environmental groups-the plants have met with stiff community resistance. Local leaders have protested that three power plants already operate in the region, and that four more facilities will harm the community's qual- ity of life, use up water, and lower property values. According to Sacha Pfeiffer (1998), writing in the Boston Globe, "In order to make the plants 'more palatable,' several power companies have offered compensation packages to Blackstone Valley communities, including money for high school scholarships, new water and sewer facilities and water conservation programs." 4. See Stotz (1991). 5. See Dasgupta, Laplante, and Mamingi (1997). 6. For detailed descriptions and analyses of PROPER, see Afsah, Laplante, Shaman, and Wheeler (1997), Afsah, Laplante, and Wheeler (1997), and Afsah and Vincent (1997). 78 COMMUNITIES, MARKETS, AND PUBLIC INFORMATION 7. During the predisclosure period, PROPER rated several addi- tional plants, finding two new blacks and five new reds. Red and black plants therefore number 118 in Figure 3.10 and 111 in Table 3.2. 8. BAPEDAL's PROPER ratings can be found on the agency's Web site: http://www.bapedal.go. id/profil/program/proper. html. 9. For a more detailed discussion, see Tietenberg and Wheeler (1998). 10. See Coase (1960). 11. See Pargal and Wheeler (1996), Hettige, Pargal, Singh, and Wheeler (1997), Hettige, Huq, Pargal, and Wheeler (1996), Huq and Wheeler (1992), Hartman, Huq, and Wheeler (1997), and Dasgupta, Lucas, and Wheeler (1998). 12. See Dasgupta and Wheeler (1996), Wang and Wheeler (1996), and Pargal and Wheeler (1996). 79 -~ m ~~~~~~~~~~~~~~Idsra Dulim Pvryan Polt n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Sore Cobi Knowledge, Poverty, and Pollution M 11s j exico's border with the United States runs from the Pacific Ocean to the Rio Grande, then winds southeast with the river until it empties into the Gulf of Mexico near Mata- moros. Along this 1,900-mile frontier, thousands of maquiladora fac- tories assemble products for duty-free export to the United States. Their explosive growth has provided employment for thousands of young Mexicans, but some factories have also contributed to pollu- tion along both sides of the border. Some of the worst pollution problems occur in the Paso del Norte region, where the Rio Grande crosses from the United States into Mexico. Here the cities of Ciudad Juarez, Mexico, and El Paso, Texas, flank the river in a high desert valley that is prone to thermal inversions. In the 1980s, air pollution mounted as growth in both cities accelerated. The U.S. Environmental Protection Agency cited El Paso's air quality as substandard, and U.S. environmental groups opposed to the North American Free Trade Agreement (NAFTA) pointed to the region's problems. Better to defer NAFTA, some ar- gued, until Mexican industry could handle pollution problems like those in the Paso del Norte. But which "Mexican industry"? Some firms handle their pollu- tion problems very well. Cemex (Cementos Mexicanos), for example, operates cement plants all over Mexico while setting a world stan- dard for environmental performance.' Cemex's Barrientos facility was the first cement plant in the Americas to receive ISO 14001 certi- GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS fication, and Mexico's Environment Ministry has publicly lauded six Cemex plants for their participation in a voluntary environmental audit program. In the border region, some maquiladora plants are undoubtedly serious polluters. But many specialize in garment and electronics as- sembly operations that are less pollution intensive than heavy man- ufacturing. For example, in 1997 about 80 percent of maquiladora factories were assembly operations (garments, electrical equipment, furniture, auto parts, etc.), 15 percent were in general manufactur- ing, and 5 percent produced chemicals.2 In fact, the worst industrial pollution in Judrez has no direct link to NAFTA or maquiladora production.3 It comes from small brick kilns, which belch filthy smoke from combustion of scrap wood, old tires, used motor oil, and sawdust laden with toxics (Figure 4.1). The kilns evoke Mexico's historic poverty, not its new prosperity. Most brick makers live near their work sites in cardboard or scrap wood shanties, in which families crowd five to a room. Some 40 percent of the house- holds report the death of at least one child. Even the kiln owners aver- age only three years of schooling, and a quarter are illiterate. Originally isolated in outlying squatter settlements, the kilns were absorbed by urban sprawl as Juarez expanded. They posed nu- merous air pollution hazards for their neighbors, principally from fine particles and carbon monoxide but also from volatile organic compounds, nitrogen oxide, sulfur dioxide, and heavy metals.4 As public awareness of such hazards grew, kiln-related incidents be- came the largest source of community complaints to the Ciudad Juarez environmental authority. Technically, the solution was clear: replace scrap fuels with propane or natural gas. However, fierce price competition prevented fuel switching in the crude brick market. Rapid bankruptcy loomed for any kiln operator who used propane, which cost 28 percent more than scrap fuels despite state subsidies. Switching fuels also entailed acquiring a new burner, learning how to use it, and modifying the kiln itself. Traditional regulation provided little incentive to switch. The local environmental authority was not only understaffed but reluc- tant to confront the brick makers, many of whom were allied with politically powerful organizations. Some 40 percent belonged to an affiliate of the Partido Revolucionario Institucional (PRI), Mexico's dominant political party. Another 19 percent, in the poorest cobo- nias, belonged to the Comite de Defensa Popular (CDP), which had formed to resist the political establishment's attempts to evict squat- 82 KNOWLEDGE, POVERTY, AND POLLUTION Figure 4.1 Fuel Use and Pollution from Kilns Two Options for Firing Bricks Smoke from Scrap Fuel I ' The Clean Option: Gas Firing Source: Corhis; Courtesy of Octavio Chavez, Southwest Center for Environmental Research and Policy (SCERPJ, Instituto Tecnologico y de Estudios Superiores de Monterrey - Campus Ciudad Juarez (ITEMS), and Salud y Desarrollo Comunitario de Ciudad Juarez (SADEC/FEMAP) ters. This tradition of resistance affected attitudes toward the local environmental agency, based partly on the fear that pollution control would bankrupt the kilns and eliminate jobs for over 2,000 of the poorest people in JuArez. Public pressure finally broke the regulatory deadlock in the early 1990s. Action was sparkedi by election of a new municipal president, whose mandate included strong action to reduce pollution from the kilns. He banned dirty fuels and routinely jailed or fined violators. Support came from local and national NGOs, led by the Mexican Federation of Private Community Health and Development Associa- tions (FEMAP). FEMAP and the city authorities launched an aggres- sive public campaign to educate the brick makers and their neigh- bors about the health risks from burning scrap. In private meetings, they persuaded the brick makers' organizations to support conver- 83 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS sion to propane. Sensing an important business opportunity, local propane suppliers offered the kiln owners free equipment to encour- age fuel-switching. Engineers from local universities provided free technical assistance, and engineers from the El Paso Natural Gas Company suggested more efficient kiln designs. Paradoxically, the controversy over NAFTA also encouraged rapid progress, as transborder pollution became an important pawn in the struggle for public opinion. Anxious to promote the free-trade agreement, the U.S. Government offered technical support for the switch to propane. Motivated partly by NAFTA and partly by rising public interest in pollution control, the Mexican Government and its local PRI affiliate tacitly supported the municipal president's cam- paign against kiln-based pollution. From 1990 to 1992, these forces converged to transform small- scale brick making in Juarez. Over half of the city's 300 brick mak- ers switched to propane firing, and pollution from the kilns dropped sharply. This transformation resonated far beyond Juarez, as it demonstrated the feasibility of pollution control by small-scale in- dustry in the informal economy. However, the triumph was short- lived, as Mexico began eliminating subsidies for basic commodities such as propane under its economic reform program. From 1992 to 1995, the 28 percent cost differential between propane and scrap fir- ing leaped to 162 percent. Bankruptcies and unemployment loomed as the remaining scrap-fired kilns easily undercut their cleaner competitors, and this seismic shock destroyed the community consensus that had sup- ported conversion to propane. As support from community groups and brick makers' organizations evaporated and the municipal gov- ernment dropped its punitive enforcement stance, the clean kilns quickly retreated to scrap firing. Still, community awareness of pollution damage remained, with residents redirecting their efforts into formal and informal pressures to discourage use of the dirtiest scrap fuels-tires, battery cases, and used motor oil. A positive response from many brick makers has kept emissions below their 1990 level, albeit higher than the level after propane conversion. The Ciudad Juarez story illustrates two ways that public action can reduce industrial pollution. First, environmental education and political mobilization can lead local communities to raise marginal expected pollution penalties (MEP; Figure 4.2). In Judrez, the com- munity fought pollution through formal channels, by filing com- plaints against the kilns and electing a municipal president who 84 KNOWLEDGE, POVERTY, AND POLLUTION Figure 4.2 Mexican Brick Makers in the '9Os: MAC vs. MEP The Ciudad Juarez Experience Pollution Decrease Pollution Increase (1990-92) (1993-95) | | MAC Decrease MEP Increase EMAC Increase Pcrea Free Equipment Rising Income ||Propane Price Increase Job Concerns Technical Extension Rising Population NAFTA Signed Improved Efficiency NAFTA Unsigned Health Education Political Pressure $ Local Pressure $ MAC MAC >E EP 1992 1990 1992 1995 1990 Pollution/Output tightened enforcement. Informally, community-based NGOs pres- sured the brick makers' organizations to support conversion to propane. Reversal occurred when national decontrol of propane prices raised marginal abatement costs (MAC) again, but MEP didn't revert to its 1990 level because environmental awareness had grown. Governments can also cut pollution by providing support for conversion to clean production, thereby lowering marginal abate- ment costs. This approach is a promising alternative to traditional regulation because it wields a carrot rather than a stick. It may also reduce pollution more cheaply than enforcement of formal regula- tions, which requires monitoring, data analysis, and, if necessary, police action-all time-consuming and expensive ways to raise MEP. However, many economists have maintained that public inter- vention to curb pollution should stop at the factory gate, for a simple and sensible reason: Factory managers know their operations better than regulators, who should simply establish incentives. When CETESB, Sao Paulo's pollution control authority, launched an ag- gressive campaign to reduce pollution of the Tiete River, program managers attributed over 50 percent of the improvement to changes in manufacturing processes rather than installation of pollution con- 85 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS trol equipment.5 Environmental agencies do not typically promote such changes directly. This arm's-length regulatory model works effectively when well- informed factory managers can easily revamp their technology as in- centives shift. But what happens when plant managers face high conversion costs with little certainty about outcomes? One response might be that fast learners will adjust and expand while others will suffer losses and leave the business. This approach has surface ap- peal, but it didn't survive the test of experience in Ciudad Juarez, as concern for jobs prompted the community to assist many kiln own- ers in taking the initial steps toward cleaner production. Polluters often must make even more fundamental changes in the way they do business to curb emissions. Is technical assistance useful and feasible in more complex situations? An affirmative re- sponse requires credible evidence that changes in plant management can significantly reduce pollution, and that governments can help firms make the needed changes at affordable cost. Recent research from Mexico suggests that developing countries can satisfy both conditions. 4.1 Helping Firms Adopt Environmental Management Certification by the International Standards Organization (ISO) offers one vehicle for pursuing such initiatives. The ISO certifies in- ternational auditing firms, which in turn scrutinize the quality of fac- tories' processes according to ISO guidelines. ISO-certified businesses, especially those seeking rapid growth in the international market- place, enjoy a competitive advantage because they can assure poten- tial customers that they maintain high quality standards. Many lead- ing firms prefer subcontractors that have satisfied ISO requirements. The latest benchmark, ISO 14001, includes new standards for environmental management systems (EMS) based on thousands of firm-level case histories. According to this benchmark, plants must take the following EMS steps to achieve ISO certification: * Perform an initial managerial review to identify environmen- tal issues of concern, such as excessive use of polluting inputs and the potential for a serious environmental accident; u Establish priorities for action, taking into account factors such as local environmental regulations and potential costs; * Establish an environmental policy statement, signed by the CEO, that includes commitments to compliance with environ- 86 KNOWLEDGE, POVERTY, AND POLLUTION mental regulations, pollution prevention, and continuous improvement; * Develop performance targets based on the policy statement (such as reduction of emissions by a set amount over a de- fined period); * Implement the environmental management system (EMS) with defined procedures and responsibilities; and * Measure performance and conduct management audits. Since the ISO published the preliminary version of 14001 in 1996, factories in both developed and developing countries have rushed to obtain certification. Some 8,000 plants achieved certifica- tion by January 1999.6 Asia leads other developing regions, but Latin America has also seen significant activity, and three African coun- tries (Egypt, Morocco, and South Africa) are represented (Figure 4.3, Figure 4.3 International Diffusion of ISO 14001 ISO 14001 Certifications, 1999 Country Indices by Region* 2300 Western Europe Eastern Europe North America 100 20 80 10 100- 0~~~~~~~~~~~~~4 0~~~~~~~ 30 20 0 Ai 10 0 - 0 20 40 Latin America 10 20 0 Source: ISO World 87 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Table 4.2).7 The developed world is split: Western Europe and Japan lead in certifications, while the United States and Canada lag behind many newly industrialized countries that depend heavily on interna- tional trade. A recent World Bank study examined a large sample of Mexican factories to determine whether factories that adopt ISO 14001 proce- dures reduce pollution (Box 4.1) 8 Almost 50 percent of these facto- ries had adopted few or no procedures necessary for ISO certifica- tion, while 18 percent had completed most or all of the required steps (Table 4.1). The research shows that process is important: Plants that have completed most of the ISO 14001 steps comply with pollution regulations far more than plants that have completed few steps.9 We also investigated the degree of "environmental mainstream- ing": whether the plants had assigned environmental responsibilities to general managers rather than specialized managers, and whether the plants provided environmental training for all workers as well as environmental personnel. The plants varied substantially in their adoption of such practices (Table 4.3). But our research shows that mainstreaming works: Providing environmental training for all plant personnel and assigning environmental tasks to general managers help plants comply with regulations. Large vs. Small Plants The Mexico survey suggests that targeting assistance to help smaller plants adopt such procedures would yield the biggest payoff. That's partly because large factories are much better equipped to monitor their own pollution (Figure 4.4). The survey shows that only 5 percent of large plants lacked that capacity. Some 60 percent of small plants could not monitor their air pollution or hazardous waste, while 40 percent lacked the ability to monitor water pollution. 10 Differences in plant size also affect the rate at which plants adopt ISO-type EMS procedures. Large branch plants with well- educated workers score 70 points higher on our EMS adoption index (Table 4.1) than small, individually owned plants operating with lower-skilled labor. The Mexico study reveals the consequences of these variations for environmental performance: The regulatory compliance rate is only about 25 percent for small, individually owned plants such as the brick kilns of Ciudad Juarez, whose workers have little educa- tion (Figure 4.5). By contrast, the compliance rate is over 70 percent 88 KNOWLEDGE. POVERTY, AND POLLUTION Box 4.1 Environmental Management and Regulatory Compliance in Mexico To learn more about the role of environmental Figure B4.1b EMS and Compliance management in promoting compliance with pollution regulations, the World Bank partici- Compliance in Mexico pated in a survey of Mexican industry with a 90 team of regulators, academics, and industrial- 80 ists."1 This team conducted confidential inter- 70 views at 236 representative factories in all size .73 60 classes for four key polluting sectors: food, 50 chemicals, nonmetallic minerals, and metals. 40 Figure B4.1a shows that about half of the l E 30 factories regularly fail to comply with Mexican regulations. The study investigated the rela- 20 tionship between compliance and four mea- 10 sures of plant-level environmental manage- 0° >75 25-75 <25 Yes No Yes No Yes No ment: (1) percent completion of steps like AdoptionScore: Environmental Main- Main- those necessary for ISO 14001 EMS certifica- Environmental Personnel? streaming streaming Management Management? Labor? tion, (2) assignment of personnel to environ- System mental tasks, (3) assignment of environmental Source: Dasgupta, Hettige, and Wheeler (1997) Figure B4.1a Mexican Polluters tasks to general managers rather than use of Mexican Factories: specialized managers, and (4) environmental Compliance with Pollution Regulations training for workers other than specialized Compliant Not Compliant personnel. 120 pronl The results (Figure B4.1b) highlight the 100 i importance of environmental management and training, particularly the adoption of ISO 80 s 14001-type procedures. Some 86 percent of aL plants with high EMS adoption scores comply 0 60 CL * 0 }with regulations, while only 24 percent of 'A 40 - plants with low scores comply. Plants that have assigned personnel to environmental 20 tasks report much higher compliance than oth- o 0 , * 1 1|1 , *, - | ers (58 percent vs. 34 percent), as do plants Very Good Fair Poor Very that have mainstreamed environmental con- Good Bad cerns among managers (71 percent vs. 47 per- Source: Dasgupta, Hettige, and Wheeler (1997) cent) and workers (59 percent vs. 34 percent). 89 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Table 4.1 Adoption Index for ISO 14001 Procedures by Mexican Factories Adoption Score Number of Plants Percent S < 25 111 47.0 25 < S<50 45 19.1 50 < S < 75 38 16.1 75 < S < 100 42 17.8 Source: Dasgupta, Hettige, and Wheeler (1997) Table 4.2 ISO 14001 Certification, 1999 by Country and Region Region/ Number Index* Region/ Number Index* Region/ Number Index* Country Certified Value Country Certified Value Country Certified Value Latin Africa America W Europe Egypt 15 21 Costa Rica 2 22 Denmark 300 175 South Africa 21 16 Argentina 37 12 Sweden 400 172 Morocco 2 6 Brazil 65 8 Ireland 80 121 Mexico 27 8 Finland 130 105 Asia Chile 4 6 Switzerland 292 93 Korea 463 95 Uruguay 1 5 Austria 180 80 Malaysia 80 82 Colombia 3 4 UK 950 78 Thailand 100 59 Peru 1 2 The Netherlands 300 75 Singapore 60 59 E Europe Germany 1100 47 Japan 1542 32 . Belgium 120 45 Philippines 23 26 Hungary 31 69 Norway 60 38 Hong Kong 40 24 Slovak Rep. 8 40 Spain 116 20 Turkey 40 20 Slovenia 6 31 France 177 12 Indonesia 43 19 Czech Rep. 12 22 Italy 100 9 India 60 16 Croatia 3 15 Portugal 7 7 China 60 7 Poland 8 6 Greece 6 5 Pakistan 2 3 Rumania 1 3 Oceania Russia I 1 North America Mauritius 2 47 Canada 90 15 New Zealand 27 45 USA 210 3 Australia 130 34 * Index = (Number Certified)/GDP, standardized to the range 1-200. 90 KNOWLEDGE, POVERTY, AND POLLUTION Table 4.3 Mainstreaming Environmental Management in Mexican Factories Management strategy Yes % No % Environmental training for nonenvironmental workers? 76 32.6 157 67.4 Environment manager also has other responsibilities? 211 93.8 14 6.2 Source: Dasgupta, Hettige, and Wheeler (1997) for large branch plants of publicly traded firms, which employ many workers with secondary-school education. As Figure 4.5 shows, plant size, firm size and worker education contribute roughly equally to compliance. Since many large plants already have EMS capability, govern- ment promotion of environmental management should focus on small- and medium-scale enterprises (SMEs). But from a public pol- icy perspective, doing so makes sense only if SMEs actually adopt EMS procedures, and if such intervention reduces pollution more cheaply than conventional regulation. For evidence on this issue, we turn to another recent Mexican project. Figure 4.4 Plant Size and Monitoring Capacity Small vs. Large Plants Little or No Ability to Monitor: Air Hazardous Water Pollution Waste Pollution 61% 59% 40% * Small Plants * Large Plants 4% 1 5% 6% Source: Wells (1996) 91 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 4.5 Plant Size and Compliance in Mexico Compliance in Mexico: Small vs. Large Plants 100 s0 Factors 60 S 40 E 20 0 iAjY Small Large Source: Dasgupta, Hettige, and Wheeler (1997) Lessons from Guadalajara Monitored experiments in promoting EMS within plants remain rare; one notable exception is a recent project in Guadalajara, Mexico, that tested whether SMEs could successfully adopt environmental management systems. Eleven large companies, many of them multi- nationals, agreed to provide assistance to 22 small- and medium-scale suppliers who were interested in improving their environmental per- formance. The project, which enlisted the private sector, local aca- demic institutions, the Mexican Government, and the World Bank, entailed several two-month cycles of intensive training, implementa- tion, and review sessions.12 After nine months of implementation, the 15 SMEs remaining in the project rated their degree of EMS adoption on a 20-point scale. In May 1997, the average adoption score was effectively zero. By Feb- ruary 1998, average scores had increased to around 16 points for en- vironmental planning and 11 points for EMS implementation. About 80 percent of the plants reported lower pollution, and nearly 50 per- cent reported improved compliance and waste handling. Many also reported improved work environments, more efficient use of materi- als, and better overall economic performance (Figure 4.6). 92 KNOWLEDGE, POVERTY, AND POLLUTION Figure 4.6 Results From Adoption of ISO 14001 Small Plants Reporting Improvements: Guadalajara Project Environmental Improvements Pollution Control Compliance Waste Handling Other Improvements Work Environment Economic Performance Input Use Efficiency 0 20 40 60 80 100 Percent of Total Plants (15) Source: Ahmed, Martin, and Davis (1998) The Guadalajara project showed that SMEs can successfully adopt EMS-with assistance. These plants will probably sustain the changes, because the project covered the fixed costs of adjusting to an environmental management system; the incremental costs of op- erating an EMS should be much lower. Participating SMEs have also altered their internal communications to provide constant feedback on environmental problems and solutions. Once formed, this new business culture will not easily disappear. The Guadalajara project reduced pollution by lowering marginal abatement costs (MAC) rather than raising marginal expected pollu- tion penalties, as in conventional regulation. The evidence suggests that this alternative approach was cost competitive. The project's total cost was around $200,000. Because about 10 plants realized significant benefits, the unit cost was approximately $20,000 per plant."3 To compare the project (an investment) with conventional regulation (an annual flow of costs), we use the discounting ap- proach. We employ a 10 percent discount rate and assume that a re- duction in MAC from EMS adoption is sustainable. This yields an annualized cost of $2,000 per factory in perpetuity. To estimate the cost of future projects in Guadalajara, we as- sume that local consultant fees are about 25 percent of fees charged by the international consultants in the pilot program. The implied 93 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS unit cost is $5,000 per plant, or $500 in annualized costs-roughly equivalent to a month's wages for a skilled worker in urban Mexico. To achieve similar results each year, conventional regulation would almost certainly require equal or greater time and costs devoted to monitoring, record keeping, and enforcement. What's more, tradi- tional regulation is less likely to yield the economic benefits of EMS stemming from greater overall efficiency. We conclude that promot- ing EMS among SMEs compares favorably with attempting to regu- late these enterprises by conventional means. Because they have been monitored and evaluated systemati- cally, both the Ciudad Judrez project and the Guadalajara project provide important new information about the power of government- supported learning to reduce pollution by SMEs. The Ciudad Juarez project revealed the feasibility of improving pollution control in informal-sector, low-technology enterprises operated by some of urban Mexico's poorest, least-educated workers. Farther south, the Guadalajara project showed the feasibility of encouraging EMS adoption by somewhat more sophisticated small- and medium-scale subcontractors to large firms. In both Ciudad Juarez and Guadala- jara, project funds financed the development of local consulting skills that will contribute to future pollution-reduction initiatives in the private sector. 4.2 Who's Complaining about Pollution? In Ciudad Judrez, effective regulation required feedback to kiln owners from the surrounding community. This is not an isolated case. Despite their putative independence, pollution control agencies respond to the demands of the political institutions that determine their budgets and, ultimately, their legitimacy. Political leaders, in turn, respond to citizen complaints from communities affected by pollution. Regulators also have an administrative reason to heed commu- nity protests: Monitoring is costly and agencies' budgets are lean in developing countries, so they cannot remain fully informed about all polluters. As a result, regulators often focus resources on responding to citizen complaints. For example, the pollution control agency of Brazil's Rio de Janeiro State devotes nearly 100 percent of its inspec- tion resources to complaints. After setting aside 50 percent of its re- sources for targeting priority polluters, Sao Paulo's pollution control agency allocates the remainder to complaints. In Indonesia, the na- 94 KNOWLEDGE, POVERTY, AND POLLUTION tional pollution control agency has few inspectors but allocates much of their time to complaints. China's provincial and local regu- lators respond annually to more than 100,000 citizen complaints.14 Although complaints are a valuable source of low-cost informa- tion, regulation based on complaints may suffer from serious biases. Plaintiffs may lack enough information to distinguish between "nui- sance" emissions and those that are truly hazardous. Colorless, odorless toxics and heavy metals may escape notice altogether. Fur- thermore, some individuals or communities may simply complain more, regardless of the objective situation. If regulators automati- cally respond to complaints, aggressive plaintiffs may capture most of the available resources. World Bank researchers have used new data from China to learn more about the links between complaints, environmental conditions, and the characteristics of communities.15 China's citizens are far from passive about pollution from neighboring factories. From 1987 to 1993 the nation's environmental authorities recorded over 130,000 complaints per year, mostly related to air, water, and noise pollution. These complaints generated action: In almost all provinces, agency response rates varied from 70 to 100 percent. Responding to so many complaints absorbed much of regulatory inspectors' time. But Figure 4.7 shows that the propensity to complain varies con- siderably across China's provinces. In 1993, Shanghai and Tianjin reported some 30 complaints per 100,000 individuals, but Gansu, Xinjiang, and Inner Mongolia reported less than 5 complaints per 100,000. The incidence of complaints was generally highest in the urban industrial centers of east China, lower in the middle provinces, and lowest in the western hinterlands, the least-developed region. To assess the factors influencing these provincial differences, we analyzed the impacts of pollution, income, and education on the in- cidence of complaints. Holding income and education constant, we found that provinces with high emissions reported 75 percent more complaints than lightly polluted provinces. However, this effect ap- plied only to highly visible pollutants such as particulates. Holding pollution and education constant, we found complaints in high-income provinces 110 percent more numerous than in low- income provinces. Another recent study shows that such complaints induce local regulators to raise the air pollution levy, thereby lower- ing industry's air pollution intensity.16 Box 4.2 illustrates the conse- quences for 50 Chinese cities: Richer areas that complain more have significantly cleaner air.17 95 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 4.7 Regional Distribution of Complaints 100,000 Population a i csn we f 16 to 32_ Source: Dasgupta and Wheeter (1996) Holding pollution and income constant, we found the complaint rate 90 percent higher in poor provinces with high literacy than in poor provinces with low literacy. Rich provinces with different liter- acy rates also varied in their incidence of complaints (Figure 4.8). Overall, the impact of literacy on citizen complaints is striking- roughly equivalent to a doubling of income or a tenfold increase in air pollution. Why the silence from low-literacy regions, even if their average incomes and environmental conditions are comparable to those in other areas? Contributing factors may include poorer information about local pollution and health damage, lower capacity to organize for political action, and reluctance to confront better-educated offi- cials. Because complaints drive much of the regulatory process, this silence has serious environmental consequences. The same effects have emerged in other countries. One study found that economic development, a proxy for income and literacy, strongly affects regulatory inspections of Indian factories. Another study documented the positive relationship between economic devel- opment and pollution control by paper mills in Thailand, India, Bang- ladesh, and Indonesia.18 A World Bank study found that pollution- intensive activities cluster in poor areas of Brazil. And in a study of 96 KNOWLEDGE, POVERTY, AND POLLUTION Box 4.2 In China the Poor Get More ... Pollution Poverty is a curse, but the poor might hope for sity rises from poor to middle-income cities and at least one compensation: Since richer areas then falls to the lowest levels in the richest have more industrial production, they might areas. well have more pollution. Unfortunately, the in- Why this tragic association between habitants of China's poor regions have no such poverty and pollution? Richer areas do have consolation. Figure B4.2 displays the relation- higher levels of industrial production, but pro- ship between average wages and emissions den- duction in those areas is also much cleaner be- sity (or emissions per unit area, a reasonable cause citizen feedback is stronger (Figure 4.8) proxy for atmospheric pollution concentration) and regulation is tighter. Also, industrial facili- in 50 Chinese cities. For suspended particulates ties in areas with unskilled workers generally (TSP), pollution density clearly rises as wages operate at lower efficiency and create more fall. The poorest cities also have higher densi- waste. The scale of industrial production in a ties of SO2 emissions than the richest cities, al- given region therefore plays second fiddle to though the graph suggests that emissions den- pollution intensity in China. Figure 64.2 Income and Air Pollution TSP Pollution vs. Income in 50 Chinese Cities S02 Pollution vs. Income in 50 Chinese Cities 3 3 2.5 3.5 t.5 3 2.5 22 * ~~~ +2 a ,.-~~~.... -4 4$. ;+ Log - ~ I.5 Wg *og W p 0.5 0.5 0 ol 3.35 3.45 3.55 3.65 3.75 3.85 3.95 3.35 3.45 3.55 3.65 3.75 3.85 3.95 Log Wage Log Wage Source: SEPA Indonesia, researchers found that factories in municipalities in the bot- tom quartile of income and postprimary schooling have organic water pollution intensity 15 times the level of plants in communities in the top quartile of income and education.19 Recent research on Brazil found that although pollution-inten- sive activities are more prevalent in poor municipalities, overall pol- lution problems are much worse in affluent municipalities because the volume of industrial production is much larger (Box 2.2).20 97 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 4.8 Literacy and Complaints China: Poverty, Literacy, and Pollution Complaints 25 - High 20 - Literacy L (90%) 15- O 10- _ ~Literacy _ 10 t 84%)_ 5 -3 0 Poor Rich Region Nevertheless, studies of Rio de Janeiro and Sao Paulo have shown that pollution-intensive industry dominates the poorer municipali- ties within these metropolitan regions. Even in cases where richer areas are more polluted, poor residents of those regions suffer the greatest exposure. 4.3 Redefining Environmental Injustice In the United States, the belief that polluters in poor, uneducated communities face laxer regulation and operate at higher levels of pollution intensity has catalyzed a political movement for environ- mental justice. Its goal is equal environmental quality for all citi- zens, regardless of their income, education, or ethnicity. The concept of environmental justice has strong intuitive appeal for developing countries as well, as so many poor citizens suffer from pollution. Yet what appears to be clear evidence of environ- mental injustice can also reflect broader economic problems. For ex- ample, environmental problems can reflect the economics of loca- tion. Decades of research have shown that urban land values are affected by a variety of factors, including residents' exposure to pol- lution. Both land values and housing rents are lower in polluted 98 KNOWLEDGE, POVERTY, AND POLLUTION areas, which are also likely to be areas where industry concentrates. Cheaper housing and industrial job opportunities give the poor a double incentive to locate in more polluted areas, even if they are fully informed about health risks. Social, political, and historical factors have generated severe in- come inequality in many developing regions. Yet where poverty is the root cause of pollution exposure, invoking environmental injustice can backfire. Suppose, for example, that a movement against envi- ronmental injustice targets a large polluting factory located in the middle of a poor residential area. Neighboring families are well aware of the pollution, but they have located there because rents are cheap and the factory offers jobs for semiskilled workers. The environmen- tal justice movement succeeds, and the plant's managers reduce pol- lution by shifting to a process that uses preassembled components and requires more educated labor. The air and water in the surround- ing neighborhood become noticeably cleaner, and pollution-related illnesses fall.21 After the victory celebrations, however, other changes occur. Be- cause the area is much cleaner, land values and rents rise. The poor- est residents have little choice but to pack their belongings and move, since they can no longer afford shelter in the area. Their job opportunities also decline, because the factory no longer needs much semiskilled labor. Some of their neighbors accept the higher rents, but they have to find new jobs in other parts of the city. To continue working, they spend more hours jammed into buses and other vehicles, risking their lives on congested, polluted roads. In the aftermath of the movement's supposed success, its intended beneficiaries are worse off, because it has confused an income- inequality problem with environmental injustice. What, then, is environmental injustice? In our view, two differ- ent concepts are well worth considering. First, the government could assume responsibility for maintaining a minimum decent standard of environmental quality for all citizens. This would resemble a public commitment to universal primary education. Failure to maintain the minimum decent standard would be defined as environmental injus- tice, warranting corrective action. This would include conventional pollution regulation, but could also mean programs that promote ac- cess to safe water and basic sanitation for poor communities. A second concept would apply where people suffer from pollu- tion because of ignorance as well as poverty. Governments should provide environmental education to all communities, so failure to 99 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS inform poor neighborhoods about dangerous pollution would qual- ify as injustice. The Ciudad Juarez case shows how effective public education can be. Very poor people in the colonias supported regula- tion of the kilns after an education campaign persuaded them that the health benefits would compensate for the additional risk of un- employment and higher rents. The program worked because it tar- geted serious polluters and provided assistance for their conversion to propane firing. Better-informed colonias continued to support cleaner production after propane price decontrol, but concern for jobs shifted their focus to pressure for cleaner scrap fuels. Ciudad Judrez demonstrated the power of public education to promote environmental change even when poverty remains unal- tered. In our view, this is an important arena in which the struggle for environmental justice should be joined. References Afsah, S., B. Laplante, and D. Wheeler, 1997, "Regulation in the In- formation Age: Indonesian Public Information Program For En- vironmental Management," World Bank, March. Ahmed, K., P. Martin, and S. Davis, 1998, "Mexico: The Guadalajara Environmental Management Project," World Bank, September. Blackman, A., and G. Bannister, 1998a, "Pollution Control in the In- formal Sector: The Ciudad Judrez Brickmakers' Project," _at- ural Resources Journal, Vol. 37, No. 4, 829-56. 1998b, "Community Pressure and Clean Technology in the Informal Sector: An Econometric Analysis of the Adoption of Propane by Traditional Mexican Brickmakers," Journal of Envi- ronmental Economics and Management, Vol. 35, No. 1, 1-21. Chavez, O., 1995, "Alternative Fuels for Brick-Makers, CD Juarez, Mexico Project," Southwest Center for Environmental Research and Policy (SCERP). Reproduced from the SCERP Web site at http://www.civil. utah.edIu/scerp/brickmakerlbrickmaking. html. Dasgupta, S., H. Hettige, and D. Wheeler, 1997, "What Improves En- vironmental Performance? Evidence from Mexican Industry," World Bank Development Research Group Working Paper, No. 1877, December. Dasgupta, S., R. Lucas, and D. Wheeler, 1998, "Small Plants, Pollu- tion and Poverty: Evidence from Mexico and Brazil," World Bank Development Research Group Working Paper, No. 2029, November. 100 KNOWLEDGE, POVERTY, AND POLLUTION Dasgupta, S., H. Wang, and D. Wheeler, 1997, "Surviving Success: Policy Reform and the Future of Industrial Pollution in China," World Bank Policy Research Department Working Paper, No. 1856, October. Dasgupta, S., and D. Wheeler, 1996, "Citizen Complaints As Envi- ronmental Indicators: Evidence From China," World Bank Policy Research Department Working Paper, No. 1704, November. Hamson, D., 1996, "Reducing Emissions from Brick Kilns in Ciudad Juarez: Three Approaches," Border Environment Research Re- ports, No. 2, Southwest Center for Environmental Research and Policy (SCERP), June. Reproduced from the SCERP Web site at http://www.vcivil.utah.edu/scerp/docs/berr2.html. Hartman, R., M. Huq, and D. Wheeler, 1997, "Why Paper Mills Clean Up: Determinants of Pollution Abatement in Four Asian Countries," World Bank Policy Research Department Working Paper, No. 1710, January. Hettige, H., M. Huq, S. Pargal, and D. Wheeler, 1996, "Determinants of Pollution Abatement in Developing Countries: Evidence from South and Southeast Asia," World Development, Vol. 24, No. 12, 1891-1904. Ostro, B., 1994, "The Health Effects of Air Pollution: A Methodology With Applications to Jakarta," World Bank Policy Research De- partment Working Paper, No. 1301, May. Pargal., S., M. Huq, and M. Mani, 1997, "Inspections and Emissions in India: Puzzling Survey Evidence on Industrial Water Pollu- tion," World Bank Development Research Group Working Paper, No. 1810, August. Pargal, S., and D. Wheeler, 1996, "Informal Regulation of Industrial Pollution in Developing Countries: Evidence From Indonesia," Journal of Political Economy, Vol. 104, No. 6, 1314 +. Petzinger, T., 1996, "Mexican Cement Firm Decides to Mix Chaos into Company Strategy," Wall Street Journal, December 13. Stotz, E., 1991, "Luta Pela Saude Ambiental: A AMAP Contra Cor- tume Carioca, S.A., Una Experiencia Vitoriosa," V. V. Valla and E. N. Stotz (eds.) Participacao Popular, Educacao e Saude, Rio de Janeiro, 133-60. 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, No. 1644, September. 1999, "China's Pollution Levy: An Analysis of Industry's Re- sponse," presented to the Association of Environmental and Re- 101 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS source Economists (AERE) Workshop, "Market-Based Instru- ments for Environmental Protection," John F. Kennedy School of Government, Harvard University, July 18-20. Wells, R., 1996, "Prevencion y Control de la Contaminaci6n en la In- dustria Mexicana: Reporte de Una Encuesta," (Lexington, Mass: The Lexington Group), December. 1. See Petzinger (1996) for a report on Cemex's innovative role in Mexican business. 2. Statistics reported by Mexico's Secretariat for Commerce and Industrial Promotion (SECOFI) at http://www.nafta-mexico.org/ export. hm. 3. Our discussion of the Ciudad Juarez case draws heavily on two papers by Allen Blackman and Geoffrey Bannister (1998a,b), which document the authors' extensive primary research and econo- metric analysis. Our thanks to Allen Blackman for additional discus- sion of Ciudad Juarez in a series of personal communications. For additional information about the brick kiln pollution problem, see Hamson (1996) and Chavez (1995). 4. Blackman and Bannister (1998a) cite a study of brick makers in Saltillo, Mexico, that finds that 47 percent of subjects tested had "abnormal" pulmonary functions. See Ostro (1994) for detailed dis- cussion of the impacts on health of particulates. 5. Interviews with CETESB staff. 6. Reinhard Peglau of the Federal Environmental Agency, Repub- lic of Germany, provided these estimates. They are reproduced by 1SO World at http://www.ecology.or.jp/isoworld/english/analy14k.htm. 7. To control for extreme scale differences, such as between Costa Rica and China, we divided each country's total certifications by its GDP, and standardized the result on a scale from 1 to 200. 8. See Dasgupta, Hettige, and Wheeler (1997). 9. The study recognizes the possibility of reverse causation: Once other factors have convinced plant managers to comply with regulations, they could implement the ISO 14001 steps as part of the improvement process. However, this would not imply that the ISO 14001 steps caused the improvement. The research (Dasgupta, Het- tige, and Wheeler, 1997) uses standard econometric techniques to adjust for this problem. 10. See Wells (1997). 102 KNOWLEDGE, POVERTY, AND POLLUTION 11. Participants included Mexico's National Environment Min- istry (SEMARNAP), the Monterrey Institute of Technology, and the Mexican National Association of Industries. Both the survey ques- tionnaire and the data are available online at http://www.woorld- bank.org/nipr/work-paper/l 877/suruey/index.htm. 12. This summary is based on a World Bank report by Ahmed, Martin, and Davis (1998). 13. The final project report estimates the cost of the pilot project to be $135,000, excluding the time and travel costs of World Bank staff. Including these factors would increase the estimated cost to around $200,000. A second approach would apportion the cost among all 15 par- ticipating plants, lowering the apparent cost per plant. However, measured benefits were apparently zero for 5 plants, so this ap- proach would yield the same result as the first, because the expected value of results per plant would be lowered proportionately. 14. Complaint-response systems in Brazil, China, and Indonesia are familiar to the authors from collaborative work with FEEMA, CETESB, BAPEDAL, and China's State Environmental Protection Agency (SEPA). 15. See Dasgupta and Wheeler (1996). Data linking environ- mental conditions, pollution complaints, and community character- istics in China are available online at http://www.worldbank.org- nipr/data/china/status. htm #Province. 16. See Wang and Wheeler (1999). 17. See Wang and Wheeler (1996) and Dasgupta, Wang, and Wheeler (1997). 18. See Pargal, Huq, and Mani (1997) and Hartman, Huq, and Wheeler (1997). 19. See Dasgupta, Lucas, and Wheeler (1998) and Pargal and Wheeler (1996). 20. See Dasgupta, Lucas, and Wheeler (1998). 21. In extreme cases, the targeted plant may simply move away. Stotz (1991) describes such a case for a tannery in Rio de Janeiro, Brazil. Environmental regulators in Rio have reported to the authors that middle-income residents led the movement against the plant; lower-income families were far more reluctant to act, because they valued the tannery as a source of employment. 103 - t | s~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -~~~~~~~~~usd an Inid Cuato Bai'SerdoMrnth 9 |~ ~ ~ ~ ~~~~~~~~~ore Calo ReaoFradscndEoPrn;C i National Economic Policies: Pollution's Hidden Half S outheast of Sao Paulo, Brazil's continental plateau crests in the Serra do Mar before dropping off sharply to the sea. Along the crest stand remnants of Brazil's fabled Atlantic Forest, one of the world's most diverse and threatened ecosystems. As the highway from Sao Paulo winds down the coastal scarp to Santos, the region's major seaport, it encounters a small, quiet river. Upstream and sur- rounded on three sides by mountains lies the industrial city of Cubatao. In the 1980s it was known as the "Valley of Death." In the go-go days of Brazil's state-led development, the valley's location made it irresistible to industrial planners. Near the port of Santos, it was a perfect place for industries like steel, petroleum, fer- tilizer, and chemicals to turn imported heavy raw materials into fin- ished products before shipping them to Sao Paulo via the long climb uphill. The river provided a source of water as well as a convenient place to dump wastes. Led by huge state corporations like COSIPA (steel) and PETRO- BRAS (oil), the Cubatao valley mushroomed into an industrial com- plex so large that it accounted for 3 percent of Brazil's GDP by 1985. Employment boomed for immigrants from Brazil's poor regions, and the future looked bright-except for two unfortunate lapses by na- ture. The small, quiet-flowing river was no match for the torrent of industrial wastewater, and the valley was a natural trap for air pollu- tion. Undeterred by local regulators, state-owned mills and their pri- vate counterparts spewed a thousand tons of pollutants into the air GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS every day. In the early 1980s the city recorded the highest infant mortality rate in Brazil, and over one-third of the residents suffered from pneumonia, tuberculosis, emphysema, and other respiratory sicknesses. By 1984, the Cubatao River was basically dead from or- ganic pollution. Downstream from Cubatao, tons of heavy metals ac- cumulated in bottom sediments and washed into the sea near San- tos. Above the valley, fallout from air pollution began killing the Atlantic Forest and denuding the mountainsides. Finally, in January 1985 the crisis became a catastrophe, as 15 inches of rain poured onto the bare hillsides in 48 hours. Hundreds of mudslides cascaded into the valley, and one broke a large ammo- nia pipeline in Vila Parisi, releasing gas that injured many residents and forced a mass evacuation. Official denial ended as the governor of Sao Paulo State declared an emergency and mandated forceful ac- tion by CETESB, the state's pollution control agency.1 Fifteen years later much has changed in the Cubatao Valley. It is still no paradise, but environmental conditions are typical of medium-sized industrial cities in Brazil. The Atlantic Forest is grow- ing back, sunny days are a reality again, children are healthier, and fish are returning to the Cubatao River, although their tissues are still laced with toxic metals. CETESB deserves great credit for this restoration. Supported by an aroused populace, its actions have made air pollution emergencies rare and cut damaging emissions consider- ably (Figure 5.1).2 Only one obstacle prevented even faster cleanup-the resistance of the state-owned factories that spearheaded the valley's develop- ment. By 1994, state enterprises contributed 42 percent of total sus- pended particulates (TSP) before end-of-pipe control but 77 percent of the actual emissions after control (Figure 5.2). The case of sulfur dioxide (SO2) was similar. State-owned factories were simply doing far less to abate pollution than private mills. Even this substandard performance had required years of targeted inspections by CETESB, embarrassing public disclosures, and shutdown threats. Managers of the state-owned plants stubbornly resisted, complaining about fi- nancial losses and invoking political support at the state and na- tional levels. Things changed abruptly in late 1993 when the government pri- vatized COSIPA, the state-owned steel company. The Cubatao plant shared in the industry's ensuing rapid modernization. From 1990 to 1996, Brazilian steel mills expanded production from 22.6 to 25.2 million tons while doubling output per worker.3 Use of materials is 106 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF Figure 5.1 Air Pollution, 1984-1998 Air Alerts 16 Cleaning Up 12 Cubatao 8 4 0 84 86 88 90 92 Fine Particles (PM1O) S02 180- 60 - j;- 160 - 50 E 140- 4- E 40- 1 12- 30- '~100 - 20- CL 80- 10- 60- 0 - 84 86 88 90 92 94 96 98 84 86 88 90 92 94 96 98 Source: CETESB Figure 5.2 Ownership down, quality is up, and there is widespread interest in the new ISO 14001 quality standard, including its environmental provisions. Emissions in Cubatao, 1994 CETESB also finds it easier to regulate the privatized Cubatao steel stateowned mill. Although Brazil's privatization program had no explicit envi- PrState-owned ronmental goals, it proved to be a godsend for hard-pressed pollu- Private tion control agencies like CETESB. BEFORE AFTER The impact of privatization on Cubatao is not an isolated case: National economic policies affect industrial emissions so strongly j 1 that they constitute "pollution's hidden half." Recent research shows TSP0o that cleaner production generally results from economic reforms- reducing barriers to international trade, privatizing state industries, developing new stock markets, eliminating subsidies for energy and 114,000* 32,000* raw materials, and deregulating domestic industries. However, such A _ reform is not a panacea. Few countries reform their economies for so, environmental reasons, so it would be a peculiar stroke of luck if all S. such actions had clean impacts. In some cases economic reform can actually increase industry's pollution intensity, and the faster growth 29,000* 17,000* sparked by more-open markets magnifies the potential for pollution. [* tons] Fortunately, numerous studies have suggested ways to antici- pate and offset such side effects while economic reform attacks pol- Source: CETESB 107 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS lution's "hidden half" on a broad front. Incomes should also rise with economic reform, increasing public support for formal and in- formal regulation of pollution. However, ensuring lower pollution re- quires close cooperation between economic reformers and environ- mentalists, as well as added resources to help regulators monitor pollution in the wake of reform. 5i1 How Trade Reforin Influences Polluters When developing countries become more "open"-cutting tar- iffs and lowering other barriers to international trade-domestic firms have better access to cleaner manufacturing technologies. Such technologies emerged rapidly in the 1970s because of stricter environmental regulation in the OECD economies.4 In steel makling, for example, continuous casting revolutionized production by elimi- nating energy-intensive intermediate stages, thereby reducing pollu- tion by about 20 percent. The use of electric arc furnaces in the in- dustry also grew rapidly, partly because the process is far less pollution-intensive. In the paper industry, thermomechanical pulp- ing greatly reduced the need for polluting chemicals. Although these new technologies were available on world mar- kets soon after they were developed, weak regulation in developing countries provided little incentive for plants to adopt them for envi- ronmental reasons. However, the new technologies also operated more efficiently than their predecessors. Lowering trade barriers made them cheaper to acquire, and potentially increased domestic firms' interest in more efficient production by opening the market to international competition. To determine whether open developing economies absorbed these technologies more rapidly than closed economies, a World Bank re- search team examined steel and paper production in 50 countries. We found that the open economies led closed economies in adoption of cleaner technologies by wide margins (Figure 5.3), and that the open economies reaped significant environmental benefits. For example, we estimated that more rapid adoption of continuous casting and electric arc technology made steel making in the open economies about 17 percent less pollution-intensive than in the closed economies.5 Some economists have also argued that greater openness to trade can encourage cleaner manufacturing because protectionist countries tend to shelter pollution-intensive heavy industries. In the early 1990s, a World Bank team found that pollution intensity was indeed higher in protectionist Latin American economies than in 108 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF Figure 5.3 Trade Policy and Adoption of Clean Technology Thermomechanical Pulping Continuous Casting Steel 16 90 a14 80- 12 70- Q> 10o F, 6- _ 8 60 ,, 4 40- 0 .. . .. 20 77 79 8183 85 87 89 91 93 95 2079 81 83 85 87 89 Electric Arc Steel 70° 60 ~ 50- 40- (0 OECD OPEN LDCs .2' 30- X 20- CLOSED LDCs 0 10 0 73 75 77 79 81 83 85 87 89 Source: Wheeler, Huq, and Martin (1993) countries with fewer barriers to trade. Another Bank team uncov- ered the same result in a study encompassing all developing coun- tries.6 Recent evidence from China suggests that greater openness to trade has reduced the share of dirty sectors (Box 5.3). However, research at the plant level in Indonesia, India, and Mexico has found that factories with more external trade links do not have lower pollution intensity, nor are they more likely to com- ply with environmental regulations.7 Thus, we can conclude that countries that are more open to trade adopt cleaner technologies more quickly, but individual factories that are export oriented have no special advantage. Open economies also seem to harbor fewer pollution-intensive industries. Of course, growing international trade may stimulate more pro- duction, swamping cuts in pollution intensity and raising a country's total pollution. We will address that problem later in the chapter. 5.2 How Input Prices Affect Posllutiot Eliminating subsidies for energy and raw materials, and break- ing up protected monopolies that produce them, change the prices 109 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS of these goods. Because industry relies heavily on such inputs, changes in their prices also exert significant effects on industrial pol- lution-sometimes in opposing directions. Much previous research, for example, suggests that materials- intensive industries also produce a lot of waste. Thus, eliminating subsidies for raw materials-thereby raising their prices-shifts pro- duction toward processes that use fewer materials and pollute less. (End-of-pipe pollution control also uses material inputs such as chemicals. Increasing their cost makes end-of-pipe control more ex- pensive and thus less likely to be used. Overall pollution declines, however, because process-level waste reduction more than compen- sates for the end-of-pipe effect.)' On the other hand, breaking up monopoly producers of raw materials should increase competition and reduce prices. Companies should respond by using more mate- rials, thereby increasing pollution intensity. Cutting energy subsidies has opposing effects at the plant and in- dustry levels. Several research projects have shown that raising en- ergy prices also tends to raise pollution intensity for individ ual plants (Box 5.1).9 Higher energy prices increase the cost of end-of-pipe treatment, thereby discouraging pollution control. They also induce substitution away from relatively clean, capital- and energy-intensive processes that generate fewer waste residuals. Electric arc furnaces and thermomechanical pulping are examples of such processes. However, raising energy prices tends to lower pollution intensity for industry as ac whole because sectors that process heavy raw ma- terials (and generate most of the pollution) are also heavy energy users. As energy becomes more expensive, demand shifts toward products that are less energy (and pollution) intensive. Higher en- ergy prices also reduce energy demand, lowering production from power plants that are often heavy air polluters. The conventional as- sumption is that these industry-level effects are greater than coun- tervailing plant-level effects, so higher energy prices should reduce overall pollution from industry. However, data scarcity has pre- vented careful research on this issue in developing countries. Even if the overall impact of higher energy prices is beneficial, some local areas may suffer from greater pollution if local factories become much more emissions intensive. In Ciudad Juarez, the impact of propane price decontrol on brick-kiln emissions provided an excel- lent case in point. Figure 5.4 illustrates possible responses to economic reforms at a plant whose managers minimize costs by equating marginal abate- 110 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF Box 5.1 Beyond Anecdotes: Building a Database through Collaborative Research Information scarcity makes research on vided access to information on plant-level industry-and-environment issues difficult in pollution and regulation. The Bank team has developing countries. National and regional developed collegial relationships with regula- environmental agencies are potentially good tory agencies in several OECD countries as sources of information, but their data on in- well. dustrial emissions and compliance are gener- The fruits of collaboration are well illus- ally not in the public domain. The World Bank trated by a cross-country study of industrial has therefore developed a collaborative pro- water pollution intensity cited in this chapter gram with several environmental agencies to (Hettige, Mani, and Wheeler, 1998). For that address the need for better information. For study, the World Bank team acquired informa- the partner agencies, the program provides in- tion from 12 countries (Figure B5.1): formation on international experience with pollution control, and technical assistance on Brazil: CETESB, the environmental designing new regulatory programs and evalu- agency for Sao Paulo State, provided informa- ating program results. For the World Bank re- tion on organic water pollution from its 1,250- search team, the partner agencies have pro- plant database. Figure B5.1 Data for Comparative Research AIL~I Data Source _ B~~~~~~~~~~~~~~~~~~~~~1 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 5.1 (Continued) China: The State Environmental Protec- Emissions Inventory System, which encom- tion Agency (SEPA) supplied information on passes 700 regularly monitored facilities. water pollution from its comprehensive data- Philippines: The Philippine Department of base on major sources of industrial pollution. Environment and Natural Resources (DENR) Our estimates are based on SEPA's 1993 data and the Laguna Lake Development Authority for 269 factories scattered throughout China. provided data on emissions from factories in Finland: The Industrial Waste Water Office the Manila area. of the National Board of Waters and the Envi- Taiwan (China): The Water Quality Pro- ronment provided data on water emissions in tection Division of the Taiwan Environment 1992 from 193 large water-polluting factories. Protection Agency provided data on emissions India: The Tamil Nadu Pollution Control from 1,800 plants. Board, which monitors air and water pollution Thailand: Seatec International, a private for all the manufacturing units in the state, environmental consulting firm in Bangkok, provided plant-level data for 1993-94. provided plant-level data on water emissions Indonesia: BAPEDAL, Indonesia's Na- from 450 facilities in two industrial estates in tional Pollution Control Agency in the Ministry Rangsit and Suksawat in 1992. of Environment, provided data on plant-level Sri Lanka: The World Bank's Metropolitan emissions. Environment Improvement Program and the Sri Korea: The National Pollution Control Lankan Board of Investment supplied data on Agency provided 1991 data on water emissions pollution and employment as part of their study from 13,504 facilities. of wastewater treatment options for the Ekala/ Mexico: The State Water Monitoring Au- Ja-ela Industrial Estate. Ekala/Ja-ela, one of two thority provided 1994 data on water emissions major estates in Sri Lanka, includes 143 indus- from 7,500 facilities in the Monterrey Metro- trial establishments with 21,000 employees. politan Area. USA: Regional databases provided infor- The Netherlands: The Ministry of Hous- mation on industrial water discharges, col- ing, Spatial Planning, and the Environment lected by the U.S. Environmental Protection provided 1990 data on water pollution from its Agency. Source: Hettige, Mani, and Wheeler (1998) ment costs (MAC) and marginal expected penalties (MEP). The plant's pre-reform pollution intensity is blue (where blue MAC = red MEP). In case 1, when energy prices rise, the cost of abating pol- lution increases to red, and so does pollution intensity (to the point where red MAC = red MEP). In case 2, with declining subsidies, the plant uses fewer raw materials, waste residuals decline, and the cost of abating pollution falls to green, as does pollution intensity (to the 112 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF Figure 5.4 Price Reform and Pollution Intensity Price Reform, Regulation, and Pollution Increasing MEP MEP e Stricter Regulation - Public Information MEP Reuig MAC-P *Case 2: Cutting\ S Materials Subsidies MAC MAC Case 1: Cutting Energy Subsidies * < : * Case 3: Ending Monopolies * . . *SBox 5,2: Devaluation / = \PwMAC * Pollution/Output point where green MAC = red MEP). In case 3, the plant's raw ma- terials supplier has lost monopoly power and lowered prices, so the plant uses more materials (generating more waste residuals), MAC increases, and so does pollution intensity. The value of a country's currency can also affect the prices of chemicals, equipment, and spare parts used for controlling pollu- tion, because developing countries often import such inputs. By making them more expensive, devaluing a country's currency-a common feature of economic reform-should raise pollution inten- sity by reducing pollution control. The study summarized in Box 5.2 suggests that Indonesia's recent devaluation had this effect. For clarity, Figure 5.4 assumes that MAC increases by the same amount for reduced energy subsidies, elimination of monopoly power, and devaluation. The plant will return to blue pollution in- tensity only if MEP is increased from red to blue by stricter formal or informal regulation. 5.3 The Impact of Plant Ownership on Pollution Economic reforms often change a country's pattern of factory ownership, which in turn affects pollution. The most obvious case is 113 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 5.2 Industrial Pollution in Indonesia's Financial Crisis A recent study analyzed the impact of Indone- Figure B5.2 Financial Crisis and Pollution sia's financial crisis on industrial production Indonesia's Financial Crisis: and emissions in a large sample of factories Changes in MAC and MEP (Afsah, 1998). The study confirms that indus- $ trial output declined significantly during the cri- sis-by 18 percent-but also finds that the in- tensity of factories' organic water pollution rose 4 by 15 percent. The study identifies two main reasons for increased pollution intensity: rapid devaluation of the country's currency, which made imported inputs for curbing pollution much more expensive (raising MAC); and dras- tic cuts in regulators' budgets (lowering MEP). m__ As plant managers responded to the new condi- Outputi tions, pollution intensity jumped from blue to red (Figure B5.2). However, the country's total Pre-Crisis emissions of organic water pollutants remained Crsis roughly constant, because declining production volume offset the rise in pollution intensity. privatization, which transfers state-owned plants to private hands, but trade reform can also decrease the role of family-owned and single-plant firms in the economy while increasing the role of large enterprises (particularly multinationals). State-owned plants world- wide have compiled an unenviable record of wasteful resource use and financial distress, which in turn means higher abatement costs, less investment in pollution control, and higher pollution intensity. Several recent studies have confirmed these effects. A study in In- donesia finds that public enterprises are considerably more pollution intensive than private factories.'( Analyses in China reveal that state enterprises have higher pollution intensity and lower regulatory compliance than other firms."1 And a four-country survey of pulp mills in Thailand, Bangladesh, India, and Indonesia shows that state-owned plants make far less effort to abate pollution than their private counterparts.12 Although the public enterprises in Indonesia's PROPER program were more compliant than private firms when the program began, 114 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF after 18 months the record of the two types of enterprises did not dif- fer significantly. We view this as a reflection of the fact-also experi- enced by CETESB in Cubatao-that state-owned enterprises are less susceptible to outside pressure, so public information exerts less influ- ence on their behavior. If this reasoning is correct, then state-owned enterprises will probably lag behind other firms participating in PROPER in the coming years. Overall, we believe the evidence clearly shows that privatizing state-owned enterprises reduces pollution. The Uncertain Effect of Family Ownership In Asia and Latin America, family enterprises have traditionally dominated many industry sectors. In Brazil, for example, two-thirds of the 33 largest private business groups are family controlled, and such groups also dominate in Mexico, Argentina, Colombia, and Chile.13 Family-controlled firms flourished during the protectionist era because state enforcement of business contracts was uncertain, the absence of international competition reduced pressure to hire professional managers, and limited domestic markets curtailed firms' need for large outside sources of capital. With more open markets and vigorous international trade, the advantages of family- owned business structures have begun to wane. Like state-owned enterprises, family-owned firms seem likely to be more pollution intensive and less compliant with environmental regulations than publicly traded firms. As we saw in Chapter 3, stud- ies in Mexico and other countries show that the stock market re- wards good environmental performance, and a complementary study of Mexican factories shows that plants in publicly traded firms com- ply with environmental regulations at significantly higher rates than family-owned factories. However, plant-level results from India do not show any compliance differential between family-owned and publicly traded firms. The authors of the India study postulate that better regulatory enforcement in Mexico prompts the Mexican stock market to value environmental performance more highly than the In- dian market. However, that interpretation is speculative, and the ef- fect of family ownership on pollution levels remains inconclusive.14 The Impact of Multinationals Conventional wisdom holds that multinationals are greener than domestically owned businesses in developing countries, because multinationals have high-level technical skills, good information about abatement alternatives, internationally competitive manage- 115 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS ment, and better access to capital. Moreover, as "guests" in the local economy, multinationals may be more susceptible to formal and in- formal pressure from regulators and communities. However, systematic evidence from studies of plant-level data in Indonesia, Bangladesh, Thailand, India, and Mexico shows that multinationals are not, in fact, greener than publicly traded domestic firms with similar characteristics.1s Still, the results of Indonesia's PROPER public-disclosure program imply a caveat. Multinationals and private domestic firms were equally compliant with regulations when the program began. However, after 18 months the former com- plied at significantly higher rates, suggesting that multinationals ad- just more rapidly to public information (Box 3.2). Because the In- donesian study does not fully control for ownership status, we do not know whether this effect stems from the fact that a firm is a multinational or that it is publicly traded. The Mexican results cited in Chapter 3 suggest that public trading is the key. In any case, by overturning the myth of green multinationals, these studies show that environmental progress is a domestic affair. Foreign investment can be valuable for many reasons, but it is un- necessary for effective pollution control. Consolidating Ownership As countries open their borders to trade and reduce the state's role in the economy, larger, more sophisticated operations that can withstand competition begin absorbing single-plant firms. As we saw in Chapter 4, these changes can boost environmental perfor- mance, because larger firms can spread the costs of in-house techni- cal services over more factories. Plant-level analyses in several coun- tries confirm this advantage. A recent study in Mexico also found that multiplant status improves regulatory compliance. (We might expect multiplant firms to attract more attention from regulators, but a recent study in India found that multiplant status has no effect on the rate at which regulators inspect a company.)j6 Reduced state intervention in the economy generally translates into larger factories in pollution-intensive industries, although this is not inevitable. The Soviet Union constructed huge steel plants with scant regard for the cost of transporting raw materials and steel, and in the post-Soviet era downsizing has reigned as transportation costs have risen. State enterprises in other economies have faced similar problems. However, economic reforms usually increase plant size. Re- cent work in the Philippines has suggested that larger manufacturing 116 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF plants are more economically efficient."7 As Box 5.3 shows, market re- form in China has led to a major consolidation of production. In India, many small plants survive only because they are protected by law. Figure 5.5 draws on four econometric studies to depict the country-wide results of consolidation: Merging two smaller hypo- thetical plants whose production and pollution levels originally total 100 units would cut overall pollution to about 70 units in Mexico and India and 80 units in China and Indonesia.18 However, even though it reduces pollution intensity, consolida- tion may inflict more overall pollution damage because large plants tend to cluster in populous areas. For example, a recent study in Brazil stratified 3,500 municipalities into 10 groups by income per capita and examined the location of 156,000 factories of various sizes. Figure 5.6 shows that large plants locate in wealthier regions, which also tend to have the largest populations (Box 2.2). Even though the large plants are less pollution intensive, more people die from air pollution because more of them live close to the plants. 19 Thus, unless regulators tighten pollution control, the net effect of consolidation may be more pollution damage. Fortunately, large plants are also natural targets for regulators with limited monitoring and enforcement capability. Strategies that focus on major polluters, such as CETESB's ABC approach, can lower pollution intensity to compensate for greater population exposure. Even where regulation is weak, local communities can more easily identify large plants and pressure them to reduce pollution. 5.4 Accounting for Pollution's Hlidden Half Overall, economic reform reduces pollution intensity by cutting subsidies for raw materials and encouraging international trade, pri- vatization of state enterprises, more publicly traded firms, and larger firms and plants. However, economic reforms do not always reduce pollution: Accelerated output growth may overwhelm cuts in pollu- tion intensity. And devaluing a country's currency, removing energy subsidies, and eliminating monopolies on production of raw materi- als may actually increase pollution intensity in some cases. Further- more, consolidation may bring more large plants to urban areas, increasing the impact of pollution on human health. The bottom line is that environmentalists can welcome most re- forms as pollution fighters, but economic reformers should recog- nize that their efforts can produce perverse environmental impacts. 117 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 5.3 Economic Reform and Industrial Pollution in China To determine the impact of China's economic Figure B5.3b Plant Size and Ownership reform on industrial pollution, we developed Output Share of Large Plants comparative statistics for five provinces in cen- 70 tral and eastern China (Figure B5.3a). Four of 0 1987 these provinces (including the Beijing and 0 1-987 e40-13 Shanghai metropolitan regions) are scattered 30 993 from north to south in China's eastern coastal 20-l i i i region. Beijing's industry has a diversity that re- 10 flects its status as the nation's political capital. Liaoning is a traditional center of heavy indus- '~ try, and many of its factories rely on old, highly polluting processes. Shanghai's industrial base Output Share of State-Owned Plants 80] Figure B5.3a Chinese Provinces 70 1987 60- ;j 50 1993 uS 40- 30 Liaoning w° tSN / / ,˘° Shanghai omy's share of large plants consistently grew from 1987 to 1993, most markedly in Shang- hai, while the share of state-owned enterprises Guangdong (SOEs) declined, particularly in Sichuan and Five Industrial Regions in Central and Eastern China Guangdong (Figure B5.3b). The econometric results reported in Das- is extremely diverse but its sheer scale assures a gupta, Wang, and Wheeler (1997) suggest that large potential pollution problem. Guangdong's both larger plants and a smaller role for SOEs development has brought rapid growth of light should reduce industry's pollution intensitv. manufacturing. Finally, Sichuan, located in the Figure B5.3c illustrates this result for organic Red Basin in south-central China, is consider- water pollution (chemical oxygen demand ably poorer than the other four provinces, and (COD)) and air pollution (TSP). At the provin- its industry is highly pollution intensive. cial level, emissions intensities for both types China's economic reforms have signifi- of pollution have declined significantly during cantly changed average plant size and owner- the period of economic reform. The greatest ship patterns in these five provinces. The econ- absolute declines have occurred in Sichuan 118 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF Box 5.3 (Continued) Figure B5.3c Pollution Intensities and Reform Figure B5.3d Dirty Sectors in China COD Intensities Share of Polluting Sectors: 70 Chinese Industry, 1977-1992 60 - _ _ _ _ _ __ _ _ _ _ _ __ _ _ _ _ _ 50 40- *1987 7 30- 1993 e 70 20- E 6h . ! hL- countr2fi S77 78 79 80 8I1i82 83 84 15 86 87 88 89 90 91 92 TSP Intensities 30 lcountry has weaker environmerital regulations 25 i and lower prices for heavy raw materials than 215 1987 many of its trading partners. To see whether zo 1 | h|h polluting industries have flourished under 10 1993 these conditions, we analyzed recent trends for o _ , 1 __ the five dirtiest sectors: chemicals, pulp and .s ° da 22 °paper, nonferrous metals, ferrous metals, and Stw0^- 4w ,, d:o / nonmetallic minerals (principally cement).* We found that the national output share of these five highly polluting sectors actually de- and Guangdong, but all five provinces have clined during the reform period (Figure B5.3d). seen pollution intensities drop. Thus, for China, opening the economy to more Concern about possible "pollution havens" trade has not produced a shift toward dirty in east Asia might focus on China, because the industries. * For identification of polluting sectors, see Mani and Wheeler (1998) and Hettige, Martin, Singh, and Wheeler (1994). Careful analysis of such effects by both economists and environ- mentalists-as well as collaboration between them-is essential. Fortunately, as we will show in Chapter 6, sophisticated use of in- formation technology can help regulators focus on the worst pol- luters and enlist communities in keeping them in line. To sustain such efforts and ensure their success, policymakers will have to de- vote some of the dividends from economic reform to improving en- vironmental information and regulation. 119 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 5.5 Plant Size and Pollution Doubling Plant Size Doesn't Double Pollution + ~~~~~~PRODUCTION POLLUTION Source: See End Note 16 Figure 5.6 Plant Size and Regional Development Brazilian Municipalities: Employment Share of Large Factories 70r 60- 20 - e40- 1 2 3 4 5 6 7 8 9 10 Income Decile Source: Dasgupta, Lucas, and WheeLer (1998) 120 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF References Afsah, S., 1998, "Impact of Financial Crisis on Industrial Growth and Environmental Performance in Indonesia," (Washington: US- Asia Environmental Partnership), July. CETESB, 1986, "Restoring the Serra do Mar." 1990, "Cubatao: A Change of Air." 1994, "Acao da CETESB em Cubatao: Situacao em Junho de 1994." Birdsall, N., and D. Wheeler, 1993, "Trade Policy and Industrial Pol- lution in Latin America: Where Are The Pollution Havens?" Journal of Environment and Development, Vol. 2, No. 1, Winter. Dasgupta, S., H. Hettige, and D. Wheeler, 1997, "What Improves En- vironmental Performance? Evidence from Mexican Industry," World Bank Development Research Group Working Paper, No. 1877, December. Dasgupta, S., M. Huq, and D. Wheeler, 1997, "Bending the Rules: Discretionary Pollution Control in China," World Bank Develop- ment Research Group Working Paper, No. 1761, February. Dasgupta, S., R. Lucas, and D. Wheeler, 1997, "Small Plants, Pollu- tion and Poverty: Evidence from Mexico and Brazil," World Bank Development Research Group Working Paper, No. 2029, November. Dasgupta, S., H. Wang, and D. Wheeler, 1997, "Surviving Success: Policy Reform and the Future of Industrial Pollution in China," World Bank Policy Research Department Working Paper, No. 1856, October. Dollar, D., 1992, "Outward-Oriented Developing Economies Really Do Grow More Rapidly: Evidence from 95 LDCs, 1976-1985," Economic Development and Cultural Change, 523-44. Economist, The, 1997a, "A Very Big Deal," Dec. 6. , 1997b, "Inside Story," Dec. 6. Hartman, R., M. Huq, and D. Wheeler, 1997, "Why Paper Mills Clean Up: Determinants of Pollution Abatement in Four Asian Countries," World Bank Policy Research Department Working Paper, No. 1710, January. Hettige, H., R. Lucas, and D. Wheeler, 1992, "The Toxic Intensity of Industrial Production: Global Patterns, Trends, and Trade Pol- icy," American Economic Review Papers and Proceedings, May. Hettige, H., M. Mani, and D. Wheeler, 1998, "Industrial Pollution in Economic Development: Kuznets Revisited," World Bank Devel- opment Research Group Working Paper, No. 1876, January. 121 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Hettige, H., P. Martin, M. Singh, and D.Wheeler, 1994, "The Indus- trial Pollution Projection System," World Bank Policy Research Department Working Paper, No. 1431, December. Mani, M., and D. Wheeler, 1998, "In Search of Pollution Havens? Dirty Industry in the World Economy, 1960-1995," Journal of Environment and Development, September. Mini, F., and E. Rodriguez, 1998, "Are SMEs More Efficient? Revisit- ing Efficiency Indicators in a Philippine Manufacturing Sector," World Bank, Operations Evaluation Department. Pargal, S., M. Huq, and M. Mani, 1997, "Inspections and Emissions in India: Puzzling Survey Evidence on Industrial Water Pollu- tion," World Bank Development Research Group Working Paper, No. 1810, August. Pargal, S., and D. Wheeler, 1996, "Informal Regulation of Industrial Pollution in Developing Countries: Evidence From Indonesia," JournaZ of Political Economy, Vol. 104, No. 6, 1314 +. Wang, H., and D. Wheeler, 1996, "Pricing Industrial Pollution In China: An Econometric Analysis of the Levy System," WA0orld Bank Policy Research Department Working Paper, No. 1644, September. , 1999, "China's Pollution Levy: An Analysis of Industry's Re- sponse," presented to the Association of Environmental and Re- source Economists (AERE) Workshop, "Market-Based Instru- ments for Environmental Protection," John F. Kennedy School of Government, Harvard University, July 18-20. Wheeler, D., M. Huq, and P. Martin, 1993, "Process Change, Eco- nomic Policy, and Industrial Pollution: Cross Country Evidence from the Wood Pulp and Steel Industries," presented at the An- nual Meeting, American Economic Association, Anaheim, Cali- fornia, January. End Notes 1. Sources for the Cubatao story include CETESB (1986, 1990, 1994) and several visits to the area by the authors. CETESB stands for Companhia de Technologia de Saneamento Ambiental. 2. For a description of CETESB's ABC targeting strategy, see Chapter 2. 3. See The Economist (1997a). COSIPA stands for Companhia Sideruirgica Paulista. 4. Wheeler, Huq, and Martin (1993). 122 NATIONAL ECONOMIC POLICIES: POLLUTION'S HIDDEN HALF 5. See Wheeler, Huq, and Martin (1993). We identified open and closed economies using a measure developed by Dollar (1992). 6. See Birdsall and Wheeler (1993), and Hettige, Lucas, and Wheeler (1992). 7. See Box 3.2, Pargal, Huq, and Mani (1998), and Dasgupta, Hettige, and Wheeler (1997). 8. For econometric evidence on materials prices and plant-level pollution intensity, see Pargal and Wheeler (1996). 9. These research projects included a 12-country study at the World Bank (Box 5.1) and country studies for India and Indonesia. See Hettige, Mani, and Wheeler (1998), Pargal, Huq, and Mani (1997), and Pargal and Wheeler (1996). 10. See Pargal and Wheeler (1996). 11. See Wang and Wheeler (1996) and Dasgupta, Huq, and Wheeler (1997). 12. See Hartman, Huq, and Wheeler (1997). 13. See The Economist (1997b). 14. See Dasgupta, Hettige, and Wheeler (1997), and Pargal, Huq, and Mani (1997). 15. References: India and Thailand: Hartman, Huq, and Wheeler (1997); Indonesia: Pargal and Wheeler (1996); Mexico: Dasgupta, Hettige, and Wheeler (1997). 16. References: India: Pargal, Huq, and Mani (1997); China: Wang and Wheeler (1999); Indonesia: Pargal and Wheeler (1996); Mexico: Dasgupta, Hettige, and Wheeler (1997). 17. See Mini and Rodriguez (1998). 18. See End Note 16. 19. See Dasgupta, Lucas, and Wheeler (1998). 123 0 Sepetiba / <:0St$0;:: Guanabara Algae ~~~~~~~~~~~~~~~~~~~~~~~~Total Coliforms* 1 15 [7 T F 00- 100 5-10 100 -500 10 -15 99K b i1000 -2000 .____ 6 2F-990 2000-25000 The Bays of Rio: FEEMA's Challenge Source: Ndlio Ricardo Aguarco, 3WayNet Assessoria de Propaganda e Marketing; Nibo Lima Fotografo Map provided by wvwv.mapquest. coa Managing and Sustaining Reform rom the top of Corcovado Mountain, Rio de Janeiro is a breath- taking study in contradictions. The view sweeps from Ipanema's luxury apartments and golden beaches in the south to the sto- ried entrance to Guanabara Bay in the east. Yet the scene also includes the mountainside shanties of the Favela Rocinha and the industrial suburb of Sao Crist6vao, both part of the "other" Rio-the crowded, polluted, poor communities that are home to most of the city's people. The top floor of the tallest building in Sao Crist6vao houses FEEMA, Rio State's environmental protection agency. On many mornings, FEEMA's staff cannot even see nearby Guanabara Bay be- cause the smog is too dense. Nor would a close-up view of the bay provide a pleasing spectacle in any case, as its inshore waters are brown and effectively dead, starved of oxygen by organic pollution from sewage and industrial waste. Once a showcase for Rio's beauty, Guanabara now languishes as a dirty backwater. South of the city, Sepetiba Bay appears to be headed in the same direction. Rio's air and water pollution seriously threaten its intended fu- ture as a center for world business and tourism, and its environ- mental decline has saddled the region's people with increased ill- ness, the loss of once-thriving fisheries, and fewer recreational opportunities. Rio's environment has deteriorated for many reasons, but FEEMA's declining effectiveness ranks high on the list.1 In the 1980s, the agency suffered from a loss of political support, reduced budgets, and an obsolete administrative system. GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS One afternoon in 1996, an impromptu street celebration marked the beginning of a new era for the agency. Employees applauded as a truck pulled away, carrying the archaic minicomputer that had housed the agency's central records. Served by a small priesthood of technicians who generated occasional reports, the machine had been effectively off-limits to the departments charged with environmental planning, factory inspections, and enforcement. Most records had been shelved in moldering folders or desultorily entered into a stand- alone PC that remained the exclusive province of a departmental manager. So the agency had operated for a decade, with minimal co- ordination and internal communication. As the minicomputer-laden truck departed, a new networked PC system began operating upstairs, symbolizing a fundamental change in the agency's approach. The network allowed departments to store their records in a common database that any unit could tap. The sys- tem reflected the agency's fresh approach to regulation: FEEMA's new president insisted on moving beyond legalistic, end-of-pipe reg- ulation to strategies that reflected assessments of overall benefits and costs for Rio State. Responding to the new mandate, agency managers began asking for reports that combined information from different departments. The effectiveness of the agency's technical staff increased as the net- worked database enabled them to analyze emissions trends, com- munity complaints, inspection reports, and readings from air and water monitoring stations, such as those in Guanabara and Sepetiba Bays. The agency's new geographic information system made a con- tribution by instantly combining maps of water quality, air quality, population centers, and pollution sources, allowing staff members to identify the worst environmental problems and polluters. Of course, FEEMA's renewal stemmed from far more than the technical fix offered by networked PCs. Like their counterparts in Ciudad Juarez, Mexico, FEEMA's leaders emphasized community participation in planning and implementing regulations-which in turn required public education on environmental quality, goals, progress, and the regulatory status of major sources of pollution. With its new system, FEEMA was well positioned to supply this in- formation in compelling, graphical formats. To capitalize on public involvement, the agency began considering a polluter rating system like Indonesia's PROPER program. FEEMA also focused on a more productive relationship with the business community. The agency had traditionally regulated through 126 MANAGING AND SUSTAINING REFORM continuous low-level negotiations with factory managers. In the new approach, FEEMA's president met with industry leaders to develop a consensus on environmental objectives for the state. Progressive CEOs supported this approach, because they envisioned Rio as an environmentally friendly headquarters city for international busi- ness. FEEMA's managers also negotiated commitments from industry groups that entailed collective responsibility for failure to meet pol- lution control objectives. Since the agency was better equipped to provide information on the condition of the environment and priori- ties for action, the business community listened to its proposals with enhanced respect. FEEMA's third outreach effort focused on developing closer rela- tionships with the World Bank and other international agencies. Like FEEMA, those institutions had begun shifting from a focus on end- of-pipe regulation to promotion of flexible regulatory instruments, government-industry cooperation, and the use of benefit-cost analy- sis to determine priorities. The Bank and other institutions provided financial and technical support for the agency's new information system, helped develop action plans, and sponsored specific projects for reducing air and water pollution in Rio de Janeiro State. Environmental reform in Rio resembles many other renewal ini- tiatives, in which regulators are using decentralized information technology to assess their options and develop cost-effective pro- grams based on multiple sources of data. New technology has cut the cost of gathering, processing, and distributing this information, enabling regulators to mediate environmental agreements between communities and businesses more effectively. More cooperative re- lationships have, in turn, encouraged stronger political support for environmental policy reforms.2 6.1 The Contribution of lnformation Systems Timely, accurate, and well-packaged information is critical to the new approach. To illustrate an effective environmental information system, consider a hypothetical case of river pollution. Emissions from diverse activities along the river-a large factory, numerous small ones, a farming district, and a riverside community-cause water to decline in quality as it moves downstream. Because staff skills and time are limited, the local environmental agency focuses on accurately tracking critical water pollutants rather than maintaining an unwieldy catalog of all possible emissions data. 127 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS The agency's priorities include heavy metals, fecal coliforms, biolog- ical oxygen demand (BOD), and phosphorus (Figure 6.1): The first two pollutants pose serious threats to human health, while the latter two damage ecosystems. The agency requires all plants along the river to submit periodic reports, certified by outside auditors, on emissions of these sub- stances. The reports are usually accurate because the agency black- lists dishonest auditors, putting them out of business. The agency uses random, surprise inspections to keep the probability of discov- ering false reports high. Monitoring devices in the river further confirm the reported emissions and measure their impact. This effort feeds standard, user-friendly database software and relies on a dispersion model, built by the agency's technical team, that tracks pollution using data on the river's flow rate, volume, temperature, and other factors. The first upstream monitor in Figure 6.1 records no significant pollution (all ratings are blue). Regulators know that a large food- processing plant downstream dumps a heavy BOD load into the river, and farther downstream a complex of small tanneries and textile mills reports substantial emissions of heavy metals and additional BOD. A second river monitor shows that these discharges signifi- cantly affect water quality, and the information system rates BOD or- Figure 6.1 Monitoring Pollution Identifying Sources of Pollution Large Plant Agriculture QMonitoring ._I ID Stations Small Plants n Ambient Quality Good BOD Fair \ Metals Poor Phosphorus= critical Coliforms 128 MANAGING AND SUSTAINING REFORM ange and metals yellow. Below the industrial area several large farms abut the river; a third monitor reveals a heavy phosphorus load from fertilizer runoff, and the system rates that element orange. The mon- itor also indicates that the river has assimilated some BOD, so its rat- ing improves to yellow, while the rating for metals becomes red. Finally, the river flows past a community that discharges un- treated sewage laden with BOD, fecal coliforms, and phosphorus from household detergents. The fourth monitoring station rates BOD and coliforms red, although metals improve to orange because some have settled to the bottom, later to appear in the tissues of fish. The rating for phosphorus becomes red from the combination of agricul- tural runoff and community wastewater. As the river leaves the monitored area, it is for all practical pur- poses dead. Contaminated by pathogens, its water is dangerous to drink or swim in, its dissolved oxygen level is too low to support many species of fish, the metals content is too high for people to safely consume any fish that remain, and algae affect the water's color and odor. Communities farther downstream will bear heavy costs from this pollution. The agency has now met its first three responsibilities: identify- ing major pollution sources, monitoring their effluent, and analyzing the effect on environmental quality. The computer system links to a geographic information system that enables users to call up tables and maps that report ambient quality at each point on the river, along with information about polluters' characteristics, emissions and compliance status (Figure 6.2). Choosing a Plan of Action Although data on pollution sources and ambient quality provide critical information, policymakers must further analyze the results to determine their most cost-effective response. Toward that end, the agency's technical team uses the database and the best available mod- els to estimate pollution-related damage to human health and losses of aquatic life, recreational amenities, and economic output. The team also uses the characteristics of polluters to identify those that can re- spond rapidly and at low cost to tighter regulation (Figure 6.3). Taking these factors into account, policymakers then develop a strategy that identifies the pollutants to be regulated, ambient qual- ity goals, a timetable for reaching them, and appropriate regulatory instruments: pollution charges, tradable permits, or emissions stan- 129 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 6.2 Data Collection The Environmental Database |Ambient Quality BOD _ & 1wl = Metals Phosphorus Coliforms c Monitors \n pr 6.~ ~ //atn CoIto s oOur CaneS BOD \ t< Metals _ PhosphoIru_ _ _ _ _ _ .| Coliforms_r _ _ dards. The regulators must also decide which sources to target, tak- ing note of factories' importance as local emplopers and other politi- cally sensitive issues. In practice, the regulators know that their ef- fectiveness depends on constant feedback and long-term support from all the groups affected by regulation. 6.2 Creatinsg Coalitions flor Chlange Credibility is the irreplaceable currency of regulation, as regula- tors' political influence and budgetary support will plummet if the public believes that an agency is corrupt or incompetent. Polluters will also resist regulation more easily if credibility falls, and the news media will discount information provided by the agency. Environmental reformers have discovered three keys to main- taining the precious asset of credibility: focus, transparency, and community participation. For resource-strapped agencies in develop- ing countries, focus provides the best protection against operational failure and loss of reputation. Agencies can avoid serious trouble if they target a small group of serious polluters, limit regulation to a few critical pollutants, effectively measure these pollutants as well 130 MANAGING AND SUSTAINtNG REFORM Figure 6.3 Analysis Regulatory Analysis } Database | ^9 | Technical Analysis P [oicy Analysis Ambient Quality Overall Damage Policy Priorities Health Ambient Goals Ecosystems Abatement Goals Emissions Damage Sources Source Targeting ii ii jE - Scale Instruments Performance Enforcement Dispersion Models Aatement Impact Models Options Costs as regulatory compliance, and publicly document their activities. Agencies whose ambitions outrun their resources may look good for awhile, but their credibility will fade as factual errors and misjudg- ments accumulate on the public record. Transparency provides the second key to credibility because it prevents corruption. Corruption at higher levels of management can fester for a long time without detection, and even if managers are honest, secrecy may tempt individual inspectors to take payoffs that may dwarf their salaries. Without public information, communities have no way of knowing whether regulators are doing their job. The solution to the secrecy problem is clear: consistent, unbiased disclo- sure of polluters' emissions, local environmental impacts, inspection results, and enforcement actions. Chapter 3 has discussed innova- tive disclosure programs in several Asian and Latin American coun- tries; Box 6.1 describes the process that ensures credibility for In- donesia's PROPER rating system. Earlier policy experiments in Brazil illustrate the importance of information integrity in pollution disclosure programs. During the late 1970s, the pollution control agencies of Rio and Sao Paulo States attempted to pressure the most serious polluters by announcing their names to the media. Both agencies were sued by the firms whose names were published. In Sao Paulo, CETESB's program sur- vived the court challenge because it could provide solid evidence to 131 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 6.1 PROPER: Building Credibility Indonesia's public rating system for polluters Figure B6.1 Steps in Developing PROPER has significantly reduced emissions and at- PROPER: Assuring Credibility tracted strong popular support. Still, the pro- gram was delayed a year, until mid-1995, be- cause BAPEDAL's leaders realized that it would succeed only if the public understood the rat- ings, trusted the underlying information, and believed in the honesty of PROPER's managers. The ratings strategy relied on public com- -_l munication. By summarizing complex infor- mation in a single five-color rating scheme, PROPER allowed the public to compare pol- luters across many locations and industry sec- Source: Shakeb Afsah; BAPEDAL tors. The ratings also made it easy for the media to show which factories complied with vetting the ratings through three checkpoints: regulations and which did not. an advisory board, with representatives from Data integrity presented a significant chal- academia, industry, other government agen- lenge. The agency had relatively little experi- cies and environmental NGOs; the Environ- ence with collecting, verifying, and analyzing ment Minister; and finally the President him- large amounts of information, so PROPER had self (Figure B6.1). The first ratings emerged some false starts before the Minister of the Envi- from these checks unscathed, and their effect ronment was satisfied with the rating system. on the business community was compounded Finally, with the assistance of technical teams by the knowledge that the President of Indone- from the World Bank, Canada, Australia and the sia had approved them. United States, BAPEDAL developed a reliable These careful measures have paid off. system for monitoring, recording, and analyzing PROPER has maintained its reputation for con- water emissions from a pilot sample of factories. sistency and objectivity, and the public has To enhance credibility with all stakehold- continued to support it despite Indonesia's po- ers, BAPEDAL developed a careful process for litical and financial crisis. support its claims. After the court rejected the lawsuit, most black- listed firms quickly reduced pollution enough to remove themselves from the spotlight. In Rio, by contrast, FEEMA's data were too weak to withstand the court challenge. The lawsuit succeeded, and the Rio disclosure program collapsed.3 Widespread participation also builds public credibility. Politics and the theater share an important trait: Once actors have appeared onstage, their existence becomes part of the drama. The audience assigns them a role that cannot be vacated without comment or jus- 132 MANAGING AND SUSTAINING REFORM tification. Successful environmental reformers exploit this principle by crowding the public stage with leaders from communities that confront major pollution problems. Once these leaders come on- stage, political opponents of environmental reform can no longer ig- nore them, and their presence helps insure that the agency itself will not be co-opted by special interests. Despite the importance of ex- panded community participation, however, not all regulators may welcome it. Power is seldom ceded willingly, and reformers' most difficult task may be persuading some of their own colleagues to support participatory approaches. An effective regulatory system promotes two-way communica- tion with participants. As communities and markets gain access to environmental information, pressure through many new channels can prompt polluters to reduce their emissions (Figure 6.4). Feed- back from the public is also critical: Effective regulators must under- stand and act upon communities' environmental concerns. Toward this end, Mexico's Environment Ministry is establishing community centers that automatically log, categorize, and route citizens' com- plaints about pollution to the appropriate authorities. Such systems allow regulators to identify areas with serious problems, and also enable citizens to monitor responses to their complaints. 6.3 The Politics of Sustaining Reform Environmental reforms are often difficult to sustain. As govern- ments change, high-level sympathy for environmental regulation Figure 6.4 Reactions to Public Information Ad M. Mh Annbienit Conditions ilm: l: . bDamages Regulatory Targets Performance Ratirgs Responses Adjustment Costs Citizens NGOs Consumers Investors o1 _ | ~~~~Loans l Stock Values 133 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS may wax and wane. Even if they support existing programs, politi- cally appointed environment ministers seldom enter office with deep knowledge of pollution problems. To remain effective, agency staff members must continually sell their programs, developing po- litical skills and long-term budgetar-y strategies along with technical competence. Securing the Budget Traditional public-finance theory acknowledges that polluition charges and fines provide useful incentives to reduce emissions, but it also holds that regulators should remit revenues to the regional or national government. By drawing oo all tax and penalty revenues, the government can fund social, educational, or environmental pro- grams with the highesr benefit-cost ratios. If the government had perfect foresight, Platonic neutrality, and seamless administrative efficiency, no one could quarrel with this approach. Spending would encompass not only current programs but also long-term investment, and public planners would assure stability in the flow of funds. Unfortunately, many regulators, espe- cially those in developing countries, do not inhabit such a world: The political process is quirky, unstable, and prone to sudden crises that drain available budgets. Regulators also face continual challenges from threatened inter- ests. While some industrialists have strategic vision and support ef- fective regulation, others remain fixated on the short-run bottom line. The most recalcitrant will lobby their political allies for cut- backs in regulators' buidgets, and may be joined by some labor lead- ers who view stricter pollution control as a threat to jobs. Confronted daily by these realities, regulators often attempt to retain control over pollution charges and fines because they are se- cure funding sources. Retaining some control also gives regulators stronger incentives to collect fines from polluters. Agencies' desire to keep revenues out of central hands sits well with local politicians and business leaders, who want to see local payments used to sup- port local environmental programs. Environmental reformers miust often heed these views because hostile businesses and politicians can block new programs. Colombia's pollution charge program moved forward when reg- ulators, industrialists, and public sewerage authorities agreed to use part of the revenues to support the regional regulatory agencies, and to invest the rest in local environmental projects (Box 6.2). Although traditional public-finance theory does not support such an approach, 134 MANAGING AND SUSTAINING REFORM Box 6.2 Sharing the Funds in Colombia Chapter 2 described Colombia's new regu- Figure B6.2 Using Pollution Charge Revenues latory system, which charges polluters for each Sharing Colombia's Regional unit of emissions. Under the standards-based Decontamination Funds system, regional environmental agencies had Environmental Education legal authority to fine plants that failed to com- Projects (Schools, ply with regulations. Reality often dictated oth- Universities) (5/) erwise, however, because enforcement proce- dures were cumbersome and susceptible to n E 5%) legal delaying tactics. The new pollution charge system jettisons criminal sanctions: Plants are free to pollute and pay, but the charges are high enough to affect managers' cost calculations significantly. Source: Colombian Environment Ministry When the program first took shape in Colombia's Environment Ministry, the design ganizations as allies on their side of the table. team focused on technical issues, working to Finally, representatives from the Environment estimate abatement costs and set charges that Ministry team, regional agencies, industrialists, would reduce pollution significantly without public works managers, and community orga- bankrupting polluters. However, once the team nizations hammered out a mutually agreeable moved to the field, it quickly found that politi- solution. The new charge program would sup- cal issues eclipsed technical ones. Polluters port "regional decontamination funds" used for themselves defined a central concern: Once local environmental projects, after some por- they paid, who would get the money? The re- tion was diverted to fund agency budgets. Fig- gional agencies laid claim to some of the funds, ure B6.2 summarizes the Environment Min- because they wanted financial insulation from istry's recommendations for use of the funds, the political funding cycle. Local business and which most regional authorities have followed. public works managers accepted that idea but Public finance theory does not counte- refused to countenance remitting the balance to nance the diversion of charge revenues to the national treasury. They weren't impressed purely environmental projects, but in reality by the argument that charges would automati- the program designers had no choice: no re- cally improve the environment by encouraging gional funds, no program. The ministry team cost-minimizing plants to reduce pollution. In- accepted the package and, in an ingenious stead, they viewed the charges as a financial twist, enlisted one of Colombia's top commer- sacrifice they would bear only if the revenues cial banks to collect the pollution charges (for were used to fund local investments in cleaner a percentage fee), administer the funds to get manufacturing and wastewater treatment. maximum interest, and disburse them to ap- Without support from industrialists and proved projects. That solution unburdened the public works managers, the charge program local agencies, which had little expertise in stood no chance of implementation. Tough ne- billing, collecting, and disbursing, while it en- gotiations loomed, and the regional environ- couraged private-sector polluters to pay in mental agencies enlisted community-based or- order to preserve their credit ratings. 135 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS the program's strengths have clearly outweighed this conceptual "flaw." Pollution charges represent a leap forward in regulatory effi- ciency for Colombia, and the local funding mechanism ensures some measure of long-term stability and effectiveness. However, accepting political reality does not imply uncritical ac- ceptance of any funding scheme. The designers of Colombia's sys- tem have stressed the application of clear benefit-cost criteria to local financing of pollution reduction projects. Useful projects may include public wastewater treatment facilities and support for im- proved environmental management in small and medium enter- prises (Chapter 4). Subsidized loans to private firms for end-of-pipe abatement are probably ill-advised in most cases. Numerous inter- national studies have shown that large firms often gain access to the funds simply because their staffs can produce good technical pro- posals. Yet such firms will generally clean up anyway, if pollution charges or other instruments provide the right incentives.4 Success Stories Three countries where regulators have adopted new programs il- lustrate the political aspects of successful reform.5 The Colombian pollution charge program has developed strong coalitions of stakeholders in many of its administrative regions. As described in Chapter 2, regulators in each region mediate negotia- tions between industries and communities on pollution reduction targets and schedules for raising charges if the targets are not met. This participatory approach has created strong community support for the program and has helped insulate it from its political and bu- reaucratic opponents. To reinforce community support, the program's promoters are also pushing for a complementary public disclosure program like In- donesia's PROPER. They view it as a powerful vehicle for environ- mental education as well as a means of addressing public distrust of governmental institutions. The program's promoters also believe that more-precise knowledge of local pollution sources will mobilize communities to confront polluters informally, as well as negotiate with them formally over pollution targets and charges. In Indonesia, where developers of PROPER also relied on com- munity support, some environmental NGOs initially feared that the program's use of the media would co-opt their traditional role as community advocates. To ensure NGO support, BAPEDAL, Indone- sia's environmental agency, invited NGO leaders to join the advisory group that reviewed industry ratings before they were publicly an- 136 MANAGING AND SUSTAINING REFORM nounced. The NGOs accepted, principally because they had a long- standing relationship with BAPEDAL's Deputy for Pollution Control. BAPEDAL also enlisted support for PROPER from progressive business leaders. Program designers were well aware that large, technically sophisticated companies like PT Indah Kiat could earn good ratings from PROPER (Chapter 3), and they expected CEOs of such firms to support the program as a source of competitive advan- tage. But the PROPER team also went out of its way to avoid antago- nizing firms whose ratings were initially subpar, by precisely identi- fying the reasons for the poor ratings, suggesting actions to improve them, and offering a grace period before formally announcing the ratings. Agency officials also met regularly with company managers to address their concerns. To further ensure long-term support, an endorsement from Indonesia's President accompanied PROPER's ini- tial ratings, and the Vice President publicly announced the highest- rated factories as part of Indonesia's Earth Day activities. The developers of Philippines' EcoWatch disclosure program pur- sued a similar political strategy. The nation's President formally an- nounced EcoWatch along with leaders of the Philippines Business As- sociation, who encouraged association members to participate in the program. The President reiterated his support in speeches and public announcements, and the program allowed poorly rated factories a grace period before public disclosure. 6.4 Living with Change Politics remains the art of the possible, and no reform of envi- ronmental policy can anticipate all untoward events. Political tur- moil is a fact of life in many developing countries, and sudden crises or larger political forces can undermine even successful programs with solid support. In Ciudad Juarez, a decision by the Mexican Government to end subsidies for propane undermined a successful program to induce small-scale brick makers to switch to cleaner fuels. Indonesia's financial crisis has made imported pollution- control inputs more expensive and forced cuts in regulators' bud- gets, reversing some of the gains made under PROPER. Nevertheless, many well-packaged reforms have proven remark- ably durable in the face of rapid political change, including all three programs discussed in this chapter (Box 6.3). We credit much of this survival to the political entrepreneurship of the programs' designers, and to their respect for three fundamental principles of innovative regulation: focus, transparency, and community participation. 137 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Box 6.3 Sustaining Reforms in the Face of Political Change Environmental policy reforms remain vulner- Economic Analysis translated this principle able to political change until they are fully insti- into policy by establishing a nationwide pollu- tutionalized. Figure B6.3 illustrates how expand- tion charge program. Implementation began ing elections in developing countries have when CORNARE, a regional agency near Me- increased the pace of political change. Neverthe- dellin, started billing local factories for emis- less, as the following examples show, strong sions in 1998. programs with wide popular appeal have repeat- The program has since operated under edly survived turnover in national governments. three different environment ministers: Jose Vi- cente Mogollon (1996-1997), Eduardo Verano Colombia de la Rosa (1997-1998), and Juan Mayr Mal- In 1993, Colombia's Congress established donado (1998-present). The latter took office the Ministry of Environment and the "polluter when Andres Pastrana Arango of the Conserv- pays" concept as a fundamental principle of ative Party defeated Horacio Serpa of the Lib- Colombian law. In 1997 the Ministry's Office of eral Party to win the presidency. Although the Figure B6.3 Elections in Developing Countries Presidential Elections * 1 1998 4 ~~~~~~~~1999 2000 Source: IFES (1999) 138 MANAGING AND SUSTAINING REFORM Box 6.3 (Continued) national administration has changed, support Philippines for pollution charges continues because the Many countries watched the development of program's local constituencies remain politi- PROPER, but none more closely than the neigh- cally potent. boring Philippines. In 1996, Secretary Victor Ramos of the Department of Environment and Indonesia Natural Resources (DENR) launched a simi- In 1993, BAPEDAL's Deputy for Pollution lar program, called EcoWatch. President Fidel Control proposed PROPER to Indonesia's Envi- Ramos publicly supported the program, which ronment Minister, Sarwono Kusumaatmadja. began by focusing on organic water pollution in After careful development, the program moved the national capital region. Within 18 months to pilot implementation in mid-1995, and was EcoWatch coverage expanded from 52 to 83 considered a major policy success of the Suharto major polluters, and their regulatory compliance government. Then in mid-1997, Indonesia's fi- rate jumped from 8 percent to 58 percent. This nancial and political crisis hit. During the ensu- successful introduction attracted widespread ing turbulence PROPER has operated under two support from the media, community leaders, new environment ministers: Wijoyo Sudarsono and environmental NGOs. and Sergir Panangian. The BAPEDAL Deputy After the 1998 elections, DENR Secretary who developed PROPER has left the agency, and Ramos left office along with President Ramos, it has endured budget cuts along with other en- and President-elect Jose Ejercito Estrada ap- vironmental programs. Nevertheless, popular pointed Antonio Cerilles as the new Secretary of support for the program remains strong and it DENR. After taking stock, the new administra- continues to operate. Original plans to rate 2,000 tion decided to continue EcoWatch because factories by the year 2000 now look ambitious, many of its constituents supported the program. but the program expects to have ratings for some DENR now plans a rapid expansion of EcoWatch 500 factories by the end of 1999. to cover major polluters throughout the country. R e P ec Hanrahan, D., M. Keene, D. Shaman, and D. Wheeler, 1998, "Devel- oping Partnerships for Effective Pollution Management," Envi- ronment Matters at the World Bank, Annual Review. IFES (International Foundation for Election Systems), 1999, "IFES Elec- tions Calendar," available at http://www.ifes.org/eleccal.htm. Lovei, M., 1995, "Financing Pollution Abatement: Theory and Prac- tice," World Bank Environment Department Paper, No. 28. Von Amsberg, J., 1996, Brazil: Managing Environmental Pollution in the State of Rio de Janeiro, World Bank, Brazil Department, Re- port No. 15488-BR, August. 139 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS , 1997, Brazil: Managing Pollution Problems, The Brown Envi- ronmental Agenda, World Bank, Brazil Department, Report No. 16635-BR, June. World Bank, 1999, Pollution Prevention and Abatement Handbook (Preliminary Version), available at http://wbln0018.worldbank. org/essd/PMExt. nsf. Wheeler, D., 1997, "Information in Pollution Management: The New Model," in Von Amsberg (1997). End Notes 1. For a comprehensive discussion of FEEMA's problems, see Von Amsberg (1996). 2. For further discussion of the new approach, see Hanrahan, Keene, Shaman, and Wheeler (1998) and World Bank (1999). 3. Interviews with staff members of CETESB and FEEMA. 4. For a detailed discussion of this issue, see Lovei (1995). 5. These cases are drawn from the authors' collaborative ex- perience with environmental agencies in Colombia, Indonesia. and Philippines. 140 muf\ J | \~~~~Nb. IPO MO \4 Greening Industry: The New Model L ooking ahead with moderate optimism, we can imagine that the Earth's population has stabilized at 10 billion by the year 2050. About half the people live in today's developing coun- tries, and a half century of 5 percent annual growth has given them incomes of $4,000 per capita-twice the level in middle-income countries at the turn of the century. People still aspire to higher in- comes, but desperate poverty has been vanquished. Who wouldn't welcome such a prospect? It could materialize, since many countries have achieved 5 percent growth since 1950. But a shadow looms when we realize that this scenario would entail a 25-fold growth in output and potentially huge increases in pollu- tion. Reacting to such numbers, some people argue that the world's poor will never enjoy material prosperity because industrial civiliza- tion will run headlong into an environmental catastrophe. The full story of environment and development has yet to be told, and we cannot guarantee a happy ending. Global warming, de- forestation, loss of biodiversity, and other problems remain daunt- ing. Our own work has focused on only one chapter of the story, and one basic question: Can societies hold local, not global, indus- trial pollution within acceptable bounds while industry continues to grow? For this question, at least, the answer appears to be yes-if we are clever and careful. We are optimistic because greening industry is not a futurist fan- tasy. In every country, no matter how poor, some factories already GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS operate at world-class environmental standards, and many profitable enterprises comply with national pollution regulations. Furthermore, extensive research has shown that sound, focused economic and en- vironmental policies can greatly increase the number of good per- formers. Some of these measures entail reform of national economic policies, some require innovative and cost-effective approaches to formal regulation, and some harness the power of communities and markets to influence polluters through informal channels. In this report, we have highlighted several innovative programs that demonstrate the potential for pollution reduction. Pilot projects are spreading as more countries decide to experiment with the new approaches, and broader experience will improve our understanding of their strengths and limitations. For the present, we can report that the results to date look promising. They suggest that coordinated ac- tion an all three fronts-economic reform, formal regulation, and in- formal regulation-can reduce industrial pollution significantly, even in very poor countries. 7.1 The Keys to Progress Sustained progress on pollution control in developing countries depends on clear evidence that its benefits and costs compare favor- ably with those of other social investments. From Beijing to Sao Paulo, recent studies have verified that abatement of critical pollu- tants is a sound investment in many urban areas. However, regulat- ing all pollutants under all conditions is neither economically defen- sible nor politically sustainable. Regulators have limited skills and resources, and they will rapidly lose political support if the public regards them as sloppy, unfair or ill informed. In Indonesia, the PROPER public disclosure program has demonstrated the impact of strategies that focus on accurately tracking and reporting a few criti- cal pollutants from large emitters. To maintain political support, environmental agencies need to marshal reliable information, educate the public about environmen- tal tradeoffs, and encourage broad participation in setting goals. Such participation plays an important role in maintaining the credi- bility of Colombia's pollution charge system, EcoWatch in Philip- pines, and Indonesia's PROPER program. In PROPER, for example, stakeholders have an opportunity to vet the program's plant-level ratings before they are disclosed, forcing the agency to discipline its system for gathering, analyzing, and reporting data. And as the case 144 GREENING INDUSTRY: THE NEW MODEL of Ciudad Juarez in Mexico shows, communities that participate in regulation will support its objectives, provide information about local polluters, and defend the environmental agency against politi- cal attack. FEEMA's sustained outreach efforts in Rio de Janeiro reveal that good relations with business leaders are crucial as well, since indus- try associations often have the political clout to veto pollution-control programs. Regulators will find natural allies among CEOs of firms whose market position depends on good environmental perfor- mance. Having already paid for cleaner production, these leaders will support measures that require similar efforts from their competitors. Leveraging Polluters' Incentives Finally and most critically, progress requires understanding that managers do not sanction pollution because they enjoy fouling the air and water, but because they are trying to minimize costs. At the factory level, the marginal expected penalty for polluting (MEP) tends to rise with emissions intensity. But when managers reduce emissions, they also increase the plant's marginal abatement cost (MAC)-the price of abating the next unit of pollution. So managers try to minimize their overall costs by adjusting emissions until MAC approximately equals MEP. Governments have many opportunities to influence this plant- level balancing act by reducing MAC or raising MEP (Figure 7.1). Governments can reduce MAC through national reforms such as lib- eralizing trade, privatizing national industries, and promoting new stock markets. Research in China, India, Mexico, Indonesia, and other developing counties has demonstrated the power of such mea- sures. But economic reform is no panacea, because some policies can produce perverse environmental impacts. In Ciudad Juarez, for example, Mexico's decision to end propane subsidies dealt a devas- tating blow to the local campaign for cleaner brick production. Na- tional economic reformers can contribute to the fight against pollu- tion by anticipating such impacts and working with environmental agencies to counter them. Appropriate measures may include strengthening formal regulation in critically affected areas, support- ing more public dissemination of environmental information, and slowing implementation of environmentally risky reforms while local environmental institutions adapt to the new demands. Coordi- nating economic reforms and environmental policies will require close contact between the relevant national ministries. A formally 145 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS Figure 7.1 Policy Options for Pollution Control Greening Industry $ MEP Reduo.g r AC _ iNrational Policy Reformts ( Nnti~r~i ~o ~ MAC Increasing MEP P3ant-Levei EMS Training, MAC / Formal Regulation Public Information M"AC MEP Pollution/Output constituted environmental advisory unit for the key economics min- isters may provide the best guarantee that economic reform pro- grams will incorporate such concerns. At the sectoral level, governments can lower MAC by supporting environmental management training for small and medium-size en- terprises. The example of Guadalajara, Mexico, suggests that such programs can provide a cost-effective complement to conventional regulation. Documented experience remains limited, however. More studies of pilot programs are needed to assess the strengths and limitations of environmental management training under different developing-country conditions. At the plant level, regulators can raise MEP through both formal and informal channels. Among formal instruments, market-based in- struments such as pollution charges reduce emissions at the lowest cost because they leave abatement decisions in the hands of factory managers. Successful experiences in China, Colombia, and Philip- pines have shown that pollution charges are feasible and effective in developing countries. Tradable pollution permits may also work well, but successful developing-country experiences have not yet been doc- umented. Although interest in market-based instruments is spread- 146 GREENING INDUSTRY: THE NEW MODEL ing, many regulators will continue to rely on standards-based regula- tion for some time. Even so, targeted programs like Sao Paulo's ABC approach have demonstrated that traditional regulation can be rea- sonably cost-effective if monitoring and enforcement are focused on large pollution sources with low abatement costs. Regulators can encourage informal regulation by publishing reli- able, easily understood information on pollution sources and their impacts. Both EcoWatch in Philippines and PROPER in Indonesia have shown that public disclosure can have a strong impact even where formal regulation is weak, because it enlists social norms and market forces in pressuring polluters to clean up. Public information programs carry the extra benefit of generating political support for pollution control, by educating communities and raising the credibil- ity of environmental agencies. .2 New odel or Controllng Po,itgr. The proliferation of these new formal and informal channels is effectively creating a new model for pollution control (Figure 7.2). In this model, regulation is much more information intensive and transparent. As an environmental agency exerts influence through nu- merous channels, it becomes more like a mediator and less like a Figure 7.2 New Dimensions for Policy The New Model 147 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS dictator. Community representatives take their place at the negotiat- ing table along with regulators and factory managers. Market agents make their presence felt as well, through decisions by consumers, bankers, and stockholders. The new model empowers policymakers because it gives them many options for improving industry's environmental performance. But the model also imposes new responsibilities-for strategic think- ing about the benefits and costs of pollution control; a strong com- mitment to public education and participation; intelligent, focused use of information technology; and a willingness to adopt new ap- proaches such as pollution charges and public disclosure. Of course, regulators will always have important responsibilities for traditional monitoring and enforcement. But in the future they will use more of their resources to raise MEP through informal regulation, to lower MAC through support for improved environmental management by small firms, and to promote sustainable economic reforms by work- ing more closely with national policymakers. The new approach pays particular attention to the problems of the poor. Recent research has shown that pollution intensity declines steadily as per capita income rises, both within and across countries. But economic development takes a long time, and the poor suffer heavily from pollution now. Evidence from Mexico, China, and else- where has shown that education provides a powerful lever for near- term improvement: Even if people are poor, they will not passively ac- cept pollution if they are well informed about its sources and impacts. Through public education and maintenance of appropriate environ- mental standards, governments can help assure basic amenities and human dignity for the poor during the transition to greener industry. 7.3 The Role of the World Bank The World Bank makes significant contributions to controlling industrial pollution on several fronts. By encouraging nations to adopt needed economic reforms, it influences pollution's "hidden half." In the long run, support for growth-oriented policies will en- courage stricter pollution control by more prosperous societies. But the Bank has learned that not all economic reforms have clean im- pacts in the near term. It has recently revised its operational guide- lines to ensure that Bank-supported reform programs incorporate en- vironmental concerns. Successfully implementing these guidelines will require sustained effort, coordination between the Bank's econ- 148 GREENING INDUSTRY: THE NEW MODEL omists and environmental specialists, and active collaboration be- tween economic ministries and environmental agencies in partner countries. The Bank has also financed decentralized environmental infor- mation systems that support the new regulatory model. Here the em- phasis should be on appropriate scale, since experience cautions against using the most sophisticated modeling and data-processing technology to address all possible environmental problems. This com- prehensive approach, which could be encouraged by the Bank's pref- erence for big loans, can easily distract regulators from confronting their communities' most critical pollution problems. And once regula- tors have lost focus and clarity of purpose, their performance and credibility soon dwindle as well. For several years, the Bank has catalyzed new thinking on pollu- tion regulation by supporting pilot projects and disseminating their lessons to the international community. Partner environmental agen- cies have taken the lead, but the Bank has provided technical assis- tance, financial backing, and public support for innovative ideas. Re- cent initiatives of this kind include the Guadalajara small business assistance project in Mexico, the pollution charge programs in Philip- pines and Colombia, state-level environmental management reforms in Brazil, and public disclosure programs in Indonesia, Philippines, Mexico, and Colombia. How can the World Bank promote the new model in the coming decade? Critical tasks include continued sponsorship of innovative pilot projects, widespread dissemination of their results, develop- ment of environmental loans that expand successful pilots to large- scale programs, and serious incorporation of environmental con- cerns into loans that support national economic reforms. The Bank can use several instruments to support pilot proj- ects, including the new Learning and Innovation Loans, Bank- administered environmental trust funds, and support of technical assistance as a "nonlending service" by the Bank's country opera- tions units. The World Bank Institute should play the lead role in disseminating new ideas, through its international policy seminars and training programs for environmental professionals. The Bank can expand pilot projects to large-scale programs through loans for development of pollution management systems that embody key principles of the new approach: focus, transparency, community participation, and regulatory instruments that leverage the economic incentives faced by polluters. To be successful, these 149 GREENING INDUSTRY: NEW ROLES FOR COMMUNITIES, MARKETS, AND GOVERNMENTS operations should promote a clear view of environmental objectives, cost-effective instruments for achieving those objectives, efficient gathering and analysis of appropriate environmental information, and, certainly not least, a strong capacity to enforce regulations when necessary. Although the Bank also provides direct financing for pollu.ion control, it has learned that subsidizing abatement investments by large, individual polluters is seldom the best way to control air and water emissions. Such polluters will generally mobilize their own re- sources to abate pollution if regulators properly leverage MAC and MEP. The major exception is construction of sewerage facilities: Household sewage remains a prime source of health damage in most poor countries, and the Bank provides support when communities cannot issue their own bonds to finance municipal sewerage and wastewater treatment systems. Further research will be needed to de- termine whether, and under what conditions, the Bank and other lend- ing institutions should also finance common treatment facilities for in- dustrial development parks and other areas where factories cluster. The Bank can also promote the new approach through its lend- ing operations that support national economic reforms. These opera- tions provide an excellent opportunity for strengthening environmen- tal agencies' ability to measure changes in environmental quality, identify serious pollution sources, and employ formal and informal regulatory instruments to counter excessive emissions. They can also promote the development of new links between economic and envi- ronmental ministries, thereby increasing the capacity of partner coun- tries to cope with the environmental consequences of future eco- nomic changes. In summary, the coming decade will offer the World Bank many opportunities to assist its partner countries in controlling industrial pollution while working to eliminate poverty. The Bank can promote the new model by encouraging innovative experiments, disseminat- ing their results, expanding the success stories to national programs, and ensuring that economic reform programs incorporate environ- mental concerns. Through all of these channels, the Bank's activities can hasten the greening of industry in many poor countries. 150 www.worldbank.org/nipr CD-ROM Enclosed "Rebutting the conventional wisdom that developing countries lack the infrastructure and the resources to control pollution, Greening Industry shows how a new combination of information-intensive, citi- zen-empowering strategies can produce impressive results. With the cost of collecting and disseminating information falling, this is a time- ly and tremendously exciting book.' _Tom Tietenberg, Mitchell Family Professor of Economics, Colby College "Greening Industry is a terrific report: it makes clear that the pes- simistic prognoses for environmental degradation as industrialization proceeds in the developing world are by no means inevitable. Based on years of research and direct experience, it provides both lucid con- ceptual descriptions of new models for addressing pollution and fas- cinating reports on their use in a number of developing countries. There is, for example, a clear explication of how pollution taxes work, along with four case studies of their effective use in Colombia, the Philippines, Malaysia, and China. Greening Industry carries an w 4 & _ important and hopeful message in terms of explicit strategies for achieving sustainable development. -Wallace Oates, Professor of Economics, University of Mary and "I think Greening Industry will prove to be an important book for a number of reasons. It is eminently realistic about environmental management in developing countries without being fatalistic. It acknowledges that there are usually a host of barriers to the applica- tion of conventional pollution control policies in poor countries, but presents compelling evidence that innovative policies can and do work despite these barriers. Finally, although the book marshals an impressive number of academic studies, the exposition is lively and accessible to a broad audience. - -Allen Blackman, Fellow, Quality of the Environment Division, Resources For the Future 1818 H Street, N.W. B ~~~~~~~~~Wasnington, D.C. 20433, U.S.A. 9 0 0 0 0 r fTelephone: 202 477 1234 Tee:Facsimile: 202 477 6391 R Telex: ~~~~~~MCI 64145 WORLDBANK MCI 248423 WORLDBANK Internet: www.worldbank.org 9 780195 211276 E-mail: books@worldbank.org Our dreaenmis a ISBN 0-19-521127-8 world free of poverty