~~~~~/3o3(, [II:;y: WORLD BANK ENVIRONMENT PAPER NUMBER 7 IfAf/f /9d Alternative Policies for the Control of Air Pollution in Poland Robin Bates, Janusz Cofala, and Michael Toman -.. . 1. W- -_r --'U- if U 'Uf RECENT WORLD BANK ENVIRONM ENT PAPERS No. I Ckleaver, MlunaNsinghe. Ds%on. Egli, Peuker, .nd lVencWiU'. tditors, Coinswria'tlitur of IVest tatl CUet rai -1frican: P a.7 i'r ctiConsrruaion tieIa: folret detse e tlfique centtralet tie It (-st No. 2 Pezzev. Sis tainzal'lr Dev-'op;ment Crnrrjt-1s: An fc[ lanrnic Analysis No. 3 NlI unasinghe, Erorrrclalr Ea rominncs an.d Stnlaival4r 1felopmel11t No. 4 Dewtevs, Trees. Land,. arid Lablor No. 5 English, Tiffen, and i Mortimrtre, Iand Resource Al-Ianatermant int Altaclaktis District. Ke,iia. 19311i}-1990 No. 6 Nider and XIunasinghe, Incorporathin Eprreiro:ne,nrtal Ce zc Cnrns inta Powetr S;ector Dlecisrtrnuaking: A Case- StUdy of Sri ImLnk-a WORLD BANK ENVIRONMENT PAPER NUMBER 7 Altermative Policies for the Control of Air Pollution in Poland Robin Bates, Janusz Cofala, and Michael Toman The World Bank Washington, D.C. Copyright ©) 1994 The Intemational Bank for Reconstruction and Development/mE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing March 1994 Environment Papers are published to communicate the latest results of the Bank's envirorunental work to the development community with the least possible delay. The typescript of this paper therefore has not been prepared in accordance 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 in this paper are entirely those of the author(E) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the dab included in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaiLies. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910.222 Rosewood Drive, Danvers, Massachusetts 01923, USA. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list (with full ordering information) and indexes of subjects, authors, and countries and regions. The latest edition is available free of charge from the Distribution Unit, Office of the Publisher, The Wc.rld Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'lena, 75116 Paris, France. Robin Bates was with the Environment Department of the World Bank at the time this paper was prepared; he is now with the Industry and Energy Department. Janusz Cofala is with the Polish Academy of Sciences and the International Institute for Applied Systems Analysis. Michael Toman is with Resources for the Future, Washington, D.C. Library of Congress Cataloging-in-Publication Data Bates, Robin W. Alternative policies for the control of air pollution in Poland / Robin Bates, Janusz Cofala, Michael Toman. p. cm. - (WorId Bank environment paper; no.7) Includes bibliographical references. ISBN 0-8213-27534 1. Air pollution-Economic aspects-Poland-Forecasting. 2. Air pollution-Economic aspects-Poland-Mathematical models. 3. Poland-Environmental conditions-Forecasting. 1. Cofala, Janusz. 11. Toman, Michael A. 111 Title. IV. Series. HC340.3.29A43 1994 363.73'926'09438-dc20 93-45657 CIP Table of Contents Abbreviations of Physical Units .................................................................. vi Foreword .................................................................. vii Acknowledgments .................................................................. vimi Abstract ................................................................... ia Chapter 1: Introduction ...................................................................I Background: Environment and Energy in Poland ..............................................................2 Plan of the ctudy ...................................................................7 Chapter 2: Conceptual Background .............................................. ..................... 8 Measurinng Emission Abatement Cost . ..................................................................8 Command Versus Incentive-Based Policies ................................................................... 8 Chapter 3: Analytical Framework .................................................................. 14 Final E ergy Demand Calculations .......................... ........................................ 14 Least-Cost Energy Supply .................................................................. 16 Measuring Social Costs of Abatement ................................... ............................... 18 Chapter 4: Design or Scenarios ........................................................... 19 Basic Economic Assumptions .................................................................. 19 Alternative Emissions Standards ........................................................................................ 20 Polish "Command-and-Control" Standards (CAC) .................... ................ 20 European Community Standards (EEC) ................................... 21 German Standards (GER) ..................................... 21 Flat Rate Reductions (FRRED) .............................................................................. 22 Comparison of Plant-Level Eniission Standards ............ ..................... 22 Altemative Economic Insnments ...................................... 22 Uniform Emissions Taxes on Large Stationary Sources (ETAX1) ................22.......... 2 S02 Trading by Large Sources (SO2TR) ................................................. .... 23 Coal Tax (COALTX) ............................................................................ 23 Uniform Emissions Taxes on All Sources (ETAX2) .......................... ........... 23 Suunmary .......................................... 24 iii Chapter 5: Model Results ............................................... 27 Base Case Scenario .................................................... 27 Energy and Emussions Dfferences Across Standards Scnios .......................................... 27 Energy and Emissions Differences Across Economic Policy Scenarios ............................... 28 Economic I mpacts of Emission Controls .................................................... 29 Chapter 6: Institutional Issues ............................................... 36 Institutional Background .................................................... 36 The Effect of Restructuring on Economic Incentives .................................................... 37 Considerations for Policy Evaluation .................................................... 39 Command and Control .................................................... 39 Emission Fees ................................................... 40 Emission Trading .................................................... 42 Chapter 7: Concluding Remarks ................................................... 46 Policy Considerations ................................................... 47 References ............................................... 48 Appendices: 1. Details on the Modeling Scenario Definitions .51 2. Details on the Model Results .61 Tables 1.1 Ambient Environmental Standards in Poland, the U.SA. and Germany .................. 3 1.2 Poland: Energy Intensity Compared with Selected Developed Countries. 5 1.3 Poland: Emissions Balances 1990. 6 1.4 Poland: Emissions Intensity Compared with EC........... 6 4.1 Characteristics of Alterative Standards Scenarios. 25 4.2 Characteristics of CAC and Altemative Economic Instrument Scenarios .26 5.1 Social Cost of Pollution Control, 1991-2015 (CAC) .30 5.2 Social Cost of Pollution Control, 1991-2015 (Economic Instuments) .30 5.3 Energy Price Indices to Final Consumers for Scenarios .33 5.4 Pollution Fees (Actual) and Taxes for ErAXI and ETAX2 ................................... 34 5.5 Costs and Profits fom Sulphur Dioxide Trading ...................................... ..... 34 Figures 2.1 Social Costs of Emissions Controls ........................................................................ 9 2.2 Defmiing a Least-Cost Envelope of Abatement Strategies ....................................... I 1 2.3 Cost-Effective Distribution of Abatement Effort ................................................ 12 iv Appendix Tables AL.l Economic Growth and Struchtral Change to 2010 ..................................... ......... 52 A1.2 Production of Energy-Intensive Products and Services ........................................ 52 A1.3 International Fuel Prices ....................................................... 53 A1.4 Fuel Prices to Final Energy Consumers in "BASE' Case ................... .................. 53 A1.5 Share of Transmissiorn/Distribution Costs and Taxes in Fuel Prices to Final Consumers in 1995, the "BASE' Case ....................................................... 54 Al.6 Own Price Elasticities Applied in the Scenarios ................................................... 55 Al.7 Polish S02 Standards ....................................................... 55 A1.8 PoLish NO, Standards ....................................................... 56 AJ.9 Polish Particulates Standards ....................................................... 56 Al-10 European Community S02 Standards ......................................... .............. 57 A1.11 European Commn unity NOx Standards ....................................................... 57 Al.12 European Community Particulates Standards ...................................................... 58 A1.13 German S02 Standards ....................................................... 58 A1.14 German NO, Standards ....................................................... 59 A1.15 Gernman Particulates Standards ....................................................... 59 A2.1 Energy Demand and Energy Intensities for the "BASE' Scenario ........................ 61 A2.2 Changes in Sectoral Energy Intensities for the "BASE" Scenario ......................... 61 A2.3 Emissions of Atmospheric Pollutants by Sectors, BASE Case ............. ................ 62 A2.4 Comparison of Scenario Results - Altemative Levels of Controls ........... ............. 63 A2.5 Emissions of Pollutants - Alternative Levels of Controls ...................................... 64 A2_6 Comparison of Scenario Results - Instruments for Large Stationary Sources ....... 65 A2.7 Emissions of Pollutants - Instruments for Large Stationary Sources .......... .......... 66 A2.8 Comparison of Scenario Results - Tax on All Fuel Users ..................................... 67 A2.9 Emission of Pollutants - Tax on All Fuel Users .................................................... 68 A2.10 Comparison of Scenario Results - Coal Tax ....................................................... 69 A2.1 i Emission of Pollutants - Coal Tax ........................ ............................... 70 A2.12 Costs of Emission Control Measures for Large Stationary and Transport Sources 71 A2.13 Total Undiscounted Energy Supply and Conversioni Investments ......................... 71 A2.14 Production of Hard Coal and Lignite ................................................................... 72 v ABBREVIATIONS OF PHYSICAL UNITS kgoe = kilograms of oil equivalent MJ = megajoules GJ = gigajoules PJ = petajoules MWth = megawatts thennal energy MWh = mregawatthours TWh = terawaihours tce = tonnes of coal equivalent pgfm3 = micrograms per cubic meter Nm3 = normal cubic meters; refers to dry gas at 00 Celcius and 1 atmosphere of pressure vi FOREWORD The decade of the 1980s witnessed a fndamental The phrase "sustainable development" has change in the way governments and development been widely used, but we are still some way from agencies think about environment and a generally-recognized, operational definition of development The two are no longer regarded as the concept. In particular. such a definition must mutually exclusive. It is now recognized that a be set in a usable analytical framework. healthy environment is essential to sustainable This paper attempts to show how development and a healthy economy. Moreover, appropriate economic incentives in the economy at economists and planners are beginning to large, and the use of specific economic recognize that economic development that erodes instruments targeted at air pollution, can make an natural capital is often not successful. In fact. important practical contribution to sustainable development strategies and programs that do not developmenL The focus here is on the use of a take adequate account of the state of critical common natural resource, namely the air we resources-forests, soils, grasslands, freshwater, breathe, and ways in which it can be protected. coastal areas, and fisheries-may degrade the lbe paper argues that, although free market resource base upon which future growth is forces alone may not achieve sustainable dependenL developmeaL they can help us to make progress Since its creation, the Vice Presidency for towards it through better resource managenent Evironmentally Sustainable Development (ESDŽ Furthermore, the analysis and simulations has placed the highest priority on the analysis of presented in the paper suggest that by intervening these important issues. Within ESD, the in markets, especially energy markets, to allow for Enviromnent Department's work in particular has externalities, those same market forces can be focused on the links between environment and harnessed to produce further environmenal development and the implications of these links improvements. Besides other interventions, the for development policy in general. The objecive paper considers systems of taxes and trading in of the Enviromment Paper Series is to make the property rights. results of our work available to the general public. Mohamed T. El Ashry Director Environment Department vil ACKNOWLEDGMENTS The simulation analysis presented in this study is the product of a team effort by numerous researchers at the Polish Academy of Sciences. The authors particularly wish to thank W. Bojarski, Z. Parczewski, A. Umer, Z. Klimont, and T. Mroczek for their contributions to the simulations. A number of other Polish experts provided valuable support, infonnation, advice, and comments during the course of the project. W1. owe a particular debt of gratitude to Stanislaw Sitnicki and Tomasz Zylicz in this regard. Jerry Sleszynski provided useful information on Polish environmental policy. Within the Environment Ministry, we wish to thank Jerzy Kwiatkowski, Rafal Milaszewski, Wojciech Jaworski, and Jerzy Janota-Bzowski. Winston Harrington (Resources for the Future) and Thomas Tietenberg (Colby College) provided in-depth and very useful conmments on the manuscript. Helpful conversations and comments on the study also were received from Richard Ackernann, Gordon Hughes. Helmut Schreiber, and Mohan Munasinghe (World Bank) and from Dallas Burtraw (Resources for the Future). All conclusions and opinions expressed in the study are the authors' alone. We owe a further debt to Kay Murphy for her extensive efforts in preparing the manuscript. The Polish conswltants were financed by the World Bank and the Government of the Netherlands. Financial support for Toman's work was provided under a contract between Resources for the Future and the U.S. Environmental Protection Agency. . .i ALTERNATIVE POLICIES FOR THE CONTROL OF AIR POLLUTION IN POLAND Abstract Like other Central European countries, standards. Third, there are clear cost savings Poland faces the twin challenges of improving from using incentive-based policy environmental quality whilc also promoting instruments, even though the exact size of the economic development. Thlis study examines savings cannot be precisely idenfied. the cost oi achieving alternative emission Finally, the impacts on invesument and energy standards a:ud the savings in abatement cost prices of environmental policies are likely to that might be achieved with policies that rely be dwarfed by the forces of economic on econonmic incentives rather than with xigid restructuring and energy price refonr. "con.mand and control" measures. The focus The analysis also suggests a dynamic is primarily on dtree pollutants arising from and mixed strategy for the iemenation of energy cornbusdion-particulate matter (PM), economic instruments. Starting with nitrogen oxides (NO.), and sulphur dioxide command and control, emission fees could be (SOx0-although carbon dioxide (C02) rincrased to encourage some additional emissions also are tacked. A central element abatement and technical innovation and to of the analysis is a dynanic model of least- provided added ievenues to be used, e.g., for cost energy supply in Poland that allows cleanup of existing environmental damages. acaminanion at a national level of the effects Howevcr, it is unlikey that fees can be of different poilution standards and policies. increased to i. level necessary to meet The simulation analysis suggests, first, current emissions standards. To accomplish that significant decreases in air pollution this goal cost-effectively, an evolutionary emissions seem likely from economic movement toward enissions trading should resuucturing and energy pricing refonns and be considered. Trading could start with that tighter (enforced) eiission standards infomWal bilateral transactions, as in the US, such as those envisaged under current Polish and become more extenive as cirmstances policy are likely to generate considerablc and the interests of polluters warrant. additional decreases in ollution. Second, the However, even limited emissions trading Polish legal standards differ in important would require clearer legal and regulatory respccts from representative West European authority, as well as continued progress in standards, but the costs of meeting the two economic restructuring. We further altemative sets of standards in Poland do not cmphasize the importance of continued appear to be widely different. The costs of progress in economic and energy price reform the strict German standards, on the other as both a complement to and a prerequisite hand. are significantly higher, underscoring a for success in envirnmental policy, so that need for corresponding benefit assessment to finms have incentives to seek out lower-cost deteniine the value to Poland of such abatement options. Chapter I INITRODUCTION Like other Central European countries, level of the effects of different pollution Poland faces the twin challenges of improving standards and policies. The simulations environmental quality while also promoting provide considerable disaggregation by type economic development. PoUution, especially of fuel and sector. With these model outputs in the air, poses serious threats to human we can calculate the costs of different health and other resources in many parts of emission standards and of different policies to the country, though the precise extent of the attain specified standards over the next 25 hazards remains controversial. At the same years. time, excessive expenditure on pollution The model deals only with categories mitigation must be avoided, given a scarcity of emitters and generic categories of emission of capital needed for raising living standards. standards rather than specifying the size, This study examiines two broad issues spatial location, and emissions standards of related to the design of air pollution policy in specific sources. Consequently, the model Poland. The first issue is the cost of cannot be used to analyze the cost of different achieving altemative emission standards that ambient air quality standards or the cost of are of potential interest for Polish different means for achieving specified environrmental policy. The second issue is the ambient standards: our policy simulations savings in abatement cost that might be ignore ambient air quality constraints and achieved if a particular set of standards is focus only on the achievement of emission implemented with policies that rely on standards. This is an important drawback, economic incentivez-particularly taxes on both from a practical perspective (Polish emissions or tradable emission pernits- regulations include ambient as well as rather than with rigid "command and control" emissions standards) and an analytical (CAC) measures. In this analysis the perspective (the design of cost-effective enmission standards themselves are taken as regulation is complicated by the need to meet given rather than being inferred from a ambient standards as well as enmission broader study of pollution damages and standards). The model also takes as given abatement costs. broad trends in the composition and energy The focus is primarily on pollutants intensity of economic activity; it does not arising from energy combustion-particulate describe the process of economic matter (PM), nitrogen oxides (NOx), and restructuring itself. Two studies now under sulphur dioxide (SO2).' Carbon dioxide way by the World Bank are designed to (CC2) emissions also are tracked. A central address these gaps. The first, financed in part element of the analysis is a dynamic model of by the U.K. Environmental Know How Fund, least-cost energy supply in Poland. The is extending the analysis of altemative policy model, developed at the Polish Academy of instruments to a local level, by examining air Sciences, allows examination at a national pollution and environmental management in the Krakow region. Air dispersion modeling provides the link between emissions and 1Emissions from industrial processes (e.g., cement ambient standards. The second, assisted by manufacturing) also are included, though they are a Norwegian financing, is considering how very small part of the total. 2 Poland is restructuring economnic incentives in 1988-89, it was the third largest emitter of general and analyzing the impact on the S02 and NOx in Europe, after the former energy sector and air pollution in particular. Soviet Union and Germany; with regard to Subject to these caveats, the simulation PM, it ranked second (Nowicki 1993); and analysis provides estimates of the potential S02 and NOx emnissior per US$ of GNP gains from relying upon incentive-based were the highest among all countries of policies over CAC for meeting emission Central and Eastern Europe, including the targets. Even in advanced industrial former Soviet Union (World Resources economiies, not al dtis potential vwill be Institute 1992). One of the characteristics of reached, because of imperfections in markets air pollution in Poland is that the sources of and policies. A prominent example of market emissions are heavily concentrated. In distortions is imperfect regulation of the Katowice and Krakow, two of the most electric power sector. Enviromnental policies polluted regLons, emissions of dust and gases may also depart from the ideal to achieve in 1990 were 5-9 times and 10-12 times the compromises with equity or other social national average per unit land area goals. Problems of this type are multiplied in respectively (Central Statistical Office 1991, transitional economies like Poland's, in which Table I)- In consequence, a disproportionate market inctitutions are stil evolving, the state number of people in these regions is affected continues to play a substantial role in the by air pollution. economy, and social goals are not clearly Upper Silesia, which contains the defined and in flux. Katowice region, has only 10 percent of the To address these points we include nation's population and 2 percent of its land some discussion of how environmental area, but it is responsible for about one-third policies are carried out in Poland and of of all SO2 and PM emissions, and one-fifth of inplementation problems under transitional NO, emissions (Nowicki 1992, p. 12). circumstances. However, we give no Katowice is home to 22 of the 80 most quantitative estimate of how institutional and polluting enterprises in Poland. As a result, in social constraints might cause environmental 1989 the annual average concentrations of policies to fall short of their theoretical S02 and NO1 in the Katowice region potential; nor do we exhaust all the exceeded the pemiissible standards by a institutional and policy issues anenable to factor of two while PM exceeded standards qualitative analysis. This area is at or near the by a factor of four (Coopers & Lybrand forefront of research in enviromnental Deloitte 1991, Tables 2.7-2.9)-well beyond economics and is ripe for further research. In the guidelmes proposed by WHO. The particular, analysis of how altemative means permissible annual average concentration in of restructuring and managing the power Poland for S02 is 32 jig/m3, and for NO1 sector might affect economiic performance and PM it is 50 pLgfn3. The WHO guidelines and environmertal quality deserves a high suggest upper limits for S02 and PM of 60 pnlonlty. pghn3 and 90 pghn3, respectively (United BACKGROUND: ENVIRONMENT Nations Environment Program 1992). AND ENERGY IN POLAND ~~~Table 1.1 compares amnbient air quality AND ENERGY IN POLAND standards in Poland with the US and Germany. The comparison makes clear that Pnviroland ntsuffers ofro centhel disastrousn the Polish standards are significantly tougher than comparable Western figures 3 Table 1.1. Ambient Environmental Standards in Poland, the U.S.A. and Germany (jLg/m3) Poland U.S.A. Germany 24 hls. anmual 24 hrs. annual 24 his. annual Pollutant average. average average average average average SO2 200W 32 365 80 i 140 NO2 150 50 -- 100 80 PM 120 50 150 50 150 21150 after 1998. bLGenan standards for these pollutants are stated in terms of the 98th percentile of te average over any 30- minute period. The figures for S02 and PM are 400 and 300 respectively. The comparable sulphur figure for Poland is 600 (440 after 1998). Sources: Polish figures are taken from the Ordinance of the Ministry of Environmental Protection. Natual Resources and Forestry on the Protecdon of Air Against Pollution, 12 February 1990. U.S. and German figures are based on calculations by Mr. Shakeb Afsah, World Bank, drawing in part on Cochran and Pielke (1992). 4 Nevertheless, the limited available evidence daily concentrations well in excess of the (including parallel studies under the World pemiitted amounts (Coopers & Lybrand Bank's Environmental Action Programme for Deloitte 1991, Tables 2.10-2.13); and Central and Eastern Europe) suggests that particular problems are created in the older ambient conditions have imposed important parts of the city of Krakow (Ministry of burdens on human health, particularly in Environmental Protection 1991). Upper Silesia and Krakow.2 The non-health Air pollution in Poland is predominantly impacts also are thought by many to be caused by energy production and use, significant, though their actual severity is especially coal. Estimates of the precise subject to dispute. extent of energy's contribution vary, pardy no Severe smog conditions and high doubt due to differences in definition. sulphur concentrations in Krakow, the old However, according to estimates made in this capital of Poland, have had a serious impact study (Appendix 2, Table A23), energy on the city's residents as well as valuable production and use in 1990 by largc historic buildings and monuments. In 1990, stationary sources, essentially for electric the Krakow region was the second most power generation and heat production. polluted in Poland, next to Katowice (Central accounted for 70-80 percent of the eniissions Statistical Office 1991, Tablc I). It of PM and S02, and nearly 50 percent of the experiences smog incidents more than any emissions of NO, The Ministry of other city in Poland, with S02 concentrations Environmental Protection, Natural Resources reaching 3000-4000 pLgfm3 ( Nowicki 1992, and Forestry estimated that the energy sector p. 16). In 1989, the average ambient was responsible for 90 percent of SO2 concentration of S02 in Krakow was 70 emissions, and 60-70 percent of dusts and jtg/n3 (Coopers & Lybrand Deloitte 1991), NOx (Mnistly of Environmental Protection morc than twice tie national standad; and 1991, p. 13). Using 1987 data, Siexpinska the annual average concentration of PM (62 (1991, p. 28) attributes 54 percent of dust ig/rm3) also exceeded the national standard. and 55 percent of gas emissions to the fiel Sinilarly, in 1989 the maxriurm daiy and power industry; while Wasikiewicz concentrations exceeded the pennitted (1991, p. 111), with 1988 data, shows 66 amounts by a factor of two (Coopers & percent of S02, 42 percent of NOX and 27 Lybrand Deloitte 1991, Tables 2.10-2.13). percent of dust coning from energy However, it should be noted that emissions production. Aside from being the main from the main polluters in Krakow have source of air pollution in Poland, energ recendy dropped, due in no small measure to provides most (80 percent in 1988) of the the restructuring of the local industrial base, enissions of carbon dioxide (Nowicki 1992, and ambient air quality has improved since p. 6). 1987. Average concentations of S02 and An importan undelying cause of PM in 1991 fell to 67 V±tM3 and 54 Rgft'3 energy's role in air pollution is the fact tht respectively (Bolek and Wertz 1992). Poland has one of the most energy-intensive Problems from NOx are believed to be les economies in the world, with heavy serious than with SO2 and PM, although the dependence on coal and lignite. According to very limited data which are availab,le point to World Bank data, in 1989 Poland's energy intensity was 1.889 kgoe per US$ of GDP (at 1987 prices), exceeded only by China at 2Seein parricul; Kimpnick. Harison, NickCll, and 1.915 (Bates and Moore 1992). Using a TDman (1993). 5 slightly different measure, the World 12 percent for oil, 9 percent for gas and 1 Resources Institute put Poland in first place, percent for other fuels (Central Statistical with 79 MJ per US$ of GNP (at 1987 prices), Office 1992). Among the larger econoniies, and China second with 76 MJ per US$ of only South Africa was more dependent on GNP (World Resources Institute 1992, Table coal for its primary energy supply in 1989, 21.2). Of panicular significance is the fact although Bulgaria, North Korea and that Poland's total primary energy Czechoslovakia were similar (World requirement per US$ of GDP is about twice Resources Institute 1992-3, Table 21.1). as much as the average for Westem Europe Coal consumption is particularly concentrated (Intemational Energy Agency 1990, p. 11). in specific sectors, one of which is electricity, Comparisons with selected developed which depended on coal for 96 percent of its countries are even less favorable, as shown in generation in 1989 (60 percent from hard coal Table 1.2. and 36 percent from lipite). Electricity and Although the above figures are heat production together accounted for more aduittedly crude indicators of energy than half the Polish coal market in 1989 intensity, there seems to be little doubt that (Intemational Energy Agency 1990, p. 162). there has been an underlying structural In consequence, within the energy sector coal element to Poland's air pollution problem. contributes 82 percent of PM, 90 percent of This is aggravated by the fact that the Polish S02, 46 percent of NO., and 75 percent of economy is not only energy intensive, but also CO2 (see Table 1.3). It is not surprising that coal-intensive. Hard coal and brown coal the emissions intensity of the Polish economy (lignite) supplied 78 percent of total primary is so much higher tan that of thc US and the energy requiremnents in 1991, compared with EC (Table 1.4). Table 1.2. Poland: Energy Intensity Compared with Selected Developed Countries (kgoe/US$ GDP at 1987 prices) Cou = 1982 1987 1989 Poland 2.272 1.995 1.889 France 0.232 0.234 0.223 Gennany 0.254 0.248 0.223 Japan 0.175 0.166 0.163 United Kingdom 0.342 0.317 0.288 United States 0.457 0.394 0.406 Source: Bates and Moore (1992), p. 43. 6 Table 1.3. Poland: Emissions Balances 1990 S02 PM NO, C02 Source 103 03 1% 03 % 103 % tonnes tonnes tonnes tonnes Coal & lignite 2,542.9 90 1,205.8 82 547.22 46 271.823 75 Petroleum 175.6 6 25.1 2 477.65 40 38,886 11 products _____ Natural Gas _ - 31.14 3 3 5 Process Emissions 97.8 3 233.1 16 82.15 7 6,250 2 Other 16.2 1 4.0 - 49.43 4 26,185 7 Total 2,832.5 100 1,468.0 100 1,187.59 100 363,577 100 Table 1.4. Poland: Emissions Intensity Compared with EC (TonnesIUS$millions of GDPj (1989) Emission: Poland U.S. EC SO2 58.4 4.0 2.6 NOx 22.1 3.8 2.2 CO2 1,872 170 Source: World Resources Institute (1992), Tables 5.2 and 5.4; and Intermational Energy Agency (1990). 7 At present, vehicles (mobile sources) 6.8 million by 2000 (172 per thousand contribute less to total emissions than do coal population) and 8.8 nillion (215 per thousand burning and industrial processes, probably population), compared with 5.3 million in due to the fact that vehicle ownership is stil 1990. However, the growth in emissions also relatively low in Poland (138 passenger cars depends on changes in the types and age per thousand population in 1990, compared distribution of vehicles, as well as on policies with 410 in France and 360 in the designed to reduce emissions, although such Netherlands). Even so, vehicles contribute changes in the vehicle stock are expected to over 30 percent of NOx emissions, 37 percent have a dampening effect on emissions.3 of carbon monoxide, 24 percent of hydrocarbons, and 35 percent of lead PLAN OF THE STUDY (Nowicki 1992, p. 22; Wasikiewicz 1991, p. 1 11). They also are probably more significant Chapter 2 provides a brief conceptual for ambient air quality than these emissions introductior. to the measurement of emission percentages suggest, since vehicle emissions abatement cost and the differences between are concentrated in areas of high population conunand-and-control policies and policies density, such as the center of Krakow, where relying upon economic incentives. Chapter 3 serious constraints are imposed on traffic provides an introduction to the sinulation planning by the configuration of the old city. framework, while Chapter 4 describes the Enissions from mobile sources are likely to scenarios we consider. Chapter 5 reports increase, as the growth rate of passenger results of the simulation analysis. In Chapter vehicle ownership is high: 7.5 percent per 6 we discuss institutional and other practical annum in Poland over the period 1980-1990, issues that arise in selecting policy options in compared with 1.6 percent per annum in Poland. Chapter 7 briefly summarizes our France. According to the results of this conclusions. study, the number of private cars will reach 3See Walls (1993) for further discussion. Note also tdat even with the projected growth in Polish vehicle ownership. he number of cars per capita will be well below Westem European experience. Chapter 2 CONCEPTUAL BACKGROUND Since a central focus in this study is assessing also will lower rents earned by limited factors the economic cost of emission reduction in of production (producer surplus). differcnt scenarios, we first define what is Figure 2.1 illustrates these effects in a cncompassed in the full social cost of simple setting with only one final good, which emission reduction. We then discuss the we can take to be total energy services. The potential differences in cost that mnay arise curves SO and SI represent the supply of with incentive-based policies compared to energy services before and after the conmmand and control. imposition of stricter pollution standards. The demand for energy services is MIEASURING EMISSION ABATEMENT represented by D. With the stricter standaras, COST and assuming that price cleans the market for energy services, the price increases from P0 The most obvious source of cost in reducing to Pl and consumption correspondingly emissions is expenditures on control declines from QO to Ql. Area A in the equipment and effort. In the context of air diagram represents the cost of compliance pollution from energy combustion these with tighter standards given the reduced use expenditures includc investments in direct of energy services. Areas B and C represent pollution control equipment (such as catalytic losses in consumer and producer surplus converters and devices for flue gas respectively from the induced drop in energy desulphurization); fuel switching or fuel services use. The procedures used for quality improvement (coal washing); and any calculating the analogues of Areas A, B, and additional cost of investment in energy C in the simulation model of least-cost energy conversion technologies that give rise to use are discussed in Chapter 3. lower emissions (such as advanced coal combustion systems for electrcity COMMAND VERSUS INCENTIVE- generation). Resources invested in these BASED POLICIES4 activities are not available for other purposes and thus constitute part of the opportunity Associated with each technology capable of cost of pollution control reducing emissions of a particular pollutant is There also are indirect effects of policy a marginal cost curve that indicates the that may be important. These arise from the incremental cost of ernission reduction using fact that, at least in a functioning market that technology. The incremental cost of system, the additional costs of poUution emission reduction for society as a whole control will be reflected in higher prices for depends not just on the costs of individual final goods and services. As a result, technologies but also on the rules goveming consumers will reduce consumption of more how technologies can be used- In particular, expensive commodities and experience a loss the incremental cost to society is likely to be in real well-being; in effect, the real command of household incomes over goods and 4For further discussion of the topics in this section services will be reduced. Reduced demands see, e.&, Pearce and Turner (1990) and Tietenberg (1992). 9 Price, Cost p1 P0~~~~~~~~~~~~ I I D Energy Services i1 oo Figure 2.1. Social Costs of Emissions Control 10 different when command-and-control policies across polluters, even if specific abatement stipulate actions by poUuters compared to technologies are not required. This is shown policies that rely upon economic incentives, in Figure 2.3. Suppose that two polluters such as taxes on emissions or tradable have different costs of abatement because of emission permits. basic differences in their technologies. Their These points are illustrated in Figures marginal costs are indicated by MACA and 2.2 and 2.3. Figure 2.2 shows the MACB in the figure. Suppose that the total construction of a least-cost marginal cost amount of abatement required to be curve (MACagg) from marginal cost undertaken by A and B is indicated by the schedules for different abatement length of the horizontal axis in the figure. technologies (MAC1, MAC2, MAC3). The The total cost of abatement by A and B is least-cost deployment of technologies is minimized by the distribution of effort where achieved by successively exhausting all lower- MACA = MACB. For any other distribution cost efforts, across all technologies, before of effort, total cost could be lowered by undertaking higher-cost efforts. This is the shifting responsibility away from the polluter "lower envelope" of the constiment costs of with higher MAC toward the polluter with different technologies shown. lower MAC. This also is the outcome that will obtain The outcome with MACA = MACB is ideally when polluters face economic the one that will be achieved (at least incentives to reduce emissions without approximately) with incentive-based policies constraints on the means of compliance. for emission reduction. With an emission tax, Under these conditions, polluters naturally emitters will only reduce discharges if the will seek out the least costly control strategies marginal cost of doing so is less than the tax: in order to minimize total costs of economic given a comrrmon tax across emitters, marginal activities, including the costs of meeting abatement cost will be equalized at the level emission limits. If, instead, policy dictates the of the tax. With emission permit trading, use of a particular technology as the "best polluters will be indifferent between reducing practicable," then the marginal cost of emiissions and transacting in permits, when abatement becomes the marginal cost the price of permits equals the marginal cost associated with the identified technology. In of emission abatement for all finns. Again, this case some lower-cost options generally marginal costs are equalized at the (common) will be foregone, and the overall marginal value of the price of permits. In comrast, a cost of abatement will be increased by the specified distribution of emission reduction policy constraint.5 effort under conimand and control is likely to Policy also can raise the overaU raise total cost relative to the outcome with marginal cost of emission abatement by more flexible incentive-based policies.6 prescribing a distribution of abatement effort 5 In many cases, required technologies are not actually 6Our focus heie is on emission control However, the specified in law, but polluters a: required to show ultimate goal of poUution control is improving that they have achieved emissions reductions at least ambient air quality. When pollutants do not mix as large as those obtained with some specified unifornly in the atmosphere, location of emissions technologies Under such circumstances. polluters matters in deternining the effect of different emission often are drawn in practice to the reference reductions on air quality This is true in varying technologies unless other options are much cheaper, degrees for aUl the poUutants we consider. As pointed if only to avoid the extra buiden of proof that comes out in Chapter 6. this considerably complicates the with deviating from the benchmark control strategies. design of effective policy. 11 Cost 3 MAC3 MAC 2 I ~~~MAC1 MAC Level of Abatement Figure 2.2 Defining a Least-Cost Envelope of Abatement Strategies 12 Cost for A Cost for B MAC B MACA I ~~~Total Abatement-> Abatement by A >~ (-Abatement by B Figure 2.3 Cost-Effective Distribution of Abatement Effort 13 In addition to static efficiency gains, supplies. With emission charges, lower incentive-based policies convey dynamic abatement cost allows polluters tO reduce benefits by rewarding the introduction of costs by engaging in more abatement effort, lower-cost pollution abatement technologies. in order to avoid the tax. [n contrast, rigid Such technical progress allows polluters to approaches provide little incentive for expand abatement in cost-effective ways and technical innovation, particularly when the reduce permit purchases or increase pemiit controls specify the means for abatement. Chapter 3 ANALYTICAL FRAMEWORK Analyzing the effects of policies to control Using these indexing conventions. energy-related air pollution requires a define: description of energy service demands and the ways those demands are satisfied. Total Akj = level of activity k in sector j; pollution discharges will reflect the total Tmkj = share of Akj achieved with demands for different energy types and the technology m; ways in which energy services are provided. Fmkj = energy intensity of technology m in We first describe the process of detenrining pursuing Akf, time paths of final energy demands in the Simkj = share of fmal energy type i used with simulations. The second part of the chapter technology m in pursuing Akj. describes the simulation of least-cost energy supply decisions. We conclude by returning An example may be useful ir to the issue of how the social cost of interpreting these symbols. Suppose j = abatement can be measured using the outputs ferrous metallurgy, and A,j = basic steel of the model.'7 output measured in tonnes (as opposed to, say, rolled products using basic ingots as an FINAL ENERGY DEMAND input). The technology index m refers to CALCULATIONS different technologies for ingot production (oxygen furnace versus open hearth), so the Total final energy demand in the economy Tmkj are the shares of total ingot output ultimately depends on the level of economic produced by different technologies and F activities in different sectors and the energy are the energy intensities of the processes. intensities of these activities. Energy The Tnj are pure numbers between zero and intensities will vary across altemative one, while the F;gk reflect energy applied per technologies that can be used to accomplish unit output (e.g., PJ/tonne). The Simnj are the economic activities under consideration. the shares of different final energy types per Projection of final energy demands requires unit of total energy application in the projection of all these component influences. processes (e.g., PJ of coal, gas, or electricity To express these ideas more formally, per PJ total energy input). Another example suppose the economy is divided into j = 1, ..., can be given in terms of transport, where the J sectors. Within each sector j there are k = Akj represent different levels of transport 1, ..., Kj activities, each involving some activities, the Tmkj represent different vehicle energy use. Suppose further that each types or modes, and the Fntj represent fuel activity k can be pursued through the use of efficiencies. m = 1, , . - Mk technologies. Finally, suppose With this notation, the following energy each technology uses some subset of final quantities can be defined: energy typesi= l,...,I1 Ei,jkj = total amount of energy type i used by technology m for activity k in sector j 7Additional discussion of the model can be found in Cofala (1985) and Cofala et al (1990). 15 = AkjTmkjFmkjSimkj (3.1) industrial economy. It is almost inconceivable for a transitional economy. Eqy = total sector j demand for energy type Thus, we have been forced to rely upon more heuristic projections of the composition of Polish economic activity, prepared by a Ki Mk variety of sectoral experts consulted by = X XEimkj (3.2) members of our research team from the k=lm=l Polish Academy. These compositional projections are combined with World Bank Ei = aggregate demand for energy type i macroeconomic forecasts and other judgments to develop the Akj. Starting with a base year of 1990, projections are made at X Ei1 (3.3) five-year intervals to 2015. j=1 Similar problems arise in projecting choices of energy-using technologies, choices Ej = total energy demand in sector j of fuel types, and trends in energy intensity. Technology and fuel type shares are specified it heuristically using sector experts' judgments. = XEiJ (3.4) Base Case energy intensities and thus total i=I energy requirements also are calculated in this way (the Base Case is described further in the Note that a time subscript has been next chapter). The expert judgments include suppressed in this formulation. In practice, assessments of how the size and composition all the components of final energy use will of energy demand will respond to higher vary over time. Economic activity and energy prices and economic restrucuring. technology choices in the industry, transport No effort is made to link changes over time in and other sectors will change in response to Base Case energy demands to assumed general economic restructuring and increased energy pnce trends and simple (point) energy prices. Fuel choices and energy elasticities, out of fear that such calculations intensity sinilarly will respond to changing would give misleading or implausible answers economic incentives-in particular to in light of the large energy price increases incentives for energy conservation. Poland has faced and the massive change in To operationalize this framework the structure of energy use that is occurring. requires projecting each component in Past experience provides little guidance in equatior (3.1) over time, which is donc here assessing these changes given the size of the in two steps. The first step is projecting price increases and the fact that conventional energy-using activity levels in the economy market responses did not operate in Poland in (the A . Projecting these activities requires the past. the ana;yst to take a view on how the entire On the other hand, our approach to structure of the Polish economy will evolve. calculating Base Case energy demands has a In principle, this could be done with a potentially serious drawback in the policy dynamic general equilibrium model dtat scenarios. As noted in Chapter 2, one included invesunent behavior, trade linkages, element of the social cost of abatement is the and macroeconomic effects. However, such a reduction in consumer surplus when ramework is daunting even for an advanced abatement expenditures raise the price of 16 cnergy services and other commodities. factors are used to express enissions as a Without some allowance for a response of function of total activity. energy demands to higher prices, the Process emission factors for S02 and capability to include this effect in the social NO, are used to describe emissions from abatement cost calculation is lost. production of agglomerate, pig iron, raw steel To ameliorate this problem, in the (from open hearth and oxygen furnaces), policy scenarios we adjust energy demands copper, cement, nitric acid, and nitrogenous relative to the Base Case by using a set of fertilizers. Process emission factors also are simple own-price elasticity assumptions for used for keeping track of particulates, since broad energy categories, along with estimates PM emissions depend on how coal is of delivered pricc, increases reflecting combusted. Emiissions from the transport abatement costs, pollution charges, or prices sector also vary across the policy scenarios, for traded permnits bome by energy service given different assumptions about the suppliers. In light of significant uncertainties imposition of control measures, e.g., catalytic about cross-price effects, we make no effort converters. These options are described in to include them. The demand adjustment is the next chapter. made iteratively. First, energy supply costs are calculated under a policy scenario with LEAST-COST ENERGY SUPPLY the Base Case demands. The calculated increases in delivered energy costs over the Many final energy types can be supplied in a Base Case are used with the elasticities to variety of ways. This is obvious in the case of adjust demands. Then the new demands are electricity and district heat, but it also applies used to rerun the energy supply module for example to coal (which can be extracted described below. In practice, only one or two from different domestic mines or imported). iterations are needed to get convergence.8 Different supply options have associated with Along with projecting final energy them different costs and air emissions. demands, this part of the analytical Moreover, there are intertemporal links framework keeps track of air emissions from among supply decisions-to cite an obvious final energy use in different economic example, a decision to invest in new power activities. (Emissions from energy conversion plant capacity has consequences for the costs are tracked in the optimization part of the of meeting future electricity demands. model described below.) In many cases, Once final energy demands are emissions follow directly from the quantity projected, the means for satisfying hese and type of fuel used. In other cases, demands are detennined as the solution of a however, emissions depend on the process dynamic linear programmming algorithm that used as well as the fuel. For example, sulphur minmizes the present value of supplying the from fuel may be absorbed in cement specified energy demands, subject to specified production, while additional sulphu beyond emission constraints and constraints inposed that in fuel may be released in copper by the capital stock. The nature of the production. In these cases, process emission emission constraints depends on the policy scenario being considered, as discussed in the 8As described in the next chapter, ;be iteraWon next chapter. process is more intricate when demands must be Te model includes as decision variables adjusted to reflect delivered cost increases and both the operation of existing plant and amission charges must be adjusted to satisfy specified equipment and investment in new equipment, emission targets. 17 including investment in cleaner new plant or prevailing enviromnental constrainmts, and a environmental retrofit. Existing plant may sequence of capital investments over dme. also be scrapped before it is fully depreciated. This sequence reflects changes in demand and Because the sequences of energy supply and environmental constraints, along with the conversion activities are the solution of a dynamics of capacity depreciation or dynamnic optinization problem, they take into scrappage and exogenously specified limnits on account the intertemporal linkages among total investment resources. decisions noted above. Minimization of the present value cost Over 70 technologies and 22 primary of energy supplies is the outcome achieved by energy types are included in the framework, an ideal competitive market. Thus, an including energy imports and exports. urtportant assumption in our use of the cost- Among the most important energy conversion nim on model is that actual energy processes considered are coal and lignite fixed markets in Poland wil perform effectively power plants, combined heat and power over the rime horizon. In pardcular, we are plants, district heating plants, and power or assumingthat prices reflect full margial costs heating plants operated by industria of energy supplies and tat individual enery concems. Enviromnental abatement options suppliers respond cost-effectively to these include hard coal cleaning, wet and dry price signals. These assumptions seem process flue gas desulphurization in coal and reasonable, as noted in Chapters 4 and 6, lignite plants, flue gas denitrification, and although some government regulation of the particulate collectors (bag filters and energy sector (especially electricity) and electrostatic precipitators). Other options subsidization of energy producers (e.g., high- include new investnent in low-emission cost coal nmines) occurs even in advanced energy conversion processes, such as gas industr countries." combined cycle plants, coal-buming fluidized It should also be noted that the energy bed plants, coal gasification, and residual oil supply and conversion model, and the energy gasification. denand projections, are deteministic. A The optimization model includes capital non-stochastic approach is a disadvantage for costs and fuel and operatng costs for both our policy analysis, in that we cannot explore primary fuel production and energy differences in the performance of emission conversion processes.9 Production and fees and permnit systems in the presence of conversion capacities constrain utilization of uncertainty about emissions and abatement facilities in any period, based on assumed costs.12 We regard this as an important capacity availability factors.'0 The model subject for further research. solution provides both a "merit order" for technology use in any period, subject to IMoAwer impont concer is whether the levels of investment calculated by the model are feasble in the 9To limit a bias in he model away from capital- capital-stapped Polish economy. Investment levels intensive outcomnes in periods toward ibe end c. the implied by tue model are discussed in Chapter 5. decision horizon, capital costs include only the 12With enissiou fees, the equilibrium marginal cost amount of investment that is depreciable under of abatement (equal to the tax) is predetermined but normal conditions within the horizon. In addition, emissions are aOL The opposite is true with permit rsults in Chapter S are reported only to 2010 to avoid trading. The outcomes with fees and pesmits end period bias. therefor diverges in the presence of uncertainties 10Tbese factors rauge from 0.74 for new coal power about emisions and costs (see Weitzman 1974 for plants down to 05 or less for combined Feat and fortber disussion). The complexity of the model power plants. used bere precluded even a ess dirct assemnewn of 18 MEASURING SOCIAL COSTS OF ACS. Using the definition of demand ABATEMENT elasticity, this loss can be approximated by Having introduced the simulation framework, 2 we now return to the issue of how the so ACS=O.5(Q0-Q')NP0IQ(I/ED) (3.5) cost of polution abatement can be calculated using the mode. To do this we refer again to where CD is the absolute value of he price Figure 2.1. The social cost in that figure is elasicity of energy dad given by the represented as Area (A+B+C). cure D. Suppose first that So rpresents the The above argument applies only to the aggregate marinal cost curve of energy fiTmsition of fixed standards on large supply in the Base Casc, without newly stationary sources in the opdmizaion model. instituted controls. The area under SO outeo However, it ts oily erized to other the Base Case consumption level QO, Area policy scenarios. For example, with emission (E+F), is the Base Case value of the objective or fuel taxes the schedule S1 is the marginal function being mninimized in the opimization cost of energy supply, including the tax. To model described above. Now suppose that detemiine the social cost of pollution control, 51 is the aggregate marginal cost curve for it is then necessary to net out government tax energy supply, with some set of revenues (a transfer paynent).'3 In addition, enneronmental standards applied to the large some pollution control measures are imposed stationary sources included in tle on final energy demands rather than on optimization model. The area under dzc prmary energy production and conversion. curve Sl out to the new consumption levcl Impoytant examples of such controls in his Q1, Area (A+E), is the new value of the stu+r are rstrictions on household coal use objective funimon in opmizadon. Tlhe and vaious transportation measures like difference betwen the two values of f catalytic converter requirments. These costs objecive finction thus is Area (A) are calculated separtly and added to control Area (F). If we add to dtis the amount costs indicated by the optimization model. PO(QO-Q1) = Area (F+C), the value of rduced energy use at the original prie P0, the result is Area (A+C). We have thus far shown that two of fte three components needed to measme the social cost of imposing abatement standards on large stationary sources-increases in costs of farginal energy supply and a drop in producer surplus fromn reduced energy use--ca be recovered direcdy from thc optimization model plus a simple aridtnedc adjustment. The third component, the consumer suplus loss, is represented by the area of tiangle B. We denote this cost by t3No sucb netng is needed with emso tading uncertain pact thogh sensitivity analysis (we when permits ae anifadhred to sourc and no Dowlatabadi and Toman 1991 for an ilustaton). revenue is transfend to the govemmeaL Chapter 4 DESIGN OF SCENARIOS We describe the basic economic assumprions economy. It shows substantial projected underlying the model runs in the first section; declines m ferrous metallurgy and relaively the different sets of environmental standards static behavior in other industres, while home exnamined in the study in the second section; construction and private vehicles expand and, in the third and final section of the considerably. These assumed changes reflect chapter, we describe the different economnic the judgments of various Polish experts based policy scenarios examined, as alternative on official national statistics, assessments of means to meet a particular set of standards. sectoral perfornance, and demographic Detailed numerical infornation on the assumptions (population change, household scenarios can be found in Appendix I (Tables formation). Al.l-Al.15). Energy prices are the other key element needed to establish the scenarios. Projections BASIC ECONOMIC ASSUMPTIONS of intemational prices for (traded) primary fuel types used in the simulations are shown Table Al.l lists the basic economic output in Table A1.3. These are broadly consistent projections on which the energy analyses rely. with World Bank and other international Total GNP in 1995 is assumed to be five projections. Over 1990-2010, world crude percent above the 1990 level, reflecting oil prices are projected to rise by about 50 recovery from the output shock of the early percent; coal prices by about 20 percent; and 1990s. GNP then grows at about 4.5 percent gas prices by around 40 percent. each year until 2000, a figure consistent with Domestic energy prices to end-users projections by the World Bank and by the depend on interaonal prices, domestic Polish Ministry of Industry and Trade primary production costs, conversion costs (Bojarski et al. 1992). After 2000, a four for electricity and district heat, delivery costs percent annual growth rate is assumed (particularly transmission and distribution (Cohen 1991), a relatively moderate costs for the network energy sources- assumption (Rolo and Stem 1992 propose 5 electrcity, district heat, and gas), excise percent annual growth in their "optimistic" taxes, and on assumptions about economic scenario). pricing of energy to end-users. Prior to 1990, Substal change in the structure of energ prices in Poland were far below the economy also is assumed. Table Al.1 economic costs. Over the subsequent two shows projections of a considerable decline in years, considerable progress has been made in industn's share of GNP, along with rationalizing energy pnces Energy prices to significant decLines in construction, while industry by mid-1992 were at or close to "other" (the service sector) grows economic costs. Mine-mouth coal prices substantially.14 Table All2 looks in more were at export parity, and liquid fuel prices detail at energy-intensive sectors of the reflected al production and delivery costs. as well as substantial taxes in the latter case. 1*4he projections are based on nationa income Prices of network fuels to households in mid- accounting convetons derived fm the previous 1992, while well above 1990 levels, still economic regime, which placed lide emphasis on meanring services. 20 needed to increase another 50-100 percent to ALTERNATIVE EMISSION reflect economic costs. STANDARDS A basic assumption underlying the study, as noted in Chapter 3, is that fiul In the Base Case scenario (BASE) no new economic pricing of all final energy sources environnmental policies are employed, but wil be in place by 1995: price is assumed to some policies in non-energy industries that reflect the full marginal cost of supply plus reflect both better resource management and any excise taxes that are applied. We also lower emissions are continued (e.g., in assume that taxes on liquid fuels, in metallurgy and cement manufacturing). In percentage terms, wil rise to Westem the transport sector, a new generation of European norms by 1995. These assumptions diesel engines, with emissions comparable to generate the projections of final energy prices current West European standards, is shown in Table Al.4; Table A1.5 records the introduced (tough existing trucks do not delivery and tax markups. The prices in face accelerated phaseout). These engines Table A1A underlie the projection of final produce 20 percent less NOx than engines energy demands in the Base Case, without curnendy used in Poland. additional environmental initiatives that Relative to the Base Case, four different elevate prices. For the network energy sets of standards are considered. We label sources, the prices are broadly consistent with these Polish "Command-and-Control" economic costs calculated by the supply Standards, European Community Standards, optinization part of the model described in German Standards, and Flat Rate Reductions. Chapter 3.15 Environmental policies that raise Polish "Command-and-Control" Standards delivered final energy prices relative to the (CAC) Base Case also reduce demands relative to the Base Case. The own-price dasticities The CAC scenano mposes plant level used to make these aemand adjusments are performance standards for large stationary shown in Table A1.6 (as alkady noted, cross- sources (expressed in pollutants per uit fuel price effects are ignored). These figures in input), consistent with limits defined in 1990 Table A1.6 are broadly consistent with other by the Ministry of Environmental Protection, estimates of long-run elasticities (Bohi 1981, Natural Resources and Forestry hi its Bates and Moore 1992). We emphasize that Ordinance on the Protection of Air Against the elasticities shown are not used to obtain Pollution. The lmits on SO2, NO., and PM the time path of Base Case demand from stationary sources are shown i Tables adjustments, given the sharp price increases Al.7-AI.9. The standards for "new' since 1990. installations (begun after the Ordinance and put into operation after 1994) start in 1995. Existing sources must meet weaker interim standards until 2000, the second year I iManOna cos isna easa defime in a mlticmreported in the projection analysis, after '5Margial cost is not easily defined in a multioniput whtichi they must meet standards thar are multiperiod linear programming model in which tich than the inte standards bta ake small changes in activity levels can have nonmargmnal effects on the dual variables. Nevenheless, the than the new-source standards (in the in Table Al .4 can be shown essentially to cover the Ordinance, the actual changeover year is fuR market costs of enargy supply in the absence of 1998). The nature of these standards is requirements for emission abatement. 2L spelled out below.'6 Note that these arc plant European Community Standards (EEC) level standards, so some intraplant emission trading across sources is assumed to be The EEC scenario uses target reductions in possible even in this case. S02 and NOx emnissions from existing In addition to stationary source sources that are embodied in the 1988 EC controls, our CAC scenario includes a Large Combustion Plant Directive (IEA number of regulations for the household and 1992). These call for SO2 reductions of 25 transport sectors which, while plausible, have percent by 1993, 43 percent by 1995, and 60 not yet been incorporated in Polish percent by 2003. NO1 reductions from environmental policy: a ban on urban coal existing sources are to be 20 percent by 1993 use by households and a requirement for and 36 percent by 1998. In the model, the catalytic converters on all cars. These S02 reductions are applied to Poland in restrictions are phased in to be 50 percent 1995, 2000, and 2005 respectively; the NO1 effective in 2000 and 100 percent effective in reductions are applied in 1995 and 2000. 2005. After 2000, an additional 30 percent In addition to existing source controls, reduction in heavy diesel engine emissions the EEC scenario includes controls on new also is imnposed, in line with US norms, and stationary sources. These standards are the sulphur content of diesel oil is assumed to country-specific, but typical standards are as drop from 0.6 percent to 0.15 percent. shown in Tables Al.10-Al.12.1 They are It is important to note that our assumed to apply starting in the frt application of the Polish large source projection year, 1995. Finally, the ban on standards is at the plant level, rather than at urban coal use and the catalytic converter the level of individual sources (e.g., boilers) requirement in CAC are carried over to EEC. as the standards are originally stated,17 To simplify computations, we set final reflecting the fact that the model is concerned demand in EEC at CAC levels, ignoring how with types of plants rather than separate icreases in final energy price relative to CAC sources within plants. In applying the affect demand. standards at the plant level we implicitly assume a capacity for trading emission German Standards (GER) reductions across sources within plants. As discussed in Chapter 6, this appears to be The GER scenario is similar to EEC but with consistent with how the Ministry's 1990 a different set of large plant constraints, as Ordinance currently is being interpreted in shown in Tables Al.13-Al.15 Existing Poland. Note also that with this definition of plants can meet interim SO2 standards until CAC, the scenario with S02 emissions 1993, after which they must be retrofitted to trading defined below should be interpreted meet new plant standards or closed. In as involving trading among as well as witiin plants. I'EC and German standards are actually expressed in mg/Nm3 of exhaust gas, which refers to cubic meters 161n practice there is a third category of plant. e.g.. of dry gas at 0°C and 1 atmosphere of pressure. To those started before the Ordinance was passed but put make the standards comparable to Polish limits, into service afterwards, which must meet the old- conversion factors of I mg/Nm3 - 035 g/GJ fuel source fuel standards as interim standards but convert input for coal plants, I mg/Nm3 = 0.30 g/GJ fuel to new-source standards by 1998. input for oil plants, and 1 mg/Nm3 - 0.31 gIGJ fuel 1'7This also applies to our scenarios with European input for gas plants were assumed. These ar Community and Gennan emissions standards approximate conversions that ignore many differeuces defined below. in fuel quality and technology. 22 adapting the S02 standards to Polish Polish NOx standards are stricter than application we apply interim standards to EC and even German standards, particularly existing plant in 1995 and tougher standards for grate firing coal plants and small gas fired in 2000 and subsequent projection years. plants. For PM, in contrast, Polish standards New plants have to meet tougher standards are much more liberal; generally, PM control starting in 1995. The GER NO, and is much less of a problem in Western particulate standards also must be met by European countries. These characteristics of 1995. As in EEC, final energy demands are the control scenarios should be kept in mind left at the CAC level. since PM often is the cheapest pollutant to abate. As already noted, tne main difference Flat Rate Reductions (FRRED) between CAC and FRRED is that the latter does not discriminate among sources based The Polish CAC and the other two sets of on size or vintage. standards impose differential constraints on These differences in plant-level large sources; for example, new sources must standards influence the total emissions meet more stringent S02 standards than projected in the different control scenarios existing sources. With FRRED, we aim for just described.20 However, they are by no the samne total emissions as in CAC (and leave means the only factor. Nor are they a reliable final energy demands unchanged), but we guide even to the comparison of total allocate allowable emnissions over time for emissions from new and old sources. Other each large source type in proportion to its influences include the distribution of plant by contribution to unabated emissions in fiel type and age, and the relative cost of BASE.'9 The comparison of CAC and different control strategies. These ERRED thus can reveal the cost compaisons are discussed in Chapter 5. consequences of differential treatrnent of large sources. ALTERNATIVE ECONOMIC INSTRUMENTS Comparison of Plant-Level Emission Standards The second set of scenarios we consider involves the application of different types of With respect to S02, EC and (especially) economic policy instruments. With one German plant-level standards are much exception, these scenarios seek to achieve the stricter than Polish standards for large new same total emissions as those projected in facilities. Much the same is true for existing CAC (based in part on 1990 Polish legal plant. In contrast, for small new facilities standards). (e.g., mechanical grate and other boilers with capacities less than 100 MWth), Polish Uniform Emissions Taxes on Large standards tend to be tougher (compare, e.g., Sationary Sources (ETAXI) the Polish limit of 22 g/GJ for new mechanical grate boilers to the German ETAX1 is based on the same final energy standard of 700 g/GJ for small boilers). demands as CAC, including the ban on urban coal use and the restictions on transport ETiissions (e.g., catalytic converter IThbis is the case for both existiang enmitters at the start of the simulation and new emitters added during 2t0hey also allow us to account for how new source the simulation. bias may affect the pattem nf plant invesmenL 23 requirements). From this starting point, emitters face a declining sequence of emissions taxes on large-source eniissions of endowments over time, consistent with the S02, NOx, and PM are found that yield the time path of eniissions in CAC (see Chapter same national emnissions as CAC. Because 5). New eminters must acquire permits to final demands are the same as CAC. this case cover all their emissions, starting from their indicates the effect on total abatement costs initial years of operation.23 of more flexible large-source emission controls compared to fixed plant-level Coal Tax (COAL TX) standards.21 COALTX, a 100 percent tax on coal, the SO2 Trading by Large Sources (S02TR) most polluting fuel, is imposed (coke, a smokeless fuel, is exempted). No regulatory SO2TR uses the samne starting point as restrictions on emissions (in energy ETAXI (CAC final energy demands, ban on conversion, households, or transport) are urban coal use, transport controls). It also imposed, and no attempt is made to achieve imposes the CAC plant-level standards for emissions parity with CAC. Total final NOx and PM. For large-source SO2 control demands also differ from CAC; they are (all the sources in the optimization model), a calculated by altering BASE demands to national market for S02 permits is simulated reflect the effects of the tax on prices of coal, by solving the optimization model with a limit electricity, and heat. only on aggregate S02 enmissions (versus plant-level standards that allow only trading Uniform Emission Taxes on All Sources among sources within plants). Allowances (ETAX2) are not bankable (so we are not simulating an intertemporally first-best SO2 trading system ETAX2 relaxes all fuxed standards and that would allow more irregular enmissions controls, including those on households and over time).22 transport, while imposing SO2, N0x. and PM To calculate compliance costs for emission taxes on all sources to yield the traders, initial permit allocations are same total emissicns as CAC. Large energy- grandfathered to existing emitters in 1990 conversion sources respond to taxes in the based on 1990 emissions shares. Existing optimization model as in ETAX 1. Household transport and other end-use emissions are 21Note that because the model does not distinguish curbed by higher fuel taxes, set to be the location of pollution sources, this scenario equivalent to the large-source emission taxes, generates uniforrn national emission tax rates foT based on the average emissions per unit of large-source emissions of NOx and PM as well as fuel use.24 Compared to ETAX1 this 502. Such uniform taxes are not socially optimal from the standpoint of pollution control since the pollutants in question do not mix uniformly at a 23Th. total emission er4owment is not quite the same national level and different regional control severities as total CAC emissions because certain small sources generally are appropriate to reflect variations in -such as municipal beating plants and some sources ambient conditions. ETAX1 should be interpreted of process emissions-are omitted from trading. only in mrlation to CAC, not as an endorsement of 24This scenario requires a two-level set of iterations. uniforn emission tax rates. For any set of final demands. enission taxes ame 22SO2TR also does not allow for trading between varied within the energy conversion optimization energy converters and final demanders (e.g., a local model until total emissions are (approximaLely) equal power plant paying to reduce emissions from to CAC levels. The "outer feedback loop" involves household coal use by subsidizing fuel substitution). successive revisions of final demand, starting at 24 scenario allows for a different distribution of SUMMARY emission control between the large stationary sources and other emitters since the latter Tables 4.1 and 4.2 summarize important also are subject to incentive-based characteristics of all the standards and instruments. Options for end-use fuel and economic policy scenarios we consider. As technology switching (e.g., voluntary shifts to alrady noted, the structure of the model catalytic converters or gas use in households) forces on us some compromises in defining that might be induced by high fuel taxes are policy scenarios that correspond to actual ignored. Such substitution possibilities lie policy design issues. Nevertheless, we beyond the scope of the model. At least in believe that the scenarios are capable of the transport sector, there is reason to suspect illustrating both the incremental costs of that such switching would not be cost- different standards and the cost consequences effective for the tax levels we impose (see of different economic policy mechanisms. Chapter 5). The comparisons are made in the next chapter. BASE levels, to eflect the effects of the emissions taxes. As a consistency check the optimnization model was run with only aggregate emission constraints (as if there were a single market for all sources to rade emissions). This outcome should be the theoretical dual to uniform emission taxes on all sources set to yield the same total emissions. In practice the result in the consistency check was essentially the same as in ETAX2, taking into account the inevitable rounding prDblems. 25 Table 4.1. Characteristics of Alternative Standards Scenarios CAC FRRED EEC GER Baseline final energy BASE CAC CAC CAC demand Demand adjustment for Yes Yes No No incremental cnergy (reflecting price increase relative different to baseline distribution of control effort) Nature of stationary Plant level Plant level Plant level Plant level source emission (Tables (Pro rata (Tables (Tables standards A1.7-AI.9) reductions A.1O-Al.12) A1.13-AI .15) relative to BASE; total emissions same as CAC) Household and Yes Yes Yes Yes transport sector controls 26 Table 4.2. Characteristics of CAC and Alternative Economic Instrument Scenarios | CAC ErAXI S02TR ETAX2 COALTX Baseline BASE CAC CAC BASE BASE final energy demand Demand Yes No No Yes Yes adjustment for incremental energy price increase relative to baseline Nature of Plant level None; CAC plant None; None (coal stationary (Tables national sids for NO ' national denand source Al.7-Al.9) emissions PM; nationa emissions taxed) emnission targets = so0 targets = standards CAC emissions CAC emissions target = CAC emissions emissions Htuitsehold Yes Yes Yes No (fuels are No (coal and taxed based use taxed) transport on average sector emissions per controls Ulit) Chapter 5 MODEL RESULTS We present in this chapter a summary of the Corresponding declines in coal and lignite most significant results derived from the shares, and increases in oil and gas, are simulations. Details can be found in obverved in primary demand. Appendix 2 (Tables A2.l-A2.14). Table A2.3 lists Base Case emissions by type and sector. Except for C02, all of the BASE CASE SCENARIO aggregate emissions are lower in 2010 tlhan in 1988. The declines by 2010 are 7 percent for Table A2.1 sumnrnarizes information about the S02, 3 percent for NO,t and 44 percent for time paths of energy demand in the Base PM compared to 1988. PM emissions decline Case. GNP in 2000 is somewhat above the monotonically, while S02 and NO, increase 1988 level, but both primary and final energy after 2000 but still do not return to 1988 demands remain considerably below 1988 levels. Concerning trends in the distribution values; energy intensity contracts by 20 of emiissions by sector, PM drops fairly percent over 1988-2000. Even in 2010. when consistently across sectors, whereas the fall in GNP is well above the 1988 level, the NO, and S02 occurs in final use and in continued drop in energy intensity of GNP industrial boilers. Some technological factors leads to a much lower growth in total energy help to explain the patterns in Table A2.3: demands (energy intensity in 2010 is 40 improved diesel engines, with 20 percent percent less than in 1988).25 Table A2.2 decreases in NOx emissions, enter the amplifies this point by showing relative scenario, and particulate emissions drop from changes in sectoral energy intensities. better control in some industrial processes Although energy intensity in the residential and a limited shift toward coal beneficiation sector retums to the 1988 level in 2000, and (even without new environmental controls). rises slightly thereafter, heated dwelling space is projected to increase over 40 percent ENER(;Y AND EMISSIONS relative to 1988, so that residential energy per DIFFERENCES ACROSS STANDARDS m3 of living area actually declines. In all SCENARIOS other sectors, energy use per unit of value added declines, even after recovery from the We consider here the impacts of different output contraction of the early 1990s. environmental standards (CAC. EEC, GER, Significant changes also are observed in and FRRED) Final energy demand in CAC the structure of energy demands. As shown (and thus in the other scenarios just listed) is in Table A2.1, the share of final demand for lower than in BASE, as a consequence of solid fuels drops sharply, while gas, liquid higher prices from environmental controls, fuels, and electricity increase their shares.26 but the difference is minor-only 2-3 percent.27 Thus. we focus here on primary 25Over 1988-2010, the elasticity of final energy demnand with respect to GNP is only 0.12. much lower elasticity of electricity demand with respect to GNP than comparable experience in Western economies. over 1988-2010 is still only 0.8. well below past 26Note that the relative electricity intensity of GNP patterns in Western economies. remains well above overall energy intensity (and even 27Recall Erom Table 4.1 that by assumpfion. final increases during the early 1990s). Nevertheless. the demands in EEC and GER are the same as in CAC. 28 energy demands, where the differences are 40-50 percent, 30-35 percent and 30 percent still modest but slightly more pronounced. below CAC respectively. Table A2.4 summarizes the results for It is also interesting to note differences primary energy demands. Coal demand in across scenarios in the sources of emission CAC is about four percent lower than in reductions in Table A2.5. For S02 BASE in 2010, while gas demand is about reductions, all the control scenarios except eight percent higher. As expected, ERRED put a greater burden relative to environmental standards make a switch BASE on new sources than existing sources: toward lower-emitting fuel more attractive, a whereas new power and combined heat and phenomenon that is even more pronounced in power plants in BASE and FRRED are FRRED. While total primary energy demand responsible for 10 percent of S02 emissions and its sectoral pattem under EEC are very in 2000 and 28 percent of cmissions in 2010, similar to CAC, there is greater switching to they generate only 2 percent of S02 lower-emitting sources with the stricter emissions in 2000 and 6-8 percent of German standards. It is interesting to note emissions in 2010 under CAC, EEC, and that the Gennan standards require more GER. primary energy than the simple pro rata enmission reductions in FRRED, to achieve the ENERGY AND EMISSIONS same final energy demands, reflecting the DIFFERENCES ACROSS ECONOMIC need for greater primary energy application to POLICY SCENARIOS achieve the tighter GER standards (e.g., through flue gas desulphurization). We next consider variations between CAC Table A2.5 shows the trajectories of and scenarios involving economic policy emissions by source and type with the various instruments. Table A2.6 indicates that the standards. We noted in the previous effects of economic incentives on primary subsection that, with the exception of C02, energy demands are minor, in the case of emissions in BASE (without new ETAXI and SO2TR. With regard to environmental initiatives) fall off considerably emissions, the main interest lies in the Telative to 1988 levels. In CAC, the new distribution of emissions across sources, since Polish legal standards produce a further the aggregates are (essentially) identical by substantial drop in emissions, again with the construction.28 A more striking feature of exception of C02. In 2000, SO2 and NOx EITAXl and S02TR, as shown in Table A2.7, ermissions decline by roughly a quarter is the alleviation of new source bias for SO2 relative to BASE, while PM declines by about in ETAXI and S02TR-new sources in a half In EEC, the S02 and NO reductions these scenarios have significantly higher S02 are similar to those in CAC out to 2000, emissions than in CAC. The other major while PM emissions are about 20 percent difference across the scenarios is that with the lower; however, by 2010, S02 emissions are greater 502 control flexibility in SO2TR and 12 percent lower than CAC while NO, is ETAX1, more SO2 abatement is undertaken about 2 percent higher. The GER controls on by power plants and less by combined heat S02, NO1, and PM are truly stringent, and power plants relative to CAC. The yielding emissions in 2000 and 2010 that are differences in NO, and PM abatement 'in some cases the iterative solution process over There are final demand differences between CAC and multiple periods does not generate exact agreement in FRRED, but they are minor. msults that in theory should be identical. 29 between ETAXI and CAC arc minor, However, emissions of S02, NOX, and PM suggesting that any cost advantages from are much higher relying just on the coal tax incentive-based control by large stationary an with more b^oad-based envirom-nental sources derive prmarily from flexibility m policies. S02 control. Tables A2.8 and A2.9 show the energy ECONOMIC IMPACTS OF EMISSION and environmental pattems for ETAX2, in CONTROLS which large stationary sources face emissions taxes, and other sources (household and Tables 5.1 and 5.2 compare the social costs transport) face (roughly) equivalent fuel of abatement across the CAC and economic taxes, reflecting the average emissions per instrument scenarios respectively, relative to unit of fuel use. Higher energy prices reduce BASE. As described in Chapters 2 and 3, final demand by about three percent relative these costs can be separated into several to CAC (five percent relative to BASE). The categories: increased costs of energy drop in primary coal demand is especially conversion and emission reduction; loss of pronounced. producer surplus; costs imposed on final Pattems of emissions differ in several energy demands in the household and significant ways between ETAX2 and transport sectors; and losses of consumer ETAXI. Recall that, in the latter case, only surplus from reduced final energy demands. large stationary source emissions are taxed, All the costs in Tables 5.1 and 5.2 are present while final demand sources (households and values over 1990-2015, calculated at a real transport) face direct controls. In ETAX2, discount rate of 12 percent.30 S02 and NO1 emissions from final users are Table 5.1 shows that, compared to higher than in ETAXI or CAC; the fudi taxes BASE, the additional cost of enission control have less effect on these emissions sources in GER is about 70 percent higher than under than direct controls. As a consequence, any of the other sets of standards (CAC, eniissions from energy conversion sources FRRED, or EEC), reflecting the sharp cuts in must be lower to meet the overall targets. S02 and NO, emissions from stationary Like ETAX I, ETAX2 has less bias against sources under GER. Given the interest in new S02 sources than in CAC. For NO., on Poland in harmonizing with European the other hand, it is cheaper in ETAX2 to Communiity standards, it is interesting that the acconunodate higher end-use emissions by sharply restricting emissions from new power and from combined heat and power plants than to restrict emissions from existing stationary sources further. The COALTX scenario is suninrized in Tables A2.l0 and A2.1 1. Predictably, the swingeing (100 percent) tax on coal households in 2000and 2010. The excise tax significandy reduces its use relative to BASE assumed here doubles those prices. and CAC: the final demand for solid fuels 3Obs discount rae is used to reflect thed .opect of drops 21-29 percent arnd pfimary demand fo r real capital scarcity and nomrivial financial risks over hardrops21-29by1724 percentand2primad de fext sevrral years. A lower discount t ae would hard coal drops by 17-24 percent.2' cause a relative incrase io the present value of costs for scenarios that involve higb capital investmncs in "Table AlA indicates that Base Case hard coal later yeahs of the decisioo horizon, such as GER and prices am $55-60/ton for industry and $93-98ton for ETAX2. 30 Table 5.1. Social Cost of Pollution Control, 1991-2015 (CAC) (US$ billions)Zi SCENARIO Cost component CAC FRRED EEC GER Large point source control 6.57 6.73 6.56 15.43 costs and loss of producer surplus _ Lost of consumer surplus 0.02 0.02 0.02 0.02 Switch to gas by urban 0.08 0.08 0.08 0.08 households Transport controls 5.89 5.89 5.89 5.89 Total 12.56 12.72 12.55 21.42 a/Discounted to 1990. 12% real discotint rate. Pollutants controlled are PM, S02, and NO,; see Chapter 4 for definitions of scenarios. Table 5.2. Social Cost of Pollution Control, 1991-2015 (Economic Jnstruments) (US$ billions)21 SCENARIO Cost corn2onent ETAXI S02TR ETAX2 COALTX Large point source control 5.30 5.87 5.51 9.76 costs and loss of producer surplus Loss of consumer surplus 0.02 0.02 0.22 1.04 Switch to gas in urban 0.08 0.08 households Transport controls 5.89 5.89 . Total 11.29 11.86 5.73 10.80 a/Discounted to 1 990. 12% real discount rate. Pollutants controlled are PM. S02, and NO.; see Chapter 4 for defiitions of scenarios. 31 cost of meeting CAC is roughly the same as incentive-based policies. In pzarticular. the EEC.31 model treats all potential tecitilologies and Table 5.2 shows the costs of enission other strategies for entissions control as heing control with taxes and (interplant) emnissions known in advance. In praclice. incentive- trading relative to BASE.32 These figures can based policies will provi(le additional m-otives be compared to the performance of CAC for developing new techlnologles and (which allows for intraplant trading) in Table strategies. as noted in Chapter 2. 5.1. For policies applied to large sources Table A2.12 expands on the diifference' (ETAXI and S021R), the cost savings over between stationary an(d tramsport controls 1v CAC are fairly modest, partly because fairly comparing the unit control cost of littfer-" expensive controls in the transport sector emission reduction strategies used in the account for almost one-half of total emission transport sector under CAC. relative to other control costs (relative to BASE) in all these options. While low-NO, diesel enigines cases. When these controls are dropped and appear to be cost-effective methods for economic incentives are applied to all emission reduction, catalytic cornveriers andi pollutants, as in ETAX2, control costs are diesel oil desulphurization are very more than halved relative to CAC.33 expensive.14 It does not follow, however, This comparison understates the that these transport controls should be potential contribution of an incentive-based avoided. The unit costs of emission controls approach since. as noted previously, our alone are not a sufficient basis for judgment. version of CAC already assumes significant since the contribution of different sources to flexibility for polluters in restricting emissions ambient air quality and the relative from separate sources. Another important harmfulness of different pollutants also affect source of understatement is that the model the social interest. Mobile sources contribute ignores dynaric efficiency benefits from onmy a small percentage of total emissions. but they could be disproportionately important in We emphasize. however, that the environmental urban centers. Moreover. their inportance consequences of the two sets of standards are not may grow over time (see Chapter Ix These likely to be the same. In particular, West European issues are beyond the scope of the present standards on S02 from large plants are tougher than study to address. Polish standards, while the reverse is true for small A plants. On the other hand, the EEC standards for PM Accordig to the model, ivestment are much tighter than in CAC. outlays for the energy supply and conversion 32The total cost of the COALTX scenario also is system over 1991-2010 are substantial, even reported in Table 5.13. The main point here is thrt without new environmental controls. In there are considerable cosrs incurred for quite modest BASE, total undiscounted outlays are emission reductions relative to BASE: the coal tax by US$65.7 billion over the period, with about itself is a very cost-ineffective policy. 33As noted in Chapter4, we do not consider the 40 percent of these outlays (US$27.1 billion) possibility that residential and transport users wouid occurring over 1991-2000. the period of voluntarily switch to lower-emission technologies as a significant economic restructuring. The consequence of the fuel taxes they face in ETAX2. Such conversion will be efficient if the present value of surplus losses from the higher taxes exceeds the 34The total cost figures in Table 5.2 indicate an even opportunity cost of the new technology investment greater spread in cost-effectiveness: comparing (including the cost of prematurely abandoning ETAXI and ETAX2. we find that the large cosi of existing equipment). Thus our social cost figure for mandated controls can be avoided Nvith onlv a modest ETAX2. which includes these surplus losses, may increase in stationary source controls along with a overstate the cost of this scenario. modest decrease in consumer surplus. 32 increases in total investnent oudays under The dramatic increases in taxes which some of the environmental policies we are implied by ETAXI and ETAX2, and the consider are shown in Table A2.13. As might resulting increases in final energy prices, be expected, the stringency of environmental would spark social controversy. standards in GER causes a substantial jump in Unfortunately, the lack of focus in the model energy supply system investments (41 percent on individual sources makes a detailed over BASE from 1991-2000 and 26 percent assessment of financial burden impossible. over 1991-2010). The increases in CAC and However, some outputs of the model do help ETAX2 are smaller, on the order of 7-9 to shed light on the issue. percent (only 2 percent over 1991-2000 for One concrete indicator is that the tax ETfAX2). increases in ETAXI and ETAX2 are Differences in standards, control calculated by the model to generate additional methods, and taxes across the scenarios also revenues with a present value of about US$9 generate variations in energy prices. To billion by 2015. This makes the total cost of illustrate, Table 5.3 shows the increases in compliance (including increased fees) in final energy prices relative to BASE for CAC, ETAXI about 55 percent higher than in ETAX2, and COALTX. The effects of fixed CAC, even though the social cost of emissions standards on final prices is abatement in ETAXI is lower (see Table relatively modest (increases range from one 5.2). The compliance cost in ETAX2 is only to seven percent)- The price increases in about 9 percent higher than in CAC. ETAX2 are more substantial, particularly for reflecting the high degree of cost- coal, reflecting the impacts of the emissions effectiveness achieved with comprehensive charges. The COALTX scenario generates changes. In contrast, the distribution of substantial increases in electricity and heat trading profits from large-source S02 prices, as well as coal, though these increases emission trading, with grandfathering, is are eroded over time by substitution away shown in Table 5.5. While new sources face from coal-fueled technologies. nontrivial costs of permit acquisition and Table 5.4 summarizes the emissions fees some existing sources (especially existing in ETAX I and ETAX2 and compares them to lignite plants) make substantial trading Polish charges as of April 1, 1993. These profits, there is no net transfer of funds away charges have risen sharply since 1990, but the from polluters and the total compliance costs levels remain well below those required even appear to be much lower than under the tax to attain interim (1995) standards and are an approaches.35 order of magnitude too small for achieving The changed energy prices in all emissions targets in 2010. Note particularly scenarios have important distributional effects the need for large increases in PM charges. on households, but they are overwhelmingly Ultimately PM and S02 fees are the same due to energy price refonns rather than order of magnitude in ETAXI and ETAX2, environmental policies. In BASE, the share whereas the current PM charge in Poland is of household expenditure allocated to fuel only half the size of the current SO2 charge. and power increases between 1992 and 2000 As a matter of interest, the NO, tax of about US$1 ,000/tonne in 2000 under ETAX2 translates into charges of about US$0.06/liter on diesel fuel and US$0.03/liter on gasoline. 35Keep in mind that ETAXI and ETAX2 involve dcarges on NOx and PM as well as S02, whereas Table 5.5 descnbes expenses only for SO2 peTnits. 33 Table 5.3. Energy Price Indices to Final Consumers for Scenarios (Base = 1.0) 1995 2000 2010 Fuel/Scenario . Industry Housing Industry Housing Industry Housing Hard coal ETAX2 1.17 1.13 1.29 1.23 1.35 1.28 COALTX 2.00 2.00 2.00 2.00 2.00 2.00 Gasoline CAC 1.01 1.01 1.01 1.01 1.03 1.03 ETAX2 1.02 1.02 1.03 1.03 1.06 1.06 Diesel oil CAC 1.01 1.01 1.02 1.02 1.04 1.04 ETAX2 1.07 1.07 1.10 1.10 1.20 1.20 Fuel oil (heavy) CAC 1.00 . 1.02 . 1.04 ETAX2 1.07 . 1.34 . 1.35 Electicity CAC 1.02 1.01 1.05 1.03 1.07 1.04 ETAX2 1.10 1.07 1.09 1.06 1.10 1.06 COALTX 1.20 1.14 1.10 1.06 1.10 1.06 District beat CAC 1.03 1.02 1.03 1.02 1.04 1.02 ETAX2 1.09 1.06 1.08 1.05 1.08 1.02 COALTX 1.23 1.09 1.14 1.05 1.12 1.03 34 r'able 5.4. Pollution Fees (Actual) and Taxes for ETAXI and ETAX2 (US$/tonne) 1995 2000 2010 Pollutant 1990 1993 (end of yr) (April) _ . ETAXI | ETAX2 ETAXI ETAX2 ETAX 1 ETAX2 So2 28 75 476 430 536 541 789 780 NOx 28 75 1029 542 1029 1029 1029 2716 PM 7 38 177 176 640 640 640 640 Table 5.5. Costs and Profits from Sulphur Dioxide Trading S02 enmissions (thousand tonnes) Trading Trading gains category or losses 1990 2000 Initial S02TR _endowment Public power plants: -new (hardcoal) 0.0 0.0 112.1 -60.5 -existing (hard coal) 671.2 511.3 540.1 -15.6 - existing (lignite) 691.8 527.0 344.7 +98.5 Public CHP plants: - new 0.0 0.0 64.1 -34.6 - existing 259.7 197.8 193.9 +2.1 Industial power plants: - new 0.0 0.0 15.6 -8.4 - existing 344.7 262.6 230.3 +17.4 Industrial heating plants: - existing 229.3 174.7 143.3 +17.0 Municipal heating plants: - existing 97.4 74.2 103.4 -15.8 Total 2315.6 1747.6 1747.6 0.0 Note: Equilibrium permit price = US$ 540/tonne of SO2 35 by 95 percent for urban salary and wage- 2010 under CAC, EEC, ETAXI and S02TR. earning households, 75 percent for rural However, the main impact on coal production households, and 80 percent for retired comes from general economic restructuring individuals. After 2000 the expenditure share and energy price refonns. Coal use dropped impacts decline steadily. In comparison to sharply between 1988 and 1990, and while it these increases, the increases from all the recovers somewhat in the BASE case, it does environmental policies are calculated to be not return to the 1988 level even by 2010. nugatory (at most an additional percentage Several of the policy scenarios further depress point for rural households and retired persons coal production, but generally only by modest in the case of the large coal tax). amounts (the maximum being four percent in The impacts of different emission 1995 in FRRED). The major reason for this reduction strategies on the coal industry merit is that the shortfall in domestic coal and special attention, given its political and social lignite demand is largely offset by expons of significance. As seen in Table A2.14, there is hard coal. Hence, even a 100 percenT coal some reduction in coal production with all the tax, which generates revenues of US$2.8-3.6 policies relative to BASE in 1995, although bilion per annum from 1995, causes coal the BASE production level is achieved by production to fall by only about 2 percent. Chapter 6 INSTITUTIONAL ISSUES The simulation results in Chapter 5 suggest INSTITUTIONAL BACKGROUND-7 that there are potential savings worth pursuing in using incentive-based control The February 1990 Ordinance sets ambient policies. This finding echoes much of the air quality standards which, as already noted, previous work on environmental policy are generally tighter than in the West. Energy design in the US and other advanced combustion sources, with installations countries, where the potential savings appear exceeding 200 kW, face emission standards to be even larger, given the inefficiencies as well (see Chapter 4). Both ambient and engendered by rigid technology-based emission standards are set nationally and standards.36 Nevertheless, the actual cannot be strengthened by local authorities, advantages of incentive-based policies may except by declaring a whole region to be a fall considerably short of the theoretical ideal, "specially protected" area a costly option because of technical problems in structuring for the locality. such policies, and compromises struck to The State Inspectorate for address political concems. Environmental PRotection-an autonomous In Chapter 6, we explore this set of unit reporting directly to the Environment institutional issues to see how they condition Minister-has the task of monitoring both recommendations for air pollution policy in discharges and ambient conditions. Most of Poland. We first briefly describe the existing the work is done by Inspectorate officers and statutory, institutional, and monitorng laboratories at the voivodship (provincial) apparatus in Poland. We then lay out a set of level. Under the 1990 Ordinance, high general criteria for evaluating policy designs emitters (more than 1200 kg S more than -CAC, charges, and emissions trading. 800 kg dust) must continuously monitor their Using these criteria, we contrast the own emissions. Some monitoring also is perfonnance of these three strategies. done by the state Sanitary Engineering Bureau, which helps maintain monitoring sites. Monitoring occurs at three spatial levels. First, a state system tracks broad _____________________ trends in S02, NOx and PM based on results 36Asurvey by Tietenberg (1990) reveals thatigid from about 70 sites throughout Poland, paid policies for air poUution control wexe likely to cost 2- for by the state budget. Second, voivodship 20 times as much as ideal least-cost control measures. monitoring often is done in cooperation with Even with less than ideal incentive-based policies, heavy polluters mentioned above. Third, at estimated cost savings are impressive (Hahn and the local level, some factories monitor toxics Hester 1989). Note that most of these studies have and metals discharges on instruction from the been concerned with meeting ambient standards rather than the emissions standards of concern here. VoWodsh). With ambient standards, total abatement cost can be reduced futher by allowing relatively larger 37Tbe material in this section is based in part on emissions in locations wbere the standards are not interviews with members of the Ministry of binding. By construction. these savings are ruled out Environment and the State Inspectorale for in our study Environmental Protection. 37 Compliance responsibilities and efforts Table 5.4 compares the emission taxes are divided between voivodship offices of the needed to anain projected future emission State Inspectorate, which monitor to check standards with the actual structure of Polish compliance and compute pollution charges, fees. The sharp increase in fees between and the voivodship offices of water and 1990 and 1993 is widely believed to have environmental management, which issue stimulated emission reductions, but the permits and colect fees and fines. Peniits amount of emission reduction due to the fee specify discharge levels (the maximum increases versus general economic permitted, and the average and annual restructuring remains unclear (exceptions to discharges pernitted) and the use of fines and standards in enforcement contribute particular abatement measures (e.g., chinmey to this identification problem). Current fees height and diameter) whose installation is apparently are too low to bring total monitored by the Inspectorate. The pemnits emissions to levels envisaged in the 1990 are based on air dispersion modeling of the Ordinance, however. Recent legal actions impact of emissions on air quality, carried out also call into question the legality of by private companies, which must follow regionally differentiated fees (as opposed to strict guidelines. If the estimated pollutants standards) in order to encourage the greatest at specified locations, as calculated by the abatement in areas of high damage (Zylicz model. exceed the regulations, action will be 1992). required before a permit is issued to the Apparently the Ordinance currently is emitter (e.g., raising the chimney height). being interpreted to allow an emission-control Polluters can argue for higher emission limits "bubble" over a whole plant (groups of by recalculating projectcd ambient effects. stacks).35 However, emissions trading among Fees for "regular" emission levels within plants is not permitted under the present law, the penrnit are codified in the current even at a very local level. New air pollution ordinance and adjusted for inflation. Fines legislation currently under consideration may are based on a half-hour "violation episode" provide some encouragement for local and are paid based on amount of measured trading. More interregional trades among exceedance and tirne ($/kg/hr) until the large sources alo are conceivable. provided polluter can prove compliance has been local ambient standards are not breached. reestablished. Flagrant violation-such as clearly not having functional abatement THE EFFECT OF RESTRUCTURING equipment-can lead to partial or total plant ON ECONOMIC INCENTIVES closure and niisdemeanor criminal proceedings against the operator, but this is Important changes in economic policy. rarely done. Some level of informal which started to take place in Poland after regulation also occurs, based on negotiation 1988, are expected to transform the way in e.g-, permits are issued and fines are which polluters respond to economic calculated at the same time where meeting the incentives. In the past, a central plannimg standards clearly is impossible. In some approach to resource allocation, relying cases, however, the emissions are too high to extensively on physical planning targets issue a permit and the regulator sinply has no combined with soft-budget constraints. leverage over the plant, which keeps seriously undermined environmental operating. 38PemsonaI communication, Professor Tomasz Zvlicz. 9 March 1993. 38 policy.39 As argued by Zylicz (1993), measures designed to facilitate the "financial instruments were doomed to restructuring of enterprises, using the state- failure in an economy where all essential owned commercial banks as key agents of inputs were allocated administratively, and change. Given the fundamental role to be plant managers had little incentive to pay assumed by the banks, the EBRP also aims at attention to price stimuli." However, recent stimulating their efficiency, including reforms in the enterprise and banking sectors (ultimately) privatization. A crucial outcome are resulting in a hardening of budget of the EBRP will be further movement from constraints and an improvement in resot -ce soft to hard budgets, in both the enterprise utilization. Furthermore, the privatization of and banking sectors. large segments of the Polish economy is Despite these changes, it is obvious introducing more effective property rights, a that SOEs will continue to be major players pre-condition for medium- to long-term in the Polish economy, at least in the productivity gains. medium term. In that context, there is An early attemnpt at restructuring the encouraging evidencc that Firms are Polish economy was taken with the Law on becoming more cost conscious. A recent Joint Ventures, in December 1988. The Law survey of 75 large state-owned manu- had some limited success, and a number of facturing enterprises (Pinto, Belka and joint ventures were forned in the following Krajewski 1993; Hume and Pinto 1993) two years, but the Govemnment of Poland shows that, even without privatization, SOEs took a major first step toward full-fledged are adjusting, restructuring and increasing privatization in 1990, when it initiated an profitability in response to the ETP. Of Economic Transformation Program (ETP). particular relevance is the finding that, "all The ETP led to the rapid privatization of firms managed to reduce the consumption of many small enterprises, although the pace materials and energy per unit of sales" was slower than initially expected for the (Hume and Pinto 1993, p. 19). Specifically, State Owned Enterprises (SOEs) (Kharas the 31 best companies reduced energy and 1991). Problems have been caused by the material consumption by 22 percent between recession, which followed the introduction of 1991-92, and the remaining ones by the ETP, and by the simultaneous collapse of approximately 17 percent (Hume and Pinto trading arrangements linked to the Council of 1993). Mutual Economic Assistance (CMEA). While these profound changes in Furthermore, the process of privatization economic policy are already in train, and proved to be much more complex and beneficial results are discemible, the timing lengthy than initially expected. In light of and speed of further change are matters of these and other factors, the Government has considerable uncertainty in Poland, as in the developed an Enterprise and Bank other economies in transition. The effects of Restructuring Program (EBRP). The EBRP decades of central planning will take years to will simultaneously address the inter-related attenuate. Nevertheless, the progress to date problems of the SOEs, which are not gives grounds for optimism that, in the servicing their debts, and of the conunercial period 1995-2000, restructuring and energy banks. The EBRP contains additional pricing reform will have proceeded far enough to justify the assumptions underlying the scenarios and simulation results discussed 39 A discussion of the effect of soft budgets in a in Chaptcrs 4 and 5. From the standpoint of central planning regime can be found in Komai actually dcsigning environmental policics, (1992). 39 the reforms make it realistic to seriously Finally, an evaluation of environmental consider incenltive-based measures. policies in an open economy like Poland's must consider the consistency between CO)NSIDERATI()NS FOR POLICY overall nationial interests and the policies of EVA,lATIOl()N other countries. Different environmental policies will have different effects on relative Economic elficiency is the criterion most prodiuct prices and the terms of trade. These commllonly invoked in comparisons of CAC effects will depend in part on ihe policies witlh incentive-based policies. There are, pursued by other countries. With these points however, other important criteria as well. in mind, wc now consider CAC. pollution The following points are grouped into two charges, and tradeable permit policies in turn. broad categories: those related tn legal and teclnical features, and tlhose relate. to equity Command and Control and other political economy concems (see also Tripp and Dudek 1989). There is a legal basis for the current system of Regarding legal and technical features, air emission and ambient standards for prospects for success with environmental stationary sources, though there is some policies are enhanced if there is a cdear legal doubt about the consistency of the two authority for taking action; the goals of policy standards. Further legislation would be are clearly stated; the effects of policy are needed to pursue measures such as fuel consistent with those goals; there is adequate restrictions on households (e.g., an urban coal capacity for monitoring effects of the policy use ban) or new controls in the transport measures; and the regulatory authorities sector (e.g., catalytic converters). As possess both the technical capacity to described above, monitoring capacity is in intexpret the evidence and the capacity to place and improving. The technical capacity enforce compliance. Regarding the political of the regulatory authorities also is growing. economy issues, environmental policy is more The legal system is theoretically capable of likely to succeed if the overall opportunity enforcing standards, but such enforcement in cost to society is kept as small as possible practice is often weak, partly because of (given the environmental objectives); and the unwillingness to deal with the social pc -y does nor create a large constituency of consequences of shutting down emnission oi sition because of adverse distributional sources and causing significant economic cf -s or other factors that make private dislocations. con iance costs high. A successful outcome Poland also is likc many countries in is albo more likely if there is greater flexibility that its environmental goals, and their for individual actors in the means of relationship to current CAC policies, are not compliance, across sources and regions; and entirely clear. The emission standards are procedural obstacles that give rise to high meant to contribute to the achievemnent of **transaction costs" for compliance are limited. ambient targets, but the latter usually take Other irnportant factors, particularly for a priority.40 It is conceivable that under the transitional economy like Poland, are robustness in the face of imperfect, evolving 40 MP mg 4A tager question beyond our scope coDcers th market institutions and incentives for cost- ambient standards themselves. These standards reducing innovations in products and should be set to reflect estimated or suspected processes. damages to human health, investment capital, historic anifacts, and natural systems. However, the absolute 40 emission standards, some sources would have policies mnay have greater effects on product to overcontrol rclative to the abatement prices and terns of trade than CAC, even necessary to meet ambient standards; or though the incentive-based policies are morc sources could face control requirements in cost-effective. ln this situation, unilateral excess of the emission standards to meet adoption of incentive-based policies in Poland arnbient limists. A technical difficulty arises could create competitive disadvantages vis-a- here in defining the level of "background" vis other trading partners that would be ermissions used to calculate a source's impact vexing tor Poland's struggling economy.4' on ambient conditions. It is not clear that these "background" emnissions are properly Emission Fees measured (to exclude the impact of the source itself); but in any case, in a situation The legal basis for emission fees in Poland is where ambient standards are already violated, well established. Fees are levied directly on it is accepted that a source will stiLl be monitored sources and indirectly on motor permitted to operate, provided its excess fuels (Zylicz 1993). This point is important, emissions are less than 20 percent of the since the use of energy taxcs based on relative standard on a 30-minute basis. pollution content may be a uscful practical Although the social cost of CAC is strategy where direct monitoring and higher than more cost-effective outcomes, as charging for emissions is problematic. The noted previously, the effect is somewhat existing system for monitoring compliance muted by the fact the Polish CAC is less rigid with standards also allows fees to be than purely t*echnology-based standards; estimated. Enforcemcnt v-f the fees has been Polish sources have more choice in abatement more problematic, particularly recently, when options, and some intraplant emission trading there were substantial dislocations in the among sources appears to be feasible. The economy and fees rose sharply. Success in current system leaves relatively undisturbed using higher fees as an enforcement tool wil the set of status quo interests in use of the be limited if there are frequent waivers or environment as waste sink (e.g., more subsidies from public budgets. However, this stringent controls are imposed on new problem is not unique to emission fees; weak sources), so there is no automatic environmental enforcement also undercuts constituency among polluters in favor of CAC. change. In fact, as noted above, the prinary The distinction between pollution constituency is in favor of weakening charges as revenue-raisers and fees as standards where short-term adjustment costs enforcement tools needs to be emphasized. are high. The CAC approach also is In Poland, as in many West European compatible with existing policies in the countries, fees contribute to government countries of the EC, notwithstanding revenue that can be used for meeting urgent acknowledgment by those countries that social needs in the cleanup of extant incentive-based policies can be more cost- environmental damage (thereby strengthening effective than CAC methods for air pollution incentives for actually collecting the fees). control. As discussed below, incentive-based This purpose is fundamentally different from the use of emission fees as enforcement tools, and relative magnitudes of these changes, both locally and nationally, remain controversial; and as noted 4tEven for nontraded goods like electricity and previously. Polish standards are very tight compared diztrict beat, indirect trade effeas through prices of to Westem nonns other commodities could be significant. 41 where the charge rates are higher and the goal whilc assuring environmental conditions, is to of policy is to shrink the "revenue base." weaken the cost-effectiveness of the fee Raising revenue for ameliorating past approach.42 environmental neglect also differs Stiff fees also create an automatic substantially from the use of fee receipts for constituency against their use by charging subsidizing new pollution control efforts. polluters for all ernissions, not just emissions While there is understandably a constituency in excess of the standards. In this situation, for the latter application of fee proceeds, such private compliance costs may exceed the efforts risk continuing or aggravating CAC level for many polluters, even if social problemns of economic restructuring (i.e.. the compliance costs (net of tax) are reduced-43 need to move toward hard budgets), and they The problem is magnified when trading can undercut the incentives for cost- parmers use CAC and their product prices minimization that make emission charges thus do not include the charges on residual more cost-effective than CAC. emissions. In this case emission fees can Emission fees not only can promote create domestic opposition based on trade flexibility in compliancc strategies but also disadvantages, since the fees raise the relative can induce innovation in pollution reduction, cost of domestic exports to foreign buyers.44 since even sources in compliance wil have an Finally, it should be noted that, in order incentive to reduce their tax liability. Fees to meet ambient standards as wel as source also can be applied even in noncompetitive emission standards in a cost-effective way, markets, an important consideration in fees have to be differentiated across sources Poland's transitional economy. When finns and regions to reflect the differing can exercise market power, the cost- contributions of sources to ambient effectiveness of a pollution tax must be conditions and the divergence among regions balanced against the impact of a tax in further in excess emissions. Clearly, setting uniforTn restricting output below competitive levels. fees high enough to meet the most stringent However, a more complicated problem arises ambient standard in the most polluted region when finns have weak incentives to minimize would involve significant overcontrol costs (Oates and Strassmann 1984). In the elsewhere. At a minimum, some "zoning" of Polish context, this can arise because of direct fees locaUy and regionally is needed to subsidies, soft finance windows for control contribute to effective attainment of ambient investments, or direct government standards. However, differential fees are intervention in enterprise operation. Again, currently not admissible, as noted above. however, these conditions also will harnper CAC. There are several aspects of fees that arguc against their use as a primary enforcement tool. Their use requires the 42wbile fees cannot assure emissions, they do put an capacity to index against inflation and to raise upper bound on the marginal abatement costs firms fees as economic growth engenders pollution will underlake. This is a useful escape valve against increases. Because fees do not provide the excessive abatement expenditures. assurance of standard inlimitl 43Chapter 5 reports the aggregate financial burden in our two tax scenarios. ETAXI and ETAX2. emissions, it is doubtful in practice whether "See Burtraw (1993) for further discussion fees would ever replace standards. However, Convesely. attempts to maintain emission fees below the effect of a binding pollution standard, levels in some Westem European countries could become a bone of contention with trading partners 42 Emission Trading Depending on the structure of the program, such trades can hold toial emissions constant The legal status of emission trading is the or require a net reduction of emissions to least clear of the strategies we are improve air quality. Intemal or extemal considering. As noted above, air pollution trades also can occur over time, as unused legislation currently under consideration is emission pennits are accumulated and understood to include a clarification of how subsequently used, a process known as individual polluting entities might trade off "banking." controls on their own contiguous sources Difficulties in organizing emission (e.g., controls on multiple sources within the trading arise in connection with the specificity same power or steel plant). The scope for and security of property rights over emission trading among nearby sources might also be allowances, and with "transaction costs," the increased. However, regional-scalc trading, general ease or difficulty of effectua:ing as in the 1990 U.S. Clean Air Act transactions. Property rights issues arise. for Amendments authorizing S02 trading, seems example, when there is ambiguity in the to be a more distant prospect. definition or measurement of baseline Eniission trading also will put some emissions against which "excess" control additional operational and monitoring burdens available for sale is reckoned, and when there on regulators. The extent of thcse burdens is amnbiguity about the status of banked depends on the scale of trading. For a full- emission credits. Insecuity about property blown permit market with many participants, rights limit supply and demand for permits a formal exchange system must evolve and and thus the volume of transactions. High continuous monitoring may be needed to transaction costs that also limit trading can ensure the integrity of trades. For more arise from onerous approval procedures for informal "swaps" of control responsibility trades or requirements that transactions among individual polluters the regulatory reduce net emissions (a form of transactions burden will not be as large, though it is stil tax). necessary to ensure that violation of the In addition to reducing overall trading, standard by one polluter is matched by such obstacles bias emission transactions in ovcrcontrol on the part of another pollutcr. favor of intemal trading, where difficulties in To better understand the prospects for identifying transaction partners, securing success with emission trading in Poland, it is property rights, and completing trades are important to keep in mind that the textbook smaller. As a consequence, some cost- model of an organized, competitive permit effective extemal trades are nmissed.46 This market is one end of a continuum. At the other end is intraplant trading across sources Qffs can involve intemal and exteral amount. Oiescnivleitmladecea by a firn (or other entity), and in between are trades. Bubbles are aggregations of pollution from all infonnal bilateral trades of various types.45 sources in a particular plant or area that require control only on the aggregates. rather than on 45Emission trading under the U.S. Clean Air Act has individual source emissions. Bubbles also can included "netting," "offsets," and 'bubbles." Netting involve intemal or extemal trades: of the three allows a firm creating a new discharge source to options listed here, they come closest to the textbook avoid stringent new source performance standards if description of emission trading. For further it reduces emissions from other plant sources. discussion see Hahn and Hester (1Q89) and Netting is always an internal trade. Offsets allow Tietenberg (1985. 1990). new sources to enter "non-attainment areas" if they 46rbe survey by Hahn and Hester (1989) documents reduce existing source emissions by an even greater the predominance of intemal trades in the U.S. Clean 43 point has been emphasized by Atkinson and 1984). However, research to date suggests Tietenberg (1991), who argue that bubble that even fairly "thin" permit markets could transactions in the US emissions trading be relatively efficient, and that significant program have been bilateral and sequential, market power over permits is relatively rather than the multilateral simultaneous unlikely if permits initially are widely exchanges needed to arrive at an efficient distributed based on historical emissions outcome. However, a significant part of this (Tietenberg 1990). trading inefficiency may stem from regulatory A bigger issue is regulation of policies that hinder the actions of self- production and pricing decisions by private or interested parties in developing better state-owned finfns that leads such firms away informational and trading institutions with from socially cost-mininiizing decisions. lower transaction costs. Regulatory distortions will occur, for Concems over both property rights and example, if firms face favorable regulatory or transaction costs stem in part from conflicts tax treatrnent of compliance expenditures between environmental objectives (improved relative to expenditures on emission air quality) and holding down compliance allowances. This is an especially serious cost burdens. T'his tension is manifested in, problem in regulating emissions from the for example, wrangling over emission electric power sector, where goverrnent baselines and regulatory checks on trades. intervention is likely to remain ubiquitous Such conflict is particularly important to keep cven after economic restructuring.48 in mind in considering the Polish situation. Another important institutional where there are strong desires both to hold difficulty in establishing emission trading (and down compliance costs and to improve air charge systems) relates to the nature of quality.47 different pollutants in the atmosphere. The Emission trading also may fail to ultimate goal of pollutant control is improving achieve its theoretical potential because of ambient air quality. When pollutants miL distortions caused by market power over relatively unifonnly over medium to large emission allowances or by regulation of firms' areas, the location of individual emissions is product output and pricing decisions (as in not that important in determining overall air the public utility sector). A firm with market quality. Emission permits supplied and power over permits could distort the demanded by dispersed polluters can be allocation of permits among other existing treated as interchangeable. Policy in these polluters and encumber entry of efficient cases can focus directly on emission control. competitors in its product markets (Hahn In contrast, when pollutants do not mix unifonrly, the location of emissions does Air Act program and provides several explanations matter in detemining the effect of emission based on propeny rights and transactioncost reductions on air quality. In this case, concemns. Tietenberg (1990) cites evidence that multiple permit systems generally are needed potential cost savings under emissions trading may be to filily reflect the impacts of parficular significant even wit intemal trading alone, but there emissions at different receptor points. Such nonetheless appeals to be a substantial loss of cost- systems are fairly difficult to structr and effectiveness from constraints on extemal trading, costy to imlement. Compromises, such as 47See also Oates, Pornty and McOartland (1989) stly sources into Comprones, often who argue that some nominal overcool under CAC stratiyig sourcs into trading zones, ofen may not be a social burden because aggregate pollution reduction below prevailing standards is "Sce Bobi and Burtiaw (1992) for discussion of this justified in light of the danages caused- issue in the U.S. contexL 44 can significantly linit cost-effectiveness by the important point is that they can exploit restricting trading options (see Tietenberg some gains. Many of the obstacles to c.ading 1985, Clhapter 4 and Tietenberg 1990 for in past US experience appear to have beea further discussion). In practice, trading with regulatorv rather than intrinsic and thus could non-unifornly mnixed pollutants generally be avoided by regulators in Poland.50 involves mostly contiguous sources in order The argument for at least limited trading to lessen these challenges. An irnportant is supported by experience from the attempt exception to this pont arises when the target to institute trading on an experimental basis in volume of aggregate emission reductions is Chorzow, a city in Upper Silesia (Dudek, very large, as in the US S02 program. In this Kulczynski, and Zylicz 1992). In Chorzow, case, it is more likely that ambient conditions significant pollution is caused by both a steel generally will improve even if emissions are mill and a power plant. The stecl plant reallocated through trading, and the locations already has undertaken some modemization of specific polluters may be less important. that reduces emissions (closing coke ovens Like ermissions taxes, emissions trading and piping in natural gas). Further wil stimulate innovation in pollution contremplated changes (closing of open hearth reduction. However, this stimulus will be furnaces) would yield additional pollution attenuated if trading options are lirmited. The reduction. The power plant is far out of domestic constituency against emission compliance and is considering ambitious but trading is likely to be weaker than with taxes expensive modernization (installation of because the initial endowment of emission fluidized bed boilers) to meet stricter rights could be adjusted to address standards likely to be faced in 1997. distributional concerns Nevertheless, Accelerating the power plant moderniization emission trading could have intemational would be cosdy, while reducing its output or trade effects if a sector that is imponant in closing the plant would have serious social trade- for example, a growing sector with consequences. few grandfathered emission rights-also must Under these circumstances, significant incur significant costs for permit acquisition.49 cost savmgs and more rapid environmental Balanced against this effect, on the other improvement were estinated to be attainable hand, is the very real possibility of cost if the power generator financed further savings because of the flexibility embodied in abatement at the steclworks. However, emissions trading. proceeding with this plan requires regulators It is important to note again that many to approve continued operation of the power of the difficulties with emission trading- plant with emission levels in excess of its monitoring, verifying trades, guarding against standard, in a region that already faces market power, reconciling emission controls serious pollution loadings. Apparently this with ambient air conditions-are mnst serious transaction so far has been blocked by lack of for full-blown permit trading on a wide scale regulatory approval. and may be less serious in more limited transactions. Such transactions are less 5sEoissions trading may actually improve effective in exploiting all gains from trade, but envirornental compliance in that some highly polluting finns that might otherwise either close down or emit with impunity, given uneveo 491Te same may be true even for sectois w.t- large enforcement of CAC, can finance abatement pennit endowments if the opportunity cost of investments under emissions trading with sales of foregone permit sales is rolled into product prices. emission reduction cedits that they generate. 45 The discussion m di chapter has instruments as mutually exclusive. All three focused on comparing CAC, emnission fees, approaches to pollution control have a role to and emission trading. However, it is play. In the next chapter we oudine an important not to see the choice of policy integrated strategy. Chapter 7 CONCLUDING REMARKS The following seem to be among the key trivial. They arise in pan from relaxing new- points implied by the simulation analysis: source bias in command approaches. Previous research suggests also that (1) Significant decreases in air pollution significant savings are possible just from enmissions are likely from economic intraplant emission trading, for which the restructuring and energy pricing refonns, model already allows, even in conmmand-and- even without major new environmental control cases. initiatives. Nevertheless, tighter (enforced) emission standards such as those envisaged (4) Further savings may be possible by under current Polish policy may generate extending incentive-based policies to more considerable additional decreases in pollution, decentralized emissions sources in lieu of particularly PM and S02. costly commnand and control (as in the transport sector in our study). However, (2) The Polish legal standards differ in these savings in the scenarios derive partly important respects from typical West from eliminating relatively costly pollution European standards; in particular, Polish PM control measures (like catalytic converters in standards are looser, while Polish NOx our cases) that might not be imposed anyway standards are tighter. However, the costs of in a more cost-conscious command-and- meeting the two altemative sets of standards control policy regime. The comparison of a in Poland do not appear to be widely very costly command-and-control scheme to different. Thus, cost would not appear to be an ideal set of incentive-based policies a major obstacle in switching to standards probably overstates the benefits of broadening more dosely attuned to the European the scope for the latter policies. Community; the larger questions are the capability of meeting any standards as tough (5) As the experiment with a coal tax as the Polish or EC standards and the suggests, pollution control policies and willingness to alter emphases on pollutants. incentives need to be broad-based to make While the Polish and typical EC effective headway in reducing emissions at standards involve roughly comparable costs, reasonable costs. Focusing on orly a few the costs of the strict German standards are sources or fuels is likely to have disappointing significantly higher. The size of this extra and expensive results. cost underscores a need for corresponding benefit assessment to detenmine the value to (6) Both command and incentive-based Poland of such standards. policies will require considerable expenditures to achieve the specified standards. However, (3) There are clear cost savings from using the impacts on investment and energy prices incentive-based policy instruments, even of environmental policies is likely to be though the exact size of the savings cannot be dwarfed by the forces of economic precisely identified. Even with limited restructuring and energy price reform that application of taxes or pennit trading to large have already been unleashed and that are a e?ationazy sources, cost savings are not 47 prerequisite for ultimate success in However, for the reasons noted previously it environmental policy. is unlikely that fees will rise to the level necessary to meet current emissions POLICY CONSIDERATIONS standards. To accomplish this goal cost- effectively, an evolutionary movement toward The argument in Chapter 6, coupled with the emissions trading among large stationary simulation results, suggests a dynamic and sources is needed.5' Trading could start with mixed strategy for the choice of economic informal bilateral transactions, as in the US, instruments. If necessary, CAC alone might and become more extensive as circumstances be relied upon, particularly for household and and the interests of polluters warrant.52 To transport sources, where emission trading is accomplish his requires in turn a clearer legal costly to arrange and taxes are technically or and regulatory authority for emnission trading, politically difficult to apply. For larger as well as continued progress in econonic sources, however, and even for smaller restructring so that firms have incentives to sources where feasible, increased reliance seek out lower-cost abatement options. would be placed on economic instuments. We emphasize again the inportance of These instruments would complement CAC continued progress in economic and energy standards rather than substituting for them. price reform as both a complement to and a In particular, the standards would provide a prerequisite for success in environmental basis for computing emission fees and for policy. In particular, it is likely to be the case structuring emission trades. that greater reliance on incentive-based A starting point for the approach we policies will require fiuther maturation of the envisage is raising emission fees above economic system in Poland-continued current levels. This would encourage some reduction of direct and indirect subsidies, and additional abatement and technical in direct enterprise decisionmaking by innovation, as well as providing added govemment. Without such progress, revenues for cleanup of existing effective enforcernent of conunand-based environmental damages (or other purposes). approaches also will face serious roadblocks. 5tNote that when fees and pennits a uwed in tandem, an increase in the former lowers the value of the latter aud causes permit prices to fail an equal amount. 52Thew may also be scope for trading between point and nonpoint sources, as is carrensly being attanpted expeimentally in the US (e.g., th redrement of old vehides in California). REFERENCES Atkinson, S. and T. Tietenberg. 1991. Burtraw, D. 1993. "The Promise and "Market Failure in Incentive-Based Prospect for S02 Emission Trading in Regulation; The Case of Emissions Europe." RFF Discussion Paper QE93- Trading," Journal of Environmental 22 (Washington, DC, Resources for the Economics and Management vol. 21, Future). no. 1 (July), pp. 17-3 1. Central Statistical Office. 1991. Ochrona Bates, R. W. and E. A. Moore. 1992. Srodowiska 1991 (Warsaw, Glowny "Conmnercial Energy Efficiency and the Urzad Statystyczny). Environment." Background Paper for World Development Report 1992, Central Statistical Office. 1992. Ochrona Policy Research Working Papers WPS Srodowiska 1992 (Warsaw, Glowny 972 (Washington, DC, The World Urzad Statystyczny). Bank, September). Cochran, L. S. and R. A. Pielki. 1992. Bohi, D. R. 1981. Analyzing Demand "Selected Intemational Receptor-Based Behavior: A Study of Energy Air Quality Standards," Journal of Air Elasnicities (Washington, DC, and Waste Management Association Resources for the Future) vol. 42, no. 12 (December), pp. 1567- 1572. Bohi, D. R. and D. Burtraw. 1992. "Utility Investrnent Behavior and the Enission Cofala, J. 1985. "A Model of the Polish Trading Market," Resources and Energy System," in A. S. Kydes and D. Energy vol. 14, nos. 1/2 (April), pp. M. Geragy, eds., Energy Markers in the 129-156. Longer-Tenn: Planning under Uncertainty (Amsterdam, Elsevier Bojarski, W., et al. 1992. "Poland - Energy Science Publishers, North-Holland). Policy and Energy Program up to the Year 2010" (Warsaw, Ministry of Cofala, J., H. W. Balandynowicz, and Z. Industry and Trade, ZPE IPPT-PAN). Parczewski. 1990. "Soenarios of Energy and Environment Future for Bolek, K. and J. Wertz. 1992. Poland," in Proceedings: Seminar on "Environmental Protection in Cracow Energy in East and West: The Polish Region" (Cracow). Case (Paris, OECD), pp. 403426. Borenstein, S. 1988. "On the Efficiency of Coopers & Lybrand Deloitte. 1991. Cornpetitive Markets for Operating Environmental Assessment of the Gas Licenses," The Quarterly Journal of Development Plan for Poland (World Economics vol. 103, no. 2 (May), pp. BankJUNDPfBilateral Aid Energy 357-385. Sector Management Assistance Program). 49 Dowlatabadi, H. and M. A. Toman. 1991. Krupnick, A., K. Harrison, E. Nickel and M. Technology Options for Electricity Toman. 1993. "The Benefits of Generation (Washington. DC, Ambient Air Quality Improvements in Resources for the Future). Central and Eastem Europe: A Preliminary Assessment." RFF Dudek, D., Kulczynski, Z., and T. Zylicz. Discussion Paper ENR93-19 1992. "Implementing Tradable Rights (Washington, DC, Resources for the in Peand: A Case Study of Chorzow." Future). Paper presented at the European Association of Environmental and Ministry of Environmental Protection, Resource Economists Third Annual Nantral Resources and Forestry. 1991. Conference, Cracow, June 16-19. The State of the Environment in Poland: Damage and Remedy Hahn, R. W. 1984. "Market Power and (Warsaw, Ministy of Environrnental Transferable Property Rights," The Protection, Natural Resources and Quarterly Journal of Economics vol. Forestry). 99. no. 4 (November), pp. 753-765. Nowicki, M. 1992. Environment in Poland: Hahn, R. W. and G. L. Hester. 1989. Issues and Solutions (Warsaw, Ministry "Where Did All the Markets Go? An of Environmental Protection, Natural Analysis of EPA's Emissions Trading Resources and Forestry). Program," Yale Journal on Regulation vol. 6. no. 1, pp. 109-153. Nowicki, M. 1993. Environment in Poland: Issues and Soludons (Dordrecht, Hume, I. M., and B. Pinto. 1993. "Prejudice Kluwer). and Fact in Poland's Industrial Transformnation," Finance and Oates, W. E. and D. L. Strassmann- 1984. Developmenr vol. 30, no. 2 (June), pp. "Effluent Fees and Market Structure," 18-20. Journal of Public Economics vol. 24, no. 1, pp. 29-46. Intemational Energy Agency. 1990. "Energy PoLicies, Poland, a Survey," (Paris, Oates, W., P. Portney, and A. McGartland. OECD/IEA). 1989. "The Net Benefits of Incentive- Based Regulation: A Case Study of Kharas, H. J. 1991. "Restructuring Socialist Environmental Standard-Setting,' Industry: Poland's Experience in 1990." American Economic Review vol. 79, World Bank Discussion Paper No. 142 no.5 (December), pp. 1233-1242. (Washington, DC, The World Bank). Pearce, W. D. and R. K. Tumer. 1990. Komai, J. 1992. The Socialist System: The Economics of Natural Resources and Political Economy of Communism the Environment (Baltimore, Johns (Princeton, NJ, Princeton University Hopkins University Press). Press). 50 Pezzey, J. 1992. "The Symmetry Between Successful Transferable Rights Controlling Pollution by Price and Progrars," Yale Journal on Regulation Controlling ilt by Quantity, Canadian vol. 6, no. 2 (Sumuner), pp. 369-392. Journal of Economics vol. 25, no. 4 (November), pp. 983-991. United Nations Environment Program. 1992. Saving Our Planer: Challenges and Pinto, B., M. Belka and S. Krajewski. 1993. Hopes (Nairobi, UNEP). "Transforniing State Enterprises in Poland: Microeconomic Evidence on Walls, M. A. 1993. "Motor Vehicles and Adjustment." World Bank, PoLicy Pollution in Central and Eastern Research Working Paper Series No. Europc." RFF Discussion Paper 1101 (Washington, DC, The World ENR93-22 (Washington, DC, Bank). Resources for the Future). Rollo J. M. C. and J. Stemn. 1992. "Growth Wasidiewicz, U. 1991. "Changes in Polish and Trade Prospects for Central and Energy Policy to Decrease Eastem Europe." Working Paper No. Environmental Pollution," in K. Gorka, I (London, National Econoniic ed., Enviromnental and Economic Research Associates (NERA). Aspects of Industrial Development in Poland: Selected Papers (Krakow, Sierpinska, M. 1991. 'The Necessity of Krakow Academy of Econonics). Introducing Raw-Material Saving and Energy-Saving Projects to Polish Weitzman, M. L. 1974. "Prices vs. Industry," in K. Gorka, ed., Quantities," Review of Economic Environmental and Economic Aspects Studies vol. 41, no. 4 (October), pp. of Industrial Development in Poland. 477491. Selected Papers (Krakow, Krakow Academy of Economnics). World Resources Institute. 1992. World Resources 1992-93 (New York, Oxford Tictenberg, T. H. 1985. Emissions Trading: University Press). An Exercise in Reforming Pollution Policy (Washington, DC, Resources for Zylicz, T. 1992. "Envirornental Taxes in the Future). Poland," Draft Manuscript (Warsaw, Ministry of the Environment). Tietenberg, T. H. 1990. "Economic Instruments for Environmental Zylicz, T. 1993. "Case Study on Poland." Regulation," Oxf9rd Review of Paper piesented to the OECD Economic Policy vol. 6, no. 1 (Spring). Workshop on Taxation and Environment in European Economies in Tictenberg, T. H. 1992. Environmental and Transition, Paris, 25-26 February. Natural Resource Economics, 3rd ed. (New York, Harper Collins). Tripp, J. T. B. and D. J. Dudek. 1989. "Institutional Guidelines for Designing Appendix 1: Details on the Modeling Scenario Derinitions 52 Table Al.l. Economic Growth and Structural Change to 2010 (1990 prices) Category 1988 199G 1995 2000 2010 Population (millions) r .9 38.2 38.8 39.5 41.0 Gross National Product (GNP) - Trillions of zloty 685 607 634 798 1181 - Index (1990=100) 113 100 105 132 195 Contribution to GNP (%) - Energy industry 6.1 6.9 7.7 7.2 5.6 - Total industry (incL. energy) 52.2 46.1 44.7 42.8 39.7 - Construction 8.8 8.7 8.7 8.2 7.0 - Agriculture 6.1 7.1 6.8 5.8 4.9 - Transport 3.9 3.9 3.9 3.9 4.2 -Other 29.0 34.2 35.8 39.2 44.2 Table A1l2. Production of Energy-Intensive Products and Services Product/service 1988 1990 1995 2000 2010 Pig iron (103 tonnes) 10260 8400 7850 7800 7650 Steel (103 tonnes) 16870 13450 11200 12550 12400 Copper (103 tonnes) 401 346 375 390 450 Aluminum (103 tonnes) 48 47 48 49 50 Sulphur (103 tonnes) 5000 4696 4900 4900 5070 Ammonia (103 tonnes) 2340 1580 2150 2500 2600 Cement (103 tonm.s) 17000 12500 13600 15020 16000 Freight transp. (109 tnme-lan)n L 144 97 107 125 164 DwelUings (103) - newbt 0 0 650 1400 4600 - retrofittedb/ 0 0 880 2200 6050 - Total 10925 11180 11730 12360 15310 Private cars (106) 4.5 5.3 6.0 6.8 8.8 A only professional land transport included b/since 1990 53 Table A13. International Fuel Pricesa/ (USS/unit) Fuel Unit 1990 1995 2000 2005 2010 Hard Coal - tc- - lump size (export price) 51 53 57 59 62 - coal fines (export price) 47 48 52 55 58 - coking coal (export price) 53 53 56 59 62 - coal fines 52 54 57 60 62 Natural Gas 103m3 -pipeline (traditional) 100 113 125 131 138 - pipeline (new) 110 124 137 144 152 -LNG 144 159 175 185 195 Liquid fuels ton _ - crude oil (heavy) 140 '60 180 199 218 - gasoline 240 274 308 343 377 - diesel 193 221 248 275 302 - heavy fuel oil 96 108 119 129 138 Nuclear fuel tceV 24 25 27 27 28 AIAI import prices are cif for imports except those noted as "export price," which are fob. /bltonne of coal equivalent Table. A1.4. Fuel Prices to Final Energy Consumers in BASE Case (US$Sunit) Fuel Unit June '92 1995 2000 2010 Hard coal - to industry tce 43 52 55 60 - to households tce 76 88 93 98 Natural gas - to industry 103m3 130 132 144 158 - to households 103m3 160 280 300 310 Gasoline tonme 654 1033 1158 1419 Diesel oil tonne 395 514 572 703 Fuel oil light tome - 310 347 425 Electricity - to industry MWh 41 51 52 53 - to households MWh 42 81 81 82 District heat - to industry GJ 3.6 4.7 4.9 5.2 - to households GJ 4.2 8.4 8.7 8.9 54 Table ALS. Sharc of TransmissiontDistribution Costs and Taxes in Fuel Prices to Final Consumers In 1995, the BASE Case Fuel Share Hard coal - to industry 20 - to households 40 Natural gas - to industry 10 - to households 60 Gasoline 75 Diesel oil 60 Fuel oil light 35 Electrcity - to industry 30 - to households 55 District heat - to industry 10 - to households 50 55 Table A1.6. Own Price Elasticities Applied in the Scenariosl/ Fueltsector Industry Transport-/ Residentiall ._________ .Commerdal Solid fuels -0.5 0 -0.4 Fuel oil -1.0 0 -0.4 Motor fuels -1.0 -1.0 -1.0 Gas -0.5 0 -0.4 Electricity -0.5 0 -0.5 Heat -0.5 0 -0.4 /lIn the transport sector, it has been assumed that only demand for motor fuels (gasoline and diesel oil) is price elastic. Little hard coal is consumed in this sector; and while electricity is used by railways, electricity costs are a small proportion of total costs. so it has been assumed that no denand adjustment will occur as a result of higher electricity prices due to stricter environmental regulations. Table A1.7. Polish SO2 Standards (g/GJ fuel input) Source A B C Coal: - fixed grate 990 720 650 - mechanical grate 990 640 200 -PF- dy bottom4/ 1240 870 200 - PF- wet bottom 1240 870 200 Lignite: PF - dry bottom 1540 1070 200 -PF - wet bottom 1540 1070 200 Coke: -fixed grate 410 410 410 - mechanical grate 500 250 250 Fuel Oil: - < 5O MWth 1720 1250 1250 ->50MWth 1720 170 170 2PF- pulverized fuel A = interim standards for existing (pre-1994) sources B = final standards for existing sources C = standards for new (post-1994) sources 56 Table A1.8. Polish NOX Standards (g$GJ fuel input) Source A B C Coal: - fixed grate 35 35 35 - mechanical grate 160 95 95 -PF- dry botom 330 170 170 PF - wet bottom 495 170 170 Lignite: - PF- dry bottom 225 150 150 - PF - wet bottom 225 150 150 Coke, - fixed grate 45 45 45 - mechanical grate 145 145 110 Fuel Oil: - < 50 MWth 120 120 90 - > 50 MWth 160 160 120 Natural Gas: - S< MWth 60 35 35 - > S0 MWth 145 85 85 Puelwood - fixed grate 50 50 50 See notes for Table Al .7. Table AL9. Polish Particulates Standards (g/GJ fuel input) Source A B C Coal: -fixed grate 1850 1370 1370 - mechanical grate 800 600 600 -PF - dry bonom 260 130 130 -PF-wetbottom 170 90 90 Lignite: -PF - dry bottom 195 95 95 -PF - wet bottom 140 70 70 Coke: - fixed grate 720 235 235 -mechanicalgrate 310 235 235 See notes for Table A1.7. 57 Table Al.10. European Community SO2 Standards (g/GJ fuel input) Source existing new > 50 MWth 25%fllreduction until 1993 43%.areduction until 1995 60%dIreduction until 2003 Coal: 100-500 MWth 84D-4xD1 >500 MW,h 140 Fuel Oil: 50-300 MWth S0O 300-500 MWth 1080-6.5xg/ >500 MWth 120 Natural Gas: >50 MWth 11 il/compared to 1980 W/x=35 for each 100 MWth of thermal capacity; e.g., if the thermal capacity is 100 MWth, the corresponding standard is 840-4*35 - 700 g/GJ L/as above but for each 100 MWth x=30; for 100 MWth unit the standard is 1080-6.5*30 = 885 g/GJ Table A1.11. European Community NOx Standards (gIGJ fuel input) Source existing new > 50 MWth 20%Wreduction until 1993 36%B/reduction until 1995 Coal: >50 MWth 228k/ Fuel Oil: >50 MWth 135 Natural Gas: > 50 MW,h 109 a/compared to 1980 blfor solid fuels wiLh less than 10% volatile compounds the value is 455 gIGJ 58 Table A1.12. European Community Particulates Standards (g/GJ fuel input) Source 50 - 500 MWt, 35 > 500 MWth 18 Table A1.13. Gerrnan S02 Standards (g/GJ fuel input) Source Coal ersting plants (tll 1993) 875 1 -o00 MWth, new 700 100-300 MW.h, new 700 and 60% removal >300,, MWt, new 140 and 85% removal 1-300 MWth, FBC 140 or 75% removal Fuel Oil: 1- 5 MWh light fuel oil only 5-100 MW,h 510 100-300 MWth 510 and 60% removal >300 MWth 120 and 85% removal Gas: Natural gas 11 Coking &- Refinery gas 31 59 Table A1.14. German NOx Standards (g/GJ fuel input) Source new existing Coal 1- 2MWtl,FBC/ 175 175 20- 50 MWh, FBC 105 105 1- 50 MWjh 175 50-300 MWth 140 >300 MWth 70 PP - dry bottomGb/ 228 PF - wet bottom 455 Olher Solids: grate firing 350 Fuel Oil: 1- 50 MWa, (fight oil) 75 1- 50 MWth (incl.primary measures) 135 50-300 MWth 90 210 >Z00 MWb 45 210 Natural Gas: < S0 MWh 62 5O-300MWt 62 155 >300 MWtb 31 155 IFBC - fluidized bed combustion b/PF -- pulverized fuel Table AL1S German Particulates Standards (g/GJ fuel input) Source new existing < 5 MWth 53 53 5- 50 MWth 18 > 50 MWth 18 44 lpgite: > 50 MWIh 18 28 Fuel Oil: > 5 MWth 1S IS Natural Gas: <100 MWtb 2 2 Appendix 2: Details on the Model Results 62 Table A2.1. Energy Demand and Energy Intensities for the BASE Scenario Parameter 1988 1990 1995 2000 2010 Gross National Product 100 88 93 117 172 (1988=100) Final energy demand (PJ)9 3582 2729 3165 3352 3873 Fuel shares (%): - Solid fuels 37.8 26.7 30.1 27.4 20.0 - Liquid fuels 14.3 16.1 14.9 15.6 19.2 - Gas 13.6 15.9 15.7 16.2 18.2 - Electricity 10.7 13.0 12.1 13-5 15.8 - Heatbi 23.6 28.3 27.2 27.3 26.8 Primar energy demand (PJ) 5387 4223 4829 5071 5848 Fuel shares (%): - Hard coal 65.9 61.1 61.3 58.4 55.1 - Lignite 11.0 13.4 12.0 10.1 8.7 - Natural gas 7.5 8.9 9.4 11.2 12.9 - Oil 13.7 15.3 15.1 17.3 18.8 - Other 1.9 1.4 2.2 3.0 4.5 Gross Electricity consumption 146 133 149 173 230 CrWh) Energy intensity of GNP 100 89 96 81 63 1j988= 100) I Electricity intensity of GNP 100 102 108 100 92 (1988=100) I . _I _ -4Non-energy use of liquid fuels not included WI1ncludes district hear and steam and hot water geneiated in industrial boilers. Table A2.2. Changes in Sectoral Energy IntensitiesW/ for the BASE Scenario (1988 = 100) Sector 1988 1990 1995 2000 2010 Industryb/ 100 109 108 96 79 Construction 100 79 70 61 50 Agriculture 100 75 94 87 84 Transport 100 87 88 78 61 Other sectors 100 95 83 69 52 Residential 100 70 98 100 107 A/For residential sector, per capita consumption. For other sectors, consumption per unit of value added, at constant prices btExcluding energy industries 63 Table A2.3. Emissions of Atmospheric Pollutants by Sectors, BASE Case (thousand tonnes except as noted) Pollutant/source 1988 1990 2000 2010 Energy conversion 2848 2287 2528 2967 - PPP and CHPi1 2019 1584 1981 2471 - Industrial boilers 670 572 414 385 - Other 159 131 133 111 Final users 979 545 664 608 - Industry 174 133 123 129 - Transport 101 80 104 144 - ResidenfiaVCommercial 704 332 437 335 Total 3827 2832 3192 3575 NO, _ Energy conversion 627 579 596 630 - PPP and CHPA/ 421 386 457 498 - Indusmal boilers 150 131 105 100 - Other 56 62 34 32 Final users 736 608 583 691 - Industry 258 125 134 133 - Transport 520 453 405 515 - ResidencaV/Commercial 57 30 44 43 Total 1363 1187 1179 1321 PM . ,_ _ Energy conversion 1353 1013 991 834 - PPP and CHPa/ 780 574 742 652 - Indusbial boilers 504 369 203 161 - Other 69 70 46 21 Final users 792 455 472 377 - Industry 347 230 196 171 - Transport 37 30 4 4 - Residential/Commercial 408 195 272 202 Total 2145 1468 1463 1211 CO2 (million tonnes) 454 364 432 490 A/PPP -- public power plants; CHP - combined heat and power plants 64 Table A2A. Comparison of Scenario Results - Alternative Levels of Controls BASE CAC _____________ ____________ 1988 1990 2000 2010 2000 2010 Primnary eneroy demand (PJ) _ -_- ___ - Hard coal11l 3549 2594 2966 3223 2834 3095 - Lignite 592 567 511 511 510 513 - Natural gas 406 374 568 755 635 814 - Oil 743 643 876 1100 865 1091 - Nuclear - Other 97 44 150 259 155 265 Total 5387 4223 5071 5848 4999 5778 Gross electricity _ _ _ consumption (TWh) 146 133 173 230 172 229 Table A2.4., continued FRRED EEC GER 2000 2010 2000 2010 20001 2010 Primary enery demand (PJ) - Hard coaibl 2769 2910 2809 3100 2860 3071 - Lignite 504 440 512 510 513 454 - Natural gas 659 847 632 810 642 820 - Oil 870 1113 865 1094 870 1105 - Nuclear - 144 - - - 82 - Other 155 264 154 267 155 264 Total 4957 5718 4972 5781 5040 5810 Gross electricity consumption (TWh) _ 170 226 172 230 177 231 B/non-energy use not included llcoke exports are included in hard coal balance 65 Table A2.5. Emissions of PoHlutants - Alternative Levels of Controls (thousand tonnes except as noted) BASE CAC Polutant/source 1988 19902 2000 2010 2000 2010 SO? - new PPpp/ - - 206 677 30 99 - existing PPP 2019 1584 1448 1329 1043 882 - new CHP DI - - 112 310 11 32 - existing CHP - - 215 155 222 155 - industrial boilers 670 572 414 385 349 286 - other energyconversion 159 131 133 III 116 98 - fmal users 979 545 664 608 546 465 Total 3827 2832 3192 3575 2317 2017 NO, I - new PPP - - 42 132 40 130 - existing PPP 421 386 308 234 193 151 -new CHP- - 22 64 18 61 -existing CHP- - 84 62 51 37 -industrial boilers 150 131 105 100 62 67 - other energy conversion 56 62 34 38 31 32 - fumin users 736 608 583 691 516 428 Total 1363 1187 1178 1321 911 906 - new PPP - - 7 48 3 36 -existing PPP 780 574 498 410 101 103 new CHP - - 2 6 1 5 -existing CHP - - 235 188 26 18 indusal boilers 504 369 203 161 100 85 - other energy conversion 69 70 46 21 49 29 -fnal users 792 455 472 377 440 353 Total 2145 1468 1463 1211 720 629 C29 (million tonnes) 454 364 432 490 423 482 66 Table A2.5., continued FRRED EEC GER PolUutant/source 2000 2010 2000 2010 2000 2010 -new PPP 133 332 24 95 21 41 -existing PPP 1001 698 973 530 149 110 -new CHP 85 170 15 45 8 33 -existing CHP 152 83 190 148 41 31 -induslnal boilers 301 197 403 386 273 262 -other energy conversion 100 72 133 109 155 224 - fial users 546 465 546 465 546 465 Total 2318 2017 2284 1778 1193 1166 CO9 (million tonnes) 417 461 421 482 427 474 A/PPP -- public power plant b/CHP -- combined heat and power plant VIData on PPP for 1988 and 1990 include CHP Plants Table A2.6. Comparison of Scenario Results - Instruments for Large Stationary Sources CAC ETAX1 S02TR 2000 2010 2000 2010 2000 2010 Primary energy demand (PJ) ___ ______ - Hard coal 2834 3095 2798 3071 28!6 3018 - Lignite 510 513 513 513 513 513 - Natural gas 635 814 642 779 642 812 - Oil 865 1091 832 1102 850 1113 - Nuclear - Other 155 265 155 264 155 264 Total 4999 5778 4940 5729 4976 5720 Gross electricity consumption (rWh) 172 229 170 227 170 226 See notes to Table A2.4 67 Table A2.7. Emissions of Pollutants - Instruments for Large Stationary Sources (thousand ton nes except as noted) C_A_C ETAXI S02TR Pollutant/source 2000 2010 2000 2010 2000 2010 SO-, _ _ _ _ _ _ _ _ |. - new ppp 30 99 118 420 112 387 -existing PPP 1043 882 937 386 885 405 -new ClHP I1 32 77 199 64 186 -existing CIHP 222 155 180 119 194 134 - industrial boilers 349 286 361 332 389 328 - oier energy conversion 116 98 94 110 128 112 - final users 546 465 546 465 546 465 Total 2317 2017 2313 2031 2318 2017 NO,, _ _ -new PPP 40 130 37 123 34 121 -existing PPP 193 151 187 146 191 147 -new CUDP 18 61 22 64 22 64 -existing CHP 51 37 49 36 49 36 - industrial boilers 62 67 68 70 65 69 - other energy conversion 31 32 30 37 27 32 -final users 516 428 516 428 516 428 ToLal 911 906 909 904 904 897 PM_ new ppp 3 36 3 26 2 27 -existing PPP 101 103 84 77 81 78 -new CHP 1 5 2 6 2 6 -existing CHIP 26 18 15 11 18 13 -industrial boilers 100 85 196 143 97 69 - other energy conversion 49 29 26 22 54 55 - final users 440 353 440 353 440 353 Total 720 629 766 638 694 601 1 C02(mlliontomes 423 482 418 478 420 476 See notes to Table A2.5 68 Table A2.. Comparison of Scenario Results - Tax on All Fuel Users CAC ETAX2 2000 2010 2000 2010 Primary energy demand (PJ) - Hard coal 2834 3095 2625 2769 - Lignite 510 513 513 513 - Natural gas 635 814 642 768 - Oil 865 1091 894 1116 - Nuclear - - - 152 - Other 155 265 155 264 Total 4999 5778 4829 5582 Gross electricity consumption (TWh) 172 229 167 225 See notes to Table A2A 69 Table A2.9. Emission of Pollutants - Tax on All Fuel Users (thousakid tonnes except as noted) CAC ETAX2 Pollutant/source 2000 2010 2000 2010 so')__ - new PPP 30 99 122 321 - existing PPP 1043 882 873 406 - new CHP 1 1 32 118 173 - existing CHP 222 155 165 106 - industrial boilers 349 286 347 335 - other energy conversion 116 98 79 111 - final users 546 465 609 543 Total 2317 2017 2313 1995 NO, _ -new PPP 40 130 28 17 - existing PPP 193 151 188 146 -new CHP 18 61 16 16 -existing CHP 51 37 49 36 - industrial boilers 62 67 64 64 - other energy conversion 31 32 28 31 - final usel s 516 428 544 08 Total 911 906 917 918 PMl - new ppp | 3 36 3 23 - existing PPP 101 103 80 76 - new CHP 1 5 2 5 - existing CHP 26 18 16 11 - industrial boilers 100 85 191 156 - other energy conversion 49 29 18 21 - final users 440 353 434 342 Total 720 629 744 634 CO (million tomes) 423 482 405 449 See notes to Table A2.5 70 Table A2.10. Comparison of Scenario Results - Coal Tax BASE CAC COALTX _ 2000 2010 2000 2010 2000 2010 Finat energy demand (P1) _ - Solid fuels 920 773 856 727 654 572 - Liquid fuels 524 744 517 723 524 743 - Gas 543 703 572 724 543 703 - Electricity 450 613 441 597 433 590 - Heat 915 1041 905 1026 874 1005 Total 3352 3874 3291 3797 3029 3613 Primary energy denand (PJ) _ __ . - Hard coal 2966 3223 2834 3095 2359 2442 - Lignite 511 511 510 513 519 522 - Natural gas 568 755 635 814 642 838 - Oil 876 1100 865 1091 1043 1193 - Nuclear - - - - - 294 - Other 150 259 155 265 156 261 Total 5071 5848 4999 5778 4717 5549 Gross electricity consumption (TWh) 173 230 172 229 166 222 See notes to Table A2.4 71 Table A2.11. Emission of Pollutants - Coal Tax (thousand tonnes except as noted) BASE CAC COALTX Pollutant/source 2000 2010 2000 2010 2000 2010 o_SO_ _ -new PPP 206 677 30 99 115 281 -existing PPP 1448 1329 1043 882 1444 1338 -newCHP 112 310 11 32 115 317 -existing CHP 215 155 222 155 231 176 - industrial boilers 414 385 349 286 531 452 -other energy conversion 133 lit 116 98 105 113 - final users 664 608 546 465 487 477 Total 3192 3575 2317 2017 3028 3154 NO, __I - new PPP 42 132 40 130 23 54 - existing PPP 308 234 193 151 308 235 -new CHP 22 64 18 61 18 61 -existing CH? 84 62 5 1 37 84 63 -industrial boilers 105 100 62 67 89 91 - other energy conversion 34 38 31 32 39 46 - final users 583 691 516 428 556 669 Total 1178 1321 911 906 1175 1288 PM -new Ppp 7 48 3 36 3 22 -existing PPP 498 410 101 103 498 410 -new CBP 2 6 1 S 2 6 - existing CUP 235 188 26 18 270 255 -industrial boilers 203 161 100 85 151 156 -other energy conversion 46 21 49 29 40 55 -final users 472 377 440 353 336 276 Total 1463 1211 720 629 1300 1180 CO (million tonnes) 432 490 423 482 386 427 See notes to Table A2.5 72 Table A2.12. Costs of Emission Control Measures ror Large Stationary and Transport Sources Measure | Pollutant Unit cost, _ reduced [US$/tonne Car catalytic converters NOX 9700 Low emission NOx 810 diesel engines Diesel oil SO2 4000 desulphurization Selective catalytic NOx 1800-2000 reduction, new power plantat Desulphurization, new S02 510 hard coal power . planti5/ _I flRoughly 35 percent more for combined heat and power plant. b/Assumes 23 GJlt calorific value and 1% sulphur content. Table A2.13. Total Undiscounted Energy Supply and Conversion Investments Period Scenario 199!-2000 1991-2010 BASE 27.1 65.7 CAC 29.2 70.2 GER 38.1 U2.8 ETAX2 27.6 71.8 73 Table A2.14. Production of Hard Coal and Lignite (million tce) ._________ 1988 1990 1995 2000 2010 BASE 141.2W 135.2W/ 133.3 132.9 136.4 CAC 132.7 132.9 136.4 FRRED 128.1 132.7 134.0 EEC 132.7 132.9 136.4 CER 129.9 132.9 134.5 IETAX1 129.3 132.9 136.4 SO2TR 131.6 132.9 136.4 ETAX2 130.9 132.9 134.9 COALTX 133.2 129.8 133.8 3/Actuals I lie Worll anmk I te.ulqua rlt'rs FIuropten f ()ffi. 7T Ikyn Office Wa.shIitig i. I DAP. 't-1Z4.I , 33 ISA 7', 1 I It Va.nr Fr.ai-i I-I I ar N mouii hi I 1 , ism tS 11irNmS4Lt.W I..k4L IlhlL, npL Il I. h1.Iis mr- 202) 4-1 1231 1'.I.'pitt in .1( ) 1 l (01141 :,.I 111111- (2112) 472 I' IM.-IOsi)j )i(1) III 4A1hii 'pi . '3: 1) 2(- i 1I\II 1 * Ii*I 4% [ tiEl|-I 1.1 1. 'RI I 'It \x 1 i'I.x t-il(l l,6 V .uusii ii z. 4 .1) |21 1 -3'i6 %%' . A I.-1 WI * RI I'iQ III.'v 2 1:; I.&ll. Ad.dlrrs' NIl HAl' Cover design by Walton Rosenquist ISBN 0-8213-2753-4