Environmental
Tax Base
Prerequisites for Increased Taxation of
Natural Resources and Chemical Compounds
Prerequisites for Increased Taxation of
Natural Resources and Chemical
Compounds
3DWULN6|GHUKROP
Associate Professor of Economics
Luleå University of Technology
THE SWEDISH ENVIRONMENTAL PROTECTION AGENCY
Internet: www.naturvardsverket.se/bokhandeln
7KH6ZHGLVK(QYLURQPHQWDO3URWHFWLRQ$JHQF\
Telephone: +46 (0)8-698 10 00, fax 08-20 29 25 E-mail: natur@naturvardsverket.se
Address: Naturvårdsverket, SE-106 48 Stockholm, Sweden Internet: www.naturvardsverket.se
ISBN 91-620-5416-3.pdf
ISSN 0282-7298 © Naturvårdsverket 2004
Cover photo: © Helt Enkelt / Johnér
Preface
This is the final report from the project “Prerequisites for increased use of taxes on raw materials and chemicals” ³)|UXWVlWWQLQJDUI|U|NDGDQYlQGQLQJDYVNDWWHUSn
UnYDURURFKNHPLNDOLHU´). The report has been commissioned and financed by the
Swedish Environmental Protection Agency and Joakim Skottheim has been the contact person throughout the project.
Today Sweden has almost 70 market-based instruments in use. According to OECD’s country review, this means that Sweden probably has PRUHHFRQRPLF
LQVWUXPHQWVLQXVHWKDQDQ\RWKHUFRXQWU\. However, environmental externalities
are in most cases not fully reflected in the prices due to the fact that the taxes are too low. This also means that significant changes in behaviors are difficult to achieve.
The economic instruments have so far not been introduced on a large scale to internalize external effects resulting from product use. The EU has discussed the importance of introducing market-based instruments on a larger scale in connection with various processes, e.g. IPP and the Thematic Strategy on the Sustainable Use of Natural Resources. This report can be seen as being an input to the ongoing work.
The opinions expressed in this report do not necessarily reflect those of the Swedish Environmental Protection Agency.
Stockholm in October 2004
Author’s preface
This final report is the result of information from various people, from which this final report has benefited. I would first of all like to acknowledge the support and the valuable inputs gained from Joakim Skottheim at the Swedish Environmental Protection Agency, who energetically has pushed the project forward.
A number of people have contributed with background reports. During the summer of 2003, Anna Christiernsson, Luleå University of Technology, and Gun-nar Lindberg, Swedish Environmental Protection Agency at the time, finalized two comprehensive literature reviews. In addition, during the fall of 2003, five students associated with the Division of Social Sciences at Luleå University of Technology were recruited to write their Bachelor and Master’s theses on the different topics covered in this project. This resulted in five theses (three in economics, one in environmental law, and one in political science), authored by Linda Andersson (economics), Anna Christiernsson (environmental law), Henrik Gåverud (econom-ics), Peter Sjöberg (econom(econom-ics), and Charlotta Söderberg (political science). The above reports and theses have together made this final report possible, and I would like to express my sincere gratitude to all these students and their respective super-visors.
Finally, comments on the work from colleagues at the Division of Social Sci-ences at Luleå University of Technology are gratefully acknowledged. Any re-maining shortcomings are of course solely mine, and should not be attributed to any other individuals.
Luleå in June 2004 Patrik Söderholm
Associate Professor of Economics Luleå University of Technology Sweden
Table of contents
Preface 3
Executive Summary 9
Summary in Swedish/ Sammanfattning 13
1 Introduction 17
1.1 Background and Motivation 17 1.2 Purposes of the Study 18 1.3 Scope and Definitions 18
1.4 Approach 19
1.5 Outline 20
2 Existing Taxes on Natural Resources and Chemical Compounds
in the OECD Countries 23
3 Taxing Natural Resources and Chemical Compounds:
A Theoretical Discussion 29
3.1 Introduction: Basic Motives for Taxing Goods 29 3.2 Taxing Natural Resources for Non-environmental Reasons 29 3.3 Taxing Natural Resources and Chemicals for Environmental Reasons 31 3.3.1 The Economy-Environment Relationship 31 3.3.2 The Economics of Taxing Natural Resources for Environmental
Reasons 32 3.3.3 The Economics of Taxing Chemical Compounds 36 3.4 Final Remarks 38
4 Evaluations of Natural Resource Taxation 39
4.1 Introduction 39 4.2 Taxing Virgin Materials to Reduce Waste and Emissions:
Some Relevant ([$QWH Evaluations 39 4.3 Taxing Aggregates: The Experiences in Sweden, Denmark and the UK 41 4.3.1 The Swedish Tax on Natural Gravel 41 4.3.2 The Danish Tax on Certain Raw Materials 44 4.3.3 The UK Aggregates Tax 45 4.3.4 Summarizing Remarks 45 4.4 Taxing Water Extraction 46 4.4.1 ([$QWH Evaluations of Water Taxes 46 4.4.2 Experiences with Water Taxes and ([3RVW Evaluations 47 4.4.3 Summarizing Remarks 49 4.5 Final Comments on Natural Resource Taxation 50
5 Evaluations of the Taxation of Chemical Compounds 51
5.1 Introduction 51 5.2 Taxes on Fertilizers 51 5.2.1 ([$QWH Evaluations of Taxes on Fertilizers 52 5.2.2 The Experiences of Fertilizer Taxation in Selected European
Countries 53 5.2.3 Summarizing Remarks 58
5.3 Taxes on Pesticides 59 5.3.1 ([$QWH Evaluations of Taxes on Pesticides 59 5.3.2 The Experiences of Pesticide Taxation in the Nordic Countries 61 5.3.3 Summarizing Remarks 67 5.4 Taxing Ozone-depleting Chemicals: The Case of the U.S. Tax 67 5.4.1 The Aim and Design of the U.S. Tax 67 5.4.2 The Effectiveness of the U.S. ODC Tax 68 5.4.3 Environmental Benefits and Costs of ODC Reduction 69 5.5 Taxes on Solvents 70 5.5.1 ([$QWH Evaluations of Solvents Taxation 70 5.5.2 Experiences of Solvents Taxes and ([3RVW Evaluations 72 5.6 Final Comments on Chemical Taxation 74
6 Taxing the Environment in a Small Open Economy:
Overcoming the Obstacles 77
6.1 Introduction 77 6.2 Environmental Policy and Competitiveness: Empirical Evidence 78 6.3 Overcoming the Impacts of Environmental Taxes on Competitiveness 79 6.3.1 Recycling of Revenues 79 6.3.2 Sectoral Exemptions 80 6.3.3 Border Tax Adjustments 81 6.3.4 Pre-announcing and Gradual Phase-in of a Tax 82 6.4 Final Remarks 82
7 European Community Policy and Law 83
7.1 Introduction 83 7.2 EC Policy Developments in the Environmental Tax Field 83 7.3 Legal Prospects for Implementing National Environmental Taxes 86
7.3.1 National Environmental Taxation when Community Measures
are Absent 86
7.3.2 National Environmental Taxation when Community Measures
Exist 88
7.4 EU Law and the Recovery of Production Residues 89 7.4.1 The Definition of Waste in EC Law 90 7.4.2 Relevant EC Waste Legislation 92 7.4.3 Summarizing Comments 94 7.5 Concluding Remarks on EC Law and Environmental Policy 95
8 Additional International and National Prerequisites
for Policy Change 97
8.1 Brief Reflections on the Process of Policy Change 97 8.2 The Role of International Institutions and Organizations 98 8.2.1 The United Nations (UN) 98 8.2.2 The Organisation for Economic Co-operation and Development
(OECD) 99 8.2.3 The World Trade Organization (WTO) 100 8.2.4 Summarizing Remarks 101 8.3 The Swedish Environmental Policy Arena 101
8.3.1 The Role of Environmental Taxes in Swedish
Environmental Policy 101 8.3.2 Do Potential Policy Coalitions Exist? 103
8.4 Summarizing Remarks 105
9 Concluding Remarks and Implications 107
9.1 Are Taxes on Natural Resources and Chemicals Desirable? 107 9.2 Are Taxes on Natural Resources and Chemicals Politically Viable? 109
10 References 113
Executive Summary
There is a growing interest among analysts and policymakers towards extending the environmental tax base to include new taxes on natural resources and chemical compounds. The main purposes of this report are to: (a) evaluate the effectiveness and the efficiency of environmental taxes on selected natural resources and chemi-cal compounds; and (b) analyze the future potential for increased implementation of such taxation policies in Sweden.
In line with theses objectives, the report comprises two major parts. The ILUVW part outlines the current use of natural resource and chemical taxes in the OECD area, and provides a survey of previous evaluations and studies of these types of taxes. This survey discusses the results from both H[DQWH and H[SRVW evaluation studies. Emphasis is put on aggregates and water taxes and on chemical taxes on pesticides, fertilizers, ozone-depleting chemicals and solvents. The VHFRQG part of the report analyzes the potential for increased use and implementation of taxes on natural resources and chemical compounds. Here we recognize that the process of policy change very seldom is straightforward; it is a dynamic process in which policy makers must overcome a number of domestic and international obstacles. For the Swedish case it is particularly relevant to analyze the potential conflict between a unilateral environmental policy on the one hand, and international com-petitiveness on the other. The roles of important international institutions, such as the EU, OECD, UN and the WTO, are also scrutinized.
From an economic efficiency point of view, taxes on natural resources are typi-cally motivated by both fiscal and environmental reasons. The ILVFDO motive can be explained by the fact that the demand and the supply of many natural resources are price-inelastic, thereby creating a stable tax base for the government. It is possible to define four different – but closely related – HQYLURQPHQWDO motives for taxing the output or the use of natural resources. These include: (a) resource depletion; (b) negative environmental externalities at the extraction stage; (c) downstream nega-tive externalities; and (d) the substitution of recycled for virgin materials. Still, in all these cases, taxing outputs or use is generally a second-best alternative, which can be used when, for instance, the monitoring of non-point source emissions and/or efficient property rights regimes are hard to implement.
The taxation of chemical compounds is typically motivated by the desire to tar-get downstream external costs in the form of harmful exposures affecting humans and/or the natural environment. These non-point source emissions are difficult and costly to control, and therefore it may be more efficient to tax the production or the use of chemical compounds upstream. However, the presence of major uncertain-ties about the marginal costs of reducing chemical use combined with sharply ris-ing marginal damages above certain critical levels, imply that command-and-control regulations (and even bans) could be more efficient for command-and-controlling chemi-cal use than would a tax regime. The above suggests that natural resource and chemical taxes provide no universal solution to be applied in all types of situations;
their effectiveness must be assessed and compared to those of other regulatory instruments on a case-by-case basis.
The practical experiences of taxing natural resources such as aggregates and water are mixed. In some cases the taxes are reported to have had significant im-pacts on resource use and on resource substitution behavior; the Swedish tax on natural gravel and the German water extraction taxes are examples of this. Low tax levels combined with low price elasticities of resource demand and sectoral exemp-tions explain why many other taxes have had only limited impacts on resource use (e.g., the Danish water supply tax). An important problem is that many natural resource taxes have multiple objectives; they combine fiscal and environmental goals and they also target a number of different environmental goals at the same time (e.g., UK aggregates tax). This not only weakens the incentive effects of the taxes, but it also makes it hard to evaluate the impacts of the taxes H[SRVW. Both existing aggregates and water taxes are largely motivated for resource depletion reasons although the issue of poorly defined property rights is likely to be a bigger problem in the case of water use than in the extraction of aggregates. There exists thus a strong case for targeting taxes at specific environmental goals (i.e., employ one policy tool for each environmental policy objective), and to make a clear dis-tinction between fiscal and environmental taxes.
Two different groups of chemicals are reviewed in the report; chemicals used in agriculture (fertilizers and pesticides), and hazardous chemicals (ozone-depleting chemicals and chlorinated solvents). A number of H[DQWH analyses of taxes on chemical compounds show that they can be effective means of reducing the environmental impacts from chemical use, even in areas where such taxes cur-rently are uncommon. These studies also illustrate, though, the trade-offs involved in either achieving a cost effective reduction in environmental damages by taxing close to the damage caused on the one hand, or employing a simple tax system with low administrative and monitoring costs by taxing inputs on the other. In practice the latter approach is the one most frequently used.
The taxation of chemical compounds in Europe illustrates the importance of – and the practical problems involved in – designing the tax so as to target the envi-ronmental damages as closely as possible. Most of the taxes implemented are truly second-best instruments as they focus on the consumption or sales of the respective chemicals. Still, the Dutch tax on fertilizers and the Norwegian taxes on pesticides show that there exist ways in which the tax system can be designed to achieve a closer proportionality to damage done. An optimal differentiation of taxes is how-ever unlikely to how-ever be achieved, especially since the damages caused by chemical use and disposal tend to vary a lot depending on the receiving environment condi-tions. In practice the volumes of chemicals used have decreased in the European countries which have implemented such taxes, but the H[SRVW evaluations that have been made indicate that the taxes’ impact on this development has been modest. Just as in the natural resource case this can be explained by relatively low tax rates as well as low own-price elasticities of demand. In fact, often the effectiveness of the chemical taxes has not rested on their incentive impacts, but rather on the uses of the tax revenues. This is particularly evident when the revenues have been
redi-rected to research and information in which the concerned industries have an inter-est.
Often there exists a conflict between international competitiveness and the use of domestic environmental taxes. The magnitude of this conflict is often debated, but overall it is still likely to be severe enough to motivate the use of strategies to overcome any negative competitive impacts. Only for industries for which markets are spatially confined due to bulky transports (e.g., aggregates), such measures may not be needed. We discuss the pros and cons of four different strategies that can be used to address the adverse competitive impacts of environmental taxes: (a) reve-nue recycling; (b) sectoral exemptions; (c) border tax adjustments; and (d) pre-announcing and gradual phase-in. Regardless of the strategy chosen there exists always a trade-off between achieving the highest environmental effectiveness and, as far as possible, avoiding negative competitive impacts. The extent to which these types of traoffs have to be made – and thus the choice of strategy – de-pends largely on the characteristics and the causes of the environmental problem that the tax is designed to address. This illustrates the importance of evaluating the environmental and competitive impacts on a case-by-case basis.
The prospects for implementing new taxes on natural resources and chemicals require not only that such a tax policy represents an efficient environmental policy instrument, but also that it: (a) gains support from international organizations (not the least from those which Sweden are members); (b) is not in serious conflict with international agreements; and (c) will be pushed forward by strong policy coali-tions. For these reasons, the report reviews the policy stances of a number of im-portant international organizations and institutions – not the least the European Community (EC) – towards the use of environmental taxes in general and natural resource and chemical taxes in particular, as well as the existence of potential do-mestic policy coalitions that may propel the use of such taxes in Sweden.
Taxation measures to improve environmental quality are promoted within the EC, but few concrete measures have been taken at the Community level given the requirement on unanimity. The member states’ discretion to adopt national envi-ronmental taxes and charges is, however, quite large. There seems to be few rea-sons why environmental taxes and charges at the national level could not be im-plemented in those areas where there is no harmonization, as long as the tax – its rate and the method of collection – is non-discriminating. In areas of law where there is explicit harmonization, however, it can be more difficult for one nation to have separate legislation. However, regardless of the distribution of competence EC law may restrict the ambition level of the tax – i.e., the tax policy must be pro-portionate – as well as the use of measures to avoid negative competitive impacts on the domestic industry.
The legal preconditions for the recovery of production residues within the European Community, especially the wide scope of the definition of waste, may restrict such recovery operations. By defining a production residue as waste, a number of obligations will follow (e.g., permitting procedures), and some of these impede the use of such substances and encourage the use virgin raw materials that are not subject to similar obligations.
Overall there is substantial support for the use of economic instruments (taxes, fees, etc.) in environmental policy among both international organizations (UN, OECD, and WTO) and among the Swedish political parties. International trade (WTO) rules set limits on the implementation of environmental taxes – especially with respect to “process taxes” – but as long as foreign products are not discrimi-nated (through higher taxes on imports than those on domestic products), these limits should not provide a major hindrance. The analysis also shows, though, that it is hard to identify any strong policy coalitions on the Swedish political arena that may initiate and push for the introduction of natural resource and chemical taxes. Taxation of natural resources is only explicitly promoted by the Green Party and the Center Party, and proposals for taxes on some natural resources (e.g., taxes on raw materials that are traded internationally) are likely to be challenged by rela-tively strong interest groups. The fact that such taxes typically address rather dif-fuse – not to say unimportant – environmental problems also make them hard to promote and thus easier to argue against. Taxes on chemicals should be easier to promote from an environmental and health perspective but they do not appear to be on the agenda of any of the political parties and few of the (national and interna-tional) organizations reviewed.
Summary in Swedish/
Sammanfattning
Många aktörer på den miljöpolitiska arenan visar ett ökat intresse för en utökning av den svenska miljöskattebasen och mer specifikt för införandet av miljöskatter på naturresurser och kemikalier. Syftet med denna rapport är: (a) att utvärdera effekti-viteten i utvalda naturresurs- och kemikalieskatter såväl teoretiskt som empiriskt; samt att (b) analysera potentialen för en utökad framtida användning av sådana skatter i Sverige.
Rapporten är indelad i två huvudsakliga delar. Den I|UVWD delen redogör kort för den nuvarande användningen av naturresurs- och kemikalieskatter i OECD-området, och ger en översikt av tidigare utvärderingar av dessa skatter. Översikten bygger på sekundärdata och på tidigare studier som utvärderat effekten av såväl hypotetiska som faktiskt implementerade skatter. Empiriskt ligger fokus på mine-ral- och vattenskatter samt på kemikalieskatter som lagts på bekämpnings- och gödselmedel i jordbruket samt på ozonförtunnande kemikalier och lösningsmedel. Rapportens DQGUD del analyserar den framtida potentialen för en utökad använd-ning av naturresurs- och kemikalieskatter. Här betonas att den politiska process som krävs för ett sådant införande sällan är okomplicerad; det är snarare en dyna-misk process där en rad internationella och nationella hinder måste hanteras. I det svenska fallet är det speciellt viktigt att analysera den potentiella konflikten mellan en ensidig miljöpolitik å den ena sidan, och internationell konkurrenskraft å den andra. Internationella organisationer och institutioner såsom EU, FN, OECD och WTO och dessas roll i sammanhanget analyseras också.
Utifrån ekonomisk effektivitet kan det vara motiverat att beskatta naturresurser av såväl miljömässiga som fiskala skäl. Det ILVNDOD motivet bygger i hög grad på att utbud och efterfrågan på de flesta naturresurser är prisokänsliga, och detta gör naturresurser till en stabil skattebas för staten. Det finns åtminstone fyra olika – men nära relaterade – PLOM|PlVVLJD skäl till att beskatta användningen eller uttaget av naturresurser. Sådana skatter kan delvis motiveras av förekomsten av: (a) re-sursuttömning; (b) negativa externa effekter vid uttaget av resursen; (c) negativa externa effekter i andra delar av produktionskedjan; samt av (d) viljan att åstad-komma substitution från jungfruliga råvaror till återvunnet material. I alla dessa fall kommer dock beskattningen av användning eller uttag att vara s.k. ”näst-bästa” lösningar, som exempelvis kan användas i de fall då det är svårt att etablera väl fungerande äganderätter för resurser eller för att kontrollera diffusa miljöeffekter där utsläppsskatter inte är tillämpliga.
Beskattning av kemikalieanvändning motiveras i huvudsak av förekomsten av negativa externa effekter längs produktionskedjan. Dessa diffusa och utspridda effekter är svåra att kontrollera direkt, och det kan därför vara motiverat att i stället beskatta användningen eller produktionen av de kemikalier som orsakar proble-men. Osäkerhet om marginalkostnaden för att reducera kemikalieanvändningen kombinerat med en ofta brant stigande marginella skadefunktioner ovanför en
kri-tisk nivå, innebär dock att kvantitativa regleringar (t.o.m. förbud) kan vara ett mer effektivt styrmedel än skatter för att påverka kemikalieanvändningen. Ovanstående innebär sammantaget att varken naturresurs- eller kemikalieskatter erbjuder några generallösningar på miljöproblematiken; deras effektivitet måste bedömas från fall till fall och då också jämföras med effektiviteten hos andra styrmedel.
De praktiska erfarenheterna av beskattning av naturresurser såsom mineraler och vatten varierar från fall till fall. I vissa fall har dessa skatter haft betydande effekter på resursanvändningen samt stimulerat till substitution; den svenska natur-grusskatten samt de tyska vattenuttagsskatterna representerar sådana fall. Låga skattenivåer kombinerat med en låg efterfrågepriselasticitet för resursen samt före-komsten av undantagsregler för vissa användare förklarar varför många andra skat-ter (t.ex. den danska skatten på vatten) endast haft blygsamma effekskat-ter på beteen-det. Syftena med många naturresursskatter är såväl fiskala som miljöstyrande, och de riktar dessutom in sig på ett antal olika miljöeffekter (t.ex. den engelska mine-ralskatten). Detta faktum försvagar dock ofta skatternas incitamentseffekter, och gör det svårt att utvärdera effekterna H[SRVW. Existerande mineral- och vattenskat-ter motiveras ofta utifrån resursuttömningsargument, men problemet med dåligt väldefinierade äganderätter är i de flesta fall ett större problem för vatten än för uttaget av mineraler. Det finns således starka skäl för att inrikta skatterna på upp-fyllandet av specifika miljömål – d.v.s. tillämpa ett policyinstrument för varje mil-jömål – samt göra en tydlig distinktion mellan fiskala och miljöstyrande skatter.
Skatter på två kategorier av kemikalier utvärderas i rapporten: kemikalier som används i jordbruket (gödsel- och bekämpningsmedel) samt ”farliga” kemikalier (ozonförtunnande medel och lösningsmedel). Modellanalyser visar att skatter på kemikalier kan vara effektiva medel för att reducera de negativa miljöeffekterna från kemikalieanvändning, också inom områden där sådana skatter sällan före-kommer. Samma analyser visar dock också att det finns en tydlig målkonflikt mel-lan att å ena sidan ha en beskattning som så nära som möjligt reflekterar miljöpro-blemet, och att tillämpa en enkel skattepolitik med låga kontrollkostnader å den andra sidan. I praktiken används oftast den senare strategin.
Den europeiska beskattningen av kemiska ämnen illustrerar betydelsen av – och de praktiska svårigheter som är involverade i – att utforma skatterna så att de så långt som möjligt fokuserar på miljöskadan. De flesta (för att inte säga alla) existerande kemikalieskatter är ”näst bästa” alternativ eftersom de läggs på kon-sumtionen eller försäljningen av respektive kemikalier. Den holländska beskatt-ningen av gödselmedel samt den norska skatten på bekämpningsmedel visar dock att det finns sätt på vilka kemikalieskatter kan utformas för att uppnå en starkare koppling till miljöskadan. En fullt ut optimal differentiering av skatter på kemikali-er kommkemikali-er dock aldrig att kunna uppnås, inte minst eftkemikali-ersom de miljöskador som följer av kemikalieanvändning tenderar att vara starkt beroende av geografiska naturförhållanden. I de länder där kemikalieskatter använts har försäljningsvoly-men minskat, försäljningsvoly-men de utvärderingar som gjorts tyder på att skatternas inverkan på dessa minskningar varit blygsamma. Precis som i naturresursfallet kan detta förkla-ras av relativt låga skattesatser och en låg egenpriselasticitet för efterfrågan på kemikalier. Ofta har kemikalieskatternas effektivitet inte varit ett resultat av deras
incitamentseffekter, utan snarare av hur intäkterna från skatterna använts. Detta är speciellt tydligt i de fall där intäkterna slussats vidare för att finansiera forsknings-aktiviteter och informationsåtgärder där den berörda industrin har ett intresse.
Det finns ofta en målkonflikt mellan internationell konkurrenskraft och intro-duktionen av inhemska miljöskatter. Konfliktens omfattning är omstridd, men överlag är den oftast tillräckligt stor för att motivera införandet av strategier för att undvika negativa konkurrenseffekter. Endast i industrier där marknaden är geogra-fiskt koncentrerad på grund av dyra transporter (t.ex. naturgrus) kommer sådana åtgärder inte att behövas. Rapporten analyserar fördelar och nackdelar med fyra sådana strategier: (a) skatteväxling och öronmärkning av skatteintäkter; (b) sektori-ella undantag och skattesänkningar; (c) gränsskattejusteringar; samt (d) förannon-sering och gradvis infasning av skatter. Oberoende av vilken strategi som väljs existerar det alltid en ”trade-off” mellan att uppnå högsta möjliga miljöeffektivitet och att, så långt som möjligt, undvika negativa konkurrenseffekter. Betydelsen av detta övervägande - och således även valet av strategi – beror i hög grad på orsa-kerna bakom och egenskaperna hos de miljöproblem som skatten är utformad för att reducera. Sammantaget visar detta hur viktigt det är att utvärdera H[DQWH såväl miljöeffektiviteten som konkurrenseffekterna av en skatt innan den implementeras.
Utsikterna för att införa nya naturresurs- och kemikalieskatter beror inte bara på deras eventuella effektivitet i förhållande till andra styrmedel, utan också huru-vida den föreslagna politiken: (a) får stöd från internationella organisationer (inte minst från de vilka Sverige är medlemmar av); (b) inte kommer i konflikt med internationella överenskommelser; samt (c) kommer att kunna stödjas och före-språkas av starka policykoalitioner. I rapporten granskas inställningen till
miljöskatter i allmänhet – och naturresurs- och kemikalieskatter i synnerhet – hos viktiga internationella organisationer (inte minst den Europeiska Unionen, EU) samt hos svenska riksdagspartier. Vi försöker också analysera den eventuella före-komsten av inhemska policykoalitioner som uttrycker ett gemensamt mål om att införa sådana skatter.
Användandet av miljöskatter premieras inom EU, men på grund av kravet på enhälliga beslut i skattefrågor har mycket få EU-gemensamma miljöskatter införts. Medlemstaternas möjligheter att anta och tillämpa nationella miljöskatter är dock stora. I de fall där det inte finns någon harmonisering finns det få hinder mot natio-nella miljöskatter så länge dessa – nivåerna och indrivningsmetoderna – är icke-diskriminerande. På de områden där det finns explicit harmonisering kan det vara svårare för ett land att anta en separat lagstiftning. Överlag gäller också att EU-lagstiftningen kan sätta gränser för skattepolitikens ambitionsnivå – d.v.s. skat-testyrmedlet måste vara ”proportionellt” – samt även för tillämpandet av åtgärder för att undvika negativa konkurrenseffekter på den inhemska industrin.
De legala förutsättningarna för återvinning av produktionsrester inom den Eu-ropeiska Unionen, främst den vida definitionen ”avfall”, tenderar att begränsa så-dan återvinning. När en produktionsrest definieras som ”avfall” följer en rad be-stämmelser om exempelvis tillståndsgivning, och vissa av dessa bebe-stämmelser verkar snarare för en ökad användning av jungfruliga råvaror.
De internationella organisationer (FN, OECD, WTO) som studerats samt flesta av de svenska politiska partierna uttrycker ett starkt stöd för användandet av miljöskatter. Internationella handelsöverenskommelser (inom WTO) sätter dock gränser på införandet av miljöskatter – speciellt i fallet med s.k. processkatter – men så länge som utländska produkter inte diskrimineras (t.ex. genom högre skat-ter på imporskat-terade varor) kommer dock dessa regler inte att utgöra betydande hin-der. Analysen visar dock också att det är svårt att i Sverige identifiera starka poli-cykoalitioner, som skulle kunna initiera och verka för introduktionen av natur-resurs- och kemikalieskatter. Beskattning av naturresurser främjas endast explicit av Miljöpartiet och Centerpartiet, och propåer om nya skatter på exempelvis mine-raler och/eller skog skulle med stor sannolikhet mötas av mycket hårt motstånd från starka intressegrupper. Eftersom naturresurskatter ofta inriktar sig på ganska diffusa – dock ej oviktiga – miljöproblem, gör också att de är svåra att ”marknads-föra” på ett effektivt sätt och därmed också lättare att argumentera mot. Skatter på kemikalier borde vara lättare att argumentera för politiskt, men sådana skatter tycks inte finnas på någon av de svenska partiernas policyagendor.
1 Introduction
1.1 Background and Motivation
Environmental policy in the 1970s was mainly based on “command-and control” regulations, such as firm-specific emission limits and technology requirements. However, since the late 1980s policy makers have given increased attention to market-based instruments in environmental policy. These include environmental taxes and charges, tradable permit schemes, and deposit refund systems. The grow-ing interest in such economic instruments stems partly from the fact that the tradi-tional systems often failed to provide cost-effective and flexible tools for emissions reductions. Another reason is that environmental policy today puts a greater em-phasis on non-point source emissions (e.g., emissions from road traffic), and these types of pollutants are hard (or at least very costly) to control on a source-by-source basis. Instead taxes on, for instance, the carbon content in fuels used have been implemented. Moreover, falling costs for measuring end-of-pipe discharges have facilitated the use of pollution charges and emissions-based tradable permit systems (e.g., the Swedish charge on nitrogen oxides, the U.S. sulfur trading pro-gram). Many of the market-based instruments in use reveal clear positive environ-mental outcomes, and appear to be cost-effective compared to “command-and-control” regulations (e.g., Ellerman et al., 2000; Stavins, 2000).
Still, in spite of this progress in some selected areas (in particular abatement of air pollution from stationary sources), threats to overall environmental quality and human health are, it is argued, growing worldwide. Industrial production and usage of natural resources to promote economic growth are important contributing fac-tors. Water scarcity is a threat to agriculture in many countries, threats to marine and coastal ecosystems, depletion of forestry and soil degradation cause losses of biodiversity and creates risks for the sustainability of the food chain, toxics accu-mulate in the food chain and the layer is degraded by emissions from ozone-depleting chemicals, to name a few examples (UNEP, 2002).
These negative trends have led many analysts and policy makers to ask whether it would be desirable to extend the scope of the use of market-based instruments in environmental policy. In this way one would achieve better environmental quality, while at the same time benefiting from the effectiveness of market-based policies. Taxes on natural resource use and chemical compounds are often mentioned as potential candidates. For instance, the Swedish Environmental Protection Agency emphasizes that new taxes on raw materials and natural resources should be con-sidered and evaluated (Naturvårdsverket, 2002), while Slunge and Sterner (2001) suggest that a tax would be a more effective means of phasing out the use of the solvent Trichloroethylene in Sweden than the current prohibition.
The most common environmental rationales for taxing natural resources and chemical compounds, respectively, concern the internalization of negative external-ities in the form of emissions and waste disposal. Taxes on the extraction of natural resources and virgin material inputs may be a second-best alternative to emission taxes. Since all materials extracted eventually become emissions to nature, current
rate of extraction equals future amount of emissions or waste. Taxing natural re-sources is thus a mean of preventing transformation of materials into waste and emissions. Natural resource taxation may also be compatible with so-called LQWH
JUDWHGSURGXFWSROLFLHV (IPP), which aim at encouraging the diffusion of
environ-mental management techniques along the entire supply chain of a product (rather than focusing on end-of-pipe solutions) (Lamming and Hampson, 1996).1 Simi-larly, in the case of chemical compounds the argument is that their use gives rise to hazardous emissions and waste products. However, also here it is difficult to regu-late the damage at the point where it arises. Taxes on chemical inputs (or the pro-duction of chemicals) may therefore be a good second-best alternative.
Before extending the scope of the environmental tax base – with an increased focus on natural resources and chemical compounds – one needs, though, to raise and attempt to answer a number of questions. )LUVW, what motivates the use of such taxes and are there alternative regulatory approaches that would be more efficient?
6HFRQG, what is the current use of natural resource and chemical taxes in the
west-ern world, and what can we learn about their efficiency and effectiveness from these experiences? 7KLUG, what are the potential legal, institutional and attitudinal obstacles towards increased implementation of taxes on natural resources and chemicals in Sweden? The last question concerns issues such as the EU legislation on the use of economic instruments and the internal market, the World Trade Or-ganization’s (WTOs) rules on international, attitudes towards the proposed taxes among key international institutions (e.g., OECD) as well as among Swedish po-litical parties and interest groups. In this paper we attempt to shed some light on these questions, and thus investigate whether increased taxation of natural re-sources and chemical compounds in Sweden is a GHVLUDEOH and a SROLWLFDOO\YLDEOH environmental policy.
1.2 Purposes of the Study
The main purposes of this report are to: (a) evaluate the effectiveness and the effi-ciency of taxes on selected natural resources and chemical compounds; and (b) analyze the future potential for increased implementation of natural resource and chemical taxation in Sweden.
1.3 Scope and Definitions
Before proceeding two important limitations of the study need to be stressed. )LUVW, the use of taxes is motivated for several reasons. These include: (a) the raising of revenues to finance public consumption and investment; (b) the redistribution of incomes (e.g., from the poor to the rich and/or over the lifetime of a person); and (c) the desire to affect behavior in any way (i.e., incentive taxes). Many taxes on natural resources (in particular in developing and post-communist countries) can be
1
There are also other (however sometimes less convincing) arguments for why environmental incentive taxes on natural resources and chemical compounds should be implemented. In chapter 3 we provide a more detailed discussion of the potential effectiveness of natural resource and chemical taxation in environmental policy.
related to the first and the second motive, but in this report we focus on the third motive, and more specifically on environmentally related taxes which aim at influ-encing environmentally damaging behavior (e.g., reducing pollution, waste etc.).2 Still, the distinction between environmental taxes and revenue-raising (fiscal) taxes is not always clear-cut. Taxes for which fiscal and environmental objectives are combined or inseparable are therefore also included in the report.
6HFRQG, the above also implies that our geographical focus is limited. We
de-vote most attention to the use of natural resource and chemical taxes in developed countries (primarily in the OECD area). In other countries the use of environmental taxes on natural resources and chemical compounds is more limited. In addition, since this report aims at drawing on the lessons learnt in other countries for imple-mentation in Sweden, we find it relevant to focus on western-world countries (e.g., western Europe, USA) with similar economic systems.3
Finally, a distinction should also be made between WD[HV and IHHV. Taxes are es-sentially compulsory unrequited payments to the state budget, while fees are ear-marked in the sense that the revenues are spent on related (typically environmental) purposes and often recycled back to the sector on which the fees were levied.
1.4 Approach
In line with the purpose of the study, this report comprises two major parts. The
ILUVW part outlines the current use of natural resource and chemical taxes in the
OECD area, and provides a survey of previous evaluations and studies of these types of taxes. There are two important issues that need to be stressed in relation to the evaluation overview.
• We follow OECD (1997) and make an important distinction between H[
DQWH and H[SRVW analyses of environmental taxes. ([DQWH evaluations are
made in advance based on economic modeling and simulations, and are best described as conditioned scenarios (Skou Andersen et al., 2001). The H[DQWH analyses referred to in this report are in most cases based on economic models that capture the essence of an economy or a market, and with whose help one can identify possible outcomes if specific tax policies are implemented, and compare these to some predetermined cri-teria. An important feature of H[DQWH evaluations is that they permit a consistent comparison between different policy designs and instruments.
([SRVW evaluations, on the other hand, are based on historical analyses
of the actual effects of existing or previously implemented tax policies.
2
It is worth noting that a group of experts from the European Commission, Eurostat and the OECD defines an environmental tax as: “a tax whose tax base is a physical unit (or proxy of it) that has a proven specific negative impact on the environment,” (ECOTEC, 2001). However, for the purpose of this report this definition is too wide as it refrains from considering the motive behind the tax. See chapter 3 for a more detailed discussion.
3
See, for instance, Söderholm (2003) who analyzes a number of institutional obstacles to environ-mental taxation in the Former Soviet Union (e.g., little knowledge among polluters about pollution abatement technologies, poorly functioning capital markets, lack of legitimacy and trust for environ-mental legislation among polluters). These obstacles are either non-existing and/or are less constraining in western European countries and in the USA.
The H[SRVW evaluations used in this report tend to rely on secondary data and compare environmental outcomes (e.g., resource consumption, emis-sions, use of recycled materials) before and after the introduction of the tax. The present report relies thus on both types of evaluations.
• Any evaluation of an environmental tax needs to define some evaluation
criterion (or criteria). That is, what should be considered a successful tax policy and what should not? Here it is important to note that the evalua-tion criteria outlined in the economics literature – economic efficiency and cost effectiveness – often differ from those used in most H[SRVW pol-icy evaluations. The latter are often based on criteria such as the fulfilling of politically set goals (e.g., [ percent reduction in resource use), which may be very loosely related to economic efficiency and/or where it has not been investigated whether other policy measures provide a more cost efficient approach to the problem at hand. The H[DQWH evaluations are more often based on economic efficiency criteria. Both of these evalua-tion criteria are used in this report.
The VHFRQG part of the report analyzes the potential for increased use and imple-mentation of taxes on natural resources and chemical compounds. Here we recog-nize that the process of policy change very seldom is straightforward; it is a dy-namic process in which policy makers must overcome a number of domestic and international obstacles. For Sweden as a small open economy, it is particularly relevant to analyze the potential conflict between an ambitious unilateral environ-mental policy on the one hand, and international competitiveness on the other. We also devote special attention to the European Union’s policy statements on envi-ronmental taxation and the role of EU law as an obstacle to increased implementa-tion of natural resource and chemical taxaimplementa-tion. The policy statements of the United Nations (UN), the Organization for Economic Co-operation and Development (OECD), the World Trade Organization (WTO) as well as of the Swedish political parties are also briefly scrutinized. This approach enables us to draw some tentative conclusions about the future potential of extending the environmental tax base in Sweden to include also natural resources and chemical compounds.
1.5 Outline
The ILUVW main part of the study comprises chapters 2-5. In chapter 2 we briefly review selected existing taxes on the use of natural resources and chemical com-pounds in Sweden and in other OECD countries. Here we also discuss the different categories of natural resources and chemical substances that are the focus of the report. Chapter 3 provides a theoretical discussion of the potential efficiency ad-vantages as well as some of the limitations of taxing natural resources and chemi-cals. In chapters 4 and 5 we review and discuss previous H[DQWH and H[SRVW evaluations of natural resource and chemical taxation, respectively. While these reviews provide lessons for how to design such taxes in the most efficient way, they also point towards the most important obstacles to their implementation.
Chapters 6-8 constitute the VHFRQG main part of the study, and focuses on the potential obstacles to increased implementation of natural resource and chemical taxation in Sweden. In chapter 6 we provide a discussion of the possible conflict between domestic environmental taxation and international competitiveness, and how such a conflict (if there) can be resolved. Chapter 7 reviews the European Union’s policies towards environmental taxation in general and natural resource and chemical taxation in particular. The relevant EU legislation is also reviewed and its implications for increased implementation of environmental taxes are ana-lyzed. In chapter 8 we also review the policy statements of the UN, OECD, WTO and Swedish political parties, and discusses to what extent it is possible to identify potential “policy coalitions” that may initiate and propel the use of natural resource and chemical taxes in Sweden. Finally, chapter 9 provides some concluding re-marks and outlines the most important implications of the study.
2 Existing Taxes on Natural
Resources and Chemical
Compounds in the OECD
Countries
A growing number of countries have implemented market-based instruments, such as taxes and charges, in environmental policy in recent years (Speck and Ekins, 2002). However, the role of environmental taxes and charges in the OECD area is still limited. Only about 6-7 percent of the total tax revenues tend to be environ-mentally related and over 90 percent of the environenviron-mentally related taxes are ap-plied within the energy and transport sector (EEA, 2000). In many cases the lion share of these latter taxes – although environmentally related – have a fiscal rather than a purely environmental motive. Within the OECD countries less than 5 per-cent of all “environmental taxes” are taxes on chemical substances, products, waste, emissions and natural resources (Ibid.). The taxation of chemical com-pounds and on natural resources is however becoming more common. For exam-ple, taxes on agricultural inputs (e.g., fertilizers and pesticides) and other chemical substances, such as solvents and PVC, and taxes on raw materials and water are becoming applied in a wider extent (Ibid.).
The Swedish situation provides an illustrative example of this overall picture. Table 2.1 shows the tax revenues (in current prices) from different environmentally related taxes in Sweden over the period 1994-2002. It displays that in 2002 the energy- and transport-related taxes constituted about 98 percent of total (environ-mental) tax revenues, while taxes on chemicals and natural resources contributed a much smaller share.4 In 2002 the taxes on pesticides and fertilizers together gave rise to tax revenues of almost 400 million SEK (0.6 percent of the total),5 and the corresponding figures for the only pure natural resource tax in Sweden, the natural gravel tax, were around 120 million SEK (0.2 percent of the total). The role of natural resource and chemical taxes has increased in Sweden since the beginning of the 1990s, albeit at a low rate. For instance, in 1994 taxes on chemicals (pesticides and fertilizers) constituted a 0.4 percentage share of the total environmental tax revenues, and the tax on natural gravel had not yet been introduced.
4
It should be noted that – as is the case in other OECD countries – many of the Swedish energy-related taxes (such as the tax on gasoline) have strong fiscal motives.
5
Another chemically related tax – which is not included in Table 2.1 – is the charge on environmentally hazardous batteries, which was first introduced in 1987. This is essentially a cost-covering charge paid by importers and manufacturers to finance the collection of used batteries in Sweden.
7DEOH(QYLURQPHQWDO7D[HVLQ6ZHGHQ6HOHFWHG<HDUVEHWZHHQ DQG PLOOLRQ 6(. 1994 1996 1998 2000 2002 Energy taxes: 42418 50060 52634 52762 59579 RXWRIZKLFK )XHOWD[HV (OHFWULFLW\WD[ 2WKHUWD[HVRQHOHFWULFLW\ &DUERQGLR[LGHWD[
Taxes on certain substances: 565 757 543 1567 1434
RXWRIZKLFK 6XOSKXUGLR[LGHWD[ 7D[RQGRPHVWLFDYLDWLRQ 7D[RQSHVWLFLGHV 7D[RQIHUWLOL]HUV :DVWHGHSRVLWWD[ Transport taxes: 5852 6721 6336 7026 7443
Tax on natural gravel: - 70 142 125 117
Total tax revenues: 48835 57068 59655 61480 68573
Source: Statistics Sweden.
As was noted above, taxes on natural resources and chemicals play a limited role also in other developed countries. Still, when looking abroad we also find other types of taxes than those identified in Sweden. Table 2.2a and 2.2b presents an overview of taxes and charges applied to natural resources and chemicals in selected countries. This overview builds on data compiled by the OECD and the European Commission.6 The taxes and fees within the European Union, Norway and Switzerland, often have environmental objectives. Taxes and fees applied within the remaining countries are defined as environmental taxes or environmental charges according to the definition by the expert group from the European Com-mission, OECD and Eurostat; however with no account taken to the purpose of the tax or the charge (see footnote #2 in section 1.3). Most notably, many of the East-ern European countries tax natural resources but almost entirely for fiscal reasons.
Tables 2.2a and 2.2b illustrate that taxation of pesticides and fertilizers exists not only in Sweden but also in other OECD countries, most frequently in the west-ern European countries. If chemical fertilizers, containing nitrogen and phosphorus, and pesticides leak from agricultural activities, different types of water quality problems arise, including include groundwater pollution, eutrophication and con-tamination of coastal and marine waters. These emissions originate from non-point sources making monitoring very costly, and for this reason the XVH of pesticides and fertilizers (rather than the emissions arising from their use) can be taxed. In this report we focus on the experiences of fertilizer and pesticide taxation in se-lected European countries and in the Nordic countries in particular.
6
The databases used are: (a) the database on environmentally related taxes in OECD countries pro-vided by OECD on its web-site http://www.oecd.org/env/policies/taxes/index.htm; and (b) the database on environmental taxes in the European Union, Norway and Switzerland found on the European Com-mission’s web-site http://europa.eu.int/comm/environment/enveco/database.htm.
7DEOHD7D[HV)HHVRQ&KHPLFDOVLQ6HOHFWHG&RXQWULHV &+(0,&$/6
&RXQWU\ 3HVWLFLGHV )HUWLOL]HUV 6ROYHQWV 2'&V
Austria Abolished X Australia X Belgium X X Bosnia-Her. Bulgaria Canada X Croatia Czech Rep. X Denmark X X X X Estonia Finland Abolished X France X F.Yugoslav Germany Iceland Hungary X X Japan Korea Latvia X Lithuania Netherlands X Norway X Abolished X Poland Romania X Slovakia X Slovenia Spain Sweden X X Switzerland X UK USA X X X
Notes: C = Chalk, ODCs = Ozone-depleting chemicals, Co = Coal, S = Sand, Cl = Clay, ST = Stone, G = Groundwater, SU = Surface water, NG = Natural gravel, and T = Tap water supply.
Sources: OECD (2003) and EA (2003).
In addition to taxes on chemicals used in agriculture – and in contrast to the Swed-ish case – taxes on ozone-depleting chemicals and solvents are quite common in the OECD countries, including the eastern European countries. The stratospheric ozone protects life on earth from harmful ultraviolet radiation. Depletion of the ozone-layer can therefore cause severe health and environmental problems, such as skin cancer, global warming and crop destruction. Several products, e.g. refrigera-tors, computers, electronics and solvents, contain ozone-depleting substances, such as halons and CFCs. Damage due to the usage of these chemicals could be pre-vented by implementing taxes that promote substitution to less harmful chemicals. In this report we highlight the experiences from and the effectiveness of the U.S.
tax on ozone-depleting chemicals, not the least since this tax often is described as a policy success.
7DEOHE7D[HV)HHVRQ1DWXUDO5HVRXUFHVLQ6HOHFWHG&RXQWULHV 1$785$/5(6285&(6
&RXQWU\ Minerals Fisheries Forestry Water
Austria Australia Belgium G Bosnia-Her. X X Bulgaria X Canada Croatia X X Czech Rep. X X Denmark S,G,Cl,C T Estonia NG X X Finland X X France Co G + SU F. Yugoslav X Germany G + SU Iceland X Hungary Japan X Korea X Latvia X X Lithuania X X X X Netherlands G + T Norway Poland X X X Romania X Slovakia X X Slovenia X Spain X Sweden NG Switzerland UK NG,S,ST X USA X X
Notes: C = Chalk, ODCs = Ozone-depleting chemicals, Co = Coal, S = Sand, Cl = Clay, ST = Stone, G = Groundwater, SU = Surface water, NG = Natural gravel, and T = Tap water supply.
Sources: OECD (2003) and EA (2003).
Chlorinated hydrocarbons are good solvents for degreasing and dry cleaning, since they dissolve fats. Many chlorinated hydrocarbons are, however, hazardous to human health since they can dissolve the fat protecting the nerve cells. Moreover, they can form extremely toxic dioxins, such as chlorofluorocarbons (CFCs), and deplete the ozone layer, if burned. They are also a major source of organic com-pounds, which affect the ozone layer as well. Also in this case, given the non-point character of the emissions, emission taxes are in most cases unfeasible. A product tax on solvents or on products containing solvents can thus be more efficient.
Taxes on solvents exist in a number of European countries, and in our review we discuss the experiences of solvent taxation and regulation in the Nordic countries (Denmark, Norway and Sweden).
In Eastern Europe taxation of different types of minerals is also wide-spread, but (as was noted above) these taxes are almost entirely fiscal, and very few envi-ronmental motives lie behind their introduction. While taxes on minerals are less common in western Europe, the few that do exist tend to be motivated by environ-mental or resource depletion arguments. Examples include the Swedish tax on natural gravel, the Danish tax on certain raw materials and the UK aggregates tax, which all are reviewed in this report. In these countries the precise motives behind the aggregate taxes are mixed; the different government authorities stress both the internalization of the negative externalities that arise in the extraction phase (e.g., the UK), as well as the encouragement of the use of recycled material (e.g., Swe-den). Taxation of forest and fisheries products is not common in the OECD area, and in the evaluative part of this report we will put little emphasis on these types of taxes as such. What we do, however, is to provide a general discussion of how taxes on virgin materials (including virgin forests) can be used to encourage the use of secondary materials (such a wastepaper or secondary metals).
Finally, taxes on water are wide-spread among the countries surveyed here. Water is a vital good which is used for drinking, cleaning and irrigation and it is also an essential input in consumption and production activities, not the least in the agricultural and food industries. It is often the case that water prices fail to reflect ecological values and in many circumstances water has been considered a free good due to the lack of individual property rights. The market mechanism can thus not always be expected to work properly enough to encourage sustainable man-agement of water resources (see chapter 3 for details). Adequate policies must therefore be implemented, and these may include water taxes or charges on the consumption or abstraction of water resources. In this report we focus solely on water taxes that target the extraction of water, and these include the tax systems in Denmark, France, Germany, and the Netherlands.
Even though the implementation of environmental taxes has increased signifi-cantly during the last decades, few analyses of the effectiveness and the efficiency of these measures have been carried out (Speck and Ekins, 2002). Even a smaller number of studies analyze the impacts of taxes and charges that fall within the scope of this report. Still, in chapters 4 and 5 we identify and review some H[SRVW evaluations of some of the taxes outlined in Table 2.2. Emphasis will be put on aggregate and water taxes (chapter 4) and on the taxes on pesticides, fertilizers, ozone-depleting chemicals and solvents (chapter 5). In some of the cases we also refer to H[DQWH evaluations in which the impacts of hypothetical taxes are modeled, and in which different tax designs can be compared in a consistent manner. Before proceeding with these reviews, however, we need to provide a brief theoretical analysis of why it would be of interest to tax natural resources and chemical com-pounds from an economic efficiency point of view. This discussion follows in chapter 3.
3 Taxing Natural Resources
and Chemical Compounds:
A Theoretical Discussion
3.1 Introduction: Basic Motives for
Taxing Goods
Government taxation of different goods, including natural resources and chemicals, is generally based on either of (or a combination of) three motives. The ILUVW motive is the fiscal one, i.e., the raising of revenues to finance public investment and con-sumption. A VHFRQG motive for taxation is the desire to redistribute income across different groups in society or across generations. These two first motives may be strongly related since fiscal taxes often are used for investing in goods and services that benefit all (not just those who pay for them directly). The WKLUG and final mo-tive for taxation is to induce specific changes in behavior. Environmental taxes such as pollution charges clearly fall into this group as they aim at encouraging pollution abatement and the diffusion of environmentally benign production tech-nologies.
This report focuses on the taxation of natural resources and chemical com-pounds as means towards promoting an environmentally sustainable society. We are therefore primarily interested in how these taxes can be used to change behav-ior in a way that supports this goal. Nevertheless, in section 3.2 we also briefly discuss to what extent taxes on natural resources and chemical compounds can be used for fiscal and redistributive reasons. Section 3.3 then outlines the most impor-tant environmental reasons for taxing natural resources and chemical compounds, while section 3.4 provides some final remarks.
3.2 Taxing Natural Resources for
Non-environmental Reasons
Taxes on chemical compounds are seldom implemented for other than environ-mental (and health) reasons (see, however, below), but this is not the case when it comes to natural resources. Natural resources are often taxed both for fiscal and redistributive reasons. The main aim of fiscal taxes is to raise sufficient revenues without distorting economic activities too much. In practice this implies that goods for which demand behavior is relatively price inelastic should be taxed higher than goods for which demand is more price sensitive. Quite often natural resources fall into the former category of goods. The demand for natural resources and virgin materials tends to be own-price inelastic, especially in the short-run (e.g., Mäler, 1997). For instance, empirical studies typically indicate that the demand for water is relatively own-price inelastic (see also section 4.4). This is partly because: (a)
few substitutes to the resource exist; and (b) the processing of natural resources often is very capital intensive (e.g., metal smelting or pulp and paper production) and substitution to other inputs may therefore be costly and take a considerable amount of time. In addition, the demand for natural resources is normally a GHULYHG demand, i.e., we seldom demand natural resources directly but instead we demand a very large number of consumer goods that are produced with the help of natural resources. Natural resources form the basis of economic activity and are likely to continue to do so in the future. As such – similar to labor inputs – they represent a very stable tax base for the government.
So far we have stressed the fact that the inelastic demand for natural resources provides one important reason for taxing them for fiscal reasons. A similar argu-ment can be made for the VXSSO\ of natural resources. Natural resources are scarce, and for this reason natural resource extraction gives rise to economic rents, that is a difference between the existing market price for a resource and its opportunity cost. Rents (in its purest form) tend to reflect firm-specific advantages in terms of lower costs or more productive inputs. For instance, in the non-renewable resource sector, economic rents may stem from the nature of ore deposits with differential qualities. Tax measures applied to the extractive industry often act to reduce the rent gener-ated by the industry. The most appealing attribute of rent taxation is its neutrality in terms of resource allocation (Heaps and Helliwell, 1985). Since no tax is collected unless a normal profit (or acceptable rate of return) is realized a tax on rent will not induce any VWDWLF efficiency losses since it does not alter production levels.7 In economic terms, the price elasticity of natural resource supply may be very low – or even zero – in the short-run. In the longer-term, however, investment activities (such as exploration) should be negatively affected by the tax on the rent in the sense that marginal projects become uneconomical. Thus, although a rent-based natural resource tax does normally not distort economic behavior given existing capital it will give rise to G\QDPLF efficiency losses.8
Finally, it is important to stress that the implementation of specific taxes on natural resources and environmental goods is often motivated by ERWK fiscal and environmental arguments. However, this type of political rhetoric may be confus-ing and misleadconfus-ing. Taxes implemented for fiscal and/or redistributive reasons should be designed and targeted in a way that secures a stable inflow of tax reve-nues. This attribute is in sharp contrast with the goal of environmental taxes: in the latter case the aim is to influence behavior and this implies that tax revenues should
7
In contrast, a tax on natural resource production (often called royalties) will cause static efficiency losses. In many developed countries mineral resource taxation through royalties is a way of extracting income from multinational mining companies for investment in local development. See, for instance, Otto (1993) for an overview.
8
An efficient fiscal tax in Sweden would be to tax the rent arising from existing hydropower plants. Such plants normally have low variable operating costs while the fixed costs are substantial. This implies that the plant owner has very few incentives to adjust production behavior to avoid the tax. In Sweden this type of tax would also have few negative dynamic impacts since new investment in large-scale hydro-power is constrained by a political decision to preserve the undeveloped rivers in the northern part of the country. Today the taxation of existing hydropower is seriously discussed as one way of replacing the revenues from the carbon dioxide tax if a tradable permit system for carbon dioxide is introduced (see, in particular, SOU 2003:120).
eventually decline as economic agents adjust. In addition, the ultimate purpose of the tax also determines how the tax should be designed. We have already noted that if the purpose of the tax is fiscal, broad and price inelastic tax bases should be tar-geted. In the next section we discuss how environmentally motivated taxes on natu-ral resources and chemicals can be motivated and how they should be designed.
3.3 Taxing Natural Resources and Chemicals
for Environmental Reasons
7KH(FRQRP\(QYLURQPHQW5HODWLRQVKLS
In order to understand the basic arguments for environmental taxation in general and the taxation of natural resources and chemical compounds in particular, it is useful to outline a simple model illustrating the relationship between economic activity and the environment. Such a model is presented in Figure 3.1. This figure first shows that the economic system is a subsystem of the environment; economic activities cannot exist without the services that the environment provides (while, on the other hand, the environment may well exist even in the absence of economic activities). Moreover, the economic system relies on the environment on two basic grounds. First, the environment provides the economic system with natural re-sources, energy and different life-support systems. Second, the economic system makes use of the environment’s (limited) ability to absorb pollution and waste products.
These two functions of the environment are in no way unrelated. According to the first law of thermodynamics all mass entering the economic system eventually become emissions to nature, either to air, land or water (e.g., Ayres and Kneese, 1969). In other words, the more natural resources that are being extracted, the more emissions and waste will emerge (at least for given technologies). In addition, in-creased emissions also hamper the environment’s ability to supply more natural resources. Thus, too much emission into nature can accumulate in the environment and cause damage both to the absorptive capacity of nature and to the future supply of natural resources. With these observations in mind Figure 3.1 also illustrates that recycling of waste and secondary materials can serve two purposes; it can relieve the demand pressure on virgin materials but it is also a way of avoiding waste to accumulate in the environment. The above clearly shows that the environment is a very valuable asset, and a major goal of policy must be to prevent undue deprecia-tion of the value of this asset.
)LJXUH7KH(FRQRPLF6\VWHPDQGWKH(QYLURQPHQW
Source: Based on Tietenberg (2002).
From an economic efficiency point of view the implementation of taxes on natural resources and chemical compounds can only be motivated if: (a) there ex-ists a market failure, thus a situation in which market forces alone cannot provide enough of environmental goods and quality; and (b) other environmental regula-tions (e.g., technology standards, emissions limits etc.) are less efficient than taxes or charges. The following sub-sections discuss these issues for natural resource taxation (section 3.3.2) and for chemical taxation (3.3.3).
7KH(FRQRPLFVRI7D[LQJ1DWXUDO5HVRXUFHVIRU (QYLURQPHQWDO5HDVRQV
Based on the above discussion it is possible to define four different but still closely related environmental motives for taxing natural resources. In this sub-section we briefly discuss each of these and analyze whether a tax is an efficient policy to address any potential market failures. Before proceeding, however, it is important to note that natural resources may be taxed in many ways (e.g., Heaps and Helli-well, 1985); the land (i.e., property tax) or the rent may be taxed. Here we focus solely on taxes levied on the production (extraction) or the use of natural resources.
The ILUVWmotive for natural resource taxation is that of resource depletion. With a tax on resource extraction, the rate of extraction will decline and the resource will not deplete as fast. This issue may be a concern for both non-renewable and renew-able resources. In both cases there is a risk of overexploitation of the resource. Nevertheless, taxing virgin materials for conservation reasons only finds limited support in the economics literature. Under well defined property rights, the owner(s) will have an incentive to use the resource as efficiently as possible (e.g.,
Eggertson, 1990; Radetzki, 2002). Thus, in this case no real market failure exists; the market is capable of signaling scarcity so that when the resource becomes scarcer the price of the resource will go up, which in turn encourages substitution to other resources and leads to lower extraction levels. In cases where property rights are not well defined, though, resource taxation may be a second-best policy to address overexploitation. The introduction of efficient property rights schemes is of course the first-best policy, but in some instances this may be very difficult. For instance, in fisheries taking place under open access conditions it may be difficult to establish property rights due to the mobility of the resource. Instead a landings tax on all fish caught may make the open access fishery behave like an efficient owner (e.g., Heaps and Helliwell, 1985).
Other market failures may also exist, which implies that the resource is ex-ploited at a faster rate than is socially efficient. For instance, in determining the time profile of a non-renewable natural resource, extraction companies employ a discount rate, which should reflect the opportunity cost of capital in the industry. The use of discount rates implies that the weights attached to future costs and bene-fits are lower than that applied to current costs and benebene-fits. For this reason the traditional Hotelling model shows that higher discount rates imply more rapid ex-haustion of a non-renewable resource and less would be left for future generations (Hotelling, 1931). If there are reasons to believe that companies employ a too high discount rate (i.e., they are too “myopic”), resource extraction will be too high and policy interventions are called for. The first-best option would be to reform the capital market, but also here extraction taxes (royalties) may serve as an efficient second-best policy.
In addition, some resources have multiple uses and the owner of the resource may not fully take all these into account in the decision making process. This im-plies, for instance, that in the case of forestry the socially desirable rotation periods may exceed those rotation periods chosen by private owners (who cares only about timber harvest and not about, say, the recreational values of the forest). In such a case, a yield tax payable when the crop is harvested could be used to induce the private owner to choose the socially optimal rotation period (Heaps and Helliwell, 1985).9
In sum, the case for taxing natural resources for conservation reasons is not particularly strong. It is, at best, a second-best policy tool for addressing market failures in the natural resource sector. However, the presence of such market fail-ures is likely to vary a lot between resources, type of owners, geographical location etc. The efficiency of such taxes should therefore be judged on a case-by-case ba-sis. It is also important to note that what we have considered so far are VWDWLFHIIL
FLHQF\ issues, but one needs also to address G\QDPLFHIILFLHQF\ concerns, i.e., the
impact of natural resource taxes on investment and technical change in the respec-tive sectors. As was noted in section 3.2, from a dynamic perspecrespec-tive a tax on natu-ral resources may decrease the (economic and, in some case, even physical) avail-ability of the resource; a tax on minerals production will make exploration
activi-9
It should be noted that this also is a second-best policy. The first-best policy would be to internalize directly the recreational values of the forest rather than to address them indirectly through a yield tax.
