I I WORLD UANK. TECHNICAL PAPER NO. 530 Environmnent Series F THE WORLD BANK WTP538 L Work in progress December 2002 for pubilc discussion D Generating Public Sector Resources to Finance K Sustainable Development s--: ---- -: Revenue and Incentive Effects -- Stfn-agoa- - . . ~~~~~~~Roberto Mlartin-Hurtado s s - . - - - ~~~~~~AMuthukwnara iVani ; - - . . ~~~~~Patricia Si/va - -- .. ... . - . . - - I w~ ~~~ . ..- _ Recent World Bank Technical Papers No 428 C Mark Blackden and Chitra Bhanu, Gende,, Growth, and Poverty Reduction Special Programl of Assistance for Africa, 1998 Status Report on Poverty in Sub-Saharan Aft ica No. 429 Gary McMa-hon, Jose Luis Evia, Alberto Pasc6-Font, and Jos6 Miguel Sanchez, Ani Eniviroznre,iial Study of Artisanal, Small, and Mediumn Minzing in Bolivia, Chile, and Peru No 430 Maria Dakolias, Court Performance around the World A Comparative Perspective No 431 Severin Kodderitzsch, Refornis in Albanizan Agriculturie Assessing a Sector in Transitioni No 432 Luiz Gabriel Azevedo, Musa Asad, and Larry D Simpson, Managermenit of Water Resources Bulk WVater Pricing in Btazil No 433 Malcolm Rowat and Jose Astigarraga, Latin Americani Insolvency Systemls A Coniparative Assessnient No 434 Csaba Csaki and John Nash, eds, Regional and lnternatioiial Trade Policy Lessons for the EU Accession in the Rural Sector-WVorld Bank/FAO Wotkshop, June 20-23, 1998 No 435 lain Begg, EU Investmiient Grants Reviezv No 436 Roy Prosterman and Tim Hanstad, ed, Legal lmpedinlents to Effective Rural Laud Relations in Eastern Europe and Central Asia A Comparative Perspective No 437 Csaba Csaki, Michel Dabatisse, and Oskar Honisch, Food and Agricultuire in the Czech Republic Fiom0 a "Velvet" Transition to the Challenges of EU Accessioii No 438 George J Borjas, Econominc Research on the Deteriniiianats of Imumigration- Lessons for the Europeami Unilon No 439 Mustapha Nabli, Finanicial Iiitegration, Vilnlterabilmties to Crisis, and EU Accession in Five Cemitral European Coun1tries No 440 Robert Bruce, loanrus Kessides, and Lothar Kneifel, Overcominng Obstacles to Liberalization of the Telecomii Sector in Estonia, Poland, the Czeclh Repuiblic, Slovenia, and Hunlgary An Overviezv of Key Policy Concem 1ls an1d Potential Ilitiatives to Facilitate the Traiisition Process No 441 Bartlomlej Kaminski, Hlungary Foreign Trade lssues in the Conte.xt of Accession to the EU No 442 Bartlomiej Kaminski, The Role of Foreign Direct Investment and Trade Policy in Poland's Accession to the Eiiropeaii Union No 443 Luc Lecust, John Elder, Christian Hurtado, Frangois Rantrua, Kamnal Siblini, and Maurizia Tovo, DeAlIlStifying MIS Guidelines for Maanagenientt 1nforniatioui Systenis in Social Fuinids No 444 Robert F Townsend, Agricultutral Incentives in Suib-Saharan Africa Policy Cliallemiges No 445 lan Hill, Forest Management in Nepal Economics of Ecology No 446 Gordon Hughes and Magda Lovel, Econotiic Refom in and Environmlemital Perforniaiice in Traiisitioii Econlomties No 447 R Maria Saleth and Ariel Dinar, Evaliiatimig Water Institutionis and Water Sector Perforniance No 449 Keith Oblitas and J. Raymond Peter in association with Gautam Pingle, Halla M Qaddumi, and Jayantha Perera, Transferring lrigation Maiiagenient to Fariiiers in Andhva Pradesh, Imdia No 450 Andres Rigo Sureda and W'Valeed Haider Malik, eds, Judicial Challenges in the Nezo Milleniuzitim Proceedinigs of the Secoiid Suiiiinit of the Ibtero-Anier-ican Supremiie Couirts No 451 World Bank, Privatizatiomi of the Power- arid Natural Gas Industries in Hungary and Kazakhstan No 452 Lev Fremkman, Daniel Treisman, and Stephen Tutov, Smibnatioianl Budgeting in Russia Preeinptiiig a Potential Crisis No 453 Bartlomiej Kaminski and Michelle Riboud, Foreign Investment and Restriictlirimig T'he Evidence fromn Huingary No 454 Gordon Hughes and Julia Bucknall, Poland Comiiplying zvith ELI Environmznemtal Legislature No 455 Dale F Gray, Assessment of Corporate Sector Valuie anid Vulnerability' Litiks to Exchange Rate and Fimiamicial Crises No 456 Salman M A Salman, ed, Groundwater Legal and Policy Perspectives Proceedings of a World Bank Semunai No. 457 Mary Canmng, Peter Moock, and Timothy Helemak, Reforming Education in the Regions of Russia No 458 John Gray, Kazakhstan A Review of Farin Restructurilng No 459 Zvi Lerman and Csaba Csaki, Ukraine Reviezw of Farm Restructuiritng Experiences No. 460 Gloria La Cava and Rafaella Y. Nanetti, Albania Filling the Vulnerability Gap No 461 Ayse Kudat, Stain Peabody, and Caglar Keydei; eds, Social Assessnlent and Agricultuiral Refori iiin Cenitral Asia an d Tuirkey No 462 'F. Rand, J Haukohl, and U Marxen, Municipal Solid Waste Iicineratioll RequirementsfJo a Successfiul Pioject (List continues on the inside back cover) WORLD BANK TECHNICAL PAPER NO. 538 Environment Series Generating Public Sector Resources to Finance Sustainable Development Revenue and Incentive Effects Stefano Pagiola Roberto Martin-Hurtado Priya Shyamsundar Muthukumara Mani Patricia Silva The World Bani Washington, D.C. i 2002 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 All rights reserved. 1234040302 Technical Papers are published to communicate the results of the Bank's work to the development com- munity with the least possible delay. The typescript of this paper therefore has not been prepared in accor- dance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is copyrighted. No part of this work may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or inclusion in any information storage and retrieval system, without the prior written permission of the World Bank. The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint, please send a request with complete information to the Copy- right Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. ISBN: 0-8213-5384-5 ISSN: 0253-7494 Stefano Pagiola is Senior Environmental Economist in the Environment Department at the World Bank. Roberto Martin-Hurtado is Junior Professional Associate in the Environment Department at the World Bank. Priya Shyamsundar is Economist in the Environment Department at the World Bank. Muthuku- mara Mani is Economist in the Fiscal Affairs Department at the International Monetary Fund. Patricia Silva is Consultant in the Environment Department at the World Bank. Cover photos from World Bank; design by Jim Cantrell. Library of Congress Cataloging-in-Publication Data has been requested Contents Foreword vi Abstract vii Acknowledgments viii Abbreviations ix 1. Introduction 1 2. Freeing Up Existing Resources 3 2.1 Reducing Energy Subsidies 3 2.2 Reducing Water Subsidies 20 Irrigation 20 Domestic water 25 3. Generating New Resources 33 3.1 Capturing Natural Resource Rents 33 Rationale 33 Forests 34 Bioprospecting 43 Estimates of Potential Resource Rent Capture 46 3.2 Charging for Services 47 Protected areas 47 Municipal solid waste 60 3.3 Imposing Green Levies 70 3.4 Developing Innovative Conservation Financing Mechanisms 85 4. Summary 91 References 93 Appendix A. International Financial Flows 103 A. 1 International Financial Flows to Developing Countries 103 A.2 Environmental Conservation Financing 105 A.3 New Mechanisms for Financing Environmental Conservation 109 A.4 Conclusions 113 Appendix B. Welfare Economics of Subsidies: The Case of Electricity 117 Appendix C. Statistical Appendix 121 Tables 2.1 Annual costs of energy subsidies, 1995-98 4 2.2 Energy subsidies in eight developing countries 5 2.3 Electricity subsidies in developing regions, 1999 7 2.4 Fuel subsidies in developing regions, 1999 10 2.5 Environmental costs of fossil fuels in six developing country cities 12 2.6 Access to electricity in selected countries, 1997-1998 14 2.7 Effects of phasing out energy subsidies in eight developing countries 15 2.8 Estimated effects of removing electricity subsidies in developing countries 17 2.9 Impacts of fuel subsidy removal and fuel taxation 18 2.10 Irrigation water charges in selected countries 22 2.11 Irrigation water charges and cost recovery 22 2.12 Average public expenditures on irrigation for a sample of countries 23 2.13 Average unit costs for irrigation 23 2.14 Water subsidies on delivery tariffs in selected countries 28 2.15 Utilities that do not meter users 29 3.1 Potential rents from timber in forest-rich countries 35 3.2 Forestry policy instruments and revenue generated in selected countries 38 3.3 Magnitude of illegal forest activities in the Asia Pacific region 42 3.4 Estimated potential resource rents, 1999 46 3.5 Tourism receipts 48 3.6 Park fees in selected developing countries 50 3.7 Estimates of foreign visitor willingness to pay for protected area visits 52 3.8 Estimates of price elasticity of demand for visits to protected areas 54 3.9 Major environmental impacts of tourism 55 3.10 Visitor numbers and fees for selected Belizean parks and protected areas 58 3.11 Implicit subsidies in municipal solid waste services, 1999 63 3.12 Options for enhancing MSWM cost recovery 65 3.13 Tipping fees in selected developing countries 67 3.14 Potential for revenue generation and the cost of expanding service 67 3.15 Environmental taxes and charges in selected economies in transition 72 3.16 Environment and natural resources in Russia's 1999 budget 76 3.17 Pollution charge collection in China, 1992-99 77 3.18 Market-based instruments in Latin America and Caribbean 80 A.1 Total flows of net resources from OECD DAC member countries 106 A.2 Debt for nature swaps, 1990-1997 111 B.1 Regional breakdown of electricity subsidies and welfare losses 118 C.1 Estimated electricity subsidies and potential efficiency gains from subsidy phase-out 118 C.2 Estimated gasoline and diesel subsidies and potential effects on public budget from price reform, 1999 124 C.3 Estimated potential resource rents by country, 1999 128 Figures 2.1 Electricity subsidies in developing countries, 1999 7 2.2 Average fuel prices in developing countries by group, 1999 10 2.3 Connection fees and cost of construction for in-home water service 26 2.4 Comparison of vendor and utility water prices in selected countries 30 3.1 Flow of timber through the forest sector and application of taxes 36 3.2 International tourist arrivals 48 iv 3.3 Intemational tourism by purpose of visit, 1998 49 3.4 Visitors and revenue in Costa Rica's national parks 57 3.5 Public expenditure in MSWM, 1994-95 62 3.6 Revenues from environmental levies in selected CEE countries 73 3.7 Tax revenues from different sources 74 3.8 Charges from environmental user fees by LLDA, Philippines 78 3.9 Institutional framework for payments for environmental services 87 A. 1 Overview of public and private financial flows 104 A.2 Total Official Development Financing to developing countries 105 A.3 Environmentally-targeted bilateral commitments 107 A.4 Environmentally-targeted multilateral commitments 107 A.5 Total Environmental ODA, 1998 108 A.6 Sources of financing for biodiversity conservation in Latin America, 1990-97 109 B.1 Welfare economics of subsidies: the case of electricity 117 Boxes 2.1 The effects of coal mispricing in India 6 2.2 The financial burden of electricity subsidies in India 8 2.3 Quasi-fiscal energy activities in the Former Soviet Union 9 2.4 Differential pricing of fuels 11 2.5 Income and health effects of access to modern energy 13 2.6 Leakage in electricity subsidies in Guatemala 14 2.7 Petroleum product subsidies and poverty in India 15 2.8 Approaches to energy subsidy reform 16 2.9 Optimal taxation of fuels 17 2.10 Charging for irrigation water 21 2.11 Irrigation in Central Asia 24 2.12 Explicit water subsidies 27 2.13 Water subsidies: the case of increasmg block tarnffs 30 3.1 "Conservation concessions" in the Guyana Shield region 40 3.2 Using revenue from fees to improve park management 51 3.3 Measuring visitor willingness to pay 53 3.4 Regional differences in landfilling practices 61 3.5 Regional differences in the use of cost recovery instruments 66 3.6 The disastrous consequences of improper MSWM in Manila 69 3.7 Definition of environmentally related taxes and charges 70 3.8 Is there a "double-dividend"? 71 3.9 Carbon taxes in Slovenia 75 3.10 Current and potential environmental taxes in South Africa 81 3.11 Polish National Fund for Environmental Protection 83 3.12 The simple economics of payments for environmental services 85 3.13 Local initiatives on payments for enviromnental services in Latin America 86 3.14 Payments for environmental services at the World Bank 86 A.1 The French GEF and ODA: Separate but complementary sources of funding 110 v Foreword Ensuring that sufficient resources will be available This paper is one of the results of that co- to finance the internationally agreed development operation. It does not aim to cover all of the goals was at the heart of the discussions at the complex issues related to Official Development World Summit on Sustainable Development Assistance (ODA) and Foreign Direct Investment (WSSD), held in Johannesburg from August 26 to (FDI). Its specific objective is to explore potential September 4, 2002. Resources will have to come avenues to generate more resources for the public from both the public and private sectors, and inno- sector to invest in sustainable development. This vative approaches will have to be designed to paper, together with UJNEP's companion paper on make financing mechanisms more effective. mobilizing private sector resources, was used to In this context, staff of the World Bank, the prepare a joint synthesis report, Financing for International Monetary Fund (IMF), and the Sustainable Development. United Nations Environment Programme (UNEP) We hope that this paper will be a valuable have worked together to explore options for input to discussions on financing for sustainable financing sustainable development. development. We welcome your comments. Kristalina Georgieva Director, Environment Department vi Abstract Achieving the objective of sustainable develop- Several main lessons emerge: ment will require considerable resources and crea- * There is substantial potential to generate addi- tive use of existing and additional resources. tional public sector resources, although the Although Official Development Assistance amounts vary considerably across sectors and (ODA) will continue to play an important role in countries. The most important potential source financing sustainable development, it is likely to of additional revenue comes not from efforts to be insufficient to meet the tremendous needs even generate new revenues, but from freeing up with with the increases in flows agreed at the availableresources byimprovingthe efficiency Financing for Development (FfD) conference. with which they are spent-in particular, by Other sources of resources must also be sought. reforming subsidies that are expensive and, Where will these resources come from and how rorming env ienthare Evenswhend, can hey e apliedto ost ffecivel addess often, envirounmentally harmful. Even when the can they be applied to most effectively address sums involved appear limited, reform can help sustainability challenges? make subsectors financially self-sustaining This paper discusses how developing coun- rather than wholly dependent on the public tries can generate some of the resources they need purse. for sustainable development. Developing country * Reforms can have important incentive effects. governments already expend significant amounts In addition to generating new resources or free- of resources on a variety of activities. But the evi- ing up existing ones, many of the reforms dence suggests that there is sometimes substantial discussed in this paper can also help reduce scope for them to generate additional resources environmental damage, by providing incen- and-perhaps more important-to free up sub- tives that tend to discourage environmentally- stantial amounts of resources which are currently harmful activities and encourage more sustain- being used inefficiently. able ones. In many cases this outcome would The primary objective of this paper is to be as or more important as the resource gene- attempt a rough scoping of the magnitude of re- ration itself. sources that might be generated or freed up by a * Reform is not anti-poor. Though many policies variety of public sector actions. The paper begins claim to be pro-poor, they often are not. by examining the potential to reform many exist- Reducing these subsidies would, therefore, not ing policies which are not only costly but often only not harm the poor, it might well benefit unsustainable and environmentally damaging. them if the resources that are freed up are used Reforming them would both free up resources for in more appropriate ways. However, any policy more valuable uses, and often directly improve reform must take considerable care not to in- sustainability. The following section then turns to advertently harm the poor. potential means for generating new financial * Reform will require political will, but also flows, by capturing a greater share of the rents good governance, capacity building, and from natural resources and by instituting 'green' investment. levies. Wherever possible, we attempt to estimate . One size does not fit all. There is substantial potential revenue flows or resource savings, variation in the needs, opportunities, and con- although data limitations sometimes prevent this. straints facing different developing countries. Numerical estimates are complemented with case Even within countries, there is substantial studies detailing the experience of specific variation in capacity to implement public countries. reform programs across regions or sectors. vii Acknowledgments This paper was prepared by a joint team from the David Hanrahan, Agi Kiss, and Gary Stuggins Policy and Economics Team in the Environment from the World Bank, Michael Keen from the Department of the World Bank and the Fiscal IMF, and Paul Clements-Hunt, Charles Arden- Affairs Department of the International Monetary Clarke, Danielle Hirsch, and Mareike Hussels Fund (IMF) headed by Stefano Pagiola. The from UNEP. members of this team, with their primary respon- This paper has also benefited from numerous sibilities, were: comments and discussion made in the course of a * Hiba Ahmed (water) month-long open "e-discussion" on the web, * Katharine Bolt (natural resource rents) organized by the World Bank Institute (WBI), * Kirk Hamilton (natural resource rents) which attracted almost 200 participants from all * Muthukumara Mani (environmental levies) over the world. We are grateful to all who * Roberthumartin-Hurtado (enelrgy, wastemana- commented, to Vesselina Hekimova and Cary * Roberto Martin-Hurtado (energy, waste mana- Anne Cadman of WBI for organizing the gement, environmental levies) e-discussion, and to Patricia Silva of the * Stefano Pagiola (user fees, innovative conser- University of California at Santa Barbara for vation mechanisms) moderating it. The paper also benefited from * Priya Shyamsundar (environmental levies, comments made by participants in the Conference forests, international resource flows) on Financing the Environmental Dimension of * Patricia Silva (user fees, forests, international Sustainable Development organized by the OECD resource flows) Global Forum on Sustainable Development in The paper benefited from detailed comments Paris on April 24-26, 2002. and suggestions provided by Anil Markandya and We are grateful for the financial support Bob Schneider. Comments and suggestions wre provided by the Government of Norway for the also provided by Kristalina Georgieva, Robert preparation and publication of this report. Clement-Jones, Charles Feinstein, Arvind Gupta, viii Abbreviations ADB Asian Development Bank MSWM Municipal solid waste management BOD Biological Oxygen Demand NGO Non-Governmental Organization CDM Clean Development Mechanism NIH National Institute of Health CEE Central and Eastern Europe NSF National Science Foundation c/kwh US cents per kilowatt-hour O&M Operations and Maintenance Ci Conservation International OA Official Assistance CO2 Carbon dioxide ODA Official Development Assistance CRS Creditor Reporting System OECD Organisation for Economic CVM Contingent valuation method Cooperation and Development DAC Development Assistance Committee OFF Other Financial Flows EAI Enterprise for the Americas Initiative OLADE Organizaci6n Latinoamericana de EBRD European Bank for Reconstruction and Energia (Latin American Energy Development Organization) EU European Union PACT Protected Areas Conservation Trust FDI Foreign Direct Investment PES Payment for Environmental Services FFEM Fonds Francais pour I 'Environnement S02 - Sulphur dioxide Mondial (French GEF) TCM Travel cost method GDP Gross Domestic Product TNC The Nature Conservancy GEF Global Environment Facility UFA Unite Forestiere d'Amenagement ha hectare (Forest Management Unit, IBT Increasing Block Tariffs Cameroon) ICBG International Cooperative Biodiversity UNCED United Nations Conference on Groups Environment and Development IDB Inter-American Development Bank UNDP United Nations Development GTZ Deutsche Gesellschaftfuir Technische Programme Zusammenarbeit (German UNEP United Nations Environment Corporation for Technical Programme Cooperation) USAID United States Agency for International IEA International Energy Agency Development IMF International Monetary Fund USDA United States Department of INBio Instituto Nacional de Biodiversidad Agriculture (National Biodiversity Institute of VAT Value Added Tax Costa Rica) WRI World Resources Institute lcd liters per capita per day WSP Water and Sanitation Program LPG Liquefied petroleum gas WSSD World Summit on Sustainable MDG Millennium Development Goal Development MINAE Ministry of the Environment and WTP Willingness to Pay Energy (Costa Rica) WWF World Wildlife Fund All monetary values are in 1999 US dollars, unless otherwise indicated. A billion is a thousand million. ix 1. Introduction The 1992 Earth Summit in Rio de Janeiro clearly The primary objective of this paper is to established the objective of sustainable develop- attempt a rough scoping of the magnitude of ment. Definitions vary, but the spirit is captured resources that might be generated or freed up by a by the notion put forth by the Development variety of public sector actions. The paper begins Assistance Committee of the OECD that by examining the potential to reform many exist- "Sustainable development entails balancing the ing policies which are not only costly but often economic, social and environmental objectives of unsustainable and environmentally damaging. the society-the three pillars of sustainable deve- Reforming them would both free up resources for lopment-integrating them wherever possible, more valuable uses, and often directly improve through mutually supportive policies and practi- sustainability. The following section then turns to ces, and making trade-offs where it is not potential means for generating new financial possible. This includes, in particular, taking into flows, by capturing a greater share of the rents account the im-pact of present decisions on the from natural resources and by instituting 'green' options of future generations." Ten years later, levies. Wherever possible, we attempt to estimate that objective remains the goal of the world potential revenue flows or resource savings, community, having been most recently restated in although data limitations sometimes prevent this. the Millennium Declaration and the Millennium Numerical estimates are complemented with case Development Goals (MDGs) and at the World studies detailing the experience of specific Summit on Sustainable Development in countries. Johannesburg. Throughout the paper, we focus on Achieving this objective will require consi- approaches that developing countries can under- derable resources and creative use of existing and take largely at their own initiative, and do not additional resources (Devarajan and others, 2002; discuss the various proposals that have been made World Bank, 2002a). Official Development for funding mechanisms that would require wide- Assistance (ODA) will continue to play an impor- spread international agreement (such as a Tobin tant role in financing sustainable development. tax). Developing countries would not need to wait Even with the increases in flows agreed at the for a global consensus to undertake the measures Financing for Development (FfD) conference in outlined here. We also focus on the potential to Monterrey, Mexico, however, ODA is likely to be generate resources within sectors that are insufficient to meet the tremendous needs. Other particularly critical for sustainable development, sources of resources must also be sought. Where and where the potential for 'win-win' solutions will these resources come from and how can they that reduce environmental damage while genera- be applied to most effectively address sustain- ting resources is greatest. ability challenges? This paper seeks to contribute In addition to the resources they might gene- to answering these questions. rate, many of the actions examined in this paper This paper discusses how developing coun- are also likely to contribute to sustainable deve- tries can generate some of the resources they need lopment directly, by changing the incentives for for sustainable development. Developing country environmental and natural resource use. Refor- governments already expend significant amounts ming energy policy, for example, would not only of resources on a variety of activities. But the evi- reduce a very significant drain on government dence suggests that there is sometimes substantial resources but also lead to reduced energy use, thus scope for them to generate additional resources diminishing the attendant pollution. A second and-perhaps more important-to free up sub- objective of the paper, then, is to examine the stantial amounts of resources which are currently nature and extent of such changes in incentives. being used inefficiently. As we will see, in some cases this effect may well be more important than revenue generation. 2 Generating Public Sector Resources to Finance Sustainable Development Many of the actions discussed in this paper per capita terms in developed countries, and to are likely to affect the poor. Indeed, fear of have equally detrimental effects on public finance harming the poor is often used as a justification to and the environment. The report, however, fo- avoid undertaking these actions. We examine cuses on developing countries as their need for these concerns in each case. As we will see, the additional resources to help finance sustainable concerns that reforms would harm the poor do not development, and to get incentives right, is tend to be borne out. Although reforms do have to particularly acute. be undertaken with some care, in many cases they It is of course impossible for a paper such as seem likely to be beneficial rather than harmful this to cover the myriad issues which affect sus- for the poor. tainable development. We focus narrowly on the Many of the paper's conclusions are equally issue of generating additional resources to finance applicable to developed countries. Subsidies to sustainable development, and do not discuss how energy, for example, tend to be much higher in these resources are to be employed. 2. Freeing Up Existing Resources One of the biggest potential sources of resources leave the institutions overseeing the sector for many developing countries would come from chronically short of funds. Consequently, their freeing up existing resources for better uses rather ability to manage, maintain, and expand the than from generating new resources. Many sector's infrastructure diminishes, until the infra- important goods and services are mis-priced as a structure crumbles from neglect-and sometimes result of policies in place. In many cases, they are collapses entirely. priced too low. Predictably, this results in over- An equally common aspect of this story is use of these goods, with consequent adverse that although many of the subsidies discussed effects on efficiency and on the environment. below are often justified as protecting the poor, Reforming them would free up substantial there is substantial evidence that they are in fact amounts of budgetary resources, as well as often regressive. This is due in part to leakage (the improving the efficiency of the sector and, often, non-poor reap some of the benefit when consu- reducing environmental pressures. ming the subsidized good; subsidized kerosene, We define subsidy policies broadly, as any for example, is also bought by the non-poor), and policies whose effect is to reduce the costs of an in part to mistargeting (the poor fail to benefit activity relative to what they would have been in from the subsidy because they do not consume the the absence of the policies. Some subsidies are subsidized good; few of the poor, for example, are explicit, such as selling electricity to consumers connected to the electric grid). Moreover, the for less than its cost of production, with parlous financial condition of public utilities that government funds covering the deficit of the results from these subsidies often prevents them electricity producer. Other subsidies are implicit, from expanding coverage of services such as such as selling fuel domestically for less than its electricity and clean water, leaving the poor using value on the world market, which results in the more expensive and often inferior substitutes. government forgoing potential income from that sale. Under this approach, the extent of subsidies 2.1 Reducing Energy Subsidies for input use is given by the difference between what users pay for that input and what they would Traditional justifications for energy subsidies have paid in the absence of all policies, multiplied in industrialized countries include ensuring by the quantity used. security of energy supply, increasing compe- A depressingly common story emerges in titiveness of industries, and protecting employ- each of the following sections: a sector is highly ment. In developing economies, the list also subsidized, either explicitly or implicitly. Typi- includes social protection and economic develop- cally, consumers are not charged at all for the ment. While energy subsidies in industrialized capi-tal costs of providing goods and services countries tend to favor producers, in developing such as electricity or irrigation water, and are only countries they tend to favor consumers. charged a small proportion of operating costs. Subsidies are not always bad; some can be This leads to two parallel-and entirely pre- successfully used to correct for market failure's. dictable-results. First, the good being supplied is However, energy subsidies are increasingly used very wastefully, since low prices provide no identified as perverse (Myers and Kent, 2001; van incentive to conserve. In addition to the inherent Beers and de Moor, 2001). It has long been inefficiency it causes, this waste can also have recognized that energy subsidies encourage waste- important adverse environmental effects: ex- ful energy use and prompt negative environmental cessive energy use results in high levels of air impacts.' In addition, energy subsidies in develo- pollution, and excessive water consumption places ping countries often fail to reach their target of pressures on aquatic ecosystems. Second, high assisting the poorest, instead being consumed by levels of use coupled with minimal cost recovery richer groups. The welfare economics of electri- 3 4 Generating Public Sector Resources to Finance Sustainable Development city subsidy are discussed in more detail in 'effective' price being lower than the long run Appendix B. The need for reform, however, marginal cost of provision, and hence to a makes itself more pressing when it becomes subsidy. In what follows, we attempt to catch evident that these subsidies are a drain of the the combined effect of low tariffs and public budget that is increasingly unsustainable. subsidized infrastructure and fuels, but will not As noted above, we define energy subsidies address the excessive losses and non-payment as measures that lead the prices of energy elements. products to be lower than they 'should' be.2 This e In the case of petroleum products (gasoline and lends itself to use of the price-gap approach to diesel), we will go one step further. These measure the size of subsidies, wherein the prices products are not usually subsidized, in the currently paid by consumers are compared to sense of their price not covering the economic reference prices. The selection of reference prices cost of production. However, differential is straightforward in theory, but contentious in taxation of petroleum products (within one practice. For internationally traded goods (such as country and across countries4) leads itself petroleum products), the relevant reference price naturally to the analysis of the potential would be the international price adjusted by revenues of matching a certain benchmark, distribution costs (augmented in the case of net oil whether this is designed to cover road transport importers, and reduced in the case of net oil sector costs, or to help financing the general exporters).3 For non-traded goods (such as, gene- budget. rally, electricity) the reference price would the Current energy subsidies long run marginal cost of provision. In the case ofelecricity,settintarEnergy is among the most heavily subsidized In the case of electricty, setting tariffs below sectors in the world. According to the figures cost of provision, supplying the sector with copldbvaBerandeM r(20) subsidized infrastructure and fuels, allowing ener accounted for over 20 ercent of world for excessive losses in distribution (whether gy p due to crumbling infrastructure or to theft), and subsidies or nearly 1 percent of world GDP in the not enforcing payment, all contribute to the mid-I990s. Developing countries have been extra- Table 2.1: Annual costs of energy subsidies, 1995-98 (US$ billion per year) OECD Non-OECD World Fossil fuels Coal 30 23 53 Oil 19 33 52 Gas 8 38 46 Subtotal 57 94 151 Electricity a 48 48 Nuclear 16 Negligible 48 Non-payments and bail-out" 0 20 20 Total 82 162 244 Per capita (US$) 88 35 44 Notes: D Subsidies for electricity in OECD countries have been attributed to fossil fuels according to the corresponding shares in electricity generation b Subsidies from non-payments and bail-out operations have not been attributed to energy sources Source. Van Beers and de Moor, 2001. Generating Public Sector Resources to Finance Sustainable Development 5 ordinarily keen on subsidizing energy, with the use. On average, end-use prices in these countries energy sector absorbing nearly 50 percent of are about 20 percent below their opportunity cost subsidies, representing some 3 percent of GDP. or market-based reference levels, despite The electricity subsector attracts the largest share substantial progress in recent years to move of subsidies (see Table 2.1). towards more rational pricing and market-based A recent lEA (1999) study confirms that policies (see Table 2.2). This results in a total of pervasive under-pricing of energy resources over US$83 billion in subsidies for energy occurs in eight of the largest countries outside the products, including over US$7 billion for OECD: China, India, Indonesia, Iran, Kazakhstan, petroleum products, US$41 billion for electricity, Russia, South Africa, and Venezuela, which US$26 billion for natural gas, and nearly US$9 collectively represent a quarter of world energy billion for coal. However, as these are all large Table 2.2: Energy subsidies in eight developing countries South China Russia India Indonesia Iran Africa Venezuela Kazakhstan Total Value of subsidies (billion US$) Gasoline 0.00 0.59 0.00 0.00 0.50 0.00 0.28 0.00 1.37 Diesel 0.00 0.00 0.00 0.30 0.76 0.00 0.06 0.00 1.13 LPG 0.00 0.00 0.51 0.00 0.16 0.00 0.04 0.00 0.71 Kerosene 0.00 0.00 1.58 0.44 0.63 0.01 0.01 0.00 2.67 LFO 0.00 0.05 0.00 0.25 0.45 0.00 0.04 0.00 0.79 HFO 0.00 0.00 0.00 0.01 0.34 0.00 0.01 0.00 0.37 Electricity 22.03 11.17 4.60 0.00 0.41 1.14 0.96 1.19 41.49 Natural gas 0.49 21.67 0.55 0.19 2.15 0.00 1.17 0.09 26.32 Steamn coal 3.28 0.00 1.24 0.00 0.00 0.04 0.01 0.14 4.71 Coking coal 3.77 0.00 0.27 0.00 0.00 0.00 n.a. 0.00 4.05 Total 29.58 33.49 8.75 1.20 5.41 1.19 2.56 1.42 83.60 % GDP 3.1 7.6 2.1 0.5 5.3 0.8 2.6 6.3 Subsidy rates (%) Gasoline 0.0 9.3 0.0 0.0 59.4 0.0 26.6 0.0 Diesel 0.0 0.0 0.0 40.2 93.9 0.0 35.9 0.0 LPG 0.0 0.0 31.6 0.0 89.7 0.0 26.1 0.0 Kerosene 0.0 0.0 52.6 55.2 89.5 2.0 4.9 0.0 LFO 0.0 1.5 0.0 45.5 82.3 0.0 19.3 0.0 HFO 0.0 0.0 0.0 7.8 88.1 0.0 39.4 0.0 Electricity 38.2 42.0 24.2 0.0 48.1 20.3 63.0 56.6 Natural gas 18.7 46.1 22.5 28.4 77.8 0.0 85.6 55.7 Steam coal 8.3 0.0 13.1 0.0 0.0 8.1 91.9 20.7 Coking coal 73.1 0.0 42.3 0.35 0.0 0.0 - 2.7 Total 10.9 32.5 14.2 27.5 80.4 6.4 57.6 18.2 Note: Figures are in 1999 US billion dollars. Original data are for 1997 (Russia, Indonesia, Iran, South Africa) and 1998 (China, India, Venezuela, Kazakhstan). LPG: Liquefied Petroleum Gas. LFO: Light Fuel Oil. HFO: Heavy Fuel Oil. Source: Adapted from IEA, 1999. 6 Generating Public Sector Resources to Finance Sustainable Development energy producers, it is not possible to extrapolate consumption. Despite the importance of coal the results to the rest of the developing world. consumption from an environmental and health In this section, we focus on two subsectors: perspective, we do not offer a treatment of coal electricity and transport fuels. Electricity subsidies subsidies. Bulky coal is not traded widely, so its represent an important share of energy subsidies. consumption within the developing world is In addition to direct subsidies, electricity also highly localized in a handful of coal-producing benefits from a significant portion of subsidies to countries-mainly India, China, and South Africa coal, oil, and gas, which lower the cost of fuels (IEA, 1999). employed in electricity generation. Box 2.1 Electricity illustrates the importance of including this upstream component in the evaluation of Electricity attracts 30 to 50 percent of energy electricity subsidies. Electricity generation subsidies (van Beers and de Moor, 2001; EA, represents 40 percent of the use of coal, 10 1999). In the IEA's sample of eight countries percent of the use of oil, and nearly 20 percent of electricity subsidies amounted to about half of the use of natural gas in developing countries total energy subsidies (IEA, 1999). In early 1991, (IEA, 2001a). In our analysis, we will attribute electricity subsidies in developing countries were this form of 'indirect subsidization' to the estimated to be US$54 billion, or US$64 billion in electricity sector, so our results will not be constant 1999 dollars (World Bank, 1997a). directly comparable to previous studies.5 For this report, we have carried out a new estimation exercise, combining tariff data for 79 Box 2.1: The effects of coal mispricing in India countries, with a conservative estimate of long run The differential pricing of fuels in India promotes marginal cost. Our tariff data come from a range consumption of the most damaging varieties. While of international sources: IEA (1999; 2001), most coal prices have been deregulated, power station OLADE (2000), EBRD (2001) and World Bank grades of domestic coal are still priced below the cost (1998a; 1999b; 2000f). Based on the estimates of production. The true value of the better grades is offered in IEA (1998) and Rodnick (2000), we use underestimated, while cleaner imported fuels are priced the following costs of generation, in US cents per too high by comparison with domestic coal. In kilowatt-hour (c/kwh): coal 3.6 c/kwh; natural gas addition, prices for the transport of goods by rail, 3.1 c/kwh; nuclear 3.2 c/kwh; hydro 1.9 c/kwh; including coal, are generally subsidized, further and oil 6.7 c/kwh. We use a mark-up for encouraging the use of the domestic product and its high ash content. As a result, (1) Indian coal is transmission and distribution costs of 2 c/kwh for preferred as a fuel for power generation over other industrial customers and 4 c/kwh for non- fuels with lower ash content, (2) the relative prices of industrial customers in our calculations. The different grades of coal give no incentive to coal difference in mark-up is due to the higher cost of producers to increase output of better-quality coals, (3) serving non-industrial customers than industrial distorted prices for transport affect the choice between customers. Industrial customers consume larger pit-head and load-center plants, shifting the location of quantities, requiring less distribution environmental impacts, and (4) demand is maintained infrastructure per kilowatt-hour consumed, and lat artificially high levels. at artificially high levels. can be served at higher voltage levels, reducing Source. WorldBank, 1999a. distribution losses. Our sample accounts for 92 percent of developing world electricity In addition to electricity, we focus on the consumption. In what follows we have scaled up pricing practices of petroleum products used for the results to account for 100 percent of transport, namely gasoline and diesel. The consumption. transport sector represents 40 percent of the use of Our crude measure of subsidy encompasses oil in developing countries (WEA, 2001 a), with Orcuemaueo usd nopse gasoline and diesel accounting for half of this the two main ways in which electricity is subsi- figure Altogether, our analysis of both electricity dized. The first is the sale of fuels (such as natural andure Altranort fuels encopsisef roughely gas and fuel oil) to electricity generators at prices and transport fuels encompasses roughly 40 below the export rice. This issue a.lies to the percent of coal consumption, 30 percent of oil caelof oi export countris The sppe to the case of oil exporting countries. The second main consumption, and 20 percent of natural gas Generating Public Sector Resources to Finance Sustainable Development 7 form of subsidy results from infrastructure costs long run marginal cost, for a total subsidy of not being fully incorporated in electricity prices. US$102 billion. This represents 2 percent of the This may result from financial transfers from the developing world's GDP. Table 2.3 summarizes government, or by not spending money in main- our results by region, and Figure 2.1 illustrates taining existing infrastructure.6 them. The Former Soviet Union, with ample According to our estimates, 51 of 79 deve- access to electricity and very low prices of the loping countries subsidize electricity. The results fuels employed in its generation, leads the indicate that in 1999 the developing world electricity subsidy league. By contrast, Sub- subsidized electricity at a rate of 46 percent of Saharan Africa, with low access to electricity, and Table 2.3: Electricity subsidies in developing regions, 1999 Subsidy Subsidy rate Subsidy (billion US$) NO) (% GDP) Sub-Saharan Africa 4.8 35 1.6 Asia 23.8 20 0.9 Central and Eastern Europe 5.8 21 1.5 Formner Soviet Union 41.3 76 13.7 Latin America and Caribbean 4.6 9 0.3 Middle East and North Africa 21.9 59 3.7 Total 102.4 34 1.7 Notes: The results are based on a sample of 79 countries representing 92 percent of consumption and have been scaled up. Estonia, Latvia and Lithuania have been included in the Central and Eastern Europe grouping. Source: Authors' estimates. Figure 2.1: Electricity subsidies in developing countries, 1999 50 - 4~0 'n 3.2 CO620W _l I, --7 S - 10 60- 6 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - -- - 4 0 - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - -- - - -- - Sub-Saharan Asia Central and Latin America Middle East and Former Soviet Africa Eastern Europe and Caribbean North Africa Union Source- Authors' estimates. 8 Generating Public Sector Resources to Finance Sustainable Development Latin America and the Caribbean, which has sufficiently large that the difference is unlikely to restructured the sector in the past decade, account be due solely to differences in methodology. This together for less than 10 percent of the US$102 is in contrast to the trend observed in other energy billion. products. For example, coal subsidies have The results of our estimations for individual experienced a rapid decrease in the 1990s, mainly countries are shown in Appendix B. Extreme care due to policy changes in China (chiefly, the partial must be taken in interpreting these results. The liberalization of the coal market) and the global benchmarks used in the estimates cannot economic downturn experienced in the 1990s by capture individual countries' characteristics. Russia and other transition countries (IEA, 1999). Although we are confident on our results at the - Although the absolute amount of subsidies global level, the degree of accuracy of the results appear to have increased, subsidy rates appear to diminishes at more disaggregated levels. In the have declined somewhat. In 1991, subsidy rates case of Sub-Saharan Africa, the results are driven were over 50 percent in transition countries, over by one country (South Africa represents over 70 40 percent in China and India, and over 30 percent percent of regional consumption), and so even in other countries (World Bank, 1997a). Our interpretation at a regional level requires figures suggest that while the situation in the consideration of the estimation's shortcomings. transition countries has not have improved, the Our approach for calculating subsidies rest of the developing world may have cannot account for cross-subsidies. Cross- experienced significant reductions in subsidy rates subsidies happen when a group of consumer is and in the size of subsidies as a percentage of charged a price above the cost of production, GDP. That the absolute amount of subsidies has while another group is charged a price below cost. nevertheless increased is due to a 25 percent In the case of electricity, industrial consumers increase in electricity consumption in developing typically cross-subsidize non-industrial ones. As a countries over the 1991-1999 period (in spite of consequence, the two effects cancel each other. the economic downturn in Eastern Europe and the Although cross-subsidization generates allocative Former Soviet Union). inefficiencies, decision-makers often justify it on equity grounds. To evaluate the importance of Box 2.2: The financial burden of electricity cross-subsidies in the power sector in developing subsidies in India countries, we compare the prices for industrial and The growing financial burden of electricity non-industrial consumers using a simple rule. We subsidies in India is unsustainable. In 1992-93, the total define cross-subsidization as occurring when the financial losses of the power sector came to US$1.7 price for non-industrial customers is below the billion. In just three years, these losses doubled. Three years later, they had doubled again. In 2001 combined rc 7 i state utility losses are estimated at somewhat more than 1.5 c/kwh. Our data indicate that while 51 US$5 billion a year. If current trends continue, in countries in our sample (or 65 percent of another three years, state utility financial losses will countries) subsidize electricity, in 65 countries of reach US$10 billion a year. To put this magnitude of our sample (or 83 percent of countries) there is losses into perspective, US$5 billion is half of what all cross-subsidization. This result, entirely expected, the states governments in India combined are spending highlights the fact that our main results can be on all levels of education every year. It is double what considered a lower bound of electricity subsidies they are all spending on health, and three times what in developing countries. they are spending on water supply. If power sector financial losses were reduced by only one-third, the Because of differences in methodologies, our savings in a single year would be sufficient to fill every results are not strictly comparable to earlier teacher vacancy in the country and provide every studies. Nevertheless, it seems likely that school with running water and toilet facilities. electricity subsidies are increasing in absolute Source: Lim, 2001. terms. Our estimate of some US$100 billion in subsidies in 1999 represents a very significant increase over the estimate of US$64 billion in 1991 reported in World Bank (1997a)8- Generating Public Sector Resources to Finance Sustainable Development 9 Box 2.3: Quasi-fiscal energy activities in the Former with electricity and other energy subsidies Soviet Union becoming an important liability for the stability of Despite a decade of transition, most countries of the the public budget and macroeconomic perfor- former Soviet Union (FSU) continue to maintain mance (see Box 2.3). largely unreformed energy sectors, characterized by Transportfuels mispricing, a high degree of government ownership, vertical integration, political interference, high levels Gasoline and diesel are not generally thought of payment arrears, cross-subsidization, large technical of as subsidized products, except in oil producing and other losses, and other operational problems. In countries. However, the subsidy bill in these many of these countries, energy companies continue to function as quasi-fiscal institutions and social safety countries can be substantial. In addition to selling nets, providing large implicit subsidies to households transport fuels at prices below production costs, and (state-owned) enterprises. In most countries of the there is the issue of externalities. Transport users former Soviet Union, quasi-fiscal activities in the generate local and global air pollution, congestion, energy sector have remained large and pervasive in and depreciation of transport infrastructure. Low recent years, often accounting for 5 percent or more of prices encourage greater consumption and larger GDP. attendant externalities. Energy sector quasi-fiscal activities create sizable For this report we have carried out two economic distortions and inefficiencies (such as estimation exercises, based on two different refer- overconsumption and underinvestment), they provide ence prices that can be used for calculating trans- opportunities for rent seeking and for nontransparent port fuel subsidies. The 'traditional' measure is and unjust (re-) distribution of income, while having adverse environmental effects. Energy sector quasi- the economic cost of producing and distributing fiscal activities disguise and obfuscate the true extent the fuels. Following Metschies (1999, 2001) we of government involvement in the economies of these use prices of gasoline and diesel of 23.8 and 21 countries, providing untargeted implicit subsidies to US cents per liter respectively for 1999 as crude consumers and enterprises. They can also pose a risk to global references. However, in the case of trans- macroeconomic stability and sustainable growth. Accu- port fuels, under the 'user pays principle' we can mulated arrears often tend to be cleared through credit devise a second reference price which includes a expansion, which fuels inflation; external imbalances mark-up to account for transport infrastructure are created by running up extemal payment arrears; depreciation costs. We call this the 'transport and fiscal policy is inhibited by the emergence of large sector' measure. The World Bank's Road Mainte- payment and tax arrears. nance Initiative suggests that developing countries Source Petri and others, 2002 should levy a fuel tax of about 10 US cents per liter to cover the cost of road maintenance.9 Note The size of electricity subsidies has that this tax would not cover environmental exter- traditionally tended to increase for two main nalities (air and noise pollution) and other exter- reasons. First, the gap between prices and nalities (congestion and accidents). We use this economic costs has tended to widen. While mark-up to generate a second reference price for production costs increase with inflation, regulated gasoline and diesel of 33.8 and 31 US cents per prices (tariffs) are usually fixed and infrequently liter respectively. When the price of the fuels is updated (for example, until a recent increase in below the reference price, the difference would tariffs, Uganda had not revised .their electricity represent a subsidy on the fuel. Neither of these tariffs for nearly a decade). Second, the quantity reference prices are fully accurate for each indi- of electricity consumed tends to increase both as vidual country, but they are useful to estimate fuel the economy grows and as consumers switch to pricing and taxation policies on a global scale. We the subsidized good (for instance, Yugoslavs find have used the transport fuel prices reported in it cheaper to heat their homes with electricity than Metschies (1999; 2001) to create average prices to invest in insulation,). India represents a case in for the year 199910 for a sample of 105 countries, point, with electricity subsidies doubling every representing nearly 100 percent of the consump- three years (see Box 2.2). For some individual tion of transport fuels in the developing world.i countries, the situation is getting out of control, 10 Generating Public Sector Resources to Finance Sustainable Development Table 2.4: Fuel subsidies in developing regions, 1999 Gasoline subsidies Diesel subsidies Total (billion USS) (billion USS) (percent of GDP) Transport Transport Transport Traditional sector Traditional sector Traditional sector measure measure measure measure measure measure Europe and Central Asia 0.3 1.6 0.5 3.5 0.09 0.6 Middle East andNorth Africa 3.1 6.2 8.2 14.8 1.90 3.5 Sub-Saharan Africa 0.2 0.5 0.2 0.6 0.23 0.36 Latin America and Caribbean 0.9 1.9 0.6 1.5 0.09 0.19 South Asia 0.0 0.0 0.0 3.1 0.00 0.54 East Asia and Pacific 0.7 3.0 3.0 7.0 0.20 0.54 Total 5.1 13.2 12.5 30.5 0.30 0.74 Notes: 'Traditional measure' refers to the size of the subsidy when the reference price is the economic cost of the fuel. 'Transport sector measure' refers to the size of the subsidy when the reference price takes into account the costs of maintaining the transport infrastructure. Source: Authors' estimates. Using these two reference prices as bench- Although these countries would not be subsi- marks, the fuel policies of developing countries dizing the fuel, they would be subsidizing road can be assigned to one of the following categories: transport. This group comprises around 25 * Subsidizing countries, where fuels are sold at countries. prices below the world market reference price. * Countries pursuing a medium price policy, This group comprises around 15 countries. with average fuel prices roughly between those * Countries pursuing a low price policy, with an of the United States and those of the European average tax of less than 10 US cents per litre. Union. Figure 2.2: Average fuel prices in developing countries by group, 1999 (US cents) 70 60 - a-Gasoline * Diesel. . 5 00 - - - - -------------- ------ -------------------------------- -- -- -- -- -- -- - - ---- -- i- - ---- ------- ---- --- l------------- 50 40 30 .. ..~~~~~~~~~~~~~~~~~~~~0 C a Lj CL ~ ~ ~ ~ E 0 .2' E a) L ) .r o~ X 8 ^ Q X .8, E8 - ' 8 e e X e m W a 0 < m < z x z E ED 3 cfl'J .O t CO z ls I 0_ CU 0 L .2c Wa) ;~ Note: Un-weighted averages. Source: Authors' estimates based on data in Metschies, 1999 and 2001 Generating Public Sector Resources to Finance Sustainable Development 11 In the sample analyzed, no developing country Box 2.4: Differential pricing of fuels pursues a high price policy such as that prevalent in the European Union. 13 Gasoline is generally taxed much more than diesel. This policy has its origin in diesel being an 'indu- Our results, shown in Table 2.4, indicate that strial' fuel (intermediate good), while gasoline is subsidies to transport fuels in developing coun- generally consumed by house-holds (final good). tries are substantial, amounting to 0.3 to 0.7 per- According to tax theory, only final goods should be cent of the developing world's GDP, depending taxed, if a number of conditions are met, including on the benchmark used. Implicit subsidies to absence of enviromnental externalities. The finan- .o.ne benchar used. US5ilici subsidies t cial advantage of diesel improves even further when gasolin amount to aboutUS$5billion,- its greater fuel effi-ciency is considered. This is un- dies to diesel to about US$13 billion, for a total of fortunate from an environmental point of view, as about US$18 billion. When transport infrastruc- conventional diesel is a more polluting fuel. Yet ture costs are taken into account, the amounts large differences in prices via taxation encourage increase to US$13 and US$31 billion respectively, fuel switching to diesel. for a total of about US$48 billion. This last figure Equivalent price of diesel after adjusting for must be taken with care. While road transport in the developing world represents 98 percent of the erg content consumption of gasoline, it only represents 80 (usc/I) percent of the consumption of diesel,'4 the rest 100 being consumed in the industrial and agricultural i sectors (IEA, 2001a). 80 - / There are significant regional differences in the prevalence of subsidies to transport fuels. The 60 - - - - Middle East and North Africa region predictably stands out (see Table 2.4). This region is respon- sible for 60 percent of transport fuel subsidies, 40 ; 1 .* representing 2 percent of regional GDP, or over , / j*? 3.5 percent of regional GDP when road transport 20 ----------- infrastructure costs are factored in. A further point to notice is that diesel is 0 priced consistently below gasoline (see Figure 2.2 0 20 40 60 80 100 and Box 2.4), despite being more environmentally Pnce of gasoline (usc/Il) harmful. We expand on this issue below. Source: Authors' estimates based on data in Energy subsidies and the environment Metschies, 1999 and 2001. In addition to the burden they place on public and (3) effects on global climate change. The finances, existing energy subsidies and pricing largest share of environmental damage is associ- policies often result in significant adverse envi- ated with the impacts of pollution on human ronmental effects by encouraging increased en- health. Every year an estimated 0.5-1.0 million ergy consumption, which results in high levels of people die prematurely from respiratory and other air pollution. Fuel prices affect the levels of emis- illnesses associated with urban air pollution sions by influencing aggregated or fuel-specific throughout the developing world, and millions demand. more suffer from these disorders (Kojima and Lovei, 2001). Globally, urban (outdoor) air pollu- Urban air pollution resulting from fossil fuel use causes substantial economic costs. The main tion is highest in China, India, and a number of damages from fuel combustion are: (I) adverse cities in Asia and Latin America. health effects of exposure to ambient air pollution A recent World Bank study (Lvovsky and in urban areas (for example, increased respiratory others, 2000) finds that marginal damage costs illness and premature deaths); (2) local nonhealth from fossil fuels use are comparable to their mar- effects (reduction in visibility; increased soiling); ket prices and, for some fuel uses, may exceed 12 Generating Public Sector Resources to Finance Sustainable Development them. In a sample of six cities (Bangkok, Krakow, 2001). According to the World Bank study, emis- Manila, Mumbai, Santiago, and Shanghai), mar- sions from transport fuels contribute 45 percent of ginal damages range from 60 percent of market the local damage of fuels in the sampled cities (or price for unleaded gasoline and 50 percent for fuel US$1.6 billion per year), and 12 percent of the oil to more than 200 percent for automotive die- global damage (or US$0.1 billion) (Lvovsky and sel. The social costs of all environmental impacts others, 2000). The environmental costs imposed totaled US$4.2 billion per year (or US$84 per by transport fuels represent 3 percent of GDP in urban citizen), of which health impacts amounted urban areas, with local costs representing 2.7 per- to 68 percent, climate change impacts to 21 per- cent of GDP. For the limited sample analyzed, the cent (using a shadow price of US$20 per ton of local damage of transport fuels adds to over carbon emissions), and local nonhealth effects US$1.5 billion. Diesel is responsible for over 70 contributed 11 percent. According to these fig- percent of this amount. ures, the environmental cost of fossil fuels repre- In contrast, the environmental costs of power sent over 7.5 percent of the sampled cities' GDP, generation represent only 7 percent of the envi- with local costs representing 6 percent of the cit- ronmental cost of fossil fuels, and are largely in ies' GDP. the form of global costs (that is, climate change). Air pollution from urban transport is of par- Although power generation represents 29 percent ticular concem. Vehicle emissions, occurring as of global damage of fossil fuels in large urban they do near ground level and in densely popu- cities in developing countries, it only represents 1 lated areas, cause much greater human exposure percent of local environmental costs (Lvovsky and to harmful pollutants in the immediate locality others, 2000). Table 2.5 reports the environmental than do sources that emit at higher elevations and costs of fossil fuels for six cities, by sector. farther away from population centers. Further- more, vehicle exhaust particles, especially those iney s diesel exhaust, are among the most damaging to Energy subsidies are often motivated as public health: they are small, and there is growing helping the poor. Access to modem energy is a evidence that some are carcinogens. Because vital dimension of poverty alleviation and devel- motorization generally rises with increasing in- opment. The poor use proportionately more wood, come, pollution abatement in the transport sector dung, and other biomass fuels in traditional ways, is likely to become an increasingly important part and less electricity and LPG, which tend to be of urban air quality management in the coming more expensive and cause adverse problems such years in developing countries (Kojima and Lovei, as indoor air pollution. TabRe 2.5: Environmental costs of fossil fuels in six developing country cities Total damage Local damage Percent of Percent of (million (million total local US$) (% GDP) US$) (% GDP) damage damage Power plants 296 0.5 29 0.05 7 1 Industrial and commercial boilers Large 713 1.3 329 0.6 17 10 Small 1,158 2.1 1,068 1.9 27 31 Households 528 1.0 460 0.8 12 14 Vehicles 1,625 2.9 1,513 2.7 38 45 All sectors 4,317 7.8 3,397 6.1 100 100 Notes. Sample of six cities (Mumbai, Shanghai, Manila, Bangkok, Krakow, and Santiago) with a combined population of 46 million people and a combined GDP of US$55 billion. Original calculations refer to 1993, but have been converted to 1999 US dollars Local damage includes health and non-health costs. Total damage includes, in addition, climate change costs. Health costs were valued using the Value of Statistical Life (VOSL) approach. Climate change costs were valued using a shadow price of US$20 per ton of carbon emissions. Source Lvovsky and others, 2000, and authors' calculations based on their data. Generating Public Sector Resources to Finance Sustainable Development 13 Box 2.5: Income and health effects of access to modern energy For cooking, the urban poor often pay more for wood solid fuels, particularly biomass (wood, dung, and crop or charcoal than they would for LPG, after adjusting residues), typically in open fires or simple stoves, for the end use efficiencies of the fuels. Kerosene mostly indoors, and rarely with adequate ventilation or lamps produce about 200 times less light than an chimneys (Lvovsky, 2002). In India, where 75 percent electric light bulb, create unhealthy particulate and of households use biomass as primary fuel, and the volatile hydrocarbon emissions, are cumbersome to exposure to indoor air'pollution causes an estimated transport, and are a serious fire hazard (ESMAP, half a million premature deaths every year, the 2001a). Families using electricity for lighting'experi- transition to cleaner fuels would reduce between a ence the ability to extend the day; children can study third and a half the mortality rate for children under longer hours, which raises educational levels; and the the age of five in rural areas (ESMAP, 2000). household has lower lighting expenditures than those Although some technical interventions to reduce using kerosene. In rural Nepal - where lighting is gen- exposure to indoor air pollution exist, the only feasible erally achieved by the use of kerosene lamps and long-term remedy is to improve access to cleaner amounts to around I percent of rural energy consump- modem energy (Smith, 1999). tion - the kerosene bill accounts for 10 percent to 20 Recent microeconometric work by the World Bank's percent of a typical family's earnings (HLF, 2001). Environment Department (Wang, 2002) suggests that Fuels such as fuelwood, charcoal, and dung are major household access to electricity has a large and source of indoor air pollution. Exposure to indoor air significant effect on reducing child mortality. Al- pollution, in turn, is a major factor contributing to the though the mechanism is not clearly understood, the global burden of disease. Epidemiological studies in effect is independent of income. According to this developing countries have linked exposure to indoor study, it would be possible to avert I under-five child air pollution from traditional fuels with at least four death-per 1,000 births by expanding household access major categories of illness: acute respiratory infections to electricity by 0.6 percent. For economic growth in children; chronic obstructive lung diseases such as alone to have the same effect, a growth rate of 6 per- asthma and chronic bronchitis; lung cancer; and still- cent would be required. births and other problems at birth. There is also evi- The detrimental health effects of traditional biomass dence of links to blindness, tuberculosis, and cardio- energy health exposure trarmful sm ass vasula dieaes Smih,199). stiats idicteenergy go beyond exposure to harmfuil smoke. In vascular diseases (Smith, 1999). Estimates indicate Nepal, rural women are so busy with their daily chores that exposure to indoor air pollution in developing that they are forced to give'local beer to children to countries causes over 2 million deaths each year keep them quiet as they gather fuel and feed the stoves. (Smith, 1996; Lvovsky, 2002), with 75 percent of Nepalese women suffer a high-incidence of uterine these deaths occurring in rural areas, and 25 percent in prolapse that is likely the result of carrying heavy urbanareas (Smith, 1996). - loads' of wood soon after delivery. - In rural India, The use of cleaner fuels such as kerosene or liquefied women spend six hours a day collecting fuelwood and petroleum gas (LPG) would have a major impact in fodder and cooking. In some areas this extreme reducing exposure to indoor air pollution and -its physical - drudgery -causes serious reproductive impact in child mortality. More than half of the problems and mental disorders in women (ESMAP, world's households cook and heat using unprocessed 2001c). The case for subsidizing energy access for in particular would be unable to pay the upfront the poor rests on two arguments, The first is that it costs involved (which include both connection would bring substantial welfare benefits, by both charges and the acquisition of suitable equipment reducing the energy costs faced by the poor and such as gas-burning stoves or electric appliances). giving them access to cleaner, less polluting fuels . The evidence indicates, however, that the (see Box 2.5). There would thus be both income bulk of energy subsidies is received by the better- and health effects. The second is that the up-front Off Most obviously, subsidies are often provided investments required to reach new customers and for energy, sources not used by the poor. This is the small ensuing revenue flows result in service most clearly the case with electricity. As shown in providers-whether public or private-having Table 2.6, the poor tend to have much lower little incentive to market energy services to the access to electricity. Many energy subsidies are poorer segments of the population. Moreover, the thus mis-targeted if their objective is to assist the poor often lack the means to pay for energy, and poor. 14 Generating Public Sector Resources to Finance Sustainable Development Table 2.6: Access to electricity in selected countries, 1997-1998 (% of population) Wealth Quintiles Country Poorest Second Middle Fourth Richest Total Kenya 0.0 0.2 0.1 1.8 56.7 11.7 Niger 0.0 0.0 0.0 0.5 38.9 7.9 Togo 0.0 0.2 0.4 6.2 67.7 14.9 Indonesia 35.0 7613 91.9 98.9 99.9 80.4 Kyrgyz Republic 98.7 100.0 100.0 100.0 100.0 99.7 Philippines 6.9 58.4 94.1 99.8 100.0 71.8 Vietnam 20.5 78.9 87.9 96.6 99.9 76.7 Bolivia 3.6 59.5 96.2 99.8 100.0 71.8 Nicaragua 2.4 48.7 92.4 98.5 98.6 68.1 Source: World Bank, 2000a. Even when the poor do use the subsidized ter-off households. The poorest 10 percent of the energy source, much of the benefit often flows to population consume slightly over one liter of the better-off. The case of Indonesia illustrates kerosene per person per month, while the richer this leakage problem. Indonesia provides a 55 per- households consume up to four or five times that cent subsidy on kerosene sales, for a total of amount (SMERU, 2001). Similarly, in Ecuador US$0.44 billion in 1997 (see Table 2.2). Most of cheap kerosene was diverted to the transport sec- the benefit of this subsidy,' however, goes to bet- tor, and much of it never reached the poor, espe- Box 2.6: Leakage in electricity subsidies in cially in rural areas (ESMAP, 1994). Thus, even Guatemala though the poor receive some benefits from kero- Following privatization of energy distribution sene subsidies, the bulk of the benefits are cap- companies, Guatemala introduced a 'social tariff' tured by better-offhouseholds. The same problem (tarifa social) which caps domestic tariffs for of leakage can be seen in many electricity subsi- households consuming up to 300 kWh per month at dies (see Box 2.6). 8 c/kwh The tarifa social costs US$50 million The poor may not only fail to benefit from annually. Although aimed at protecting the poor, energy subsidies-they may often actually be the tarifa social largely fails to reach poor house- harmed. While better-off households receive sub- holds. Low connection rates among poor house- .. .. holds combined with a high consumption threshold sidized electricity, the poor are often forced to result in about 65 percent of beneficiaries being continue using much more expensive and often non-poor households. Non-poor households cap- much dirtier fuels. In India, for example, pricing ture about 90 percent of the total value of the subsi- distortions and mounting subsidies for kerosene dy. 60 percent of poor households receive no bene- and LPG create perverse incentives for illegal fits at all from the scheme, as they do not have an diversions to the commercial and transport sec- electricity connection. A much more pro-poor po- tors, thereby limiting the availability of modem licy would be to use the resources allocated to the fuels for cooking, in a country where 75 percent tarifa social to expand electricity coverage to un- of households bum wood, dung, and crop residues served households. Households without electricity ( . . (70 percent of whom belong to the poorest segment (seeB 2. thicase of tricys eve more of the population) pay implicit prices of more than prominent. Artificially low tariffs lie at the root of 1,100 c/kwh (more than 100 times the price of the poor financial performance of many state- electricity) to illuminate with candles and wick owned energy companies in the developing coun- lamps and to power appliances with dry cell batte- tries. In turn, this poor performance seriously re- ries, compared with full cost electricity tariffs of 11 duces the ability of those companies to invest to to 15 c/kwh. An estimated 50,000 additional new meet the increasing demand, especially to those connections each year could be financed with the consumers who do not yet have access to com- current expenditure of the tarifa social mercial energy. Low tariffs also reduce the incen- Source: Foster and Araujo, 2001. tives for utilities to expand coverage, as each Generating Public Sector Resources to Finance Sustainable Development 15 there are few data on the distributional impacts of Box 2.7: Petroleum product subsidies and poverty these environmental costs, there are reasons to be- in India has a two-tier pricingsystemunderwhich lieve that the urban poor bear the brunt of air India has a two-tier pricing system under which pollution. The poor often live in densely popula- state-owned suppliers sell kerosene and LPG at sub- tiorhoor often lve in densely porri- sidized prices to domestic users, and at market prices ted neighborhoods located close to traffic corni- to private sector distributors. This structure leads to dors and industries, and they travel in open vehi- substantial leakage, as subsidized fuels are diverted cles or walk and spend a great deal of hme out- from domestic to commercial users. The poor, for doors (Lvovsky, 2002). The poor are also expect- whom these subsidies were designed, do not usually ed to suffer disproportionately from the impacts of use kerosene or LPG for cooking (although they use climate change (IPCC, 2001). To the extent that kerosene for lighting). A 1994 survey of households energy subsidies contribute to preventing the in Hyderabad 1994 showed that 63 percent of the expansion of coverage, they also contribute to the subsidy went to the richest 40 percent of the popula- toll imposed on the health of the poor by conti- tion, while only 17 percent of it went to the poorest nued use of low-quality fuels (see Box 2.5 above). 40 percent. Although these across-the-board subsi- dies are an enormous drain on the government bud- Movement up the energy ladder'5 to cleaner get-US$1.1 billion for LPG and US$1.9 billion for and more efficient fuels is historically a natural kerosene in 1999-2000-their overall effect may be trend, and policies to encourage and accelerate it to reduce the accessibility of these fuels to low are amongst the most promising ways of reducing income households. Mounting subsidies do not give exposures and health effects. Although there are incentives to the govermment to expand the distri- successes, finding ways to efficiently encourage bution of kerosene. The response of the Government movement up the energy ladder remains a chal- of India to mounting losses has been to freeze the . . . total amount of kerosene distributed, and decrease . the allocation where possible. More effective policy costly for governments, and they often backfire by interventions to support the use of cleaner fuels distorting markets while providing little actual would be to promote local microcredit facilities to benefits to households (Smith, 1999). finance the up-front costs of fuel switching, and to Reforming energy subsidies target subsidies to enable switching. Sources: ESMAP. 2000. 2001b Energy subsidies have a well-earned reputa- tion of being perverse, because they not only additional connection would automatically mean increased losses. ~~~encourage wasteful consumption of a natural increased losses. resource but also cause important negative exter- The poor are also harmed disproportionately nalities-both local (air pollution, congestion) and by the adverse environmental consequences of global (climate change through carbon emissions). high energy use spurred by subsidies. Although Removing these subsidies, therefore, would not Table 2.7: Effects of phasing out energy subsidies in eight developing countries Energy savings Economic efficiency gains Emissions reductions (%O) (Yo of GDP) (NO) (million tonnes) China 9.4 0.37 13.44 421 Russia 18.0 1.54 17.10 249 India 7.2 0.34 14.50 128 Indonesia 7.1 0.24 10.97 28 Iran 47.5 2.22 49.45 141 South Africa 6.4 0.10 8.11 28 Venezuela 24.9 1.17 26.07 38 Kazakhstan 19.2 0.98 22.76 29 Total sample 12.8 0.73 15.96. 1,058 World 3.5 - 4.59 Notes: Data for 1997 (Russia, Indonesia, Iran, South Africa) and 1998 (China, India, Venezuela, Kazakhstan). Sources: From data in IEA, 1999. 16 Generating Public Sector Resources to Finance Sustainable Development only free substantial budgetary resources for other African countries, the challenge is to structure uses, but also increase efficiency in energy use subsidies to alleviate poverty and raise living and reduce environmental damage. Table 2.7 standards by improving access to modem forms of gives a sense of the potential gains that could be energy in an environmentally acceptable and obtained by phasing out energy price subsidies in financially viable way. In Asia, the recent finan- the eight countries analyzed by the IEA (1999). cial crisis and surge in international prices, which Primary energy consumption would fall by 13 have led to higher end-user prices and put enor- percent; GDP would increase by almost I percent mous strain on government finances, have height- through higher consumption; and carbon dioxide ened the necessity for a move to more market- (CO2) emissions would fall by 16 percent. based energy pricing while highlighting the need Despite its potential benefits, removing for continued support to poor people. And most energy subsidies has proven a difficult challenge. Latin American countries recognize the necessity Although it would bring high overall benefits for to reform energy subsidies and have already enga- society, some groups would lose. Nigeria and ged in reform processes where access to electri- Indonesia have repeatedly experienced the roll- city seems to be the overarching theme. back of announced reform, following the mobili- Effects of electricity subsidy removal zation of the expected losers. Box 2.8 describes the possible approaches to energy subsidy reform. The effects on consumption of electricity subsidy removal would likely be very important. Box 2.8: Approaches to energy subsidy reform Our calculations indicate that increasing tariffs to Two distinct approaches are possible to bring cover long run marginal cost would generate a energy prices in line with economic costs. In a 'big cash flow of some US$35 billion, and will save bang' approach the intention is to close the gap quickly US$67 billion in production costs. A most rele- by means of large increases in prices, while in a vant concept from an economic perspective, 'gradualist' approach the gap is closed gradually, in a however, is efficiency gains (see Appendix B). series of steps involving a longer transition period. The Our calculations indicate that these would amount big-bang approach capitalizes on political will to Ourcalcullion. reform, while the gradualist approach risks reforms to some $20 billion. being reversed as political will erodes with time and Assuming a -0.5 elasticity of demand, a affected constituencies get organized to oppose the constant elasticity inverse demand function, and a reform. However, the big-bang approach can have maximum reduction of consumption of 50 per- greater social costs (as consumers do not have time to cent, removing electricity subsidies entirely would adapt to the new situation), and risks provoking social result in an average decrease in electricity con- unrest that may be able to stop the reform (as hap- pened, for example, in Indonesia and Nigeria). Experts sumption of 22 percent (see Table 2.8). This remain divided on which approach is preferable. For figure iS driven by the situation in the Former the gradualist removal of subsidies, it has been sug- Soviet Union, where very low tariff levels and gested that price increases, however frequent, should widespread non-payment have resulted in impor- be kept below 10 percent each time, even if this repre- tant inefficiencies. sents a small increase in absolute terms. Other imple- The environmental effects are more difficult mentation measures could include informing consum- to ascertain. The impact of reduced consumption ers that they are being currently subsidized (for on carbon emissions varies from country to coun- instance, by explicitly itemizing subsidies on electricity onycarbonremissonstvaries,fromcountryutoncoun bills), to make changes in prices automatic (by some try and from year to year, depending upon the kind of indexing), and to link price increases to service generation mix. For instance, in countries that rely improvements. primarily on hydroelectricity, the reduction in consumption would not have significant effects on The issues involved in reformmIg energy sub- greenhouse gas emissions. We generated country- sidies differ across regions (IEA/UNEP, 2001). based carbon emission coefficients to calculate the Efforts in transition economies need to focus on potential reduction in emissions. Our results indi- raising energy prices to market levels and impro- cate that phasing-out subsidies to electricity would ving collection rates: non-payment of energy bills reduce carbon dioxide emissions by about 195 constitutes a major source of implicit subsidy. For million tons. Generating Public Sector Resources to Finance Sustainable Development 17 Table 2.8: Estimated effects of removing electricity subsidies in developing countries Decrease in consumption Reduction in CO2 Efficiency gains emissions (TWh) (O) (% of GDP) (million tonnes) Sub-Saharan Africa 56 25 0.22 17 South and East Asia 235 13 0.08 59 Central and Eastern Europe 56 13 0.21 10 Former Soviet Union 426 49 3.41 65 Latin America and Caribbean 52 7 0.04 5 Middle East and North Africa 193 41 0.86 37 Total 1,017 22 0.33 193 Source: Authors' estimates. The difficulties in reforming electricity sub- can be generated; without improvement in service sidies should not be underestimated. In the case of increase in revenues may be politically unfeasible. electricity, at least four elements need to be com- bined in a successful reform: increase of tariffs, Effects of fnel subsidy removal and fuel improvement of collection payments, improve- ment of service, and strengthening disconnection Fuel subsidy removal and fuel taxation can capacity. The optimal combination of these ele- have remarkable effects on both public finance ments is not clear. Cash-strapped utilities struggle and the environment. The public finance aspect is to improve service, consumers are unwilling to especially important in developing countries, pay for unreliable services, and (given the social where weak institutional capacity makes many costs of disconnecting hospitals, schools, and even taxes difficult to collect (Bacon, 2001). The char- households in cold climates) utilities frequently acteristics of fuel taxes-straightforward to admi- lack the capacity to enforce payments. Utilities nister, relatively price inelastic, and income elastic need cash to improve both service and collection; -make them specially attractive to these coun- without increase in tariffs and collection no cash tries. However, designing fuel taxes when there Box 2.9: Optimal taxation of fuels Tax theory suggests that where the government's more harmful than those from gasoline, but encourag- sole aim is to raise revenue for public expenditure, ing the use of diesel-powered mass transit may be goods for which demand is least sensitive to price desirable as a way of relieving congestion. Kerosene is increases should tend to bear the highest tax rates. particularly problematic in developing countries, since Goods that are close substitutes should be taxed at it can be used to adulterate both gasoline and diesel. similar rates to prevent demand from switching from As a result, setting lower taxes on kerosene (to reduce the higher- to lower-taxed good, reducing government the cost of lighting and cooking fuels for the poor) can revenue.. In addition, where equity is an important erode the total tax collected (necessitating an increase consideration, goods accounting for a larger share of in the general tax level to produce a given revenue). budgets for the rich than for the poor should be taxed And where kerosene replaces gasoline, lower taxes more heavily. Goods that produce large negative exter- lead to higher emissions and worse vehicle perform- nalities (such as emissions from automotive fuel use or ance. Higher taxes on kerosene can hurt poor house- congestion from excessive road use) should also be holds, however, as they tend to spend a larger share of taxed at high rates, to discourage their consumption their budgets on this fuel than better-off households. and reduce social harm. For fuels, these principles do But this also means that if governments wish to offset not always work in the same direction, so the relative the effect of higher kerosene taxes on poor households, importance of each principle needs to be evaluated in they can do so through targeted assistance rather than each case. For example, emissions from diesel fuel are across-the-board kerosene subsidies. Source: Bacon, 2001. 18 Generating Public Sector Resources to Finance Sustainable Development Table 2.9: Impacts of fuel subsidy removal and fuel taxation Reduction in Reduction in Net effect on Reduction in C02 consumption consumption public budget emissions (Mt) (percent) (billion US$) (million tonnes) Impacts of eliminating subsidies to gasoline Eastern and Central Europe 0.8 1.5 0.3 1.9 Middle East andNorthAfrica 13.6 33.8 3.1 31.7 Sub-Saharan Africa 0.5 3.4 0.2 1.1 Latin America and Caribbean 2.2 3.1 0.9 5.2 South Asia 0.0 0.0 0.0 0.0 East Asia and Pacific 1.5 2.0 0.7 3.5 Total 18.6 7.0 5.2 43.3 Impacts of bringing gasoline prices to regional average Eastern and Central Europe 9.8 17.5 11.0 22.8 Middle East and North Africa 20.8 51.8 8.7 48.5 Sub-Saharan Africa 2.3 16.1 2.2 5.3 Latin America and Caribbean 9.7 13.8 9.8 22.7 South Asia 0.2 2.3 0.3 0.4 East Asia and Pacific 16.2 21.3 17.4 37.8 Total 58.9 22.3 49.4 137.4 Impacts of eliminating subsidies to diesel Eastern and Central Europe 1.6 2.4 0.5 3.8 Middle East and North Africa 59.8 82.0 8.3 139.7 Sub-Saharan Africa 0.4 3.1 0.2 1.0 Latin America and Caribbean 2.2 2.8 0.6 5.1 South Asia 0.0 0.0 0.0 0.0 East Asia and Pacific 11.3 8.4 3.0 26.4 Total 75.4 18.1 12.6 176.1 Impacts of bringing diesel prices to regional average Eastern and Central Europe 11.1 16.5 4.3 25.8 Middle East and North Africa 53.7 73.6 8.0 125.5 SubSaharan Africa 1.8 12.5 0.7 4.2 Latin America and Caribbean 3.0 3.8 0.8 6.9 SouthAsia 5.1 10.2 2.9 11.9 EastAsia and Pacific 21.7 16.1 5.9 50.6 Total 96.3 23.1 22.6 225.0 Impacts of bringing gasoline prices to global average Total 18.5 7.0 37.8 43.2 Impacts of bringing diesel prices to global average Total 148.4 30.3 41.0 346.5 Source: Authors' estimates. are multiple policy objectives to be attained is not generated when poverty reduction objectives and simple (see discussion in Gwilliam and others, the possibility of interfuel substitution are added. 2001). Box 2.9 illustrates some of the insights of We have estimated the effects of removing tax theory for fuel taxation, and the complications fuel subsidy and increasing fuel taxation. Table Generating Public Sector Resources to Finance Sustainable Development 19 2.9 presents the results of a simulation that assu- A recent review concludes that the impact of mes a -0.4 price elasticity of demand and a con- energy sector reform on the poor has been positive stant elasticity inverse demand function. The in countries where the reform is most complete effects of eliminating gasoline and diesel subsi- and mature (Albouy and Nadifi, 2000). While the dies (as conventionally understood) would lead to direct impact is often a tariff increase for many a reduction in consumption of 7 percent and 23 users, in these cases the poor were protected percent respectively; a positive effect on the pub- against increases or benefited as group from the lic budget of US$18 billion; and a reduction in improvements that reform brought about in the CO2 emissions of nearly 220 million tons. For sector-better access and service-in government countries with fuel prices below regional aver- finances and in the economy at large. Reforms in- ages,'6 raising prices to match those averages directly benefit the poor in three ways: (1) would generate US$71 billion. In this case, con- enabling energy delivery mechanisms (whether sumption for both fuels would fall by 23 percent, off-grid electricity, LPG bottles, or other) that and CO2 emissions would fall by nearly 360 expands energy access, voice, and choice for con- million tons. When using the global average sumers; (2) freeing fiscal resources for high prior- prices'7 as benchmarks, the net effect on the ity fiscal spending; and (3) reducing the health public budget reaches some US$79 billion. impacts of energy supply to which the poor are The high environmental costs of fossil fuels most exposed. highlight the substantial social gains, mainly in Finally, the real impact of energy subsidy public health, that could be achieved by eliminat- removal on the poor should come from the use of ing price distortions caused by subsidies, protec- the funds liberated. In countries where lack of tion of domestic oil or coal monopolies, and un- access is an important issue, some form of assis- balanced taxation of similar products with little tance may be required to help poor households regard for their true social costs (Lvovsky and obtain higher quality energy services. Such assis- others, 2000). Eliminating fuel subsidies and tance should be directed at encouraging access to imposing higher taxes on dirtier fuels would services rather than subsidizing the operating reduce their consumption and the associated envi- costs of providing the services. In countries where ronmental costs. Removing these subsidies would affordability is the main issue, as in Eastern produce domestic environmental benefits, Europe and Central Asia, safety nets (such as including reduced local air pollution, as well as raising pensions or providing targeted assistance) positive effects on a global scale. Energy must be put in place in parallel to the tariff consumption would be cut by 3.5 percent, thus increases. When the temperature is -40°C, energy improving world energy intensity significantly. prices obviously cannot be raised without concern World CO2 emissions would fall by 4.6 percent. for affordability. This highlights the fact that sub- Impact on the poor sidy reform must be contemplated in the context of broader policy reform, not merely as a sector A priori, it might seem evident that removing intervention, and must be tailored to the specific energy subsidies will affect the poor adversely. circumstances of each country. However, it is important to distinguish cases Conclusions where the poor do not have access to the energy carrier from those where they do. In the first case, Removing energy subsidies would support which is typical of electricity, the poor simply will the three principal aims of sustainable develop- not be directly affected by subsidy removals. In ment: social welfare, environmental protection, the second case, when the population has grown and economic growth. Funds supporting subsidies used to extremely low prices-such as Indonesia could be directed to social benefits and income and Nigeria, for fuels, and the Former Soviet redistribution. Environmental benefits accrue Union, for electricity-even the poor are energy- from proper pricing, which could reduce both hungry, and as a consequence the sudden removal local and global pollution (including CO2 emis- of subsidies could severely hurt them, especially sions). Economic growth would be boosted in the absence of safety nets. through improved efficiency and reduced budget 20 Generating Public Sector Resources to Finance Sustainable Development costs. Subsidy removal can also have a longer- leakage and mistargeting would need to be term impact. It contributes to per capita welfare addressed. Still, phasing out subsidies would over time by eliminating one stimulus for over- require the establishment of safety nets, such as consumption (which leads to the rapid depletion increases in pensions. This would make assisting of the stock of natural capital) and stimulates the poor much cheaper, but the technical and technologies capable of enhancing sustainable institutional demands should not be understated. development (IEA, 2001b). Another major issue is that of access to modem Revenue impacts. Given the current size of energy. With the poor currently paying higher energy subsidies, it is not surprising that the reve- prices for lower quality energy, at least reforming nue potential from phasing them out should be electricity subsidies to address the capital costs of substantial. For the developing world, revenue expanding service coverage and taking them away from phasing out subsidies to gasoline and diesel from subsidizing consumption should go a long could be as much as US$18 billion. Moreover, if way in eliminating incentives to over consumption countries with low taxes on those fuels where to and better targeting the poor. increase them to regional levels, the positive effect in the public budget would jump to over 2.2 Reducing Water Subsidies US$71 billion. Electricity subsidies alone account for some US$102 billion. This figure encom- Water is indispensable, both as an input into passes the opportunity cost of fuels, and the non- agricultural production and for consumption. In incurred costs of keeping up infrastructure, adding both the irrigation and the domestic water supply to some US$67 billion, and the additional cash sectors, however, similar patterns emerge: only a flow that will be generated, estimated to be very small fraction of the costs of providing water US$35 billion. An important consideration is that are collected from users. Although based on well- of the financial sustainability of subsidies, espe- meaning policies, these patterns have typically cially electricity ones. These subsidies, under a had dismal results: water is used wastefully by combination of growth-driven increase in demand those who have access to it, even while many and fixed tariffs that are quickly outpaced by suffer from lack of access; governments bear sub- inflation, are growing to the point of threatening stantial financial burdens to sustain water delive- public finance stability. ries; nevertheless, water providers are chronically Environmental impacts. The removal of short of funds, preventing them from maintaining energy subsidies, by eliminating pricing distor- their infrastructure-which sometimes decays to tions, would encourage energy savings that would the point of collapse-and from expanding it to be translated in environmental improvements. In reach unserved users. Ironically, although low addition to local benefits regarding air pollution, water prices are often justified as protecting the there are important global benefits to be gained in poor, their impact tends to be regressive: available the field of climate change, as reduced consump- subsidies are captured mainly by the better off, tion would translate into reduced carbon dioxide while lack of funds prevents the extension of emissions. Phasing out subsidies on electricity, water deliveries to the poor. Reducing water sub- gasoline and diesel could reduce those emissions sidies could, therefore, both alleviate a very sub- by roughly 0.6 billion tons of CO2. This represents stantial financial burden on governments (freeing 4.6 percent of the developing world emissions, up resources for other uses, including better ways and 2.4 percent of global emissions. The potential to provide water to the poor) while increasing the trading of those emissions reductions can be efficiency of the sector and reducing its adverse valued at US$1.7 billion. environmental impact. Poverty impacts. Energy subsidies in the Irrigation developing world are often justified in terms of protection of the poor. However, there is ample Investments in irrigation infrastructure have evidence that they are regressive, with the non- played an important part in increasing food pro- poor capturing up to 90 percent of subsidies. For duction in developing countries. Continued expan- subsidies to benefit the poor, the problems of sion of irrigated agriculture is essential for future Generating Public Sector Resources to Finance Sustainable Development 21 food security. A 24 percent increase in net irriga- infrastructure is in poor conditions. ted area of the world by the year 2025 is necessary Throughout the developing world most irri- to keep supply and demand for food and cereals in gation, especially large-scale irrigation, is publicly equilibrium at prices that are affordable for the financed. With some exceptions, private irrigation poor (Gonzalez, 2000). Accordingly, irrigation in developing countries is on a very small scale will require US$175-250 billion of productivity and is a family affair for most farmers (Briscoe, enhancing investments between 1997-2020 to - . . . . . r meet future food demand-not taking into account 1999b). Supporting irrigation IS costly and requi- res substantial financial resources on the part of the costs of operating and maintaining existing the public sector. Every year, an estimated US$10 systems (Rosegrant and others, 2001). But this to 15 billion are spent for public investments in need for increased irrigation is running into nume- irrigation in developing countries. However, rous obstacles. under current pricing policies for irrigation water, * In recent years, irrigation expansion has slow- very few countries recover irrigation costs. ed dramatically due to sharply lower invest- . . ments (Gonzalez, 2000). Investments in new Charges for irrgation irrigation projects are expected to continue Countries have different experiences with their decline, based on capital commitments of charging users for irrigation water (Box 2.10). For developing country governments and intema- agriculture, all developing countries set water tional donors. charges on the basis of average rather than margi- * Increased demand for irrigation water faces the nal costs of supply. Authorities generally calculate reality that water scarcity is increasing year charges by dividing the average cost of service by after year. Expanding agricultural production area irrigated, often adjusting the results by sea- to meet the future food demands of increased son, type of crop, or type of technology used. populations will require nearly 20 percent more Charges are not generally adjusted by region with- water within the next 25 years. Yet water is in a country or for the technology used for water already scarce in many countries (Gleick, supply, even though both can result in different 1996; Seckler and others, 1998). costs of supplying water (Dinar and Subramanian, * Moreover, much of the existing irrigation 1997). Box 2.10: Charging for irrigation water Charges for irrigation water can be either volu- Under either approach, the water prices can be set metric or non-volumetric. either administratively or via market forces. In prac- * Volumetric charges are based on a measure of the tice, most countries set water prices administratively. volume of water consumed. They tend to encourage Although cost recovery is highly preferred it is efficient water use, but their implementation costs worth noting that establishing the right fee to cover are relatively high. They require the water authority irrigation costs is not an easy task, especially for irri- to establish consumption measurement equipment, gation systems that provide different services to multi- and then set the price, monitor use, and collect fees. ple water users. The assessed total cost of water supply * Non-Volumetric methods are used when informa- (and drainage services) must be calculated and then tion on water consumption is inadequate. Possible distributed among different water users such as far- approaches include output pricing (charges a fee mers, villages that receive water supply from irrigation for each unit of output produced by water users), canals, and those benefiting from flood controls down- input pricing (charges users for water consumption stream or from hydropower. With such complexity, a through a tax on inputs), and area pricing (charges specific target revenue, and hence a target price range on a per unit area basis, often with different rates for irrigation water, can be identified with difficulty. depending on crop choice, area irrigated, method, Moreover, O&M costs often vary over time or by and season). These methods tend to be easier to season or in the face of major events such as floods implement and administer, as they are usually and other problems. Changing the pricing schemes in based on already-monitored indicators. However, the face of these changes might not be easy to do. they provide little incentive for efficient water use. 22 Generating Public Sector Resources to Finance Sustainable Development Table 2.10: Irrigation water charges in selected countries Non-volumetric charges Volumetric charges (US$/ha or USS/season) (US$/m3) Algeria 3.79 - 7.59 0.019 - 0.22 Brazil 3.5 0.0042 - 0.032 India 0.164 - 27.47 Israel 0.16 - 0.26 Madagascar 6.25 - 11.25 Namibia 53.14 0.0038-0.028 Pakistan 1.49 - 5.8 Portugal 0.0095 - 0.0193 Sudan 4.72 - 11.22 Taiwan 23.3 - 213.64 Tanzania 0.26 - 0.398 Tunisia 0.02 - 0.078 Source Dinar, 2000. Table 2.11: Irrigation water charges and cost recovery Basisfor water charges Cost recovery Country Volume Area O&M Capital Comments China Yes Partial No Labor contribution by users Egypt No No No charges, but users are responsible for O&M India Yes Yes Partial Partial Very different methods for charging across the country Indonesia No No No charges, but users are responsible for O&M Iraq Yes Government owned Jordan Yes Partial No Water supplied on demand Mexico Yes Yes Partial No Sophisticated pricing methods applied Morocco Yes Mostly Partial Nigeria Yes Partial No Pakistan Yes Partial Peru Yes Partial Partial Charges only in scarcity regions Philippines Yes Partial Additional water rights fee charged Zimbabwe Yes Partial No New uniform charges based on crop profitability Source: Tsur and Dinar, 1995. Table 2.10 shows water charges in a sample however, water charges are extremely low. Table of developing countries. Both volumetric and non- 2.11 illustrates some of the main issues in charg- volumetric charges are used in developing coun- ing users for irrigation water in a sample of devel- tries. In some countries, such as Algeria and oping countries. Use of volumetric charges is Brazil, these two types of charges are practiced in limited in practice. In most cases, area charges are combination, forming a two-part tariff. As can be used as the basis for setting water charges. seen, there is considerable variation in water Table 2.11 also highlights the most important charges, even within a country. In many cases, problem of charging users for irrigation water: Generating Public Sector Resources to Finance Sustainable Development 23 cost recovery being mostly partial or non-existent. low levels. Moreover, they have tended to keep Few countries even attempt to recover capital them unchanged over many years, resulting in a costs. But even the limited objective of recovering cumulative decline in real terms. In 1997, for annual O&M costs is seldom achieved. A review example, Punjab went so far as to withdraw all of World Bank projects found that only 30 percent charges for water and power for irrigation. of the projects recovered O&M costs (World Bank, 1994). For example, receipts covered only Impacts on expenditures 13 percent of costs in Pakistan, 25 percent in As a result of low water charges, the public China, and 10 percent in the Philippines (Briscoe, sector incurs substantial financial burden when 1999b). In India, revenues from water charges supporting irrigation systems throughout the accounted for 10 percent of the states' expendi- developing world. Table 2.12 shows public spen- tures on irrigation (World Bank, 1998b). Although ding on irrigation for a sample of countries. As India's National Water Policy states that "water can be seen, the average total public spending on rates should be adequate to cover the annual irrigation varies substantially among different maintenance and operation charges and a part of countries, from US$26 to US$275 per irrigated the fixed costs", all states have set charges at very hectare.'8 In countries with large irrigation sec- Table 2.12: Average public expenditures on irrigation for a sample of countries Irrigated area, 1998 Average public expenditure Country (Million ha) (US$ million/year) (USS per irrigated ha) India 59.0 5,000 85 Pakistan 18.0 475 26 Mexico 6.5 250 39 Bangladesh 3.8 160 42 Egypt 3.3 200 61 Viet Nam 3.0 80 27 Morocco 1.2 135 113 Colombia 0.8 35 44 Nigeria 0.2 55 275 Ethiopia 0.1 20 200 Source: Calculated using data from ICID and public expenditure reviews. Table 2.13: Average unit costs for irrigation Unit Cost Unit Cost Activity (US$/ha) (US$/ha) By type of irrigation system: By region: Gravity Irrigation 5,600 East and South Asia 2,800 Pump Irrigation 3,800 East Asia 4,300 Mixed 3,700 South Asia 1,400 New Construction 7,700 India 1,400 Rehabilitation 1,600 Middle East 5,100 Rehabilitation and Extension 3,200 Africa 13,000 North Africa 5,000 Sub-Sahara Africa 18,300 Latin America and the Caribbean 3,900 Source: Jones, 1995. 24 Generating Public Sector Resources to Finance Sustainable Development tors, such as India and Pakistan, total expenditure mined to be the major reason for deferred mainte- can be substantial. nance, which has been threatening the operational Investments in irrigation can be expensive. safety of the irrigation system. Actual funding An evaluation of 208 World Bank funded irriga- available to O&M of system canals was 49 to 29 tion projects shows that the unit costs (calculated percent less that realistic O&M escalating needs as actual project costs divided by the number of identified for proper working of the irrigation hectares of completed irrigation area) can reach system in 1994-1998 (Pakistan National Program, US$7,700 per hectare for some irrigation activi- 1998). ties (Table 2.13). World Bank experience indi- cates that unit costs are generally low where Box 2.11: Irrigation in Central Asia minor improvements are made to existing Substantial investments in irrigation during the systems. New construction projects are much Soviet era have led to a massive dependence on irri- more expensive than rehabilitation and/or exten- gated agriculture in the Central Asian republics in the sion projects. Gravity irrigation costs are 48 per- Aral Sea Basin. Agriculture, almost all of which is cent higher than pump irrigation costs in Bank irrigated, provides 2040 percent of GDP and employs funded irrigation projects worldwide. Unit costs some 28 million people. None of these irrigation funded arrgantionprjet wrldid. Unit cos systems charged more than nominal water fees, result- also vary substantially by region. Table 2.13 also ing in extremely high levels of water use, with water provides average unit costs per hectare in World applications per hectare 50 percent higher than compa- Bank-financed projects in several regions. The rable countries such as Pakistan. The environmental high figure for unit costs in Sub-Saharan Africa consequences of this system have been well docu- reflects the difficulties of implementing and com- mented. They include, most spectacularly, the drying pleting irrigation projects. This makes justifying up of the Aral Sea, but also substantial salinization irrigation investments in the region very hard as problems that affect downstream agriculture and the high on-farm benefits are needed to justify such health of riparian populations. Over the last decade, high unit costs. lack of funding has resulted in plummeting investment in irrigation and drainage systems, and a near-collapse Further costs are then incurred for the opera- of maintenance. As a result, as much as 70 percent of tions and maintenance (O&M) of existing irriga- water abstracted for irrigation is wasted before it tion systems. The high budgetary requirements reaches fields, and many drainage systems are almost involved in establishing and then operation irriga- inoperable. Unreliable or scarce water supply have tion systems are not matched by any inflow of reduced the area irrigated substantially, usually affect- resources, however. ing poorer households disproportionately. The lack of public revenues from providing Source: World Bank, 2002b. irrigation services has substantial consequences Impacts on water use on the sector's performance. In many countries, the irrigation sector is caught in a vicious cycle of Low water prices also result in very high downward-spiraling performance in which the rates of water use. At present, farmers account for lack of public sector resources results in inade- about 80 percent of water use (Perry, 2000). In quate operation and maintenance, leading to poor many parts of the world, the water volume being irrigation and drainage services and eventually to extracted for agriculture and other uses is already farmer dissatisfaction. Dissatisfaction then trans- beyond sustainable yield. Increased demand for lates into low collection rates of irrigation service irrigation services, coupled with a reduction in fees, further weakening irrigation budgets (Perry, new irrigation investments and increased water fe, fGonzalez,n2000). scarcity, makes increasing the productivity and 2000; Gonzalez, 2000). efficiency of existing irrigation systems essential Examples of this cycle abound. Central Asia to improve food security and reduce poverty. The is perhaps the most prominent (see Box 2.11). In main challenge is to increase water productivity Pakistan, one main reason for irrigation system's so as to achieve more crop per drop. deterioration is the deferred maintenance due to a consistent shortfall in maintenance finding for the One desirable effect of charging users for O&M of canals. Inadequate funding was deter- irrigation water would be to discourage waste and reduce demand. To achieve such responses the Generating Public Sector Resources to Finance Sustainable Development 25 charge for irrigation water must be directly related Zeid, 2001). Setting the right price is also a very to the volume delivered. Charging schemes that complex matter in practice, as it requires substan- do not influence water input directly, area pricing tial regulatory arrangements, operational require- or output pricing, lead to inefficient allocations of ments, and financial commitments (Perry, 2000; irrigation water. When fixed charges are in Johansson, 2000). place-for example, where there is a fixed water charge by crop or area-the marginal water price Domestic water is often zero and does not vary with the quantity Domestic water service has serious effects on of water used. In this situation, even if the full the health, economic productivity, and well-being costs of the service are recovered there is no of nations (Fass, 1993; oED, 2000; Ahmed, 2000; incentive for farmers to save water; if the fixed Imschoot, 1992). Recognizing this, efforts to price allows a profit, they will use as much water Improve 1992) cogniz ing corts as they want, or if not they will not irrigate at all Improve service coverage in developing countries (Perry,2000). ~~~~~~have increased substantially in recent decades (Perry, 2000). (WHO and UNICEF, 2000). Impacts on the poor Domestic water service, however, is often As with many other subsidies, low irrigation provided with a philosophy that embraces two water charges are sometimes justified based on concepts: that the right of access to water service their supposed effect on the poor. However, the must be guaranteed to all, and that most people in extent to which water pricing methods can affect developing countries are either too poor or income redistribution among different farmers is unwilling to pay for water service at a price that limited. Farm income disparities are due mainly to reflects its cost of provision (Briscoe, 1999b). such factors as farm size, location, and soil qual- This welfare philosophy is often used to justify ity, rather than to water (or other input prices). keeping water tariffs artificially low through a Water prices have been found to be a poor means range of explicit and implicit subsidies (Foster to reduce income inequality among farmers (Sea- and others, 1999), with significant impact on graves and Easter, 1983; Tsur and Dinar, 1995). public revenues, water resources, and the poor. On the other hand, where irrigation charges are Water subsidies pose substantial burdens for low, or not collected, the direct beneficiaries of public revenues and cause the sector to perform irrigation, who typically are a privileged group in poorly, in terms of coverage and quality of most agrarian economies, receive their service at service, in most developing countries (Idelovitch the expense of the economy in general (Perry, and Ringskog, 1995; Briscoe, 1999b; Kessides, 2000). 1997). Water subsidies also result in enormous Conclusion water waste, by both water users and water utili- ties. Low tariffs, especially when non volumetric, The low prices charged for irrigation water do- little to encourage water conservation by users are a major factor in the difficulties facing the (Cestti and others, 1997), while limited revenues sector. Not only do they impose high budgetary lead to inadequate maintenance of distribution burdens on the govemment, but they also starve systems (Ahmed 2000; Briscoe, 1996, 1999a; the sector of the resources it needs for operations, 1999b; Saghir and others, 2000). Despite their maintenance, and expansion, and it induces stated purpose of helping the poor, water subsidies unsustainably high rates of water use. Thus pric- also tend to adversely affect the poor. Inadequate ing irrigation water is often seen as vital to design means that water subsidies have often increase irrigation system efficiency, to reducing benefited the middle and upper classes rather than the budgetary burden of supporting irrigation the poor-who remain unconnected to the service departments, and to helping conserve and improve (Walker and others, 2000). Poor service imposes water usage in the sector. Irrigation pricing substantial losses of time, health, effort, and reform, however, is a sensitive issue in most of money, especially for the poor, who are continu- the world, with complex political, historical, ally adapting to such inadequate service (Rada social, cultural, and economic dimensions (Abu Research, 1996; Bhatia and Falkenmark, 1993). 26 Generating Public Sector Resources to Finance Sustainable Development Figure 2.3: Connection fees and cost of construction for in-home water service Chil 0,, Chile ° Mean cost of construction in | Philippines w e .72 Latin America (US$144/home) Pakistan 15 Nepal 0 Mongolia 0 Maldives 0 __________ Mean cost of Malaysia - 38 construction in Asia Laos o s8_ Indonesia 1 62 India 21 Bhutan - 70 Bangladesh :49 Seychelles 0 Namibia 0 Mauritius 0 Mean cost of construction Malawi 0 1|3jMean connection fee (US$) in Africa (US$102/home) Lesotho 0 Ethiopia 0 Cote d'lvoire 0 Cameroon 0 Benin 0 0 25 50 75 100 125 15 US$/home Source: WSP, WHO, UNICEF data. As in the previous section, we use a tariff- through delivery tariffs). But water utilities gap approach to measure the size of water subsi- seldom collect sufficient revenues to cover even dies, wherein the water tariff currently paid by their operating costs, let alone capital costs. consumers is compared to a reference per unit cost Connection fees of water service provision.19 A review of 15 water utilities in Africa and Current water subsidies nine utilities in the Middle East yielded a connec- Water utilities face two major costs in tion fee that is close to zero (WSP, 1998). Figure providing service: capital costs, and operation and 2.3 shows the connection fees charged for in- maintenance (O&M) costs. To be financially sus- home water service connections in 21 countries tainable, utilities need adequate resources to cover for which data are available. Within this sample, their capital investments (often through connec- 13 countries provide in-home connections free. tion fees) as well as their O&M costs (often Data on the cost of construction for in-home Generating Public Sector Resources to Finance Sustainable Development 27 connections was not available for individual subsidies often are often quite high, reaching 93 countries, but the figure shows the regional ave- percent of the cost for some countries. rage. Even in countries that charged for connec- Table 2.14 also estimates the total magnitude tions, such as most Asian countries, the fees fell of these subsidies, based on the volume of water short of the cost of providing the connection. In used. These estimates show that even when unit countries where many households enjoy in-home subsidies are low, they can result in substantial connections, these subsidies represent a signifi- financial outflows because of the substantial cant financial burden. amount of water used. Subsidy costs in this For other types of water facilities, such as sample of countries alone reach almost US$8 stand posts, the situation is similar. In most devel- billion a year. oping countries, the public sector fully covers the The impact of low tariffs on revenues is cost of construction, with no fees charged to con- exacerbated by low collection rates. Poor records sumers. The average construction cost per person on the number of actual consumers and the served by stand posts in large cities is about amount of water actually consumed, combined US$31 in Africt', US$64 in Asia, and US$41 in with inadequate billing and collection practices, Latin America and the Caribbean (WHO and create substantial financial losses to domestic UNICEF, 2000). water supply agencies. Most water utilities actu- Water charges ally cover more customers than they know or have on records since most of those unconnected Throughout the world, few developing eventually get their water from the system indi- countries charge users the total cost of providing rectly. Unreported shared connections and illegal water service (Briscoe, 1999b). A recent global connections often exist without the knowledge of assessment showed that more than half of deve- the water utility (Rada, 1996). Low collection loping countries impose water charges that are rates in some countries often result from laws less than the total cost of providing water (WHO prohibiting discontinuing water services in case of and UNICEF, 2000). In the Middle East, for customer payment default, because water is con- example, most countries subsidize charges for sidered a basic human need (Idelovitch and urban water supply. Only in Tunisia, Morocco, Ringskog, 1995). In some countries, only 15-30 and Iran do revenues cover O&M costs and gene- percent of billings are actually collected (WSP, rate a modest surplus necessary to partially cover 2001). capital costs (Saghir and others, 2000). Similarly, while few utilities in Central America have expli- Box 2.12: Explicit water subsidies cit subsidy policies for piped water, a comparison In addition to implicit subsidies resulting from between existing water charges and the estimated under-pricing of water, many countries provide a range efficient cost of providing water shows that most of explicit subsidies. A review of water supply policies undercharge substantially, leading to a generalized in Panama, for example, revealed three different expli- implicit subsidy and to accelerating de-capitali- cit subsidy mechanisms: zation of their systems (Walker and others, 2000). * A subsidy benefiting some 20,000 residential custo- In Asia, cost recovery of working expenses of mers in the metropolitan area who are regarded as rural water supply was reported to be as little as 'social cases'. They represent about 7 percent of the 1.8 percent and less than 1.3 percent of total costs customer base. Eligible households have their entire (including capital costs) in some countries. In bill paid directly by the Ministry of Health at a total India, for example, no state collects tariffs that cost to the state of US$3 million per year. cover O&M apart from isolated district or village * A 33 percent discount on the tariff, benefiting about projects (WSP, 2001). Table 2.14 shows median 60 percent of residential customers at a cost of water tariffs for a sample of countries for which about US$1.6 million per year. data are available and compares them to the * A 25 percent reduction in the tariff (up to a maxi- median unit cost of providing water (WHO and mum monthly expenditure of US$10) for pension- UNICEF, 2000). As can be seen from the table, ers. The total cost of this subsidy was not reported. Source: Foster and others, 1999. 28 Generating Public Sector Resources to Finance Sustainable Development Table 2.14: Water subsidies on delivery tariffs in selected countries Subsidy Domestic Total subsidy Water tarif Cost withdrawals (million Country (US$/m3) (VS$/m3) (US$/m3) (% of cost) (million m3/year) US$/year) Africa Burundi 0.17 0.66 0.49 74 36 18 Egypt 0.07 0.30 0.23 77 3,100 713 Lesotho 0.52 0.55 0.03 5 n.a. n.a. Malawi 0.20 0.36 0.16 44 95 15 Namibia 0.29 0.78 0.49 63 71 35 Nigeria 0.02 0.16 0.14 89 1,125 158 Senegal - 0.66 0.99 0.33 33 68 22 Sierra Leone 0.30 0.54 0.24 44 26 6 Tanzania 0.20 0.30 0.10 33 101 10 Zimbabwe 0.10 0.30 0.20 67 171 34 Asia Bangladesh 0.09 0.20 0.11 55 1,704 187 Georgia 0.01 0.20 0.19 94 728 136 India 0.04 0.20 0.16 80 25,000 4,000 Indonesia 0.19 0.20 0.01 4 4,729 47 Iran 0.10 0.20 0.10 50 4,395 440 Nepal 0.09 0.20 0.11 55 246 27 Latin America Barbados 0.35 0.93 0.58 62 62 36 Bolivia 0.20 0.30 0.10 33 124 12 Colombia 0.25 0.30 0.05 17 5,233 262 Costa Rica 0.04 0.30 0.26 86 757 195 Cuba 0.07 0.30 0.23 77 2545 585 Ecuador 0.04 0.30 0.26 88 2,100 554 El Salvador 0.29 0.30 0.01 3 246 2 Honduras 0.13 0.92 0.79 86 61 48 Paraguay 0.35 0.96 0.61 64 65 39 Venezuela 0.20 0.30 0.11 35 1,800 198 Notes: Tariffs shown are median tariffs. Cost estimates were derived in different ways, and may not be completely comparable. Some were obtained by estimating a univariate cost function using regression analysis, and some by averaging reported production costs from different utilities in the same country. Where no country-specific data were available, the median regional unit production cost of urban water supply is used. Source: Authors' calculations based on data from World Bank, WSP, and WHO and UNICEF. This analysis is unavoidably incomplete. A explicit and implicit water subsidies requires sub- reliable estimate of the magnitude of subsidies in stantial financial outflows. As a result, water the water sector would requires data on connec- systems often fall into a low-level equilibrium tion fees, delivery tariffs, capital costs, and O&M trap in which utilities provide limited or low- costs for each type of water facility in a given quality service because of insufficient resources, country. Such data are rarely available. Neverthe- and inadequate service leads to few resources less, enough data are available to conclude that being collected. Inadequate service is often Generating Public Sector Resources to Finance Sustainable Development 29 reflected in the lack of new connections or in getting rid of wastewater will be low and discour- unreliable service. Both have serious conse- ages reducing water waste. Households that are quences for service users' resources, especially not connected to sanitation systems and pay high those of the poor, who often incur substantial costs for getting rid of waste water consume less costs to seek alternatives. water and are more careful about water leakage and waste than those who are connected to the system and pay low sanitation tariffs (Rada Low water tariffs provide little incentive to Research, 1996). conserve water and result in substantial over-use. Further pressure on water resources comes Low tariffs do little to encourage consumers to from inefficient distribution. Due in part to the reduce their waste and fix leaky faucets and pipes. lack of financial resources to perform routine Studies have found that a large share of the water maintenance and replace old and leaky pipes, used by some buildings-around 350 liters per public water utilities are often inefficient in terms capita per day (lcd)-is due to faulty toilets valves of their use of water. For example, about 45 per- and leaky faucets (Cestti and others, 1997). cent of water is unaccounted for in Bogota, 58 The impact of low tariffs is particularly percent in Manila (Briscoe, 1996), 65 percent in marked when they are non volumetric, as there is Damascus, and around 50 percent cities such as then no incentive whatsoever to limit consump- Amman, Algiers, and Sana (Saghir and others, tion. In developing countries water tariffs are not 2000). In many utilities the situation is so bad that only low but also mostly non-volumetric. Most losses are controlled by having water in the distri- cities and rural areas in the developing world are bution system for only a couple of hours a day, not metered. Table 2.15 provides a review of 111 and by keeping pressure very low (Ahmed, 2000; utilities all over the world and shows that around Hoehn and Krieger, 1996). In Madras, it is esti- 95 percent of utilities do not meter their customers mated that if supply was to increase from 2 hours (WSP, 1998). In Asia, cities with un-metered a day at 2 meters of pressure, to 12 hours a day at water supply have total consumption levels of 10 meters of pressure, leaks would account for about 600-630 lcd, well above consumption in almost three times the current city supply metered cities, which range between 200-400 lcd (Briscoe, 1999b). In Cairo, reducing domestic (Cestti and others, 1997). A similar pattern exists leakage by 10 percent, system-wide leakage by 3 in Latin America, where un-metered communities percent, and institutional leakage by 15 percent, use around 400 lcd, while consumption in metered would allow up to seven hours of additional cities like Santiago is only 240 lcd. In Panama, service to be provided to those connected to the metering lowered consumption by over 20 percent water system. Since most households have a high in four months (Cestti and others, 1997). willingness to pay for continuous water service, In addition, if low water charges are accom- reducing leakage and using the saved water to panied with low sanitation charges, the cost Of improve service could result in substantial bene- fits-as much as US$3 million per month, Table 2.15: Utilities that do not meter users Number of utilities Region Number of countries Number of utilities with no metering Africa 9 15 15 East Asia and Pacific 19 41 35 Eastern and Central Europe 11 27 27 Latin America the Caribbean 1 3 3 Middle East and North Africa 6 11 11 South Asia 7 14 14 Total 53 111 105 Source WSP data. 30 Generating Public Sector Resources to Finance Sustainable Development depending on how the saved water is allocated (Ahmed, 2000). Figure 2.4: Comparison of vendor and utility water prices in selected countries Impacts on the poor Although protecting the poor is put forward Turkey - as an important rationale for the current tariff Peru __ structure, in most cities of the developing world ; current tariffs tend to heavily subsidize the rich Ivory Coast (Briscoe, 1999b). A recent study shows that there is little income-related differentiation in effective Ecuador . piped water tariffs. Each of the cities studied had Colombia volume-based tariffs, which would have generated cross subsidies from the rich to the poor if the rich 0 4 8 12 16 20 were to consume more water. However, the Prvate vendor water pnce as demand for water in the cities studied varied little multiple of utility water price with income. As a result, the richest 60 percent of the households captured most of the subsidy Source. From data in Bhatia and Falkenmark, 1993. (Walker and others, 2000). become fewer and fewer) and quality (due to contamination from poor sanitation and industrial Box 2.13: Water subsidies: the case of increasing affluent) (Bhatia and Falkenmark, 1993). In most block tariffs developing countries, the poor pay very high For households that have private and metered water prices for alternative water supplies (Briscoe, connections, charges are often based on Increasing 1 999b). Water vendors often charge prices that are Block Tariffs (IBTs), in which two or more prices are many times higher than those charged by public set for water, with each price applying to consumption utilities in some cities, as illustrated in Figure 2.4. within a defined block and prices rising in each succes- In Nigeria, for example, water vendors were sive block (Boland and Whittington, 2000). A survey found to collect about 24 times as much revenue of urban water utilities in Asia, for example, found that 20 out of the 32 utilities surveyed used IBTs (ADB, as the public water utility (Whittington and others, 1993). Some experts argue that IBTs have the potential 1991). Poorer households also have more diffi- of implementing marginal cost pricing principles (Hall culty in adapting to unreliable service. For exam- and Hanemann, 1996). However, in practice develop- ple, few of Cairo's poorer households have water ing countries often subsidize such tariffs by applying pumps that increase pressure, as such pumps can IBTs so that the first block price does not cover the cost as much as 100 percent of annual household cost of providing the service. Moreover, utilities in income (Rada, 1996). such countries find it difficult to limit the size of the initial block. About 4-5 cubic meters per month are Reforming water tariffs required for basic needs (based on a family of five), but Reforming water tariffs would seem strongly most utilities set the first block in excess of 15 cubic meters per month (Boland and Whittington, 2000; justified from both a financial and an environ- Asian Development Bank, 1993). This means that most mental perspective. However, reform needs to be connected households can include all their consump- undertaken carefully. Policy changes require tion within the first subsidized block. paying attention to consumers' ability and will- Not only do the poor generally derive few ingness to pay for water. The removal of subsidies benefits from water subsidies, they also tend to involves some implementation costs. Removing suffer disproportionately from the poor service subsidies on delivery charges can be costly, as it that results from utilities' financial difficulties. requires water utilities to implement a mechanism The poor are often at the end of the line in terms to measure water usage, most probably through of new connections and reliability of existing metering, and to enforce its new pricing policy. service. As a result, they continue to depend on Metering can be costly and its benefits need to be traditional sources of supply, which are often evaluated (Cestti and others, 1997)i Removing afflicted by declining access (as new sources subsidies on connection fees is usually rather less Generating Public Sector Resources to Finance Sustainable Development 31 complicated. Having to pay substantial lump sums at once may well 'cause problems to many con- large costs for those governments and benefits to sumers, however. those in other countries. Conclusion Previous studies (such as IEA, 2001a) aim to measure total subsidies in the energy sector, and so The reformulation of tariffs and subsidy poli- exclude the subsidies to products employed in cies is crucial to reforming water services in electricity- generation in order to avoid double developing countries. Reforms have proposed to counting. We are here specifically interested m the improve the provision of such services through a total subsidies embedded in the electricity combination of measures that encourage private 6subsector. sectorpn(to improve efficiency), the This latter form of subsidy is not reflected in a sector participation v financial burden to the government in the current removal of subsidies (to permit for sustainability), . . . ~~~~~~year, but in increased future costs. Put another way, improving volumetric pricing (to allow for less our results reflect an 'economic' measure of water waste), and better targeting of existing subsidies, rather than a 'financial' one. tariffs (to ensure that reform does not harm the 7 This mark-up is intended to reflect the higher costs poor). The removal of subsidies will have sub- of serving non-industrial customers, as explained stantial impacts on public revenues, the water previously. We use a mark-up of 1.5 c/kwh instead resource, and the poor. of 2 c/kwh as a conservative estimate, in order to avoid including in the cross-subsidization results countries where the differential distribution costs Notes between industrial and non-industrial customers is particularly narrow due to individual circumstances. 8 Here and elsewhere, estimates from other reports Unless natural gas were to be subsidized more than have been adjusted to 1999 prices for compara- other fossil fuels. In theory, the use of the cleaner bility. fuel could offset the environmental impact of 9 This is the estimated amount that would cover all greater consumption. expenditures for road maintenance, depreciation, 2 We follow the International Energy Agency (IEA) debt servicing, and new highway construction in the in using the broadest possible definition of subsidy: United States. However, it would pay little more (if "An energy subsidy is any government action that at all) than road maintenance in developing coun- concerns primarily the energy sector and that tries, where traffic density is much lower. In addi- lowers the cost of energy production, raises the tion, since heavy good vehicles with few axles, price received by energy producers, or lowers the typi-cally powered with diesel, do most of the price paid by energy consumers" (IEA, 1999: 43). damage, our 'transport sector measure' results may 3 It can be argued that, for OPEC countries, the over-estimate the 'subsidy' for gasoline, and under- existence of export quotas reduces the opportunity estimate the 'subsidy' for diesel. cost of fuels from the level of international prices to ° Metschies (1999, 2001) reports the prices for that of production costs. We follow IEA (1999) in gasoline and diesel for over 160 countries, in a day assuming implicitly in the analysis that each OPEC in the last quarter of 1998 and 2000, respectively. country could leave (implicitly or explicitly) the To create an average price for 1999, we have cartel and sell in the international markets the combined prices for 1998 and 2000, giving the a amount of oil consumed domestically without 0.75 weight to those in 1998 and 0.25 to those in affecting the international price of oil. Since this 2000. assumption cannot be sustained if a significant " Data on the consumption of gasoline and diesel in number of OPEC countries decide to leave the cartel, our global estimates of subsidies - obtained every sngle country do not exist. Our sample by aggregating the subsidies computed for indivi- world's GDP, and is likely to represent a similar dual countries - would represent an over-estima- percentage of transport fuels consumption. For Sub- tion. In what follows we do not attempt to correct Saharan Africa, howevert our sample represents forthse'glba euilbrumeffects. 'aaa Afia oee,or aperpeet for those 'global equilibrium' effects. only around 85 percent of the region's GDP. We 4 An important aspect of inter-country tax differences have decided not to scale up the results, since we do is fuel smuggling. Extensive smuggling of fuels out not have a direct measure of the non-represented of Indonesia or Nigeria, for example, represent consumption and it is likely to be a small figure. 32 Generating Public Sector Resources to Finance Sustainable Development 12 United States: gasoline 36.5 usc/l, diesel 33.7 usc/l. 17 These are the unweighted average of the 105 European Union (unweighted average): gasoline economies in the sample: 49.6 usc/I for gasoline 103.3 c/l, diesel 83.7 usc/l. and 37.7 usc/I for diesel. Note that average prices in 13 Although four countries (Namibia, Sri Lanka, all regions except the Middle East and North Africa Turkmenistan, and Ukraine) present gasoline prices are above these prices. Note also that these prices in the range of those in the European Union (that is, do not even reach 50 percent of the average prices higher than 85 usc/l), the same does not happen in the European Union. with diesel (in none of them diesel reaches 70 18 International donors have also made substantial usc/I). financial commitments. Irrigation is the largest 4 Numbers range from around 90 percent in Latin recipient of public agricultural investment in deve- America and the Middle East to around 70 percent loping countries. Active World Bank commitments in Sub-Saharan Africa and the former Soviet Union. (covering 229 projects) for irrigation totaled over 5 It must be noted that the concept of the energy US$7 billion in 2000, for example (Gonzalez, ladder is contentious. Households do not abandon one fuel completely to move up the energy ladder i9 Ideally, long run marginal costs of water supply until their income rises to very high levels. Rather should be used. However, data constraints preclude they tend to use multiple fuels, especially if this. biomass is abundant. 16 See Figure 2.2 above for average regional gasoline and diesel prices. 3. Generating New Resources Improving the use of existing resources is sector would encourage more sustainable practi- important, but will likely not be sufficient by ces. Clearly, this potential is limited to forest-rich itself. In this chapter, we examined the potential countries, particularly in East Asia and Latin for generating new resources. This potential is America. However, such rents could provide very substantial, but varies significantly across coun- significant revenue sources for a number of tries. We examine here three main approaches, in smaller countries. sectors which are particularly relevant for sustain- This section begins with a brief overview of able development. We begin by examining the the theory and rational for natural resource rent potential for capturing a greater share of natural capture, with a focus on minerals and mineral resource rents, focusing on the potential in the fuels. It then concentrates on forest rents and the forestry sector. We then examine the possibility of means to capture them. It also briefly review charging for services which are currently provided efforts to capture rents from bioprospecting, free or nearly so, such as access to protected areas which was once thought to have the potential to and waste management. Next, we discuss the provide a significant flow of additional funds. As potential for a range of taxes and charges. Finally, will be seen, these hopes have largely been we briefly review the potential for new mecha- disappointed. The final part of the section nisms, such as Payments for Environmental provides some rough estimates of the overall Services (PES), to generate financing for specific potential for natural resource rent capture from conservation activities. different sectors. 3.1 Capturing Natural Resource Rationale Rents All governments tax economic activity in The value of a natural resource is defined by order to finance social and economic policies. its value when extracted-the resource rent- However, there are two main reasons why govern- which is the difference between its market value ments should tax natural resources in particular: and the full cost of extracting it. Full extraction costs include wages and salaries, depreciation of na lesource u b t the capital stock (production machinery and government, and structures), and the opportunity cost of the capital 2. over-exploitation of natural resources can employed (typically an assumed 'normal' rate of occur i the absence of fiscal incentives. return on capital). Total resource rent is therefore Natural resources are typically in the public equal to the economic profit of extraction. domain, but are exploited by the private sector. The bulk of potential revenue from rents is, This is the prime motivation for governments to of course, from mineral fuels and minerals. These capture some portion of the economic profits rents are probably already being mostly captured, (rents) from resource extraction. There can also be however. While there may be some scope for important incentive effects when resources are some marginal improvements in rent capture, inadequately taxed as well: the private operator these resources provide limited scope for' gene- will be tempted to exploit the richest ores or forest rating additional revenue. In contrast, the potential concessions as quickly as possible, before the for additional rent capture is substantial in the government wakes up, leading to dynamic forest sector. The potential revenues from this inefficiencies. sector are clearly much smaller, amounting to Under perfect competition, economic profits about US$9 billion, and evidence indicates that can be reduced to zero with a potentially unlimi- only a small part of these potential revenues is ted number of agents entering the market. currently being captured in developing countries. However, natural resources do not permit unlimi- Moreover, increasing rent capture in the forestry 33 34 Generating Public Sector Resources to Finance Sustainable Development ted entrants, given the fixed number of resource investment within the tax holiday period, it may deposits known at any point in time. Under these be optimal for them to shut down rather than pay circumstances economic profits can persist simply the deferred taxes, leaving the government with because supply cannot be increased indefinitely. no rental payments. In an ideal world, these economic profits or Two common approaches to resource taxa- resource rents will go to the resource owner, the tion involve production sharing and specific taxes government. The focus of this section is on the (royalties). Under a production sharing agreement, capture of resource rents through fiscal instru- a share of the gross proceeds of the sales of the ments. resource is paid to the government. While this The capture of rent involves a series of approach shares the market risk between govern- considerations. For instance: When should the ment and the exploiting firm, it represents in payment be made? On what basis should the pay- effect a tax on gross proceeds. Such a tax can ment be made? How much should be paid? There severely impact a company's bottom line when exists a myriad of policy instruments available to market conditions are unfavorable or production a developing country government that fall under costs increase. the broad heading of resource taxation. Royalties, typically administered as a speci- Uncertainty is a key aspect of the extraction fic tax on physical quantities of resource (per ton of subsoil assets simply because they are under- or cubic meter), are among the oldest of fiscal ground, 'unseen' and fixed in supply. While the instruments. However, these also represent a tax extent of forest resources is more readily measu- on gross proceeds, with the attendant problems red, there can be considerable uncertainty con- from the point of view of the resource exploiting cerning resource quality and harvest cost. Uncer- firm. In addition, as a fixed unit tax, the govern- tainty regarding available resource rents therefore ment is not in a position to benefit from buoyant exists at various levels: (1) the quantity and qua- resource prices without changing royalty rates. lity of subsoil resources; (2) extraction costs; and A balanced approach to resource taxation (3) future resource prices. often combines instruments, in particular the com- These uncertainties create a degree of risk bination of a royalty and a profits tax. The royalty facing the extraction companies and the govern- is set low, to limit risks to firms, while the profits ment. A combination of instruments can be an tax captures at least some portion of the economic effective way of sharing the risk between public rents from exploitation. sector and private resource-extracting firms. From There are several guiding principles for desi- the companies' point of view there is the risk that gning fiscal regimes for natural resources. Fiscal extraction costs may be too high and resource instruments should ensure: prices too low, to the point where profits are zero * an adequate return to goverment, reflecting its or negative. From the government's perspective, role as resource owner and the sovereign there is the risk that the revenue streams derived taxing authority; from the taxation of extraction companies may be * timely receipt of revenue through the life cycle highly variable. Dtml eep frvnetruhtelf yl of the mineral deposit or forest concession; Resource taxation regimes that defer r risk sharing between the private sector and the taxation-a tax holiday of 10 years for example- government that permits efficient development are commonly seen in developing countries. of the sector without exposing the budget to While this approach is used as an incentive to excessive revenue variability. attract investment in the resource sectors, it raises issues for the realization of government revenues. Forests Pri-vate firms operating under such a regime may be tempted to 'high-grade' the resource (exploit This section discusses some key issues the richest sources first, often to the detriment of involved in resource taxation related to the the exploitability of the remaining resource). If forestry sector. We examine the extent to which these firms have written off their capital developing country governments have been able Generating Public Sector Resources to Finance Sustainable Development 35 to extract rents from the use of forest resources. generation and by the contribution of these We analyze some country case studies and iden- resources to GDP. As the table shows potential tify revenues obtained from forests in these coun- forestry rent in forest-rich countries ranges from tries. We also examine some of the constraints US$3 million to over 'US$2 billion. If this rent faced by governments in increasing the efficiency could be captured by governments, several coun- of forest resource use. Our analysis is selective tries would have significant additional resources due to the limited amount of information available available for development activities. For example, and the variation in forestry issues from country to governments from large countries such as China country. and Brazil may be able to generate over US$1 A basic justification for increasing rent billion annually from forest rents. In many smaller capture in the forestry sector is because doing so countries, potential forest rents make up a non- will reduce the above normal profits currently trivial component of GDP. enjoyed by many logging companies. A re- Our review below suggests that only a very distribution of rent from the private to the public small portion of these potential rents from forest is sector will free up resources for sustainable deve- currently being captured by developing country lopment. Further, reducing the profits logging governments. In many cases, forest rent accrues companies obtain, through appropriate pricing primarily to logging firms, or is being simply thar reflects the scarcity value of trees, can pro- allowed to dissipate because of inefficiencies.2 In vide an incentive for more efficient timber ex- the following pages, we discuss some of the traction and decrease waste (Gillis, 1990). When reforms being undertaken by governments which governments collect part of the rents of har- could increase rent extraction from the forestry vesting timber accruing to private companies, sector. arguably there is less incentive to indiscriminately Taxes and fees applied to the forestry fell trees.' sector Table 3.1 shows potential rents that could be collected from timber in the top 10 developing Governments have a variety of fiscal policy countries, ranked by size of potential resource instruments that they can use to capture rents from Table 3.1: Potential rents from timber in forest-rich countries Top 10 countries by potential revenue Top 10 countries by % GDP USS % US$ % Country million GDP Country million GDP China 2,249 0.2 Solomon Islands 16 6 0 Brazil 1,489 0.3 Central African Republic 18 1.8 Indonesia 703 0.5 Papua New Guinea 69 1.8 India 635 0.1 Guinea-Bissau 3 1.6 Malaysia 488 0.5 Gabon 57 1 2 South Africa 394 0.3 Guyana 8 1.1 Mexico 252 0.0 Lao People's Democratic Republic 19 1.1 Chile 222 0.3 Chad 16 1.1 Poland 207 0.1 Paraguay 69 0.9 Nigeria 194 '0.5 Mongolia 8 0.8 Notes: 1999 data. Assumes that 50 percent of total available resource rent can be captured. Omits fuelwood. Source: Derived from World Bank data, see Appendix Table C.3 for details. 36 Generating Public Sector Resources to Finance Sustainable Development Figure 3.1: Flow of timber through the forest sector and application of taxes Government Taxes Forest Fees Forest Owner (o Concession Fees Pre-Harvest Fees: Sta gVolume v-Fees determined before S extraction of timber e Allowable Cut Fees l o Development Taxes . k * e Perfornance Bonds o Corporate Profit > Timber Extraction Taxes Industry * Sales Taxes H o Transport Taxes > Harvest Fees: o Volume Based Fees Fees determined by l Per Tree Based Fees / - quantity of timber * Value Based Fees * Export Taxes EXPORT extracted from the ( o Reforestation Fees forest * Log Export Fees * Corporate Profit Timber Processing Taxes Industry * Sales Taxes 1 Transport Taxes } Post-Harvest Fees: e Processed Product Fees determined Fees / -following processing of * Processed Product * Export Taxes EXPORT timber I Export Fees o Corporate Profit Taxes Wood Product e Sales Taxes 5 Retailers * Transport Taxes Source: Jiwanji, 2001. the forestry sector. Figure 3.1 provides an over- timber harvested (per cubic meter, m3), the view of the types of fees that apply to the forestry number of trees harvested, or the value of the trees sector. As the Figure shows, governments can harvested. Each type of fee has its own extract rent through pre-harvest, harvest, or post- advantages. However, fees based solely on the harvest fees. A number of routinely used taxes are quantity of timber extracted may lead to 'high- also levied on the forestry sector. grading'-the practice of selectively extracting The most common type of fees that apply to only the most valuable tree species, leaving har- forest sector activities are stumpage fees or vestable timber in the forest. This is economically royalties based on the amount of timber harvested. inefficient in the long run, because it creates disin- These types of fees can be based on the volume of centives for silvicultural practices that would Generating Public Sector Resources to Finance Sustainable Development 37 maximize the net present value of log production and do not require measuring and classifying the under sustained production-that is, the entire harvested timber. However, such fees could result value of areas set aside for logging is not being in inefficient harvesting of low value timber and captured. It can also result in increased wasteage un-necessary destruction of forest biodiversity. because trees are often damaged or destroyed in Setting the correct fees may require a considerable the process of getting to high-value trees scattered amount of information on the quantity and quality in large forest segments. In general, stumpage fees of timber at specific sites, unless the government and royalties have the potential to capture a high auctions concessions. Auctioning concessions proportion of rents if they are differentiated may require less information, since governments enough across species and are updated frequently can establish a minimum price for the concessions to reflect market conditions. Their key disadvan- and the burden of evaluating the full value of such tage is that they can be quite complex to deter- a concession falls on bidding companies. Under mine, and can be a considerable burden on under- competitive conditions, auctions can be a mecha- staffed and poorly-funded forestry agencies that nism to extract a high proportion of the rents are in charge of measuring and classifying timber available. However, the auctioning process is not harvest. free from political maneuvering problems, as Fees can also be applied to timber after it has shown in the Cameroon case study below. been extracted and processed into other wood In addition to fees specific to the forestry products such as sawn wood, veneer, and ply- sector discussed above, other types of government wood. Fees on processed products are classified as taxes and charges may apply to activities in this post-harvest fees, as opposed to the harvest-based sector. Just like other types of economic activities, fees discussed above. One of the main advantages the forestry sector may be subject to sales, export, of post-harvest fees is that they can be applied to and corporate profit taxes. While the primary logs that are illegally harvested. In countries motive for such taxes is to raise revenues, they do where a large proportion of timber is processed capture some amount of rent if forest-specific fees for exports, and illegally harvested timber makes are below efficient levels. These taxes can some- up a large portion of processed wood products, times be easier to collect, since they rely on insti- post-harvest fees can raise signifi-cant revenues. tutions and procedures that are a part of the They are also administratively easier to apply normal tax raising functions of a government. But since it is relatively simple to measure the quan- because such taxes are not specifically designed to tity of products manufactured or exported. Post capture resource rents, they are usually inefficient harvest fees are quite common and many govem- mechanisms for this purpose, and have little or no ments apply post-harvest fees (or even outright incentive impact on the logging and wood pro- export restrictions) on un-processed logs to encou- cessing industry. rage domestic processing of timber. However, Table 3.2 provides some examples of forestry because post-harvest fees are often estimated taxes currently collected in selected developing based on the amount of timber that would be countries. As can be seen, most countries rely on required to produce a given amount of processed several instruments-stumpage fees, reforestation wood, these fees can penalize more efficient pro- fees, area taxes, export taxes, profit taxes, and so cessing firms. on-to raise revenue from the forestry sector. Governments have increasingly adopted pre- Annual revenues from the sector range from some harvest fees, such as concession fees and areas US$12 million in Laos PDR to over US$300 fees, in addition to harvest-based fees. There are million in Brazil. While these revenues are small some advantages to using pre-harvest fees from a in relation to total GDP, they still represent scarce revenue perspective. Pre-harvest fees are indepen- resources that developing country governments dent of the amount of timber harvested, and can put to good use. Further, comparing these encourage an intensive use of the forest with high numbers to the potential rents in Table 3.1 above recovery rates of timber from all valuable tree shows that in all cases, except for Cameroon, the species.3 They also tend to be simpler to collect, rents available to governments can be increased since fees are defined in the concession agreement considerably by improving the tax mechanism. 38 Generating Public Sector Resources to Finance Sustainable Development Table 3.2: Forestry policy instruments and revenue generated in selected countries Revenue % of (million % of Potential Country Current Policy US$) GDP Rent Brazilb * Regulations limit amount of area that can be harvested in 345.0 0.07 23 private lands * ICMS, a value-added tax, is primary source of revenue Cameroon' e Stumpage fees: 2.5% of FOB price 59.5 0.65 104' o Export taxes: US$12 to US$24/m3 * Area taxes: US$2.40 to US$3.90/ha/year e Transfer taxes: US$0.16/ha Gabond o Processed wood fees: US$0.3 to US$0.6/ m3 31.7 0.75 56 o Export taxes: 5% (processed wood) and 1 1% (logs) of FOB price, treasury tax of 0.2% of export taxes and duties, port tax of US$0.9 per metric ton * Area tax: US$0.006 to US$0.03 Indonesiae o Stumpage fees: US$16 to US$20/ m3 199.0 0.14 28 e Area tax: US$3 to US$10/ha/year * Export taxes: 20% on logs and sawn wood Laosf * Royalties 12.6 0.87 66 * Reforestation Fees o Provincial/Local Development Fees Russiag o Government sets minimum stumpage prices, for which n.a., see municipalities can charge fees (in 1997 these fees notes amounted to less than US$1/rm3, on average) o Competitive auction of lease rentals have been recently introduced and bids have been as high as US$10/ m3 * Profit tax is the primary source of revenue (export taxes were abolished in 1996) Notes and sources All figures have been adjusted to 1999 US dollars FOB = free on board (that is, the market value of the tree) n.a. = not available ' Derived by dividing actual rents by the potential rents shown in Table 3.1. b Revenue figures are for 1998 and only for the Amazon region, which includes the states of Para, Mato Grosso, and Rondonia. These states provide 93 percent of the timber production. Source: World Bank, 2000b. C Revenue figures are for 1997 Export taxes apply to volume of raw logs exceeding allowed quota and in 1997 generated almost 80 percent of revenues from forestry taxes. Source: Cameroon, 1998. d Revenue figures are for 1997. Other fees include: pre-harvest fees ($0.60 per hectare), timber marking tax (US$0.3 to 4.20), reforestation tax applied to wood exported, to name Just a few. Export taxes accounted for 85 percent of revenues from forestry taxes in 1997. Source: Global Forest Watch, 2000a. ' Revenues figures are for 1998 and include only revenues from the Dana Reboisasi (DR) and the Iuran Hasil Hutan (IHH), the two largest forestry fees. Revenues were significantly lower than in 1996 (US$340 million) due to the devaluation of the Rupiah. Sources: World Bank, 2000c, 2000e; D W. Brown, 2001. fRevenues are from Royalties for 1998/99. Source: World Bank/Sida/GOF, 2001. S A World Bank study estimated the potential tax revenue from the forestry sector in 1996 was between US$900 million (under the most conservative estimate) and as high as US$5.5 billion, depending on assumptions about wood production levels and profitability of harvesting companies and wood processors. The calculations assumed stumpage fee rates ranging from US$3 to US$10 per cubic meter and a profit tax rate of 35 percent. Sources: European Forest Institute, 2000; World Bank, 1997b. h In calculating potential rent, it is estimated that the government can capture 50% of total estimated rent. This figure shows that the Cameroonian ezovemment is able to obtain more than 50% of rent. Generating Public Sector Resources to Finance Sustainable Development 39 Case studies recent round of concession allocations (in June 2000) were far more transparent than in the past. ln this section, we review several case studies Th chne aeaeepctdt euti h that illustrate the potential and pitfalls of attempt- Thle changes made are expected to result in the tha ilusrae he otntal ndpitals f atept government earning approximately US$6.5 ing to capture resource rents in the forest sector. goemn eaig apoiaeyU.. million in revenue per year. In fact, revenue per In general, the case studies illustrate the need for h forest sector reform and careful monitoring of obtained in 1997 (Collomb andtBikie, 2001). newly instituted policies. Cameroon Gabon Gabon, like Cameroon, is a major exporter of Reform of Cameroon's forest policy was tropical logs from Central Africa (ITTO, 2000). linked to structural adjustment lending conditions Gaopials drivestmos ofrits fOr sector that followed the onset of the economic cn'sis in Gabon also derives most of ItS forestry sector revenue from export taxes, as Cameroon does. Yet the late 1 980s. The reforms introduced four major Gaocletdonyhfasmhinrvus changes: allocation of concessions through an Gabon collected only half as much ev revenues aucton sste, ne prcingand axig meha- from the forestry sector in 1997 even though it auction system, new pricing and taxing mecha- produced 2.7 million cubic meters of logs, appro- nisms, requirements for management plans, and ximately the same volume of logs as Cameroon's provisions for community forestry. These reforms 3 million cubic meters. Total forestry taxes in have increased the amount of revenue the govern- Gabon were $32 million in 1999 dollars, com- ment collects from the forestry sector. Efforts to pared to $60 million in Cameroon. Area taxes introduce auctioning of logging licenses proved were 75 times lower than in Cameroon. Thus, particularly troublesome in the implementation of forest reforms in Gabon are likely to pay high new reforms. An audit conducted by the Global dividends in terms of increased revenue to the Forest Watch (2000b) found that 56 percent of govermment. licenses in 1997-98 were operating irregularly- licenses had either expired, were operational in a There is considerable scope for change in protected area, or had been improperly allocated Gabon's forest sector. Most forestry taxes in after 1994. Also, 21 of the 31 concessions (Unites Gabon have not changed in the past 25 years Forestieres d'Amenagement, UFAs)4 allocated did despite high inflation. The more than 12 different not go to the highest bidder. It is estimated that types of forestry taxes that exist also make the Cameroon government lost at least US$2.5 monitoring difficult. In addition, forestry agents million, or 4 percent of potential forestry tax police on average over 800 square kilometers of revenues, by not allocating 14 UFAs to the logging concessions per agent, which further highest bidders in 1997. As a result of auction aggravates the problem of rent collection (Global problems in 1997, allocation of UFA concessions Forest Watch, 2000a). were temporarily stopped. Logging companies Reforn of Gabon's Forestry Law began in either obtained a vente de coupe5 or an auto- 1996. The new laws propose long and medium risation de recuperation,6 to continue operation in term management plans for some types of logging the profitable logging sector or engaged in illegal concessions, allocate logging permits through an harvesting. auction to pre-qualified companies, and simplify Recent evidence suggests that continued the existing tax structure (Global Forest Watch, pressure on Cameroon's government to clarify the 2000a). Thus far, implementation of the new rules of the bidding process has had a positive measures has proceeded quite slowly. impact on the forestry sector. Public disclosure of Brazil information from an independent observer, in particular, appears to be leading to stronger enfor- The forestry sector in Brazil, while exten- cement of the new regulations. Evidence of this is sively regulated, is characterized by a weak insti- the government's recent suspension of licensed tutional environment and weak enforcement. activities of 32 logging companies for failure to Regulations governing the exploitation of natural pay taxes amounting to US$4.1 million (ITTO, forests have followed more of a command and 2001). Further, latest reports suggest that the most control approach than a pricing approach. Regu- 40 Generating Public Sector Resources to Finance Sustainable Development Box 3.1: "Conservation concessions" in the Guyana Shield region Conservation concessions are a new conservation can countries combined. The reserve created is part of mechanism that allow conservation groups to lease the Guyana Shield region, one of the largest un- forest tracts from the govermment in the same way a fragmented tropical forest blocks which extends from timber company would - except that these areas are the northern part of Brazil to Venezuela and Colombia. used for conservation instead of logging. Suriname, and Guyana are two of the most "forest Environmental NGO Conservation Intemational (CI) rich" countries (90 percent and 80 percent of their land established Guyana's first 'conservation concession' in area are under forests, respectively) whose economies its upper Essequibo/Takutu Region in 2000. CI depend heavily on natural resource extraction. Both approached the government of Guyana for the countries have been under pressure to exploit these establishment of this concession after it carried out a forest resources which until recently were largely similar project in Suriname, where CI established a undisturbed. Export of timber generated a mere US$6 trust fund to support the management of the new million for Suriname in 1995 and less than US$1 Central Suriname Nature Reserve. The trust fund has million for Guyana, much lower than the estimated so far received pledges of over US$18 million from "forest rent" potential in these countries. "Conserva- multilateral donors. These resources will enable tion concessions" present a feasible altemative to Suriname to manage protected areas covering nearly generating revenues from a conservation-based deve- 15 percent of the country's total area. Suriname lopment strategy, as opposed to an export-based deve- harbors more rain forest than all of the Central Ameri- lopment strategy. Sources: Hadden, 1999; Conservation Intemational, 2000a, 2000b; Hardner and Rice, 2002. lation of the forestry sector falls primarily into six be somewhere between 92 and 112 million areas: environmental impact assessments, burning hectares (World Bank, 2000e). With such an and clearing permits, property-specific cutting res- abundant endowment, it is inevitable that some trictions, geography-specific cutting restrictions, deforestation will occur in the pursuit of economic forest management requirements, and restrictions development. In fact, the government of Indonesia on the export of forest products. For example, the has exploited its natural resources in an export-led Forest Code requires that 80 percent of forested development strategy, which resulted in a area on private lands in the Amazon must remain sustained and rapid rate of economic expansion under forest cover, although small holders may be lasting nearly three decades. However, accounting exempted from this requirement. These and other for the loss of natural capital may reduce regulations of logging activity, when enforced, Indonesia's performance from impressive to amount to an implicit tax on the industry. How- ordinary (Hamilton and Clemens, 1999). ever, these restrictions are not much of constraint Evidence that the rate of forest exploitation since over 80 percent of logging in the Amazon was unsustainable began accumulating by the late region is illegal (Lele and others, 2000). 1990s (Seymour and Dubash, 2000). As the Tax evasion in the forestry sector is economic crisis and devastating fires of 1997-98 estimated to be very high. The primary source of showed, Indonesia's economy was built on ques- revenue from the forestry sector comes from a tionable foundations. Nowhere is this weak foun- value-added tax. When calculations based on the dation more evident than in the forestry sector, rate of uncollected taxes in the state of Para (88 where timber concessions were used for political percent) are extended to the entire Amazon patronage. The traditional rights of indigenous region, the amount of revenue loss would be in the forest dwellers and communities dependent on order of US$225 million in 1998 (Lele and others, forest resources were subordinated to the indu- 2000). strial interests of the APKINDO plywood mar- Indonesia keting cartel, which received incentives that gene- rated large economic rents for license holders Indonesia is endowed with the second largest (World Bank, 2000e). It is estimated that the rents expanse of tropical moist forest in the world. The flowing to plywood producers have soared since actual extent of forest cover remaining is not the economic crisis, from an estimated US$25- known for lack of reliable data, but is believed to $80/m3 to US$172/m3 in the first few months of Generating Public Sector Resources to Finance Sustainable Development 41 1999. In addition, windfalls occur to timber 0.3 percent to 1-2 percent of the national forest producers because the government of Indonesia area (Lao PDR, 1999). collects stumpage fees at artificially low exchange Log production in Laos has been growing at rates. In all, these policies are estimated to have an increasing rate since the 1980s, particularly cost the government of Indonesia over US$10 from the mid 1990s onwards. In 1998, 34 percent billion dollars in forgone tax revenue from 1990 of total export value came from forest products. to 1999 (Brown, 1999). As other exports are heavily based on imported Reform of the forestry sector was made a part inputs, the importance of the forestry sector is of the second adjustment lending package that the even larger in terms of net foreign exchange IMF and World Bank provided to restore stability earnings. Royalties on logs have been an impor- in the Indonesian economy in 1998. The reform tant source of revenue to the government of Laos. package included measures to increase taxation In 1998-99, the government of Laos earned about from the forestry sector by raising stumpage fees, US$12.6 million from timber royalties. However, auctioning concessions, implementing perfor- it is estimated that the government of Laos lost mance bonds, and reducing marketing and invest- some US$114 million in revenues from 1994 to ment restrictions in the forestry sector. The results 1998/99 due to low royalty collection (World of these reforms have been mixed. Some progress Bank/Sida/GOF, 2001). has been achieved in dismantling forest product Payments for logs consist of three marketing monopolies. But spirit of the reforms components: a royalty and a reforestation fee that appear to have been undermined in several ways are payable to the national treasure, and a local (Seymour and Dubash, 2000). For example, the development tax/fee that varies across provinces. proceeds collected from the performance bond Wood buyers may be exempted from the still go into what is called a 'reforestation fund', reforestation fee if they plant trees. However, in which has been primarily used to subsidize general, the scale of planting does not match the plantations of exotic hardwood species. The amount harvested. Royalty rates also differ export tax was lowered from 200 percent to 30 according whether the log is processed or percent in 1998 as stipulated. However, export exported. This system creates inefficiencies and taxes have not generated any significant amount high-grading, with even medium quality fully of revenue because the Ministry of Trade and utilizable logs left behind in forests (World Industry's delays in approving export licenses Bank/Sida/GOF, 2001). effectively acts as a non-tariff trade barrier (World Bank, 2000e). In fact Indonesia appears to have Recently, changes have been introduced to recently re-instituted the log export ban. Thus, the regulatory system that may increase revenues lack of good governance and consequent revenue and provide some flexibility. A recent Prime lakof gcontinue to be challenges in the forestry Ministerial Order allows provincial authorities to losses in to establish contract prices in negotiations with the buyer. Thus, a minimum royalty is fixed at the Lao PDR federal level and the provinces have an Forests cover nearly half of Lao PDR. Of opportunity to fix a higher royalty and reap the this, nearly 2.5 million ha are officially designated additional benefits. This system is yet to be tested as production forests. Production forestry is and ils worth careflly watching as it gets considered one of the country's few potential implemented. sources of sustainable economic growth, and Conclusions poverty reduction. In fact, forestry is estimated to account for 5 percent of GDP. However, like The main lesson to draw from the case elsewhere, Laos forest resources are threatened by studies and the information on potential rents unplanned and undisciplined exploitation (World from forests is that the forestry sector offers Bank/Sida/GOF, 2001). Annual deforestation, due significant opportunities for resource-constrained to shifting cultivation, encroachment, fire, and governments to raise domestic revenues. unsustainable logging, is estimated to be between Currently, a large proportion of forestry rents is being captured by the private sector. Rent 42 Generating Public Sector Resources to Finance Sustainable Development Table 3.3: Magnitude of illegal forest activities in the Asia Pacific region Country Assessment Malaysia One third or more of forest exports was illegal in the early 1990s. Forest products exports to Japan were underdeclared by 40% in the early 1990s. Cambodia Households and enterprises perceived corruption as the leading problem for citizens and enterprises. Illegal exports were coming through the Thai border. In 1997, a minimum of US$184 million worth of timber was felled in Cambodia, much of it received by corrupt officials. In 1997, only 10% of logging was legal. Cambodia, Substantial illegal logging trade exists between the two countries. Vietnam Indonesia 95 percent of exports were illegal in the early 1990s. Malaysian companies were accused of logging illegally and smuggling logs to Malaysia. In the mid 1990s, 84% of timber concession holders failed to obey the law. Losses due to illegal logging was estimated to be US$3.5 billion per year in the mid 1990s. 84% of timber concessionaries did not follow the law. Illegal logging was taking place in national parks. As much as 40% of pulp and paper wood supplies came from undocumented sources. Myannar In 1995, some 276,000 m3 valued as US$86 million, or alnost half of the country's forest exports were undeclared. Papua New Foreign companies bribed politicians and leaders, illegally logged and exported timber, and Guinea transferred funds abroad illegally. Senior officials obtained logging rights in exchange for bribes. Fraudulent activities in the forestry sector led to an estimated loss in national income equivalent to the annual aid the country received from Australia: more than US$ 180 million in 1998. In 1994, Forester Minister Tim Neville estimated that US$1 million was lost daily to illegal practices. Foreign corporations gave bribes to local leader, national ministers, members of the Parliament, and to at least one secretary of the Department of Forests. Philippines 16 million ha of forest have shrunk to 70,000 mainly due to illegal logging. During the 1980s, the country lost US$1.8 billion a year due to illegal logging. A large ADB-financed Afforestation project reportedly was affected by widespread corruption, with money diverted to finance the agendas of local politicians. Russia The deputy head of the Federal Forestry Service, Dmitry Odinstov, recently complained about the rise in illegal operations, noting that there are "some timber procurement offices controlled by organized criminal groups engaged in illegal timber exports." Poachers cut down about 20% of the official harvest in Russia's Far East. About 20% of timber logged in Russia violates the law. As much as 50% of logging in the Primorsky and Khabarovsk regions may be illegal. There is substantial export trade from Siberia to China. Source Contreras-Hermosilla, 2001. collection would result in social re-distribution of lead to more efficient use of forest resources, rent between the state and the private sector and which may indirectly lead to reduced defo- free up resources for social and environmental restation. This is another reason for promoting investments. forest taxes. However, the theoretical results on Economic theory very broadly suggests that the links between rent capture and resource effi- increasing rent capture in the forestry sector could ciency are not definitive. Definitions of rent vary, and different taxes and competitive mechanisms Generating Public Sector Resources to Finance Sustainable Development 43 penalize certain conservation practices and reduction strategy that can be accomplished encourage others. Further, most "experts" in the through forest policy reform. field conclude that the specific economic and Data quality and availability will continue to institutional context of individual countries be key challenges in the process of making infor- determines whether and to what extent resource med policy decisions in most developing coun- taxation enhances resource use and harvesting tries. Such limitations, however, should not be efficiency. seen as excuses for not paying particular attention Low rent capture in developing countries is to the effects of forest policy reform, especially on closely tied to illegal harvesting and implicit and the poor who depend on forest resources for their explicit contractual agreements that benefit a few livelihood. It is also important that fiscal reforms interest groups. Illegal logging is a major problem in the forestry sector be accompanied with moni- for nearly all forest-rich countries in the toring of logging activity to ensure that revenue developing world. This is illustrated in Table 3.3, enhancing reforms also improve efficiency of which presents a summary assessment of illegal resource use. timber extraction in selected countries. In such situations, increasing taxes may provide a higher Bioprospecting incentive for illegal logging and may prove counterproductive. In the early and mid-1990s, biodiversity prospecting ('bioprospecting') was expected to Good governance can make a huge difference provide an important new source of financing for to sustainable development by increasing reve- forest conservation (Farsworth and Soejarto, nues available and by discouraging illegal forestry 1985; McAllister, 1991; Principe, 1989; Pearce activities. Recent evidence from Cameroon shows and Puroshothaman, 1992; Reid and others, that sound reforms that enhance transparency can 1993). It has not tumed out that way. While some result in significant increases in rent to the state. A pharmaceutical companies have proven willing to key message here is that forest policy reforms pay for access to samples of genetic material, the must take into account the willingness and ability sums involved have been much smaller than anti- of governments to implement these reforms. How- cipated (Laird and ten Kate, 2002). ever, it is important to note that this reform was an internal initiative, and not driven by outside Interest in evaluating the extent to which sources. In the latter case, success may be more bioprospecting might generate resources has led to difficult to achieve or may take longer, as the several efforts to quantify the value of biodi- Indonesia case study shows. It will be important versity. The methodology of these studies and the for countries such as Gabon and Laos, which are values they sought to estimate varied considerably in the process of adopting new forestry measures and, not surprisingly, produced a wide range of to take these lessons into account. estimates. Some early studies focused on esti- mates of the 'value of undiscovered drugs' and Finally, the Impact of forest policy regulation tended to find very large values. The metho- (including tax policy) on the availability of timber des used tpay involved Ti n the and non-timber forest products for poor rural prie ofad bythqaniy onsumto tai communities is another important consideration. prc of a rgb h uniycnueooti comunitieso is another ipoort oensidheatiy on. an average value per 'plant based' drug or un- A quarter of the world's poor depend heavily on tested plant species, estimating the patent value of forest resources for construction materials, fuel, asd or s ing the value of fodder,~~ ~ fod n eiie. Soeie 10to2 a drug, or using the value of a statistical life times fodder, food, and medicines. Sometimes 10 to 25 percent of income comes from non-timber forest the number of lives saved per drug or the avoided percnt f momecoms trm nn-tmDe Ioest illness costs (Farnsworth and Soejarto, 1985; products for low-income farming households Ritenbe 1989; Princ 19; callister, (Scherr and others, 2001). The development of 1991; Pearce and Puroshothaman, 1995). The markets for forest products where small poor pro- reslt oese sdudishvare fo US$28.4 duer my av acopeitveadanag results of these studies varied from US$28.4 ducer may ave acomptitiv advatage millilon to US$190 per year per untested species (because of their lower opportunity costs or lower ( r 1999 T e w e r e isden primi transportation costs) can be an important poverty assumptions made abte succes ratelo by assumptions made about the success rate of 44 Generating Public Sector Resources to Finance Sustainable Development discovering new drugs and how the value of a 18 biodiversity "hot spots." By relaxing the successful drug is determined (that is, drug sales, assumption in the Simpson, Sedjo, and Reid value of a statistical life, or patent value). model that bioprospecting firms conduct random The approaches followed by the above testing of leads, Rausser and Small (2000) use a studies have several problems. First, the value of a sequential search model that yields rents estimates drug should not be entirely attributed to the plant that vary from US$0 to US$9,177 per hectare for material it is derived from, especially when other the same areas. In their models, high probability substitute inputs are available (alternative to leads provide information rents, which are in plants or synthetic materials). Second, only a addition to the scarcity rents that depend on the handful of research leads will result in valuable total number of leads available for testing. How- discoveries, and then only after a considerable ever, the authors themselves acknowledge that the amount of time-usually measured in decades. information rents could be large enough to pro- Third, these early studies estimated average vide an incentive for conservation in only some values. Later studies have focused on the marginal special cases. Criticisms to the models developed value of genetic material. Finally, studies should in these papers point out that different drugs may carefully distinguish whether private or social be used to treat the same condition, such as the values are being estimated and which is the use of taxol in combination with other cancer appropriate value to answer the desired question. drugs that attack the cancer via different mecha- The private value to pharmaceutical companies, nism, and therefore one discovery does not reduce estimated through drug sales or patent values, is the value of remaining leads for the same condi- useful in determining whether these companies tion. Alternatively, more than one drug may be have incentives to conserve these plant resources. derived from the same lead, as is the case with The social value takes a broader view of benefits taxol and taxorete (Day-Rubistein and Fisvold, and addresses the question of whether conser- 2001). vation is justified from the viewpoint of society. Craft and Simpson (2001) reconsider the As most authors agree that the social value of question of the value of biodiversity in pharma- biodiversity is large enough to justify efforts to ceutical research when products can be diffe- preserve it (especially when considering the full rentiated to address the concerns discussed above. range of values, and not just the direct use value The two competition models developed attempt to provided by drugs derived from natural com- bridge the gap in the early literature, which in pounds), later studies have focused on estimating calculating the average value of species failed to the private value of bioprospecting to pharma- take into account competition between different ceutical companies. The focus of values estimated products derived from different sources, and the in later studies, whether private or social, also latter literature, which is unsatisfactory in treating changed from the value of the drug to the value of different products derived from different species a research lead or discovery per hectare of land as either perfect substitutes or being completely conserved. Their estimates, therefore, are not unrelated. While the models specified give diffe- directly comparable to those from these early rent estimates of social values, the authors find studies (Pearce, 1999). that the private incentives for biodiversity preser- The values estimated in the later vation are small in both cases. No numerical bioprospecting studies are also very sensitive to calculations or estimates are carried out, however. model assumptions. The benchmark study in this Most of the literature cited above focused on category is Simpson, Sedjo, and Reid's 1996 determining whether the rents from bio- paper, "Valuing Biodiversity for Use in prospecting are large enough, in principle, to Pharmaceutical Research." The authors find that induce resource conservation. Even if they are, the value of the 'marginal unit' of genetic resour- this is not by itself sufficient to ensure that ces is small due to uncertainty in the discovery resources are conserved. Equally or even more process and redundancy among leads. Their esti- important is that these rents are distributed in a mates of the scarcity rent of genetic resources way that induces local inhabitants to undertake range from US$0.20 to US$20.63 per hectare for conservation, because "if the rents do not accrue Generating Public Sector Resources to Finance Sustainable Development 45 to local land users who ultimately make (Crook, 2001). However, most industry analysts conservation or conversion decisions, the debate believe that royalties are in the one to three surrounding the size of bioprospecting rents is percent range of profits from successful irrelevant" (Barret and Lybbert, 2000). The long compounds developed. term potential of bioprospecting to provide Another successful bioprospecting venture incentives for conservation is also diminished due has been established by the International to the "quasi-option" nature of the value of biodi- Cooperative Biodiversity Groups (ICBG), which versity for pharmaceutical research. As know- was the outcome of a international workshop on ledge of species, or other potential scientific drug development, biodiversity conservation and progress in developing cures for diseases, such as economic growth held in 1991. The National combinatorial chemistry, evolves, the conserva- Institute of Health (NIH), National Science tion value will diminish. This will also happen if Foundation (NSF), and US Agency for Interna- the genetic material which provides the lead is tional Development (USAID) granted six awards then synthesized instead of being harvested from of approximately US$500,000 per year for five the source to produce the medicine. year research projects in Latin America, West The much publicized bioprospecting agree- Africa, and Southeast Asia. In 1995 USAID with- ment between the National Biodiversity Institute drew from the partnership, and the US Depart- of Costa Rica (INBio) and Merck Pharmaceutical ment of Agriculture (USDA) joined. In 1998, the Ltd. stands as one of the very few successful ICBG granted similar awards for a new five-year examples of bioprospecting's potential to generate cycle, some for continuing the earlier projects and funds for conservation (INBio, 2002). Announced others for new projects (ICBG, 2001). Funding in 1991, the agreement between INBio and Merck level in 1999 was US$3.7 million (Rosenthal and was the first of such kind. Merck paid INBio others, 1999>. Each ICBG is a consortium of aca- US$1 million for the right to evaluate the com- demic institutions, local and international private mercial potential of a limited sample of plant, voluntary organizations, and private pharma- animal, and microbial samples collected from ceutical companies. Six of the eight projects Costa Rica's 11 conservation areas. Merck also involve pharmaceutical partners, such as Bristol provided equipment and technical assistance for Meyers Squibb, Monsanto, and Glaxo Wellcome, processing samples at INBio. The agreement which make mostly contributions in kind to stipulated that 10 percent of research budgets and develop the technological capacity to developing 50 percent of any future royalties be awarded to countries to process samples. the Ministry of the Environment and Energy The ICBG program has three principal goals: (MINAE) for reinvestment in conservation. The first, to improve the human health through the agreement between INBio and Merck was discovery of new therapeutic agents to treat renewed twice and INBio has also signed similar diseases of importance to both developed and agreements with other American pharmaceutical developing countries; second, to conserve biodi- companies, such as Bristol Myers Squibb, versity through valuation of diverse biological Akkadix Corporation, and Givaudan Roure as organisms and the development of local capacity well as European companies. INBio reports that to manage these natural resources; and third, to since it started its activities in 1991, the bio- promote sustainable economic activity in less prospecting deals it has entered have generated developed countries by sharing the benefits of the more than US$2.5 million in direct financial drug discovery and conservation research pro- contributions to other divisions in INBio, the cesses (Rosenthal, 1997). Like the INBio/Merck conservation areas, MINAE, and national agreement, ICBG agreements include equipment, universities. This is a small sum compared to the training, and infrastructure for capacity deve- amounts generated in 1993 by the' forestry sector, lopment in the host country, in addition to royalty US$28 million, and the tourism industry US$421 and advance payments for access and collection of million. Royalty rates are kept in strict secrecy, samples. Each ICBG is also establishing or asso- partly so that INBio may be able to negotiate ciated with a trust fund for conservation. In more favorable terms in future agreements Suriname, for example, the ICBG established a 46 Generating Public Sector Resources to Finance Sustainable Development benefit sharing plan where the monetary benefits taxes are too low then the government is giving of advance payments and royalties are to away the patrimony and creating incentives for contribute to the "Forest Peoples Fund" and five over-exploitation of the resource. local governmental and non-governmental organi- In estimating potential resource rents, there- zations. The agreements also require that locally fore, an assumption as to risk sharing is required. important, but understudied, diseases and indi- We assume in what follows that potential rent genous therapies be investigated by commercial capture is 50 percent of the total available and scientific partners. The ICBG approach is not resource rent. without critics, who argue that financing pharma- Estimates of total resource rents are taken ceutical research and development is a round- from the breakdown of resource rents published in about way to encourage conservation (Simpson, the World Development Indicators (World Bank, 2001'. 2001d). There are two adjustments that need to be made to the WDI data, however. First, the WDI reports only rents on net forest depletion, where capture harvest exceeds growth in a given country. Se- As the preceding discussion has indicated, cond, the WDI includes rents on fuelwood as well when it comes to taxing resource revenues, the as commercial timber. The figures reported in public and private sectors generally share the risks Table 3.4 on potential resource rents are adjusted and uncertainties associated with resource extra- to reflect total rents on commercial timber. Coun- ction and harvest. If resource taxes are set at too try by country estimates of available resource high a level or are assessed on gross revenues then rents are presented in Appendix Table C.3. the private sector will view resource sector invest- The rent estimates presented below exclude ments as being too risky to undertake; if resource two key resources: fish and diamonds (which has Table 3.4: Estimated potential resource rents, 1999 Total Total Mineralfuel Mineral resource resource rents rents Timber rents rents rents ($ million) (S million) ($ million) ($ million) (% GDP) World 190,245 12,864 31,580 234,689 0.76 By income group Low income 19,734 1,715 2,357 23,805 2.30 Middle income 113,782 7,186 6,424 127,393 2.31 Lower middle income 57,896 2,371 3,130 63,397 2.43 Upper middle income 55,886 4,815 3,294 63,995 0.22 Low and middle income 133,516 8,901 8,781 151,198 2.31 High income 56,728 3,963 22,799 83,491 0.34 By region East Asia and Pacific 14,317 2,295 3,927 20,539 1.08 Europe and Central Asia 32,235 205 353 32,793 2.99 Latin America and Caribbean 26,808 4,443 2,393 33,644 1.64 Middle East and North Africa 49,743 346 53 50,142 8.17 South Asia 3,575 693 748 5,017 0.86 Sub-Saharan Africa 6,619 872 1,124 8,614 2.66 Notes: Assumes that potential rent capture is 50 percent of the total available resource rent. Regional aggregates include developing countries only. Source: Derived from data in World Bank, 2001 d, see Appendix Table C.3 for full country breakdown. Generating Public Sector Resources to Finance Sustainable Development 47 particular ramifications for countries such as Protected areas Botswana, Namibia, South Africa). Both are excluded for reasons of data availability. In the Natural habitats provide a variety of benefits. case of fish there are additional considerations- Many important benefits are indirect: regulation typically governments do not tax small-scale of hydrological flows, for example. Others are fishing and, in any event, the available rents have derived from direct use, such as recreation and been driven to zero in many fisheries owing to extraction. At present, many of these benefits are excess fishing effort. not priced, or are priced at very low levels. Better pricing for the use of these resources could Table 3. tells a interesing stor. First generate significant financial flows, and help potential resource rents globally are a small beter m ant resouces themselves his percentage of world product, 0.76 percent. The setion manage the resources themselves. Thfs bulk of these rents (81 percent) are from mineral fuels. Over 13 percent of total rents derive from to capture part of the benefit that recreational timber harvest, while over 5 percent derive from users gain from the use of protected areas. The minerals. Within the timber harvest figures, over potential flow of financial resources from such twice the value of rents accrue to softwood har- measures is not huge, but it could at the very least make a significant contribution to offsettig the vest in high income countries compared with the m 7 rents on hardwoods in low and middle income costs of conservation. countries. Visitors receive significant benefits from the Potential rents, principalyonmrecreational opportunities provided by protected Potentialhrents,kprincipallyson miner fuels, areas. In many cases, however, there is little or no makeoupnthies bmoulk to of ertimes the rents in middcharge for the enjoyment of these benefits. As a lowicmcountries.admon towover, aties tpercentse i result, protected areas themselves are chronically low income esounteriets However, as percentagelow short of funds for conservation and protection. Of GDP the resource rents exceed 2 percent in low Intuigusrfecolgnraeudsht and lower middle income countries, compared would hel fits bud gap. athough dat with low figures of 0.22 percent in upper middle scarce, the evidence suggests that there is scope to incomntries. cutisn0.4pcninhhiom substantially increase user fees in many protected areas. This would both generate revenue and The regional distribution of potential provide a valuable management tool. It is resource rents is also of some interest. South Asia important, however, to realize that user fees and East Asia and the Pacific exhibit the lowest cannot be simply raised arbitrarily and without ratio of total rents to GDP, with moderately higher limit. Visitors will resist excessive fees and go levels in Latin America and the Caribbean. elsewhere. Perhaps most important, the ability to Relative high ratios to GDP, over 2.6 percent, can charge high fees depends on offering higher levels be seen in Sub-Saharan Africa and Eastern Europe of services than are currently available-one and Central Asia. Finally, the Middle East and needs to spend money to make money. User fees, North Africa have potential rents of over 8 percent therefore, are not a blank check. of GDP owing to the high dependence on exports This section begins by reviewing available of crude oil in the region. information on user fees charged in protected areas in developing countries. These include not 3.2 Charging for Services only entrance fees but a growing array of ancillary fees. It then reviews the literature on visitors' Many services are currently being provided willingness to pay for the recreational benefits at little or no charge to their users. By charging provided by protected areas. A comparison shows for use of these services, developing country that visitors appear to be willing to pay governments could at the very least reduce the substantially more than they are currently budgetary burden they bear to provide them. We charged. It is difficult to estimate how much focus here on two such services: the recreational additional resources might be generated by raising services provided by protected areas, and the fees, as data on visitation rates to protected areas management of municipal solid waste. 48 Generating Public Sector Resources to Finance Sustainable Development Table 3.5: Tourism receipts (as percent GNP and exports, and per capita) Percent Percent Per capita Country GNP Country Exports Country (TS$) Maldives 99.1 Antigua and Barbuda 83.3 Antigua and Barbuda 4,062 Antigua and Barbuda 83.3 St. Lucia 63.4 Austria 1,877 St. Lucia 45.9 Maldives 57.4 Barbados 1,788 Seychelles 31.0 Barbados 56.3 St. Kitts and Nevis 1,595 Barbados 27.4 Dominican Republic 48.6 Malta 1,578 Jamaica 26.7 Seychelles 43.2 St. Lucia 1,333 St. Vincent and the 24.9 St. Vincent and the 41.5 Switzerland 1,109 Grenadines Grenadines Belize 24.4 Belize 38.0 Denmark 728 Malta 21.8 Grenada 38.0 Spain 567 Grenada 18.1 Jamaica 37.0 Belize 540 Gambia, The 15.3 Dominica 25.8 Maldives 496 Fiji 14.8 Egypt 25.2 Iceland 496 Vanuatu 14.3 Fiji 24.3 St. Vincent and the Grenadines Dominican Republic 13.9 Gambia, The 24.0 Ireland 463 Source: IFC/World Bank/MIGA, 2000. Figure 3.2: International tourist arrivals (1996-98 average) 100 - 80-. .- .- c~ 6 0 . ... : . 0 40 - . .... 0 **-- ,- 1.,_-~ Africa East Eastem Labn America Middle East South Asia Asia/Pacific Europe and Caribbean Source: World Tourism Organization data. is either scarce or non-existent in most countries. able as a source of foreign exchange. Although Some crude estimates are made based on interna- events such as the September 11 attacks as well as tional arrivals rates, to provide a very rough order local and regional problems can lead to sharp of magnitude. The potential for the use of user fluctuations in tourist numbers, the overall trend is fees as a management tool for protected areas is expected to continue to increase. Table 3.5 pre- discussed next. sents some basic information on tourist numbers, while Figure 3.2 shows the number of arrivals in Current use of user fees the different regions of the world. Tourism has been growing rapidly world- The World Tourism Organization predicts a wide. For many countries, it is already one of the substantial increase in international tourist arri- principal income sources, and particularly valu- vals, with the percentage of tourists taking part in Generating Public Sector Resources to Finance Sustainable Development 49 "ecotourism" being one of the fastest growing * Multi-tiered pricing. Most countries with segments of the industry (Figure 3.3).8 Nature well-developed nature-based tourism industries travel, which is a broader category than "eco- charge different fees to different categories of tourism", has been growing at a 10 to 30 percent visitors. Higher fees for foreign visitors have annual rate, according to several estimates been defended for several reasons, including (UNEP, 2001). The increasing environmental con- that foreign visitors do not pay taxes to support sciousness worldwide, nature films, and publica- the parks visited, do not bear the opportunity tions on specific destinations have been the most cost of preserving these natural resources, and important factors affecting the increasing demand that they demonstrate a higher willingness and of the nature based tourism sector. Visiting pro- ability to pay higher fees than national citizens. tected areas is often a major attraction for a majo- * Park-specific fees. Several countries also rity of leisure tourists. differentiate fees across parks, sometimes to Figure 3.3: International tourism by generate more revenue from popular parks, purpose of visit, 1998 sometimes in an attempt to steer visitors to less popular parks and even out the impact of 100% nature based tourism on these sensitive eco- systems. This may be the result of a national policy, such as the categorization of parks by 80°/ -the Kenya Wildlife Service or perhaps the re- Leisure, recreation, sult of a decentralized system where each park holiday (62%) management authority decides on its own fees. A noticeable exception in this group is Costa 60% Rica, which instituted a set of differential fees across parks in late 1995, but later adopted a uniform fee for all national parks. 40% * Additional fees. Another common pricing Visiting friends and strategy at several parks and protected areas is relatives, religion, to charge additional fees for vehicles, cameras, health, others (20%) and guided tours. In India, for example, the 20% _- actual cost of visiting a park may be two to d w Business, three times the cost of the entrance fee at the professional (18%) gate. Entrance fees do not include the cost of 0% activity related fees, such as a jungle walk or elephant ride, which can vary widely by park. Source: World Tourism Organization data. While most parks charge generally low fees, A survey of protected areas in the early there is one exception, at the popular Corbett 1990s suggests that only about half of the world's Tiger Reserve. The reserve is divided into five protected areas charged entrance fees (Giongo and zones, each with its own separate entrance fee. others, 1994). Table 3.6 shows the fees charged The most popular visited zone charges an by parks and other protected areas in a selection entrance fee of US$8, valid for three days. In of developing countries identified as popular eco- addition, there is a vehicle fee of US$3. In tourism destinations in a survey of travel industry Madagascar, visitors are required to hire a experts (Mercer and others, 1995; Weaver, 1999) guide to accompany them on their visits. and the availability of fee information, gathered Travelers' guides also note that while from various travel guides and country's national Uganda's national park entrance fees are lower park service or tourism promotion websites. than neighboring parks in Kenya and Tanzania, additional fees at Uganda's parks can make the As seen in the above table, countries practice total cost of a visit sometimes more expensive a variety of pricing strategies at national parks and than in similar parks in Kenya and Tanzania. other types of protected areas. 50 Generatng Public Sector Resources to Finance Sustainable Development Table 3.6: Park fees in selected developing countries (US dollars) Country Entrancefees charged Otherfees charged Belize $2-5 Costa Rica $6 for national parks; $8-12 for private reserves; $105 Guided tours at private reserves $ 12-15; Coco Coco Island Island also charges a $28 dive tax. Guatemala $5 in general Brazil $ 1.65-3.30 (Lindberg and Halpenny 2001) Ecuador $5-20 for most parks; $ 100 Galapagos Island India $0.35-4.65 for most parks; $8+ for Corbett Tiger Most parks also charge a vehicle fee and a Reserve (see conmments) camera fee, so actual cost of visit is 2 to 3 times the entrance fee Indonesia $0.30 to $0.50 Nepal $7 Royal Chitwan NP (largest and most visited park); $15 Sargomatha NP (base camp of Everest) Thailand $5.30 most national parks; $2.10 Thai Elephant Less visited parks charge between $0.25 and Conservation Center, at Doi Phukha NP $1.50 for admission and camping/hiking Kenya $15-27 depending on park category; $10 mountaineenng fee Tanzania $20-100 Uganda $15 Bwindi , Mgahinga, Murchinson Falls and Queen $250 gorilla permit fee at Bwindi and Mgahinga Elizabeth NP; $7 for all other protected areas in addition to entrance fee. Other additional fees apply (see discussion) Benin $8.12, Parc National de la Pendjari $1.62 camera fee Burkina $8.12, Parc National d'Arli $1.62 required tour guide fee Faso Cote $3.25, Canoe NP; $32.50, Tai NP Fee for Tai NP includes lodging and park guide d'Ivoire (required for visit) Gambia $2.63, Abuko Natural Reserve; Free, Kiang West NP Ghana $ 10, Mole NP and Kakum NP Additional $2-3 fee to hike specific trails Niger $6.50, Parc National du W Nigeria $1, Yankari NP $1 camera fee Senegal $3.25, Oiseaux du Djoudj NP, Alikolo-Koba NP and $8.12 vehicle fee Delta du Saloum NP Botswana $26 entrance fee for most national parks $6.50 campmg fee, $2.15-10.80 vehicle fee Nanmbia $4.91, EtoshaNP; $1.63-4.91, Namib-NaukluftNP $3.27 vehicle fee and $11.45-21.28 camping fee (fees vary by gates) South Africa $1.96-10.64 Swaziland $2.45-3.37 for most NP and nature reserves; $40.91 $1 car fee, $13 game drive fee at some parks annual entry permit Zimbabwe $2 25 to $20 Madagascar $7.95 (valid for 3 days) at National Parks; $7.95 Kaleta $3.98-12.73 required guided tour fee at NP. private reserve; $68-154 per person at Berenty private reserve (most visited reserve in the country) Notes. Entrance fees reported for individual foreign adults in most cases. Not enough information to determine if fees are differentiated between foreigners and nationals in the following countries: Guatemala, Brazil, India, Indonesia, all West African countries, Namibia, South Africa, and Zimbabwe. Sources. Central America: Lonely Planet; South America: Let's Go-Peru, Ecuador, Bolivia (2002 edition), Lindberg and Halpenny (2001) for Brazil ; Southeast Asia. Lonely Planet-Southeast Asia (October 2001 edition), Let's Go- Southeast Asia (2002 edition), Let's Go-India and Nepal (2002 edition); East Africa: Lonely Planet-East Africa (June 2000 edition), KWS website; West Africa: Lonely Planet-West Africa (April 1999 edition); Southern Africa: Let's Go- South Africa and Southern Africa (2002 edition); Madagascar: Lonely Planet-Madagascar (May 2001 edition). Generating Public Sector Resources to Finance Sustainable Development 51 * Indirect fees. Some countries use indirect The use of fees, however, has been a rather taxes rather than user fees to capture revenue underutilized tool (Laarman and Gregersen, from visitors. For example, the Turks and 1996). The concept that 'nature' is a free good has Caicos Islands have established a 1 percent often created the expectation of free access to hotel and meal value surcharge that is ear- national parks and protected areas. In the US, marked to finance all marine parks. The reve- national park fees were prohibited by law for nue generated by this tax was about US$0.5 many years. While very little information on million in 2000-less than what is needed for actual revenues raised from park fees is available the purpose (no entry fees are charged at MP). for developing countries, we notice that in The Cook Islands have a US$10 airport tax, 20 general, because these countries cannot afford to percent of which is earmarked for the Environ- subsidize park recreation, tourism has become a ment Protection Fund (Lindberg and Halpenny, large source of income for park agencies.9 The 2001). Belize's US$3.75 airport tax generates South African National Park System, for example, approximately US$500,000 a year, in addition recovers 80 percent of its budget costs from fees to user fees charged at national parks and ma- and tourism business it operates in parks (Eagles, rine parks, to support protected areas (Spergel, 2001). 2001). Potential for higher user fees Box 3.2: Using revenue from fees to improve park A review of the economics valuation management literature finds that foreign visitors are willing to In 1993 and 1995 the World Conservation pay (WTP) considerably higher amounts than the Monitoring Center (WCMC) surveyed over 600 pro- fees currently charged for visits at developing tected areas agencies throughout the world to obtain country natural areas (Lindberg and Aylward, data on their budgets and staff level. The differences 1999). For developing countries in particular, between developed and developing country statistics where most protected areas are viewed as 'paper are significant. The mean budget for protected areas parks' since they lack the funds to support a was US$2,059 per km2 in developed countries and management strategy, user fees can become an US$157 per km2 in developing countries. Priority coun- ma nt strce of income an tries identified by the study for increased financial assi- important source of income to fund conservation, stance, based on low budget inputs and high biological protection, and visitor impact activities. See box richness, included the countries in the Congo river above. basin region of Africa, the Indo-China Peninsula, and Table 3.7 shows available estimates of Meso-America. The extent to which user fees can be foreign visitor willingness to pay for visits to used to generate revenue for parks in these areas will likely depend on the amount of investment these coun- protected areas i developmg countries. tries can make to attract tourists. Even a casual comparison of these estimates If sufficient resources are available, parks can be an to those of fees currently being charged suggests effective way to preserve biodiversity. A recently study that there is considerable scope for increased use of 93 protected areas in 22 tropical countries found that of user fees. However, several caveats must be the most significant factors determining the effecti- bome in mind: veness of parks in protecting biodiversity are the Restricted sample. Almost all available esti- number of guards, demarcation of borders, and exi- mates of tourist WTP have been made in a few stence of compensation mechanism to the community. magship parkst a fe countrie such The assessment concludes that the amount of environ- flagship parks in a few flagship countries, such mental degradation in parks are much lower than in the as Costa Rica, Kenya, and Ecuador. Whether unregulated surrounding areas and therefore a modest these estimates are applicable to different increase in funding to these protected areas, to hire parks, and to parks in different countries is an more guards, establish management plans, and so on, open question. It is likely that not all parks are would increase the parks' effectiveness in protecting sufficiently attractive to sustain fees as high as biodiversity. some of those shown here. Parks able to charge Sources: Green and Paine, 1997; Bruner and others, very high fees such as the Galapagos 2001. (Ecuador), Coco Island (Costa Rica), and some East African safari parks are exceptions rather 52 Generating Public Sector Resources to Finance Sustainable Development than the rule. Location, uniqueness of attra- methods. Both are subject to important ctions, and other factors influence the demand methodological limitations (see Box 3.3). elasticity and therefore the revenue raising po- Moreover, the quality of their application has tential of fees charged. Most attractions within been uneven. a country would not command such high fees o Interpretation. Most studies have focused and likely have a limited potential to raise almost solely on valuation, and have not taken revenue for conservation. the further step of examining the impact of Methodology. These estimates are based on specific pricing mechanisms. They often two main methodologies: travel cost methods provide little guidance to park managers and (TCM), which derive estimates of WTP from national governments making decisions on observed behavior, and Contingent Valuation what fees, if any, to charge (Lindberg and methods (CVM), which are based on survey Aylward, 1999). Table 3.7: Estimates of foreign visitor willingness to pay for protected area visits (1999 US$) Site Mean WTP Survey question/design Source Monteverde Cloud $48.7 Entrance fee implied by dividing per capita Menkhaus and Forest, Costa Rica consumer surplus to travel to the country by Lober, 1996 the number of protected areas (TCM) Poas Volcano and $24.6 WTP higher fee for improved infrastructure and Shultz and Manuel Antonio $15 services at the site (CVM) others, 1998 NP, Costa Rica Irazu Vocano, Poas $23.3 Highest fee WTP to visit the site (CVM) Chase and Volcano, and $23.1 others, 1998 Manuel Antonio $26.6 NP, Costa Rica Lake Nakuru NP, $80-88 (TCM) Recreational value of flamingo viewing, per Navrud and Kenya visitor per day Mungatana, $134-140 (CVM) Highest increase in total trip cost WTP to still 1994 visit Lake Nakuru Beza Mahafaly $324-422 Regression of visitation rate on airfare cost Maille and Special Reserve, (TCM) Mendelsohn, Madagascar 1995 Mantadia NP, $30.5 (TCM) WTP for a 10 percent increase in quality at the Mercer and Madagascar park others, 1995 $72 (CVM) WTP to add a visit to the park Komodo NP, $11.70 WTP for access to the park (CVM) Walpole and Indonesia others, 2001 Montego Bay MP, $21.4 first time WTP entrance fee to preserve area in its current Dharmaratne Jamaica visitors state (CVM) and others, $10.7 repeat visitors 2000 Barbados NP $37.4 first time WTP entrance fee to create a NP to protect area Dharmaratne visitors in its current state. Fees currently charged for and others, $8.6 repeat visitors activities in the proposed park area (CVM) 2000 Notes NP = national park MP = marine park WTP = willingness to pay TCM = travel cost method CVM = contingent valuation method Source Adapted in part from Lindberg and Aylward, 1999, other sources as indicated. Generating Public Sector Resources to Finance Sustainable Development 53 Box 3.3: Measuring visitor willingness to pay The Travel Cost method (TCM) was first applied by the value of a visit to Lake Nakuru, and therefore infer economist Harold Hotelling, who suggested that the the recreational value of viewing the lake's main use value of a recreation site can be inferred from the attraction, the flamingos. It must also be noted that the travel costs, that is, transportation, food, lodging, and TCM has been developed as an indirect way to infer so on, that one incurs to visit the site. This approach the recreational value of a specific site in question and was later refined by taking into account that visitors to take the estimated value of an annual visit to one from different distances may incur different cost and site, as Menkhaus and Lober (1996) do, to infer an an inverse relationship between the number of visitors entrance fee for all parks in a country may be quite a and the travel cost could be used to estimate a demand stretch in the application of the methodology. curve. From the estimated demand curve, an indivi- While the TCM method uses revealed information, the dual's willingness to pay (WTP) for a visit to a recre- Contingent Valuation method (CVM) relies on ation site can be established. One of the advantages of stated information. The CVM uses visitor responses to the TCM is that it is based on observed market hypothetical scenarios about fee types and amounts to behavior. However, the survey methods used to elicit infer how much they value a visit to a specific this information from visitors to a site may not draw a recreations site. The values reported are contingent on representative sample of the population in question, the specific scenario described and in order to elicit Also problematic, in the context of international tou- meaningful responses from those surveyed the rism, is the fact that international travel fares do not scenario must be realistic and the survey question to necessarily vary in proportion with the distance from elicit the information must be carefully designed. Due the site visited and how to attribute travel expenses of to the hypothetical nature of CVM, the values elicited multi-purpose visits to one specific site visited are subject to a number of biases, which can be (Menkhaus and Lober, 1995). classified into three main types of biases: (1) The studies surveyed that apply the TCM have used incentives to misrepresent responses; (2) implied value different approaches to overcome these difficulties. cues, such as a starting point bias or range bias when Maille and Mendehlson (1993) used of a sample of using payment cards; and (3) scenario misspecification visitors from 11 different nations (so as to get some bias. The effects of those biases on survey responses variation in airfare cost, the only independent variable have been widely debated in the economics literature, in the regression equation estimated), but have been however, a recent expert panel has judged that by criticized for using a rather simplistic approach, based adhering to specific set of guidelines, these surveys only on a small sample to only one site to generalize can produce reliable starting points for valuation the value of all ecotourism to Madagascar. Other stu- purposes. Using both the TCM and CVM has also dies, such as that by Navrud and Mungatana (1994), been suggested, as a way to provide a check on the attempt to separate the value of a trip to Kenya from validity of the methods (Navrud and Mugatana, 1994). Potential for increased revenue Even when this information is collected, problems Poteneatiafor arise due to different procedures for collecting generation data at different entrances or across parks. Most Substantial anecdotal evidence suggests that parks do not have enough staff covering all nature-based tourism has been increasing, but "it entrances at all times of the day and all months of is not possible to comprehensively report on the the year, and very often shoulder season total volume of recreational use in recent years or information is poorly documented. The World its change over time," apart from individual Commission on Protected Area, a subgroup of the country reports (Eagles, 2001). Estimating the po- World Conservation Union, released the first tential revenues that might be generated with user guideline for measurement of public use of parks fees requires information on visitation to protected and protected areas in 1999.10 areas. Unfortunately such data either do not exist Potential revenue generation from user or are largely inaccessible. Most national tourism fees. In the absence of information on actual agencies do not keep statistics on the market visitation rates to protected areas, we carry out sectors, such as nature-based tourism or park- some admittedly very crude calculation to gain a based tourism. Most parks in developing countries sense of the order of magnitude of financial are poorly designed to document visitation levels, resources that might be generated. If we assume 54 Generating Public Sector Resources to Finance Sustainable Development 25 per cent of all international tourists visit at estimates of net new financial resources that least one park or protected area during their visit might be generated, but of the gross total financial and assume a nominal US$10 fee is charged, then resources that might be generated. the amount of revenue collected would be about Elasticities. In order to determine the US$600 million; a US$20 fee would generate impacts of increasing user fees at national parks about US$1.2 billion, assuming inelastic demand and protected areas, estimates of the price elas- for visits (see below). Greater numbers of visitors ticity for various types of recreational activities at would obviously result in proportionally greater these sites are needed. Previous research suggests revenues. These are clearly very crude estimates. that, in most cases, the demand for nature based In countries that draw a lot of visitors due to their tourism is price inelastic. One reason one can natural attractiveness, the percentage of arrivals expect demand to be inelastic is that fees are a corresponding to visitors to protected areas may relatively small component of the overall trip b)e much higher. A survey of visitors carried out price. The uniqueness of the site visited will also by the Costa Rican Tourism Institute reports that affect the elasticity-the more substitutes that are approximately two third of all visitors to Costa available, the more responsive visitors are likely Rica visit at least one national park (Ecotourism to be to price changes. Table 3.8 shows some esti- Society, 1998). Moreover, many visitors might mates of price elasticity of demand for visits to visit more than one protected area, and thus pay protected areas. user fees at several. As the studies cited in Table 3.7 show, some areas can probably charge fees User fees as visitor management tools substantially higher than US$10 or US$20. Under The increasing number of visitors to pro- rieasonable assumptions of number of visitors and tected areas and other sensitive ecological habitats reasonable fee levels, therefore, one can estimate have been a cause of concern due to the negative the potential revenue that might be generated as environmental impacts that may result. About half being between about US$1 billion and about of Manuel Antonio National Park, one of Costa US$3 billion. Note that where fees are already Rica's most popular national park, has been being charged, some proportion of this amount is closed to tourists in an attempt to protect the eco- already being captured. These estimates are not system that is home to one of the last surviving Table 3.8: Estimates of price elasticity of demand for visits to protected areas Location Elasticity estimates Costa Rica Chase and others (1998) Lindbergh and Aylward (1999) Poas Volcano -2.87 -0.051 Irazu Volcano -1.05 -0.296 Manuel Antonio -0.96 -0.238 Bostwana -1.35 Lodge users -0.68 Campsite users -0.93 Both Kenya (Lake Nakuru) -0.17 to -0.84 Foreigners -1.77 to -2.99 Residents Notes An elasticity of -2.0 would imply that a 10 percent increase in user fee would result would reduce the number of visitors by 20 percent, meaning that demand is very responsive to a price change and therefore total revenue would fall. Elasticity values closer to -1.0 would mean that the percentage increase in price is proportional to the percentage decrease in visitor numbers (so total revenues raised from fees would not be significantly affected), while an elasticity closer to zero would mean that demand is not very responsive to price and therefore increasing the fee would significantly increase revenues. The Chase and others elasticity estimates are based on the CV method, while the Lindberg and Aylward estimates are based on actual data. Generating Public Sector Resources to Finance Sustainable Development 55 populations of spider monkeys in that country. revenues generated by fees will give higher prio- Visitors' WTP for a visit to the park is several rity to budgetary considerations. times the current US$6 fee charged. Economic In other cases, higher fees have not decreased theory would suggest an increase in the entrance the number of visitors. For example, the fee as one mechanism to control visitation levels. Galapagos Islands have substantially increased However, as will be discussed in the case study fees to foreign visitors since the 1990s and have section below, political opposition to fee increases continually received higher number of foreign often undermine a government's ability to take visitors." It has been suggested that visitation advantage of this tool. There is also the possibility levels on the archipelago have reached critical that when confronted with difficult decisions tra- levels. The unwillingness to further increase fees ding off additional revenue for reduced visitation may be in part due to the significance of tourism levels for ecosystem protection (that is, when sector, which employs approximately 80 percent demand is elastic), park management agencies or of the people living on the islands and generates other government agencies that depend on the 60 percent of all tourism revenues earned by the Table 3.9: Major environmental impacts of tourism Social and cultural Pressures on other local pressures relating to Pressures on natural resources (energy, food, conservationlsustainable resources raw materials, etc) Pollution and waste use of biodiversity Land and landscape: Harm to wildlife habitat Land: solid waste and Social and cultural sand mining, beach and and habitats, with litter (tourists produce on impacts: disturbance of sand dune erosion, soil associated biodiversity average around lkg of local way of life and social erosion, urbanization, loss. waste a day). structures, changes in road and airport Biological resources: Freshwater: pollution by traditional practices that building, resulting land disruption of wildlife sewage. contribute to conservation degradation, loss of habitats, clearance of Mrnwaesnd (including conservation of wildlife habitat, vegetation for tourism coastal areas: sediment dversimpac landscape deterioration, development, increased run-off, pollution from Adverse impacts on Marine resources: pressure on endangered land-based hotels and livelihoods: lack of benefit recreational impact species due to trade and marinas, wasted and sharing with those who (scuba diving, hunting, increased litter associated with bear both human and snorkeling, sport demand for fuelwood, marine sports and cruises environmental costs. fishing), damage to coral forest fires. (in 1995, it was Resource use conflict: reefs and subsequent Ecologically fragile estimated that cruise competition between impacts on coastal areas such as ships in the Caribbean tourism and local protection and fisheries. rainforests, wetlands, alone produced over populations for limited Atmosphere: high mangroves, coral reefs, 70,000 tonnes of waste water, sanitation and energy use by tourism sea grass beds: if not per year). energy resources, facilities and properly planned and Air: at local level, air competition with traditional transportation. managed, nature tourism pollution from land uses (especially in Freshwater: overuse of threatens the world's transporting tourists; heavily used areas such as critical water resources most ecologically fragile global impacts, coastal zones). by hotels, swimming areas including parks especially from carbon pools and golf courses; and natural World dioxide emissions, this is of particular Heritage sites. relating to energy use in concern in regions such transportation, air- as the Mediterranean conditioning and heating (where water is scarce of tourist facilities, etc. and tourists consume . Noise: from ground and over 200 liters a day). air transportation. Source: UNEP, 2001. 56 Generating Public Sector Resources to Finance Sustainable Development Ecuadorian government (Honey, 1999). Critics ment of Costa Rica will reportedly spend US$0.8 have pointed to the increased dependence on the million to improve facilities at the Liberia airport, tourism economy, which has slowly replaced in order to attract more scheduled flights to the traditional productive activities of the island northwestern region and relieve the main airport inhabitants, as well as generated increased in the capital city. The privately run San Juanita migration and construction in the region (Parra- International Airport in San Jose will receive Bozano, 2001). Therefore the use of entrance fees US$120 million in loans from the World Bank for visitor management is somewhat constrained and other international banks to increase the num- by other political and economic concerns, that ber of gates from 6 to 16 and continue improve- may not always have conservation and protection ments at the facility-a much needed investment as its primary objectives. Table 3.9 summarizes for a country of 3.7 million that wishes to accom- some of the environmental impacts of tourism modate I million visitors a year. Investments will often discussed in the literature. also be made to triple the size of Manuel Antonio, Case studies one of the most popular parks with about 150,000 visitors a year. The additional area will allow for Three case studies, based on countries for more dispersion of visitors along the park's pris- which better data are available, serve to illustrate tine shoreline and along its overused trails. some of the potential and some of the difficulties In the 1990s, international donations of funds involved in increasing user fees for protected to protected areas decreased. Faced with a finan- areas. It should be stressed that all three of these cial emergency that required a reduction in the case studies are based on countries with well- budget allocated to park services, Costa Rica developed nature-based tourism industries. As began experimenting with increases in entrance such, both the numbers of visitors involved and fees to national parks. While park fees changed the fees that they are able to charge will be only slightly in the early 1990s, in September substantially higher than in most other countries. 1994 Costa Rica dramatically increased its fees to Moreover, all three of these countries devoted all national parks from US$1.57 to US$15 for substantial amounts of resources to achieving their foreigners. Revenues nearly tripled, but foreign current status, both in developing tourism infra- visitor numbers fell by approximately one third. structure and in marketing themselves to potential Political opposition to the new fees was fierce, visitors. Other countries attempting to follow in particularly from the tourism industry. Despite the their footsteps will not be able to do so overnight, substantial increase in the amount of revenues nor for free. generated, Costa Rica changed its fees again in Costa Rica 1995, creating a complicated system of differen- tiated fees. A ticket at the gate for foreigners Costa Rica has established itself as Central remained at US$15 for all national parks, but America's 'special jewel', a stable democracy tickets purchased in advance cost only US$10 for with a reputation for its enlightened conservation the most popular parks and US$7 and US$5 for policies. National parks cover 12 percent of the parks with medium and low visitation rates. A land area, and forest reserves and Indian reser- pass, allowing visits to up to four national parks vations bring the total amount of land protected to was also available for US$29. Tour operators paid 27 percent. The country's natural attractions in- only US$5 for bulk ticket purchases. Political clude active volcanoes and tropical forests (cloud, opposition remained strong, especially as a black dry, and moist), which provide habitat for a vari- market for tickets developed. In addition, the ety of wildlife, including 850 bird species, many inability of the Park Services to handle advance types of monkeys, sloths, armadillos, jaguars, and purchases led Costa Rica to set a US$6 entrance tapis. The high density of biodiversity attracts fee for foreigners at all national parks in April of nature lovers from all over the world. 1996. Fees for residents remained at about US$1. The booming tourism industry, however, Foreign visitation rates have increased by an ave- requires spending on infrastructure and to main- rage of 10 percent every year after the fees were tain quality at the most visited sites. The govern- lowered to US$6. Visitation by nationals has also Generating Public Sector Resources to Finance Sustainable Development 57 Figure 3.4: Visitors and revenue in Costa Rica's national parks Revenue Visitors (US$ million) 2,000,000 - 30 1,800,000 - ----| ~ Foreigners - .- 2.5 1,600,000 - ----------- = Nationals -- . 2.0 1,400,000 - ----------- -|-Revenue - -1 5 1,200,000 -- - - 1.0 1,000,000 ----- 0.5 800,000- - - - * . - 0.0 600,000 - - -0 5 400,000 -10 200,000 ~ t.-1 5 0 ' "-2.0 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 Sources- Visitor numbers from Eagles (2001), revenue calculated by the authors. been steadily increasing, although at a declining Competition from other travel destinations may rate, over the past five years.'2 act as a constraint on increasing fees to the levels Costa Rica's three most popular protected suggested by valuation studies. areas, Poas Volcano, Manuel Antonio, and Irazu Belize Volcano, have been the subject of several valu- ation studies (see Tables 3.7 and 3.8 above), Tourism is one of the most important sectors allowing the consequences of fee increases on of Belize's economy, accounting for 18 percent of revenues and visitation rates to be examined more GDP and 25 percent of foreign exchange earnings closely. In 1996, these three parks accounted for in 1999. It is estimated that tourism provides one almost 60 percent of all visits to protected areas in four jobs (Ceballos-Lascurain, 2001). Belize's (Weaver, 1999). Using the elasticity estimates marine parks have been the country's major obtained from actual visitation data, Lindberg and tourist attraction. The Meso-American Caribbean Aylward estimate that an increase in the entrance Reef, the fourth largest reef system in the world fee from US$5 to US$10 would increase revenue and home to a broad range of marine life, has at these three parks from US$0.9 million to some of the best diving in the world. Blue Hole, US$1.6 million, or slightly less than doubling. If Half Moon Caye, and Turneffe Islands are the fees increased to US$15 (still below the values only three coral atolls in the western hemisphere. suggested by WTP studies), revenues would Although travel inland is difficult because the increase to US$2.3 million. Whether such fee country has only three paved roads, the dramatic increases are feasible, however, is another matter. Mayan ruins and 'ecolodges' in the dense jungle The previous attempt to increase fees to US$15 in also attract many visitors. While only 28 percent 1995 was met with resistance by the tourism of tourists visited a national park, 60 percent industry and foreign visitors alike. Unless visitors reported participating in snorkeling and 30 per see an improvement in the quality of a visit to cent in scuba diving (C. Brown, 2001). these sites to justify the higher fees, visitation The Conservation Division of the Forest rates may fall and the tourism industry that de- Department, which oversees most of Belize's pends on their visits may suffer substantial losses. national parks and protected areas, receives no '58 Generating Public Sector Resources to Finance Sustainable Development Table 3.10: Visitor numbers and fees for selected Belizean parks and protected areas Fee for Estimated foreign Visitors Revenuec visitors Park (US$) 1997 1998 1999 ('000 US$) Guanacaste NP' 2.55 2,582 2,567 2,788 5,322 Crooked Tree WSa 4.00 2,074 1,484 1,619 4,857 Cockscomb Basin WSa 5.00 3,488 4,078 3,603 13,511 Blue Hole NP' 4.00 5,017 7,098 6,162 18,486 Half Moon Caye NMa 5.00 n.a. 7,310 7,940 29,775 Belize Zoo and Tropical Education Center n.a.d 37,029 40,855 39,838 -- Mountain Pine Ridge n.a.d 32,262 17,896 25,835 -- Community Baboon Sanctuaryb n.a.d 4,931 4,676 4,011 -- Hol Chan Marine Reserve 2.50 41,380 38,737 37,954 71,164 Total 143,115 Notes: a Managed by Belize Audubon Society b Managed by an association of village landowners, earning revenue from renting rooms or selling meals to tourists. Assumes three quarters of all visitors are foreign visitors d There is a US$10 night tour fee at the Belize Zoo, as well as other activity fees (canoeing trips, lodging, etc) that raise revenues for the zoo, in addition to membership/donation fees. NA = Not applicable. Source: C. Brown, 2001, Belize Audubon Society, and travel guides; revenues from own calculations. governmental appropriation. It did receive Kenya US$118,000 to pay the salaries of four employees, some administrative costs, and some equipment Kenya is the heart of the African safari maintenance in 2000-2001. Parks managed by the country, with the most diverse collection of wild Conservation Division do no have management animals on the continent. The Masai Mara, one of plans, infrastructure, or onsite management, there- the most popular game parks in Kenya, is a 320 fore visitors may enter and leave as they please, km2 reserve where visitors can witness the even though in some protected areas public access migration of some 2 million wildebeest, zebras, is prohibited (C. Brown, 2001). Parks and pro- and gazelles in their pilgrimage for food and lected areas in Belize do receive financial support water. From the vast plains of the south, to the from the Protected Areas Conservation Trust bamboo forest of the Mount Elgon and Mount (PACT), but competition for funds is fierce and Kenya, and the beaches on the east coast of grants are limited to US$17,500. PACT also re- Mombasa, Kenya offers a variety of attractions for ceives 20 per cent of revenues earned by parks tourists. Kenya got an early start in the managed by other NGOs, such as the Belizean development of nature tourism and ecotourism, Audubon Society (BAS). BAS manages two na- relative to its neighbors, such as Tanzania and lional parks, two wildlife sanctuaries, two natural Uganda, and by 1987 tourism had become monuments, one nature reserve, and one private Kenya's number one foreign exchange earner. nature reserve. BAS is authorized by the govern- The 1990s, however, have not been a good inent of Belize to charge and collect entrance, decade for tourism in Kenya. The number of concession and other activities fees. The remain- tourists to East African fell during the Persian ing profits that are not allocated to PACT are Gulf War in the early nineties. Kenya also began spent on managing the protected areas and build- to face competition from Tanzania, Uganda, ing infrastructure and security for these areas. Zimbabwe, and South Africa. Kenya's share of Generating Public Sector Resources to Finance Sustainable Development 59 the African tourism market fell from 9.3 percent p. 319). However, problems developed because of in 1989 to 7.1 percent in 1993, and its receipts a lack of transparency on how much and where re- dropped from US$420 million to US$413 million venues generated were spent and corruption. As during the same period. Political unrest, crime, the number of tourists began to fall in the 1990s, and the bombing of the US embassy in the late the deterioration of tourism infrastructure due to a 1990s continued to hurt the tourism industry lack of investment became evident. Disputes over (Honey, 1999). The government's recent decision the distribution of revenues from park entrance to require visas from all incoming visitors, which fees and tourist operations escalated. The quality costs tourists US$50 in addition to the US$40 of tourists' experience declined, while tourists departure airport tax, has also had an impact on themselves were becoming more environmentally the number of tourists. Safari operators, the hotel aware and more discriminating in their choice of industry, and airline employees have all comp- destination (Honey, 1999). lained about the new law, that the difficulty in obtaining applications overseas, and slow-moving Summary lines at entry points. Potential for resource mobilization. Once Kenya has received a considerable amount of widely neglected, the potential for resource foreign aid in order to develop its tourism indus- generation from protected areas has been the sub- try, which among other things required financial ject of increasing attention in recent years. support for the creation and maintenance of natio- Throughout the world, user fees at protected areas nal parks and game reserves. Following a policy have been instituted and/or raised. At least some that wildlife must 'pay its way' the government of the potential for revenue generation from this set up experimental programs at the Masai Mara source is already being exploited, although studies Game Reserve and Amboseli National Park to indicates that scope remains for more. Never- address poaching, hunting, and cropping in the theless, several caveats must be borne in mind. private and communal lands around protected First, there is a need for realism on how high fees areas. With 60 to 75 percent of its wildlife living can be raised. Although visitor price elasticities outside national park boundaries, it was necessary are likely to be low, they are not zero. In particu- to elicit community participation in the manage- lar, they depend heavily on the availability of ment of these areas, as well as to provide compen- substitutes (both in-country, in other countries). sation for restrictions imposed on the use of Only the most unique protected areas are likely to resources belonging to local communities. support very high user fees.'4 It should also be T'he Maasai are pastoralists who have tradi- borne in mind that user fees are only part of the tionally grazed and watered their herds on the benefits provided by tourists to a nation's econo- lands surrounding the Maasai Mara and Amboseli. my. Increasing park revenue by higher user fees may cause shortfalls elsewhere, due to lower visi- "in 1961 theeragvernmet ofese Kenye rcogzed that tor numbers. Unfortunately, few studies on the net "it was imperative thatese peoe come tos effect of park pricing policies exist. Second, high wildlife as astive eno benef, w rthmpro user fees can only be sustained if: (1) the protec- tecting es and established and reaed an a e ted area itself remains in good condition, which that these and a few other reserves would be ma reur sinfcn. xedtrso osra managed by the local district or county councils may require significant expenditures on conserva- managed bya the lcntral districnt orHcounty cou s tion activities; and (2) supporting infrastructure rather than the central government (Honey, 1999, (roads, hotels, and so on) exists to handle the visi- p. 309). The income derived from tourism by the tors. A si any s, ou have to spnd moey commnit wasvieed s th ke fator n rdu- tors. As in many things, you have to spend money cing poaching activity and contrary to the situa- to make money. These factors together suggest cioning poahing acstivityand,cont the situa-ofele s that user fees from protected areas are unlikely to tion in the rest of Kenya, the number of elephants provide a significant income source for any coun- try. They could, however, provide a great deal of Maasai put it, the agreement to share revenues with the local community acted as if "the national the mselves ts r tha brdenron park has gained two thousand extra pairs Of eyes ssestesle,tu euigta udno to help gaine two toandera profeyes the goverment's budget. For this to occur, how- ever, the revenue generated from user fees needs 60 Generating Public Sector Resources to Finance Sustainable Development to remain in the protected area system rather than historically been given little or no attention in being siphoned off to the central treasury, as is many developing countries, environmental quality usually the case. in many urban areas in developing countries has Direct incentive effects. The possible impact deteriorated to such a degree that it can no longer of higher user fees on visitation also creates the be ignored (Diaz, 2001). In cities such as Dar es possibility of using prices as tools to encourage or Salaam, mismanagement of solid waste is discourage visitation to particular areas. To date, increasingly becoming a public health threat and ihis potential remains under-utilized. User fees environmental burden (Mato, 1999). have seldom been used with the main purpose of This situation is driven, among other factors, visitor management. Restrictions on where visits by financing problems. Solid waste management are pernitted within a park and on the range of consumes substantial financial resources, activities permitted, and limits on the total number accounting for an important share of municipal of visitors or on visitation at particular times of expenditure across the developing world. At the the year remain the preferred approaches. These same time, cost recovery levels are very low. The approaches sacrifice considerable potential reve- main consequence of the resulting financial nue, however. constraints is the lack of coverage of the Poverty implications. A frequently cited collection system, which affects primarily to the objection to higher user fees for protected areas is poorer sectors of the population. In principle, cost that they would limit access by poorer sectors of recovery could be enhanced in several ways. society. As with many other such arguments, Although poor households may not be willing to however, the truth is that the bulk of the benefits pay the full cost of basic waste collection and are captured by better-off members of society, as disposal services-in contrast to services such as they are the ones who typically take advantage of drinking water-there is scope for improvement the recreational opportunities offered by protected in the generation of revenues to finance expanded areas. Moreover, two-tiered pricing systems or service. off-peak pricing schemes could be used to Municipal solid waste management in guarantee access to target groups. The other developing countries important poverty dimension of protected areas is that they were often created by fiat, and often cut Municipal Solid Waste Management off local communities from resources they had (MSWM) is a major responsibility of local depended on. Higher user fees will not change government. MSWM encompasses the functions this, but offer the potential to provide resources to of collection, transfer, treatment, recycling, compensate local communities for the resources resource recovery, and disposal of municipal they have lost. Again, this depends on the waste. Municipal solid waste is defined to include resources generated by user fees remaining in the refuse from households, non-hazardous solid protected area system and being used for this waste from industrial, commercial, and purpose. Such efforts would not only have an institutional establishments (including hospitals), equity dimension, but also a practical one: by market waste, yard waste, and street sweepings increasing the benefits that local communities (Schiubeler and others, 1996). derive from protected areas, it would reduce their Although it is essential to environmental incentive to encroach upon them. protection and public health management, solid waste management is highly unsatisfactory in Municipal solid waste most developing country cities. Although much of the solid waste generated is collected and Ineera cient collection methods, insficien disposed of through collected incineration or coverage of collection systems, and improper burial in sanitary landfills, substantial amounts disposal~ ~ of muiia.oi at r motn continue to be burned in the open or dumped threats to environmental quality and public health r ini developing countries. Although po . haphazardly. Such practices are putting an controlopmg envronental manageenpollution increasing pressure on land, air, and water quality, and posing threats to human health. These threats Generating Public Sector Resources to Finance Sustainable Development 61 will be exacerbated by projected increases in total for example, provide collection services in waste generation. In developing countries, muni- exchange for the opportunity to extract recyclable cipal solid waste is often disposed of with ash, materials and food waste for resale. In Manila, human waste, medical waste, and industrial waste. collection workers routinely take with them on For this reason, municipal solid waste in deve- their routes scavengers who pick out and sell loping countries is sometimes more dangerous to recyclable materials and share the proceeds with human and ecological health than it is in the collection workers. And scavengers sift industrialized countries (Beede and Bloom, 1995). through waste at transfer stations and final dump- Most MSWM systems have three basic sites. It is estimated that about 7,000 scavengers components: work at the MSW dumps in Manila, 8,000 in Jakarta, and 10,000 in Mexico City (Cointreau- * Collection and transport: gathering and Levine, 1994). Third, MSWM is often inefficient. removing MSW from its point of generation to In addition to the excessive labor force in public safeguard public health, limit congestion, and service providers, collection techniques are in- preclude unpleasant odors and aesthetically efficient. For example, when crews sweep up offensive sights. discarded food waste in the streets of the old * Processing: transforming the physical quarters of Moroccan cities residents, some of it characteristics of MSW by recycling, compos- spills back onto the streets; and in Shanghai, un- ting, burning, or compacting in order to reduce covered collection trucks also spill some of their the threat it poses to human health and ecosys- loads back onto the streets (Beede and Bloom, tems, improve its disposability, and possibly 1995). capture value from the waste. * Disposal: isolating and containing the residual Box 3.4: Regional differences in landfilling practices waste that is left after processing. In many African countries, much of the basic Some MSWM systems ignore or incompletely infrastructure for solid waste management has yet to be implement one or more of these key components. established. The majority of the African countries use Typically, developing country cities only collect open dumps to dispose of solid waste. While decision- 50 to 70 percent of MSW (Cointreau-Levine, makers in the region are aware that their countries have 1994). Landfill disposal usually involves discar- to upgrade open dumps to sanitary landfills, this is not ding the waste in open dumps (Bartone and regarded as a priority in most countries. At the national Bernstein, 1993) (see Box 3.4). This practice is and municipal levels, few countries have taken steps inadequate because sitting landfills in areas with a towards constructing, maintaining, or operating land- high water table or constructing them without clay liners may lead to the formation of leachate that Countries in the East Asia and Pacific region are can seep out of the landfillandpollactively investing in solid waste projects, thanks to can seep out of the landfill and pollute ground- partnerships with lenders and donors. Landfills in the water and surface water. To the extent that hazar- region are being upgraded from open dumps to sanitary dous waste is present in the MSW stream, landfills. While most capital cities in the region are ser- leachate could seriously contaminate the water viced with some level of landfill practice, the majority supply. of the waste in the region is still disposed in open MSWM in developing countries displays dumps. three other important characteristics. First, it is Source: Johannessen and Boyer, 1999. labor intensive, partly because labor is relatively cheap and capital relatively expensive, and partly There are limited economies of scale availa- due to over-hiring for political reasons. Second, ble in MSW collection. Although urbanization recycling is widespread. Labor-intensive collec- raises the concentration per square meter of tion and processing of recyclable materials are MSW, which may lower the average cost of found throughout the developing world. House- collection, it also tends to increase the cost of holds bring their recyclables to redemption MSWM because low-income urban areas often centers, and small-scale entrepreneurs purchase have narrow or congested streets that cannot recyclables door to door. The Zabbaleen in Cairo, support large collection trucks (Cointreau-Levine, 62 Generating Public Sector Resources to Finance Sustainable Development 1994). Disposal and treatment seem to display may clog open drains, creating breeding grounds greater economies of scale associated with tran- for malaria- and dengue-transmitting mosquitoes, sfer stations, landfills, and other MSW mana- or causing floods in rainy seasons, which may gement facilities. This suggest that collection ser- increase human contact with pathogen-infected vices are best provided on a decentralized basis, feces contained in the waste (Beede and Bloom, whereas it may be more cost-effective for disposal 1995). Losses in productivity due to a decline in and treatment facilities to be consolidated at a public health have reached alarming levels in regional or metropolitan area level (Bartone and some developing countries (Hemelaar, 1999). In Bernstein, 1993). addition, improperly vented dumps can cause The public health implications of inadequate explosions and fires. MSWM are important. Poor collection or disposal Solid waste problems are growing throughout practices attract and promote the breeding of the developing world (Beede and Bloom, 1995). insects, rodents, and pathogens that can cause and This is partly because of population growth and transmit diseases, particularly several of the partly because rising per capita incomes have led diseases in the tropical cluster: schistosomiasis, to rising per capita generation of waste. Urbani- South American trypanosomiasis, and Bancroftian zation combined with rising per capita income in filariasis (Beede and Bloom, 1995). The burden to many developing countries, have led to burgeon- developing countries from these diseases alone ing concentrations of solid waste in metropolitan was 8 million disability-adjusted life-years in areas. These trends drive up the average cost of 1990, or about two life-years per 1,000 people. An collection, processing, and disposal and strain the estimated 25 percent of these might have been administrative capacities of city governments. averted through "feasible interventions" (World Bank, 1993), such as covering the waste delivered Financal burden of MSWM in to a dumpsite with fifteen to thirty centimeters of developing countries soil at the end of each day. Although the direct Solid waste management represents a heavy contribution of inadequate MSWM to the burden financial burden for local governments in develo- of disease in developing countries is modest, the ping countries. MSWM expenditures account for indirect contribution is larger. For example, waste an estimated 0.4 percent of GDP (World Bank, Figure 3.5: Public expenditure in MSWM, 1994-95 1 0 0.8 * Vietnam EL z , 0.6 Bangladesh 0a *+ India * Colombia 0.4 * Philippines - Malaysia a Latvia * Estonia Hungary 0.2 Ghana * ~~~* Romania 0.0 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 GNP per capita ($) Source: Elaborated from data in World Bank, 1999c. Generating Public Sector Resources to Finance Sustainable Development 63 1999c). In most developing countries, the mana- vironmentally desirable level of solid waste mana- gement of municipal solid waste has traditionally gement, while general fund subsidies recognize been a primary responsibility of local govern- the public good and equity aspects of solid waste ments. In many medium-sized municipalities, management. MSWM can use between 20 and 50 percent of the Two factors complicate the task of estimating total municipal budget (Diaz, 2001; World Bank, the size of the implicit subsidies embedded in 1999c; Schubeler and others, 1996). MSWM services in developing countries. As in Municipalities in low and middle income other cases, the absence of detailed statistical data countries allocate the majority of their solid waste is an important constraint. In addition, MSWM management budget to collection and transport- has both private and a public (environmental and ation services. Collection and transfer make up 70 health externalities) good components. The public percent of the cost, of which 80 percent is labor good component justifies the existence of sub- costs (World Bank, 2001b). Final disposal costs sidies in MSWM, in the form of public sector are minimal because disposal is usually accom- financing of the service. Thus any discussion of plished through open dumping. subsidies in MSWM must refer exclusively to the Cost recovery levels from service users are private good component of MSWM services. very low. Analysis of the financial records of Given these constraints, some crude assumptions many developing country cities shows that current are needed to arrive at a ballpark figure. First, we practices for cost recovery for solid waste are very assume as first cut that the private and public weak. Recovery rates of less than 10 percent are good components of MSWM each represent 50 not uncommon (World Bank, 2001b). percent of the benefits provided by MSWM services. We also assume that 20 percent of Government cost-sharing arrangements municipal budgets is spent in MSWM and that 15 include matching grants from higher levels of percent of these costs are recovered. We further government and general fund subsidies from local assume that 15 percent of MSWM expenditure is governments. Matching rants are used to induce undertaken by the private sector (Hanrahan, pers. local governments to provide a socially and en- comm., 2002). Combining regional data on muni- Table 3.11: Implicit subsidies in municipal solid waste services, 1999 MSWMsubsidies Per urban Costs Total citizen Share of GDP Region (billion US$) (billion US$) M$) (%) Middle East and North Africaa 1.6 0.6 3.8 0.10 Sub-Saharan Africab 0.9 0.3 1.2 0.10 Asia Pacificc 44.3 15.5 12.6 0.70 Latin America and the Caribbean 8.0 2.8 7.2 0.14 Eastern Europe and Central Asiad 24.6 8.6 22.6 1.03 Total 79.4 27.8 11.4 0.47 Notes: a Excludes Iran, includes Sudan b Excludes Sudan c Includes Iran, excludes Mongolia, Lao, and Cambodia d Excludes Turkey, includes Slovenia, Mongolia, Lao, and Cambodia. Assumptions: (a) 20 percent of municipal budget is spent in MSWM, (b) 15 percent cost recovery rate, (c) 15 percent expenditure in MSWM is undertaken by the private sector, (d) MSWM is 50 percent a private good and 50 percent a public good. Source Authors' calculations, using data on urban population and municipal expenditure from UNCHS, 2001. 64 Generating Public Sector Resources to Finance Sustainable Development cipal expenditure and urban population from (electricity is best) so that the threat of cutoff is UNCHS (2001) with those assumptions, we esti- credible. For example, in Colombia and Eastern mate implicit subsidies embedded in MSWM Europe, some cities have a single utility bill that management in developing countries to be about covers multiple services such as water, sewerage, US$27 billion per year (see Table 3.11). telephone, electricity, and solid waste (Bartone, Improvin fin a spers. comm., 2001). There is little economic justi- fication for linking garbage services to other There are two ways of improving the finan- utilities, but it is a pragmatic solution. It can be cial sustainability of MSWM services: phasing- argued that electricity or water consumption is a out what we have termed "MSWM subsidies" by proxy for income and consumption leading to improving cost recovery, and reducing and con- waste generation, and hence a reasonable surro- trolling costs. We briefly analyze these measures, gate. In the case of large single point producers along their poverty, environmental, and incentive such as industrial or commercial enterprises, effects. volume or weight-based charges may be more Enhancing cost recovery. Any effort to appropriate. This has the advantage of linking enhance cost recovery encounters the obstacle of waste revenues to the actual volume of services low willingness to pay for improved MSWM provided (Schubeler and others, 1996). service. This may be either because households Finding successful models of cost recovery place little value on improved management or in solid waste management is difficult. However, because they do not believe that the supposed a few interesting cases suggest that there is room levels of service will be achieved (Beede and for greatly improving cost recovery. Private sector Bloom, 1995). The costs of MSWM can be reco- operations show that certain urban areas can vered either from user charges or general reve- generate revenue to cross-subsidize others, and nues. There are three types of user charges: that there are successful experiments in linking "garbage taxes", volume or weight based fees, and waste charges to other (more valued) utilities. tipping fees. The direct collection of garbage taxes e In Dhaka, Bangladesh, there are more than 250 is very expensive (so they are usually linked to small scale waste collection programs, each property taxes or to utilities), while volume or covering 200 to 1,000 families. They are weight based fees require a very sophisticated totally private and un-regulated and charge fees refuse collection system and lead to dumping of US$0.3 to US$1 per month per family. Most behavior. Other potential sources of revenues are: of them operate in high and middle income property taxes, business licenses, utility surchar- residential and commercial areas. More than 90 ges, and general fund subsidies (including trans- percent of users have been paying the fees. fers). These other methods do not comply with the Several of the programs have been operating polluter pays principle, and their incentive effects for more than 4 years (Ali, pers. comm., 2001). are limited. Choosing among them depends upon In Lusaka, Zambia, the private sector is regu- the relative importance of various criteria: lated through the Environment Council of whether revenues are adequate and easily coll- Zambia (ECZ), while the size and powers of ected, whether the polluter pays for the damage the conventional city council has been reduced. inflicted, whether the option is politically accep- The private sector only serves the commercial table, and whether payment of the revenue can be sector and some middle and high income resi- enforced (World Bank, 2001b). Table 3.12 exa- dents. The fees and profit margins are very miines the available options. high and un-controlled. Low income areas do In over 90 percent of cities worldwide where not receive any service at all (Ali, pers. comm., there is a garbage fee, it is collected with the 2001). property tax bill, usually as a separate item o In Ecuador, the central government allows (B3artone, pers. comm., 2001). Since property cities to attach a surcharge of 10-12 percent to taxes are so weak in developing countries, the cost electricity bills. This has allowed sufficient recovery rates are quite low. A better practice is to cost recovery to greatly improve the service collect garbage fees with another utility bill (Bartone, pers. comm., 2001). Generating Public Sector Resources to Finance Sustainable Development 65 Table 3.12: Options for enhancing MSWM cost recovery Adequate Politically Revenues? Easily Collected? Polluter Pays? Acceptable? Enforceable? User charges "Garbage tax" Can generate Often collected Only to extent Requires political Difficult to adequate with property that rate depends will to set and withhold service revenues if rates tax; Direct on surrogate for update rates for non-payment; are set based on collection is waste generation, Inefficient when costs, and are expensive (about like lot size, collected with updated as 10-13% of total property value property tax needed costs) Volume or As long as fees No, requires Yes No Leads to Weight Based are set based on sophisticated dumping Fees costs and updated refuse collection behavior without as needed system; local inspection and enforcement capacity Tipping Fees (for Yes, if based on Yes, if weigh- Yes, if fees are Municipalities Yes, but must use of disposal full costs of bridges are passed on by are often verify that trucks facilities) investment and utilized hauler to waste reluctant to pay go to disposal operation generator fair share site Other revenues Property Taxes Suffers from Yes No, except to Requires Evasion typical extent that political will to commonly weakness in property tax is collect property observed property tax surrogate for tax and update collection waste rates generation Business Yes Yes No Yes Yes Licenses Utility Yes Yes No, except to Not easy to get Yes Surcharges extent that utility energy use is cooperation; surrogate for often legally waste challenged generation General Fund Yes, but at Yes No Low political No guarantee Subsidies expense of other priority for that funds will (including needs disposal be allocated transfers) services Source: World Bank, 2001b. There are significant regional differences in Latin America several cities are experimenting the use of cost recovery instruments (see Box 3.5). with charges associated to utilities. Africa relies heavily on central government trans- In addition, limited evidence suggests that fers; East Asia/Pacific is trying deposit-refund there is room for improving cost recovery in dis- systems and volume-based fees; South Asia posal operations. Comparisons on tipping fees makes ample use of property taxes; Eastern (see Table 3.13) show that countries with similar Europe is turning to the private sector; and in levels of GNP per capita charge very different 66 Generating Public Sector Resources to Finance Sustainable Development Box 3.5: Regional differences in the use of cost recovery instruments In Africa, the central government generally finances ble for collection services in most cities. Municipal MSWM and other municipal activities through taxes managers are discussing the introduction of fees, de- collected by the Treasury. Although direct municipal centralized modes of payments, and higher specific charges are an emerging instrument for cost recovery charges in African cities, their use is not yet widespread. In Eastern European countries are finding it difficult to general, residents pay via indirect local taxes. User finance the centrally controlled model of MSWM fees are based on flat rates for collection services to service delivery and are turning increasingly to the households and commercial establishments. Examples private sector to finance MSWM activities, including of cities using municipal user fees include Accra, collection and disposal. In some cities, residents are Conakry, and some suburbs of cities in South Africa. paying fees directly to private collectors. Special taxes are also used to raise revenue for MSWM services. In Dakar an MSW collection tax is In most countries of Latin America, households pay imposed by the central government and is collected by for MSWM services through property taxes. Under the Treasury. In Bamako a cleaning tax is levied based this system, costs are seldom recovered, especially for on property value and direct user fees are also the poorer districts of the cities, because it requires the employed in some areas. municipality to have an updated register of Tbroughout the EasAsa/aciihouseholds. In some cities-mainly in Bolivia, Throughout the East Asia/Pacific region there is a Colombia, and Ecuador-charge fees through other range of methods by which costs are beig recovered utility bills (usually through the electricity bill, but via user fees. The deposit-refund system for many sometimes also through the water bill, as in Panama). recyclables is being expanded In cities like Bangkok, A number of studies show that this system is fair Singapore Tokyo, and Jakarta. Volume-based fees are because the volume of waste generated correlates also being tried out. reasonably highly with electricity and water In the Indian subcontinent, cities and towns have to consumption. This system is highly efficient and rely upon municipal taxes-mainly property taxes- generally finances most, if not all, solid waste for MSWM funds. The property tax base for municipal operations. Unfortunately, this system is not always services has had implications for equitable coverage: feasible, particularly in countries where utilities have illegal settlers, such as squatters, are not deemed eligi- been privatized. Source: UNEP, 1996. amounts. This happens both when comparison are rations exists, but it is limited. Some specialists made across regions or within the same region: point out the most effective way to ensure finan- Indonesia and Philippines (around US$1,100 GNP cial sustainability may be through cost reduction per capita) charged US$9.7 and US$1.3/ton; Peru (for instance, Schiubeler and others, 1996). There and Colombia (around US$2,200) charged 5 and are almost always opportunities to significantly 11 US$/ton; South Africa (around US$3,140) reduce the operational costs of MSWM services. charged US$12/ton while Malaysia ($4,300) char- Cost reduction measures include providing appro- ged US$1.2/ton. Furthermore, considerable diffe- priate levels of service, allowing for a greater role rences within countries can also be found. of the informal sector, and embracing public- Sometimes, the barrier to enhancing cost private partnerships. recovery is not one of affordability but of the cre- To ensure the long-term economic sustain- dibility of the providing agency. For instance, in ability of MSWM systems, investments in system B3angkok, a proposed raise in fees from 4 to 40 development should correspond to the level of Thai baht (US$0.14 to $1.4) per month encoun- resources which the society can make available tered bitter opposition, even though many resi- for waste management. Development aid has in dents pay 20 to 30 baht (US$0.7 to $1) per month many cases financed high-service capital inten- directly to the waste handlers to ensure removal, sive waste management systems-mirroring those (Blore and Nunan, 1997). of the donor countries-that are expensive to Reducing and controlling costs. The poten- maintain and are not adapted to the characteristics tial for increasing revenues from solid waste ope- and service needs of the recipients, the existence Generating Public Sector Resources to Finance Sustainable Development 67 Informal waste recovery and/or scavenging also Table 3.13: Tipping fees in selected contributes to cost savings by reducing the developing countries volume of waste which needs to be transferred and disposed (Schubeler and others, 1996). For Tipping Fees 1996 GNP per low-income countries, one option may be to Country ($/ton) capita ($) remove sanctions on informal sector collection Argentina 5-18 8,410 and recycling enterprises, integrate them with Chile 5-17 4,920 other MSWM strategies, and explore ways that Brazil 5-18 4,360 these enterprises can economically divert more municipal solid waste from landfills (Beede and Malaysia 1.2 4,300 Bloom, 1995). Mexico 4-17 3,640 In addition, important cost reductions may be South Africa 12 3,140 achieved by introducing competition through Peru 5 2,410 public-private partnerships for waste manage- Colombia 11 2,190 ment. Private enterprises are highly motivated to Philippines 9.7 1 190 lower costs and may introduce innovations and PhIlippinesi3 1,190 efficiency-raising measures to this end. At the Indonesia 1.3 1,090 most fundamental level, cost reduction implies a China 2.5 750 better utilization of available manpower and Source: Johannessen and Boyer, 1999. equipment, improved maintenance of equipment, introduction of appropriate technologies and the of livelihoods related to recycling, and the limited elimination of inefficient bureaucratic procedures financial resources of the municipalities. (Schiubeler and others, 1996). Some evidence Public waste collection costs may be reduced shows that public agencies bidding against private through the participation of residential commu- firms for contracts can be about as efficient as nities in local solid waste management (Schiubeler their competitors; the efficiency gains from and others, 1996). In most cases, this involves competition do not depend on the form of hiring of small scale enterprises or informal waste ownership of the collection service (Beede and collection workers directly by the households. Bloom, 1995). Table 3.14: Potential for revenue generation and the cost of expanding service Revenue generated Expenditure Expenditure by increasing cost Current level needed to needed to recovery rate to offormal expandformal expandformal 50% disposal disposal to 75% disposal to 90% Region (billion US$)' (0/6) (billion US$)b (billion US$)b Middle East and North Africac 0.7 44 0.9 2.1 Sub-Saharan Africad 0.3 31 0.8 1.2 Asia Pacific' 21.5 59 10.6 20.3 Latin America and the Caribbean 2.7 66 0.7 1.8 Total 25.2 13.0 25.4 Notes: Assuming zero elasticity of demand. b Assuming formal disposal is a linear function of expenditure. c Excludes Iran, includes Sudan. d Excludes Sudan. ' Includes Iran, excludes Mongolia, Laos, and Cambodia. Source: Authors' calculations. Current level of fornal disposal from UNCHS (2001). 68 Generating Public Sector Resources to Finance Sustainable Development Actual data on MSWM services and financial efforts have emphasized increased private sector expenditures is difficult to come by. In Table 3.14 participation for urban waste services, with full we report data on current levels of formal disposal cost recovery from the users. However, the city- from UNCHS (2001) as a proxy for overall wide cost recovery models may not be appropriate MSWM services coverage. Combining this figu- for all income groups due to a number of reasons res with our estimates on current levels of expen- (Ali, pers. comm., 2001), including: diture, allow us to calculate the cost involved in * The large-scale private sector is reluctant to reaching two service benchmarks. Comparing this serve the low income areas, unless the service cost with the potential revenue generated by is subsidized. increasing cost recovery rates to 50 percent in * While users of the service are non-homogenous currently serviced areas, illustrates that there is grouseoftheservicerenon-homogenous significant potential for expanding service finan- groups of people with different capabilities and significant potncrease infost rexp g es,ervicen fin assets, the service fee is often set on the basis ced by an increase in cost recovery rates, even if of an imputed 'ability to pay' as indicated by cost recovery in the newly serviced areas-where 'leitiae sessc spoet aus the poor live and where willingness to pay may be l egitimate' assets such as property values, thelpoower-fiveandss h erth5e rcillingnessnto paymaybincome, or plot size, and does not take into lower-falls short of that 50 percent. account human and social assets. Solid waste management, cost G The current working practices overlook that recovery, and the poor poor people could successfully undertake the There are two main issues linking MSWM management and operation of a significant and poverty: access and affordability. Lack of proportion of waste services in their own areas. finance for full waste collection, combined with The poor could offer useful recycling services political economy issues, mean that low income taat not only reduce the amount of waste but areas are often left without waste collection also provide a useful source of livelihood. services. In many developing countries, poorer * The formal models are generally weak in residents cannot afford the fees though they have satisfying customers through increased partici- a relatively high willingness to pay (Korfinacher, pation, consultation, improved image of the 1997). While the economic demand for waste provider, transparency, and accountability. collection services may cover primary collection Attention to these aspects could lead to impro- costs, it seldom covers full transfer, treatment and ved cost recovery. disposal costs, especially among low-income Efforts to improve cost recovery could help groups (Schubeler and others, 1996; Porter, 1993). the poor. The argument runs as follows. The poor Due to the environmental and health externalities suffer a lack of coverage of MSWM services. involved in the case of solid waste, the failure of Alongside political economy issues, lack of willingness-to-pay to reach cost recovery levels financial resources is the main reason for this lack does not imply that service should not be of service. Increasing cost recovery rates to 50 expanded. To achieve equitable service access, percent, in order to match the private good com- some degree of cross-subsidization and/or finan- ponent of MSWM services, would liberate sub- cing out of general revenues will be required. In stantial financial resources that could be used to many low-income residential areas, community- improve the service. Since the poor are the ones based solid waste management is the only feasible that suffer more the shortcomings of current solution (Schiubeler and others, 1996). At the MSWM practices, they could benefit from such a same time, systems for improving cost recovery change. by attaching waste collection charges to the bill- Conclusions ing of another service may be made progressive, in the sense that large users would pay a higher In practice, municipal government perfor- rate per volume of collected waste than small mance in the collection of waste service fees is users. often quite poor. People are reluctant to pay for The issue of privatization cuts across the municipal waste collection services which are per- iissues of access and affordability. Recent MSWM ceived to be unsatisfactory; at the same time, poor payment performance leads to a further deterio- Generating Public Sector Resources to Finance Sustainable Development 69 ration of service quality, and a vicious circle may achieving a cost recovery rate of 50 percent, to arise. With current urbanization and lifestyle cover for the private good component of MSWM trends, the business-as-usual scenario on waste services, would generate an increase in revenues management finances would be increasingly un- of US$25 billion in developing countries (exclu- sustainable. The situation, in short, is that a large ding countries in transition). Investing these finan- number of poor remain without the service, local cial resources on expanding coverage would boost authorities and private sector remain short of the level of service. expected cost recovery, and adverse environ- The potential for improving waste manage- mental and health consequences remain unabated. ment financing through cost recovery needs to be As Box 3.6 illustrates, those consequences may qualified. Social and institutional issues-such as reach disaster levels, and when that happens the low rates of property tax collection and distrust of poor are likely to be more affected. government-are limiting factors for improving cost recovery. Besides, the option of linking waste Box 3.6: The disastrous consequences of improper 'pricing' to utilities pricing will be hampered MSWM in Manila down the road by the privatization of the utility Manila, the capital of the Philippines, is home to companies. the infamous Smokey Mountain, a dumpsite right in the center of the city. Improper waste management at Poverty implications. The poor are willing this dumpsite has long affected air and water quality in to pay for MSW services, but this may not be the area. Available data showed that Smokey Mountain sufficient to cover costs. Hence, enforcing full was a major source of pollutants, such as chemical cost recovery on an individual basis may hurt the oxygen demand, chloride ions and lead. Ninety percent poor. Although some cross-subsidization may be of drinking water samples in the dumpsite revealed needed, it is in principle preferable to make bacterial contamination. After public protests, Smokey subsidies transparent. Full cost recovery for Mountain was closed and leveled in the mid-i 990s. better-off segments of the urban population, In July, 2000, an enormous wall of garbage combined with improved cost recovery for the collapsed in Manila's main dump in Payatas, crushing poorest, should have beneficial impacts on the a row of shacks and killing over a hundred people. The poor via better coverage. Current practices leave victims were impoverished squatters, including many the poor without service and, as a result, at greater children, who lived in the area and picked through the health risk. Improving cost recovery would gene- dump to collect items they can sell. The Payatas dump- rate funds that would allow to extend service. site was to be closed permanently in December 2000, but plans were postponed. Direct incentive effects. Reaching adequate These examples illustrate the tragic human and levels of cost recovery may require the use of environmental consequences of improper waste mana- effective instruments, such as associating a gar- gement. bage tax to the electricity bill. In this case, no Source: World Bank, 2000g. direct incentive effects can be expected. Setting tipping fees in line with costs would offer Potential for resource mobilization. User incentives to industrial and commercial clients to charges can generate substantial revenues and reduce waste generation. However, it would also provide incentives to minimize waste, especially increase the risk of open dumping. Experience in if structured so that those who pollute more, pay many countries has shown that charging the full more ('polluter pays principle'). Although user costs of disposal may create incentives for charges can be imposed at different stages of solid littering and open dumping, especially if the waste management (including collection and enforcement of regulatory standards (that is, no disposal), in many cities they do not cover the full dumping) is weak and entities can avoid paying costs of solid waste management activities. While the user charge by disposing of the waste citizens and enterprises are generally willing to themselves (World Bank, 2001b). Finally, imple- pay for solid waste to be collected, they are often menting a scheme where poor people undertake unwilling to pay the full cost of disposing of the tasks of management and operation would waste in a sanitary manner (Porter, 1993; World increase recycling, reducing the amount of waste Bank, 2001b). For current levels of service, to landfill. 70 Generating Public Sector Resources to Finance Sustainable Development 3.3 Imposing Green Levies mental levy, often referred to as a Pigouvian charge, is set at the level where marginal costs The 1980s and 1990s saw a rapid increase in from pollu-tion abatement equal marginal benefits the use of economic instruments to improve from abatement, thereby 'internalizing' the full environmental management. The United States social marginal costs.'5 If it is cheaper for a firm implemented tradable emission permits and to abate pollution than to pay the levy, it will do environmental charges, while European countries so. The theoretical expectation is that firms will instituted several environmental taxes. Environ- continue to abate pollution up to the point where it mental taxes in a broad sense (including energy is cheaper for them to pay the levy rather than taxes, transportation taxes, pollution charges, and bear the cost of abatement. At this point, overall fees) represent a relatively small, but growing, pollution will have declined to the socially opti- source of revenue in many OECD countries. The mal level at which any social benefits from an advances made on this front in developing coun- extra unit of pollution match costs. Box 3.7 sum- tries have not been as significant. Environmental marizes some of the taxes and charges that are taxes and charges or, more broadly, 'environ- broadly defined as environmental levies. mental levies' make the tantalizing promise of helping to generate revenue while simultaneously Box 3.7: Definition of environmentally related taxes addressing problems of resource and environ- and charges mental degradation. In this section, we discuss the Emission charges or pollution charges are fiscal and environmental implications of environ-- payments that are directly related to measures (or mental levies in developing and economies in estimates) of pollution, whether emitted into air, into transition. We identify case-studies and success water, or onto soil. Emission charges generally deal stories where taxes and charges have worked to with one type of emission at a time. For example, there improve the environment. We also attemptto pro- may be a tax per ton of sulfur dioxide (SO2) emitted into the air and a tax per unit of Biological Oxygen vide estimates of the revenue gains from institu- Demand (BOD, a measure of water pollution) emitted ting such levies. Different methods of recycling into the water. Thus, there may be more than one the revenues from these levies can have very emission charge related to any given activity. different efficiency consequences. In practice, User charges are payments related to the service however, revenues from environmental levies are delivered. Only households/individuals/industries often earmarked to particular spending programs. connected to the relevant public service are charged. Often, political feasibility may require some ear- The revenues raised are generally used to provide a marking and compensation for payers. service, such as the collection and public treatment of effluents or sewage. Product taxes may be levied to price Taxes and charges are seen as key instru- environmental effects correctly, and could be used to ments to decrease pollution because they can have correct externalities other than pollution. A product tax the same environmental impact as a regulatory may be levied on the units of harmful substance intervention but at lower cost to society. While contained in products: for instance a carbon tax is regulation can to some extent ensure basic com- based on the carbon content of each particular fossil pliance with minimum environmental and social fuel. The product (or excise) tax may also be levied per unit of the product, if the objective is to reduce usage requirements (although enforcement is usually far of the product generally. This may be in the form of from perfect), use of positive incentives can un- consumption taxes (for example, on disposable leash creativity and entrepreneurship which can products or non-fuel efficient cars) or production taxes lead to results far beyond the minimum. Also, (for example, on coal, diesel, or plastic). regulations are usually set based on what is A pure Pigouvian charge would create a viewed as feasible with today's technology where- strong incentive for individuals and firms to as incentives can stimulate the development of decrease pollution by choosing least cost options. tomorrow's technologies. In reality, implementing a Pigouvian charge is The theory underlying the use of environ- difficult since it would require full information on mental levies is well known. An optimal environ- social benefits and costs associated with changes Generating Public Sector Resources to Finance Sustainable Development 71 in pollution. For example, a Pigouvian charge on national governments, the majority of such carbon would entail taxing units of carbon revenues should be viewed as short-term revenues emissions and knowing how a unit increase in -since they will exist only until firms and carbon would affect social welfare and abatement individuals introduce pollution abatement techno- costs borne by industries. Such information is not logies. Revenues will decrease over time and can usually available. As a compromise, environmen- be sustained only at a lower level. Non-Pigouvian tal levies are generally of two types: (1) emissions charges or taxes targeted mainly at revenue gene- charges that allow an industry to meet a certain ration may have smaller incentive effects, but tax regulatory standard; or (2) product taxes, gene- revenues are likely to be sustained over a longer rally imposed on goods that produce pollution period of time (see Box 3.8). rather than on the pollutants themselves (for Green levies in practice example, on diesel rather than the sulfur content of diesel). Since the early 1990s, revenues from green taxes in all OECD countries have been increasing Box 3.8: Is there a "double-dividend"? slowly. In 1997, they accounted for slightly less In the 1980s a new literature emerged focusing on than 7 percent of total taxation and 2-3 percent of the revenue implications of environmental taxes and GDP on average (OECD, 1999).16 Energy and charges and hinted at the potential for getting a transport taxes accounted for more than 90 "double-dividend" from green taxes. In addition to percent of the total revenue from environmental reducing environmental problems, environmental taxes in most countries. In the Eastern European levies, it was suggested, could also increase overall economies in transition, our estimates based on economic efficiency if the revenues they generated data from the Regional Environental Center for were used to reduce rates on other, more distortionary Central and Easter Europe (REC 2001) suggest taxes (OECD, 2000d; Bosquet, 2000). While the that anuEs f ro pe (REs plus suggest double dividend hypothesis gained considerable ground that revenues from green taxes plus charges are in in the 1980s, a number of recent analytical papers have the range of US$7 billion. Amounting to about 8 cast doubt on its validity. The work of Bovenberg and percent of total tax revenues on average, green de Mooij (1994) and others (Parry, 1995; Oates, 1995; levies are clearly a significant source of tax reve- Goulder, 1995; Ligthart, 1998) has shown that this nue in these emerging economies. While less hypothesis ignores an important source of interaction information is available on the developing world, between environmental taxes and pre-existing taxes. environmental taxes appear to be still in their They argue that since environmental taxes raise the infancy in these countries. Even in China, with cost of producing output they tend to discourage labor supply and investment, and thereby exacerbate the one of the most extensive pollution charge sys- efficiency costs associated with tax distortions in labor tems, green tax revenues account for only about and capital markets. The costs from this interaction 1.1 percent of the total tax revenue.17 However, effect may dominate any efficiency benefits from these environmental levies do seem to contribute recycling environmental tax revenues in other tax critical resources for environmental purposes in reductions. Nonetheless, the possibility of revenue- several countries. neutral tax reforms has made environmental taxes A number of developing countries have politically feasible in many European countries. In introduced market-based instruments such as some countries-particularly in Finland and Sweden, intronmed taxes-and ion ch in and to a lesser extent in Germany-green tax reform environmental taxes and pollution charges in has been revenue negative: that is, increases in green recent years. However, there is no comprehensive levies have been accompanied by decreases in other survey (or database) of the use of these instru- taxes with the net effect being a decline in the overall ments comparable to those done for OECD (1999) tax burden to tax payers (Bosquet, 2000). and transition economies (REC 2001). Neverthe- The relationship between the revenue less, based on a review of the literature on implications of environmental levies and their environmental policy in Latin America and environmental impacts is not straight-forward and Caribbean, Asia, and Africa a few observations can often be at odds. Although a pure Pigouvian can be made. charge may have a strong impact on the The primary motivation for environment- environment and provide revenues to local and related taxes in most developing countries is 72 Generating Public Sector Resources to Finance Sustainable Development revenue generation. The basic environmental Taxes and charges in economies in policy framework is a regulatory one and environ- transition mental taxes are generally designed as comple- ments to regulation. Thus far, emission charges After the collapse of the Communist systems seem to dominate in most countries, with product the countries in Central and Eastem Europe (CEE) charges also fairly common. Deposit-refund initiated a transition from centrally-planned to schemes are beginning to operate in some high- market-oriented economies. As a part of the income countries. However, carbon taxation is process, the governments liberalized prices and almost non-existent. Some countries have also initiated the breakup and privatization of state- used innovative schemes such as public disclosure owned enterprises. The transition (among other to complement fiscal and regulatory policies. For things) led to a recognition of the broad-scale example, Indonesia and the Philippines classify environmental degradation that took place during factories based on their reported emissions and the Communist era. Most countries in the region broadcast the results widely (using simple rating sought ways to implement more efficient environ- systems), regulators are enabling communities to mental policies to remedy for the past degrada- identify serious polluters and pressure them to tion. Several countries introduced legislation to clean up . It is argued that such public disclosure impose environmental taxes/pollution charges on programs can curb pollution at modest cost and is polluters. CEE countries have also been largely better than explicit taxation (World Bank, 2000g). motivated by European Union accession policies In other cases, governments are also discovering and the need to adapt domestic laws and practices that working with the private sector (in creating to fit into the EU framework. the incentive structure) to manage pollution may Most CEE economies apply environmental be more cost-effective than a traditional command levies to a wide range of sectors, such as energy and control approach. In the absence of a compre- and transport, and many also have introduced hensive database, we review some regional pollution charges on air emissions, water efflu- examples of specific environmental levies here. ents, and waste disposal (see Table 3.15). Aside Table 3.15: Environmental taxes and charges in selected economies in transition (US$ million) Natural Resources Percent of Air (Mining and Tax Percent Energy Pollution Water Transport Conservation) Wastes Other Total Revenues of GDP Albania n.a. n.a. n.a. 22 n.a. I n.a. 23 n.a. 0.6 Bulgaria 489 0 1 19 n.a. n.a. n.a. 509 15 4.1 Croatia 449 n.a. n.a. 17 2 n.a. n.a. 469 6 2.3 Estonia 91 1 5 21 5 3 n.a. 125 9 2.4 Hungary 1,605 2 22 129 61 122 0 1,941 12 4.0 Latvia n.a. n.a. n.a. n.a. n.a. 1 n.a. 1 0.05 0.0 Lithuania 78 5 1 79 31 n.a. 11 204 8 1.9 Poland 2,487 n.a. n.a. 105 0 21 n.a. 2,612 6 1.7 Romania 57 0 5 n.a. n.a. n.a. n.a. 62 n.a. 0.2 Slovakia na 7 163 n.a. 17 122 n.a. 309 5 1.6 Slovenia 621 n.a. 18 n.a. n.a. n.a. n.a. 639 8 3.2 Total 5,878 15 213 392 116 269 11 6,894 Notes: n.a. not available. Source: SOFIA database 2001 / World Bank Database (SIMA) 2001. Generating Public Sector Resources to Finance Sustainable Development 73 Figure 3.6: Revenues from environmental levies in selected CEE countries 16 a) c 12 a) F x 0) .,.-- 0 - 1i. * ( = 0 c X = (j) C 02 o .> 0 w 0Ca ~ 0 0 Source: SOFIA data / World Bank SIMA database. from excise duty on fuels, a number of taxes tar- behavior. While carbon taxes or other environ- geted at fuel use and motoring are in place. The mental taxes on energy sources have been discu- pollution charge system has been motivated in ssed in many countries, only a few have so far large part by the need to raise revenues for implemented any systematic energy tax along environmental conservation projects. environmental lines due to inadequate admini- CEE governments obtained a total revenue of strative provisions. But this might change quickly approximately US$7 billion from environmental as countries prepare themselves for EU accession. taxes and charges in 1999. As Table 3.15 shows, Until recently, taxation of motor vehicles and the country which used environmental taxes most motor fuels-the second largest component of extensively was Poland, with over US$2.6 billion environmental taxes, accounting for about 6 per- in revenues. Hungary follows with a revenue of cent of total revenue also focused primarily on almost US$2 billion. These numbers amount to raising revenue. A few countries are now begin- about 2 percent of GDP in Poland and 4 percent in ning to implement coordinated fiscal and environ- Hungary. More importantly, these numbers equals mental policies, aimed at reducing motor vehicle 6 percent and 12 percent of tax revenues in Poland related pollution.18 and Hungary (see Figure 3.6). These are substan- Air and water pollution charges in CEE tial revenues for resource constrained govern- countries are relatively low and account for less ments, accounting on the average for about 2-3 than 4 percent of total environment-related tax percent of GDP. Latvia, Albania, and Romania are revenues. In general, a base charge rate is applied stragglers in this impressive effort to use environ- for permitted emissions (of air and water) and a mental taxes and charges. penalty rate applied if firms and industries are not Figure 3.7 shows that taxation of energy in compliance. In some countries polluters that products continues to be a major source of reve- invest in cleaner technology or pollution control nue in most countries, accounting for over 90 per- equipment have received subsidies in the form of cent of the total revenue from environment-related tax allowances. The low revenues from emission taxes-a level which is very comparable to OECD and effluent taxes are attributed to lack of enfor- numbers (Barde and Braathe, 2002). In general, cement and failure to adjust for inflation.'9 None- energy taxes have been levied primarily as part of theless, it is useful to note that with the exception general fiscal measures intended for revenue- of countries with advanced eco-tax programs, raising purposes, rather than as attempts to modify some tax rates in CEE are comparable with rates in Western Europe in real terms. 74 Generating Public Sector Resources to Finance Sustainable Development Figure 3.7: Tax revenues from different sources Air Pollution 0.2% Water 3.1% Transport /\5.7% Natural Resources (Mining and Conservation) Energy 1.7% 85.3% Wastes 3.9% Other 0.2% Source: SOFIA database 2001. A number of CEE countries are considering mental impacts can be attributed to a variety of levying explicit carbon taxes: that is, taxes which factors, such as limited information on marginal are related directly to the carbon content of fuels. abatement costs, low charge rates, lack of infla- Since there are essentially fixed coefficients tion adjustment, non-compliance, and lack of between CO2 emissions and carbon content, for monitoring capacity (Vincent and Farrow, 1997). example, the difficulties of monitoring CO2 However, assessing the environmental impacts of emissions can be avoided, at little loss, by taxing economic instruments in CEE is difficult, because directly the carbon content of the fuels that of the lack of baseline data and the number of generate CO2 when burned.20 Thus the tax would different factors that have contributed to environ- be specified as some specific amount per tonne of mental improvements, such as increased commu- coal, per barrel of oil, or per million cubic feet of nity pressure and donor assisted programs. Some gas, the amounts being such as to equalize implied specific instruments, such as air pollution charges taxes on carbon content.2' While carbon taxes are in Poland, do appear to have resulted in improve- gaining prominence in OECD countries, among ments and the same is expected of recently CEE countries (and even among developing coun- increased motor taxes in several countries in the tries in general) only Slovenia has imposed carbon region (REC/SIEI, 2001; Anderson and Fiedor, taxes with reasonable success (see Box 3.9). The 1997). In general, water pollution charges seem to tax raised US$83 million in 1999, representing an be more effective than air pollution charges be- estimated 0.4 percent of GDP, indicating its cause of higher charge rates and the greater ease potential as a source of revenue for developing of monitoring (Vincent and Farrow, 1997). countries. The equity and distributional impacts of Do environmental taxes and charges in CEE environmental levies are important issues, but adequately help the environment? Some analyses they have not been discussed much in the lite- suggests that the pollution charge systems in CEE rature. This is an issue of some concern, as many have often failed to induce investment in aba- of these taxes are likely to be regressive. A recent tement technologies, and that the effects on study indicates that CEE countries have increased emission levels have been rather small their motor fuels taxes substantially over the last (Soderholm, 1999; Jackson, 2000; Vincent and ten years, in order to be in line with EU. As a Farrow, 1997). As the Russian case illustrates (see result, CEE citizens currently pay some of the below), their inability to achieve desired environ- highest motor fuel taxes in Europe (REC/SIEI, Generating Public Sector Resources to Finance Sustainable Development 75 2001). Similarly, water and waste related charges fees, continental shelf, forest fees, water fees, and have also substantially increased and amount to land tax) accounted for over 98 percent of green almost 10 percent of household expenditures for taxes, and pollution charges slightly over 1 per- some households. The regressive nature of some cent. Within the federal budget, green taxes tota- of these taxes are best addressed through targeted led US$424 million, making up for 1.7 percent of policy packages that can help the poor. the total federal budget, with pollution charges accounting for about 4.4 percent of the total Box 3.9: Carbon taxes in Slovenia federal green tax revenues (see Table 3.16). At the beginning of 1997, a CO2 air emission tax At thefederal level, an economic strategy has was introduced with the adoption of an official decree. been adopted that intends "to increase the fiscal The decree stipulates that the tax is paid for the use of importance of taxes related to use of natural fuels and combustion of organic substances. The tax is resources and also of property taxes, which should calculated according to the content of carbon in various become the basic source of revenue for regional fuels: (1) 30 percent of the base rate for coals if they are used for heat and power generation with a total g ( i s utilization rate exceeding 55 percent; (2) 50 percent of the regional level, provinces such as Komi and the base rate for coals if they are used for electricity Samara have started differentiating some natural generation in thermal plants with a utilization rate resource user fees in order to reflect and capture a exceeding 33 percent; (3) 100 percent for the base rate greater share of the rent. We describe below the for co-incineration of combustible organic substances more specific case of the Russian pollution charge in combustion units and industrial furnaces and for system. incineration in waste incinerators; (4) 100 percent of the base rate for all other fuels. Calculation of the tax is Pollution charges in Russia have a more based on a sum of pollution units of a purchased recent history than resource fees, with the first quantity of fuels or pollution units of combusted charges being introduced in 1991. The current substances. The tax is paid by a seller (or importer) of system of pollution charges covers air and water solid, liquid; or gaseous fuels to end consumer. The tax pollution and waste disposal, relying on a mix of is added to the retail price paid by the consumer. quality-based and technology-based standards. Where combustible organic substances are used and Maximum and temporary emissions permits toge- co-incinerated in heating units and industrial furnace, ther determine the corresponding pollution char- or waste incinerators, the tax is paid by the manager of ges. The Russian system is characterized by a the heating unit, industrial furnace, or waste incine- gres Teriab n systeis chara cterz byia rator. Introduced at a rate of 2.2 Slovenian Tolars (SIT) great variablity i pollution charge rates, which per litre of petrol, 2.6 SIT/litre of diesel, and 3.1 range from less than US$0.01/tonne for within- SIT/litre of fuel oil, the rates were tripled in 1998. The limit emissions of carbon monoxide or disposal of current tax rate is equivalent to about US$15/ton CO2 non-toxic waste generated by the mining industry, and the tax raised US$82.6 million in 1999 repre- to some US$386,000/tonne of lead released in the senting an estimated 0.4 percent of GDP. air outside of temporary limits. In addition to Sources: Speck and others, 2000. water pollution charges, Russian municipalities also collect fees for water supply, sewage and Environmental taxation in Russia wastewater treatment. These fees are paid to local water utilities on a cost-recovery basis, and rates Enviromnental tax reform has yet to make vary extensively across regions or municipalities, significant inroads into Russian tax legislation and with firms traditionally cross-subsidizing house- practice. However, a few recent developments holds (Bosquet, 2001). signal a renewed interest for enhancing the fiscal roeo th eniomn an natua eore Although the current pollution charge system (Bosqeofthe 2001).iAo rding tod nathl Rssianes is motivated in large part by the need to raise .Bosquet, 2001). Accrevn ro t ae revenues, the system has been largely ineffective Ministry of Finance, revenue from green taxes in doing so. This is typically attributed to the lack (that is, pollution charges and natural resource of enforcement. It is also noted that taxes and fees) totaled US$1.8 billion in 1999, making up fines have not been adjusted for inflation despite a 3.7 percent of total revenues in the consolidated budget. Of this, natural resource fees (subsoil user rapidl wetakeing c cy.E en wit ind ation, the total Impact of the charges has been 76 Generating Public Sector Resources to Finance Sustainable Development Table 3.16: Environment and natural resources in Russia's 1999 budget Consolidated budget Federal budget Share Million Million Share Million Million (percent) Rubles US$ (percent) Rubles US$ Total revenues 100 1,197,454 48,284 100 611,710 24,666 Total "green taxes" 3.7 44,575 1,797 1.7 10,496 423 Subsoil user fees 2.5 30,077 1,213 1.2 7,190 290 Continental shelf 0.0 25 1 0.0 0 0 Forest fees 0.1 1,727 70 0.1 447 18 Water fees 0.1 1,035 42 0.03 198 8 Pollution charges 0.04 461 19 0.1 461 19 Land tax 0.9 10,939 441 0.4 2,189 88 Other user fees 0.03 312 13 0.0 10 0 Other revenues 96.2 1,152,878 46,487 98.3 601,214 24,242 Source: Ministry of Finance, Russia, in Bosquet, 2001. decreasing dramatically, because of high inflation Pollution charges in China rates and the lack of profits reported by enter- prises in the targeted sectors (Kozeltsev and China has had a system of pollution charges, M\4arkandya, 1997). In terms of environmental im- assessed as non-compliance fees, since 1979. pacts, pollution charges have not had the desired Charges are levied on both the amount and con- effects, as enterprises often failed to obtain techni- centration of emissions.22 The system is back- cal equipment and other physical resources stopped by a system of penalties to encourage needed for reducing emissions, which in turn compliance with standards.23 Thus far, some requires that capital markets function well. In 300,000 factories have paid for their emissions. In Russia these markets are not sufficiently develo- 1999, some US$600 million in revenues were ped or are too thin to provide financing of collected from environmental charges (see Table environmental projects. Moreover, the awareness 3.17). About 80 percent of these funds have been among firms of the various technological options used to finance pollution prevention and control, for pollution control and their associated costs is accounting for about 15 percent of total invest- generally very low (Soderholm, 1999). ment in these activities. A recent World Bank generally verlo(dh 199)study (Wang and Wheeler, 1999) suggests that The Asian experience with market-based although charges provide insufficient economic instruments incentives for compliance (since they are often too Environmental charges have traditionally low to induce abatement to the legally required been used in several Asian countries to raise level), they have proven highly potent in fighting government revenues. China and Malaysia, for pollution and cutting pollution intensity. For example, established a system of pollution levies example, each 1 percent increase in water pollu- over twenty years ago. More recently, there has tion levy has led to a 0.8 percent drop in the inten- been a more serious effort to fix charges such that sity of organic water pollution and each 1 percent they will actually help the environment. Four rise in the pollution levy has cut the pollution examples of environmental taxes from Asia are intensity of suspended particulates by about 0.4 discussed below. percent.). There is, however, considerable varia- tion in terms of the actual charges collected across the provinces.24 This is explained by two factors: the price a community places on pollution dama- ge, and the community's capacity to understand Generating Public Sector Resources to Finance Sustainable Development 77 Table 3.17: Pollution charge collection in China, 1992-99 (US$ million) Year 1992 1993 1994 1995 1996 1997 1998 1999 Total Pollution Charge 434.3 465.1 359.4 444.6 492.7 548.0 592.1 600.4 1. Above Standards 331.1 337.7 250.0 297.8 315.5 342.6 314.2 318.6 Water 215.2 213.1 153.4 180.1 186.5 198.1 197.7 200.5 Air 92.3 97.2 74.8 89.0 80.9 81.7 79.1 80.2 Solid Waste 5.6 6.4 3.7 5.7 4.5 6.0 5.3 5.4 Noise 16.1 20.7 18.1 22.8 25.7 29.4 31.9 32.3 Radiation 1.8 0.0 0.1 0.2 0.2 0.2 0.1 0.1 2. Wastewater Discharge 15.4 21.9 23.2 30.4 34.6 36.8 34.2 34.7 3. S02 emissions 17.6 27.3 62.2 63.1 4. Four pieces 87.8 105.5 86.1 116.4 142.5 168.6 181.5 184.1 Source: State Environmental Protection Agency (SEPA) and World Bank staff estimates in Wang and Wheeler, 1999 and act on local environmental problems. The res- forces (environmental user fees), public disclosure ponse to pollution levies so far suggests that as the of information, and engagement of the public levy rises, the pollution could be reduced far through community participation in environmental faster than anticipated (Wang and Wheeler, 1996). management. In terms of the sheer magnitude, the current A good example of an environmental levy in Chinese system may be without peer in the world. the Philippines is the user fee system first imple- However, it has been criticized on two principal mented in 1997 by the Laguna Lake Development grounds. First, critics have claimed that there are Authority (LLDA) (Nepomuceno, 2001). Its aim marked differences in the degree of enforcement was to curtail pollution of the Laguna lake and to across regions, making the current system seem provide incentives for factories to install cleaner somewhat arbitrary. Second, the incentive proper- technologies. While the Philippines has long ties of the system have been called into question.25 maintained a traditional regulatory system, this Using more recent data from 29 provinces, Wang was an innovative attempt to target and clean-up and Wheeler (1999), have countered this criticism, the Laguna bay which is the second largest and arguing that the stylized facts offer little support one of the most polluted water bodies in the coun- for the view that the levy system is arbitrary and try. The user fee system had two parts: polluters ineffective. While effective levies do vary greatly pay a small flat fee to cover the administrative in China, their geographic distribution is corre- costs, and an additional charge per unit of emi- lated with provincial rates of urbanization and ssions that meet the standards and a higher unit industrialization. Furthermore, recent increases in charge for emissions above the standard. Figure effective levies have been accompanied by large 3.8 shows the charges levied under the system. reductions in water pollution intensities and loads. The system now covers 659 industries or Environmental user fee system in establishments, including 100 percent of all facto- Philippines ries discharging waste water into Laguna Lake. Eventually, the system is expected to cover all The Philippines has recently introduced a polluting activities from industrial, commercial, mix of new environmental policies to ensure that domestic, and agricultural sources in the country its economic growth is environmentally sustain- (World Bank, 2000d). Twenty percent of the able. These instruments complement the tradition- revenues are shared with local government units al command and control system by using market in the program area. The system has led to 78 Generating Public Sector Resources to Finance Sustainable Development Figure 3.8: Charges from environmental user fees by LLDA, Philippines (US$ millions) 2 5 - E _ _ Barging, survey, and shoreland management fees 2.0 - . W Administrative fines - F | I Polluton 1.5 - E Discharge permit fees (environmental user fees) 1.0 - --. .. ... . --- W | Processing fees - F ^1 - , 'lindustnes (air) 0.5 --*-* . > - . | Fishpenlfishcage fees F _ _ [ and surcharges 0.0 1997 1998 1999 2000 Source: LLDA Database, Philippines. significant environmental improvements. Within 1977, the Department of the Environment decided the first two years of implementation, total BOD to combine the effluent charge system with dis- discharges from plants dropped by about 88 charge standards and the first discharge fees were percent. Revenues from the program have also collected in 1978.26 As the quantity of waste dis- provided more resources for water quality mana- charged increased, the standards became more gement programs and to strengthen the existing stringent and the discharge fee also increased.27 monitoring and enforcement capability of the The results have been very encouraging. Despite a LLDA. The government expects to build on the 50 percent increase in the number of palm oil experience of this first phase to mainstream the mills between 1978 and 1982 and a subsequent program throughout the country and across sectors increase in palm oil production, the total BOD and media. In particular, the Department of load released in public water bodies dropped Environment and Natural Resources hopes to use steadily from 22 tons per day in 1978 to 5 tons in the results in drafting the legislation required for 1984. Overall, the regulatory package was clearly applying user fees outside the LLDA area (World effective in reducing pollution and improving the Bank, 2000d). quality of Malaysia's rivers. The Malaysian effluent charge system The Malaysian system, however, is often faulted for being economically inefficient, be- The Malaysian effluent chargesystem, onesof cause its charges are not based on marginal the oldest in a developing country, was instituted environmental damage costs but rather on the cost by the Environmental Quality Act, which was of capital for pollution abatement. Also, the adopted in 1974. The Act included provisions for charges are based on BOD rather than on the using economic incentives and disincentives, in volume of waste discharged, thus providing an the form of effluent charges, in support of regula- incentive for some firms to dilute their effluent in tions on discharges. It required that all dischargers order to avoid paying the charge. In some cases pay a fee to obtain a license to discharge waste the surcharge for effluents is too low to act as a into public water bodies. The fee varied according deterrent.28 It has been argued that Malaysia might to: (1) the class of the premises; (2) the location of have reached the same goals more cheaply by the premises; (3) the quantity of wastes dischar- relying almost exclusively on pollution charges, ged; and (4) the existing level of pollution. In Generating Public Sector Resources to Finance Sustainable Development 79 since they would have allowed plant managers the after which it increases to 75 Rupees (US$1.20) freedom to minimize pollution-related costs (Khwaja, 2001). (World Bank, 2000d). However, despite its weak- Other financial incentives that are being nesses, the Malaysian mixed system of command- considered for inclusion in the program to meet and-control along with market-based instruments the national environmental quality standards in- provides valuable lessons for developing countries clude low import and sales tariffs, duty exem- that are planning to introduce market-based instru- ptions, and provision of accelerated depreciation ments to support environmental legislation. It also on pollution abatement equipment. The funds shows that efficient enforcement of emission stan- collected from these charges are likely to be set dards can get the job done if public institutions aside in provincial environmental trust funds that function well, as in Malaysia. may receive matching grants from the state, User fees and pollution charges in providing additional funds for environmental Pakistan objectives. Thus, Pakistan is on the verge of instituting a more comprehensive set of incentive- In Pakistan, environmental taxes and charges based fiscal instruments for environmental mana- have traditionally been designed to meet fiscal gement. goals or to recover costs of providing services. In recent years, growing awareness of environmental Market-based instruments in Latin issues and increasing pressures on natural re- America and the Caribbean sources-annual environmental losses were esti- Direct regulation traditionally has been the mated in the early 1990s to amount to 4 percent of most common approach to environmental prob- GDP (Brandon, 1995)-have highlighted the lems in Latin America and the Caribbean. usefulness of financial instruments in directing However, during the last decade the use of econo- environmental policy. mic instruments for environmental management In the late 1990s, taxes, charges, and fees has started to gain acceptance, especially in from environmentally linked sources such as countries with greater institutional development. water, forestry, and natural gas contributed bet- The use of fees for environmental services and ween 4 and 8 percent of annual revenues.29 Most contributions for public expenditure incurred in of these charges are at rather low levels and the the provision of such services are becoming more sectors they are associated with, such as irrigation, common. Colombia, for example, applies pollu- tend to be heavily subsidized. Irrigations charges tion and compensatory taxes to waste disposal and are the most substantive of all the environmental emissions; Brazil charges for water extraction and levies and accounted for approximately 38 percent for the disposal of industrial effluent; Chile char- of all user charges and 6 percent of total non-tax ges for waste disposal; Mexico charges fees for revenues in 1999-2000.3 In this sector, water fees wildlife utilization as well as fees for industrial and user charges are again driven by the need to waste disposal; Argentina levies taxes on the dis- raise revenues for meeting government expen- charge of residual waste water and Venezuela ditures. charges a fee by volume of industrial wastes More recently, there have been efforts to among others (ECLAC/UNDP, 2001). All of these manage pollution and air quality problems by approaches, in their fashion, attempt to internalize basing policies on the "polluter pays" principle. environmental costs. Table 3.18 presents exam- Pollution levies and fiscal incentives are seen as ples of the different types of some of the market- important components of a recently approved based instruments used in the region. program implementing national environmental A recent review by the World Bank (Huber quality standards. The program plan calls for and others, 1998) suggests that the primary role of pollution charges at an exponentially increasing the market-based instruments in the Latin rate over a number of years, starting with a low America and Caribbean region has been to raise base rate that then increases after two years. The revenue. Other potential objectives-such as proposed base rate is 50 Rupees (US$0.80) per reduction of environmental impacts or improving unit of pollution discharged for the first two years the cost-effectiveness of regulations-have been 80 Generating Public Sector Resources to Finance Sustainable Development Table 3.18: Market-based instruments in Latin America and Caribbean 0 0 > -6 .0 . 0 Instruments m c m u o i > 2 2 . F Credit subsidies X X X X X Tax/tariff relief X X X X X X X Deposit-refund schemes X X X X X X X X X X X Wastefeeand levies X X X X X X X X X X X Forestry taxation X X X Pollution charges X X X X Renewable resource taxes X X X Earmarked conventional tax levy X X x Tradable permits X X X Eco-labeling X X X X X Liability instruments X X x Source. Huber and others, 1998. underemphasized or not attained. In most cases, CORNARE, the pollution control authority in the institutional weaknesses such as under-funding, Oriente Antiqueno region has been a pioneer in in-experience, unclear jurisdiction, or lack of instituting this program. The successful collabo- political will, have often limited the effective ration between the agencies and the local business implementation of these instruments. Their and communities in arriving at the desired successful use requires adequate legislation and environmental goals has been a cornerstone of this financing, capable institutions, and effective program. monitoring and enforcement. Since its inception, the pollution tax has Colombia's pollution charge system become a source of very substantial revenues in the country. In the period 1997-2000, environ- Colombia's tax on water pollution is one of mental authorities collected 17.9 billion pesos the region's relatively successful environmental (US$15 million) for the national environmental programs. It has provided funding for environ- system through pollution taxes, over double the mental activities and generated adequate incen- total allocations of 8.7 billion pesos (US$6 tives for reducing water pollution. A salient fea- million) they received from the national budget ture of this program is that each region starts by (ECLAC/UNDP, 2001). The Ministry of the envi- setting its own pollution reduction goals, impo- ronment is currently working to establish regional sing the national base charges, and tracking total funds for cleaning up the environment, which will discharges for six months. The regional regulatory be financed with the proceeds of water clean-up agency then has complete flexibility in terms of projects the method it adopts for reducing pollution and its options for minimizing costs through less expen- Environmental taxes in Africa sive clean-up solutions. The tax is applied pro- At present, environmental taxes are little gressively over five years, starting with the mini- used to address environmental and resource mana- mum rate and then increasing every six months by gement problems in Africa, due to the limited pre-established amounts until agreed regional scope of the existing tax system and inadequate environmental quality goals are achieved, collection mechanisms. Although user charges of Generating Public Sector Resources to Finance Sustainable Development 81 some kind are being used in some countries, they degree of arbitrariness in the setting of rates and a are often too low to address either revenue issues general lack of uniformity and clarity. or environmental and resource management issues. Also, the base of the principle underlying Box 3.10: Current and potential environmental charges is not clear, resulting in charges set taxes in South Africa haphazardly and not being transparent. The * Aviation uel levy: 0.15 US cents per litre levy on charges are often kept low for equity purposes. aviation fuel, used to fund the Civil Aviation They are also seldom adjusted, so that their 'real' Authority. value is eroded (Arntzen, 2001). Nevertheless, * Electrical energy levy: Estimated at 0.1 USc/kWh here we offer some examples of prevalent over the next ten years (about 5 percent of the ave- environmental taxes and charges in Africa. rage bill) on the net electrical energy sold by licensed generators and importers at the wholesale South Africa. The most widely used en- level, used to fund a National Electrification Fund. vironmental taxes in South Africa are those, imposed on energy products. Among*existing and 9 Electricity-Regulatory levy: A dedicated levy of imposed on energy products. Among existing and 0.000986 USc/kWh on electricity sales by genera- possible future taxes, fees, levies and charges to tors, used to finance the regulatory services pro- be imposed in South Africa (SACOB, 2000), vided by the National Electricity Regulator. eleven can be classified as "environmental" in . Fuel levy: The levy rates applicable from April terms of the expected results, even if their purpose 2000 are: petrol leaded 8.66USc/litre; un-leaded is explicitly revenue generation for non-environ- 8.04USc/litre; gas oil and diesel oil 7.61 USc/litre; mental purposes (see Box 3.9). Energy products kerosene and hydrocarbon mixtures 8.66 USc/litre. account for as many as six out of the eleven * Natal Parks Board levy: Levies range from US$1 "potential environmental taxes", and four out of per person, per camp for hutted accommodation to the five existing "environmental taxes" (the natu- US$0.1 per person per entrance gate for gate entry, ral parks levy being the fifth one). used to assist neighboring communities (capacity building and development). Tanzania. Tanzania has imposed environ- * Electricity tax: The White Paper on Energy Policy mental taxes on a range of sectoral activities, in proposes that an "excise tax" be introduced on addition to the traditional taxes on energy pro- electricity (specific customer segments may be ducts (Kahyarara, 1999). In the primary sector, the excluded from the tax) to fund local govermment. most important tax change may have been the * Environmental levy: The White Paper on Energy removal of subsidies to fertilizers and the impo- Policy announces that the government will investi- sition of a 5 percent tax, which has resulted in a gate the introduction of an environmental levy on decline in fertilizer use of about 40 percent. The energy sales, the proceeds of which would be dedi- environmental consequences have been mixed, cated to supporting more environmentally benign however, with a reduction in the negative effects options, including energy efficiency. of fertilizers on soil being accompanied by a * Climate change levy: A policy discussion docu- disastrous encroachment by farmers in forest ment issued by the Department of Environmental areas, forced by lower crop yields. Local govem- Affairs and Tourism in 1998 hints at the possible arets, forced b low erop yies Loa gen- future use of a charge levied in respect of green- ments adminster a number of taxes whose main house gas emissions. purpose is revenue generation, but that also * Mineral rights tax: The Green Paper on Minerals attempt to mitigate environmental degradation, and Mining Policy of 1997 discusses a proposal for such as taxes on livestock, fishing, and mining. imposing a tax on privately held mineral rights. The general picture is one of limited behavioral * Pollution/waste charges: The White Paper on Inte- effects due to the low rates of the taxes. Tariffs for grated Pollution and Waste Management proposes water consumption (ranging from US$0.31 per to use various "market-based" instruments designed cubic meter for domestic use to US$0.65 for to improve environmental performance. irrigation) are also in place, as is a monthly * Water resource use and protection-related charges: sewerage charge. In general, environmental taxes The National Water Act empowers the relevant in Tanzania are gaining ground, but they are Minister, with the concurrence of the Minister of administered mainly by the local governments- Finance, to establish a water use pricing strategy. with a lack of qualified personnel-leading to a Source Jacob, 2000. 82 Generating Public Sector Resources to Finance Sustainable Development Kenya. The overriding concern in Kenya's developed banking systems and capital markets, tax policies has been the generation of revenue. weak or ineffective enforcement of environmental However, key tax reforms in 1994 have been policy, and uncertainties in fiscal systems. It is significant for the environment. They eliminated sometimes argued that in situations where public the VAT on some fuels and established a road budgets are volatile and public investments often maintenance levy. Taxation on gasoline and diesel cut from year to year, due to the need of macro- was increased while kerosene taxes were reduced.. economic stabilization, earmarked funds can A beach management levy was introduced as a provide secure and sustained financing for mechanism for mobilizing resources for clean-up. environmental programs. Even in countries that These changes have had a number of environ- are relatively more stable and developed, such as mental impacts, the most noticeable ones being: Poland or the Czech Republic, public and private reduced air pollution, through reduced vehicle use sector resources for the environment are con- brought by the vehicle sales tax and the road strained. license fee, and improved quality of the coastal Earmarked "Environmental Funds" are being environment (Ayoo and Jama, 1999). used extensively in transition economies, to Earmarking of environmental funds address broad environmental objectives including education and awareness building, research and Earmarked revenues from environmental institution building, and public and private pollu- taxes or charges are increasingly viewed as an tion abatement. In 1997, for example, the aggre- important mechanism for financing environmental gate revenues of sixteen "national"' funds of CEE projects in many parts of the world. In most and former Soviet Union Republics amounted to countries, pollution charges and waste discharge about US$800 million. While their size and level or treatment fees are becoming more closely of development vary widely from country to aligned with the cost of providing basic environ- country, earmarked funds of this sort have become mental services, and these revenues are increa- quite substantial in some transition economies. In singly being retained by the bodies that are res- Poland for example, environmental funds ponsible for financing those services. At a poli- accounted for 33 percent of total environmental tical level, earmarking is gaining support because conservation expenditure, whereas the comparable it is found that such a "cost-recovery" approach share for the funds in Hungary, Slovenia and may make it easier to build consensus, remove Lithuania was about 20 percent (See Box 3.11 on barriers, and guarantee budget resources to the Polish experience). In Russian Federation, finance environmental programs (Huber and however, funds provided less than 5 percent of others, 1998). finance for environmental projects. Although earmarking of revenues is normally Based on extensive reviews of existing funds avoided in public finance, there are certain condi- (OECD, 1995; Lovei, 1999; Huber and others, tions under which it can be justified. These 1998) it is now generally accepted that financing circumstances usually arise when: (a) transactions through earmarking of funds are effective only it and administrative costs of environmental pro- the underlying reasons for the environmental grams are very high, (b) revenues are related to problems are simultaneously tackled at the policy the explicit provision of a service for a fee, or (c) level. Although incentive effects are important by- collection of revenues is politically difficult products of earmarked direct environmental taxes, without clear indication of how the moneys will efforts to strengthen enforcement capabilities of be used. Earmarked funds provide a means of environmental agencies and upgrading inforna- financing environmental improvements when tion systems are preeminent in successful funds. rnarket, policy and institutional failures impede Without strengthened environmental regulations the emergence of the financing mechanisms and enforcement, environmental funds may end characteristic of more developed market econo- up contributing to existing distortions. Also, the rnies. In transition and developing economies, benefits of earmarking are found to be most pro- these problems can include severe financial nounced when environmental revenues are ear- constraints on enterprises and households, poorly marked to decentralized programs, as in Colombia Generating Public Sector Resources to Finance Sustainable Development 83 Box 3.11: Polish National Fund for Environmental Protection The Polish National Fund for Environmental projects in the fields of special concern. Recipients Protection and Water Management was created in often include municipalities, industrial enterprises, 1989 as an independent financial institution to support research and development institutions, and NGOs. protection of the environment. Pollution charges, fees, Generally, non-commercial organizations receive and fines paid by polluters for exceeding current grants, while commercial enterprises receive loans. standards are the Fund's main source of income. Since The projects are first evaluated by an assessment com- its creation, it has participated in financing more than mission, which issues an opinion on the project's 5,000 projects, accounting for 20 percent of total funding eligibility. The final decision is issued by the expenditures on environmental protection in Poland. Minister of Environment, or the Board of Directors. The National Fund spent US$448 million in 1999 on Projects must comply with the national environmental environment-related projects. The main objective of policy (Anderson and Zylicz, 1999). this National Fund is to finance environmental The National Fund has played a leading role in protection investments, including projects described in environmental project financing and serves as an the "National Environmental Policy" adopted by the excellent case of a well-managed earmarking program. Polish Parliament in 1991 or specified in the The discharge of pollutants into the water and air and "Implementation Programme for the National the accumulation of wastes have decreased consi- Environmental Policy by the year 2000". Their imple- derably recently. For example, 5.3 million cubic mentation is supervised by the Minister of Environ- meters of wastewater per day (86 percent of all indu- mental Protection, Natural Resources and Forestry. strial and municipal wastewater discharged into the National Fund resources are used to finance natio- water) was being treated in 1999 (REC, 2001). In the nal and regional public infrastructure projects, support long-run, pollution abatement expenditures are ex- local projects (such as the construction of wastewater pected to be financed from private resources in res- treatment plants and potable water systems) whose ponse to the incentive effects of environmental taxes costs exceed the capabilities of local budgets, and back and the improved enforcement of regulations. (see above). To date, experience in Latin America thus serves as a basis for decision making at the and the Caribbean has demonstrated that ear- national level (Francis and others, 1999) marking programs are most successful where: (a) Earmarking of environmental funds to over- taxes or incentives are linked to existing come or mitigate numerous challenges has come a collection mechanisms, and (b) revenues are made long way in the last decade or so. In many CEE available to decentralized authorities for environ- and Latin American countries, the funds have mental programming or for institutional strength- accelerated environmental improvements, leve- ening (1-luber and others, 1998). Earmarked funds raged additional finance for environmental invest- have also been found to be effective when they ments, and strengthened domestic capacity for also strengthen the self-financing capacity of project preparation. Experience so far shows that beneficiaries and tackle the causes of financial the evolution, effectiveness, and potential of funds constraints. Thus co-financing emerges as an im- is closely linked with the broader progress asso- portant requirement for disbursement of the funds. ciated with economic and political reforms, as Finally, transparency and accountability in well as with developments in national environ- the operation of earmarked funds are essential to mental policy frameworks. avoid ad hoc political influence and mismana- gement of public funds. Mechanisms for stake- Conclusions holder participation in the decision making can go Environmental taxes and charges have a long way in improving the transparency and increasingly found a place in government efforts efficiency of the operations. For example, both in to control pollution and raise revenues in several Poland and Czech Republic, local authorities play countries around the world. Our definition of very important roles in the establishment of prio- environmental taxes and charges has included rities and in the selection of projects to be funded. those instruments specifically instituted for pollu- The priorities identified by the local authorities tion control purposes as well as those that have come into existence to raise revenues. 84 Generating Public Sector Resources to Finance Sustainable Development Emerging economies in Central and Eastern In developing countries, data on charges and Europe have embraced green levies as a policy environmental impacts is very limited. There are, instrument with considerable enthusiasm. however, specific examples of cases where pollu- According to our calculations, for the 11 CEE tion charge systems have worked very well. In countries for which we have data, revenue both Colombia and the Philippines, water charges amounting to 2 percent of GDP can be attributed have resulted in fairly significant improvements in to environmental taxes and charges in CEE. In water quality. The Chinese pollution charge sys- specific countries such as Bulgaria, the revenues tem, despite various inefficiencies, has resulted in obtained are as high as 15 percent of tax revenues environmental improvements. In Malaysia, a mix- and 4 percent of GDP. These substantial revenues ture of command and control and economic suggest that taxes and charges are a well used instruments has had a positive environmental regulatory tool in CEE transition economies. effect. In general, the charge system is more likely Developing countries have not made as much to be efficient when the primary objective of the progress with the use of economic instruments in tax or charge is to curb pollution, the charge is set general. Certainly several countries have instituted at a high enough level, there is adequate moni- 'environmental taxes and charges with revenue toring, and there is a fair amount of understanding collection being a very important goal. The exam- and acceptance among all stakeholders. ples of China and Colombia show that using Another important issue that needs to be environmental levies is feasible in developing raised relates to the distributional impact of countries. Colombian authorities collected over environmental taxes and charges. The impact of $15 million between 1997 and 2000 from pollu- environmental taxes on the poor can be gauged by tion charges and China collected over $600 looking at: (1) the direct impact when the price of million in 1999 alone. However, huge opportu- the taxed good rises and this affects household nities remain in the developing world for obtain- expenditures; (2) the indirect impact of price ing revenues from environmental taxes. increases when the taxed good is used in the The potential for environmental taxes and production of other goods; (3) the net impact charges in developing countries can be partially given any improvement in the environment and its assessed by considering the situation in OECD impact on welfare and given behavioral changes countries. In OECD countries, environmental that result from changes in the price of the taxed taxes alone (not including charges) make up good. Studies that have analyzed the incidence of approximately 7 percent of total tax revenues. Tax environmental levies on the poor in developing revenues in OECD countries tend to be twice the countries are limited. Given the concentration of level of taxes in developing countries. Even cars ownership in the upper-income classes, gaso- accounting for these differences, it is clear that a line taxes tend to be progressive. Taxes on diesel lot more resources can be raised to finance sus- have been found to be mildly regressive in tainable development in developing countries. developing countries (Gwilliam and others, 2001).13 As identified in a previous chapter, stu- Thve beenvironmixentalnCEE ip ctsofgre liesa dies on water user charges suggest that the poor have been very mixed. In CEE countries and already pay high rates for clean water from other Rlussia, the charges have been rather low and their sore;tu,ndrctanicmtnesni- impact~~~~~~~ o polto qutlite. Poadi sources; thus, under certain circumstances an in- impat onpolltionquit limted.Polad 15 crease in user fees may be acceptable if it means considered an exception. Here pollution charges better access to clean water have been adjusted for inflation and have resulted in positive environmental outcomes. In general, Finally, while ear-marking and environ- lack of monitoring and enforcement capacity, lack mental funds have helped speed the pace of envi- of information on abatement costs and techno- ronmental improvements in many countries logy, and low standards have contributed to the especially in CEE, they should not be seen as sub- disappointing environmental record of green stitutes for more fundamental reforms. They levies in transition economies. should actively support the development of more market-based mechanisms, ultimately leading to stricter, more direct implementation of market- Generating Public Sector Resources to Finance Sustainable Development 85 based instruments. For these reasons, as the use of environmental funds grows, it becomes in- Box 3.12: The simple economics of payments for creasingly important to assess the experience with environmental services different funds, to achieve a better understanding This figure shows the simple logic of payments of their strengths and weaknesses, and to define for environmental services. Land users receive few the types of practices which can best ensure that benefits from forest conservation-often, less than they play a positive role as part of the framework the benefits they would receive from alternative of national environmental policies. land uses, such as conversion to pasture. But the latter uses can impose costs on downstream popula- tions, who no longer receive benefits of environ- 3.4 Developing Innovative mental services such as water filtration. A payment Conservation Financing Mechanisms from the downstream beneficiaries of the service can help make conservation more attractive for land The environmental services provided by na- users. This payment must obviously be more than tural ecosystems are often taken for granted until the difference between the benefit to land users of they are lost-the hydrological services provided conservation and conversion to pasture (or they by forests, for example, are only noticed after would not change their behavior) and less than the value of the benefit to downstream populations (or deforestation results in flooding or reduced water they would not be willing to pay for it). quality, increasing the vulnerability of down- stream populations and threatening their health m 0 0 0: and livelihoods. That such services should be lost . g 8 despite their value is easy to understand: Land cD users typically receive no compensation for the 0 3 o services their land generates for others, and so do o C gg not take them into account in making land use decisions. Responses to this problem have tended to - rely either on regulations which attempt to dictate (D - particular patterns of land use, or on remedial measures, such as repairing the damage caused by =. X flooding or the construction of civil works aimed X at protecting downstream communities from -. flooding. Neither of these approaches has proved .00 a effective. Remedial measures are often imperfect o V % g 0c. and expensive-often far more expensive than preventive measures. And regulations are extre- 3 mely difficult to enforce because of the spatial dispersion of land uses. Moreover, it may impose Source: Pagiola and Platais, forthcoming. high costs on poor land users by restricting them I from undertaking privately profitable land uses. chanisms worldwide (Landell-Mills and Porras, Recognition of this problem and of the 2002), and the list is constantly growing. failure of previous approaches to dealing with it Among the mechanisms being developed, has led to widespread experimentation with one of particular interest here is that of Payments market-based mechanisms to address these for Environmental Services (PES). The central problems. Many believe that market-based principle of PES is that those who provide approaches can provide powerful incentives and environmental services should be compensated for efficient means of conserving forests and the doing so, and those who receive these services public goods they provide, while at the same time should pay for their provision (see Box 3.12). This offering new sources of income to support rural approach seeks to align the incentives faced by livelihoods (Pagiola and others, 2002). A recent land managers with those of society as a whole by review found almost 300 examples of such me- paying land managers for the environmental ser- 86 Generating Public Sector Resources to Finance Sustainable Development Box 3.13: Local initiatives on payments for environmental services in Latin America * Costa Rica pioneered the PES approach in 1997 by The municipal water authorities of the cities of developing a country-wide PES system, financed by Quito and Cuenca, in Ecuador, are allocating part of payments from a fuel tax, sales of carbon emission their revenues to protect their water sources. In credits, contributions from water users such as HEP Quito, the money is used to finance improved mana- producers, and a GEF payment for bio-diversity gement in several protected areas from which they conservation (Pagiola, 2002). By mid-2000, the receive the bulk of their water (Echevarria, 2002). In program covered over 200,000 ha of forest. Cuenca, the municipal water authority has acquired X In Colombia, many water user groups pay for parts of the upper watersheds that supply the city watershed services-sometimes by buying the entire and kept them under forest. upper watershed. Power companies pay a percen- a Municipalities downstream of El Impossible Natio- tage of their revenues from hydroelectric power nal Park in El Salvador have agreed to make a fi- generation to the regional corporations that are res- nancial contribution to park management as payment ponsible for watershed management. for the watershed services they receive. Box 3.14: Payments for environmental services at the World Bank Helping countries to find innovative solutions to 3 The Regional Integrated Silvopastoral Ecosystem problems such as the loss of environmental services, Management Project, which is piloting the use of which overlap issues of livelihood, vulnerability, and PES as a means of encouraging a shift from un- health, is a key element of the Bank's Environment sustainable agricultural practices to sustainable Strategy (World Bank, 2001c). The World Bank is silvopastoral practices in Colombia, Costa Rica, and working with several countries to develop PES sys- Nicaragua. tems that would help substitute for these missing a On-going project preparation work in El Salvador, markets-especially in Central and South America Ecuador, and the Dominican Republic aimed at (Pagiola and Platais, forthcoming). Bank-supported developing pilot PES programs. operational work on PES includes: a Assistance to Mexico in carrying out a survey of * The Ecomarkets Project, which supports Costa land management practices in the ejido sector Rica's PES program. This project includes a (which includes most of the remaining forest area) to US$32.6 million loan from the World Bank to help help design a PES system and provide a baseline to the government ensure current levels of environ- monitor its implementation. mental service contracts, and US$8 million grant The World Bank's training and capacity-building arm, from GEF for the biodiversity services provided the World Bank Institute (WBI) has also developed a through the program. training course on PES targeted at technical personnel in ministries, conservation agencies, and NGOs invol- ved in implementing such programs. vices they generate. It finances these payments by Identifying environmental services. The charges to the beneficiaries of these services, thus benefits sought from land use changes typically relieving the government budget from the burden include (but are not limited to): of doing so. It has the further advantage of provi- o Hydrological benefits: Controlling the timing ding additional income sources for poor land and volume of water flows, and the quality of users, helping to improve their livelihoods. water. Several countries are already experimenting o Reducing sedimentation: Avoiding damage to with such systems (see Box 3.13). Interest has downstream reservoirs and waterways and been particularly strong in Central and South hence their uses (HEP generation, irrigation, America, where the effects of Hurricane Mitch in recreation, fisheries, domestic water supplies). 1998 made the need to protect environmental e Disaster prevention: Preventing flooding and services clear. The World Bank has been landslides. particularly active in supporting such efforts . . o Bcodiversity conservation: Land uses can be (Pagiola and Platais, forthcoming) (see Box 3.14). moeo les frenl to bdvrsy. Generating Public Sector Resources to Finance Sustainable Development 87 * Carbon sequestration: Many land uses pared to the cost of alternatives), and on the size sequester substantial amounts of carbon. of the group. Both qualitatively and, especially, quantita- Once beneficiaries have been identified, a tively, we often know less about the services means must be devised to capture part of their generated by different land uses than we think we willingness to pay. This is obviously easiest when do. This reflects partly the diversity of conditions they are easily identifiable and already organized, encountered (hydrological benefits depend on the so it is relatively easy to negotiate with them. rainfall regirne, on the type of soil and vegetation, Should an agreement be reached for them to pay and on topography, for example) and on the for the water services they receive, they already objectives being sought (regulating waterflows to have the capability of collecting the required avoid flooding and dry season deficits may funds from their members. An additional fee require different interventions than maximizing could be added to water bills paid by municipal total water volume). water users, for example. Populations in flood- prone areas, in contrast, are not organized (except Financing environmental service pay- to the extent that they are included in some of the ments. For PES programs to survive, they must otherethree groups) a re is nomeai to hav seuesucsoiacn. Thi isese other three groups) and there IS no mechanism to have secure sources of financing. This IS espe- collect payments from them. cially important if payments must be on-going, as discussed in the previous section. The basic prin- Developing systems of environmental service ciple is that beneficiaries pay for the services they payments that work. PES programs will only have receive. This requires identifying the beneficiaries the desired effect if they reach the land users in and the specific services they receive. Bene- ways that motivates them to change their land use ficiaries do not receive generic "ecosystem ser- decisions. In general, several principles are clear: vices"-rather, they are interested in very specific * Make payments on-going. The benefits being services. Even within specific services, there are sought will generally be enjoyed year after differences. Domestic water supply systems re- year, as long as appropriate land uses are main- quire a constant flow and high quality, but hydro- tained. For this to occur, land users must re- electric power producers with reservoirs usually ceive payments as long as they maintain the prize total volume and care little for water quality, land use. except for the absence of sedimentation. The * Target payments. An undifferentiated pay- willingness to pay of any given group of bene- ment system that pays everyone the same will ficiaries will depend on the specific service they be much more expensive than a targeted receive, on the value of that service to them (com- scheme. It will also make it difficult to tailor Figure 3.9: Institutional framework for payments for environmental services |Governanice structure| Beneficiary Land user -OBeneficiary Land user Beneficiary Financing Payment . Land user = Beneficiary Mechanism Mechanism Land user - Beneficiary Land user Beneficiary Land user Environmental services Source: Pagiola and Platais, forthcoming. 88 Generating Public Sector Resources to Finance Sustainable Development interventions to the particular requirements of Notes given situations. However, a targeted payment scheme may be more expensive to implement It is important to note that it is not lowering profits than a non-targeted one. A balance needs to be that will change incentives but rather found between the efficiency advantages and er se the higher costs of better targeting. increasing the price of each tree felled-a profit tax the higher costs ofrbetter targetimg. would have no incentive impact (only a scale * Avoid perverse incentives. For example, impact). payments for reforestation can encourage land 2 This is not a phenomenon seen only in developing users to cut down standing trees so as to countries. Many developed countries subsidize qualify. logging activities or fail to charge appropriate Establishing the institutional framework. stumpage fees. Revenues forgone from low stump- PES programs require a supporting institutional age fees and raw logging export bans in British infrastructure. Figure 3.9 summarizes the fun- Columbia, Canada, alone have been estimated to be ctions that must be performed by PES programs. between US$564-1,164 million. However, this is a c portions . . more pressing problem in developing countries, tal portincofthe benefits received by enuronedan- which can ill-afford such revenue losses, and where tal service beneficiaries must be captured and the environmental impact of mis-pricing forest re- channeled to land users to provide incentives to sources is more severe due to weak logging regu- protect forests. These systems depend on several lations (Sizer, 2000). prerequisites. Market participants must have 3 Note that while pre-harvest fees can maximize the access to information on the value and volume of value of timber extraction, they do not necessarily services being exchanged. Participants must have maximize the total value of the forest when forest opportunities for finding and negotiating pay- benefits other than timber are also taken into ments. Property rights over service commodities account. need to be clearly defined and ownership 4 UFAs are concession licenses that allow logging in assigned. Monitoring and enforcement mecha- non-permanent forest domains. They are allocated nisms are required. A network of 'supporting for 15 years and for a maximum of 200,000ha regulatory and institutional arrangements may be (Global Forest Watch, 2000b). necessary for markets to function effectively. 5 Vente de coupe are more limited logging licenses. Establishing such market infrastructure is not These are allocated for one year and for a maximum easy, and rarely cheap. area of 2,500ha. However, the tax burden on these licenses is much lower than on the larger Enhancing the poverty alleviation impacts. concessions, creating an incentive for loggers to Many of the potential supplier of environmental obtain these licenses. Lack of enforcement of services are expected to be poor; the upper water- existing regulations also suggests that substantial sheds which are critical sources of water services, illegal logging occurs with this type of license. for example, are often inhabited by poor subsis- According to one estimate, for every vente de coupe tence farmers. For these land users, payments for that is logged legally, four are logged illegally environmental services represent an important (Seymour and Dubash, 2000). potential additional source of income. This will 6 Autorisation de recuperation allow only 30m3 of not happen automatically, however. Working with extraction from non-permanent forest domains for a many small, dispersed farmers imposes high tran- maximum of three months. saction costs, so particular efforts are needed to 7 The potential for bioprospecting fees to capture part ensure that the poor have access to the new oppor- of the benefits of extracting valuable genetic tunities created by PES programs. In Costa Rica, information from natural habitats has already been for example, a system of collective contracting discussedabove. has been developed, in which groups of small 8There are several definitions of "ecotourism" in the farmers can join the PES program as a group literature. The two most accepted are by the amers can - l. Intemational Ecotourism Society, which defines ecotourism as "responsible travel to natural areas that conserves the environment and sustains the well-being of local people," and by the International Union for the Conservation of Nature, which Generating Public Sector Resources to Finance Sustainable Development 89 defines ecotourism as "environmentally responsible able to become financially self-sustainable on the travel and visitation to relatively undisturbed basis of diving, snorkeling, and boating fees natural areas, in order to enjoy and appreciate charged. nature (and any accompanying cultural features- 15 This is often referred to as the "polluter pays both past and present) that promotes conservation, principle," which recognizes that polluters should has low negative visitor impacts, and provides for pay for any enviromnental damage they create. beneficially active socio-economic involvement of 16 This estimate only includes taxes, for examples local populations." excise duty on petrol, and excludes environmental 9 While revenue information may very well exist, it is charges such as water pollution charges, which not generally reported, except perhaps as part of yield substantial revenues. larger aggregates. 17 This does not include revenue from other sources 10 This manual provides standardized terminology for such as energy and transport taxes, but these taxes park tourism, outlines a five level system for are not very high. measurement, provides guidance on measurement 18 techniques and technologies, and provides examples For examl Rovenia a introdd a m r of the use of tourism data in park management C02 tax and Romania a fuel road tax. (Eagles, 2001). 9 Poland appears to be an exception, since pollution "1 In the early 1990's foreigners were charged a charges are adjusted annually to compensate for US$40 fee and the number of foreign visitors was inflation (REC/SIEI, 2001). about 25,000. In April of 1993 the fee was raised to 20 This makes a carbon tax more appealing than a tax US$80, and number of foreign visitors reached on S02 emissions, (the source of acid rain) for 37,000. In 1997 fees were raised to US$100, while instance, as the latter would need to be monitored the number of foreign visitors continued to increase directly. to just over 50,000 (Honey, 1999). 21 Of the three major fossil fuels, coal produces the 12 Revenue figures are estimated using average prices most carbon per unit of energy, followed by oil and for a given year and multiplying by the number of then natural gas. visitors for that year. 22 Factories are charged only for pollution in excess of 13 There were incidents of retaliation, however, where standards, and the charge is levied only on the the Maasai people at Amboseli began spearing single air or water pollutant that most seriously vio- wildlife in response to a proposed government lates regulatory standards for each medium. The policy in 1971 to take over large tracts of land sectors that contribute the most to the pollution within the Maasai water basin. The government's charge system are smelting, chemical, textile, failure to keep compensation agreements reached electricity, and coal (World Bank, 2000). with the community continued to be a source of 23 China's regulators do impose serious penalties, problems. including shutdowns, for factories that persistently 4 There are some examples cited in the literature of violate standards. higher fees having no impact on visitation rates 24 The official rate is supposed to apply uniformly (Belize, Brazil, Egypt, Indonesia, Italy, Netherlands across China. Antilles (Saba MP), Papua Guinea, and Suriname, 25 Local collection incentives are also affected by the reported by Lindberg and Halpenny 2001) and even officially-mandated use of the levies collected. The leading to higher visitation rates due to site regulations require the use of 80 percent for improvements (Moremi Game Reserve in Bostwana subsidies or loans to co-finance abatement by pollu- study by Barnes, cited by Lindberg and Aylward, ters who have paid a levy. Local environmental 1999; and Bonaire Marine Park, in the Netherlands authorities can use the remaining 20 percent and Antilles, where increased fees have been used to any other penalty fees to support their own opera- maintain coral reefs and have attracted more divers, tions. For further discussion of the allocation of and St. Lucia's Soufriere MMA in Lindberg and levy funds, see Yun (1998) and Yang and others Halpenny, 2001; Galapagos National Park charges (1997). US$100 to foreign visitors and still sees the number 26 O l of visitors increasing year after year, beyond what (US$45) fee, plus a charge of M$10 (US$4.50) per its estimated carrying capacity, several sources). In tonne of organic pollution discharged into water Lindberg and Halpenny's review of marine and a surcharge of M$100 (US$45) per tonne for protected areas, they cite many parks that have been BOD discharges beyond the allowable limits. Since 90 Generating Public Sector Resources to Finance Sustainable Development the DOE at that time had no way of valuing actual mental taxes issue (Brett and Keen, 2000; damages from pollution, it intended this charge to Marsiliani and Renstrom, 2000; Fredricksson, be high enough to provide some abatement 1997). incentive without being burdensome. 33 This work does not discuss any benefits to the poor 27 In the ensuing years, the government abandoned the from a tax-resulting decrease in pollutants. surcharge, maintaining only the M$10 (US$4.50) per tonne discharge fee and specified that the standards would be mandatory. 28 Some companies find it cheaper to pay the fine than to treat their effluent sufficiently to meet the standard. 29 In 1999-2000, charges and fees amounted to 5.4 percent of revenue receipts, but was higher at 7.7 percent in 1997-98. Estimate based on World Bank data for Pakistan. 30 Includes receipts from services provided in the areas of agriculture, fisheries, forestry, fuel and power, industrial and mineral resources, printing, industries, transport, and communications. The estimate is based on World Bank data for Pakistan. 31 Environmental funds generally derive their reve- nues from taxes and charges on pollution. These receipts are set aside for environmental purposes instead of being transferred to the general govern- ment budget. The funds then are used to provide finance and assistance for investments and other activities to achieve environmental objectives. 32 In the public finance literature the conventional view is that taxes collected from various activities should be paid into a general fund, from which government expenditures are financed. In other words, there should be no "earmarking" of taxes. The reason for this position is that earmarking intro- duces inefficiencies and constraints in the way that a government can allocate funds to the different expenditure categories. If environmental taxes are earmarked for expenditure on environmental invest- ments, the level of those investments will be dicta- ted by the tax revenues rather than by the merits of the investments; or the tax rates will be dictated by the abatement expenditure requirements and not by estimates of the environmental damage. The former (expenditures determined by tax revenues) is more common and can result in too low a level of abatement, or too high, depending on what taxes are collected. Also, from a fiscal policy perspective, for example, earmarking has potential dangers: allo- cating and disbursing revenues outside the govern- ment budget may result in economic inefficiencies, particularly over the long term. In addition, if funds are not effective in supporting environmental policy goals, they lose their main rationale. There is a rich theoretical literature on the earmarking of environ- 4. Summary Several main lessons emerge from this the poor, it might well benefit them if the discussion: resources that are freed up are used in more * There is substantial potential to generate appropriate ways. However, any policy reform additional public sector resources, although or other effort must clearly take considerable data limitations preclude a comprehensive esti- care not to inadvertently harm the poor. mate. The amounts vary considerably across * Reform will require political will, but also sectors and countries, however. The most im- good governance, capacity building, and portant potential source of additional revenue investment. In many cases, it will be necessary comes not from efforts to generate new reve- to spend money in order to make money. Cost nues, but from freeing up available resources recovery in domestic water, for example, will by improving the efficiency with which they require establishing water metering and crea- are spent-in particular, by reforming subsi- ting or strengthening the institutional capacity dies that are expensive and, often, environmen- to read meters. Good governance is critically tally harmful. Even when the sums involved important for revenue generation from natural appear limited, reform can help make sub- resources. Recent evidence from Cameroon, sectors financially self-sustaining rather than for example, shows that sound reforms that en- wholly dependent on the public purse. hance transparency can result in significant in- * Reforms can have important incentive creases in forest rents to the state. A key mes- effects. In addition to generating new resources sage here is that reforms must take into account or freeing up existing ones, many of the re- the willingness and ability of governments to forms discussed in this paper would also help implement these reforms. reduce environmental damage, by providing in- * One size does not fit all. There is substantial centives that tend to discourage environmen- variation in the needs, opportunities, and con- tally-harmful activities and encourage more straints facing different developing countries. sustainable ones. Indeed, in many cases this Even within countries, there is substantial vari- outcome would be as or more important as the ation in capacity to irnplement public reform resource generation itself. For example, in both programs across regions or sectors. Although a Colombia and the Philippines, water charges- paper of this size obviously cannot do justice to although motivated primarily by revenue gene- this diversity, we have tried to point out diffe- ration-have resulted in fairly significant im- rences where appropriate. provements in water quality. Likewise, the Chinese pollution charge system has resulted in environmental improvements. * Reform is not anti-poor. Though many policies claim to be pro-poor, they often are not. Electricity and water subsidies are of no assistance to poor people who almost always lack access to these services. Reducing these subsidies would, therefore, not only not harm 91 References Abu-Zeid, M. 2001. "Water Pricing in Irrigated Countries." Resources, Conservation and Recycling, Agriculture." International Journal of Water 8, pp. 43-54. Resources Development, 17:4, pp. 527-538. Bayon, R., J.S. Lovink, and W.J. Veening. 1999. Ahmed, H. 2000. "Three Essays on Water Service: A "Financing Biodiversity Conservation." Sustainable Case Study of the Residential Sector of Cairo." Ph.D. Development Department, Environment Division. Dissertation. East Lansing: Michigan State Washington: Inter-American Development Bank. University. Beede, D.N., and D.E. Bloom. 1995. "The Economics Albouy, Y., and N. Nadifi. 2000. "Impact of Power of Municipal Solid Waste." The World Bank Sector Reform on the Poor: A Review of Issues and Research Observer, 10:2, pp. 113-150. the Literature." ESMAP Technical Paper No.002. Bhatia, R., and M. Falkenmark. 1993. "Water Resource Washington: UNDP and World Bank. Policies and the Urban Poor: Innovative Approaches Anderson, G., and B. Fiedor. 1997. "Environment and Policy Imperatives." Water and Sanitation Charges in Poland." In R. Bluffstone and B.A. Currents Working Paper. UNDP-World Bank Water Larson (eds), Controlling Pollution in Transition and Sanitation Program. Washington: World Bank. Economies, Theories and Methods. Cheltenham: Blore, I., and F. Nunan. 1997. "Public Valuation of Edward Elgar Publishing Ltd. Solid Waste Impacts: A Case Study in Bangkok." In Anderson, G., and T. Zylicz. 1999. "The Role of Polish C. Kirkpatrick and N. Lee (eds.), Sustainable Environmental Funds: Too Generous or Too Development in a Developing World. Cheltenham: Restrictive?" Environment and Development Edward Elgar. Economics, 4, pp. 413-448. Boland, J., and D. Whittington. 2000. "The Political Arntzen, J. 2001. "Economic Instruments and Natural Economy of Water Tariff Design in Developing Resource Management in Southern Africa." Policy Countries: Increasing Block Tariffs Versus Uniform Brief No. 1. Pretoria: Resource Accounting Network Price with Rebate." In A. Dinar, ed, The Political for Eastern and Southern Africa (processed). Economy of Water Pricing Reforms. New York: Asian Development Bank (ADB). 1993. Water Utilities Oxford University Press. Handbook: Asian and Pacific Region. Manila: ADB. Bosquet, B. 2000. "Environmental Tax Reform: Does Ayoo, C., and M.A. Jama. 1999. "Environmental it Work? A Survey of the Empirical Evidence." Taxation in Kenya." In T. Stemer, ed., The Market Ecological Economics, 34, pp. 19-32. and the Environment. Cheltenham: Edward Elgar. Bosquet, B. 2001. "The Role of Natural Resources in Bacon, R.W. 2001. "Petroleum Taxes." Viewpoint, Fundamental Tax Reform: Theory, Practice and BaonR.W.40 01.Wahntn "Peroleu Taxes. Vi onProspects." Ph.D. Dissertation. College Park: No.240. Washington: World Bank Uiest fMrln Barcena, A., C. Miguel, and J. Acquatella. 2001. University of Maryland. "Financing for Sustainable Development in Latin Bovenberg, A.L. and R.A. de Mooij, 1994. America and the Caribbean." Background paper for "Environmental Levies and Distortionary Taxation." the Regional Preparatory Conference of Latin American Economic Review, 84, pp. 1085-1089. America and the Caribbean for the World Brandon, C. 1995. "Valuing Environmental Costs in Conference on Sustainable Development, Rio de Pakistan: The Economy-Wide Impact of Janeiro, Brazil, October 23-24, 2001. Environmental Degradation," South Asia Region Barde, J., and N.A. Braathe. 2002. "Environmentally intemal discussion paper No.IDP-164. Related Levies." Paper presented at the Conference WashingtonWorld Bank. on Excise Taxation, The Hague, April 11-12, 2002. Brett, C., and M. Keen. 2000. "Political Uncertainty Barret, C., and T. Lybbert. 2000. "Is Bioprospecting a and the Earmarking of Environmental Taxes." Viable Strategy for Conserving Tropical Journal ofPublic Economics, 75, pp. 315-340. Ecosystems?" Ecological Economics, 34, pp. 293- Briscoe, J. 1996. "Financing Water and Sanitation 300. Services: The Old and New Challenges." Water Bartone, C.R., and J.D. Bernstein. 1993. "Improving Supply, 14:3/4, pp. 1-17. Municipal Solid Waste Management in Third World Briscoe, J. 1999a. "The Changing Face of Water Infrastructure Financing in Developing Countries. 93 94 Generating Public Sector Resources to Finance Sustainable Development International Journal of Water Resources Conservation International. 2000a. "Conservation Management, 15:3, pp. 301-308. International Creates New Market Mechanism to Briscoe, J. 1999b. "The Financing of Hydropower, Conserve Global Forests." Press Release, September Irrigation and Water Supply Infrastructure in 25, 2000. Washington: Conservation International. Developing Countries." Water Resources Conservation International. 2000b. "Conservation Development, 15:4, pp. 459-491. Trust Brings Economic Opportunities to Suriname." Brown, C. 2001. "Visitor Use Fees in Protected Areas: Press Release, April 8, 2000. Washington: Synthesis of the North American, Costa Rica and Conservation International. Belizean Experience." Ecotourism Program, Contreras-Hermosilla, A. 2001. "Illegal Forest Technical Report Series No.2. Washington: The Activities in the Asia Pacific Rim." Markets for Nature Conservancy. Forest Conservation, September, pp. 1-8. Brown, D.W. 1999. "Addicted to Rent: Corporate and Craft, A.B., and R.D. Simpson. 2001. "The Value of Spatial Distribution of Forest Resources in Biodiversity in Pharmaceutical Research with Indonesia; Implications for Forest Sustainability and Differentiated Products." Environmental and Government Policy." Report No.PFM/EC/99/06. Resource Economics, 18, pp. 1 -17. Jakarta: ITFMP. Crook, C. 2001. "Biodiversity Prospecting Brown, D.W. 2001. "Why Governments Fail to Agreements: Evaluating the Economic and Capture Economic Rent: The Unofficial Conservation Benefit in Costa Rica and Peru." Appropriation of Rain Forest Rent by Rulers in Unpublished Ph.D. Dissertation. Toronto: University Insular Southeast Asia Between 1970 and 1999." of Toronto. Ph.D. Thesis, University of Washington. Day-Rubenstein, K., and G. Frisvold. 2001. "Genetic Bruner, A., R. Gullison, R. Rice, and G. Fonseca. 2001. Prospecting and Biodiversity Development "Effectiveness of Parks in Protecting Tropical Agreements." Land Use Policy, 18, pp. 205-219. Biodiversity." Science, 291, pp. 125-128. Deacon, R., and P. Murphy. 1997. "The Structure of an Cameroon, Government of. 1998. "Contribution du Environmental Transaction: the Debt-for-Nature Secteur Forestier a I'Economie Nationale." Yaounde: Swap." Land Economics, 73, pp. 1-24. Ministere de L'Economie et des Finances. Devarajan, S., M.J. Miller, and E.V. Swanson. 2002. Castro, G., and I. Locker. 2000. Mapping Conservation "Goals for Development: History, Prospects, and Investments: An Assessment of Biodiversity Funding Costs." Washington: World Bank. in Latin America and the Caribbean. Washington: Dharmarantne, G., F. Sang, and L. Walling. 2000. Biodiversity Support Program. "Tourism Potentials for Financing Protected Areas." Ceballos-Lascurain, H. 2001. "Integration of Annals of Tourism Research, 27, pp. 590-610. Biodiversity Conservation Planning into the Tourism Diaz L. 2001. "An Overview of Solid Waste Sector." Industry and Environment, 24:3-4, Special Management in Economically Developing Issue on Ecotourism and Sustainability, pp. 38-41. Countries." Paper presented at the A Bird's Eye Cestti, R., G. Yepes, and A. Dianderas. 1997. View on Waste Regional Conference on Integrated "Managing Water Demand By Urban Water Solid Waste Management, Cape Town, 25 October Utilities: A Review of the Literature." Washington: 1999. World Bank and SIDA (processed). Dinar, A. 2000. "Political Economy of Water Pricing Chase, L., D. Lee, W. Schulze, and D. Anderson. 1998. Reforms." In A. Dinar (ed.), The Political Economy "Ecotourism Demand and Differential Pricing of of Water Pricing. New York: Oxford University National Park Access in Costa Rica." Land Press for ihe World Bank. Economics, 74, pp. 466-482. Dinar, A., and A. Subramanian. 1997. "Water Pricing Collomb, J.-G., and H. Bikie. 2001. "1999-2000 Experiences: An International Perspective." Allocation of logging permits in Cameroon: Fine Technical Paper No.386. Washington: World Bank. Tuning Central Africa's First Auction System." Donge, L., M. Kato, and C. Maurer. 2001. "An Washington: Global Forest Watch. Environmental Analysis of Recent Trends in Cointreau-Levine, S.J. 1994. Private Sector International Financial Flows with a Special Focus Participation in Municipal Solid Waste Services in on Japan." Washington: World Resources Institute. Developing Countries. Vol. 1, The Formal Sector. Eagles, P. 2001. "International Trends in Park Washington: World Bank. Tourism." Paper presented at the EUROPARC 2001 Conference, October 3-7, Matrei, Australia. Generating Public Sector Resources to Finance Sustainable Development 95 Echevarria, M. 2002. "Financing Watershed Foster, V., A. Gomez-Lobo, and A. Halpern. 1999. Conservation: The FONAG Water Fund in Quito, "Designing Direct Subsidies for Water and Ecuador." In S. Pagiola, J. Bishop, and N. Landell- Sanitation Services: The Case of Panama." Santiago: Mills (eds), Selling Forest Environmental Services University of Chile (processed). Market-based Mechanisms for Conservation and Francis, P., J. Klarer, and N. Petkova. 1999. Development. London: Earthscan. Sourcebook on Environmental Funds in Economies Economic Commission for Latin America and the in Transition: A Regional Overview and Surveys of Caribbean (ECLAC)/United Nations Development Selected Environmental Funds in Central and Programme (UNDP). 2001. "Financing for Eastern Europe and the New Independent States. Sustainable Development in Latin America and the Paris: OECD. Caribbean." Paper prepared for the Regional Fredricksson, P.G. 1997. "The Political Economy of Preparatory Conference of Latin America and the Pollution Taxes in a Small Open Economy." Journal Caribbean for the World Conference on Sustainable of Environmental Economics and Management, 33. 1, Development. pp. 44-58. Ecotourism Society. 1998. "Ecotourism Statistical Fact Freund, C., and C. Wallich. 1997. "Public-Sector Price Sheet." Burlington: International Ecotourism Reforms In Transition Economies: Who Gains? Who Society. Loses? The Case Of Household Energy Prices In Energy Sector Management Assistance Programme Poland." Economic Development and Cultural (ESMAP). 1994. "Ecuador Energy Pricing, Poverty, Change, 46, pp. 35-59 and Social Mitigation." Washington: World Bank Gillis, M. 1990. "Forest Concession Management and Energy Sector Management Assistance Programme Revenue Policies" in N.P. Sharma, ed, Managing the (ESMAP). 2000. "World Bank and IAP Related World's Forest- Looking for Balance Between Activities in India." Indoor Air Pollution Newsletter Conservation and Development. Dubuque: No. 1. Washington: World Bank Kendall/Hunt Publishing Co. Energy Sector Management Assistance Programme Giongo, F., J. Bosco-Nizeye, and G. Wallace. 1994. "A (ESMAP). 2001 a. "Energy Services for the World Study of Visitor Management in the World's Poor." Washington: World Bank National Parks and Protected Areas." Report for by Energy Sector Management Assistance Programme the Colorado State University, The Ecotourism (ESMAP). 2001b. "Expanding Access to Modem Society, The World Conservation Union, and The Energy." Indoor Air Pollution Newsletter No.4. World Conservation Monitoring Center. North Washington: World Bank Bennington: The Ecotourism Society. Energy Sector Management Assistance Programme Gleick, P. 1996. "Basic Water Requirements for (ESMAP). 2001 c. "Health of Women." Indoor Air Meeting Human Activities: Meeting Human Needs." Pollution Newsletter No.3. Washington: World Bank Water International, 21:2, pp. 83-92. European Bank for Reconstruction and Development Global Environmental Facility (GEF). 1998. "GEF (EBRD). 2001. Transition Report 2001: Energy in Evaluation of Experience with Conservation Trust Transition. London: EBRD Funds." Monitoring and Evaluation Team. European Forest Institute. 2000. "Policies for Washington: World Bank. Sustainable Forestry in Belarus, Russia and Global Forest Watch. 2000a. "A First Look at Logging Ukraine." Leiden: European Forest Institute. in Gabon." Washington: World Resource Institute. Farnsworth, N., and D. Soejarto. 1985. "Potential Global Forest Watch. 2000b. "An Overview of Consequences of Plant Extinction in the United Logging in Cameroon." Washington: World States on the Current and Future Availability of Resource Institute. Prescription Drugs." Economic Botany, 39:3, pp. Gonzalez, F. 2000. "Rethinking Irrigation." Rural 231-240. Development Department." Washington: World Fass, S.M. 1993. "Water and Poverty: Implications for Bank (processed). Water Planning." Water Resources Research, 29:7, Goulder, L.H. 1995. "Environmental Taxes and the pp. 1975-1981. Double Dividend: A Reader's Guide." International Foster, V., and C. Araujo. 2001. "Poverty and Modem Tax Public Finance, 2:2, pp. 157-183. Utility Services." Washington: World Bank Green, M.J.B, and J. Paine. 1997. "State of the (processed). World's Protected Areas at the End of the Twentieth Century." Cambridge: World Conservation Monitoring Centre. 96 Generating Public Sector Resources to Finance Sustainable Development Gwilliam K., R. Bacon, M. Kojima, and K. Lvovsky. ICBG website: 2001. "Transport Fuel Taxes and Urban Air http:/www.hih.gov/fic/programs/icbg.htmnl. Quality." Pollution Management in Focus Discussion International Energy Agency (IEA). 1998. Projected Note No. 11. Washington: World Bank. Costs of Generating Electricity. Update 1998. Paris: Hadden, P. 1999. "Forestry Issues in the Guyana Shield IEA. Region: A Perspective on Guyana and International Energy Agency (IEA). 1999. World Suriname."European Union Tropical Forestry Paper Energy Outlook Looking at Energy Prices: Getting No.3. London: Overseas Development Institute. the Prices Right. Paris: IEA. Hall, D., and M. Hanemann. 1996. "Urban Water Rate International Energy Agency (IEA). 200 la. Energy Design Based on Marginal Cost." In D. Hall, ed, Statistics of Non-OECD Countries. Paris: Advances in The Economics of Environmental OECD/IEA. Resources: Marginal Cost Rate Design and International Energy Agency (IEA). 200 lb. Toward a Wholesale Water Markets. Volume 1. Greenwich: Sustainable Energy Future. Paris: OECD/IEA. JAI Press.SutialEnryFtr.PisOCDEA 1-ramilton, K., and M. Clemens. 1999. "Genuine International Energy Agency/United Nations S Rates in Developing Countries." World Enviromnent Programme (IEA/UNEP). 2001. Savings Economic Revew1pp. 333-356. "Energy Subsidy Reform and Sustainable. Bank Economic Review,13, pp.33-36.Development: Challenges for Policymakers. Annex Hardner, J., and R. Rice. 2002. "Rethinking Green to the Synthesis Report." New York: UNEP Consumerism." Scientific American, 286:5, pp. 89- (processed). 95. International Tropical Timber Organization (ITTO). lIemelaar, L. 1999. "Financing Waste Management in 2000. "Prices for African Timber Have Withstood Developing Countries." Warmer Bulletin, 64, pp. 11- the Asian Financial Crisis and Demand Remains 13. Strong." ITTO Newsletter, 10:2. Himalayan Light Foundation (HLF). 2001. "The International Tropical Timber Organization (ITTO). Challenge of Rural Energy Service in Nepal." 2001. "Cameroon Suspends Logging by 32 Katmandu: HLF (processed). Companies." IITO Newsletter, 11:2. Hoehn, J., and D. Krieger. 1996. "Cairo Water and Instituto Nacional de Biodiversidad (INBio). 2002. Wastewater Economic Benefit Assessment." Cairo: "Biodiversity Prospecting." INBio website USAID (processed). http://www.inbio.ac.cr/en/ (accessed June, 2002). Honey, M. 1999. "Ecotourism and Sustainable Jackson, N. 2000. "Environmental Taxes: Current Development: Who Owns Paradise?" Washington: Priorities in Central and Eastern Europe." Newsletter Island Press. on Green Budget Reform, 7. Website: Huber, R.M., J. Ruitenbeek, and R.S. Da Motta. 1998. www.rec.org/siei. "Market Based Instruments for Environmental Policy Jiwanji, M. 2001. "Resource Taxation: The Subsoil and Making in Latin America and the Caribbean: Forest Sectors." Washington: World Bank. Lessons from Eleven Countries." Discussion Paper (processed). No.381. Washington: World Bank. Johannessen, L.M., with G. Boyer. 1999. Idelovitch, E., and K. Ringskog. 1995. "Private Sector "Observations of Solid Waste Landfills in Participation in Water Supply and Sanitation in Latin Developing Countries: Africa, Asia, Latin America." America." Washington: World Bank (processed). Washington: World Bank (processed). International Finance Corporation (IFC)/World Johansson, R. 2000. "Pricing Irrigation Water: A Bank/Multilateral Investment Guarantee Agency Literature Survey". Policy Research Working Paper (MIGA). 2000. "Tourism and Global Development." No.2449. Washington: World Bank. Washington: World Bank. Jones, W.I. 1986. The World Bank and Irrigation. A Imschoot, M.V. 1992. "Water as a Source of World Bank Operations Evaluation Study. Employment." International Labor Review, 131: 1, Washington: World Bank. pp. 125-138. Kahyarara, G. 1999. "Economic Instruments for Intergovernmental Panel on Climate Change (IPCC). Environmental Management in Tanzania: The Case 2001. Climate Change 2001: Mitigation. Cambridge: of Environmental Taxes." In T. Sterner, ed., The Cambridge University Press. Market and the Environment. Cheltenham: Edward International Cooperative Biodiversity Groups (ICBG). Elgar. 2001. "Programs and Initiatives." Available at the Generating Public Sector Resources to Finance Sustainable Development 97 Kaiser, J., and A. Lambert. 1996. Debt Swaps for Feasibility of 'Win-Win' Outcomes." IMF Working Sustainable Development. Paper No.98/75. Washington: IMF. IUCN/SCDO/EURODAD. Gland: IUCN. Lim, E.R. 2001. "Need for Reforms." Paper presented Kessides, K. 1997. "World Bank Experience with the at the Conference on Distribution Reforms, New Provision of Infrastructure Services for the Urban Delhi, October 12-13, 2001. Poor: Preliminary Identification and Review of Best Lindberg, K., and B. Aylward. 1999. "Price Practices." Issues Note. Transportation, Water, and Responsiveness in the Developing Country Nature Urban Development Department. Washington: Tourism Context: Review and Costa Rica Case World Bank (processed). Study." Journal of Leisure Research, 31, pp. 282- Khwaja, M.A. 2001. "National Environmental Quality 299. Standards and Plan of Action." SDPI Research and Lindberg, K., and E. Halpenny. 2001. "Protected Area News Bulletin, July-August, 8:4, pp. 5-6 Visitor Fees." North Bennington: The Ecotourism Kojima, M., and M. Lovei. 2001. "Coordinating Society. Environment, Transport, and Energy Policies for Lovei, M. 1999. "Environmental Funds." Pollution Cleaner Air." Pollution Management In Focus Management in Focus Discussion Note No. 1. Discussion Note No. 10. Washington: World Bank. Washington: World Bank. Korfmacher, K.S. 1997. "Solid Waste Collection Lvovsky, K. 2002. "Environment, Health, and Systems in Developing Urban Areas of South Africa: Poverty." Environment Strategy Note No. 1. An Overview and Case Study." Waste Management Washington: World Bank. & Research, 15:5, pp. 477-494. Lvovsky, K., G. Hughes, D. Maddison, B. Ostro, and Kozeltsev, M., and A. Markandya. 1997. "Pollution D. Pearce. 2000. "Environmental Costs of Fossil Charges in Russia: The Experience of 1990-1995." Fuels: A Rapid Assessment Method with Application In R. Bluffstone and B.A. Larson (eds), Controlling to Six Cities." Environment Department Paper Pollution in Transition Economies* Theories and No.78. Washington: World Bank. Methods. Cheltenham: Edward Elgar Publishing Ltd. Maille, P. and R. Mendelsohn. 1993. "Valuing Laarman, J., and H. Gregersen. 1996. "Pricing Policy Ecotourism in Madagascar." Journal of in Nature-Based Tourism." Tourism Management, Environmental Management, 38, pp. 213-218. 17, pp. 247-254. Marsiliani, L., and T.I. Renstrom. 2000. "Time Laird, S.A., and K. ten Kate. 2002. "Linking Inconsistency in Environmental Policy: Tax Biodiversity Prospecting and Forest Conservation." Earmarking as a Commitment Solution." Economic In S. Pagiola, J. Bishop, and N. Landell-Mills (eds.), Journal, 110:462, pp. C123-C138. Selling Forest Environmental Services: Market- Mato, R.R.A.M., and M.E. Kaseva. 1999. "Critical based Mechanismsfor Conservation and Review Of Industrial and Medical Waste Practices in Development. London: Earthscan. Dar es Salaam City." Resources, Conservation and Lampietti J.A., A.A. Kolb, S. Gulyani, and V. Recycling, 25:3-4, pp.271-287. Avenesyan. 2001. "Utility Pricing and the Poor. Reycling, 25:3-4 , p .271-287. Lessons from Armenia." World Bank Technical May, P.H., F. Veiga Neto, V. Denardin, and W. Laessons.49m Washington: World Bank. Loureiro. 2002. "The 'Ecological' Value-Added Tax: Paper No.497. Washington: World Bank. Municipal Responses in Parana and Minas Gerais, Landell-Mills, N., and 1. Porras. 2002. Silver Bullet or Brazil." In S. Pagiola, J. Bishop, and N. Landell- Fools' Gold? A Global Review of Markets for Forest Mills, eds, Selling Forest Environmental Services: Environmental Services and Their Impact on the Market-based Mechanisms for Conservation. Poor. London: IIED. London: Earthscan. Lao PDR, Government of. 1999. "Government's McAllister, D.E. 1991. "Estimating the Pharmaceutical Strategic Vision for the Agricultural Sector." Draft, Values of Forests, Canadian and Tropical." October 1999. Vientiane: Government of Lao PDR. Canadian Biodiversity, 1:3, pp. 16-26. Lele, U., V. Viana, A. Verissimo, S. Vosti, K. Perkins, Menkhaus, S., and D. Lober. 1996. "International and S.A. Husain. 2000. "Brazil-Forests in the Ecotourism and the Valuation of Tropical Rainforest Balance: Challenges of Conservation Development, in Costa Rica." Journal of Environmental Evaluation Country Case Study Series." Washington: Management, 47, pp. 1-10. World Bank. Mercer, E., R. Kramer, and N. Sharma. 1995. "Rain Ligthart, J.E. 1998 "The Macroeconomic Effects of Forest Tourism-Estimating the Benefits of Tourism Environmental Taxes: A Closer Look at the Development in a New National Park in 98 Generating Public Sector Resources to Finance Sustainable Development Madagascar." Journal of Forest Economics, 1, pp. Environmental Services Market-based Mechanisms 196-210. for Conservation and Development. London: Metschies, G.P. 1999. Fuel Prices and Taxation. Earthscan. Eschborn: GTZ. Pagiola, S., J. Bishop, and N. Landell-Mills (eds). Metschies, G.P. 2001. Fuel Prices and Vehicle 2002. Selling Forest Environmental Services: Taxation. Second edition. Eschborn: GTZ. Market-based Mechanisms for Conservation and Myers, N., and J. Kent. 2001. Peverse Subsidies: How Development. London: Earthscan. Tax Dollars Can Undercut the Environment and the Pagiola, S., and G. Platais. Forthcoming. Payments for Economy. Washington: Island Press. Environmental Services. Washington: World Bank. Navrud, S., and E. Mungatana. 1994. "Environmental Pakistan National Program. 1998. "Managing Irrigation Valuation in Developing Countries: The Recreational for Environmentally Sustainable Agriculture in Value of Wildlife Viewing." Ecological Economics, Pakistan: A Case Study of the Punjab Irrigation 11, 135-151. Department." Consultancy Report No.C-12. Lahore: Oates, W.E. 1995. "Green Taxes: Can We Protect the International Irrigation Management Institute. Environment and Improve the Tax System at the Parra-Bozzano, D. 2001. "Galapagos: Ecoturismo Same Time?" Southern Economic Journal, 61, pp. versus Conservaci6n." Industry and Environment, 914-922. 24:3-4, Special Issue on Ecotourism and Organization for Economic Co-operation and Sustainability, pp. 30-32. Development (OECD). 1995. Environmental Funds Parry, I.W.H. 1995. "Pollution Taxes and Revenue in Economies in Transition. Paris: OECD. Recycling." Journal of Environmental Economics Organization for Economic Co-operation and and Management, 29, pp. S64-S77. Development (OECD). 1999. Economic Instruments Pearce, D. 1999. "Valuing biological diversity: issues for Pollution Control and Natural Resources and overview." Paper presented at the Workshop on Management in OECD Countries: A Survey. Paris: Benefit Valuation for Biodiversity Resources OECD. organized by the OECD, Paris, on October 18-19, Organization for Economic Co-operation and 1999. Development (OECD). 2000a. "Aid Targeting the Pearce, D., and S. Puroshothaman. 1992. Protecting Rio Conventions: First Results of a Pilot Study." Biological Diversity. The Economic Value of Paris: OECD. Pharmaceutical Plants. Global Environmental Organization for Economic Co-operation and Change Working Paper No.92-27. London: Development (OECD). 2000b. Development CSERGE, UEA, and UCL. Cooperation Report 1999. Paris: OECD. Pearce, D., and S. Puroshothaman. 1995. "The Organization for Economic Co-operation and Economic Value of Plant-Based Pharmaceuticals." Development (OECD). 2000c. "Donor Support for In T. Swanson (ed.), Intellectual Property Rights and Institutional Capacity Development in Environment: Biodiversity Conservation an Interdisciplinary Lessons Leapied." Paris: OECD. Analysis of the Values of Medicinal Plants. Lessons Learned" Paris: OECD.Cambridge: Cambridge University Press. Organization for Economic Co-operation and PerryC g20 Caring f riger:tThe Development (OECD). 2000d. Greening Tax Mixes Persy, C.J. 2000. "Charging for Irrigation Water: The in OECD Countries: A Preliminary Assessment. Issues and Options with a Case Study from Iran." Environment Directorate, Directorate for Financial, Research Report No.52. Colombo: Intemational Fiscal and Enterprise Affairs. Paris: OECD. Water Management Institute. Operations Evaluation Department (OED). 2000. Petr;, M., G. Taube, and A. Tsyvinski. 2002. "Energy "Rural Water Projects: Lessons from OED Quasi-Fiscal Activities in BRO Countries." IMF Evaluations." OED Working Paper Series. Working Paper No.02/60. Washington: IMF. Washington: World Bank. Porter, R.C. 1993. "Providing Urban Environmental Organizacion Latinoamericana de Energfa (OLADE). Services in Developing Countries." Environmental 2000. Estadisticas e Indicadores Econe mico and Natural Resources Policy and Training Energiticos de America Latina y el Caribe. Quito: Project/Midwest Universities Consortium for OLADE. International Activities. Policy Brief No 4. E'agiola, S. 2002. "Paying for Water Services in Central Arlington: USAID. America: Learning from Costa Rica." In S. Pagiola, Principe, P. 1989. The Economdc Value of Biodiversty J. Bishop, and N. Landell-Mills (eds), Selling Forest Among Medicinal Plants. Paris: OECD. Generating Public Sector Resources to Finance Sustainable Development 99 Principe, P.P. 1991. "Valuing Diversity in Medicinal Ruitenbeek, H.J. 1989. "Economic Analysis of Issues Plants." In V. Heywood, 0. Akerele, and H. Synge and Projects Relating to the Establishment of the (eds.), Conservation of Medicinal Plants. Proposed Cross River National Park (Oban Division) Cambridge: Cambridge University Press. and Support Zone." London: World Wide Fund for Rada Research. 1996. "Environmental Economic Nature. Program: Qualitative Research Study." Cairo: Rada Saghir, J., M. Schiffler, and M. Woldu. 2000. "Urban Research. Water and Sanitation in the Middle East and North Rausser G.C., and A.A. Small. 2000. "Valuing Africa Region: The Way Forward." Middle East and Research Leads: Biosprospecting and the North Africa Region, Infrastructure Development Conservation of Genetic Resources." Journal of Group. Washington: World Bank (processed). Political Economy, 108:1, pp. 173-206. Scherr, S., A. White, and D. Kaimowitz. 2001. Regional Environment Center (REC). 2000. "Making Forest Markets Work for Low-Income Environmental Taxes in an Enlarged Europe: An Producers," Policy Brief. Washington: Forest Trends Analysis and Database of Environmental Taxes and and Bogor: Center for International Forestry Charges in Central and Eastern Europe. Ady Endre: Research. REC. Schtlbeler, P., with K. Wehrle and J. Christen. 1996. Regional Environment Center (REC) / Sofia Initiative "Conceptual Framework for Municipal Solid Waste Economic Instruments (SIEI). 2001. "Environmental Management in Low-Income Countries." Taxes: Current Priorities in Central and Eastern UNDP/UNCHS/World Bank/SDC Collaborative Europe." Newsletter on Green Budget Reform, 8. Programme on Municipal Solid Waste Management Website: www.rec.org/siei. in Low-Income Countries. Working Paper No 9. Reid, W.V., S.A. Laird, R. Gamez, A. Sittenfeld, D.H. New York: UNDP. Janzen, M.A. Gollin, and C. Juma. 1993. "A New Shultz, S., J. Pinazzo, and M. Cifuentes. 1998. Lease on Life." In W.V. Reid, S.A. Laird, C.A. "Opportunities and Limitations of Contingent Meyer, R. Gamez, A. Sittenfeld, D.H. Janzen, M.A. Valuation Surveys to Determine National Park Gollin, and C. Juma. Biodiversity Prospecting- Using Entrance Fees: Evidence from Costa Rica." Genetic Resources for Sustainable Development. Environment and Development Economics, 3, pp. Washington: World Resources Institute. 131-149. Rodnick, H. 2000. "Desregulaci6n del Sector Electrico Seagraves, J.A., and K.W. Easter. 1983. "Pricing en America Latina: Exitos y Dificultades." Paper Irrigation Water in Developing Countries." Water presented at the Symposium of the IEEE Power Resources Bulletin, 4, pp. 663-671. Engineering Society, Quito, Ecuador, 22 September Seckler, D., A. Upali, D. Molden, R. De Silva, and R. 2000. Barker. 1998. "World Water Demand and Supply Rosegrant, M.W, M.S. Paisner, S. Meijer, and 1990 to 2025: Scenarios and Issues." Research J.Witcover. 2001. "Global Food Projections to 2020: Report No. 19. Colombo: International Water Emerging Trends and Alternative Futures." Management Institute. Washington: IFPRI. Seymour, F., and N. Dubash. 2000. "The Right Rosenthal, J.P. 1997. "Equitable Sharing of Conditions: The World Bank, Structural Adjustment, Biodiversity Benefits: Agreements on Genetic and Forest Policy Reform." Washington: World Resources." In OECD, Investing in Biological Resource Institute. Diversity: Proceedings of the Cairns Conference. Simpson, R. 2001. "Bioprospecting as a Conservation Paris: OECD. and Development Policy: Overview and Insights Rosenthal, J.P., D. Beck, A. Bhat, J. Biswas, L. Brady, From Three Cases." Paper presented at the K. Bridbord, S. Collins, G. Cragg, J. Edwards, A. International Workshop on Market Creation for Fairfield, M. Gottlieb, L.A. Gschwind, Y. Hallock, Biodiversity Products and Services, Paris, January R. Hawks, R. Hegyeli, G. Johnson, G.T. Keusch, 25-26, 2001. E.E. Lyons, R. Miller, J. Rodman, J. Roskoski, and Simpson, D., R.A. Sedjo, and J.W. Reid. 1996. D. Siegel-Causey. 1999. 'Combining High Risk "Valuing Biodiversity for Use in Pharmaceutical Science With Ambitious Social and Economic Research." Journal of Political Economy, 104:1, pp. Goals.' In J.P. Rosenthal (ed.), Drug Discovery, 163-85 Economic Development and Conservation: The Sizer, N. 2000. "Perverse Habits: The G8 and International Cooperative Biodiversity Groups. Subsidies that Harm Forest and Economies." Forest Special Issue of Pharmaceutical Biology, 37, pp. 6- Notes. Washington: WRI. 21. 100 Generating Public Sector Resources to Finance Sustainable Development Smith, K.R. 1996. "Indoor Air Pollution in Developing Trusts and Endowments." Center for Development Countries: Growing Evidence of its Role in the Information and Evaluation. Washington: USAID. Global Disease Burden." In S. Yoshizawa, K. van Beers, C., and A. de Moor. 2001. Public Subsidies Kimura, K. Ikeda, S. Tanabe, and T. Iwata, (eds.), and Policy Failures. Cheltenham: Edward Elgar Indoor Air '96: Proceedings of the 7th International Vincent, J., and S. Farrow. 1997. "A Survey of Conference on Indoor Air Quality and Climate, Nagoa Jpan21-26 July 1996. Tokyo: Institute of Pollution Charge Systems and Key Issues in Policy Nagoya, Japan, Design." In R. Bluffstone and B.A. Larson eds, Public Health. Controlling Pollution in Transition Economies* Smith, K.R. 1999. "Indoor Air Pollution." Pollution Theories and Methods. Cheltenham: Edward Elgar. Management In Focus Discussion Note No.4. Walker, I., F. Ordonez, P. Serrano, and J. Halpem. Washington: World Bank. 2000. "Pricing, Subsidies and the Poor: Demand for Social Monitoring & Early Response Unit (SMERU). Improved Water Services in Central America." 2001. "Kerosene: Government Subsidies and Policy Research Working Paper No.2468. Household Consumption." The SMERU Research Washington: World Bank. Institute Newsletter, No.2. Jakarta: SMERU. Walpole, M., H. Goodwin, and K. Ward. 2001. Soderholm, P. 1999. "Environmental Policy in "Pricing Policy for Tourism in Protected Areas: Transition Economies: The Effectiveness of Lessons from Komodo National Park, Indonesia." Pollution Charges." Paper presented at the Annual Conservation Biology, 15, pp. 218-227. Meeting of the Association of Evolutionary Wang, H., and D. Wheeler. 1996. "Pricing Industrial Economics, New York, January 2-5, 1999. Pollution in China: An Econometric Analysis of the South Africa Chamber of Business (SACOB). 2000. Levy System." Policy Research Department "List Of Existing and Possible Future Taxes, Fees, Working Paper No.1644. Washington: World Bank. Levies, Charges to be Imposed." Gauteng: South Wang, H., and D. Wheeler. 1999. "Endogenous Africa Chamber of Business (processed). Enforcement and Effectiveness of China's Pollution Speck, S., J. McNicholas, and N. Jackson. 2000. Levy System." Policy Research Department "Energy Taxation and Green Tax Reform in Central Working Paper No.2000. Washington: World Bank. and Eastern Europe." Paper presented at the Wang L. 2002. "Health Outcomes in Poor Countries Conference on Enhancing the Environment by and Policy Options: Empirical Findings from Reforming Energy Prices, Pruhonice, June 14-16, Demographic and Health Surveys." World Bank pe0.2 Working Paper No.2831. Washington: World Bank. Spergel, B. 200 1. "Raising Revenues for Protected WtradSntto rgan(S) 98 Areas: A Menu of Options." Center for Conservation Water and Sanitation Program (WSP). 1998. Areas:nAcMenu Washn tions." Center Wlifor Conseond. "Workshop on Willingness to Pay for Drinking Finance. Washington: World Wildlife Fund. Water Supply and Sanitation: Report." Washington: Tanzi, V., and H. Zee. 2001. "Tax Policy in WSP. Developing Countries." Economic Issues No.27, Water and Sanitation Program (WSP). 2001. "Rapid Washington: IMF. Assessment of Government Subsidy to Rural Water Tsur, Y., and A. Dinar. 1995. "Efficiency and Equity Supply In India." Paper presented at the World Bank Considerations in Pricing and Allocating Irrigation Infrastructure Forum, Washington, May 2-11, 2001. Water." Policy Research Working Paper No.1460. Weaver, D.B. 1999. "Magnitude of Ecotourism in Washington: World Bank. Costa Rica and Kenya." Annals of Tourism United Nations Center for Human Settlements Research, 26:4, pp. 792-816. (UNCHS). 2001. The State ofthe World Cities Whittington, D, D. Lauria, X. Mu. 1991. "A Study of Report. New York: UNCHS. Water Vending and Willingness to Pay for Water in United Nations Environment Programme (UNEP). Onitsha, Nigeria." World Development, 19:2/3, 1996. International Source Book on Environmentally ppS.521-529. Sound Technologies for Municipal Solid Waste World Bank. 1994. "A Review of World Bank Management. New York: UNEP.WolBak194"ARveofordan Management. New York: UNEP. Experience in Irrigation." Operations Evaluation United Nations Environment Programme (UNEP). Department. Report No.13676. Washington: World 2001. "Ecotourism and Sustainability." Industry and Bank. Environment, 24:34, Special Issue. World Bank. 1997a. Expanding the Measure of Wealth. United States Agency for International Development Environmentally Sustainable Development Studies (USAID). 1996. "Win-Win Approaches to Development and the Environment: Environmental Generating Public Sector Resources to Finance Sustainable Development 101 and Monograph Series No.17. Washington: World North Africa Region. Washington: World Bank Bank. (processed) World Bank. 1997b. "Russia: Forest Policy during World Bank. 2000g. Philippines Environment Monitor Transition." Country Study Series. Washington: 2000. Washington, DC: The World Bank. World Bank. World Bank. 2001 a. "Global Development Finance World Bank. 1998a. "Electricity Tariffs in Southern 2001," Washington: World Bank. Africa: March 1998." Washington: World Bank. World Bank. 200 lb. "Urban Waste Management." (processed). Washington: World Bank (processed). World Bank. 1998b. "India - Water Resources World Bank. 2001 c. Making Sustainable Management Sector Review." Report No.18416-IN. Commitments An Environment Strategy for the Washington: World Bank. World Bank. Washington: World Bank. World Bank. 1 999a. "Meeting India's Future Power World Bank. 2001 d. World Development Indicators Needs: Planning for Environmentally Sustainable 2001. Washington: World Bank. Development." Washington: World Bank World Bank. 2002a. The Environment and the (processed). Millennium Development Goals. Washington: World World Bank. 1999b. "Electricity Price Comparison." Bank. Washington: World Bank (processed) World Bank. 2002b. "Irrigation in Central Asia: Where World Bank. 1999c. "What a Waste: Solid Waste to Rehabilitate and Why." Washington: World Bank Management in Asia." Washington: World Bank (processed). (processed). World Bank/Sida/Government of Finland. 2001. "LAO World Bank. 2000a. "Access to Electricity in PDR. Production Forestry Policy, Status and Issues Developing Countries." Poverty Thematic Group. for Dialogue." Vol. I Main Report. Washington: Washington: World Bank (processed) World Bank/Sida/Ministry of Foreign Affairs, World Bank. 2000b. "Brazil-Forests in Balance: Government of Finland. Challenges of Conservation with Development." World Health Organization (WHO) and United Nations Washington: World Bank. Children's Fund (UNICEF). 2000. Global Water World Bank. 2000c. "The Challenges of World Bank Supply and Sanitation Assessment 2000 Report. Involvement in Forests: Evaluation Country Case Delft: IRC International Water and Sanitation Study Series." Washington: World Bank. Centre. World Bank. 2000d. Greening Industry: New Roles for Yang, J., D. Cao, and J. Wang. 1997. "The Air Communities, Markets, and Governments. Policy Pollution Charge System in China: Practice and Research Report. Washington: World Bank. Reform." in OECD, Applying Market-Based World Bank. 2000e. "Indonesia-The Challenges of Instruments to Environmental Policies in China and World Bank Involvement in Forests: Evaluation OECD Countries. Paris: OECD. Country Case Study Series." Washington: World Yun, P. 1998. "The Pollution Charge System in China: Bank. An Economic Incentive?" Economy and World Bank. 2000f. "MNA Energy Strategy Note." Environment Program for South East Asia Infrastructure Development Group, Middle East and (EEPSEA) Research Report. Ottawa: International Development Research Center. Appendix A. International Financial Flows This Appendix looks at how the landscape of financial mechanisms used in the 1990s, such as development financing for environmental purpo- debt-for nature swaps and environmental funds. ses has changed since the United Nations Conference on Environment and Development A.1 International Financial Flows to (UNCED) in 1992 in Rio de Janeiro, Brazil. The central issue of the Rio Conference concerned mechanisms for financing sustainable develop- Foreign aid is intended to transfer resources, ment. The Conference also recognized the need beyond those the recipient country can mobilize for global cooperation, given increasing economic either domestically or through trade, for the pro- globalization and simultaneous degradation of motion of economic development. Official public global environmental public goods. Five resolu- sector flows to developing countries from bilateral tions were passed at Rio, which represented the and multilateral institutions fall into three cate- most universal and coordinated policy response at gories: official development assistance (ODA) are that time to worldwide concerns about the en- loans and grants to the poorest countries, official vironmental dimensions of development.1 assistance (OA) are loans and grants to middle Financial resources to address environmental income countries, and other financial flows (OFF) problems are limited in developing countries, as are items such as export credits and loan government budgets are generally severely con- insurance. Figure A. 1 provides an overview of the strained by social and economic demands. Thus, sources of public and private financial flows to one of the commitments made by developed coun- developing countries. The primary focus of this tries at the Rio Conference was to increase the paper will be on flows of ODA targeted for amount of their official development assistance environmental projects. relative to their GDP-a reaffirmation to fulfill Statistics on the flow of aid to developing the target of 0.7 percent of GDP set by the United countries are collected by the Development Nations in 1970. In Agenda 21, it was agreed that Assistance Committee (DAC) of the OECD "official development assistance ought to be a through two reporting systems: the annual aggre- major source of external financing for developing gated DAC statistics and the activity-specific countries, particularly the most vulnerable among Creditor Reporting System (CRS). DAC members them." However, five years later an appraisal of now provide over 95 percent of all ODA.2 The progress made since the Rio Conference at the DAC statistics provide an overall picture of the nineteenth special session of the General geographic distribution or purpose of aid, and the Assembly concluded that little headway had been relative importance of each recipient country, re- made with regard to the Rio resolutions (OECD, gion, or purpose in the total. The CRS statistics 2000b). Ten years later, we reach the same con- permit examination of the geographical and pur- clusions by examining available data on interna- pose breakdown simultaneously (OECD, 2000b). tional development assistance. Figure A.2 shows total commitments of In section A. l, we examine trends in the total official development financing to developing flow of financial resources from OECD countries countries from bilateral and multilateral sources to developing countries. Section A.2 discusses the over the last decade.3 The amount of money extent to which these flows support environmental flowing from rich to poor countries has remained projects. In section A.3, we focus on specific relatively constant, despite commitments made by instruments such as the Global Environmental developed countries to increase assistance to Facility (GEF) and the contribution of Non- developing countries. When comparing the ratio Governmental Organizations (NGOs) in mobili- of ODA to GDP, we find that flows of ODA have zing resources for environmental causes in deve- yet to reach the 0.7 percent of GDP target.4 Dis- loping countries. We also discuss some of the new aggregated data from DAC statistics (not shown 103 104 Generating Public Sector Resources to Finance Sustainable Development Figure A.1: Overview of public and private financial flows Sources of Development Finance in Developing Countries Domestic 1ntemational (public + private) | Pnvate Finance l Multilateral and Bilateral Private Finane l Concessionary Finance Equity | DetLending Other Foreign Bank Grants Loans (export Portfolio Direct Port- and credits, loans equity lInves- folio trade insurance, tment debt lending etc.) Capital Market Flows ODA/OA OFF Project Finance combines different sources and types of capital to finance a project activity (e.g., build a power plant, road, or water treatment plant). Sources: World Bank, 2001 a, and OECD DAC-Online 2001. below) indicate that total ODA declined from 0.33 this period, official development financing, of percent of donor countries GDP in 1992 to 0.22 which ODA is the largest component, remained percent in 1997. This was followed by a slight in- more or less at the same level in absolute terms. crease in the years immediately after, which The environmental implications of the in- brought the figure up to 0.24 percent in 1999, only crease in private sector flows are difficult to to drop once again to 0.22 percent in 2000. If assess since there are no reporting mechanisms for ODA flows had remained at the peak level of 0.33 this type of activity. It is also much harder to in- percent of GDP, there would have been an addi- fluence private investment to take into account tional flow of US$20 billion a year to developing environmental concers, given the public nature countries. of environmental variables. Preliminary evidence ODA is only one component of the total suggests that most FDI flows have been directed amount of resources flowing to developing coun- at natural resource-extraction industries, such as tries. Table A.1 below shows the total official petroleum and minerals (World Bank, 2001a). public and private flows to developing countries Another concern is that most private resource over the last decade. In the 1990s, private flows to flows have bypassed most poor developing coun- developing countries, dominated by foreign di- tries. The top 10 recipients received approxi- rect investment (FDI), increased considerably. mately 74 percent of FDI flows going to develo- FDI accounted for 18 percent of all resources ping countries.5 While total FDI flows have grown flowing to developing countries in the early steadily over the last decade, the share of FDI 1990s, but by 1998 it increased to almost half of received by developing countries has actually resources flowing to developing countries. During declined over the same period (Donge and others, Generating Public Sector Resources to Finance Sustainable Development 105 Figure A.2: Total official development financing to developing countries 100 |C Bilateral m Multilaterl 80 ----. _ _ O 60 f 20 40- 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Sources: OECD International Statistics online, CRS ODA/OA Commitments. 2001). Therefore, flows of public funds as official measures to prevent further environmental degra- development financing will remain an important dation; and (3) apply environmental impact instrument in addressing poverty/development assessments for activities that have a significant concerns and environmental outcomes in the pro- impact on the environment.6 In order to achieve cess of development. these objectives and implement the Rio Conven- In general, there has been little change in the tions on climate change, biodiversity conser- total amount of public aid flowing to developing vation, and to combat desertification, the UNCED countries during this last decade. Most developed Secretariat estimated that an additional annual countries have failed to keep their promise to transfer of resources amounting to approximately increase their contributions as a percentage of US$125 billion would be necessary (OECD, their GDP. We also find that official development 2000a). While it is clear that the level of finan- financing to developing countries has decreased in cing for environmental purposes is nowhere near importance relative to private sector financing. the goal established, it is important to assess what The implication of this development for environ- is the current situation. mental concerns is uncertain due to lack of infor- Official development financing for the mation on the purpose of private flows. The rest environment over time of this Appendix looks more closely at public resource flows to developing countries and exa- Financial data collected by OECD allows us mines the environmental impact of these flows. to examine environmentally targeted bilateral commitments.8 Environmentally targeted financial A.2 Environmental Conservation flows are defined as activities consistent with:9 (1) Financing the prevention or control of the release of sub- stances with negative consequences on the health Developed countries provide loans and grants and habitat of living organisms, (2) the reduction exclusively for environmental purposes as part of of consumption or demand for non-renewable their ODA. Three formal commitments made natural resources, and, (3) the protection of eco- during the Rio Conference aimed to: (1) make systems, endangered wildlife, or global environ- environmental protection an integral part of the mental commons (Donge and others, 2001). There development process, not something to be consi- are no accepted standards on how to classify aid dered apart from it; (2) apply the 'precautionary flows that support environmental protection and principle' in the face of serious threats or irre- natural resource management. We use the World versible damage and not use the lack of scientific Bank's classification of environmental projects evidence as a reason to postpone cost-effective within its lending portfolio, as well as a broader 106 Generating Public Sector Resources to Finance Sustainable Development Table A.1: Total flows of net resources from OECD DAC member countries 1991 1995 1999 1991 1995 1999 (current US$ billions) (% of total) A. Official Development Financing 84.5 87.6 85.9 61.2 33.2 34.2 1. Official development assistancea 57.1 59.1 52.1 41.4 22.4 20.7 Bilateral 41.1 40.6 37.9 30.0 15.3 15.1 Multilateral 15.8 18.4 14.2 11.4 6.9 5.6 2. Official Assistance 6.6 8.4 7.8 4.8 3.2 3.1 Bilateral 5.0 7.1 4.9 3.6 2.7 1.9 Multilateral 1.6 1.3 2.9 1.1 0.5 1.2 3. Othertypesofofficial financing 20.8 20.1 26.1 15.1 7.6 10.3 Bilateral 13.1 14.0 10.4 9.5 5.3 4.1 Multilateral 7.7 6.1 15.6 5.6 2.3 6.2 B. Total Export Credit 0.6 5.6 4.0 0.4 2.1 1.6 C. Private Flows 53.0 170.7 161.1 38.4 64.7 64.2 1. Direct Investment (DAC) 24.8 59.6 145.6 18.0 22.6 58.0 2. Offshore centers 6.5 6.3 37.9 4.7 2.4 8.5 3. International bank loansb 10.7 76.9 -79.6 7.7 29.1 -31.7 4. Total loans in bonds 28.7 24.7 28.8 3.5 9.4 11.4 5. Other types of loans (including equity)C 7.1 3.5 59.5 5.2 1.3 23.7 6. Subsidies from non-governnental 5.4 6.0 6.7 3.9 2.3 2.7 organizations Total flows of net resources (A+B+C) 138.1 263.8 251.0 100 100 100 Notes: a Does not include non-official debt cancellation for 1991 or 1992. b Does not include bank loans (included in C) or guaranteed financial credits (which are included in B) c Incomplete reports for several DAC countries; includes Japan from 1996 on. Sources: OECD, 2000b. classification suggested by the World Resources development financing commitments for environ- Institute (WRI) (Donge and others, 2001).1o We mental purposes following the 1992 Rio Conven- use OECD data available from the CRS database tion. Environmental commitments reached peak rather than the DAC statistics database because levels in 1996 at US$1.8 billion (World Bank the former allows more detailed screening accor- screen) and US$6.9 billion (WRI screen). How- ding to the identified purpose of the project sup- ever, under both screens, this increase is almost or ported with official development flows."l totally reversed by the end of the decade. As a Figure A.3 shows the total amount of percentage of total commitments, under the World environmentally targeted bilateral official Bank definition, we see environmentally targeted development financing commitments using two aid has stayed relatively constant, averaging definitions. The figure also shows these screened around 3 percent of total official development values as a percentage of the total official deve- financing commitments. The broader WRI defi- lopment financing commitments. As the Figure nition fluctuates between 9 and 18 percent, ave- shows, there was only a slight increase in official raging about 13 percent of total commitments. Generating Public Sector Resources to Finance Sustainable Development 107 Figure A.3: Environmentally targeted bilateral commitments 20%- O0 -_WB % -U-WRI % ,.!E 1 5 % - __J .....;... ---------------.--------- ujE cE o 810% 8 6-~~~~~~~~~~~~~~~~~~~~~~~------- ------... .2 v 2 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - o 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Source* OECD International Statistics online, CRS ODA/OA Commitments: Form I-All details. Figure A.4: Environmentally targeted multilateral commitments 'B .l 25% -B-eIWB % -U-WRI % X E 20% - -------- ---;.,----------------------------------------- ---------------------- -------- cE 0E 5% - L,o 5%) 0- -- ------------ - -- --.-- -- -. Cu CoI 0% 8 - | _-WB screen 6 2 - ---------- .2 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Source: OECD International Statistics online, CRS ODA/OA Commitments: Formn I-All details. A slightly different pattern emerges when we multilateral financing commitments toward the examine multilateral commitments towards the environment amounted to almost US$2 billion and environment (Figure A.4).12 In this case, environ- nearly US$7 billion under the two screens. As a mentally targeted multilateral commitments saw a percentage of total commitments, under the World sharp decline in 1993, as did total commitments, Bank definition, multilateral aid has stayed relati- and have been on a somewhat declining trend vely constant just below five percent. Under the after rebounding in 1994. At the peak in 1994, WRI definition, multilateral aid has fluctuated a 108 Generating Public Sector Resources to Finance Sustainable Development Figure A.5: Total environmenal ODA, 1998 4.0 - _ _ _-_-__-_ _- _ __ O Other * Desertficaton 3.0 - .-------------------------------0 Climate change --------------------- ------ 0 Biodiversity i 0 * 2.0 L .=L r- - -I. 0.0 - _ ,___I __, .. .,i ....___ Agriculture Energy Fisheres Forestry Transport Water General Environment Protecton Source: OECD, 2000a. bit more, hovering around 15 percent over the last Official Development Assistance few years. targeting the Rio Conventions Interpretation of the patterns discussed above need to take into account existing difficulties in The information in Figure A.5 provides a classifying financial flows into specific categories snapshot of the relative distribution of ODA in and the lack of an acceptable definition on how to 1998 for different environment related activities, use existing classification schemes to reflect envi- indicating their sectoral distribution and the pro- ronentl cnsieraion. Hwevr, n gnerl,it portion that was specifically* targeted at meeting rodnental considerations. However, in general, the goals of the Rio Conventions.14 The transport is hard to escape the conclusion that there has sector was the largest recipient of funds (over been little increase in bilateral or multilateral US$3.5 billion), followed by the energy and the financing of the enviromnental sector following wtrscos hc eevdaotU$. the commitments made at Rio.'3 water sectors, which received about US$2.5 billion each. Of these totals, only a relatively Bilateral flows to specific small amount was specifically targeted at meeting environmental problems and regions the goals of the Rio Conventions, although some of the other expenditures undoubtedly also contri- In this section, we examine a pilot study btd carried out by the OECD to measure the extent to buted. wvhich official development assistance contributed The study found 313 operations for climate- to the different specific agreements made in Rio change related activities in 1998, representing a (OECD, 2000a). This allows us to examine how total value of US$1.8 billion.'5 Three-quarters of total environmental assistance was distributed bet- this assistance went to Asia, and most was allo- ween different sub-categories. We also present re- cated to the energy and transportation sectors. The sults from study that identified investments over US$778 million used for biodiversity related acti- the last decade in support of one the key Rio con- vities is spread across the agriculture, forestry, ventions (biodiversity) in the Latin America and water, and general environmental protection sec- the Caribbean region. While both studies provide tors. Here too the main recipient of assistance (50 only a partial picture of how financial flows have percent) was Asia. Africa, on the other hand, targeted toward Rio goals, examining this data received half of the US$562 million in ODA allows us to get a better understanding about what devoted to activities to combat desertification. regions and sectors received funds. These funds have been allocated primarily to- Generating Public Sector Resources to Finance Sustainable Development 109 wards improving agricultural practices, and to a Multilateral and bilateral agencies shared lesser extent to management of water resources.'6 almost equally the bulk of the financing of Investmentnbiodiversity in Latin biodiversity projects (see Figure A.6). Thirteen InAestmentca ind theCaribbprincipal financing sources accounted for 77 America and the Caribbean percent of total funding, including the World Substantial investments have been made to Bank (17 percent), the Inter-American Develop- forestall the deterioration of biodiversity in the ment Bank (IDB) (11 percent), the German Latin America and Caribbean region. The United Corporation for Technical Cooperation (GTZ) (9 States Agency for International Development percent), USAID (6 percent), and the GEF (6 (USAID), the World Bank, and the Biodiversity percent). Projects relating to management of natu- Support Program designed a survey to compile ral resources and of protected areas accounted for information on funding available for biodiversity over 70 percent of financing. Of the projects that conservation at the project level (Castro and could be classified by eco region, tropical and Locker, 2000).17 The results show that a cumula- subtropical broadleaved forests accounted for 66 tive total of US$3.3 billion of financing was pro- percent of funding (Castro and Locker, 2000). vided between 1990 and 1997 for 3,489 conserva- Although there is no doubt that some headway has tion projects. Brazil and Mexico received the most been made on the issue of biodiversity conserva- funds, together accounting for 45.5 percent of tion in the last two decades, this has been insuffi- funds. The countries to receive most funding per cient, as threats to biodiversity have grown and square kilometer, however, were Venezuela, the become more complex. Central American countries, and the Dominican The analysis in this section shows that the Republic, Ecuador, Haiti, and Jamaica. The south- amount of official development financing target- em cone countries and Cuba received the smallest ing environmental purposes is only a small share amounts. of total financial flows to developing countries (generally between 5 and 15 percent) and has re- Figure A.6: Sources of financing for mained relatively constant over the last decade. biodiversity conservation in Latin America, This is despite the greater awareness of environ- 1990-97 mental and natural resource management pro- blems and commitments made since the Rio Con- Other ference. The analyses of financial flows for 3,000 - (US$46 million) environmental purposes is severely constrained by Foundations lack of specific data and consensus over defini- 2,500 (US$107 tions of what can be considered "environmental 2,500 - NGOs financing." We clearly need more in-depth (US$164 million) studies, such as the one carried out for biodi- 2,000 Bilaterals versity investments in Latin America and the (US$1,160 million) Caribbean, to fully assess the extent of financial 1,500 support for environmental programs. co 1,500- A.3 New mechanisms for financing 1,ooo - | li!environmental conservation Multilaterals (US$1,340 million) The Rio Conventions called for 'new and 500- additional' funds for protecting the global environment. For example, the Convention on 0 Climate Change stipulated that the developed countries would provide developing countries with the "new and additional" financial resources Source: Castro and Locker, 2000. necessary to meet the needs arising from the implementation of the Convention. This guarantee 110 Generating Public Sector Resources to Finance Sustainable Development was one of the decisive factors in having projects that address land degradation problems developing countries agree, at least in principle, to that fall into above focal areas. the process of preventing the risks threatening the Support for the GEF included a US$2 billion global environment. It is difficult, however, to pledge by 34 nations in 1994, followed by another measure the creation of "new and additional" pledge of US$2.75 billion by 36 nations in 1996. resources over and above commitments of official Negotiations are in place for a third replenishment development aid, given that many countries are of the fund in 2002. From 1991 to 1999, the GEF not even fulfilling their pledge to target 0.7 allocated US$991 million to biodiversity projects, percent of their GDP to ODA. Box A. 1 discusses US$884 million to climate change projects, some of the difficulties involved in this process of US$360 to international waters, US$155 million accounting for these two separate commitments. to protection of the ozone layer, and US$350 Box_A_1__The_French_GEF_and_ODA:_Separate but million to combat soil degradation. The GEF Box A.l: The French GEF and ODA: Separate but finances on average 29 percent of the total costs of country level projects, with the percentage The French GEF (Fonds FranVais pour varying by region, and 36 percent of global pro- I'Environnement Mondial, FFEM) was established in jects. Therefore, in addition to GEF allocations, 1994 and has a portfolio of 99 projects. US$67 million mobilization of resources in the amount of US$4.7 was allocated to the FFEM for the 1994-1998 period, billion for climate change and US$1.5 billion for and the fund was replenished with an identical amount for the 1999-2002 period. As far as France is biodiversity-related projects took place in co- concerned, the spirit of the Rio undertaking is adhered financed projects (Barcena and others, 2001). to since these "new and additional" resources have GEF funds in biodiversity-related projects have been made the subject of a specific fund, a specific pro- been particularly important in the establishment of cedure, and specific institutions. The FFEM has conservation trust funds, as will be discussed deliberately been kept separate from the development below. aid circuits. There are three reasons why it is difficult to Debt for nature swaps evaluate the application of the principle of additional NGOs pioneered the approach of exchanging resources. First, resources allocated to protecting the external debt from developing countries for global environment are supposed to contribute to environmental purposes. The first such 'debt for development projects rather than cover the costs of nature' swap was conducted for Bolivia by purely environmental projects. As a result, the "visibility" of these resources tends to fade away at the Conservation International (CI) in 1987. This field level. Second, in certain countries the new and approach was quickly followed by similar initia- additional resources may follow the same channels and tives negotiated by the National Parks Foundation be governed by the same institutions as the of Costa Rica and World Wildlife Fund (WWF) development aid funds. This was one of the reasons for Ecuador. As a mechanism, debt for nature why France developed the FFEM. swaps flourished during the debt crisis of the late 1980s and early 1990s, facilitating debt can- The Global Environmental Facility cellation in Latin American countries amounting The Global Environment Facility (GEF) has to a nominal figure of over US$90 million been a significant source of new funding for the (Deacon and Murphy, 1997). These transactions environmental projects in developing countries. have continued to be conducted and are estimated The GEF was created in 1991 as an experimental to have leveraged some US$1 billion to date for facility and was restructured after the Rio conservation purposes (Barcena and others, 2001). Conference to act as the implementing financial While the pace of such transaction declined in the mechanism for the Convention on Biological later half of the nineties, recent funding Diversity and the United Nations Framework opportunities created by the United States may Convention on Climate Change. The GEF area of result in a new wave of debt for nature swaps. focus also includes international waters and ozone In a debt for nature swap, the debtor govern- depletion. GEF funds can also allocated to eligible ment commits to mobilizing domestic resources in local currency for an agreed environmental Generating Public Sector Resources to Finance Sustainable Development 111 Table A.2: Debt for nature swaps, 1990-1997 (US$ millions) Cost to Nominal value of Conservation Year Country Purchaser or donor donor debt swapped funds 1993-97 Mexicoa Ci 3.1 3.80 3.43 1994 Madagascarb WWF 0.0 1.34 1.07 1993 Madagascar WWF 0.9 1.87 1.87 1993 Philippines WWF 13.0 19.00 17.70 1992 Boliviab TNC, WWF, CI 0.0 11.50 2.76 1992 Panama CI,TNC 7.5 30.00 30.00 1992 Brazil CI, TNC 0.8 2.20 2.20 1992 Chile EAI n.a. 15.90 1.40 1992 Guatemala CI/USAID 1.2 1.33 1.33 1992 Philippinesc WWF 5.0 9.85 8.82 1992 Mexico CI/USAID 0.4 0.44 0.44 1991 Guatemalad TNC 0.1 0.10 0.09 1991 Jamaica TNC, USAID 0.3 0.44 0.44 1991 Jamaica EAI n.a. 271.00 9.20 1991 Bolivia EAI n.a. 38.40 1.80 1991 Nigeria Nigerian Conservation Fund 0.1 0.15 0.09 1991 Ghana' DDC, CI, Smithsonian Institution 0.3 1.00 1.00 1991 Mexicod,f CI 0.0 0.25 0.25 1991 . Mexicof CI 0.2 0.25 0.25 1991 Mandagascar5 CI 0.1 0.12 0.12 1991 Costa Ricah RFA, MCL, TNC 0.4 0.60 0.54 1990 Costa Rica' Sweden, WWF, TNC 1.9 10.75 9.60 1990 Dominican Conservation Trust of Puerto 0.1 0.58 0.58 Republic Rico, TNC 1990 Madagascar WWF 0.5 0.92 0.92 1990 Poland WWF 0.0 0.05 0.05 1990 Philippines WWF 0.4 0.90 0.90 Notes: Cl: Conservation International; DDC- Debt for Development Coalition; EAI: Enterprise for the Americas Initiative; MCL: Monteverde Conservation League; RFA: Rain Forest Alliance; USAID: US Agency for International Development; WWF: World Wide Fund for Nature; TNC: The Nature Conservancy; n.a.: not applicable. a Figures for Mexico aggregated from 10 separate transactions for the time period. b Debt donated by JP Morgan. 'Nominal value of debt includes US$0.2 million debt donation by Bank of Tokyo. d Debt donated by Bank of America. eInvolves buying blocked local currency funds from multinational. Also includes Midwest Universities Consortium for International Activities and U.S. Committee of the International Council of Monuments and Sites. f Total amount of program is US$4 million g Total amount of program is US$5 million. h Purchase of Central American Bank for Economic Integration Debt. 'WWF contributed US$ 1.5 million to this deal on top of the swap. Sources: Barcena and others, 2001; Kaiser and Lambert, 1996. purpose in exchange for cancellation of external plishes the twin goals of reducing the country's hard-currency loan payments. The swap accom- external debt and increasing the amount of 112 Generating Public Sector Resources to Finance Sustainable Development resources available for environmental conser- Debt for nature swaps have been reasonably vation. These agreements can be negotiated bilate- effective tools for financing conservation pro- rally between debtor and creditor govermnuents or, grams in developing countries and emerging eco- as is more often the case, through an intermediary nomies. While they have so far made little dent in such as a local or international environmental con- the trillions of dollars of debt owed by these servation organization. The intermediary organi- countries, they offer a tested mechanism for inte- zation often contributes its own funds to acquire grating sustainable development into future debt part of the debt, which is purchased in secondary reduction efforts. markets at a discounted rate. Conservation trust funds More than 19 countries throughout the world have been involved in debt for nature swaps and Conservation trust funds have emerged in their number continues to grow (Deacon and several developing countries as a mechanism for Murphy, 1997). Table A.2 provides some details financing environmental programs in a sustainable on debt swaps carried out from 1990 to 1997. As manner. These funds have been established in the Table shows, debt for nature exchanges in the more than 40 developing countries over the last 1990s were dominated by Latin American coun- decade for the purposes of protecting either a spe- tries. cific area or the country's entire protected area system, conservation of a particular species, or The large number of debt for nature swaps In fiacn sml grnt tolclcomnte Latin America is attributable to the Enterprise for fiancig small grants to local communities the Americas Initiative (EAI) launched by the carrying out conservation projects, among other United States in 1990. This scheme resulted in a things. Several of the debt for nature swaps listed partial reduction in debt owed to the United States above, have been used to set up a conservation partal edutionin ebtowedto he nite Sttes trust fund to manage or establish protected areas. for Latin American countries that subscribed to the Brady Plan. A portiou of the capital debt was The GEF has been a major source of capital cancelled and the interest, paid in local currency, funds for many existing conservation trust funds, was allocated to a trust for environmental pro- particularly "Park" funds that set up a national tection and sustainable development. Most natio- park or finance existing park operations. NGOs nal environmental funds created in Latin Ameri- have made financial contributions to establish can (including funds in Argentina, Bolivia, Chile, conservation trust funds, in addition to playing an Colombia, El Salvador, Jamaica, and Uruguay) important role in mobilizing donations from other derived from this proposal. The EAI cancelled donors. NGOs have also made substantial in kind US$876 million of US debt and generated some contributions by providing technical expertise US$650 million in project support for Latin (legal assistance in setting up and managing funds America (Bayon and others, 1999). established) and paying for the staff and travel costs incurred to hold negotiation workshops. The pace of debt for nature swaps slowed in the mid 1990s (except for several deals carried out The most common type of a conservation by CI in Mexico). More recently, debt for nature trust fund is the endowment fiud. The initial swaps have regained appeal with the passage of contribution to the fund-the capital or principal the United States Tropical Forest Act in 1998. amount-is invested in a variety of financial This Act extended the EAI debt for nature swap to instruments, such as bank deposits, government the protection of highly significant tropical forests treasury bonds, and corporate stocks and bonds. located in developing countries through out the The investments generate a flow of annual interest world. This legislation was due to expire in 2002, earnings to fund activities specified by fund. One but was extended until 2004. The Program, with a of the main advantages of endowment funds is budget of US$50 million for 2002, US$75 million that they provide sustained long-term funding for for 2003, and US$100 million for 2004, is likely a specific purpose, such as the maintenance of to result in a new wave of nature-debt swaps parks or protected areas. Some conservation funds (Barcena and Acquatella, 2001; Bayon and others, have been created to accomplish a specific goal 1999). and are therefore managed as a sinkingfund. This type of fund spends the income generated, as well Generating Public Sector Resources to Finance Sustainable Development 113 as a portion of their capital each year. A revolving financing (ODA, OA, and OFF) from bilateral and fimd, on the other hand, receives revenues from multilateral sources has averaged around US$85 earmarked taxes or user fees and in general spends billion per year over the last decade. On the other all of their revenue in a given year. Sometimes hand, private sector financing has increased signi- these funds set aside a small percentage of ficantly, going from around US$53 billion in the revenues generated to create a reserve fund that early 1990s to approximately US$160 billion by may be tapped in case of an economic downturn the end of the decade. Private financial resources, or sudden political events. Belize's Protected however, tend to be concentrated in a few Areas Conservation Trust Fund, financed by a countries and sectors and cannot be viewed as US$3.75 airport tax generates about US$600,000 substitutes for public financial aid in the each year, 5 percent of which goes into a reserve development process. fund (Spergel, 2001). The analysis of official development Experience with conservation trust funds has financial flows targeting environmental concerns been limited. Early evaluations of the funds in shows that these flows have more or less followed existence have identified some issue that deserve the stagnant pattern of overall development finan- attention in monitoring the future performance cing. Commitment to environmental concerns in and the impact of conservation trust funds (GEF bilateral aid averaged at 3 percent of total com- 1998; USAID, 1996). mitments under the World Bank classification of * Governance. Most funds have a mixed public- environmental projects. Using the broader WRI private control board, which usually is set up definition, environmental commitments fluctuate as a non-governmental institution. It is impor- between 9 and 18 percent, averaging about 13 tant to ensure that the governing board is free percent of total commitments. Multilateral from political influences and that has usually environmental aid averaged between 3 and 15 per- meant governments have one or two repre- cent of total commitments, depending on the defi- sentatives and do not have the majority vote, nition used. On average, bilateral and multilateral sentativs Mandgemenothavethe "Pajork"ifund uy venvironmental aid each ranged between US$1-5 * Program Management. "Park" funds usually billion per year. Some disagreement over the defi- have a clearer mandate of their mission than "grant" funds, which need to find their niches nitions applied to categorize projects as environ- to fulfill the fund's established objective. A mental is possible, making the determination of fund's effectiveness in achieving its objective the exact dollar amount of support debatable. will depend upon how well the fund's However, no matter what screen is used, the objectives intersect with ongoing government message is clear-financial support to the envi- agencies' efforts in a given situation. ronment is not consistent with the pledges made ageinancies' effortsanagient. siFuationd difby developed countries to support the implemen- * Financial Management. Funds differ widely tation of the Rio Conventions. in the amount of initial endowment that they receive and how they are structured. Some An in-depth analysis of financial flows for a funds have been innovative in their manage- single year (1998) is indicative of how bilateral mentstrategy, fundraisingto reinvest operating aid is distributed among different parts of the expenses to grow the initial endowment, seek- developing world and the three Rio Conventions ing new sources of earmarked revenue from -Climate Change, Biodiversity, and Desertifica- government sources, etc. A strong commit- tion. The data that are available show that climate ment to continued fundraising is often nece- change received 1.5 times more resources than ment to ensure long-term financial stability. biodiversity and twice as much resources as ssary to ensure long-term financial stability. desertification. Asia was the major recipient of A.4 Conclusions climate change and biodiversity resources, while most of the desertification related assistance went Official development financing has not to Africa changed significantly since the Rio Conference, While flows of official development despite commitments made by the developed financing have been the focus of attention in this countries to increase them. Official development chapter, we also reviewed the use of other 114 Generating Public Sector Resources to Finance Sustainable Development innovative mechanisms over the past decade to finance environmental objectives. NGOs pio- Luxembourg, the Netherlands, New Zealand, neered the use of debt for nature swaps in the late Norway, Portugal, Spain, Sweden, Switzerland, the 1980s and early 1990s. They have also been United Kingdom, the United States and the heavily involved in raising funds and providing Commission of the European Communities, technical expertise in the setting up of many together with the International Monetary Fund, the environmental/conservation trust funds. The GEF United Nations Development Programme (UNDP), envronentl/onsrvaio trust- and the World Bank as permanent observers. had also played a major role in leveraging funds Thes fgres as comitmentos,rnor ne and upprtin th imlemetaton o th Ri These figures show commitmnents, not net and supporting the implementation of the Rio disbursements, which are presented in Table A. 1. Conlventionls.4 C There are marked differences between donor In March, 2002, heads of state meeting at the countries in this respect: in 2000 ODA ranged from [nternational Conference on Financing for 1.06 percent of GDP by Denmark, and over 0.8 Development in Monterrey, Mexico, adopted the percent by the Netherlands, Sweden, and Norway, "Monterrey Consensus", a commitment by nations to allocations of less than 0.2 percent by Greece and to provide the means to attack poverty worldwide. Italy, and just 0.1 percent by the United States. All The agreement set guiding principles for develop- DAC members except Switzerland and the United ment finance, stressing the need to combine aid States have accepted the 0.7 percent of GDP target. with sound economic policies and freer trade. The 5 The top ten recipients are China, Brazil, Mexico, Conference also saw increased aid commitments Argentina, Malaysia, Poland, Chile, South Korea, by major donors, with EU countries committing Thailand, and Venezuela. themselves to reach an average ODA equivalent 6 These commitments are embodied in Principles 4, to 0.39 percent of national output by 2006, with 15, and 17, respectively, of the Rio Declaration on individual countries reaching at least 0.33 percent the Environment and Development. and the US pledging to increase its ODA by 50 7 It is far from clear how the UNCED Secretariat percent. These commitments would represent an arrived at this figure, which surpasses the total increase in ODA flows of at least US$12 billion a yearly flow of official development financing to year by 2006 (an additional US$7 billion from the developing countries. EU, and about US$5 billion from the US). These s Data for bilateral official development financing increases will provide an important additional includes only DAC member countries and consists resources, but will not by themselves be sufficient primarily of ODA and OA commitments for most to meet the fancing needs f sustainable countries. Only for the United Kingdom are OFFa to meet the financing needs of sustainable significant source of financing, about 10 percent of development. total aid since the mid 1990's. 9 The analysis is limited by the way the information is classified in the OECD International Statistics Notes database. While the information available suggests _"general environmental protection" assistance has 1 The five resolutions agreed to at the Rio Conference grown since 1990, it is difficult to gauge the full are: the Rio Declaration on Environment and amount of assistance for the environment using this Development, Agenda 21, the Forest Principles, the category because it only includes activities without UnitekConvention on Climate sector allocation. An alternative category, the United Nations Framework on olimal "memorandum items" allow classification of aid Change and the Convention on Biological acrigt oiyojcie,sc sgne Diversity. Other significant multilateral instruments according to policy objectives, such as gender were agreed upon following the summit, such as the equality, aid to the environment, and direct United Nations Convention to Combat assistance to poor people. Under this classification Desertification, the Kyoto Protocol and the both sectoral projects (such as those classified Cartagena Protocol, and the Global Programme of under energy generation and supply or water supply Action for the Prevention of Marine Pollution from and sanitation) as well as "general environmental LActindbs Sourt es. Prevention ofMarinePollutionfromprotection" can be classified as memo items for "aid to the environment." However, reporting by most 2 The DAC members are: Australia, Austria, countries comply only with the sectoral classifi- Belgium, Canada, Denmark, Finland, France, cation of aid and only scattered data is available Germany, Greece, Ireland, Italy, Japan, under memo items. Generating Public Sector Resources to Finance Sustainable Development 115 '° The World Bank classification is broader than the 15 Data provided by 15 donor countries (Germany, OECD memo items "aid to environment" category, Australia, Austria, Belgium, Canada, Denmark, since it takes into account projects in the energy and Spain, Finland, France, Japan, Netherlands, forestry sector with clear environmental compo- Sweden, Switzerland, United Kingdom, and United nents. This definition is a close approximation of States). the how the World Bank classifies its "core envi- 16 It should be emphasized that this study only ronmental portfolio." The classification suggested analyzed flows of bilateral ODA, which represented by WRI is broader, incorporating projects in the approximately 40 percent of official development water and sanitation, energy, and agriculture (inclu- financing in 1998. ding sub-sectors such as forestry and fishery), 17 The survey was distributed to 118 leading donor among~~~~~~~~~~~~~~~ ~ othe sectorsdstiutdto18.edigdoo among other sectors. organizations and did not included public spending Because the data reported to CRS reflects commit- within countries or profit-seeking investments. The ments rather than disbursements, at best what we results are based on replies from 65 sources of can assess is the extent of support for the environ- financing (including the most significant ones). mental objectives. 12 The data for official development financing commitments by multilateral institutions is less comprehensive than the data for bilateral donors. The CRS database includes the major regional development banks (African Development Bank, Asian Development Bank, and Inter-American Development Bank), as well as the World Bank Group. Most of the financial flows from these institutions are in the form of OOF, except for the IDA and special development funds of regional development banks 3 There are considerable differences among DAC donors on the relative shares they provide through multilateral institutions, as opposed to their own bilateral assistance programs. The eight largest donors put widely different emphasis on the share they devote to bilateral aid. Italy and the United Kingdom provide more than 40 percent of their assistance multilaterally. Canada, Germany and the Netherlands all have major bilateral programs, but provide a share of multilateral assistance at or above the DAC average. Japan, the United States, and France all have substantial bilateral programs in which geographic concentration is important, and have decided that perhaps their interests are best served by spending a greater share of their development assistance than the DAC average through the bilateral channel. 4 Seventeen out of 23 DAC Members contributed to the 1998 study. Countries that provided the data accounted for 96 percent of total bilateral ODA commitments by DAC bilateral donors in 1998. The pilot study sought to test the feasibility of refining the reporting as required. For one year's worth of commitments, DAC members were asked to classify and score individual aid activities against the 'climate change', 'desertification', and 'biodi- versity' markers, defined in collaboration with the Convention secretariats. Appendix B. Welfare Economics of Subsidies: The Case of Electricity Energy subsidies across developing and for the q units consumed) and what they actually transition economies add up to billions of dollars pay (p times q), and is known as consumer each year. As well as being a drain on the surplus. treasury, subsidies also entail net losses of social If the government decides to cap the price of welfare. This appendix explains the underlying electricity at the level p*, consumers will demand economics and provides some numerical results q* units. The electricity company will not be for electricity. We will assume, for the moment, willing to supply quantity q* at price p*, since the that the government is the owner of the electricity revenue generated (p* times q*) will not be companies, and that any losses of the electricity enough to cover the production costs (p times q*). companies are financed through the general The difference between those two figures budget. represents the subsidy required by the electricity Assume for simplicity that the long run company to supply q* units at price p*, and is marginal cost of producing electricity (generation given by the area B+C+D. As explained in plus transmission and distribution) is p for every Section 2.1, we have calculated that across production level. As a result the supply curve is developing and transition economies this flat, at the level p. Assume as well that the amounted to some US$102 billion in 1999. demand curve is linear. Under these conditions, When the subsidy is in place, consumers are the quantity of electricity produced and consumed paying p* times q* for electricity, represented by will be q, and the price paid for every unit of the area E+F. If the subsidy is removed, then electricity will be p, as seen in Figure B.1. The consumers will be paying p times q. Since the welfare generated in this transaction is reaped by elasticity of electricity demand is generally consumers and is given by the area A. This is the assumed to be lower than I (in our case we have difference between what the consumers are assumed it to be 0.5), the result will be an increase willing to pay (the area under the Demand curve in revenue the size of the area B-F. We have Figure B.1: Welfare economics of subsidies: the case of electricity A Supply p B D Dmand q q* 117 118 Generating Public Sector Resources to Finance Sustainable Development Table B.1: Regional breakdown of electricity subsidies and welfare losses (US$ billions) Welfare Loss Potential increase (potential efficiency Subsidy in cash-flow gain) Sub-Saharan Africa 4.8 1.4 0.7 South and East Asia 23.8 8.9 2.1 Central and Eastern Europe 5.8 2.2 0.9 Former Soviet Union 41.3 14.6 10.3 Latin American and the Caribbean 4.6 1.3 0.7 Middle East and North Africa 21.9 6.9 5.1 Total 102.4 35.2 19.8 Note "Potential increase in cash-flow" refers to revenue increases from subsidy elimination. Source: Authors' calculations. calculated that across developing and transition of the 'subsidy' is assumed by taxpayers (higher economies this adds up to some US$35 billion. In taxes to pay for the direct transfers), future consu- addition to the increase in revenue, there is a de- mers (higher prices in the future to pay for up- crease in production costs, since fewer units of dating infrastructure), or oil companies share- electricity need to be produced. The decrease in holders (lower profits), the sizes of the subsidy production costs is given by the area F+C+D, and and the welfare loss are not affected. in our study amounts to some US$67 billion. The If the govemment wishes to transfer re- increase in revenue plus the cost savings sum to sources between groups in society (for example, the area B+C+D, US$102 billion, the amount of from high income taxpayers to low income hous- the subsidy. From a public finance perspective, eholds), an electricity subsidy is a highly in- the total amount of the subsidy is the relevant efficient way of doing it. In the case of electricity figure. in 1999, govemments are spending $102 billion In terms of welfare, the establishment of the while consumers only receive $82 billion. It subsidy implies that the welfare of consumers would be more efficient (by $20 billion) to simply increases by the area B+C. Since this increase in transfer this amount of money directly. If the welfare has cost the govemment the amount government wants to increase the welfare of the represented by the area B+C+D, there has been a poor then electricity subsidies have two added net welfare loss for society of D. This is problems. Since the non-poor also consume technically known as a deadweight loss. We have electricity (and, it can be argued, much more than calculated that across developing and transition the poor), only a fraction of the $82 billion actu- economies this welfare loss amounts to some ally reaches the poor (this is known as leakage). US$20 billion. From a welfare economics And, since only a fraction of the poor have access perspective, the welfare loss is the relevant figure. to electricity, many poor households will not By definition, this US$20 billion represents the benefit at all (this is known as mis-targeting). In efficiency gain from removing the subsidy. Table addition, if the poor pay some taxes and the B. 1 summarizes these aggregate figures by region. electricity subsidy is at least partly financed by the Let us return briefly to our assumption that general budget, then the poor who do not consume the government finances the losses of the electricity end up worse off under the subsidy electricity company. In reality, the price of scheme. electricity is maintained below the long run Our estimate of welfare loss represents a marginal cost of production by a combination of lower bound, as in practice there are two other direct transfers, under-investment in infra- sources of welfare loss. First, since governments structure, and subsidized fuels. Whether the cost rarely transfer to the electricity companies the Generating Public Sector Resources to Finance Sustainable Development 119 resources to fully finance the subsidy enjoyed by environmental effects associated with electricity consumers, these utilities end up without the production. To account for those externalities, the resources to extend their distribution network. In price of electricity should be established above p fact, there is an incentive not to serve new custo- at the level at which the revenue generated would mers, since additional customers represent addi- compensate those suffering the externality. tional losses. So consumers who have a willing- Strictly speaking, the calculation of the subsidy ness to pay higher than the cost of supplying them and associated welfare loss should be undertaken remain unserved, and their consumer surplus is using this higher price as a reference. In graphical unrealized, representing a welfare loss. terms, this would imply that the supply curve The final source of welfare loss is the exi- would be placed at a higher level, making area D stence of environmental externalities. The main (the welfare loss) larger. text describes the negative local and global Appendix C. Statistical Appendix Table C.1: Estimated electricity subsidies and potential efficiency gains from subsidy phase-out Subsidy Efficiency gains Economy U$ million % LRMC % GDP US$ million % GDP Albania 80 53.1 2.2 18 0.5 Algeria 707 63.4 1.5 177 0.4 Angola 50 68.8 0.8 13 0.2 Argentina 0 0.0 0.0 0 0.0 Armenia 96 40.3 5.2 8 0.4 Azerbaijan 1,032 81.8 22.6 258 5.7 Bangladesh 0 0.0 0.0 0 0.0 Belarus 887 50.4 3.3 186 0.7 Bolivia 0 0.0 0.0 0 0.0 Bosnia and Herzegovina 17 0.0 0.4 0 0.0 Brazil 0 0.0 0.0 0 0.0 Bulgaria 730 46.2 5.9 133 1.1 Chile 0 0.0 0.0 0 0.0 China 10,609 18.4 1.1 568 0.1 Colombia 0 0.0 0.0 0 0.0 Costa Rica 0 0.0 0.0 0 0.0 Cote d'Ivoire 0 0.0 0.0 0 0.0 Cuba 35 3.3 n.a 0 n.a. Czech Rep. 1,096 34.0 2.0 127 0.2 Congo, Dem. Rep. of 0 0.0 n.a. 0 0.0 Dominican Rep. 0 0.0 0.0 0 0.0 Ecuador 287 54.6 1.5 69 0.4 Egypt, Arab Rep. of 1,611 40.4 1.8 238 0.3 El Salvador 0 0.0 0.0 0 0.0 Estonia 29 8.7 0.6 1 0.0 Georgia 105 26.6 3.7 9 0.3 Guatemala 7 2.5 0.0 0 0.0 Haiti 0 0.0 0.0 0 0.0 Honduras 0 0.0 0.0 0 0.0 Hungary 510 24.3 1.1 38 0.1 (.. .continues) 121 122 Generating Public Sector Resources to Finance Sustainable Development Table C.1 (...continued) Subsidy Efficiency gains Economy U$ million % LRMC % GDP US$ million % GDP India 6,361 25.9 1.4 515 0.1 Indonesia 3,110 62.7 2.2 778 0.6 Iran, Islamic Rep. of 4,106 66.0 4.1 1,027 1.0 Jamaica 0 0.0 0.0 0 0.0 Jordan 228 41.1 2.8 35 0.4 Kazakhstan 1,553 60.9 9.2 388 2.3 Kenya 0 0.0 0.0 0 0.0 Kyrgyz Rep. 371 89.7 29.7 93 7.4 Latvia 0 0.0 0.0 0 0.0 Lebanon 105 13.8 0.6 4 0.0 Lithuania 86 19.0 0.8 5 0.0 Macedonia, Former Yugoslav Rep. 70 20.4 1.1 4 0.1 Mexico 2,124 17.9 0.4 110 0.0 Moldova 82 43.6 7.0 14 1.2 Morocco 0 0.0 0.0 0 0.0 Mozambique 0 0.0 0.0 0 0.0 Namibia 0 0.0 0.0 0 0.0 Nicaragua 0 0.0 0.0 0 0.0 Oman 179 35.0 n.a. 22 0.1 Panama 0 0.0 0.0 0 0.0 Paraguay 26 11.7 0.3 1 0.0 Peru 0 0.0 0.0 0 0.0 Philippines 0 0.0 0.0 0 0.0 Poland 551 8.9 0.4 13 0.0 Romania 542 27.1 1.5 46 0.1 Russian Federation 29,083 79.7 15.1 7,271 3.8 Saudi Arabia 7,084 81.4 5.0 1,771 1.3 Senegal 16 15.5 0.3 1 0.0 SlovakRep. 233 16.1 1.2 11 0.1 Slovenia 0 0.0 0.0 0 0.0 South Africa 4003 39.9 3.1 580 0.4 Syrian Arab Rep. 537 60.2 3.4 134 0.8 Taiwan, China 357 3.5 0.1 3 0.0 Tajikistan 675 99.2 62.2 169 15.6 Tanzania 0 0.0 0.0 0 0.0 (.. .continues) Generating Public Sector Resources to Finance Sustainable Deveiopment 123 Table C.1 (...continued) Subsidy Efficiency gains Economy U$ million % LRMC % GDP US$ million % GDP Thailand 1,208 21.5 1.0 78 0.1 Trinidad and Tobago 135 50.9 2.0 29 0.4 Tunisia 36 6.4 0.2 1 0.0 Turkey 0 0.0 0.0 0 0.0 Turkmenistan 283 92.2 8.6 71 2.1 Ukraine 4,783 65.9 15.1 1,196 3.9 Uruguay 0 0.0 0.0 0 0.0 Uzbekistan 2,382 89.4 27.5 595 3.5 Venezuela 1,948 59.3 1.9 487 0.5 Vietnam 269 22.3 0.9 18 0.1 Yemen 100 49.5 1.4 20 0.3 Yugoslavia, Federal Rep. of (Serbia/Montenegro) 1,736 86.6 17.6 434 4.4 Zambia 0 0.0 0.0 0 0.0 Zimbabwe 174 26.9 3.2 15 0.3 Notes: See text for a brief explanation on methodology, interpretation of results, and caveats. Subsidies refer to "economic" as opposed to "financial" ones. The methodology used takes account to a very limited extent to individual countries characteristics, and it is not intended to replace in-depth country evaluations as, for instance, those in IEA (1999). These are preliminary results, where no correction measure has been taken to deal with outliers. Source: Authors' estimates. 124 Generating Public Sector Resources to Finance Sustainable Development Table C.2: Estimated gasoline and diesel subsidies and potential effects on public budget from price reform, 1999 Gasoline and Net effect on public budget from Gasoline subsidy Diesel subsidy diesel subsidy gasoline and diesel price reform (US$ million) (US$ million) (% GDP) (US$ million) (%o GDP) Economy GBI GB2 GBI GB2 GBI GB2 GA RA GA RA Albania 0 0 0 0 0.00 0.00 671 98 18.2 2.7 Algeria 0 72 177 514 0.37 1.23 539 194 1.1 0.4 Angola 0 0 20 87 0.32 1.41 29 28 0.5 0.5 Argentina 0 0 0 0 0.00 0.00 4 3 0.0 0.0 Armenia 0 0 0 1 0.00 0.07 66 51 3.6 2.8 Azerbaijan 0 0 0 38 0.00 0.84 51 32 1.1 0.7 Bahrain 0 25 3 16 n.a. n.a 267 95 n.a. n.a. Bangladesh 0 0 0 63 0.00 0.14 356 297 0.8 0.6 Belarus 0 0 41 223 0.15 0.83 179 219 0.7 0.8 Benin 0 0 0 0 0.00 0.00 41 18 1.7 0.8 Bolivia 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Bosnia 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Brazil 0 0 0 0 0.00 0.00 24 14 0.0 0.0 Brunei 0 1 4 17 n.a. n.a 38 44 n.a. n.a. Bulgaria 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Camneroon 0 0 0 0 0.00 0.00 299 0 3.3 0.0 Chile 0 0 0 0 0.00 0.00 5,434 1,066 8.0 1.6 China 0 944 0 538 0.00 0.15 7,145 8,715 0.7 0.9 Colombia 0 176 0 178 0.00 0.41 1,077 1,500 1.2 1.7 Congo, Rep. of 0 0 0 0 n.a. n.a 8 1 n.a. n.a. Congo, Dem. Rep. of 0 0 0 0 0.00 0.00 53 0 2.4 0.0 Costa Rica 0 0 0 0 0.00 0.00 48 41 0.3 0.3 Cote d'Ivoire 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Croatia 0 0 0 0 0.00 0.00 83 63 0.4 0.3 Cyprus 0 0 0 42 0.00 0.46 0 0 0.0 0.0 Czech Rep. 0 0 0 0 0.00 0.00 266 0 0.5 0.0 I)ominican Rep. 0 0 0 98 0.00 0.56 374 128 2.1 0.7 Ecuador 0 0 0 156 0.00 0.82 1,793 823 9.4 4.3 Egypt, Arab Rep. of 0 125 690 1,416 0.77 1.73 570 164 0.6 0.2 El Salvador 0 0 0 0 0.00 0.00 15 11 0.1 0.1 Eritrea 0 0 0 6 0.00 0.97 5 2 0.8 0.2 (.. .continues) Generating Public Sector Resources to Finance Sustainable Development 125 Table C.2 (...continued) Gasoline and Net effect on public budget from Gasoline subsidy Diesel subsidy diesel subsidy gasoline and diesel price reform (US$ million) (US$ million) (%o GDP) (USS million) (% GDP) Economy GBI GB2 GBI GB2 GBI GB2 GA RA GA RA Estonia 0 0 0 0 0.00 0.00 86 64 1.7 1.3 Ethiopia 0 0 0 28 0.00 0.44 21 19 0.3 0.3 Gabon 0 0 0 0 0.00 0.00 43 33 1.0 0.8 Ghana 0 20 0 23 0.00 0.56 171 166 2.2 2.1 Guatemala 0 0 0 0 0.00 0.00 127 77 0.7 0.4 Haiti 0 0 0 0 0.00 0.00 14 0 0.3 0.0 Honduras 0 0 0 0 0.00 0.00 52 0 1.0 0.0 Hong Kong 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Hungary 0 0 0 0 0.00 0.00 0 0 0.0 0.0 India 0 0 0 2,273 0.00 0.51 5,571 2,247 1.2 0.5 Indonesia 672 1,562 2,570 4,373 2.29 4.20 6,441 7,314 4.6 5.2 Iran, Islamic Rep. of 1,723 2,764 3,726 5,612 5.47 8.41 3,062 7,992 3.1 8.0 Iraq 645 935 1,045 1,581 n.a. n.a 263 2,885 n.a. n.a. Jamaica 0 0 0 0 0.00 0.00 75 55 1.1 0.8 Jordan 0 0 73 195 0.90 2.41 314 45 3.9 0.6 Kazakhstan 0 0 0 114 0.00 0.67 549 605 3.3 3.6 Kenya 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Korea Dem. Rep. of 0 0 0 0 0.00 0.00 24 0 n.a. n.a. Korea, Rep. of 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Kuwait 98 267 38 94 0.46 1.21 597 360 2.0 1.2 Kyrgyz Rep. 0 0 0 3 0.00 0.27 37 34 2.9 2.7 Latvia 0 0 0 0 0.00 0.00 3 0 0.0 0.0 Lebanon 0 0 0 118 0.00 0.72 451 0 2.7 0.0 Lithuania 0 0 0 0 0.00 0.00 25 0 0.2 0.0 Libya 20 207 105 353 1.17 5.25 974 305 n.a. n.a. Macedonia, Former Yugoslav Rep. 0 0 0 0 0.00 0.00 7 0 0.2 0.0 Malaysia 0 375 287 962 0.36 1.69 2,809 2,774 3.6 3.5 Malta 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Mexico 0 0 0 0 0.00 0.00 3,234 2,421 0.7 0.5 Moldova 0 0 0 0 0.00 0.00 26 22 2.2 1.9 Morocco 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Mozambique 0 0 0 0 0.00 0.00 1 0 0.0 0.0 Myanmar 18 48 112 236 n.a. n.a. 354 327 n.a. n.a. (.. .continues) 126 Generating Public Sector Resources to Finance Sustainable Development Table C.2 (...continued) Gasoline and Net effect on public budgetfrom Gasoline subsidy Diesel subsidy diesel subsidy gasoline and diesel price reform (US$ million) (US$ million) (%l GDP) (US$ million) (° GDPJ Economy GB1 GB2 GBI GB2 GBI GB2 GA RA GA RA Namibia 0 0 0 0 0.00 0.00 43 26 1.2 0.7 Nepal 0 0 0 10 0.00 0.20 32 10 0.6 0.2 Netherland Antilles 0 0 0 0 n.a. n.a. 3 0 n.a. n.a. Nicaragua 0 0 0 0 0.00 0.00 15 4 0.7 0.2 Nigeria 218 517 163 406 1.09 2.63 1,367 1,799 3.9 5.1 Oman 0 20 0 32 n.a. n.a. 247 29 n.a. n.a. Pakistan 0 0 0 685 0.00 1.18 1,239 757 2.1 1.3 Panama 0 0 0 0 0.00 0.00 94 46 1.0 0.5 Paraguay 0 0 0 40 0.00 0.51 120 0 1.5 0.0 Peru 0 0 0 0 0.00 0.00 87 0 0.2 0.0 Philippines 0 0 0 416 0.00 0.54 1,333 967 1.7 1.3 Poland 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Romania 0 0 0 0 0.00 0.00 118 48 0.3 0.1 Russian Federation 0 1,107 60 2,458 0.03 1.85 9,495 11,229 4.9 5.8 Saudi Arabia 560 1,525 1,890 3,609 1.76 3.68 6,524 3,494 4.7 2.5 Senegal 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Singapore 0 0 0 0 0.00 0.00 55 0 0.1 0.0 Slovak Rep. 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Slovenia 0 0 0 0 0.00 0.00 0 0 0.0 0.0 South Africa 0 0 0 0 0.00 0.00 832 475 0.6 0.4 Sri Lanka 0 0 0 29 0.00 0.18 172 34 1.1 0.2 Sudan 0 5 0 52 0.00 0.59 179 149 1.8 1.5 Syrian Arab Rep. 0 0 391 931 2.37 5.63 1,255 273 7.6 1.7 Taiwan, China 0 0 0 0 0.00 0.00 361 48 n.a. n.a. Tajikistan 0 30 0 6 0.00 3.32 224 318 20.7 29.3 Tanzania 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Thailand 0 79 0 261 0.00 0.28 2,442 1,600 2.0 1.3 Togo 0 0 0 0 0.00 0.00 14 8 1.0 0.6 Trinidad and Tobago 0 0 2 26 0.03 0.38 94 75 1.4 1.1 T'unisia 0 0 0 0 0.00 0.00 130 0 0.6 0.0 l'urkey 0 0 0 0 0.00 0.00 0 0 0.0 0.0 l'urkmenistan 78 126 114 181 5.81 9.30 252 461 7.6 14.0 United Arab Emirates 4 152 38 153 n.a. n.a. 601 200 n.a. n.a. (... continues) Generating Public Sector Resources to Finance Sustainable Development 127 Table C.2 (...continued) Gasoline and Net effect on public budget from Gasoline subsidy Diesel subsidy diesel subsidy gasoline and diesel price reform (US$ million) (US$ million) (% GDP) (US$ million) (% GDP) Economy GB I GB2 GB I GB2 GB I GB2 GA RA GA RA Ukraine 0 0 0 0 0.00 0.00 648 533 2.1 1.7 Uruguay 0 0 0 0 0.00 0.00 0 0 0.0 0.0 Uzbekistan 193 353 241 442 2.54 4.66 913 1401 5.3 8.2 Venezuela 894 1761 576 1,003 1.42 2.68 3,488 5,941 3.4 5.8 Vietnam 0 0 0 179 0.00 0.62 728 456 2.5 1.6 Yemen, Rep. of 0 74 108 184 1 58 3.78 369 192 5.4 2 8 Yugoslavia, Federal 0 0 0 0 n.a. n.a. 0 0 0.0 0.0 Rep. of Zimbabwe 0 0 0 0 0.00 0 00 100 67 1.8 1.2 Notes: See main text for a brief explanation of methodology and meanings GBI: Traditional measure GB2: Transport sector measure GA: Global average RA: Regional average n.a. not available Source. Authors' calculations based on data in Metschies, 1999, 2001. 128 Generating Public Sector Resources to Finance Sustainable Development Table C.3: Estimated potential resource rents by country, 1999 Mineralfuel Mineral Timber Total resource rents rents rents rents Fconomy ($ million) ($ million) ($ million) ($ million) (% GDP) Albania 13 1 1 15 0.40 Algeria 4,728 22 10 4,759 9.94 Angola 747 0 18 765 8.95 Argentina 1,168 85 60 1,313 0.46 Armenia 0 0 0 0 0.00 Australia 1,046 2,384 363 3,793 0.94 Austria 46 0 173 219 0.11 Azerbaijan 666 0 0 666 16.64 B3angladesh 56 0 16 73 0.16 E3elarus 0 0 0 0 0.00 Belgium 0 0 0 0 0.00 lBenin 0 0 7 7 0.29 Bolivia 54 28 9 91 1.09 Bosnia and Herzegovina 0 0 0 0 0.00 Botswana 0 4 2 6 0.10 Brazil 3,164 2,179 1,489 6,832 0.91 Bulgaria 54 22 29 105 0.85 Burkina Faso 0 0 11 11 0.41 Burundi 0 1 5 5 0.74 Cambodia 0 0 23 23 0.75 Cameroon 240 0 57 297 3.23 Canada 8,314 413 4,172 12,899 2.03 Central African Rep. 0 0 18 18 1.69 Chad 0 0 16 16 1.02 Chile 5 1,430 222 1,657 2.46 China 6,896 1,336 2,249 10,481 1.06 Colombia 2,261 56 15 2,332 2.69 Congo, Dem. Rep. of 0 0 40 40 0.71 Congo, Rep. of 283 0 13 297 13.38 Costa Rica 0 0 30 30 0.20 Cote d'Ivoire 0 0 64 64 0.57 Croatia 55 0 0 55 0.27 Cuba n.a. n.a. 7 n.a. n.a. Czech Republic 23 0 0 23 0.04 Denmark 55 0 16 72 0.04 (...continues) Generating Public Sector Resources to Finance Sustainable Development 129 Table C.3 (...continued) Mineralfuel Mineral Timber Total resource rents rents rents rents Economy ($ million) ($ million) ($ million) ($ million) (%/6 GDP) Dominican Republic 0 39 0 39 0.23 Ecuador 981 4 105 1,090 5.74 Egypt, Arab Rep. of 965 27 3 995 1.12 El Salvador 0 0 12 12 0.09 Eritrea 0 0 0 0 0.00 Estonia 0 0 0 0 0.00 Ethiopia 0 1 39 40 0.62 Finland 0 13 780 793 0.61 France 73 0 523 596 0.04 Gabon 330 0 57 387 8.89 Gambia, The 0 0 2 2 0.59 Georgia 0 0 0 0 0.00 Germany 178 0 551 729 0.03 Ghana 0 38 25 63 0.81 Greece 1 19 13 33 0.03 Guatemala 61 0 9 70 0.38 Guinea 0 46 13 59 1 70 Guinea-Bissau 0 0 3 3 1.60 Haiti 0 0 4 4 0.10 Honduras 0 4 15 20 0.36 Hong Kong, China 0 0 0 0 0.00 Hungary 56 0 30 85 0.18 India 2,962 693 635 4,290 0.96 Indonesia 4,361 677 703 5,740 4.03 Iran, Islamic Rep. of 11,425 89 20 11,534 10.41 Iraq n.a. n.a. I n.a. n.a. Ireland 0 28 39 68 0.07 Israel 0 25 2 27 0.03 Italy 259 0 66 325 0.03 Jamaica 0 58 5 63 0.91 Japan 29 0 336 366 0.01 Jordan 0 36 0 36 0.45 Kazakhstan 1,708 0 0 1,708 10.78 Kenya 0 0 31 31 0.30 Korea, Dem. Rep. of n.a. n.a. 27 n.a. n.a. Korea, Rep. of -3 20 38 56 0.01 (.. .continues) 130 Generating Public Sector Resources to Finance Sustainable Development Table C.3 (...continued) Mineralfuel Mineral Timber Total resource rents rents rents rents Economy ($ million) ($ million) ($ million) ($ million) (% GDP) Kuwait 6,163 0 0 6,163 20.84 Kyrgyz Republic 0 0 0 0 0.00 Lao People's Dem. Rep. 0 1 19 20 1.40 Latvia 0 0 0 0 0.00 Lebanon 0 0 0 0 0.00 Lesotho 0 0 0 0 0.00 Libya n.a. n.a. 2 n.a. n.a. Lithuania 0 0 0 0 0.00 Macedonia, Former Yugoslav 0 0 0 0 0.00 Rep. Madagascar 0 0 2 2 0.05 Malawi 0 0 8 8 0.46 Malaysia 2,245 16 488 2,749 3.48 Mali 0 0 8 8 0.32 Mauritania 0 86 0 86 9.01 Mauritius 0 0 0 0 0.00 Mexico 9,651 218 252 10,120 2.09 Moldova 0 0 0 0 0.00 Mongolia 0 33 8 41 4.50 Morocco 0 107 12 119 0.34 Mozambique 0 0 21 21 0.53 Myanmar n.a. n.a. 75 n.a. n.a. Namibia 0 4 0 4 0.14 Nepal 0 0 16 16 0.33 Netherlands 0 0 14 14 0.00 New Zealand 160 16 0 176 0.32 Nicaragua 0 1 4 5 0.24 Niger 0 0 8 9 0.43 Nigeria 4,993 0 194 5,187 14.80 Norway 1,142 8 121 1,271 0.83 Oman 2,388 0 0 2,388 15.96 Pakistan 557 0 64 621 1.07 Panama 0 0 1 1 0.01 IPapua New Guinea 179 154 69 402 11.21 Paraguay 0 0 69 69 0.89 Peru 140 213 32 385 0.74 Philippines -1 42 79 121 0.16 (.. .continues) Generating Public Sector Resources to Finance Sustainable Development 131 Table C.3 (...continued) Mineralfuel Mineral Timber Total resource rents rents rents rents Economy (S million) ($ million) ($ million) ($ million) (% GDP) Poland 202 171 207 580 0.37 Portugal 0 17 132 149 0.13 Puerto Rico 0 0 0 0 0.00 Romania 383 12 86 481 1.41 Russian Federation 25,721 0 0 25,721 6.41 Rwanda 0 0 5 5 0.27 Saudi Arabia 27,257 0 0 27,257 19.56 Senegal 0 5 16 22 0.45 Sierra Leone 0 0 3 3 0.42 Singapore 0 0 0 0 0.00 Slovak Republic 0 0 0 0 0.00 Slovenia 0 0 0 0 0.00 South Africa 22 564 394 980 0.75 Spain 18 30 299 346 0.06 Sri Lanka 0 0 17 17 0.10 Sudan 0 3 31 34 0.35 Sweden 0 95 1,168 1,264 0.53 Switzerland 0 0 71 71 0.03 Syrian Arab Republic 1,839 7 1 1,847 9.53 Tajikistan 2 0 0 2 0.09 Tanzania 0 3 24 26 0.30 Thailand 69 6 57 133 0.11 Togo 0 5 6 11 0.78 Trinidad and Tobago 433 0 1 434 6.32 Tunisia 198 60 4 262 1.25 Turkey 219 46 184 449 0.24 Turkmenistan 705 0 0 705 22.01 Uganda 0 0 29 29 0.45 Ukraine 1,180 0 0 1,180 3.05 United Arab Emirates 5,640 0 0 5,640 11.94 United Kingdom 2,238 0 155 2,393 0.17 United States 31,366 915 13,805 46,086 0.50 Uruguay 0 0 32 32 0.16 Uzbekistan 1,466 0 0 1,466 8.28 Venezuela 8,889 128 20 9,037 8.84 Vietnam 570 10 91 671 2.34 (.. .continues) 132 Generating Public Sector Resources to Finance Sustainable Development Table C.3 (...continued) Mineralfuel Mineral Timber Total resource rents rents rents rents Economy ($ million) ($ million) ($ million) ($ million) (% GDP) West Bank and Gaza 0 0 0 0 0.00 Yemen, Rep. of 942 0 0 942 13.81 Yugoslavia, Federal Rep. of n.a. n.a. 0 n.a. n.a. Zambia 0 32 13 45 1.44 Zimbabwe 3 79 4 85 1.52 Notes: Potential timber rents are calculated assuming that 50 percent of total rents can be captured. Total rents are calculated as follows: Industrial Roundwood Production x Average Regional Price x Rental Rate. The regional price is calculated using FAO data on regional export values and quantities for industrial roundwood. Due to the difficulties in collecting production cost data, rental rates were estimated. The rental rate is defined as (market price-unit cost)/ market price. Based on a review of the literature and discussion with World Bank staff, rental rates of between 0.3 and 0.55 were used. Data is from FAO for 2000. n.a. not available Source: Derived from data in World Bank, 2001d. Recent World Bank Technical Papers (conitiniued) No 463 Stephen Foster, John Chilton, Marcus Moench, Franklin Cardy, and Manuel Schiffler, Groundwater in Rural Development Facing the Challenges of Supply and Resource Sustactiability No 465 Csaba Csaki and Zvi Lerman, eds, Structur-al Change in the Farming Sectors in Central and Eastern Europe Lessonsfor EU Accession-Second World Bank/ FAO Workshop, Juie 27-29, 1999 No 466 Barbara Nunberg, Readyfor Europe Plibic Adnuitustmationi Reform and European Union Accession in Central and Eastern Europe No 467 Quentin T Wodon with contributions from Robert Ayres, Matias Barenstein, Norman Hicks, Kihoon Lee, William Maloney, Pia Peeters, Corinne Siaens, and Shlomo Yitzhaki, Poverty and Policy in Latin Amnerica and the Caribbean No 469 Laurian Unnevehr and Nancy Hirschhorn, Food Safety Issues in the Developing World No 470 Alberto Valdes, ed , Agricultuiral Support Policies in Trainsition Economirles No 471 Brian Pinto, Vladimir Drebentsov, and Alexander Morozov, Dismantling Russia's Nonpaymnents System Creating Conditionisfom Growth No 472 Jit B. S Gill, A Diagnostic Franie7oorkfor Revenue Administration No 473 Esen Ulgenerk and Leila Zlaoui, Froni Transitioni to Accession. Developing Stable and Competitive Finanlcial Markets in Bulgaria No 474 loanrus N Kessides, ed., Hungary A Regulatory and Structural Reviezw of Selected Infrastriuctiure Sectors No 475 Csaba Csaki, Zvi Lerman, and Sergey Sotnikov, Farm Sector Restructurinig in Belarus Progress and Constraints No 476 Katherine Terrell, Czech Republic Labor Market Report No 481 Csaba Csaki, John Nash, Achim Fock, and Holger Kray, Food amid Agriculture in Bulgaria The Challemige of Pieparingfor EU Accession No 482 Peter Havlik, Trade amid Cost Competitiveness in the Czech Republic, Hiungary, Polanid, and Slovenia No 483 Mojmnr Mrak, Commniuinal Infrastrictuire in Slovenia Survey of Investment Needs and Policies Aimed at Encouraging Private Sector Participation No 484 Csaba Csaki and Laura Tuck, Rural Development Strategy Easternl Eumiope and Central Asia No 488 Nina Bubnova, Governance Imtpact on Private Investment No 489 Tim Schwarz and David Satola, Telecomniinuications Legislation in Transitional and Developing Economiles No 490 Jesko Hentschel and Radha Seshagiri, The City Povem ty Assessnient A Primer No 491 Damel MIuller-Jentsch, The Devlopmnent of Electricity Markets in the Euro-Mediterranean Area No 492 Tuntivate Voravate, Douglas F Barnes, and V Susan B3ogach, Assessiiig Marketsfor Reneioable Energy in Rural Areas of Not thwesterni China No. 496 Jerry Lebo and Dieter Schelling, Design and Appraisal of Rural Transport Infrastructure Ensuiriuig Basic Access for Rural Comntiuniiiities No 497 Julian A Lampietti, Anthony A Kolb, Sumila Gulyaru, and Vahram Avenesyan, Utility Pricing amid Poor Lessonsfronm Armtenia No 498 Gillian Perkins and Ruslan Yemtsov, Arnienia. Restructuiring to Sustain Universal General Education No 499 Rogrigo A Chaves, Susana Sanchez, Saul Schor, and Emil Tesliuc, Financial Markets, Credit Comistraints, and Investmenit in Rural Romnanma No 500 Zvi Lerman and Karen Brooks, Turknlienistanl' An Assessment of Leasehold-Based Farm Restructiuring No 501 Aldo Baietti, Private Infrastructure in East Asia Lessons Learned in the Aftermath of the Crisis No 505 Ali Hashim and Bill Allan, Treasuryi Reference Model No 506 Omer Gokcekus, Nick Manning, Ranjana Mukherjee, and Raj Nallari, Institutional En?viron?ment and Public Officitils' Performance in Guyana No 507 Ranjana Mukherjee, Omer Gokcekus, Nick Manning, and Pierre Landell-Mills, Bangladesh The Experience and Perceptions of Public Officiuils No 509 World Bank, Kosovo Econonumc and Social Refornsfor Peace and Reconciliation No 510 Anatoly Vinokur, Joana Godinho, Christopher Dye, Nico Nagelkerke, The TB and HIV/AIDS Epidemnics itn the Russinal Federation No 512 Geremia Palomba, Milan Vodopivec, Financing, Efficiency, and Equihy in Albaniami Eduction No 513 Thomas O'Brien, Chiistian Fulipov, The Current Reglulatory Framezvork Govermiing Business in Bulgaria THE WORLD BANK ISIS H 3treet, N.W Vasbingtu, 10.G. 2433 USA TeieWiie: 202-473-WOOf Facsirnmle. 22-477-6391 uernle%: wwwwortdbank.arg F-mail: Fkedbavi@wsordlniak.org F' ' MRU 9,4f143364-5