Sarah Keay-Bright. European Environmental Bureau. Brussels, December 2000. EEB document N 2000/021

A critical analysis of the voluntary fuel economy agreement, established between the European automobile manufacturers and the European Commission, with regard for its capacity to protect the environment. Sarah Keay-Bright European Environmental Bureau Brussels, December 2000 EEB document N 2000/021 1

A critical analysis of the voluntary fuel economy agreement, established between the European automobile manufacturers and the European Commission, with regard for its capacity to protect the environment. By Sarah Keay-Bright European Environmental Bureau December 2000 Author: Editor Responsible: Sarah Keay-Bright John Hontelez 34 Boulevard de Waterloo 34 34 Boulevard de Waterloo 34 B1000 Brussels B1000 Brussels Belgium Belgium Tel: +32 2 289 1300 Tel: +32 2 289 1300 Fax: +32 2 289 1099 Fax: +32 2 289 1099 Email: cleanair@eeb.org Email: info@eeb.org Reproduction of all or part of the document is encouraged with acknowledgement of the source 2

Author s Foreword This paper aims to explore the evolution and form of the voluntary fuel economy agreement, established between the European automobile industry and the European Commission, to better understand the factors influencing the decision-making and with regard for its capacity to safeguard the environment. Much of the data has been obtained through personal communication with the actors involved in the negotiating process. Thanks is extended to all those who gave their time to be interviewed for this study. While the general thrust of the paper supports the position of the European Environmental Bureau on the voluntary fuel economy agreements, established between the European Commission and the automobile industry, certain aspects may be solely the views of the author. 3

ACEA CAIR CEC CEMT CLEPA CO CO2 CONCAWE COP DETR DG 3* DG 4* DG 7* DG 11* DG 12* DG 17* DG 21* ECMT EU EUCAR EUROPIA FT FoE GATT IPCC JAMA KAMA MEP MIRA MVEG NAFTA NGO NOx R&D SEM UNFCC WHO WTO List of Acronyms European Automobile Manufacturers Association Cardiff Automotive Industry Research Commission of the European Communities (French Equivalent for ECMT) European Association of Automotive Suppliers Carbon monoxide gas Carbon dioxide gas The oil companies' European organisation for environment, health and safety. Conference of the Parties (to the UNFCC) Department of Environment, Trade and the Regions. Commission of the European Communities Directorate-General for Industry Commission of the European Communities Directorate-General for Competition Commission of the European Communities Directorate-General for Transport Commission of the European Communities Directorate-General for Environment, Nuclear Safety and Civil Protection. Commission of the European Communities Directorate-General for Science, Research & Development. Commission of the European Communities Directorate-General for Energy Commission of the European Communities Directorate-General for Customs and Indirect taxation. European Conference of Ministers of Transport The European Union The European automotive R&D programme established by ACEA European Oil Industry Association Financial Times (newspaper) Friends of the Earth General Agreement on Trade and Tariffs Intergovernmental Panel on Climate Change Japanese Automobile Manufacturers Association Korean Automobile Manufacturers Association Member of European Parliament Motor Industry Research Association Motor Vehicle Emissions Group North American Free Trade Association Non-governmental organisation Nitrogen oxides Research and Development Single European Market United Nations Framework Convention on Climate Change The World Health Organisation The World Trade Organisation *NOTE: As many of the events discussed in this paper took place before the re-naming of each Directorate- General (DG), the previous numerical format (as above) has been used to refer to the Commission s DGs throughout this paper. 4

Executive Summary Objectives and methods This paper aims to explore and analyse the evolution and form of the voluntary fuel economy agreement, established between the European automobile industry and the European Commission. By such analysis, factors influencing the decision-making process were identified and the capacity for the automobile industry s voluntary agreements to safeguard the environment was evaluated. Using the case study of the voluntary agreements established by the Commission with ACEA (European automobile manufacturers), JAMA (Japanese automobile manufacturers) and KAMA (Korean automobile manufacturers), conclusions were drawn as regards the appropriateness and effectiveness of voluntary agreements as a tool for environmental policy making. Relevant documentation was analysed and supplemented by personal communication with many individuals directly involved, or with a relevant interest, in the voluntary agreement negotiations. Background: Evolution of the voluntary agreements From 1991 to 1992, the Commission investigated methods for regulating CO 2 from cars. The two main proposals under consideration involved a purchase tax and an annual circulation tax. The Commission was divided on these proposals and Council rejected them both. The Commission then decided to address CO 2 from cars in 1995 through a three pillar strategy (COM(95)689, CEC 1995) composed of a voluntary agreement, a fiscal framework and a consumer information scheme. The Council called for a 35% improvement in the automobile industry s average fuel economy by 2005 while the Parliament called for a 35% improvement by 2005 and a 50% improvement by 2010. The Commission, however, pursued a 25% improvement from industry and decided the Council s demand for an extra 10% would be achieved through the fiscal framework and consumer information package. In 1997, ACEA offered the Commission an 11% improvement by 2005, which was rejected by all three EU institutions. In March 1998, ACEA and the Commission agreed to a 25% improvement (i.e. 140gCO 2 /km) in the fleet s average fuel economy by 2008 and an intermediate target of 165-170 gco 2 /km by 2003. The terms of the agreement were then agreed. As part of the agreement with ACEA, the Commission initiated similar negotiations in 1998 with the Korean and Japanese manufacturers, KAMA and JAMA respectively. JAMA and KAMA agreed to similar commitments to those of ACEA, although KAMA has until 2004 to achieve the intermediate target, JAMA s 2003 intermediate target range is wider at 165-175 gco 2 /km, and both JAMA and KAMA have an extra year to achieve the final 140gCO 2 /km target. The Commission and the automobile industry will jointly monitor the commitments of the latter. If factors are thought to be preventing progress, one of three options can be pursued: a remedy can be sought; the objectives can be renegotiated; or the situation can be monitored. If there is disagreement between the industry and the Commission as to the factors which have prevented the industry from fulfilling their commitments, either the agreement can continue as before while the situation is monitored and assessed by technical studies, or legislation can be introduced. Factors influencing the evolution of the voluntary agreements Council enthusiastically initiated the debate on targets to be set for the automobile industry, calling for a target of 120gCO 2 /km as early as 1992. Despite a change in test-cycles, which meant that 120gCO 2 /km under the new test-cycle was more ambitious than under the old-test 5

cycle, Council managed to hold on to its proposed target of 120gCO 2 /km. However the threat to industry was later watered down when the Commission designed the three pillar strategy such that the fiscal framework and the consumer information scheme would contribute to achieving this target and thus relieve some of the burden on manufacturers. At the same time, the Member States enthusiasm was not coupled with action. In the early nineties, Member State experts could not agree on the various proposals for regulating CO 2 emissions from passenger cars and accused each other of favouring national industries. Member States also failed to give their political support to the Commission s proposals for a purchase tax or an annual circulation tax. Perhaps the Commission should have foreseen that such fiscal proposals were unlikely to be adopted by Council as the voting procedure requires unanimous agreement from all Member States, but DG 11 s excessive support of fiscal measures was fuelled to a considerable extent by inter-dg competition. Following the signing of the Maarstricht Treaty and the publication of the 5 th environmental action programme, the Commission and the Council were seen to be preoccupied with voluntary agreements as a policy tool for the new approach or third way. Such enthusiasm for the voluntary approach may be an explanation as to why the pros and cons of alternative measures to regulate CO 2 from passenger cars were not more fully investigated at this time, including that of the voluntary agreement. The failure of the Commission to achieve consensus on a policy method to legislate CO 2 from passenger cars in the early nineties, not only set back the intention to tackle the issue by several years, but weakened the Commission s position with respect to the voluntary agreement negotiations as the Commission had no stick with which to threaten industry. The automobile industry knew that the Commission was very keen to ensure that the voluntary agreements would be a success and that there existed no alternative in the way of a binding legislative proposal. These factors, in addition to the lack of technical support for the Commission desk officers, opaque technical negotiations dominated by industry technical experts and the Commission s absence of commitment from senior officials, culminated in a particularly unambitious offer from ACEA in 1997 which was immediately rejected by the EU institutions. In reaction to ACEA s unambitious offer in 1997 of 167gCO 2 /km (an 11% improvement) the Commission adopted a non-technical political approach and began negotiating at higher level, involving the Commissioners, their cabinets and directors. Due to commitment from the top, the Commission was able to achieve a more ambitious target in March 1998 of 140gCO 2 /km by 2008 (a 25% improvement). The voluntary agreement negotiations were also given extra momentum due to the EU commitment, agreed under the UNFCC Kyoto Protocol 1 in December 1997, to reduce EU greenhouse gas emissions by 8% relative to 1990 levels by 2008-2012. The Kyoto Protocol also provided the negotiations with a timeframe. The voluntary agreement negotiations were also considerably influenced by the personalities of the negotiators involved and their inter-relationships with one another. The influence of the individuals involved was also much greater than might normally be expected as so few people were involved in the negotiations. Major concerns As the EU s Kyoto Protocol commitments require an 8% greenhouse gas emission reduction below 1990 levels by 2008-2012, the passenger car sector s shortfall will have to be met by other sectors and/or Member State transport policies and/or some form of emissions trading. Estimates from different sources, suggest that the agreement will stabilise emissions at somewhere between 20% and 30% above 1990 levels for CO 2 from 1 The EU has committed to the UNFCC COP 3 (Kyoto 1997) to reduce greenhouse gas emissions by 8% below 1990 levels by 2008-2012. The Kyoto Protocol has yet to be ratified. 6

passenger cars, depending on the assumptions used. However, such estimations are overly optimistic, as simulations carried out have not taken into account the unrealistic driving simulation of the official test-cycle as well as the extra weight that is added to the model used for the test-cycle, in the way of accessories such as air-conditioning and electric windows, before it is sold to the consumer. If the voluntary agreements would be binding, they would be enforceable. As they are nonbinding they can only be enforced by applying penalties, sanctions or sticks of some sort. However, the Commission does not have adequate tools at its disposal and has not even prepared a binding legislative proposal - which could be an effective stick - to replace the voluntary agreements in case they should break down. The voluntary agreements are clearly a risky instrument to use for addressing CO 2 from passenger cars, particularly as the European Union s commitments under the Kyoto Protocol are binding, once ratified. In addition, the impacts of global warming may not be reversible. The voluntary agreements established between the Commission and the automobile industry raise some legal concerns. First, provisions already exist for bringing into force binding legal measures to regulate CO 2 emissions from cars (e.g. Article 175 of the EC- Treaty (ex-article 130s), Article 5 of Directive 91/441/EC). Rather than use these mechanisms the Commission chose to use non-binding Recommendations. These seem to entail underlying commitments from the Commission. As such they would appear to have potentially binding elements and so it could be argued that they should not have been brought into force by the use of non-binding Recommendations. These underlying commitments are evidenced by the fact that, for example, the Commission will only legislate should the voluntary agreements fail and that it will obtain similar commitments from other non-acea manufacturers. Second, there is a gap in EU law with regards to the negotiation and legal framework for Community level voluntary agreements of this type. The adoption of such an important policy measure, using an untried legal formula could be open to criticism, especially as this strategy effectively limited the involvement of the Council and Parliament. Only Parliament and NGOs raised concerns with the need to establish post-2012 targets when far greater emissions reductions will be necessary in order to minimise the damage of global warming. The auto-industry needs to be given the right signals so that it can begin planning well in advance due to the lengthy lead-times necessary. Another major shortfall of the voluntary agreement is the lack of transparency and democracy as stakeholders were hardly involved. NGOs and Parliament were bypassed, and to a lesser extent Council was also excluded from the negotiations. The co-decision procedure is a far more democratic decision-making procedure than the nature of the voluntary agreement negotiations carried out between the Commission and the automobile industry. As the Commission used a Recommendation of the Commission, it passed up the opportunity for greater involvement of the Council and Parliament as it could have developed a Recommendation of the Council and Parliament. Additional concerns The agreed 2008 target and terms of the agreement are designed to support internal combustion technologies and do little to support alternative powertrains and technologies, or fuels other than petrol or diesel. The objectives of the agreement are vulnerable to renegotiations as there is a high risk that the assumptions of the agreement will not hold true. The Commission and the automobile industry will need to reach agreement during the monitoring procedure as to whether the agreements assumptions are being borne out: 7

- Account must be taken of legislated vehicle-related policies which might neutralise fuel economy improvements, as they may add extra weight to the average car which results in an increase in fuel consumption. - EU-wide availability of higher quality fuel than has been legislated by Auto-Oil 1 is assumed. The agreement s monitoring process will have to take into account the impact on industry s 2003 and 2008 targets, of the variation in fuel quality throughout the EU relative to ACEA s assumptions. However, should the sulphur content of fuel be reduced further from 50ppm, it must be noted that the monitoring procedure may award the automobile industry with CO 2 emissions reductions due to the improved fuel even though the CO 2 emissions have been transferred to the refinery sector which emits more CO 2 in order to produce the lower sulphur fuel. - As the automobile industry is to achieve 90% of its Commitment through technology, 10% can be achieved through non-technological means such as downsizing. However gains attributable to the latter will have to be separated out from gains due to the fiscal framework and the consumer information scheme. - The monitoring procedure will have to consider any factors which hamper the diffusion of technologies into the market. - Impacts on financial performance, competition and employment due to the voluntary agreements will have to be taken into account. - Come 2008, ACEA may also regard KAMA and JAMA s extra year to achieve the 140gCO 2 /km target as an unfair advantage and thus call for an extra year to achieve their target. - Not only will the Commission and the automobile industry have to achieve consensus on many issues, which like the agreement negotiations may be influenced by the personalities of the various individuals, but there is considerable scope for statistical discrepancies between the data collected by the Commission and by industry. The Commission will need considerable capacity and resources to adequately carry out the monitoring process, especially if technical studies are required to verify claims of industry. The area of the Commission responsible for monitoring the agreement s progress currently lacks adequate capacity and resources and this situation is not foreseen to improve. While the Commission will not want to see the targets and timeframes watered down due to concern to maintain its credibility in the face of pressure from Parliament, Council and NGOs, the Commission will be more concerned that the voluntary agreements do not collapse. Collapse of the agreements with the auto-industry could significantly damage the reputation of voluntary agreements as an EU policy instrument. The Commission s concern to prevent the collapse of the agreements could potentially reduce its bargaining power during the monitoring negotiations. If the Commission has to belatedly resort to legislation, this could also be to the detriment of environmental protection. The Commission will find it difficult to know when to deploy legislation, but it is likely to be near to 2008 when it is clear that the target will not be met. Environmental protection could be weakened by the additional time needed to allow for the development and implementation of enforceable legislation i.e. two or more years. As the Commission still has no legislative proposal prepared in case the voluntary agreements should fail, even more time may be needed for the development of such a proposal. The capacity of such a legislative proposal to protect the environment will depend on whether the objectives are further tightened during the legislative procedure and when the legislation is deployed. 8

Addressing CO2 from passenger cars: the voluntary approach Vs binding measures Proponents of the voluntary approach argue that the instrument provides incentives to the business sector for the development of efficient, innovative and environmentally-friendly solutions. It is also argued that voluntary agreements are much quicker to develop than legislative acts which must pass through each of the EU institutions, in accordance with lengthy procedures. However, this paper argues that the legislative procedure (i.e. in most cases for the achievement of environmental objectives, the co-decision procedure is used) is no less time-consuming or resource intensive and is more democratic than a voluntary agreement. Proponents also argue that binding legislation can dictate technology. This paper argues that ACEA in fact dictated the technological development of the automobile industry through the establishment of certain terms for the agreement. It is also the case that binding legislation can be designed so as not to dictate technology. Some Member States enforce non-binding agreements by issuing sanctions or penalties. DG ENV is currently developing a Regulation on environmental agreements which shall provide environmental agreements with a legal framework. At the same time the issue of enforceability of at Community-level voluntary agreements will also need to be resolved, as the Commission does not currently have adequate tools at its disposal to redress breach of agreements. Some argue that Parliament and Council would not reach agreement under the co-decision procedure for legislation addressing CO 2 from cars, but this argument is unfounded. Both Parliament and Council have clearly demonstrated a positive attitude to the issue of regulating CO 2 from cars by supporting more ambitious targets than those agreed to by the Commission and by calling for proposals for binding legislation in case the negotiations should fail. The barrier to agreement between the three institutions is more likely to be the method to be used, as an alternative to the voluntary agreement, to regulate the fuel economy of new passenger cars. With adequate investment of capacity and resources, the various options could be better analysed which should result in a satisfactory solution. The negotiations between the Commission and the automobile industry, as described by this paper, provide many examples of distrust between the negotiating parties. The defensive approach of industry was largely due to its highly competitive nature. A major concern for the automobile industry was that actions to regulate fuel economy might threaten their most profitable market segments which generally consist of cars that are more luxurious, larger, more powerful, faster and therefore more fuel inefficient than the average car. The automobile industry is so competitive that ACEA s members were divided as to what contribution each would make (and thus what the collective target would be). The division within ACEA nearly brought the negotiations to a halt in 1997, and the specific contributions to be required from each manufacturer have still not been agreed. Voluntary agreements should not be applied to industries whose market trends are moving in the opposite direction to that which is desired in order to achieve environmental objectives, 9

especially if they can not be enforced. Once ratified, the Kyoto Protocol will be binding and some of the impacts of global warming may be irreversible. Thus, unenforceable voluntary agreements should also not be applied to problems which require guaranteed environmental objectives. Alternatives to the voluntary approach for addressing CO 2 from passenger cars By considering the nature of the automobile industry and the global warming problem, targets need to guarantee environmental objectives and should be sufficiently stringent to move the industry to the limit of existing technological potential within an appropriate timeframe that allows for planning. In addition to the 140gCO 2 /km target set for 2008, with a view to achieving 120gCO 2 /km by 2012, long-term targets or indicators to achieve the ultimate environmental objectives deemed necessary to stay within safe ecological limits should have also been set, so that manufacturers can begin planning now. Studies show that fiscal measures are effective, to varying degrees, in reducing car use, influencing car purchase choice and in curbing the rebound effect. Thus fiscal measures can be useful complements to vehicle fuel economy regulation, especially if combined with consumer information schemes. It is unlikely that fiscal policies can be used as stand-alone measures to legislate CO 2 from cars, as severe fiscal measures can suffer from political unacceptability, as witnessed recently throughout Europe by the protests against high fuel prices. Unfortunately it seems unlikely that effective Community-level fiscal policies will be adopted in the future if the unanimous voting procedure continues to be used for deciding fiscal policies. While binding limit value legislation would guarantee a specific fleet average fuel economy, it can affect car types differently and will therefore be unpopular with manufacturers and Member States. A tradeable credit scheme applied to manufacturers could be a more flexible and particularly cost-effective alternative for regulating the fleet s fuel economy but it would not necessarily guarantee achieving a specified fleet average fuel economy (depending on the effectiveness of implementation). In most cases, regulating vehicle fuel economy by setting CO 2 emission limits according to a certain parameter can give rise to perverse effects as cars can be designed to avoid the legislation s intention while not yielding the required environmental improvement. But even enforced limit value legislation, aimed at achieving a certain averaged fuel economy for manufacturers sales, can not guarantee a specified CO 2 emissions reduction from the EU passenger car sector. This is due to the trends of increasing mileage per car owner and increasing car ownership as well as an increasing market share of more powerful, faster and larger cars. It is the author s view that the Domestic Tradeable Quotas (DTQs) scheme, developed by Fleming, is a possible alternative to regulating the average fuel economy of the car fleet. The DTQ scheme is an economic policy tool based on the idea of per capita carbon budgets aimed at reducing national carbon emissions. The nation implementing the scheme would set a carbon budget to be reduced over time. The carbon units making up the budget would be issued to individuals and organisations. Each individual would receive an equal and unconditional entitlement of carbon units; organisations would acquire the units they need from tender (a form of auction modelled on the issue of government debt). Low users could then sell their surplus on the national market to higher users. 10

The DTQ policy tool could provide a transparent and equitable alternative to vehicle fuel economy regulation, overcoming the problems of perverse effects and the unequal treatment of different car types (manufacturers). The scheme would also provide the much needed market driving forces for improved fuel efficiency and renewable energies, which for the transport sector, would drive R&D agendas and the need for fuel efficient cars, public transport, as well as facilities for cycling and pedestrians. The policy tool should also suffer less consumer resistance compared to fiscal measures as the DTQ scheme is transparent and is based on the principle of equity. However, the DTQ concept requires and deserves further research and development. 11

Table of Contents 1. INTRODUCTION... 14 2. THE EVOLUTION OF THE VOLUNTARY FUEL ECONOMY AGREEMENTS... 16 2.1 THE ORIGINS OF AN INITIATIVE TO ADDRESS CO 2 FROM PASSENGER CARS... 17 2.2 ACEA S RESPONSE... 19 2.3 THE COMMISSION UNITES ON A THREE PILLAR STRATEGY COM(1995)689... 19 2.4 1995-1997: FRUITLESS VOLUNTARY AGREEMENT TECHNICAL NEGOTIATIONS... 20 2.5 THE BIRTH OF AN AGREED TARGET... 25 3. ANALYSIS OF THE ACEA VOLUNTARY AGREEMENT... 27 3.1 THE TARGET AND CO 2 ACCOUNTING... 27 3.1.1 Is 140gCO 2 /km technically ambitious?... 27 3.1.2 Industry s commitments in CO 2 terms... 32 3.1.3 The CO2 test cycle: How a vehicle s fuel economy is over-estimated... 34 3.2 TERMS OF THE AGREEMENT... 37 3.2.1 Fuel quality... 38 3.2.2 Fiscal measures... 41 3.2.3 Vehicle related legislation that may affect fuel economy... 43 3.2.4 Competition issues... 44 3.2.5 The commitments of non-acea members... 45 3.2.5.1 The negotiations with KAMA...46 3.2.5.2 The negotiations with JAMA...47 3.2.6 Monitoring... 48 4. EVALUATION OF THE VOLUNTARY AGREEMENT... 52 4.1 THE CAPACITY OF ACEA S VOLUNTARY AGREEMENT TO ENSURE CO 2 REDUCTIONS FROM THE PASSENGER CAR SECTOR... 52 4.2 THE VOLUNTARY AGREEMENT AS AN ENVIRONMENTAL POLICY TOOL FOR APPLICATION AT EU LEVEL... 55 4.2.1 The legal aspects... 55 4.2.2 The Role of the Commission, Parliament, Council and NGOs... 57 4.3 METHODS TO REGULATE CO 2 FROM CARS... 60 4.3.1 Economic and regulatory measures to influence the motorist... 60 4.3.2 Regulating vehicle fuel economy standards... 64 4.3.2.1 The case for binding regulation as opposed to the voluntary approach...64 4.3.2.2 Methods of regulating vehicle fuel economy...65 4.3.2.3 Achieving guaranteed CO 2 emission reductions from passenger cars...68 5. BIBLIOGRAPHY... 73 6. ANNEX 1 THE ACEA COMMITMENT... 82 12

Table of Figures FIGURE 1 MAJOR EU SOURCES OF EU CO 2 EMISSIONS FROM FOSSIL FUELS 1985-1995... 15 FIGURE 2 FACTORS WHICH HAVE OFFSET FUEL ECONOMY IMPROVEMENTS FROM 1983-1997... 15 FIGURE 3 ACEA MEMBERS... 15 FIGURE 4 THE EVOLUTION OF THE VOLUNTARY AGREEMENT... 16 FIGURE 5 OUTLINE OF THE CONTENT FOR THE TECHNICAL DISCUSSIONS BETWEEN THE COMMISSION SERVICES AND ACEA.... 21 FIGURE 6 MARKET SHARE (%) OF NEW PASSENGER CAR MANUFACTURERS FOR 1999 EU REGISTRATIONS... 23 FIGURE 7 THE FUEL ECONOMY FLEET AVERAGE FOR THE EU 1980-1995.... 24 FIGURE 8 CO 2 REDUCTION GAIN ESTIMATIONS FOR THE EU PASSENGER CAR SECTOR (GRAPH)... 33 FIGURE 9 AUTO OIL II EMISSIONS FORECAST... 34 FIGURE 10 THE NEW TEST CYCLE OF DIRECTIVE 93/116/EEC IN COMPARISON WITH DIRECTIVE 80/1268... 35 FIGURE 11 HIGHER FUEL QUALITY ASSUMPTIONS... 38 FIGURE 12 NON-ACEA MEMBERS... 46 FIGURE 13 FACTORS TO BE TAKEN INTO CONSIDERATION BY THE MONITORING PROCEDURE... 51 FIGURE 14 THE CARBON BUDGET... 53 FIGURE 15 PROPORTION OF EXTERNAL AND INFRASTRUCTURE COSTS COVERED BY REVENUES IN TRANSPORT (1991)... 63 FIGURE 16 CONTRACTION AND CONVERGENCE... 69 FIGURE 17 THE MARKET FOR DOMESTIC TRADABLE QUOTAS... 70 FIGURE 18 ENVIRONMENTAL CITIZENSHIP... 71 FIGURE 19 THE ACEA COMMITMENT... 82 Table of Tables TABLE 1 MEMBER STATE LEGISLATIVE PROPOSALS TO REDUCE CO 2 FROM PASSENGER CARS.... 17 TABLE 2 POTENTIAL FUEL ECONOMY IMPROVEMENT ESTIMATES... 27 TABLE 3 THE COMMISSIONS REPORT ON TECHNOLOGICAL POTENTIAL AND COST... 28 TABLE 4 THE OPERATIONAL AND CAPITAL COSTS OF ALTERNATIVE TECHNOLOGIES COMPARED TO EXISTING TECHNOLOGIES... 29 TABLE 5 CONTRIBUTIONS FROM ALTERNATIVE POWERTRAINS AND NEW FUELS TO ATTAINING THE 2008-9 TARGET... 31 TABLE 6 CO 2 REDUCTION GAIN ESTIMATIONS FOR THE EU PASSENGER CAR SECTOR (TABLE)... 32 TABLE 7 EXAMPLES OF MEASURES TO REGULATE CO 2 FROM CARS BY INFLUENCING CONSUMER USE AND CHOICE... 62 13

Introduction 1. INTRODUCTION The car occupies an incredibly powerful force in our society as it has given us much freedom and can satisfy not only our needs but some of our desires too. Society assumes this newfound freedom to be a right. At the same time, the onset of global warming implies that the world needs to control its carbon budget. This also applies to the EU passenger car sector, which now has some 700 million vehicles on the road, a number that continues to increase. But efforts to improve the fleet s fuel economy fly in the face of market trends which show a shift towards larger, faster, more powerful and therefore more fuel inefficient cars. Fuel economy is not high on the list of many car purchasers priorities, so is not a market driving force. This is the dilemma that policy-makers face in attempting to regulate CO 2 from passenger cars. The EU passenger car fuel economy fleet average improved after the oil shocks of the 1970s due to increased oil prices, but has increased since the 1980s due to factors that that have added weight to the vehicle (see Figure 2). Thus ACEA estimates that some 70% of the improvement in EU passenger car fuel economy between 1983 and 1997 has been offset by factors such as EU legislation relating to noise, safety and emissions but also due to consumer demands such as comfort, electric motors (windows), air conditioning, extra power and faster speed 2. Growth in EU transport emissions has now outstripped the economic growth of the EU due to factors such as, inter alia, higher standards of living, a shift away from public transport and the establishment of the Single European Market (CEC 1998b). Carbon dioxide is the most important greenhouse gas as regards global warming potential, both in general and from passenger cars 3. The percentage of the EU s total CO 2 emissions attributable to the transport sector has increased from 19% in 1985 to 26% in 1995 (CEC 1998b). Passenger cars account for 50% of the EU s transport related emissions and 12% of total EU CO 2 emissions (CEC 1999a). The transport sector s CO 2 emissions, as the fastest growing major source of EU CO 2 emissions (see Figure 1 below), are expected to increase 22% above 1990 levels by 2000 and 39% by 2010 (CEC 1999a). While EU greenhouse gas emissions had decreased by some 4% below 1990 levels by 1994, such emissions have again risen to 1990 levels (CEC 1998a). Thus steep reductions will have to be sought from all EU CO 2 producing sectors if the current CO 2 growth trend is to be reversed. With an EU commitment to the UNFCC Kyoto Protocol (1997) to reduce greenhouse gas emissions by 8% below 1990 levels by 2008-2012, the European Commission has attempted to tackle CO 2 from passenger cars with a three pillar strategy as set out by COM(1995)689 (CEC 1995). This policy package comprises a voluntary fuel economy agreement between the European Commission and the vehicle manufacturers selling cars in the EU (including non- European manufacturers), a fiscal framework for Member States and a consumer information scheme. This paper aims to explore and analyse the evolution and form of the voluntary fuel economy agreement, established between the Commission and the automobile industry represented by ACEA (European manufacturers), JAMA (Japanese manufacturers) and KAMA (Korean manufacturers). By such analysis, it is intended that factors influencing the decision-making process will be identified and the capacity for the voluntary agreement to safeguard the environment shall be evaluated. It is hoped that lessons can be drawn from this case study as 2 Vans and SUVs doubled their European market share reaching 5.2% in 1998.(FT Automotive Quarterly Review 1999). There is a linear correlation between engine capacity and carbon dioxide emissions. Car engines have been getting bigger and specific output per litre engine capacity has been increasing in recent decades (Holman 1992). 3 Carbon dioxide is not the most powerful greenhouse gas but is the most volumous and so contributes 58% to global warming (IPCC 1996). While carbon monoxide, volatile organic compounds and nitrous oxide compounds are also released through hydrocarbon combustion and all directly or indirectly contribute to global warming, carbon dioxide has the greatest global warming potential from passenger car exhausts (Wade et al 1994 ). 14

Introduction regards the appropriateness and effectiveness of voluntary agreements as a tool for environmental policy making at EU-level but also with regard to effectively regulating CO 2 from cars. Relevant documentation has been analysed and supplemented by personal communication with many individuals directly involved, or with a relevant interest, in the voluntary agreement negotiations. Figure 1 Major EU sources of EU CO 2 emissions from fossil fuels 1985-1995. Source: DG7 EUROSTAT (CEC 1998g) Figure 2 Factors which have offset fuel economy improvements from 1983-1997 Source: ACEA website http://www.acea.be Publications: addressing climate change Figure 3 ACEA members BMW AG; Daimler-Benz-Chrysler AG; Fiat Auto S.p.A.; Ford of Europe Inc.; AB Volvo (became part of Ford 1998); General Motors Europe AG; Dr. Ing. H.c.F.Porsche AG; PSA Peugeot Citroen; Renault SA; Volkswagen AG 15

The Evolution 2. THE EVOLUTION OF THE VOLUNTARY FUEL ECONOMY AGREEMENTS Figure 4 below outlines the evolution of the voluntary agreements established between the Commission and automobile industry, since their origins in 1991. Figure 4 The evolution of the voluntary agreement 1991 Council Directive 91/441 EEC calls for proposals to reduce CO 2 from passenger cars 1991 DG 11 conducts analyses of legislative proposals 1992 Purchase tax proposed by MVEG sub-group to MVEG is rejected 1992 Circulation tax proposed by MVEG to European Council is rejected 1992 Consensus within Commission that CO 2 from passenger cars is too difficult to legislate 1995 Voluntary agreement proposed as part of three pillar strategy COM (1995) 689 Early 1996 Commission enters into technical discussions with ACEA June 1997 ACEA offers 167gCO 2 /km which is rejected by the European institutions December 1997 UNFCC COP3, Kyoto, Japan February 1998 Strasbourg workshop brings together negotiators to discuss regulating CO 2 from passenger cars March 1998 140gCO 2 /km accepted, but Commission expresses caution with regard to conditions June 1998 Auto-Oil 1 legislation passed. Parliament uses co-decision procedure to bring implementation forward by five years. July 1998 ACEA and Commission finalise terms of agreement July 1998 ACEA presents a Commitment and the Commission responds with COM (1998) 495 presented 5 th February 1999 Commission presents ACEA with Recommendation 1999/125/EC May 1999 Negotiations with KAMA completed. KAMA given one extra year to achieve intermediate and final targets May 1999 Commission and JAMA dispute over the term market mix. JAMA s Commitment proposal rejected. July 1999 JAMA and Commission agree on an extra year for final target of 140gCO 2 /km and a wider intermediate target range of 165-175gCO 2 /km as opposed to 165-170gCO 2 /km 12 th October 1999 Environment Council supports the Commission s intention to accept the agreements of JAMA and KAMA 13 th December Adoption of the Directive on Fuel efficiency labelling of cars Directive 1999 1999/94/EC 13 th April 2000 The Commission presents Recommendations for JAMA and KAMA, 2000/304/EC and 2000/303/EC respectively. August 2000 The Decision on Monitoring of CO 2 emissions from new passenger cars D1753 (2000) is adopted. 2003 ACEA must be within the intermediate target range of 165-170gCO 2 /km and JAMA within 165-175gCO 2 /km. ACEA and JAMA will also undertake a review of the progress made and the potential to achieve 120gCO2/km by 2012. 2004 KAMA must be within intermediate target range of 165-170gCO 2 /km. KAMA will also undertake a review of the progress made and the potential to achieve 120gCO 2 /km by 2012. 2008 ACEA must achieve a fleet average of 140gCO 2 /km 2009 KAMA and JAMA must attain a fleet average of 140gCO 2 /km 16

The Evolution 2.1 The origins of an initiative to address CO 2 from passenger cars Article 5 of the Directive 91/441 EEC (Council of the European Union 1991) requires that the Council shall decide on measures designed to limit CO 2 from passenger cars, through acting by a qualified majority on a proposal from the Commission. Member States were therefore invited by the Commission to suggest policy options and the Motor Vehicles Emissions Group 4 (MVEG) was entrusted by the Commission with the task of putting forwards a politically acceptable policy proposal (Ends Report 1992 No. 215). Table 1 Member State legislative proposals to reduce CO 2 from passenger cars. Country Proposal Comment Germany (The ENDS Report 215 Dec 1992) France (The ENDS Report 215 Dec 1992) Italian Delegation Proposals to MVEG 18.7.91 UK Delegation to MVEG1991 Dutch Delegation to MVEG 10.1.91 CO 2 emission limits to be imposed on different size bands of car Absolute emission limit Average CO 2 emission standard. Companies exceeding this average would pay a fine, while those beating it would be offered some financial benefit (ENDS Report 207 April 1992). Variable car purchase tax based on actual CO 2 emissions. Below a threshold set initially at 100g of CO 2/km in 1994, no tax would be payable on a new car. Above a ceiling set initially at 400g/km no extra tax would be payable. Between these two values the tax rate would rise in steps exponentially. Tradable emission credits which would be bought and sold among manufacturers according to whether their models met a specified and progressively tightened fuel efficiency standard. 1. The covenant type approach setting of stepwise strengthened emission requirements over a period of time 2. The emissions requirement type approach - on basis of vehicle weight or engine size, possibly supplemented by other factors Would do nothing to discourage consumers from purchasing larger high emission vehicles (which happen to be produced in volume in Germany). Legislating by bands based on a particular parameter e.g. weight, is also vulnerable to borderline effects as a manufacturer might opt to increase the weight of a model so it might move up into a heavier category where it would suffer less penalisation As the standard would be tightened so the permits would become more expensive thus increasing the incentives for improving fuel efficiency. Similar to tradable permits - see UK. Difficulty in setting limit as internal industry information is needed. Also difficult to set fines and rebates at an effective level. High transaction costs. The tax may have to be extremely high to be effective and would thus be politically unacceptable. Manufacturers of heavier cars would argue unfair discrimination with falling sales and that consumer needs design the car i.e. a family of five needs more than a Ford fiesta. A tradable permits system would be extremely complex. High transaction costs. The system has worked well in the US, but cars are a very different system to stationary sources of industrial plants. Depends on creating a true competitive market, setting the correct limit values so that companies have real incentives to improve fuel economy. Also might be difficulty to police the system and may be difficult to set fines at an effective rate. Fluctuations in cost or availability of credits might make planning very difficult. Such a scheme may also conflict with WTO/GATT. Difficulty on setting limits - requires internal knowledge of industry to be truly effective. As for Germany - setting limits by a certain criteria may fall vulnerable to borderline effects. Several competing options emerged from the Member States. It became clear that the diversity of national market composition characteristics and conditions complicated attempts 4 MVEG was composed of a group of national officials and motor industry representatives drawn together in the mid-1980s to address road transport emissions. 17

The Evolution to develop legislative options. The options presented to the Commission were particularly criticised for supporting national manufacturing interests (ENDS Report 1992 No. 207) (see Table 1)). Each proposal had its positive and negative points and consensus on a single policy option was not achieved. In November 1992, a MVEG sub-group, which had been formed to discuss fiscal measures, proposed a purchase tax, based on a combination of the weight of each new vehicle and its CO 2 emissions per kilometre, in order to achieve a 40% fuel economy improvement for new cars by 2005 5 (The ENDS Report, No. 225). There were objections from Member States, particularly from the UK, that MVEG lacked the competence to discuss fiscal issues (The ENDS Report, No. 215). The proposal presented in December 1992, championed strongly by DG11, received little support from DG3, DG21 and MVEG, as rates would have needed to be unacceptably high 6 to achieve the desired effect (Interview 1; The ENDS Report, No. 225). There was also concern that such rates, as experienced in Denmark, would slow new car sales and age the vehicle car parc, thus defeating the policy objective of improving the average fuel economy of the entire fleet (The ENDS Report, No. 215). The Commission then invited the MVEG sub-group to consider the option of an annual circulation car tax based on CO 2 emissions. The issue of the difficulty in attaining tax harmonisation came to the fore, especially in the context of requiring unanimity from Council 7 and taking account of the diversity of current vehicle fiscal policy throughout the EU. All Member States would therefore have to agree to the high tax rates deemed necessary to achieve the environmental objectives. In addition, some states such as Luxembourg and the UK, would probably have vetoed such a proposal in support of the principle of subsidiarity on fiscal measures. This was well demonstrated by the EU s attempt to introduce the carbon tax in 1997 (ENDS Daily 13.3.97). Member States also have different quantitative commitments under the UNFCC Kyoto Protocol which will require varied policy measures. The circulation tax proposal was rejected at the Environment Council meeting on 9 th December 1992 (The ENDS Report 1992, No. 215). Not only was there division between Member States on the issue of legislating CO 2 from passenger cars with harmonised fiscal policies, but the Commission was divided too. While DG11 was strongly championing both of MVEG s tax proposals, DG21 rejected both proposals and DG 3 was not especially keen on either but supported the circulation tax more than the purchase tax (Interviews 1 and 2). Thus there emerged a competitive issue as to which DG would win (Interviews 1 and 2). The failure of the Commission to reach consensus on how to regulate CO 2 from passenger cars significantly reduced the Commission s bargaining power in the voluntary agreement negotiations with ACEA which followed in 1996 and 1997. As a result, the Commission could not threaten ACEA with binding legislation and it was clear to all involved, including industry, that the Commission did not want to return to the fruitless discussions of the early nineties. 5 The proposal involved purchase charges that would be weight and gco 2/km based, set at zero in 1995 for cars operating at no more than 160gCO 2/km and reduced each year by 5g so that 110gCO 2/km would be the limit for 2005 (the limit of current technology). The legislation would be implemented by a Directive stating the minimal rates but giving the Member States flexibility in application (The ENDS Report 1993 No. 225). 6 Rates in the region of 4000-8000 for a car emitting two and half times as much CO 2 as the limit set were envisaged (The ENDS Report1993, No.225). 7 EU fiscal measures require unanimity from Council (under Article 100A of the EU Treaty (Horspool 1998)) which means a Member State can veto a proposal. By adoption of a framework with such minimum limits, many Member States could then under the principle of subsidiarity choose to tax at higher levels to achieve their environmental aims. This scenario is at odds with the principle of fiscal harmonisation and the establishment of the single market, which is strongly supported by ACEA. 18

The Evolution 2.2 ACEA s response In immediate response to the Commission s interest to address CO 2 from passenger cars, ACEA offered to reduce the sales weighted CO 2 emissions of the EU new car fleet by 10% on a voluntary basis within the period 1993-1995 (ACEA 1991; Europe Environment 7.1.92). This was set against demands such as those of the UK s transport secretary of the time, Mr Rifkind, and the UK s Royal Commission of Environmental Pollution (RCEP 1994), urging manufacturers to pursue a 40-50% improvement in fuel efficiency by 2005 (The ENDS Report No. 207). ACEA was even at this time placing much emphasis on the CO 2 reduction benefits of direct injection diesel technology (ACEA 1992a). While arguing that fuel efficiency improvements had been offset by various factors outside their control (see Figure 3, page 15), ACEA was also promoting non-technical ways to reduce CO 2 from cars i.e. driver behaviour campaigns, improved public transport, technical inspection of in-use vehicles, cleaner fuels and traffic management (ACEA 1992b). In response to the MVEG proposals for a purchase tax and circulation tax, ACEA proposed a CO 2 emission tax levy to be fully harmonised in the Member States (ACEA 1992b). ACEA argued that legislation based on weight would be unfair as the spread of CO 2 emissions can reach up to 100% in each weight class for various reasons 8. The automobile organisation demanded that the tax would 1) replace existing taxes, 2) be solely CO 2 based and 3) would not be penal on particular segments of the market. However, such a tax would face the same problems as a purchase tax or circulation tax as proposed by MVEG. As mentioned previously, a harmonised tax rate would be dictated by the Member States proposing the least ambitious demands such that unanimous agreement among the Member States 7 could be achieved. Further, replacing other taxes would remove taxes that aim to internalise various other environmental and social costs. And finally, a tax addressing fuel economy can not avoid penalising larger cars, as they are generally more fuel inefficient than smaller cars because they are heavier. 2.3 The Commission unites on a three pillar strategy COM(1995)689 The idea of a voluntary agreement was initially suggested and strongly championed by DG3 (Interviews 1 and 2). This raised the inter-dg competivity issue again, but this time between DG3 and DG11 as the Danish Commissioner for the Environment, Bjergaard, initially took the Danish line of weak support for the voluntary approach (Interviews 1 and 2). Eventually deadlock was broken as DG11 feared returning to the complex legislative option debate (refer to chapter 2.1, page 17) having realised that fiscal policies could not be used as stand-alone measures to regulate CO 2 from passenger cars (Interview 1). The adoption of the voluntary approach by the Commission was regarded as a personal victory for DG3 (Interviews 1 and 2) but meanwhile DG11 took control of the board by proceeding with the development of the three-pillar strategy which was very much Bjergaard s initiative (Interview 2). The strategy, described in the communication COM(1995)689 (CEC 1995), encompasses a voluntary agreement with industry, a framework for fiscal measures and a fuel economy consumer information scheme The Commission was prepared to accept a 25% improvement by 2005 (para 34 of COM(1995)689, CEC 1995), but the Council was demanding a 35% reduction by 2005 (Council of the European Union 1996b). The Commission therefore expressed its intention to achieve the extra 10% though the fiscal framework and consumer information scheme (COM(1995)689; CEC 1995). 8 Factors such as inter alia, engine displacement and output, transmission configuration, aerodynamics, seating and loading capacity and performance. 19

The Evolution Between the Commission s failure to reach consensus on a method to legislate CO 2 from passenger cars in the late eighties and the year 1995 when the three pillar strategy was proposed, important events took place which altered the Commission s approach to environmental policy-making. In 1992, the Treaty of Maarstrict was signed and ratified. The Treaty amendments gave the principle of subsidiarity much greater emphasis and the Council made clear that the interpretation of its provisions would see that the Community would only legislate if it would be more effectively carried out at Community level than at national level, to the extent necessary and as simply as possible (Pallemaerts 1999). The Council also made clear that Directives would be preferable to Regulations and where appropriate non-binding measures would be preferred to legally binding ones (Council of the European Union 1992). The early 90s also witnessed the publication of the 5 th Environmental Action Programme which stated the Community s aim to broaden the range of instruments used for policymaking. The Commission also states in this publication that perhaps too great a reliance was placed on regulation of the command and control type and that the legislative approach may not always be the best choice as the first step. Council s strong support of the new voluntary approach meant that the Commission was provided with a politically supported alternative option for legislating CO 2 from cars. The Commission was at the same time presented with an opportunity to put the new Community philosophy into practice. But while the Commission at last managed to move forwards on legislating CO 2 from passenger cars, its declared over-enthusiasm, encouraged by Council, to increasingly use new instruments such as environmental agreements as part of its new approach or third way, significantly contributed to diluting the Commission s bargaining power. The conclusion of a study of voluntary agreements in Germany by Rennings et al (1997) makes reference to precisely the shortcomings that the Commission would later experience throughout the negotiations due to such over-enthusiasm and the lack of an alternative legislative binding proposal: If a decision to give preference to voluntary solutions in general is made or if a decision in favour of such solutions is taken at an early stage, this too is counter productive, because the substance of negotiated solutions, the governmental potential for threats, is weakened and delays in the form of a stamina contest are provoked. (Rennings et al 1997) 2.4 1995-1997: Fruitless voluntary agreement technical negotiations The initial attitude of ACEA to an EU-wide voluntary initiative was not overly enthusiastic. ACEA made its displeasure known that the Commission made public its first contact with ACEA regarding such a Commitment (Interview 1). While ACEA reminded the Commission of its 1991 voluntary offer, those manufacturers who had in the meantime made national voluntary offers such as Germany, France and Sweden 9, saw no reason why they had to go any further with an EU-wide commitment (Interview 3). Meanwhile, despite the inter-dg deadlock following the discussions in the early nineties on how to regulate CO 2 from passenger cars, unity within the Commission developed once the voluntary approach had been accepted by DG11 (Interviews 1 and 2). Two desk officers from DG11 and DG3 were made responsible for implementing the three-pillar strategy at working level. The inter-dg working relationship has been generally regarded as an excellent example of inter-dg co-operation (Interviews 1, 2, 3 and 4). 9 PSA and Renault offered the French government 150gCO 2/km by 2005; Volvo offered the Swedish government a fleet fuel economy improvement of 25% by 2005; and the German manufacturers offered the German government a fleet fuel economy improvement of 25% by 2005 (CEC 1995); Environment Watch 4.10.96; ENDS Daily 20.4.98. 20

The Evolution Figure 5 Outline of the content for the technical discussions between the Commission Services and ACEA. Source: (CEC 1997c) 1. The structure of an agreement The structure of an agreement as regards monitoring, addressing free-riders, intermediate targets and the issue of transparency and objectivity in establishing contributions to be made by individual manufacturers. The parties have identified a need to outline framework conditions which would make a fuel economy/co 2 emission objective conditional on certain factors affecting its attainment. This would provide a basis for a review of the objective in the event that the framework conditions are not met. It has been noted that Community and Member States carry a responsibility to ensure coherence between different policy objectives, and to put into place initiatives which increase the demand in the market for more fuel efficient cars. It has been understood that in reaching an eventual overall CO 2 objective, other key environmental and vehicle safety objectives will not be prejudiced. 2. The CO2 emission objective Industry has expressed reservation of Council s 120gCO 2 /km by 2005. The parties have recognised the need for a global impact study covering economic, industrial and social aspects related to technical possibilities which should be conducted as soon as possible. Industry would put forward information on the potential for a CO 2 reduction. The Commission intends to draw on the various experts to assess information and would ensure the assessment is carried out in a transparent and objective way. 3. Free riders Importers and other foreign manufactures i.e. KAMA and JAMA. In talks with ACEA the need to include the most important importers and other foreign manufacturers in an equitable burden sharing arrangement and to avoid any free-riding has been emphasised. 4. The form of the agreement The legal form has not been examined in detail yet and the Commission is waiting response to the communication on environmental agreements COM(96)561. It has also became apparent that some of the issues to be addressed in an agreement and certain potential solutions are technically complex and commercially sensitive and will require further study. The need for internal co-ordination within ACEA, which is of course necessary for the success of the agreement, means that the Commission may have to show flexibility in its approach to the negotiations. The working level meetings were held on average once a month throughout 1996 and 1997, between the two desk officers of DG3 and DG11 and two Ford engineers representing ACEA. The technical discussions were based on the framework proposed by the Commission as outlined in Figure 5 above (CEC 1997c). This framework covered objectives, agreement structure and terms. With just two Ford engineers representing ACEA, it appears that ACEA members were inadequately represented. In addition, neither NGOs nor Member State experts were involved in the discussions or negotiations. Further, it was not until the agreement with ACEA had been 21

The Evolution established, that the Commission initiated negotiations with the non-eu manufacturers KAMA and JAMA (see chapter 3.2.5, page 45). Specific investigations and studies, seemingly orientated towards ACEA s defence, were carried out e.g. studies to challenge Commission calculations; analyses to prove Greenpeace s SmiLe car was no different from an ordinary prototype and would not be suitable for commercial production (refer to The ENDS Report No259 August 1996); investigations to question global warming uncertainties (Interviews 5 and 6). Other issues were also discussed at the request of industry, such as: anti-diesel tax campaigns, air quality requirements, the necessity for cleaner fuel, safety and recycling policy requirements (Interview 5). As the Commission could not attest ACEA s findings and as it had conducted no technical studies of its own, its bargaining power was considerably diminished. ACEA admitted that the Commission desk officers would have benefited from technical support. At the same time the two Commission desk officers did not receive much in the way of higher-level political support and were very much left to their own devices (Interviews 1, 2 and 4). In addition, the voluntary agreement had to compete for priority with the air quality Directives that were high on the Commission s agenda during this time (Interview 3). Due to time and financial constraints, DG 11 believes the Commission could not have conducted its own technical and economic analyses, but moreover believes that attempting to establish consensus between differing technical analyses of the Commission and industry would be near impossible and likely to result in deadlock (Interviews 1 and 5). The Commission also viewed that it would have been difficult to ensure the independence of such technical studies as it is the automobile industry that possesses the technical knowledge (Interview 5). While the working relationship of the DG3 and DG11 desk officers may be a good example of inter-dg co-operation, the effectiveness of the arrangements for the technical discussions from 1995 to 1997 is questionable. The two Commission desk officers found that ACEA s representatives adopted a very technical and bottom-up approach, far removed from the political way of the European institutions. Meanwhile ACEA s representatives, who were engineers by profession, found the lack of understanding of the automobile industry by the Commission extremely frustrating. While the expectations of both sides were well understood by both parties, both negotiating parties agreed that bridging the gap by accepting each other s approach was the most difficult aspect of the negotiations. According to the negotiators, both the industry and the Commission knew throughout 1996 the key elements and numbers to ensure a deal, including the 140gCO 2 /km target. An agreement was nearly reached at the beginning of 1997 as the Commission was particularly keen to give DG11 Commissioner Bjergaard something to take with her to the UNFCC Third Conference of the Parties (COP3) in Kyoto in December of that year. However, the deal fell through in early 1997 as some manufacturers, including the Germans, believed that the outcome of COP3 would not require them to act (Interviews 3 and 6). As a result, ACEA decided that Ford and Renault would work on estimating a target (before COP3) through technical studies which would be based on the use of existing technology, an unaffected economic situation and would take into account factors which could negate fuel efficiency improvements. 22

The Evolution The automobile industry is particularly competitive (see Figure 6) and is suffering from considerable over-capacity 10. With an EU market approaching saturation, and squeezed profit margins, the industry will not likely welcome additional spending if a competitive advantage is not to be gained. Manufacturers are also likely to be very protective of their most profitable market segments which generally consist of more luxurious cars which are often larger, faster, more powerful, and therefore more fuel inefficient than the average car. In effectuating the technical studies on behalf of ACEA, it would have been in the competitive interest of Renault 11 and Ford, and indeed all car manufacturers, to withhold information as to what could truly be achieved under various time-frames or economic scenarios. Further, as Renault and Ford were acting on behalf of ACEA members, they also needed to defend the lowest common denominator. Figure 6 Market share (%) of new passenger car manufacturers for 1999 EU registrations Source: ACEA website Ranking Manufacturer % market share 1 Volkswagen 19 2 PSA Peugeot Citroen 12.1 3 Ford 11.7 4 JAMA 11.5 5 General Motors 11.4 6 Renault 10.9 7 Fiat 9.9 8 DiamlerChrysler 5.5 9 BMW Rover 4.6 10 KAMA 3.1 11 Other 0.3 TOTAL 100 In June of 1997, ACEA offered the Commission a fleet average fuel economy target of 167gCO 2 /km by 2005 with certain conditions relating to: a joint strategic research programme; future legislation in relation to vehicle safety, environmental standards and fuel quality; the market share of diesel cars; and fiscal matters (CEC 1997b). The offer was equivalent to a 10% improvement on the baseline of 186gCO 2 /km of 1995 i.e. almost identical to the percentage improvement offered by ACEA in 1991 (ACEA 1991a; see Figure 7). DG11 insists that the aggregate of the ACEA member national targets 9 already offered exceeded ACEA s offer, while ACEA contests that the difference was not significant. The EU institutions all rejected ACEA s offered target and terms which were clearly geared towards the business-as-usual scenario (Interviews 1, 2 and 5). The conditions accompanying 10 Financial Times 15.1.98. Survey of world motor industry: Growth runs into a jam - overcapacity still hovers at around 25-30% with too many car companies making too small profits. 11 Renault assisted ACEA in carrying out the technical studies. 23

The Evolution the target would have inappropriately restrained the Commission particularly with respect to its ability to develop any other policies relating to vehicles (CEC 1997b). One of the main reasons for the unambitious target and conditions of the June 1997 offer was due to the intense competition between ACEA members as they could not agree to the individual contributions each would make (CEC 1997b). The Commissioner for the Environment (DG 11), Bjergaard, therefore declared she would write to the environment ministers to urge them to raise the issue with their national manufacturers (ENDS Daily 16.10.97). The effort was fruitless for ACEA s second attempt to produce an offer at an ACEA board meeting on the 1 st December of 1997 also failed. It was then uncertain how and whether the ACEA position would develop, thus the Environment Council proposed a deadline of March 1998 (Council of the European Union 1997b). Despite the attainment of an official agreement, there is still no clear consensus within ACEA over individual contributions to be achieved by each manufacturer (Interviews 3 and 4) and both the Commission and ACEA still remain unclear as to what constitutes ACEA member equivalent efforts. Figure 7 The fuel economy fleet average for the EU 1980-1995. Source: The European Environmental Bureau With no technical basis provided by any of the EU institutions for their proposed targets 12, and no technical studies to challenge or verify ACEA s 167gCO 2 /km offer of June 1997, the automobile organisation stuck to its unambitious offer (Ends Daily 18.11.97; ACEA 1997). ACEA published its justification for the offer in its own newsletter: While it is technically possible to achieve 120gCO 2 /km it is impossible for the whole fleet to achieve 120gCO 2 /km, such a change would imply radical downsizing of the whole fleet with cars that would be neither affordable nor meet the requirements of most car users... it would be possible to achieve an average 150-160gCO 2 /km by 2005... and the Commission has admitted that a longer time horizon may be necessary. ACEA welcomes increased research. 12 The Commission was calling for a 25% improvement by 2005, the Council was pushing for 120gCO 2/km by 2005 or 2010 at the latest and Parliament set targets of 120gCO 2/km by 2005 and 90gCO 2/km by 2010 with Austria stating that Parliament s 90gCO 2/km should be a maximum (ENDS Daily 16.10.97). 24

The Evolution We believe that this time should be put to good use in order to investigate the many scientific uncertainties concerning the nature and reality of global warming and climate change. (quoted in ACEA newsletter No 46 October 1997) By late 1997 a report was presented by Arthur D Little on behalf of ACEA, illustrating the impacts of trying to achieve 120gCO 2 /km by 2005 with findings including loss of jobs, adverse trade balances, contraction of the industry and necessary restructuring (Interview 3). The report was shared with the Commission and governments in early 1998. 2.5 The birth of an agreed target There is no doubt that the establishment of the UNFCC Kyoto Protocol at COP3 in December 1997 was a major driving force which helped to bring the negotiations to their final conclusion in July 1998 (Interviews 1 and 2). This is because COP3 provided the negotiations with additional momentum and a timeframe. In late 1997, following the rejection of ACEA s unambitious offer, the negotiations moved from working level to higher level and several new actors joined the scene. The presidency of ACEA changed hands from Renault to BMW, with Pischetsrieder taking the helm from Schweizter. The Commissioner for the Environment, Bjergaard, and the Commissioner for Industry, Bangemann, with the support of their cabinets, were now communicating regularly with Pischetsrieder and his team of auto-industry Director-level representatives. There existed, amongst these new actors, both the skills and the political will to overcome the problems. However, the Commission now had the advantage, as the negotiating approach was now political and non-technical. Pischetsrieder represented BMW; a company which is not a market-leader and would be vulnerable to the demands of limit-value legislation (refer Figure 6, page 23). Pischetsrieder therefore recognised the need to finalise a deal that would be favourable for the German manufacturers (Interview 3). The German industry was also familiar with the voluntary approach. Despite remaining divisions within ACEA - as some companies such as Volkswagen and Peugeot, were more in favour of legislation than others such as BMW and Daimler-Chrysler (Interviews 1 and 7) the negotiations progressed rapidly. Several feel that Pischetsrieder s own personality and approach made a significant contribution to moving the negotiations forward (Interviews 1, 2 and 3). In 1999, the ACEA presidency passed from Pichetsrieder to Piëch of Volkswagen. Piëch adopted a more confrontational approach compared to Pischetrieder. This was illustrated when Piëch persuaded the German Chancellor Schroeder, a former Volkswagen supervisory board member, to stall the passing of the end-of-life vehicle legislation which Council had already agreed to (Financial Times: Simonian 29.6.99 and Smith 16.6.99). Schroeder acted on the concerns of Piëch and made use of the German presidency of Council to postpone the Directive in March of 1999. In June, the Germans were joined by the UK and Spain and the Directive was blocked. This caused outcry among other Member State ministers and received disbelief, particularly within Germany, as Germany has a reputation for taking a very tough line on environmental issues (Interview 1). DG11 reports, however, that ACEA s approach to the voluntary agreement has not altered with changes in ACEA s presidency. Following the UNFCC COP 3 negotiations of December 1997, a workshop was held in Strasbourg on February 18 th 1998 to address the various issues that concerned both the Commission and the automobile industry with regard to establishing the voluntary agreement 13. The meeting was held in the context of the Commission s expectation of a new 13 (CEC 1998i) Workshop on CO 2 emissions from passenger cars - Strasbourg, 18 th February 1998. 25

The Evolution offer from ACEA by the March 1999 Environment Council. The workshop - attended by all those directly or closely involved in the negotiations, a quarter of which have been interviewed for the aims of this paper - aimed to gain a shared and deeper understanding of the issues surrounding the fuel economy agreement and to apply considerable pressure to ACEA (Interviews 1 and 2). Legislative options were therefore presented by DG11 to send a strong political message to the automobile industry (Interviews 1 and 8). This was the first time that legislation had been seriously discussed since the brainstorming sessions of MVEG in the early 1990s. A concrete legislative proposal, however, was not developed. The importance of ensuring the voluntary agreement s success was emphasised at the Strasbourg workshop, as its future use would depend on it. The message was particularly strong from the Commissioner for Industry and the UK s environment minister Meacher (Interviews 10 and 2). As the UK whole-heartedly supported the concept of the voluntary approach, Meacher was keen to make use of the UK s presidency of Council. He therefore addressed the manufacturers at the Strasbourg workshop, wrote to Pischetsrieder in an attempt to secure the agreement (Interview 10; Kampfner, Financial Times, 16.2.98) and to assist his calls for action, he advocated the extension of the timeframe of 2005 to 2010 at the latest. On rejection of ACEA s 1997 offer of a 10% improvement, the Commission reaffirmed its demand for a 25% improvement and made clear that 140gCO 2 /km would be the acceptable minimum (Interview 1). Having declared the minimum it would accept, the Commission was unlikely to achieve any more than the stated 140gCO 2 /km. Yet, the Commission refused to compromise and the 140gCO 2 /km target for 2008 was officially agreed in March of 1998. The commitment also included an intermediate target range of 165-170 gco 2 /km by 2003 and some members of ACEA have committed to introducing models to the market by 2000 with a fuel economy of 120gCO 2 /km or less. However, the achievement of the latter commitment was already assured when the agreement was established. ACEA s commitments will be jointly monitored by the Commission and ACEA. A joint review in 2003 shall assess the industry s progress relative to the intermediate target range and at the same time ACEA shall review the potential to achieve the Community s goal of 120gCO 2 /km by 2012. There was general satisfaction from DG3 and DG11 that 140gCO 2 /km moved beyond business-as-usual and was politically acceptable (Interviews 1, 2 and 5; CEC 1998k; CEC 1998l). However, concessions were clearly awarded to industry through the drafting of the terms of the agreement, which is later discussed in chapter 3.2. 26

The Analysis 3. Analysis of the ACEA voluntary agreement 3.1 The target and CO 2 accounting 3.1.1 Is 140gCO 2 /km technically ambitious? There is a general consensus among academic studies (although they use various assumptions and bases for their estimations) that a 40-50% improvement in an average car s fuel economy would technically be possible using existing commercial technologies by 2010, at little extra cost over 10-15 years (Martin and Michaelis 1992; De Cicco and Ross 1993; Binsbergen et al 1994; Michaelis 1996; see Table 2 and Table 3 below). Thus the target of 140gCO 2 /km by 2008 (25% improvement), as agreed by the Commission and the automobile industry, appears to be technically unambitious. Table 2 Potential fuel economy improvement estimates Study Poulton (1997) Johansson (1995) UK DETR (in evidence to RCEP 1994) EC MVEG (quoted in Fergusson and Wade 1993) US Congress, office of technology assessment (OTA 1995) Lovins et al (1996) Plotkin & Greene (1997) Potential improvement for new cars Timescale Country concerned Comments 33% 1995-2010 UK New car fuel economy of 4.8litres/100km achieved 50% 1995-2015 Sweden Fuel efficiency of 6.4 litres/100km could be achieved without increasing vehicle production costs. No reduction in vehicle size or performance. 40% Next 10-15 years 25-40% in average new car fuel efficiency 0-3.5l/100km (70-80mpg) Up to 75%, 2.4l/100km 5.6-6.0l/100km by 2005 and 4.7l/100km or lower by 2015 25%: 2000 40%: 2005 UK EU Technology already commercial or at an advanced stage in design. 2005-2015 USA The OTA study is based on both automakers view and a combination of literature review, interviews with suppliers and examination of prototypes. USA Employs a hybrid drive system, uses ultralight materials. Requires a paradigm shift that would render existing tooling and equipment obsolete. Significant costs due to 2005 and 2010 USA radical change. Improvements in aerodynamics, rolling resistance, weight reduction, and existing advanced engine technologies on market place. Performance and amenities of 1995. 27

The Analysis However, technical feasibility is not the same as economic feasibility. The Commission states in COM(95)689 (see below) that a 40% improvement in the average fleet fuel economy would involve a cost increase of 940-2270 Euro per vehicle. At the same time, smaller energy efficiency improvement costs are much smaller i.e. a reduction of 30% in specific fuel consumption gives a price increase that is almost half as high as the cost estimates for 40% (source: a report prepared for the UNFCCC based on an overview of cost estimates in the literature quoted in Phylipsen et al 1996). The cost of improving a vehicle s fuel economy can vary for each vehicle depending on its type and design (e.g. weight, engine size). As vehicles have different fuel economy reduction potentials for the same cost, the cost-effectiveness for achieving fuel economy improvements depends on the way in which the car fleet is regulated. This is discussed later in chapter 4.3.2 (page 64). The external costs of passenger cars to society and the environment are currently far from being adequately internalised. Thus paying for improved fuel economy technology is one way of internalising such costs. It is also more likely that consumers will be more accepting of paying for the cost of improved technology than of paying taxes aimed at regulating their driving and purchasing behaviour. The issue of effectiveness and political acceptability of fiscal measures is also briefly discussed later in chapter 4.3.1 (page 60). Table 3 The Commissions report on technological potential and cost Technical report of COM(95)689 (CEC 1995a) Technological potential The technological potential and costs: on the basis of figures by the US National Research Council adapted where necessary to the characteristics of the car fleet in the EU, on the specific technologies available to reduce passenger car fuel consumption, the following can be stated: Improvements in engine and transmission technologies, rolling resistance, aerodynamics as well as a vehicle weight reduction by 10% through the use of new materials together can improve the fuel efficiency for petrol cars by some 40% using mid-range estimates. The technologies considered are already commercial or at an advanced stage of design. This reinforces the estimate by the Commission s MVEG. Costs On the basis of the same study it is expected that the total costs of a fuel economy improvement of 40% is between ~ 940ECU and 2270 ECU per vehicle. The payback (UK study @ 8% discount rate) estimates that a 14-32% improvement would enable recovery in the vehicle s lifetime or an improvement of 5-28% could acquire a three year payback (no average annual mileage stated). CO2 saving 40% between 1996 and 2005 would reduce total end of pipe CO2 from EU passenger cars by 17.5% compared to current trends in 2005 and by 30.1% in 2010. Due to the growth in the vehicle fleet and mileage, CO2 from passenger cars would increase by 4.9% by 2005 as compared to 1990 and decrease by 6.9% by 2010 as compared to 1990. 28

The Analysis Some also argue that increased initial purchase costs may slow the renewal of the fleet, therefore counteracting the desired environmental objectives. However, this is a muchdebated issue as studies show that the environmental gains can be positive if the environmental impacts of the entire life cycle of the car are properly considered (Wells & Newenhius 1997). At the same time, ACEA expects various factors to offset such fuel economy improvements. For example, ACEA says that while the fuel economy of the EU fleet is shown by statistics to have improved by 8% between 1983 and 1997, it had actually improved by 28% but was offset by factors (e.g. safety, emissions requirements) that added weight to the vehicle (although the figures have not been verified by independent sources). In calculating the 1997 offer of 167gCO 2 /km, ACEA declared that 50% of the possible improvement would be offset by various factors and that this was taken into account in the calculation. As the Commitment is not binding, offsets claimed by the automobile industry and accepted by the Commission, will be taken into account by the monitoring procedure. Therefore the 25% improvement could be eaten away. Thus a major problem seems to lie with the offsets to fuel efficiency gains. Of the various offsets, consumer demands is of major importance but can be tackled by reversing the current trend of upsizing which is supported by consumers choosing to buy larger and more powerful vehicles. As regards vehicle specifications associated with say noise, safety, toxic emissions or waste disposal, weight can indeed be gained. Notwithstanding, the internal combustion engine should not be excused from attaining fuel economy targets due to other vehicle-related legislated requirements, because this discriminates against other technologies that may be able to address various environmental problems in addition to improved fuel economy. Table 4 The operational and capital costs of alternative technologies compared to existing technologies Source: CEC 1998b, COM(1998)204 Operational and Maintenance Costs (estimates)# Relative Capital Costs of Alternative Vehicle Technologies Transport technology 1995 (ECU 90/y) 2010 (ECU 90/y) 1990 (ECU 90/y) 1995 (ECU 90/y) 2010 (ECU 90/y) Gasoline car 456 456 100 100 100 Diesel car 470 470 114 114 114 LPG car 481 481 105 105 105 Methanol car 473 473 104 104 104 Ethanol car 473 473 104 104 104 CNG car 676 676 113 113 113 Electric car 700 420 - - 181 Fuel cell 980 420* - 195 117 hydrogen car Hydrogen 676 676 - - 128 combustion car Fuel cell - 748 254* hydrogen COM(1998)204 Source: Potential from changes in Fuel and New vehicle technologies for cars (CES KU Leuven, IFP, NTUA) *Significant decrease expected by 2030. # It is not known on what basis the costs of hydrocarbons in 2010 have been estimated. Oil prices of the next century are very uncertain, particularly when demand will exceed production and the market becomes controlled by the Far East. 29

The Analysis The last few decades have been abundant in prototypes with alternative powertrains, materials and fuels which do offer significant advantages in the way of for example, fuel economy, toxic emissions and reduced noise. Such models are now in commercial production, but in extremely small volume and find it difficult to compete with the wellestablished internal combustion engine mass market (e.g electric vehicles of General Motors, Peugeot, Fiat; Toyota hybrids; BMW natural gas vehicles). In addition, the problems of higher cost and lack of supporting infrastructure still delay the infiltration of such technologies. Although maintenance costs and capital costs have already or are predicted to converge with the traditional gasoline and diesel model (see Table 4), internal combustion technology has been incrementally improved to such an extent that the advantages offered by alternative technologies are increasingly diminishing. Thus it seems that a technological paradigm shift can not be justified. This latter point is attested by authors Wells and Newenhius who argue that a move away from the all-steel body and internal combustion engine will still be necessary (Wells & Newenhius 1997, p208): (as regards environmental concerns) Despite the continual improvement in steel-making technology, design and manufacturing processes, and despite the significant cost per unit advantage for steel in high volumes, all-steel cars are too heavy and the production technology too capital-intensive and inflexible. It is clear that the target of 140gCO 2 /km is not sufficiently ambitious to make use of the available fuel economy potential that technologies already offer today. The terms of the agreement are also biased towards the needs of internal combustion engine technology. For example, while it is assumed that there will be unhindered acceptance of alternative technologies, the terms of the Commitment are moreover designed to assist the maintenance of petrol/diesel internal combustion engine technology: higher fuel quality, unhampered diffusion of technology (aimed at increasing diesel share) and achievement of 90% of the fuel economy target solely by direct injection diesel and gasoline technology. Another study of the European automobile voluntary agreements, has explored the probability of the industry fulfilling their commitments (Kageson 2000). By analysis of current market trends, manufacturer intentions and the opportunities for market entry of more fuel efficient technologies, Kageson estimates that the automobile industry will fall short of its 140gCO 2 /km target by 5gCO 2 /km i.e. 10.7% (see Table 5, page 31). Interestingly, Kageson estimates that only 18.4% of the 2008(9) target will be met through a shift to diesel and more common rail diesel technologies, and just 12.8% will be achieved through a shift to direct injection petrol engines. While Kageson acknowledges the potential of certain factors to offset fuel economy gains (especially due to consumer demands), these factors are not taken into account in this estimation. Kageson concludes that without additional financial incentives and disincentives, manufacturers will only make use of a minor part of the available potential for fuel economy improvement and that taxes on fuel must be raised in line with the fleet s fuel economy improvement in order to curb the rebound effect 14. If the assumptions underlying Kageson s study hold true, then the European automobile industry will not be far from honouring its commitments as the statement taken from ACEA s Commitment, below, is interpreted by the Commission and industry to mean that some 10% of the final target can be achieved through non-technical means such as downsizing (see Point 2 of the ACEA Commitment, Figure 19, page 82; CEC 1998h): 14 Several studies have observed a rebound effect (i.e. people driving more due to the lower cost of driving) due to improved fuel economy after the oil shocks of the 1970s, particularly as a result of CAFÉ in the US (Greene 1997). However, the size of such an effect is uncertain (Michaelis 1996), and depends on the policies used. 30

The Analysis This target will mainly be achieved by technological developments affecting different car characteristics and market changes linked to these developments. In particular, ACEA will aim at a high share - to the point of 90% - of new cars sold being equipped with CO 2 efficient direct injection gasoline and diesel technologies. Table 5 Contributions from alternative powertrains and new fuels to attaining the 2008-9 target Comparison with a 1997 average emission of 183gCO 2 /km. Source: Kageson (2000) Measure Reduction in average CO 2 emissions of all new cars Contribution in per cent 1) Shift to diesel + more common rail diesel engines 7.9 18.4 2) Shift to direct injection petrol engines 5.5 12.8 3) Shift to electric hybrids 4.5 10.5 4) Shift to alternative fuels 1.3 3.0 Total average 19.2 44.7 5) To be achieved by other measures: High-powered ignition systems; improved fuel injectors; computer controlled engine management; improved compression at low engine loads; variable valve timing; continuously-variable transmission to improve gearing efficiency; reduced mechanical friction; reduced air drag and rolling resistance. 23.8 55.2 Shortfall estimated of 140gCO 2 /km target* 5 11.6 Assumptions numbered according to the measures as numbered in the table: 1) For ACEA members the diesel share is currently 27% of new sales. Assuming that the diesel engine can realistically increase its total share to 35% by 2008, the contribution to the car industry s commitment would be around 3.1gCO 2 /km if the reduction is evenly split on all new cars (all else equal). This is equivalent to 7.2 % of the 43g needed by the Commitment. Considering that a further shift to common rail diesel engines is likely to take place, it is estimated this will improve the average efficiency of all new diesel cars by 8% or minus 4.8g if the reduction is spilt on all new cars. 2) While it is not clear how widespread the availability of petrol with less than 30ppm suplhur will become throughout the EU over time, it is estimated that 20% of all cars (corresponding to 30% of all new petrol cars) produced for the European market in 2008 are equipped with these engines. 3) Assumes the entire industry will manage to sell 30 000 hybrids in Europe in 2001 (most of which will be Toyotas) and will be able to increase the combined output of hybrid and fuel cell cars by 50% per year up to 2008, the total production that year will be just above 1 million. This is equal to close to 5% of an estimated 20 million new registrations in 2008 (up close to 4% per year compared to an average 5% increase 1994-1999). Assumes the average fuel consumption of these cars is 55% of today s conventional petrol-fuelled cars. Contribution of battery electric vehicles is neglected due to no expected battery-efficiency revolution. 4) Assumes the total contribution of biofuels, including low blend mixes with petrol and diesel, will amount to 0.7% of Europe s consumption of car fuels in 2008 5) Assumes traditional injection petrol car will still account for 45% of market in 2008. Assumes that general fuel efficiency improvements and lighter materials could improve fuel economy by 10% in 2008 compared to 1997. 31

The Analysis 3.1.2 Industry s commitments in CO 2 terms With respect to the automobile industry s commitments, there is little information available in the way of CO 2 modelling for passenger cars. However, one study carried out for the Netherlands Presidency in 1997 (Phylipsen et al 1997, see Table 6), estimated that attaining the Council s target of 120gCO 2 /km by 2005 would not even stabilise CO 2 emissions from passenger cars at 1990 levels. Table 6 CO 2 reduction gain estimations for the EU passenger car sector (table) The Expert Group s work on EU Common and Coordinated Policies and Measures edited by GJM Phylipsen, K Blok, H Merkus for the Netherlands s Presidency 1997 SFC=specific fuel consumption; EU target=120gco 2 /km The scenarios 2 and 3 are considered to resemble most the intentions of the Council conclusions of June 1996. The study is subject to a number of assumptions: 1. In the period up to 1986 specific fuel consumption of new cars dropped by 2% a year. Since then this specific fuel consumption has more or less stabilised 15 (ECMT 1996; Schipper 1995). 2. Passenger car transportation grows by 2% per annum 16. 3. The average lifetime of cars is 12 years. The actual savings will depend on the policy measures put in place. These calculations do not include the rebound effect 17, which can be avoided through use tax e.g. fuel duty. There are also notable differences between countries (Blok and Phylipsen 1996): car ownership rates (between 200-500 per 1000 inhabitants) average car size (between 900 and 1300kg per vehicle) average car mileage (ranging at least between 13,000 and 18,000km/yr) turnover of cars present fuel economy levels (between 6.5 and 8.21 per 100km) Scenario Reference Scenario 1 Scenario 2 Scenario 3 Scenario 4 Description The SFC remains constant at the present level upto 2010 The SFC decreases linearly from 2002 to the EU target in 2010. The SFC decreases linearly from 1998 to the EU target in 2010 The SFC decreases linearly from 1998 to the EU target in 2005 and remains constant thereafter The SFC decreases linearly from 1998 to the EU target in 2005 and decreases further onwards. Mtonnes CO 2 % above 1990 2005 2010 1990 CO 2 levels in 2010 380 486 536 41 380 478 479 26 380 459 453 19.2 380 440 419 10.2 380 440 401 5.5 15 This may not be the case for all Member States. 16 This leads to a growth of CO 2 emissions in the reference scenario of 2% a year (no energy efficiency improvement). This is a somewhat higher growth rate of CO 2 emissions in passenger transport than the growth rate of total transport emissions. No separate figures for passenger transport are available. Note that this growth rate can be influenced by policy measures. 17 Several studies have observed a rebound effect (i.e. people driving more due to the lower cost of driving) due to improved fuel economy after the oil shocks of the 1970s, particularly as a result of the fuel economy regulation system, CAFÉ, used in the US (Greene 1997), but the size of such an effect is uncertain (Michaelis 1996), and depends on the policies used. 32

The Analysis Figure 8 CO 2 reduction gain estimations for the EU passenger car sector (graph) 550 500 450 400 350 Reference Scenario 1 Scenario 2 Scenario 3 Scenario 4 300 1990 2005 2010 In reality, the C0 2 emissions reduction scenario for the passenger car sector is unlikely to follow that of Scenario 3 or 4 of Figure 8 above, as ACEA has stated that the improvement is likely to be non-linear - slow at first with an acceleration later (Interview 4; Point 3 in the Technical Annex of the Commitment, Figure 19, page 82). Thus a scenario that is somewhere in-between Scenarios 1 and 2 is most likely. At the same time, it should be noted that while the automobile industry may attain its 140gCO 2 /km target by 2008 and the 120gCO 2 /km target may be achieved by 2010 due to the efforts of either the automobile industry or through the Member States fiscal policies (or a combination of both), the car parc takes a decade or so to renew itself (varies from country to country). Thus the impact of the industry s commitments will continue to 2020. Even the most ambitious Scenario 4, which assumes a linear reduction to 120gCO 2 /km by 2005 and a further reduction beyond 120gCO 2 /km to 2010 (if linear the end target would be 72gCO 2 /km), can not reduce the CO 2 emissions from the EU passenger car fleet to below 1990 levels by 2010. The base case modelled for the Commission s Auto-Oil II programme (CEC 2000), estimates that emissions from road transport (50% of which are attributable to passenger cars) would stabilise from 2005 at levels 10-15% above 1995 levels i.e. 22-27.5% relative to 1990 levels This is based on the assumptions of an increase in passenger car demand of some 26% and of freight transport by some 29%. The automobile industry s commitments have been included in the simulation and while the freight/commercial vehicle sector is not legislated and has not agreed to commitments like the European passenger car manufacturers, the Commission has optimistically assumed a significant improvement in this sector s fuel economy. The results of the Auto-Oil II simulations are similar to Phylipsen et al s most likely modelled outcome (i.e. a combination of Scenarios 1 and 2 will result in CO 2 levels 19.2-26% above 1990 levels). However, both simulations fail to consider the unrepresentative nature of the test-cycle which over-estimates a vehicle s fuel economy. Not only does a vehicle driven on the road carry considerably more weight compared to the test model (e.g. due to the extra weight of accessories such as air conditioning, electric motors for windows, radio, luggage, passengers) but the test-cycle s simulated driving test is also unrepresentative of real-life driving. This issue is discussed in the next chapter. 33

The Analysis Figure 9 Auto Oil II emissions forecast 120 100 80 60 1980 1990 2000 2010 2020 2030 Year Source: CEC 2000 According to the Commission (COM(95)689, CEC 1995; Table 3, page 28), a 40% improvement would reduce emissions by 6.9% compared with 1990 levels, which is more in line with the EU s commitments to the UNFCC Kyoto Protocol. This estimation takes into account growth in the vehicle fleet and mileage, as without such considerations the 40% fuel economy improvement would reduce CO 2 emissions by 30.1% in 2010 compared to 1990 levels. The EU is committed to its Kyoto Protocol commitment of an 8% reduction of greenhouse gas emissions below 1990 levels by 2012 at the latest. The above evidence shows that the automobile industry s commitments will not bring CO 2 emissions for passenger cars even near to 1990 levels, despite the technological potential to do so. Thus Member States will have to achieve savings through transport polices or the carbon deficit will have to be offset by other means, such as emissions trading or reductions from other sectors. However the IPCC (1996) has warned that a reduction of some 70% in carbon emissions world-wide will be required (probably more for industrialised nations), to keep the accumulation of global warming gases within a ceiling of 450ppm (IPPC 1996). It is therefore likely that much more will be required from the passenger car sector in the future. 3.1.3 The CO2 test cycle: How a vehicle s fuel economy is over-estimated Directive 93/116/EC 18 (Council of the European Union 1993), replacing Directive 80/1268/EEC, defines the CO 2 test procedure and has been applicable to all new engines introduced since 1 st January 1996 and to new cars sold from 1 st January 1997. The new testcycle, as it is called, led to an average artificial increase of 9% in the CO 2 figures measured according to the old test-cycle (see Figure 10 below). Throughout the negotiations there has been much dispute, for the purposes of bargaining, over whether a figure was old test-cycle or new test-cycle (p.c. Interviews 1, 2 and 3). For example, the 167gCO 2 /km proposed by ACEA in 1997 was originally proposed as 155gCO 2 /km according to the old test-cycle, and this was later translated to 167gCO 2 /km on the new test-cycle (see Figure 10 below). The target of 120gCO 2 /km, adopted by the Council (Council of the European Union 1996a), originated from the French and German ministers 18 Previously Directive 80/1268/EEC required fuel consumption to be measured under laboratory conditions at a steady 90kmph as on country roads and a steady 120km/h as on highways, and on an urban cycle. The engine was warmed up before these tests. Directive 93/116/EU requires a cold start. Acceleration and breaking procedures are accompanied by gearshifts intended to reflect normal driving styles. 34

The Analysis Klaus Toepfer and Segolene Royale during a conference in 1992 (Interview 1). At this time the Directive for the new test-cycle had not even been published. Figure 10 The new test cycle of Directive 93/116/EEC in comparison with Directive 80/1268. Source: ECMT 1997 Note with Figure: There is no broad empirical basis available yet which would allow derivation of a precise correlation between the fuel consumption based on the old and new test-cycles. It can be assumed that each car reacts in different ways to the change in test-cycle. The new test-cycle moves the figures for fuel consumption closer to the actual fuel consumption of vehicles in use (ECMT 1997). As the Council and the Commission stuck to the 120gCO 2 /km target through the transition from the old to the new test-cycle; it seems the institutions successfully moved their original goal posts, because 120gCO 2 /km is more ambitious on the new test-cycle. ACEA pointed this out at the CO 2 passenger car workshop in Strasbourg in February of 1998, but the Council continued to stick to 120gCO 2 /km (CEC 1998i). However, the three pillar strategy COM(95)689 made clear that part of the Council s 120gCO 2 /km target would not be achieved by industry but through the fiscal framework and consumer information policies. The new test-cycle according to Directive 93/116/EEC is more representative than the old test-cycle according to Directive 80/1268/EEC as it includes cold starts. Nevertheless, the new test-cycle is still criticised for being unrepresentative of real driving conditions. The Green Car Guide (July 1999 p4) of What Car? heavily criticises the EU test for unrepresentative loading and acceleration: In particular, acceleration is ludicrously gentle. In the urban section the maximum acceleration required is 0-50kph (0-31 mph) in a leisurely 26 seconds with two gear changes; in the extra urban section the maximum acceleration is 0-70kph (0-44mph) in a comatose 41 seconds, this time with three shifts. 35

The Analysis The test also gives inaccurate results for the measurement of CO 2 emissions from direct injection petrol engines, which the automobile industry intends to largely rely upon for the achievement of its commitments. For example, based on type approval values, the Mitsibushi s Carisma 1.8 LX GDI emits 18% less CO 2 than the Carisma 1.8 GLX. A comparative test of the two performed by Swedish MTC according to the European test-cycle (NEDC) resulted in a 10% fuel reduction, and when driven according to the American FTP-75 test-cycle, just an 8% reduction (Ahlvik (1998); cited in Kageson 2000). Meanwhile, a Dutch test which is designed to simulate real-world conditions, gave just a meagre 2% reduction in CO 2 emissions (Huigen, 1998 cited in Van den Brink and Van Wee, 1999; cited in Kageson 2000) 19. The unrepresentative nature of the test-cycle is especially important as regards CO 2 because there is a trade off between CO 2 and NOx. While manufacturers wish to design engines to give a favourable CO 2 result, NOx emissions will still need to satisfy regulations. However, the balance between the two pollutants can vary according to operating conditions. This is also important with respect to the fact that manufacturers are capable of designing their cars to beat the test-cycle. Not only can cars be designed to give optimal results according to the known conditions of the test-cycle but modern electronic equipment can adapt the engine to recognise when the car is being driven according to a specific test cycle such that the combustion and operating conditions can be adjusted accordingly 20 (Kageson 1998). The agreement covers passenger cars with up to eight seats and most four-wheel drive vehicles are also to be included. Yet the tested vehicles are not loaded with any baggage, passengers or trailers. Further, additional features or accessories are not included in the test procedure such as air-conditioning and electronic gadgets. For example, air-conditioning equipment tends to increase fuel consumption by 0.5-1.0 litre/100km when operating (ECMT 1997). The French energy and environment agency, Ademe estimates that air-conditioning increases fuel consumption by some 15-20% and that in the year 2000, over 20% of new cars will be sold in the EU with air-conditioning. Due to the addition of such accessories to the car after the test-cycle, there can be a significant difference in weight between the vehicle tested and the vehicle that is driven on the road such that the latter can be 10 to 20% heavier than the former, which will negatively impact fuel economy. As discussed in the previous chapter, the unrepresentative test-cycle can introduce significant errors into CO 2 forecast simulations. In addition, the official fuel economy rating of a car, which is sold with many accessories, may mislead consumers that are seeking to purchase a fuel efficient car. In an attempt to reconcile the problem of inadequate information to the public, Parliament and Council agreed during the legislative process of Directive 1999/94/EC 21 to call for the provision of information to the public on the effects of additional options such as air conditioning on fuel consumption. Despite Parliament s efforts during the readings on the proposal for a Decision on monitoring CO 2 from passenger cars (CEC 1998c), the Member States will not be required to transmit data concerning accessories such as air conditioning. The Commission persuaded Parliament that calculating the impacts of air-conditioning on fuel consumption is particularly complex as it is difficult to estimate how much the device is used. In addition, it is argued that there are 19 Present generation direct injection technology uses a NOx storage catalyst in combination with a three way catalytic converter which is needed to bring the engine s high emissions of NOx into line with the EU emission limit value for petrol cars. The NOx storage catalyst stores NOx during lean burn engine operation, when fuel consumption is relatively low. The reduction can only take place in a three-way catalyst under stoichiometric conditions, therefore Mitsibushi has made the engine operate stoichiometrically above 30% of maximum engine torque. Under such conditions, the direct injection engine uses as much fuel as an indirect injection engine (Kageson 2000). 20 Six truck manufacturers, including two European firms (Volvo and Renault), recently used software to defeat the US EPA s pollution control. As a result, emissions of NOx from highway driving increased by 300% (Kageson 1998). 21 Directive 1999/94/EC The availability of information on fuel economy of new passenger cars 36

The Analysis many other factors which can impact fuel economy such as driver behaviour, car maintenance, driving conditions, opened windows and (pre-) heating systems etc.. 3.2 Terms of the agreement Following the adoption of the 140gCO 2/ km target in March 1998, the drafting process for the Commitment took some 4 months, during which time the Commission amended the document drafted by ACEA (Interview 2). The final agreement was officially presented to the Commission by ACEA on 28 th July 1998 and the Commission responded the following day with the Communication COM(1998)495 (CEC 1998d). The Commission officially welcomed ACEA s agreement through a Recommendation in February the following year (1999/125/EC CEC 1999b). On the establishment of the agreement with ACEA, the Commission then began negotiations with the Japanese (JAMA) and Korean (KAMA) manufacturers with the intention of obtaining similar commitments (see chapter 3.2.5, page 45). In drafting the Commitment text, the first bone of contention centred upon ACEA s demand for certain conditions: no negative measures against diesel fuelled cars; full availability of improved fuels by 2005 especially with low sulphur content; importers to make equivalent commitments and major manufacturing countries to implement similar policies (Interview 2). These conditions were similar to those demanded by ACEA in June 1997. In an effort to keep the right to apply its own initiative in developing policies and to allow more flexibility, the Commission redefined conditions as assumptions (COM(1998)495, CEC 1998d). Confusion then developed as to the differentiation between condition and assumption (Interview 3). It appears that assumptions may provide flexibility over conditions, as alternatives and solutions to problems can be sought before resorting to a re-negotiation of the agreement (Interview 1). Nevertheless, it is clear that ACEA s commitments are subject to certain terms and essentially the objectives will have to be renegotiated if negative impacts are proven to nullify the fuel economy improvements. This is of particular concern as some of the terms of the agreement are vague catch all expressions which can provide the basis for a review of the situation, and necessary adjustments in good faith. Examples of such catch-all expressions describing situations that the Commission will take into account include: any measures which might hamper the diffusion process of either of the CO 2 technologies (Assumption C of the ACEA Commitment; Figure 19, page 82) and...if the impacts of this Commitment, particularly its employment situation and its global competitive environment, are detrimental,... (notes under Monitoring section of the ACEA Commitment; see Figure 19, page 82). One of the conditions proposed by ACEA for both the 1997 and 1998 offers concerned funding for research. ACEA had also hoped that extra R&D funding might form the fourth pillar of the COM(1995)689 strategy (CEC 1995), but Point 3 of COM(1995)689 clearly states that efforts would be aimed at capitalising from the technological potential that already exists. The European automobile industry s EUCAR 22 R&D programme would therefore continue as before, with proposals having to compete for funding from the 5 th Framework R&D budget (Interviews 2 and 3). At the same time, while investment in automobile research is necessary, there is much evidence that the technology to achieve the fleet average of 140gCO 2 /km already exists (see chapter 3.1.1, page 27). Much could also be gained from efforts to curb the consumer trends of an increasing proportion of sales of faster, larger and more powerful cars, which are generally heavier and therefore less fuel efficient. 22 EUCAR is the European Council for Automotive Research and Development established to act as an agent for expanding collaborative effort on basic research 37

The Analysis 3.2.1 Fuel quality When Pischetsrieder presented the target, of 140gCO 2 /km by 2008, to the Commission on March 10 th of 1998, it was conditional on the EU-wide availability of higher quality fuel than that which was being negotiated for Auto-Oil 1 23 legislation at the time. Figure 11 Higher Fuel Quality Assumptions ASSUMPTION A) Availability of enabling fuels Given the outstanding importance of improved fuels for CO 2 reductions ACEA assumes the full market availability of fuels with a sufficient quality to enable the application of technologies needed for the industry to achieve its CO 2 commitments during the lifetime of this Commitment (s. Technical Annex, Point 2 Fuel Specifications). Technical Annex Point 2) Fuel Specifications Characteristics of the fuels are key factors in car CO 2 emission reductions: A) to achieve further emission reduction together with lowered CO 2 emissions the fuel efficient lean burn technology will be combined with special exhaust gas after-treatment devices capable to reduce NOx under lean burn conditions. But those systems are only working with fuels meeting specific requirements, in particular a low sulphur content; B) low sulphur fuels ease the NOx/CO 2 trade-off in favour of CO 2 emission reductions; C) low aromatics in gasoline and a high cetane number in diesel lead to CO 2 emission reduction too. ACEA acknowledges the outcome of the conciliation procedure between the Council and the European Parliament on 26.6.1998 and upholds its 140gCO 2 /km commitment by 2008. However, ACEA is expecting that fuels of the following better quality might be available in the market due to technical reasons, commercial competition as well as possible national policies: I) Some gasoline (e.g. Super-Plus, 98 octane as agreed in Germany) and some diesel plus with a maximum sulphur content of 30ppm are provided in 2000 on the whole EU market in a sufficient volume and geographical cover. II) In 2005 full availability of fuels on the whole EU market which satisfy the following: gasoline with a maximum sulphur content of 30ppm and of a maximum aromatic content of 30%; diesel with a maximum sulphur content of 30ppm and a cetane number of minimum 58. Any problems which might arise with respect to fuel quality will be considered in the monitoring procedure. Source: ACEA Commitment (see Figure 19, page 82) Following the automobile industry s calls for low sulphur fuel in 1998 prior to the adoption of new legislation of fuel quality requirements (Auto-Oil ; 98/70/EC), DG17 attacked the autoindustry, backing the arguments of the European oil industry association (EUROPIA) and called for evidence to justify the need for higher quality fuel (Interview 13). The oil industry argued that the overall CO 2 balance would be worse if refineries would be forced to remove even more sulphur (Interview 13). EUROPIA also presented the argument that some technologies are more sulphur tolerant than others. ACEAs justification for fuel with a low sulphur content was that the fuel would enable the introduction of advanced technologies which would be fuel efficient as well as capable of meeting EU vehicle emission limits. ACEA claimed that sulphur poisons exhaust gas cleaning technology such as NOx storage catalysts and particulate traps or filters. Particulate matter has until now been a problem only for diesel cars. However, petrol engines may also be targeted for regulation in the future as the operating conditions of lean burn petrol engines, which manufacturers hope to use to reduce CO 2 emissions, give rise to elevated levels of 23 Auto-Oil 1 legislation sets fuel quality standards and automotive emission limits for 2000 and 2010 38

The Analysis ultrafine particulate matter. With new evidence strengthening the link between particulate matter and increased morbidity and mortality 24, the problem is set to continue such that particulate matter may need to be further regulated, especially ultrafine particles. Due to lack of convincing evidence, the Auto-Oil 1 standards were not further tightened during the June 1998 conciliation procedure between Parliament and Council, but Parliament did manage to use its powers under the co-decision procedure to bring forward the date of compliance with the Directive s provisions by 5 years to 2005. In this particular case, the automobile industry was glad to see Parliament exercise its powers through co-decision procedure, but the event was a clear warning to the automobile industry that if a voluntary agreement was not finalised, legislation would be the only other option and Parliament s position on passenger car fuel economy was known to be strongly supportive of stringent regulation. The outcome of Auto-Oil I, which delayed the voluntary fuel economy agreement by more than one month (Interview 1), resulted in the inclusion of various assumptions in the Commitment (see Figure 11 above). DG17 insists that these assumptions will not be borne out: (Points I and II of Point 2 of ACEA s Commitment) I. Some diesel plus with a maximum sulphur content of 30ppm are provided in 2000 on the whole EU market in a sufficient volume and geographical cover II. In 2005, the full availability on the EU market of gasoline with a maximum sulphur content of 30ppm and of a maximum aromatic content of 30%, and of diesel with a maximum sulphur content of 30ppm and a cetane number of minimum 58. The Auto-Oil 1 legislation (98/70/EC) requires EU-wide availability of fuel with a sulphur content of 150ppm for petrol and 350ppm for diesel by 2000, and 50ppm for both petrol and diesel by 2005. Due to the differences between the Auto-Oil 1 legislation and ACEA s assumptions, DG 17 accuses ACEA of trying to manipulate the outcome of the Auto-Oil 1 negotiations and of using the issue of fuel quality as a tactic to protect itself from the Commitment (Interviews 13 and 15). While, DG17 explicitly accuses DG3 and DG11 of being subject to the influence of ACEA (Interviews 13 and 15), DG3 and DG11 believe DG17 was heavily influenced by EUROPIA (Interview 2). The Member States are currently at different stages as regards sulphur content in fuel. Several Scandinavian countries have already legislated for low sulphur fuels (e.g. Sweden 10ppm sulphur), while countries such as Spain, Portugal and Greece state that they will struggle to reach 50ppm by 2005 as required by Auto-Oil 1. The latter countries claim that their national refining industries suffer lack of investment and are economically vulnerable, especially as the industry suffers from considerable over-capacity (Simonian, Financial Times 20.8.98; Boulton 17.2.98; Financial Times 4.6.99). Meanwhile, Germany has recently proposed a target of 10ppm and called on the Commission to legislate for 10ppm sulphur by 2007. The agreement s monitoring process will have to take into account the impact of the variation in fuel quality throughout the EU relative to ACEA s assumptions, when assessing the automobile industry s achievement of their targets in 2003 and 2008. 24 Various studies show that particulate matter, particularly finer particulate matter (<2.5µm), is associated with increased morbidity and mortality (Pope et al 1995; Dockery et al 1993; Ghio and Samet 1999; Wilson and Spengler 1996). Diesel soot has been classified as a probable carcinogen by the International Agency for Research on Cancer and a toxic substance by the California Air Resources Board based on a wide variety of evidence e.g. railroad workers exposed to diesel exhaust have a 40% increase in lung cancer risk (Speizer, Franck 1994). 39

The Analysis The Commission is currently considering the manufacturers continued calls for near zero sulphur fuel (Europe Environment No. 567 4.5.00), which is defined as fuel with a maximum sulphur content of 10ppm. The Commission launched a consultation in May 2000, to investigate whether (Europe Environment No 569 1.6.00) sulphur levels should be lowered even further after 2005. The Commission received some 26 submissions of evidence from various organisations and governments, which have been collated by consultants and independently peer-reviewed (Marsh et al 2000). The Commission shall consider the evidence before taking any initiative to draft a Directive. ACEA s evidence emphasises the need for low sulphur fuels to enable the introduction of advanced technologies which would be fuel efficient as well as capable of meeting EU vehicle emission limits (ACEA 2000b). However, the summary of the oil industry s submission by CONCAWE (CONCAWE like EUROPIA represents the European oil industry) states that compared to 50ppm sulphur fuels, ultra-low sulphur fuels do not appear to enable new technologies but rather enhance some of those that would already be able to work at 50ppm. CONCAWE also claims that ultra-low sulphur fuels would not bring any direct benefits in terms of air quality and that certain after-treatment systems release increased quantities of other pollutants such as ammonia and the greenhouse gas nitrous oxide. But the German government rebuffs this claim stating that: Sulphur free fuels give greater scope for restricting other emissions such as ammonia, nitrous oxide and methane. However, the independent peer-review concludes that low sulphur fuel will result in emissions reduction and air quality benefits but that these benefits are difficult to quantify (Marsh et al 2000). The issue of the CO 2 emissions balance between the CO 2 gains obtained through vehicle operation and the CO 2 losses due to extra energy for refining the oil, has also been investigated. The UK government concluded that the overall impact on CO 2 emissions by 2010 for 10ppm sulphur fuel could range from being broadly neutral to a small net increase, depending on the assumptions made on future vehicle fuel economy trends and the projected refinery impact 25. The Dutch government draws similar conclusions to that of the UK (The Netherlands 2000). The UK s findings are also generally confirmed by those of CONCAWE which state that, The benefits in terms of CO 2 emissions (from cars) would not necessarily surpass the CO 2 debit due to extra refinery processing (CONCAWE 2000). However, the conclusions of the German government are to the contrary: The German Federal Environment Agency estimates that CO 2 emissions from refineries are visibly lower than the CO 2 savings for cars. The conclusions of the independent peer-review state that in the longer term there will be an annual net reduction in greenhouse gas emissions (including CO 2, nitrous oxide and methane) with the use of near zero sulphur fuel as opposed to 50ppm sulphur fuel but also caution that the implications with respect to the CO 2 balance need to be better evaluated (Marsh et al 2000). Should a Directive for low sulphur fuels be drafted, the proposal would probably be available by Spring 2001 (Interview 18). However, it may take some two years to pass the proposal through Parliament and Council. The attainment date for the introduction of low sulphur fuel will depend on many factors but 2008-2010 may be possible and the phase in of the higher quality fuel may need to be linked to the phase-in of the advanced technologies (Interview 18). With respect to phasing-in near zero sulphur fuel, the independent peer-review recommends, amongst other things, that account be taken of the short-term emissions benefits, the needs of the low sulphur fuel with respect to achieving Euro 4 (light duty vehicles) and Euro 5 (heavy duty vehicles) emissions standards and the 140gCO2/km fuel economy target for passenger cars. 25 The calculations were based on the assumption some 8% improvement in fuel efficiency, as measured on the standard EU test cycle, may be expected from gasoline direct injection vehicles using fuel of a 50ppm sulphur content and that an extra 2 to 4% improvement would be gained from using fuel of a zero sulphur content. 40

The Analysis Thus there is a chance, but no guarantee, that fuel of 30ppm or even 10ppm sulphur content will be introduced by 2008, when the automobile industry is to achieve its target of 140gCO 2 /km. However, the industry requests full EU-wide availability of fuel with a 30ppm sulphur content by 2005, which can only be achieved if all Member States take their own initiative to ensure fuel of such quality is available or are forced to by market pressures. If fuel with a maximum sulphur content of 30ppm is not fully available EU-wide by 2005, ACEA and the Commission will have to reach agreement as to the impact on the auto-industry s commitments. But if the sulphur content of fuel is to be further reduced below 50ppm by 2008, it should be noted that the automobile industry may be awarded CO 2 emissions reductions for emissions which are transferred to the oil refineries that emit more CO 2 emissions in order to produce the fuel of improved quality for the automobile industry. 3.2.2 Fiscal measures ACEA believes it can attain 140gCO 2 /km without any help from Member State fiscal policies:...this Commitment provides complete and sufficient substitute for all new regulatory measures to limit fuel consumption or CO 2 emissions, and for any additional fiscal measures in pursuit of the CO 2 objectives of this Commitment. (Point 2 of Introduction and Principles of the Commitment (Figure 19, page 82)). Meanwhile the Commission is in the process of developing the fiscal framework, which ACEA assumed would be introduced after the review in 2003 and not before, when it would be known whether or not ACEA could achieve the 120gCO 2 /km target by 2012 through technology only (Interview 12). Automobile manufacturers are largely not supportive of the use of fiscal measures to influence consumers. ACEA argues that fiscal measures, which encourage downsizing, will negatively impact the automobile manufacturers of larger cars who are also making significant investments in fuel economy improvements. In addition, the organisation argues that the diversity of the product range will be reduced, thus harming manufacturers competivity. Notwithstanding, the auto-industry s main concern with regard to downsizing is that the profit margins are greater for the larger and more luxurious cars which are more powerful, faster and therefore more fuel consumptive than smaller cars. As there is an increasing trend for consumers to purchase such cars (FT Automotive Quarterly Review 1999) downsizing could threaten profits. The benefits of downsizing in addition to improved fuel economy also includes safety. It is the relative difference in the size of cars, which largely dictates the consequences of a crash. Logically, the downsizing of the entire car fleet would reduce the fleet disparities in size and weight and would favour safety (p.c. Wells, Newenhius and Interview 16). Thus, with regard to passenger car fuel consumption, regulators, manufacturers and consumers are all pulling in different directions. Purchase taxes are also unfavourable with the automobile industry as they can potentially slow the parc turnover, as in Denmark. The automobile industry also takes a very strong line against fuel duty if it is unfavourable to diesel. Diesel has generally benefited from lower tax rates than petrol in the EU, but this trend is beginning to reverse. With the automobile industry s expectation of moving towards a 40% diesel market share 26 (Interview 3), anti-diesel initiatives such as raised taxes on diesel (e.g. as recently the case in France and the UK (Owen 1998)) could prove problematic for the automobile industry. The 26 Which is near to the maximum share possible due to the natural cap provided by the average composition of oil (Interview 3). 41

The Analysis high diesel share anticipated by ACEA, concerned the Commission especially as diesel cars are currently responsible for most particulate emissions and are noisier than petrol cars. Although technology is improving, old diesel cars are still a major problem. In addition, diesel contains more carbon per litre of fuel than petrol. To target CO 2 emissions equitably, fuel should be taxed according to its carbon content, which would mean taxing diesel 13% above petrol (Kageson 2000). In an attempt to protect the industry from anti-diesel fiscal measures, Pischetsrieder initially called for no negative measures against diesel-fuelled cars (Interview 2). Due to pressure from ACEA, the Commission agreed to include in the text a vague reference to unhampered diffusion of technologies in order to cater for ACEA s concerns regarding fiscal measures: This Commitment is based on the assumption of...an unhampered diffusion of car CO 2 efficient technologies into the market via competition amongst ACEA members and other market participants which is expected to result in market mix changes. (Assumption C of ACEA Commitment, refer to Figure 19, page 82). There is sure to be conflict between the Commission and the automobile industry as to what constitutes unhampered diffusion or hampered diffusion. While allowance is made in the Commitment for measures that might hamper diffusion, the Commission again bowed to ACEA pressure and has expressed its willingness to review the Commitment under certain circumstances should fiscal measures interfere with the particular fuel efficient technologies which the manufacturers are trying to promote: It is the Commission s and ACEA s common understanding that this does not restrict the Community s and the Member States freedom to use fiscal or regulatory measures. However, such measures would be considered in the monitoring of the Agreement and could be grounds for its review under certain circumstances. Paragraph 3 of the Communication COM(1998)495 (CEC 1998d). At the same time, some ACEA members wish to contribute to the Commitment through downsizing, and this is acknowledged by the Commission. ACEA argues that an element of downsizing should be allowed for within the scope of the agreement so that certain ACEA members can contribute to achieving the 140gCO 2 /km target by selling a higher proportion of smaller fuel efficient cars than they do at present. (Staff working paper SEC(1998) 1047 (CEC 1998h)). Thus the 10% gain that ACEA expects from non-technological means (Point 2 of the Commitment, Figure 19, page 82), will have to be separated from the Commission s efforts to encourage downsizing through its fiscal framework and consumer information scheme aimed at achieving the extra 20gCO 2 /km called for by Council (Council of the European Union 1996a). The Commission has admitted this may cause difficulty 27. Further, the impact on the industry s competivity of Member States fiscal policies, information schemes or any measures which hamper the diffusion of technologies, will have to be separated out from other nonagreement-related factors impacting economic performance. As explained above, the Commission has granted the automobile industry major concessions through the terms of the Commitment which make allowance for many factors which might negatively impact the automobile industry s progress in fulfilling its commitments. This is particularly detrimental with regard to the promotion of environmentally optimal technologies. 27 The Commission has stated 26.10.98: It is also recognised that the decision to whether a market change is linked to new technology or not will not always be easy to make, in response to questions from JAMA concerning details of the agreement. 42

The Analysis If Member States introduce fiscal measures to address, for example, noise or adverse health effects, such measures could be regarded as discriminatory against the technologies which ACEA has chosen to promote, even if they are less fuel efficient than the more environmentally optimal technologies. 3.2.3 Vehicle related legislation that may affect fuel economy Fuel economy improvements can be off-set by various factors such as, inter alia, policies addressing safety, recycling, noxious emission reduction and other measures that can affect the weight of a car and thus its fuel economy (see Figure 2, page 15). Debates surrounding prioritisation of policy objectives are likely to arise in the future. The difficulty will be that vehicle-related policies to address problems such as air quality or safety will be developed as binding legislation. Unlike the voluntary fuel economy objectives, such binding legislation can not be so readily adjusted. The Commission expresses its concern that other vehicle-related policy objectives might be compromised by the fuel economy objectives, not vice versa, and states in Paragraph 7 of its Communication COM(1995)689 (CEC 1995): The attainment of certain fuel economy targets should not counteract other policy objectives, especially with regard to the reduction of noxious emissions from motor vehicles and vehicle safety. An overly ambitious fuel efficiency target combined with a bad choice of policy instruments could compromise these other objectives. The Commission s Communication COM(1995)689 does not refer to win-win scenarios that can be obtained by optimising transport related policy objectives such as safety, noise and toxic emissions at the same time as addressing fuel economy. ACEA claims that between 1983 and 1997, fuel economy improved by 28% but due to offsets (e.g. consumer demands, specifications for safety, noise and emissions) the measurable improvement was just 8% (see Figure 2, page 15). In calculating the 1997 offer, ACEA declared that the target was just half of what could actually be achieved because allowance was made for the various offsets which would counteract fuel efficiency. Thus ACEA s 2008 target of 140gCO2/km could similarly be eaten away due to the impacts of new vehicle-related legislation. While the impact of binding legislation on fuel economy will have to be taken into account by the monitoring procedure, there is also the risk that the flexible voluntary agreement could in the future be responsible for watering down other vehicle-related legislation. ACEA is already noting offsets due to legislation developed since the adoption of the Commitment and recently remarked on the end-of-life vehicles Directive. The Directive requires manufacturers to take back vehicles free of charge from the last owner 28 and lays down provisions for collecting, processing, recycling, and recovering vehicles. ACEA has stated that the end-of-life vehicles Directive will have a significant effect on fuel economy as the recycling requirements, for example, will mean that manufacturers can not use certain light-weight materials (Interview 4). In a press release issued at the time, ACEA stated: Other provisions have to be considered in the further steps of the procedure. For example, the currently envisaged recycling quota for the year 2015 could hamper the development of 28 The last holder/owner of a vehicle is entitled to deliver a vehicle to a treatment centre free of charge and producers are required to bear all or most of the costs for applying these measures and/or to take back the end-of-life vehicles without there being any costs involved for the last holder/owner. The European Parliament and the Council agreed to a start date of 1.1.01 for vehicles put on the market after this date, and starting from 1.1.07 for vehicles put on the market before 1.1.01 43

The Analysis new material technologies that can considerably reduce car weight and lower CO 2 emissions. The car industry recommends proceeding to an in-depth examination of future car material composition before setting this level. (ACEA 1999) The Commission argues that the above argument is not justified, as there are alternative materials that could be used (Interview 1). It is already clear that there may be a need during the monitoring process, for independent studies to verify the automobile industry s offset claims. This could be an expensive exercise for the Commission. 3.2.4 Competition issues The automobile industry s voluntary agreements must comply with the EC Treaty competition rules and must be compliant with GATT/WTO rules. The agreements satisfied the demands of the EC-Treaty by giving formal notification to the Commission under Article 85 29 and each obtained exemption under Article 85(3) (CEC 1998o). DG11 and DG3 intended that ACEA, KAMA and JAMA operate as three bubbles such that the representative organisations, ACEA, JAMA and KAMA, would be responsible for ensuring acceptable contributions from the individual manufacturers. DG4, however, has made it clear that there will be no burden sharing as this could contravene competition rules if companies collaborate and arrange for some manufacturers to contribute more than others to their own benefit. Thus all the manufacturers will be required to demonstrate an equivalent effort. The issue of defining equivalent efforts initiated a conflict within ACEA as to whether a percentage improvement or an absolute target value should be pursued. BMW and Mercedes- Benz, as producers of heavier than average cars, were advocating a percentage improvement, as an absolute target of say 140gCO 2 /km would be well out of their range and would require an improvement of more than 25%. The French (PSA, Renault) and Italian (Fiat) manufacturers, however, were advocating an absolute target, over a percentage improvement, as the former would be easier for them to achieve 9. For example, PSA and Renault had a fleet average fuel economy of 162gCO 2 /km in 1996 (Environment Watch: Western Europe 4.10.96), thus the attainment of 140gCO 2 /km constitutes a fuel economy improvement of just 13.5%. Smaller manufacturers also argued that producers of larger than average cars, which generally yield greater profits, would be able to pass the extra cost for advanced technology on to the consumer more easily (Interview 1, 3 and 4). Adequacy of manufacturer contributions as regards equivalent effort will be left to the subjective judgement of the ACEA board and the Commission. ACEA members have still not discussed the contributions expected from each other (Interviews 3 and 4). According to ACEA, it is not possible to define equivalent efforts so contributions will be defined on an adhoc basis: Manufacturers contributions are not quantifiable. They will depend on many factors. (Interview 4) As explained earlier (see chapter 3.2.4 Competition issues, page 44), ACEA could not reach an agreement on the contributions that each manufacturer would have to make. Thus, a situation could later arise where manufacturers might not agree that each is making an equivalent effort and this may be a reason for some to claim economic harm due to the 29 Article 85 paragraph 1 of the EC Treaty prohibits all agreements between undertakings, decisions by association of undertaking and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the common market. But Article 85 paragraph 3, provides for an exemption where such agreements, decision on practices provide compensating benefits to the economy and to the consumer. 44

The Analysis Commitment. However, such economic harm is to be taken into account by the monitoring procedure: On the basis of the outcome of the monitoring, or if the impacts of this Commitment, on the European automotive industry, particularly its employment situation and its global competitive environment, are detrimental, ACEA and the Commission will review the situation and make any necessary adjustments in good faith. (Notes under Monitoring section of the ACEA Commitment; Figure 19, page 82). Thus, economic harm to manufacturers due to the Commitment (which can include fiscal measures), will also have to be separated from other factors that may affect sales. This will be no easy task. The automobile industry s commitments also require that the Community will use its best efforts to ensure car CO 2 reduction equivalent efforts from other manufacturing countries (e.g. in B Distortion of Competition, 2 nd paragraph of Assumptions in Commitment, see Figure 19, page 82). This is an unrealistic demand as the percentage CO 2 reduction required by the UNFCC Kyoto Protocol varies considerably from 5% for the USA, 7% for Japan to 0% for Korea. Therefore, the various countries will focus differently on each of their CO 2 producing sectors and may choose to apply policies unique to their own situation. 3.2.5 The commitments of non-acea members ACEA s Commitment is also based on non-acea members committing to equivalent CO 2 reduction efforts for their sales in the EU (Under B Distortion of Competition of the Commitment, Figure 19, page 82). As is clear from this paper, achieving agreement with ACEA was no easy task for the Commission. For ACEA to hold to its Commitment, the Commission needed to obtain equivalent commitments from KAMA and JAMA. It was originally intended that the negotiations with ACEA would be in parallel with non-acea members. Both JAMA and KAMA state that they expressed willingness to participate when they were contacted in 1996. However, the negotiations with KAMA and JAMA did not begin until late 1998, after an agreement had been established between ACEA and the Commission (Interview 10). JAMA and KAMA were displeased with such an arrangement as it would mean take it or leave it for their members (Interview 10 and 17). Initially there was considerable confusion throughout the Commission, as well as JAMA and KAMA, as to what constituted an equivalent effort. KAMA had a 1995 fleet fuel economy average of 206gCO 2 /km while JAMA had a fleet average of 193-202gCO 2 /km (Interview 1, 2, 10 and 17). To achieve 140gCO 2 /km, both KAMA and JAMA would therefore need to reduce their fleet average fuel economies by more than 25%. Both JAMA and KAMA had the opinion that an equivalent effort would mean a 25% improvement by 2008 (Interview 10 and 17), but the Commission was insisting on a mirror-image agreement of 140gCO 2 /km within the same time-frame, with the same assumptions, and with variations only if justified (Interview 1, 2 and 17). The issue therefore centred upon how much beyond a 25% fuel economy improvement KAMA and JAMA would be required to contribute and if the absolute target of 140gCO 2 /km was a fair demand. Indeed ACEA members will not all contribute the same amount in percentage terms (e.g. 25%) and ACEA will decide what constitutes an adequate contribution from their members. The manufacturers of JAMA and particularly KAMA also argued that with far fewer members than ACEA, they did not have as much flexibility as ACEA (Interviews 10 and 17). 45

The Analysis The Commission believed that if voluntary commitments could not be obtained from non- ACEA manufacturers, they would be separately legislated (Interview 12). As the negotiations with JAMA encountered difficulties, DG11 prepared a legislative proposal to legislate JAMA, with the intention to maintain ACEA and KAMA under the voluntary agreement (Interview 12). However, it is improbable that such proposals would have complied with GATT/WTO rules. Indeed the proposal encountered technical difficulties in the Commission s legal department, but the proposal was in the end shelved as JAMA and the Commission reached consensus on the terms for a voluntary agreement (Interview 12). Figure 12 Non-ACEA Members KAMA, The Korean Automotive Manufacturers Association: Daewoo and Hyundai. JAMA, The Japanese Automotive Manufacturers Association: Toyota, Mitsibushi, Mazda, Nissan, Honda (+other small companies). Other non-acea manufacturers, such as for example Lada and Proton, compose just 0.1% market and there is consensus that the share is insignificant. By the end of 1998, the US manufacturer Chrysler, was under the umbrella of Daimler-Benz. 3.2.5.1 The negotiations with KAMA KAMA was reluctant to agree to an average of 140gCO 2 /km. KAMA argued that the Korean industry was going through some painful restructuring 30, was suffering financially, did not have the technology and did not want to commit to more than 25% (Interviews 1, 2 and 17). But the Commission argued that: the Korean models were inefficient compared to European models; that the Koreans had previously bought technology from companies such as Nissan, General Motors, Mercedes and others and could do so again; and that Daewoo and Hyundai, as two of the richest companies in Asia, were planning to expand considerably throughout Europe (Interviews 1, 2, 4 and 10). The Commission also pointed out that the Koreans had already broken competition rules in an attempt to avoid import duty by disassembling models made in Korea, exporting them to Slovakia where they were reassembled and exported to Europe duty-free (Interview 1). In addition Daewoo and Hyundai with its limited model range would only have to introduce a mini or hyper-efficient car to achieve most of its target (Interview 10). For several weeks the negotiations were under deadlock, which caused ACEA great anxiety as negotiation breakdown could have meant the possibility of general legislation (Interview 1 and 2; SEC(1999)364, CEC 1999c). The Koreans assumed wrongly that an agreement had already been reached with JAMA, and representing just 3.1% of the European market felt considerable pressure to close a deal (Interview 10 and 17). The Koreans eventually offered 140gCO 2 /km by 2010 (Interview 17). A compromise of 2009 was swiftly reached (Interviews 1, 2 and 17). The Commission was very satisfied with the outcome and ACEA was content (Interview 1). However, ACEA has already pointed out that the one extra year to achieve 4gCO 2 /km could be a significant competitive advantage for KAMA over ACEA when the 2008 target of 140gCO 2 /km is approached (Interview 4). The Koreans would not be required to commit to the production of a 120gCO 2 /km vehicle by 2000 (Interviews 1, 2 and 17) as lead times for production are a minimum of 3 years. In any case this requirement was already fulfilled by ACEA when the Commitment was drafted in 30 (The Economist 20.8.99) South Korea s government will allow Daewoo, the country s second-largest chaebol, to be broken up. Many of its businesses will be sold off in the hope of realising at least half of its debts of around $50 billion. 46

The Analysis 1998. As not all ACEA members will be producing a 120gCO 2 /km car by 2000, it would appear unfair to expect the same from the two manufacturers represented by KAMA. 3.2.5.2 The negotiations with JAMA JAMA and the Commission could not agree upon the 1995 Japanese fleet average fuel economy value. The Commission had used data supplied by the French automotive association AAA to calculate ACEA s market average, hence they were keen to use the same data source for JAMA s calculation. The Commission pressurised ACEA to interfere with the AAA but to no avail; the influence of PSA and Renault is believed to have been responsible for ACEA s lack of action (Interviews 3 and 10). It was clearly in the interests of ACEA to assist in the negotiations with JAMA as negotiation breakdown would mean legislation, but it seems that the suppression of the information was intended to manipulate the Japanese or stall the negotiations. For four months the AAA would not provide the data for JAMA. JAMA therefore hired consultants to carry out the calculations. JAMA s consultants estimated a fuel economy average of 202gCO 2 /km, after which AAA issued a figure of 193gCO 2 /km. As a result consensus was not achieved on a single figure (Interviews 1, 2 and 10; CEC 1999d). To make allowance for ACEA and JAMA s different starting points, due to the Japanese EU market composition distortion because of trade barriers 31 (the latter of which is expected to change and approach the ACEA market composition with the lifting of trade barriers in 2000), it was suggested by the Commission that the target of 140gCO 2 /km be based upon an index linking the current Japanese market mix to the ACEA market mix (Interviews 2, 10 and 12). By this method the difference between the ACEA and Japanese market compositions could be linked by a weighting index and could thus be compared (Interviews 2, 10 and 12). In May 1999, the Commission pressurised JAMA for an offer. JAMA felt the term ACEA market mix had not been clearly defined (Interviews 2, 10 and 12). Japanese culture prefers details are finalised and worked through by a bottom-up approach before the final product is rubber stamped at higher level in contrast to the top-down approach of the Commission, used in negotiations with ACEA (Interviews 10 and 12). JAMA was therefore extremely reluctant to ask its members to sign up to an agreement with vague assumptions and details. Nevertheless approval was reluctantly obtained on 20 th May 1999 from the chief executive officers of the Japanese manufacturers (Interviews 2 and 10). When the final offer was presented to the Commission on the 27 th May 1999, in the faith that the contents were the requirements of the Commission, the representatives of the various levels within the Commission were unsure and confused after several meetings as to the acceptability of the offer with regard to JAMA s definition of ACEA market mix, (which was more detail than the Commission had wanted at this stage) (Interviews 1 and 2). The offer was shown to ACEA, much to the chagrin of JAMA (Interview 10). A loophole was recognised by ACEA and seen as a threat to European competivity that could undermine the environmental objectives. The shortcoming was acknowledged and accepted by the Commission and JAMA (Interviews 2 and 10); CEC 1999d). Indeed, it would have been the case that if the Japanese market mix did not move towards the ACEA market mix by 2008, the Commitment as written could have permitted JAMA to officially achieve the 140gCO 2 /km target but in reality achieve no more than possibly 160gCO 2 /km. 31 Due to the EU trade restrictions imposed by the EU on Japanese imports for the last 10 years, the Japanese market mix is distorted with a bias towards larger luxury vehicles which provide a greater profit margin than smaller vehicles (Interviews 1 and 10). With transplants set up throughout Europe and the restrictions due to be lifted in 2000, it is expected that the Japanese market mix will approach the ACEA market mix by 2008 (Interviews 2 and 10). 47

The Analysis While JAMA insisted that it had not realised the shortfall of the text (Interviews 2 and 10), the Commission negotiators were split in their support for JAMA. The Japanese manufacturers pointed to their already fulfilled intention to expand their small and medium sized sectors throughout Europe, as several European transplants are already established, in preparation of the lifting of trade barriers in 2000. It had also been expected by most parties that it would be relatively easy to establish an agreement with JAMA as the automotive association had adopted a positive attitude from the beginning of the negotiations to the concept of a commitment (CEC 1999c). On 22 nd July 1999, the Commission and JAMA came to a compromise agreement with the only differences of a target of 140gCO 2 /km by 2009 instead of 2008 to allow for JAMA s distorted market mix and a higher intermediate target than ACEA of 165-175gCO 2 /km. The agreement was reached in September 1999 and the Commission presented JAMA with its Recommendation 2000/304/EC on the 13 th April 2000. As with the KAMA negotiations, the extra year to achieve 140gCO 2 /km may result in ACEA calling for a timeframe extension to 2009 as their 2008 deadline approaches. As Japanese manufacturers are planning to increase their market share which is already considerable at 11.5% despite trade restrictions (compared to the Korean market share of 3.1%), ACEA will become increasingly protective of its own competivity (refer to Figure 6 page 23). The competition within the automotive industry is also enhanced by the fact that the European market is reaching saturation point and running at over-capacity 32 (Financial Times 15.1.98; Wells & Newenhius 1997). Further, due to well-established Japanese regulations designed to promote smaller, more fuel efficient cars, Japanese technology is advanced and R&D programmes are well developed (Interview 10). Indeed, Toyota launched the world s first commercial hybrid in 1997. The ambiguous terms of the Commitment, the insensitive attitude from parts of the Commission and the aggressive attitude of ACEA, all contributed to reduce JAMA s support for the voluntary approach. At one stage the negotiations were close to breaking down; it has been indicated by the negotiators that there was a 50% chance that JAMA might have withdrawn from the negotiations and opted for legislation (Interview 10). 3.2.6 Monitoring The proposal for a Decision on monitoring CO 2 emissions from new passenger cars was adopted in summer of 2000 (COM(98)348, CEC 1998c; Decision (2000) 1753, CEC(2000b)), following conciliation talks between the Parliament and Council. The holistic and statistical monitoring are to be administered jointly by both ACEA and the Commission in order to monitor ACEA s progress and allow scrutiny of the assumptions underlying the Commitment (see Figure 13, page 51). The Commission will also report each year to Parliament and Council on the data received. The aim of the proposed Decision is: 1. To provide data to allow assessment of the changes in the new car fleet which could impact on other Community environmental objectives e.g. air quality, acidification, this is particularly with reference to changing diesel/gasoline market mix. 2. To assess the contribution of individual manufacturers 3. To collect data regarding the distribution of CO 2, mass, power and engine capacity. Such information can assess the potential impact on road safety, the future demand for motor fuels, competitiveness of the automotive industry etc. 32 (Nakamoto 15.1.98) Japan carmakers renew their offensive: report of surging Japanese imports into Europe with transplants and a concern with over-capacity in Europe and the ability of the Japanese to bring cars to the market much quicker than Europeans. 48

The Analysis No Community-wide scheme to monitor the CO 2 emissions from new passenger cars currently exists nor has any Member State put into operation such a scheme at national level. The most comprehensive scheme to monitor CO 2 emissions from new passenger cars has been developed by the European Conference of Ministers of Transport (ECMT 1997). The Parliament was successful in introducing several environmental improvements to the Decision proposal and stressed its position on voluntary agreements and the need to use the Decision as a basis to monitor the automobile industry s fulfilment of their commitments. Most particularly, Council agreed to accept Parliament s proposal to urge the Commission to look at the need for a legal framework, including the issue of sanctions, for voluntary environmental agreements to be entered into in the future. However, such a legal framework would not be applied retroactively to existing agreements such as that with the automobile industry. At the same time, the Council did not agree to include vehicle dimensions among the data to be transmitted by the Member States, largely due to lobbying from industry which is sensitive to such information revealing market trends on vehicle shape and size. Nevertheless, such trends are quite relevant with respect to fuel economy e.g. aerodynamics. Parliament also intended to extend the Decision to all other types of motor vehicle including light commercial vehicles, but its proposal was not supported by Council and thus not adopted. The Commission persuaded Council and Parliament not to pursue this latter amendment as they believe these other vehicles are reputed for being efficient on a consumption level (trucks), or because their overall share in CO 2 emissions is low. However, alternatively powered vehicles will be included in the scope. Parliament was also persuaded by the Commission and industry, not to propose that air conditioning be included in the data reported due to the complexity of taking such a parameter into account (see Chapter 3.1.3, page 34). Parliament also proposed that Member States report on how CO 2 emissions change and whether reductions are due to manufacturer technical measures or changes in consumer behaviour. The Commission shall report to the European Parliament and to the Council by the end of 2002 at the latest on the operation of the monitoring scheme. The data collected under the monitoring system from the year 2003 onwards shall serve as the basis for monitoring the auto industry s voluntary obligations to reduce emissions of CO 2 from motor vehicles. The Commission will also have to report annually to Council and Parliament on the data submitted by Member States to the Commission. ACEA insists that the statistics for individual manufacturers should not be quoted publicly (Interview 4) 33 and the Commission supports this view and has expressed its intention to keep its distance from the internal politics of ACEA (Interview 5). Nevertheless, disclosure of information for individual manufacturers would allow the public and NGOs to compare the manufacturers and thus encourage competivity and proactivity 34. The Commission should have an EU database established by 2001, based on statistical data obtained from Member States who will need to modify or set up national databases (Interview 1). ACEA, JAMA and KAMA, will obtain data from its members. Obtaining the same data from different sources is intended to satisfy the Commission s guidelines on environmental agreements (CEC 1996b) which requires the data to be independently verified. But in some cases, data obtained separately by the Commission and the industry will have originated from the same source (Interview 4). According to COM(1998)348 (CEC 1998c) there are several potential sources for error. These include human error in the conversion of data contained in the paper type-approval documentation to a digital form, the selection of the correct version-specific data from such 33 The Commission will be receiving data on individual manufacturers but it will not necessarily make such data public. 34 A member of the public or an NGO could, although only with considerable effort, gather the information necessary to illustrate the progress of the individual manufacturers. NGOs and the public may, however, be able to obtain access to the information relayed by the Commission to the Parliament and Council through the annual reports required under 2000D1753 (CEC 2000c), but the data may not necessarily include data specific to manufacturers. 49

The Analysis electronic databases of type-approval information, human error during the transfer of data from the paper Certificate of Conformity to the electronic registration file during vehicle registration and the transfer of incorrect data from manufacturers and dealers during automated vehicle registration (ECMT 1997). Due to the various agreed terms underlying the industry s commitments, the Commission and industry will have to reach consensus on the many factors which may have to be taken into account by the monitoring procedure (see Figure 13 below). Offsets due to vehicle-related legislation adopted since the adoption of the Commitment will have to be taken into account. In addition gains or losses due to the availability of low sulphur fuel will also have to be considered, as the Commitment assumptions require EU wide fuel quality which is superior to that currently legislated (chapter 3.2.1, page 38). The impacts on financial performance, competition and employment of the industry are also to be considered, which may include the impacts of fiscal measures for example. Nonetheless, it will be difficult to separate out the many factors that could be responsible for declining profits. The 10% gain that the industry expects from non-technological means (Point 2 of the Commitment, Figure 19, page 82) will also have to be separated from the Commission s efforts to encourage downsizing through its fiscal framework and consumer information scheme aimed at achieving the extra 20gCO 2 /km called for by Council (Council of the European Union 1996a). At the same time, the negative impacts of fiscal measures on both the industry s progress with respect to its commitments as well as its economic situation are to be considered. For example, the monitoring scheme will also have to take account of any factors that hamper diffusion of technologies into the market, which might include fiscal measures (e.g. Assumption C of ACEA Commitment, refer to Figure 19, page 82). While there is much scope for statistical discrepancies between the monitoring procedures of the Commission and industry, the joint holistic monitoring is likely to encounter more difficulties as consensus will need to be sought on the interpretation and acceptance of the findings that affect the status of the assumptions ( Figure 13, page 51). The monitoring procedure requirements will also be resource intensive for the Commission, especially if the Commission has to carry out studies to verify automobile industry s claims for factors which are negating progress. The effectiveness of the monitoring procedure will also depend on existing actors and new faces to the scene. While the turnover of Commission officials is varied (from very short term to long term), the ACEA president rotates annually by manufacturer, in contrast to MEPs and Commissioners that are elected every 5 years 35. 35 1999 has witnessed a change of Parliament and Commissioners. Bjergaard of DG11 has been replaced by Wallstrom and Bangmann of DG3 by Liikanen 50

The Analysis Figure 13 Factors to be taken into consideration by the monitoring procedure 51

The Evaluation 4. Evaluation of the Voluntary agreement 4.1 The capacity of ACEA s voluntary agreement to ensure CO 2 reductions from the passenger car sector Studies estimate that the agreement will only stabilise emissions at some 20-30% above 1990 levels for CO 2 from passenger cars (see chapter 3.1.2, page 32). This is also an overly optimistic estimation as simulations fail to take into account the additional weight (e.g. accessories such as air-conditioning, luggage, passengers) added to the test model before the car is sold, and the unrealistic driving conditions simulated by the test cycle. As the EU s Kyoto Protocol commitment requires an 8% greenhouse gas emission reduction below 1990 levels by 2008-2012, the passenger car sector s shortfall will have to be met by some combination of contributions from other sectors, Member State transport policies and/or through some form of emissions trading. The agreed target of 140gCO 2 /km (i.e. 25% improvement of fleet s average fuel economy) for 2008, with only a possibility of achieving 120gCO 2 /km by 2012, is not ambitious. Based on the known existing economically feasible technological potential and the potential to be gained from downsizing, a far more ambitious target could have been set. If the automobile industry s voluntary agreements would be binding they would also be enforceable. As they are non-binding they can only be enforced by applying penalties, sanctions or sticks of some sort. However, the Commission does not have adequate tools at its disposal and has not even prepared a binding legislative proposal - which could be an effective stick - to replace the voluntary agreements in case they should break down. As the achievement of the agreements objectives cannot be guaranteed, it is therefore a risky and inappropriate policy instrument to use, particularly as the European Union s commitment to the UNFCC Kyoto Protocol 36 is binding once ratified. The greater the delay in action to reduce greenhouse gas emissions, the greater the emissions reductions will have to be in a certain period of time in order to stabilise greenhouse gas concentrations in the future. At the same time, delayed action increases the risk of damage to ecosystems as they will need to adapt to the rates of change in temperature and sea-level rise resulting from this lack of action. The institutions were therefore rightly keen to attain a result within the shortest time-frame possible. The Commission s initial communication for the three pillar strategy, COM(1995)689 (CEC 1995), and the responses of Council and Parliament to this Communication point clearly to a fixation on the year 2005. The Council later adjusted its time-frame to 2010 at the latest, much to the criticisms of Parliament who called for 120gCO 2 /km by 2005 and 90gCO 2 /km by 2010. It was due to the establishment of the Kyoto Protocol in December of 1997 that the time-frame of 2008-2012 and a quantitative target for EU greenhouse gas emission reductions were introduced into the negotiations. A longer time-frame was one of the factors which ACEA claimed enabled the manufacturers to accept a more stringent target (Interview 4). At the same time, it can be argued that the lengthening of the timeframe was appropriate to allow manufacturers to adequately plan. However, as previously mentioned, the lengthened timeframe could have allowed for the introduction of a more stringent target with respect to environmental objectives and with respect to the technological potential that currently exists. Indeed, many policy experts support the idea that it is not the stringency of the regulation that is the problem but rather the way it is designed and implemented (Porter & Van der Linde 36 The EU has committed to the UNFCC COP 3 (Kyoto 1997) to reduce greenhouse gas emissions by 8% below 1990 levels by 2008-2012. The Kyoto Protocol is yet to be ratified. 52

The Evaluation 1995a, 1995b; Clarke 1994). Considering the long planning times that automobile manufacturers require, it is necessary to indicate to industry the long-term environmental objectives that are required. Parliament was alone in calling for post-2012 targets. There were no discussions between the automobile industry and the Commission with regard to setting post 2012 targets (if the review of 2003 extends the agreement to this date). The Commission believes that post-2012 is too far away and the technological and economic situation may change considerably leading up to this time (Interviews 1 and 2). In order to react to the warning of the IPPC (IPPC 1996) that a reduction of some 70% in carbon emissions world-wide will be required to keep the accumulation of global warming gases within a ceiling of 450ppm, the use of economically recoverable fossil fuels will have to be phased out within this timeframe (see Figure 14). This will require significant contributions from all sources. Yet, ACEA indicates that it has already been discussing such issues at board level in anticipation of future legislation (Interview 4). Figure 14 The Carbon Budget Source: Greenpeace (Hare 1997) Hare (1997) has estimated that a carbon budget of 225 billion tonnes of carbon is the amount which can be burned without exceeding the ecological limits of 1 C above pre-industrial global average temperature and a 20cm sea level rise above 1990 levels. It is assumed that major action is taken to halt deforestation, stabilising the role of forests at current levels, which would involve a significant global reafforestation programme next century. The carbon budget of 225 billion tonnes is equivalent to one quarter of the economically recoverable reserves and less than 5% of the world s total estimated reserves. With carbon use currently at approximately 6 billion tonnes of carbon per annum, the 225 billion tonnes of carbon will not last 40 years at current rates of use. The terms of the agreement are designed to support internal combustion engine technologies and do little to support alternative technologies or fuels e.g. higher quality fuel specifications, consideration of fiscal measures. The terms of the agreement also lay the target and timeframe particularly vulnerable to renegotiations. The monitoring procedure, conducted jointly by the automobile industry and the Commission will assess whether the terms of the 53

The Evaluation agreement or assumptions hold true or not. Many of the assumptions are catch all expressions that will allow ample opportunity for industry to excuse itself. Assumptions which are particularly at risk include those that take into account binding vehiclerelated policies which might neutralise fuel economy improvements (i.e. if legislated policies add extra weight which increases fuel consumption), and the demand for some EU-wide availability of fuel with a 30ppm sulphur content by 2000 and full availability by 2005, rather than 50ppm as legislated by Auto-Oil 1 for 2005 37. Thus, the agreement s monitoring process will have to take into account the impact of the variation in fuel quality throughout the EU relative to the industry s assumptions, when considering industry s achievement of its targets in 2003 and 2008(9). If the sulphur content of fuel is to be further reduced from 50ppm, it is important to note that the automobile industry may be awarded CO 2 emissions reductions which have in fact been transferred to the oil refinery sector which emits more CO 2 due to the production of improved quality fuel. As the automobile industry is to achieve 90% of its Commitment through technology, 10% can be achieved by non-technological means such as downsizing. At the same time, gains attributable to the latter will have to be separated out from gains due to fiscal policies and fuel economy information schemes, designed to close the gap on the Council s target of 120gCO 2 /km. Further, the monitoring procedure will have to consider any factors which hamper the diffusion of technologies into the market. Such measures could include, for example, fiscal measures. In addition, ACEA may also regard KAMA and JAMA s extra year to achieve the 140gCO 2 /km target (i.e. 2008 instead of 2009) as an unfair advantage and thus call for an extra year to achieve their target, especially if KAMA or JAMA should increase their market share. Not only will the monitoring procedure have to consider negative impacts on the automobile industry s commitments, but it will also have to separate out and take into account impacts on financial performance, competition and employment due to the voluntary agreement. The Commission and the automobile industry will also have to achieve consensus on issues raised during the monitoring procedure, which like the agreement negotiations will be influenced by the personalities of the various individuals. In addition, there exists considerable scope for statistical discrepancies between the data collected by the Commission and the industry. Consensus may not always be easy to achieve as the Commission will be reluctant to give anything away in order to ensure the success of the voluntary approach. Watered down targets or lengthier timeframes will not meet with approval from Parliament, Council and NGOs. At the same time, the Commission will not want the voluntary agreement to collapse as this could significantly damage the reputation of voluntary agreements as an EU policy instrument. This latter point could seriously weaken the Commission s bargaining power in the monitoring negotiations. If the Commission has to belatedly resort to legislation, this will also be to the detriment of environmental protection. The Commission will find it difficult to know when to deploy legislation, but it is likely to be near to 2008 when it will become clear if the target will not be met or not. Environmental protection could be weakened by the additional time needed to allow for the development of legislation, which might require an extension of the timeframe by some two or more years. However, there will be an opportunity to set more stringent environmental objectives, but their achievement will depend on when the legislation is implemented and on the effectiveness of the legislative method used (see chapter 4.3.2.2, page 65). It should be noted that the Commission still does not have a legislative proposal prepared in case the agreements should break down. Legislative possibilities have been 37 The automobile industry insists that low sulphur fuel is necessary for the introduction of more fuel efficient technologies and to enable technologies to meet emissions standards. 54

The Evaluation discussed but efforts have not been invested to solve some of the problems or to further explore alternative legislative options. 4.2 The voluntary agreement as an environmental policy tool for application at EU level 4.2.1 The legal aspects It has already been stated in the above section that the voluntary agreement is a risky and inappropriate policy instrument to use for regulating CO 2 from cars, particularly as the European Union s commitment to the UNFCC Kyoto Protocol will be binding once ratified and CO 2 emission reductions will need to be guaranteed. By adopting a non-binding agreement, the Commission has ignored the lessons to be learned from the Member States considerable experience, especially in the Netherlands and Germany. Indeed, it has ignored its own advice as to the effectiveness of binding agreements as stated in Paragraph 19 of its Communication COM(1996)561 (CEC 1996b): Binding agreements provide in general better safeguards in terms of achieving environmental objectives. Since the publication of COM(1996) 561, six environmental agreements have been established at Community level. These have included the three agreements with ACEA, JAMA and KAMA, and as is common to all six, not one has been binding in a strictly legal sense. This failure to develop and establish binding measures is surprising. Council Directive 91/441/EEC of June 1991 38 requires that the Council shall decide on measures designed to limit CO 2 emissions from motor vehicles, acting by a qualified majority on a proposal from the Commission. In addition, Article 175 of the EC-Treaty states that the Council, in accordance with the co-decision procedure, and having consulted to Economic and Social Committee and the Committee of the Regions, shall decide what action is to be taken by the Community in order to achieve the environmental objectives of the EC-Treaty. It would appear that the Council should decide on measures to regulate CO 2 emissions from motor vehicles, while it can also be argued that as these measures aim at achieving Kyoto targets (an environmental objective), that the Parliament should be implicated under Article 175 of the EC Treaty. Instead the Commission, with the support of Council, developed the voluntary agreements with ACEA, JAMA and KAMA. At the same time, the lack of an agreed legislative method also reduced the Commission s bargaining power during the voluntary agreement negotiations, as an alternative binding legislative proposal would have provided a stick with which to threaten industry. The Commission s adoption of voluntary agreements at Community level raises legal concerns. It might be suggested that the Commission has usurped the power of the Council by taking the voluntary route and not using the apparently available legal regime of Article 175 or Directive 91/441/EC. However, the matter is complex. The Commission experienced technical and political difficulties in the early nineties when attempting to draft binding legislative proposals which would be applicable to all cars sold in the EU. Thus the Commission decided to consider setting an average fuel economy target for the entire fleet of new passenger cars sold in the EU, leaving it up to the groups of manufacturers (i.e. ACEA? JAMA and KAMA) to decide how this would be achieved. But there were difficulties in making such a measure binding. The use of a Regulation, which is of general effect, or a Directive, which is addressed to Member States would probably not be appropriate if the target was a 38 Council Directive 91/441/EEC of June 1991, amending Directive 70/220/EEC on the approximation of the laws of the Member States relating to measures to be taken against air pollution by emissions from motor vehicles 55

The Evaluation discrete group of car manufacturers. The Council and Parliament could have addressed a Decision(s) to ACEA, JAMA and KAMA. But a Decision, as with any proposed Regulation or Directive would have needed binding targets for the sake of legal clarity. This apparently could not be achieved and so the non-binding route was chosen. However, as is argued below, if the Commission has arrived at an agreement which is in effect binding then the legality of the voluntary route becomes more doubtful. The Commission s decision to adopt voluntary agreements with the automobile industry may pose additional questions of legality. As a policy tool, the scope for such non-binding agreements had already been foreseen. According to COM(1996)561, these can take a previously used format i.e.: 1) industry commitments recognised by the Commission i.e. unilateral non-binding commitments 2) any other form of a non-binding understanding (e.g. exchange of notes, letters of intent, declarations signed in the presence of a Member of the Commission) However, doubts remain as to their proper legal basis. The EC Treaty is silent on the mandate required by the Commission to negotiate an agreement with an industry sector, and while this is not fatal to their formation, the scope of their negotiation and use falls within a grey zone of community law. By analogy, Article 300 of the EC-Treaty imposes a prior mandate, according to the provisions of this article, for the negotiation and conclusion of international agreements. However, it cannot be argued that these six Community level voluntary agreements fall within the scope of Article 300. The Commission s response to this apparent lacunae in EU law is to adopt the environmental agreements by use of Recommendations. Article 211 of the EC-Treaty gives the Commission the absolute right to adopt Recommendations if it considers necessary, yet it should be noted that under Article 249 such Recommendations will have no binding force. A serious concern with the automobile industry s agreements is that there appears to be some form of underlying commitment from the Commission. For example, in the preamble to the Recommendation 1999/125 concerning ACEA s Commitment, it is stated that the Commission will: only legislate should the voluntary agreement prove unsuccessful; obtain commitments from other non-acea car manufacturers; jointly, with industry, monitor and review industry s progress; acknowledge that no further fiscal measures are needed beyond the Commitment and so will take their impacts into account (CEC 1999b). The question is whether these seeming commitments on behalf of the Commission have in fact produced a de facto binding measure. Has the Commission, using non-binding Recommendations, effectively tied the hands of the European Union as to the action it may take? This is a question that only the European Court of Justice as the guardian of the Treaties can answer definitively. However, from a political perspective, it is likely that should the Commission fail to uphold its commitments under the Recommendation, that there would be an adverse reaction on behalf of the automobile industry. If it is accepted that these voluntary agreements are not binding, it may still have been more appropriate to have arranged their introduction through a Recommendation of the Parliament and the Council rather than of the Commission. This would have allowed for the involvement of all three institutions in the development and adoption of the voluntary agreements. In an indirect attempt to improve the situation, Parliament proposed an amendment to the Commission s proposal outlined in COM(98)348 (CEC 1998c) for a Decision on Monitoring of CO 2 from passenger cars which urged the Commission to put forward as soon as possible a 56

The Evaluation legal framework for the agreement entered into with car manufacturers organisations, including measures to be taken should the agreement fail to work. As a result, the new Commissioner for the Environment, Wallström, declared in November 1999 that she would not negotiate or propose any further environmental agreements until the framework for their operation had been clarified by DG ENV s (new title for DG11) proposal for a Regulation on environmental agreements. The Regulation will clarify the legal framework for environmental agreements established at Community level. Parliament insisted that the legal framework should also be applied to existing agreements, but its wishes were rejected. It is a basic rule of law that new legislation will not act retrospectively. However, the Parliament and Council did agree to a compromise text under the conciliation process in May 2000 based on the explicit undertaking of the Commission to submit a legal framework for environmental agreements to be entered into in the future with organisations. At the same time, DG TREN (the newly combined DG7 and DG17) has taken a different position to DG ENV and has recently initiated negotiations with the maritime industry for a voluntary agreement on maritime safety (Environment Watch 15.9.00). It is intended that the Regulation for environmental agreements will further define the role of the Parliament and the Council in deciding commitments, and that monitoring and enforcement aspects will be addressed. At present, it seems the proposal is moving in the direction of the model for voluntary schemes used in the US, whereby environmental objectives are set and incentives are provided for participation and sanctions for breach of conditions. However, application to participate would remain voluntary while the terms and conditions of operation would be legally binding (Eammon Bates Issue Tracker February 2000). 4.2.2 The Role of the Commission, Parliament, Council and NGOs Council clearly had the political will to address CO 2 from cars as it was Council that initiated the debate on targets to be set for the automobile industry, calling for a target of 120gCO 2 /km by 2005. Despite a change in test-cycles, which meant that 120gCO 2 /km under the new testcycle was more ambitious than under the old test-cycle, Council managed to hold on to its proposed target of 120gCO 2 /km. However, the automobile industry was relieved of 20gCO 2 /km of the burden, as the Council approved of the Commission s 1995 three pillar strategy proposal to achieve the extra 20gCO 2 /km through the fuel economy information scheme and fiscal framework. At the same time as promoting ambitious targets, the Member States were failing to agree on methods to limit CO 2 emissions from cars, partly due to bias towards their own national industries. Member States were also failing to give their political support to the Commission s proposals to regulate CO 2 from cars through a purchase tax or an annual circulation tax. The Commission should have foreseen that the proposals for fiscal measures would be unlikely to achieve unanimous approval in Council. The explanation that the potential problem was not identified ahead of time may be because the use of fiscal measures was pushed excessively from within certain parts of the Commission, which appeared to be fuelled by inter-dg competivity. However, it is clear that more effort should have been invested into alternative policies to regulate CO 2 from cars, instead of relying on the political bravery of Member States to apply fiscal policies. This also highlights the fact that it is unlikely that effective fiscal policies will be implemented at Community level for the achievement of environmental objectives while the unanimous voting procedure continues to be used for deciding on fiscal policy. 57

The Evaluation The benefits of developing a package of measures to regulate CO 2 from cars was not recognised until 1995. However, when the Commission began to develop its three pillar strategy, the various alternative methods to regulate vehicle fuel economy seemed to be too readily dismissed without allocating adequate resources to weighing up the pros and cons of each, including the voluntary agreement. The various proposals submitted to the Commission s MVEG group in the early 1990s could have provided a useful starting point. Council s preference for non-binding measures to legally binding ones is clear from statements in documents such as the Council Conclusions of 1992 (Council of the European Union 1992) and the 5 th environmental action plan, and through the ratification of the Treaty of Maarstricht, which placed much emphasis on the principle of subsidiarity. This may be part of the reason why little effort was given to exploring the merits of different schemes to regulate CO 2 from passenger cars. The failure of the Commission to achieve consensus on a policy method to legislate CO 2 from passenger cars in the early 1990s (see chapter 2.1, page 17) not only set back the intention to tackle the issue by several years, but weakened the Commission s bargaining position in the voluntary agreement negotiations. With no developed legislative proposal, the Commission had no stick with which to threaten industry. This power imbalance contributed to the Commission s failure to achieve an ambitious target in 1997. Other factors which also played a role in the Commission s failure included: the lack of technical support for the Commission desk officers; opaque negotiations dominated by industry technical experts; an absence of commitment from the top in the Commission; and the Commission and Council s preoccupation with the voluntary approach as a third way. The Council demonstrated its commitment to a meaningful target by rejecting ACEA s offer of 167gCO 2 /km in 1997 as quite inadequate and insisted the motor industry take action to ensure a satisfactory outcome of the negotiations (Council of the European Union 1997). In retaliation to industry s unambitious offer of 167gCO 2 /km (by 2005) coupled with the pressure of the EU s agreed greenhouse gas reduction target and timeframe under the Kyoto Protocol, the Commission began negotiating at higher level. The higher level negotiators adopted a strictly non-technical and political approach. Due to a much stronger commitment from the top and a changed approach, the Commission was able to achieve a much more ambitious target in March 1998 of 140gCO 2 /km (by 2008), which the Council noted with interest (Council of the European Union 1998a). Having achieved the target, ACEA and the Commission then had to finalise the drafting of the assumptions of the Commitment. Recognising the difficulties, the Council stressed the need to resolve ambiguities and outstanding issues in the ACEA draft Commitment, and urged the Commission and industry to urgently conclude their negotiations (Council of the European Union 1998b). While the Commission and Council were evidently keen to establish a voluntary agreement with the auto-manufacturers, internally the Council did take a tougher line on the targets and terms of the agreement to which the Commission could accede. The Council repeatedly requested the Commission to prepare binding legislation should the negotiations break down at any time (Council of the European Union 1997b, 1998c) and stressed the importance of the shortest possible timeframes, intermediate targets, a monitoring scheme and a fiscal framework (Council of the European Union 1996b, 1998c). The Council also requested that the Commission evaluate in 2003, the progress towards reaching 120gCO 2 /km by 2005 or 2010 at the latest, rather than by 2012 which is presented as a possibility in the Commitment. 58

The Evaluation The Commission looked to the Council for guidance and approval but did not always take its concerns or demands fully into account, particularly with respect to drafting an alternative binding proposal. However, it is clear the Commission gave even less consideration to the position of Parliament. While the Parliament generally supported the thrust of the Commission s three pillar strategy (Interview 14), the various Parliamentary Committee reports reveal much discontent as regards the adequacy of a voluntary agreement as a policy instrument. The Parliament s concerns included inter alia (Interview 14; European Parliament 1997a, European Parliament 1998; Europe Environment 9.7.96): a) The exclusion of Parliament from the decision-making process b) The lack of an ambitious target c) The lack of post-2012 targets d) The lack of arrangements for a continuation with the commitment should one or more of the assumptions on which ACEA and the Commission have based it not hold true e) The imprecise nature of the intermediate target which is too weak as a sole indicator of progress f) The lack of procedure details for a revision in 2003 g) The inadequacy of a reference to future arrangements for monitoring to be agreed through an exchange of letters h) No provision made for the eventuality if ACEA members fail to comply Parliament therefore called for an amendment to Directive 70/220/EEC to ensure an EU fuel economy fleet average of 120gCO 2 /km by 2005 with a further step to reduce average CO 2 emissions to 90gCO 2 /km 39 by 2010 (European Parliament 1997a). The NGO community also echoed many of Parliament s concerns, and foresaw, ahead of other parties, many of the shortcomings that would result from the establishment of the voluntary agreement. For example, as early as 1996, the EEB had issued a Resolution on environmental agreements (EEB 1996) which clearly warned against establishing voluntary agreements at EU level because of the Commission s inability to apply sanctions. Of the many points listed in the resolution, several were especially relevant to the ACEA agreement, including: - Council and Parliament should set environmental targets or objectives which can not be negotiated by industry, - The ultimate targets should be quantitative and long-term while intermediate targets and deadlines must be defined. (Despite having an intermediate target of 165-170gCO 2 /km for 2003, ACEA s 140gCO 2 /km target for 2008 is still an intermediate target with regard to achieving long-term environmental objectives.) - It must be possible to re-negotiate the rules and targets of the agreement after a reasonable period of time if necessary (from an environmental perspective). Tacit extension of an agreement must not be possible. The EEB closely followed the progress of the negotiations. The main message of the NGO federation, with respect to the Commission s negotiations with ACEA, was that other sectors of the passenger car sector would have to make up for the passenger car sector s shortcomings with respect to the EU s Kyoto Protocol commitment and the derogation awarded to the passenger car sector would send the wrong signals to other sectors. 39 Originally proposed by the Donnelly report on the EU Automobile industry JOC 269 16.10.95. 59

The Evaluation Prior to ACEA s second offer in March of 1998, the EEB wrote to the Commissioner for the Environment, Bjerregaard, to urge the introduction of binding limit values as ACEA had missed the opportunity to present a serious commitment in 1997 (EEB 1998a). Following the drafting of the terms of the agreement, the EEB warned in a press release in June of 1998 (EEB 1998c), that acceptance of the agreement and its assumptions would raise major doubts about the effectiveness of environmental agreements as an EU policy instrument and described the assumptions of the agreement as a direct threat to democracy. The EEB issued its final warning one month later in a press release which strongly criticised the Commission s inability to draft legislation as an alternative to the agreement and which estimated that CO 2 emissions from passenger cars could in fact rise by as much as 40% above 1990 levels in 2018 due to increasing mileage and ownership, and the 10 years it would take for ACEA s improvements to infiltrate the entire car fleet. Thus a major shortfall of the ACEA voluntary agreement was the lack of transparency and democracy as stakeholders were hardly involved and their opinions were not considered. NGOs and Parliament were bypassed, and Council was also excluded but to a lesser degree than the former. The exclusion of Council, Parliament and NGOs meant that with regard to negotiating the targets and terms of the agreements, most of the institutional power lay with the Commission. In reality, the ACEA voluntary agreement negotiations were conducted by a select few individuals who influenced the negotiations to varying and sometimes significant degrees. In addition, just as the negotiations were shown to be both positively and negatively influenced by the personalities, approaches and inter-relationships of the negotiating actors, the monitoring procedure will likewise be subject to similar influences. The co-decision procedure, which according to Article 175 of the EC-Treaty is required for the deciding of what actions are to be taken to achieve EU environmental objectives, is a far more democratic decision-making procedure in comparison to the voluntary agreement negotiations. As mentioned earlier, the situation was worsened as the Commission chose to adopt a Recommendation of the Commission and not a Recommendation of the Council and Parliament. The latter option would have allowed for the involvement of all three institutions in the development and conclusion of the agreements. 4.3 Methods to regulate CO 2 from cars The following chapter briefly discusses measures which can be used to influence motorist behaviour and to regulate vehicle fuel economy standards in order to regulate CO 2 from passenger cars. As this paper concentrates on the use of the voluntary agreement to regulate vehicle fuel economy standards, this chapter centres largely on alternative methods to the voluntary agreement to regulate vehicle fuel economy standards. 4.3.1 Economic and regulatory measures to influence the motorist There are many policy measures that can be applied to influence the motorist s behaviour and so reduce fuel consumption. Fiscal measures or economic incentives assist to internalise external costs such as climate change, congestion, accidents, noise, air pollution and infrastructure costs, to name a few. External costs to society and the environment need to be paid by motorists such that other more environmentally-friendly modes of transport can economically compete. The outcome of the debate to legislate CO 2 from cars in the early 1990s was that fiscal measures could not be used as a stand-alone measure to legislate CO 2 from cars. Fiscal policies as stand-alone measures may require high rates to be environmentally effective such 60

The Evaluation that the rates would risk being politically unacceptable. The recent backlash of motorists against high fuel prices throughout Europe reinforces this point (The Guardian 16.9.00; Ends Daily 6.9.00). Various studies show that fiscal measures can effectively influence motorist behaviour. Thus it could be beneficial to introduce harmonised Community-level fiscal measures which would also provide a level-playing field for industry, but this is unlikely to be achieved while the unanimous voting procedure continues to be used for deciding on fiscal policy. Fuel taxes have been shown to reduce fuel use and to avoid the rebound effect 40 which occurs when people drive more due to the lower cost of driving (Holman 1992; Greene 1997; Barde & Button 1990). Road pricing and congestion charging are also measures which can have a significant influence on car use and can be designed to link to car fuel economy, distance or other environmental factors (Vleuget et al and Giaoutzi et al in Barde & Button 1990). Annual circulation vehicle taxes and purchase taxes can be used to push people into more fuel efficient or environmentally optimal cars (Holman 1992; Barde & Button 1990). The various policies that can be applied to regulate or influence the purchase choices or the driving behaviour of motorists are briefly outlined in the table below (Table 7). The effectiveness of economic measures can be improved by linking them to consumer information schemes (e.g. labelling, advertising, information booklets). However, based on market trends in Sweden and the UK, fuel economy information schemes are unlikely to have an influence on the consumer s purchase decisions unless they are used in combination with adequate economic incentives or disincentives (ECMT 1997). Nevertheless, consumers can respond well to labelling schemes, which may be influenced by various factors, as proven by a study assessing consumer reaction to the impact of the EU labelling scheme for the energy efficiency of cold appliances (Winward et al 1998). Authors such as Greene (1997) argue that economic policies or fiscal measures are essential if policies are to be economically efficient but they are most efficient and effective if used in conjunction with fuel economy standards. This is the basis upon which the Commission has developed its three pillar strategy to address CO 2 from cars. The automobile industry s voluntary agreements are intended to improve the general fuel economy of the fleet through improved technology, while the fiscal framework will provide the economic incentives for the motorist and the consumer information scheme will enable consumers to make an informed choice. 40 Several studies have observed a rebound effect (i.e. people driving more due to the lower cost of driving) due to improved fuel economy after the oil shocks of the 1970s, particularly as a result of CAFÉ in the US (Greene 1997). However, the size of such an effect is uncertain (Michaelis 1996), and depends on the policies used. 61

The Evaluation Table 7 Examples of measures to regulate CO 2 from cars by influencing consumer use and choice Policy Instrument Fuel pricing Energy product tax Purchase tax Annual circulation tax Congestion charging Road pricing Insurance system reform Speed limits Inspection and maintenance Labelling and consumer information Car parking charges Driver education Application Disadvantages Advantages A tax charged on fuels such as petrol or diesel Tax added to the purchase price of a car. Can be based on fuel economy and/or other environmental parameters. To prevent the slowing of fleet renewal, inefficient cars pay a fee and efficient cars obtain a rebate. The tax levied should be non-linear to reflect the price of the car. A registration tax paid annually. Can be based on fuel economy and/or other environmental parameters. A tax charged to road users to reduce congestion. Rates can vary according to the amount of congestion, time of day, environmental performance of the car, the number of people in the vehicle etc. A tax charged to road users according to the distance driven, location, vehicle type and/or other environmental parameters. Insurance premiums adjusted to distances driven. Part of the insurance premium could be collected through the price of fuel while the other part (e.g. for theft, fire) could be paid directly to the insurance company. Applied to roads and enforced by police officers or speed cameras Programmes to ensure that ageing vehicles are properly maintained such that their environmental performance remains optimal. Labelling of cars and provision of literature in the way of posters, guides etc. Charges applied to cars for parking on the road or in car parks. Can be implemented through training courses, driving tests, literature, advertisements, videos Rates have to be very high to have a significant effect. Can be politically unpopular. Rates have to be very high to have a significant effect. Can slow the renewal of the fleet as consumers are dissuaded from purchasing new cars. Does not affect used car buyers. Rates have to be very high to have a significant effect. Can move traffic to other roads or areas. Alternative modes of mobility must also be provided to allow for a modal shift. Motorists may resist electronic charging due to privacy rights. Can move traffic to other roads or areas. Alternative modes of mobility must also be provided for to allow for a modal shift. Motorists may resist electronic charging due to privacy rights. Complex to implement as insurance contributions collected through fuel sales will have to be distributed equitably to all insurance companies. Difficult to enforce. Motorists may accelerate and decelerate in order to beat cameras, and therefore use more fuel. Difficult to implement and enforce effectively. Difficult to implement. Many different factors contribute to the purchase decision of a consumer. Rates have to be high to be effective. Car parks take up space. Difficult to implement. Incentives are needed. Can be used to curb the rebound effect. The tax can influence annual mileage, vehicle choice and driving behaviour. Can have a strong influence on the purchase choice of a new car. Can influence the purchase choice of new car buyers and used car buyers Can be effective in reducing congestion and encouraging a modal shift. Can be effective in reducing road use or encouraging a modal shift. Charge can be suited to the location, local or regional conditions. An equitable measure as it targets people who drive greater distances and are therefore responsible for greater costs to the environment, society, infrastructure deterioration etc. Provides an incentive to reduce distance driven. Vehicles consume considerably more fuel at higher speeds. Reduces the need for cars with high speed capability. Reduces accidents. Costs are relatively moderate and potential benefits are high Can help the consumer to make an informed choice and may influence the consumer. Provides an incentive for motorists to leave the car at home and to use public transport. Has significant potential to reduce fuel consumption 62

The Evaluation Figure 15 below shows the proportion of cost recovery achieved for road transport externalities in Member States for 1991, thus implying the estimates for full cost recovery. However, the methodology and assumptions used are complex and open to debate. Many costs to the environment can not always be identified, while some that can be identified can not be quantified and it may not always be possible to economically evaluate those which are quantified. It is indeed stated by the European Environment Agency (EEA) in the accompanying text to Figure 15 below, that with adequate data and methods, the use of land, solid waste generation, water pollution, fragmentation of human and animal communities, and the aesthetic impacts of infrastructure and traffic, could also be included in the externality cost estimations (EEA 2000). Despite the shortfalls, it is clear from Figure 15 that the current combination of all vehicle-related economic measures for each Member State is far from full cost recovery, with Denmark achieving the greatest internalisation of costs at 50%. As Member States are far from adequately internalising external costs to the environment and society, additional measures are clearly needed to complement fiscal measures, which can suffer from political unpopularity. Figure 15 Proportion of external and infrastructure costs covered by revenues in transport (1991) EU 15 United Kingdom Sweden Spain Portugal Netherlands Luxembourg Italy Ireland Germany Greece France Finland Denmark Belgium Austria 0 10 20 30 40 50 60 Cost recovery % Source: European Environment Agency 2000 (EEE, 1999b, using data from UIC, 1994 and ECMT, 1998) Note with chart: External costs are those that transport users inflict on others, such as noise, air pollution,accidents, climate change, congestion and infrastructure costs. With adequate data and methods, the use of land, solid waste generation, water pollution, fragmentation of human and animal communities, and the aesthetic impacts of infrastructure and traffic, could all be included (EEA 2000). 63

The Evaluation 4.3.2 Regulating vehicle fuel economy standards 4.3.2.1 The case for binding regulation as opposed to the voluntary approach While the Commission believes that the voluntary approach offers greater flexibility to manufacturers and avoids dictating technology 41, it can be argued that objectives set by both binding and non-binding legislation can be either elastic or inflexible. Flexibility depends on the objectives set, how they are set, and according to what time-frame. To the contrary, it can be argued that the voluntary agreements established with the automobile industry are a good example of how technology can be dictated by industry through non-binding commitments: This target will mainly be achieved by technological developments affecting different car characteristics and market changes linked to these developments. In particular, ACEA will aim at a high share - to the point of 90% of new cars sold being equipped with CO 2 efficient direct injection gasoline and diesel technologies. (ACEA Commitment: Commitments Point 2) The major advantage of binding legislation is that it can guarantee the achievement of environmental objectives because it is enforceable. Carefully designed binding legislation can also avoid dictating technology and can achieve environmental objectives cost-effectively (refer to discussion in chapter 4.3.2.2 and 4.3.2.3). Proponents of the voluntary approach argue that it provides incentives to the business sector for the development of efficient, innovative and environmentally-friendly solutions. However, much of today s technology for reducing CO 2 from cars, has been available for many years. In addition, much could be gained if consumers simply purchased less powerful, smaller cars with less modern conveniences. Instead, market trends are going in the opposite direction and fuel efficiency gains through technology are being offset by market shifts towards larger, faster and more powerful cars. The case of the voluntary agreements established with the automobile industry demonstrates the difficulties in applying the voluntary approach to an industry whose market trends are moving in the opposite direction to that which regulators would like to see. For example, the negotiations were dogged with events revealing lack of trust between both the Commission and the automobile industry but also between manufacturers themselves due to the competitive and thus defensive attitude of the automobile industry. The negotiations nearly came to a halt as manufacturers could not agree as to what equivalent effort would mean for each of them, and this problem is still not solved. The lack of trust was well illustrated by industry s unambitious offer in 1997, which was the result of two years of negotiations between the industry and the Commission and which was rejected immediately by all three EU institutions. As many writers claim that trust is essential in establishing effective agreements between industry and a regulating body (EEA 1997; Gouldson and Murphy 1998; Glasbergen 1998; Rennings et al 1997; Sergesen and Miceli 1998), voluntary agreements clearly need to be compatible with market forces. The Commission claims in Point 9 of its communication on environmental agreements COM(1996)561 (CEC 1996b) that the conclusion of environmental agreements can be considerably quicker than the adoption of legislation. It has taken some four years to develop and conclude the three voluntary agreements with the automobile industry. Yet the legislative process in accordance with the co-decision procedure (see Annex 2) often takes less than four years from its origins in the Commission to the final agreement between Parliament and Council. In addition, claims that agreements bear less administrative costs are also not 41 The Commission states in COM(1996)561 (CEC 1996b) that an important benefit of Environmental Agreements is that they leave greater freedom to industry at company or sectoral level to decide on how to reach the environmental targets than does legislation which prescribes or implies, for instance the use of a certain technology. 64

The Evaluation founded, as the negotiating process was a time-consuming, labour-intensive and costly exercise (see chapter 1). If CO 2 emissions from passenger cars are to be legislated in the future, the ultimate legislative compromise of the institutions is unlikely to be less stringent than that already agreed through the voluntary agreement. This is largely because the Council was in pursuit of 120gCO 2 /km by 2005 or 2010 at the latest, while Parliament was advocating 120gCO 2 /km by 2005 and 90gCO 2 /km by 2010. It was the Commission that proposed the target of 140gCO 2 /km, with the extra 20gCO 2 /km to be achieved through the fiscal framework and the consumer information scheme. In addition, the Council stated in 1996 that the three pillar strategy might not be sufficient to adequately reduce CO 2 from cars and so reserved the right to consider the need for additional measures, if necessary, which could include binding measures (Council of the European Union 1996a). In addition, the legislative procedure to reduce CO 2 from passenger cars would be according to the co-decision procedure as required by Article 175 of the EC- Treaty such that powers between the Commission, Council, Parliament would be reasonably balanced and a compromise between the institutions would have to be reached. If binding legislation is to be developed and adopted before 2008(9), when the automobile industry is to achieve its target of 140gCO 2 /km, the institutions have to their advantage that they have already discussed the issues in considerable depth, but still the Commission and Member States experts would have to agree to the legislative method to be used. The setback of the objectives will depend on exactly when the legislative instrument is deployed and on the level of ambition of the objectives. Manufacturers demand a minimum of three years for planning and the nearer to 2008 that legislation is deployed, the greater the time-frame extension may have to be beyond 2008. Some argue that consensus within Council would be difficult to achieve, in developing fuel economy legislation, due to the desire for Member State governments to protect national industries. Yet half of the Member States have no industries to protect and all EU Member States have greenhouse gas reduction commitments under the Kyoto Protocol to fulfil. There are also concerns that the presidency of Council will delay readings of legislative proposals but as the presidency rotates every six months, this is likely to prove only a temporary impediment. 4.3.2.2 Methods of regulating vehicle fuel economy Limit values can be applied to the fleet in a number of different ways. Car CO 2 emissions can be regulated by setting limits on, for example, CO 2 emissions according to a parameter such as kw engine power, per cc of engine volume, per tonne of vehicle weight, per unit of inner volume, per km etc. Alternatively CO 2 emissions can be regulated less directly through measurement of engine power per unit of engine volume or engine power per tonne of vehicle weight. But many of these options suffer shortfalls as manufacturers can construct the vehicle or tune the vehicle s performance to achieve the limit value with sometimes negative or counter-productive effects. For example, relating CO 2 emissions to vehicle weight could be counter-productive as some manufacturers may respond by increasing the weight to make room for more power (Kageson 2000). Consumer trends towards increasing vehicle performance, speed, size and power can also be controlled by capping top speeds, engine power cylinder capacity, emissions per km etc.. However, such measures are only useful in halting negative trends as they provide no incentive for cars falling below the limit or cap. While binding limit values can guarantee a specified average fuel economy for the car fleet, they suffer the disadvantage that certain car makes may have to be removed from the market 65

The Evaluation if they are unable to comply. Thus binding limit values are not popular with car manufacturers and may therefore not be regarded as politically acceptable by Member States. To overcome this problem, flexibility can be introduced into such schemes by using economic incentives which can be applied to either the consumer or the manufacturer. The consumer purchasing the car could pay a penalty for cars non-compliant with the limit value and could also receive rebates when purchasing cars depending on how much they are below the limit. Alternatively manufacturers could be provided with economic incentives through, for example, fines (as used for the US CAFE system) or under a tradeable credit permit system. The US has used its Corporate Average Fuel Economy tax (CAFE) since 1978 to set minimum standards for fuel economy. Under the US CAFE system, manufacturers must ensure that the average fuel economy of their sales meets a specified fuel economy limit. Fines of $5 per vehicle for every 0.1 mpg below the established standard are levied on manufacturers failing the limit value, but to be efficient the fine levied should exceed the extra profit gained due to selling a more fuel inefficient car. Various authors have written about the effectiveness of the CAFE standards with some arguing for its success and others claiming detrimental effects such as economic harm to manufacturers, lighter vehicles that resulted in increased traffic injuries and fatalities, a rebound effect due to increased driving and slower scrappage rates. Greene (1997) has reviewed the literature available and concludes that CAFE was effective in improving fuel economy, despite losses due to the rebound effect, which could have been curbed by economic incentives such as a fuel tax. However, it is generally thought that the CAFE system could not be as efficiently applied to European manufacturers as their fleet profiles, or rather fleet average fuel economy figures vary too much, such that manufacturers would be unevenly affected. The UK delegation to MVEG proposed a tradeable credit system to regulate CO 2 from passenger cars during the preliminary discussions in DG 11 in the early 1990s, and the Commission has since discussed the idea. Tradeable emission credits would be bought and sold among manufacturers according to whether their models met a specified and progressively tightened to a fuel efficiency standard. To prevent manufacturers from withholding credits from sale to competitors, part of the credits would be reserved for an EU authority which would be responsible for auctioning them and paying the revenue to the original credit holders. The success of a tradeable credit system would depend on creating a true competitive market and setting the correct limit values so that companies would have real incentives to improve fuel economy. However, there are drawbacks and these were raised in the MVEG group. It is feared that a tradeable permit system could be extremely complex and involve very high transaction costs. It might also be difficult to police the system and to set fines at an effective rate. Fluctuations in cost or availability of credits might also make planning very difficult and such schemes may come into conflict with WTO/GATT. However, many argue that the cost benefits could far outweigh the administrative complexities (Hockenstein et al 1997; UK Delegation to MVEG 1991). Kageson (2000) argues that the problems of market failure, such as insufficient credits reserved for auction, could be overcome if the system for dealing with credits would be of efficient design. To be politically acceptable, the design would also have to appear equitable to all manufacturers such that larger cars are not favoured more than small cars and vice versa. At the same time it can be argued that small cars should be favoured because they are inherently more fuel efficient than larger cars and they are less able to pass extra costs on to the consumer compared with manufacturers of larger cars. 66

The Evaluation During the voluntary agreement negotiations, both the Council and Parliament insisted that the Commission develop a binding legislative proposal to regulate CO 2 from passenger cars should the voluntary agreement negotiations fail but also to provide the industry with a threatening stick to assist the negotiations. Yet the Commission failed to produce such a proposal. It was not until after the rejection of ACEA s offer in 1997 that the Commission began to put some thought into such a proposal. At the Commission s workshop held in Strasbourg in February 1998, which was attended by negotiators and stakeholders with an interest in the voluntary agreement, the Commission presented five options for regulating vehicle fuel economy 42 (CEC 1998j). The options included: binding limit values graduated according to a certain parameter (e.g. engine capacity) designed to yield a fleet average fuel economy of 140gCO 2 /km and 130gCO 2 /km by 2010; indicative standards, complemented with economic disincentives for the consumer, graduated according to a specific parameter and lowered over time (fee-bate system or economic disincentive system); indicative standards plus tradeable credits to be applied such that each car falling below the standard earns credits relative to the degree by which it falls under the standard and cars above the standard similarly earn a debt of credits; a scheme whereby Member States would be recommended to ensure that by a certain date, a certain share of the fleet would consist of Enhanced Fuel-Efficient Cars (EFEC) e.g. with a fuel efficiency of less than 80g/km. The Commission s basic assessment of the various options to regulate the fuel economy of manufacturers' car sales, concluded that binding limit value legislation would be of limited cost effectiveness, be technology-fixing and would not provide incentives for overachievers. The uneven impact of binding limit values on manufacturers and certain car types was also recognised, especially as binding limit values could result in the banning of certain cars from the market. The Commission presented fee-bates, economic incentives and tradeable credits as having the potential to significantly improve the cost effectiveness of regulation compared to binding limit values, but that such measures would not necessarily guarantee a set fleet average fuel economy target. Several weeks after the Strasbourg workshop, DG 2 presented a discussion paper to a Commission ENVECO meeting 43 which showed that the potential economic cost savings of tradeable credits could be large. The studies showed that a tradeable credit system compared with the second-best alternative option of graduated limits coupled with economic disincentives for the consumer, would save between 300-800 million Euro annually. In the end, the Commission abandoned all options. The Commission believes its action is justified as efforts to investigate or establish binding alternatives would no longer be necessary as an agreement with ACEA was established. When the negotiations with JAMA became difficult, the Commission began to draft a legislative proposal (for JAMA only) which encountered legal difficulties. However, the Commission stalled the development of the legislative proposal when the agreement with JAMA was established in late 1999. As discussed by this paper, there is still a significant chance that an alternative to the voluntary agreement, in the way of a legislative proposal, may be necessary should the agreements with the automobile industry break down. In the meantime, the Commission has launched a 350 million Euro research project in early 1999 aiming to produce prototypes consuming 3 litres per 100km (90gCO2/km) by 2003 (ENDS Daily 7.1.99). This initiative is in line with the Enhanced Fuel-Efficient Car (EFEC) concept discussed at the Strasbourg workshop in February 1998. More recently, the 42 The Commission stressed clearly that the scenarios presented at the workshop were based on preliminary data and work by the Commission Services and were subject to substantial uncertainties. It was also made clear that the scenarios did not present any decisions taken by the Commission Services. 43 The paper entitled, " Flexibility for efficiency. Achieving a fuel-efficient car fleet in Europe: the potential for a tradable credit system" was presented at an ENVECO meeting, 23-24 April 1998 in Brussels. ENVECO meetings are held roughly twice a year. For these meetings, the Commission's DG ENV and ECFIN invite Environmental economists from the national ministries plus the OECD on an informal basis. 67

The Evaluation Commission launched another 5.2 million Euro programme aimed at achieving a car which can travel 100 kilometres on one litre of diesel by 2004 (ENDS Daily 26.9.00). However, the EFEC scheme can only be considered as a supplement to other measures such as those provided by the three pillar strategy. 4.3.2.3 Achieving guaranteed CO 2 emission reductions from passenger cars While binding limit value legislation may guarantee a specific fleet average fuel economy, it does not guarantee CO 2 emission reductions from the passenger car sector because of factors such as increasing car ownership and car mileage, as well as trends towards more vehicle accessories (which add more weight to the car) and driving behaviour not represented by the test-cycle. A method is likely to be needed, particularly in the future when greater emission reductions will be called for, which can guarantee greenhouse reductions. At the same time, such a method has to be politically acceptable to consumers such that the government can implement it. Due to the psychology associated with car choice, driving behaviour and marketing (Marsh & Collett 1986), many consumers often purchase a car with specifications that exceed their needs. The main reason both manufacturers and regulators are struggling to achieve fuel economy improvements, is that improving fuel economy is at odds with market trends which are in the direction of increasing sales of larger and more powerful vehicles 44. Improving fuel economy implies downsizing and extra cost for technical improvements which are passed on to the consumer for no improvement in performance (engine capacity, utility, size) and a technical feature that is not required by the consumer if there are no adequate incentives. The importance that voluntary agreements conform to the market to ensure success is pointed out by studies such as Glasbergen s studies (1998) on Dutch long-term agreements on energy efficiency improvement. Globe International advocates the contraction and convergence scenario which looks to a deep reduction in global carbon emissions, together with a convergence to national allocations based on an equal per capita entitlement (see Figure 16). Based on the idea of per capita carbon budgets aimed at reducing national carbon emissions, Fleming has developed an economic policy tool which he refers to as Domestic Tradeable Quotas (DTQs) (Fleming 1996, 1997, 1998a, 1998b). The nation implementing the scheme would set a carbon budget to be reduced over time. The carbon units making up the budget would be issued to individuals and organisations. Each individual would receive an equal and unconditional entitlement of carbon units; organisations would acquire the units they need from tender (a form of auction modelled on the issue of government debt). Low users could then sell their surplus on the national market to higher users. It should be stated at this point that the DTQ scheme is not necessarily supported by the EEB as being the solution to regulating CO 2 from cars. However, it is the author s opinion that the DTQ concept merits discussion. 44 Vans and SUVs doubled their European market share reaching 5.2% in 1998.(FT Automotive Quarterly Review 1999). 68

The Evaluation Figure 16 Contraction and Convergence Fleming s DTQ model (see Figure 17) uses carbon units defined as one kilogram of carbon dioxide (or CO 2 equivalents in the global warming potential of nitrous oxide and methane) which are issued according to a set budget. A computer database holds all the individual carbon accounts (the Register). Consumers surrender credits to the energy retailer, accessing their quota account by using their (carbon) credit card, by direct debit or by other means. The retailer then gives up carbon units when buying energy from the wholesaler. The loop is closed as the primary energy provider returns units back to the Register when the company pumps, mines or imports the fuel. Credits are issued on a rolling annual basis, with an initial issue for one year and a top-up each week. As the carbon budget is set annually with step-wise reductions to intermediate goals and longterm indicative targets, the scheme provides a long-term signal which is difficult to achieve with taxes as economic cycles are so variable. The carbon budget is more flexible than most economic measures or taxes as it provides automatic rewards or punishments depending on how effectively emissions are reduced from the sector. In addition, the DTQs should adjust to oil price rises. In response to oil price rises, it would be likely that the price of credits would reduce, as people would use less fuel, such that the net price (i.e. fuel + quota) would be relatively stable. While carbon credits would indeed be subject to daily market variations, their 69

The Evaluation price would be less prone to short-term price instability than might be expected because the price of a quota would be based on an expected long-term demand. Figure 17 The Market for Domestic Tradable Quotas Source: (Fleming 1998a). Some might argue that if consumers would make no effort to reduce emissions use at all, or if they felt they had no opportunity to reduce their emission use, that credit prices would rise rapidly and many hardship stories would unfold. The system could risk a political backlash from consumers and industry. However, any other economic instruments aimed at influencing consumer behaviour (e.g. an energy product tax) would be as politically vulnerable to failure. As already mentioned in chapter 4.3.1 (page 60), fuel taxes are currently suffering from political unpopularity. Economic incentives of various sorts would have to be much stronger than at present in all EU Member States to adequately internalise external costs to the environment and society. Thus the DTQ scheme provides an equitable and transparent alternative. Such a system should also be acceptable to industry as consumers could choose to purchase fuel-inefficient vehicles for reasons of utility or aesthetics, but their use would be limited unless savings could be made elsewhere (e.g. household electricity from renewable energies) or unless extra credits could be purchased on the market. In addition, the DTQ scheme guarantees CO 2 reductions from the passenger car sector and does not suffer from any of the perverse effects of the limit value legislative options (as explained in chapter 4.3.2.2 (page 65)) which relate emissions to one or more vehicle parameters. Based on personal responsibility, the scheme would provide the market driving forces necessary to introduce advanced technology and to drive R&D agendas more effectively because fuel efficiency would become a dominant need. In the case of the transport sector, such market forces would increase the demand for public transport and provisions or facilities for cyclists and pedestrians. The concept indeed requires further development and evaluation but its potential to efficiently reduce CO 2 emissions is clearly significant. The areas of the DTQ scheme that need to be looked at more carefully are numerous. The terms of reference for the carbon budget and its 70

The Evaluation institutional arrangements need to be further developed, as do the market structure and information technology requirements of the scheme. The economic consequences of setting a rolling ten-year quantity budget will need to be further analysed and the set-up and running costs of the scheme need to be estimated. It will also be necessary to investigate how the DTQ will inter-link with other mechanisms for reducing CO 2 emissions e.g. international emissions trading. The optimal scale of the scheme will also need to be identified as well as the most appropriate method of distributing credits among individuals and commercial entities. Further investigation will be necessary into the motivational structure of the DTQ model, the link between private motivation and public need, the issue of public acceptance, as well as the claim that DTQs fulfil the criteria of effectiveness, equity and efficiency. At the same time, additional economic measures will still be required to supplement the scheme in order to internalise costs to the environment and society that are not internalised through the DTQ scheme e.g. congestion, toxic emissions, noise, accidents, infrastructure deterioration. However, through effective application of the DTQ scheme, less fossil fuel would be burned and many of the problems associated with these aforementioned additional external costs would at the same time be considerably reduced. Governments may fear that introducing a scheme based on per capita carbon quotas would be politically unacceptable. Yet any policy aimed at regulating individual behaviour will experience public resistance to a degree. Relative to other measures, the DTQ scheme should be more politically acceptable as it is based on the principle of equity. The willingness of the public to accept policies which directly affect their behaviour, depends on society s awareness, acceptance of environmental problems, acceptance of the contribution of the individual to such problems and the preparedness of individuals to adjust their behaviour for community benefits. Figure 18 Environmental Citizenship 71

The Evaluation A developed sense of citizenship is absolutely necessary to challenge the conflict between individualism and society such that individuals are prepared to support and accept regulation intended to protect their environment and future generations. Increased environmental awareness will not necessarily mean that a person will voluntarily change their behaviour, as individual efforts may feel hopeless or insignificant. Nevertheless, individuals may be more prepared to accept regulatory measures which will require behavioural change if all persons will be equally affected, and if the result will be significant. To achieve such societal change, the various bodies which are in a position to influence an individual s outlook on life (e.g. the family, peers, religious institutions, educational institutions; see Figure 18), must build on their own ecological conscience and take on the responsibility to further develop and increase their influence. This philosophy has been explored by many writers but is well summarised by Calicott (Earth s Insights 1997: p.3): Similarly, in the environmental arena, if a mutually enhancing relationship between human civilisation and the natural environment is to evolve, environmental law and regulation must be complemented and supplemented by environmental moral sensibility and what Aldo Leopold calls an ecological conscience. 72

Bibliography 5. BIBLIOGRAPHY ACEA(1991) ACEA voluntary self-commitment on passenger car emissions/co2 October 1991. Brussels. ACEA (1992a) CO 2 emissions: The point of view of the European Vehicle Manufacturers. CVTC/25/92 Rev 1. 26 th March 1992. Brussels. ACEA (1992b) ACEA comments to MVEG on CO 2 emissions from passenger cars. VE/ACEA/5. 27 th April 1992. Brussels. ACEA (1997) ACEA newsletter No 46 October 1997 ACEA (25 th June 1999) Discussions will continue on the proposal Directive on end-of-life vehicles. http://www.acea.be ACEA (2000a) Vehicle manufacturers call for sulphur free fuels. 27 th April 2000 http://www.acea.be ACEA (2000b) ACEA answer to the Commission s Call for evidence regarding the appropriate level of sulphur content of petrol and diesel used in the Community 28.7.00 http://www.aeat-env.com/sulphur_review/sr-responses.html ACEA website (http://www. acea.be) AutoPacific website (http://www,autopacific.com) Barde, J.; Button, K. (1990) Transport Policy and the Environment: Six case studies. London: Earthscan Publications Ltd. Bell, S.; Ball, S. (1997) Bell and Ball on Environment Law: The Law and Policy relating to the Protection of the Environment, 4 th edition. London: Blackstone. Bennett, C. (6 th July 2000) Pump up those prices. The Guardian (UK) Binsbergen A.J.van, Erkens A.; Hamel B., (1994). Long Term Efficiency Improvement for Transport Technology Assessments, Faculteit Civiele Techniek, Technische Universiteit Delft, Delft, Delft, The Netherlands. Boulton, L.; Smith, M. (17 th February 1998) Oil v. autos in fight over cost of pollution: Control of car emissions is not just an environmental issue. Financial Times: Europe. Burton, J.; Simonian, H. (29 th September 1995). South Korea stacks the cards against importers - Western car-makers blame draconian barriers for puny sales, Financial Times: World Trade News. Callicott, B. J. (1997) Earth s Insights. London: University of California Press. CEC Commission of the European Communities (1991) Discussion paper: motor vehicles fuel consumption/co2 emissions. RJP/PN/eb, 27 th June 1991. CEC Commission of the European Communities (1995). A Community strategy to reduce CO 2 emissions from passenger cars and improve fuel economy. COM(95)689 final, 20 th December. Brussels. CEC Commission of the European Communities (1996a) Proposal for a European Parliament and Council Directive relating to measures to be taken against air pollution by emissions from motor vehicles and amending Council Directives 70/156/EEC and 70/220/EEC. 96/0164 (COD), Brussels. CEC Commission of the European Communities. (1996b) Communication for the Commission to the Council and the European Parliament on Environmental Agreements. COM(96)561, 27 th November. Brussels. 73

Bibliography CEC Commission of the European Communities. (1997a) Information note by the Commission Services for the Environment Council on 16.12.1997. Agenda item 3C.: CO 2 from cars. Brussels. CEC Commission of the European Communities. (1997b) Information note by the Commission Services for the Environment Council on 16.10.1997. Agenda item 9D.: CO 2 from cars. Brussels. CEC Commission of the European Communities. (1997c) Report on the Commission s Discussions with the Automobile Industry on an Agreement to Reduce CO 2 Emissions from Passenger Cars. Staff working paper SEC(97)176, 28 th January. Brussels. CEC Commission of the European Communities (1998a) Communication from the Commission to the European Parliament. Climate Change - Towards an EU Post-Kyoto Strategy. COM(1998)353 final, 3 rd June 1998. Brussels. CEC Commission of the European Communities. (1998b) Communication for the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions on Transport and CO 2 : Developing a Community Approach. COM(98)204 final, 31 st March. Brussels. CEC Commission of the European Communities.(1998c) Proposal for a Council Decision establishing a scheme to monitor the average specific emissions of carbon dioxide from new passenger cars. COM(98)348 final, 12 th June. Brussels. CEC Commission of the European Communities. (1998d). Implementing the Community Strategy to Reduce CO 2 from Cars: An Environmental Agreement with the European Automobile Industry. COM(98)495 final, 29 th July. Brussels. CEC Commission of the European Communities (1998e) Proposal for a Council Directive relating to the availability of consumer information on fuel economy in respect of the marketing of new passenger cars, COM(1998)489 final. Brussels. CEC Commission of the European Communities and ACEA (1998f) CO 2 Emissions from Cars: The EU Implementing the Kyoto Protocol. Pocket booklet. CEC Commission of the European Communities (1998g) EU Transport in Figures: Statistical Pocketbook, DG7, Eurostat. CEC Commission of the European Communities. (1998h) Environmental Agreement with the European Automobile Manufacturers Association (ACEA) on the Reduction of CO 2 from Passenger Cars. Staff working paper SEC(1998)1047, 19 th June. Brussels. CEC Commission of the European Communities (1998i) Workshop on CO 2 emissions from passenger cars - Strasbourg, 18 th February 1998. Internal document, Brussels. CEC Commission of the European Communities (1998j) Discussion Paper: Binding CO 2 limit values as a basis for a re-orientated CO 2 /cars strategy. 13 th February, Brussels. CEC Commission of the European Communities (1998k) Reduced CO 2 emissions from cars: Commission acknowledges new proposal from ACEA. Commission press release, 11 th March 1998, Brussels. CEC Commission of the European Communities (1998l) Commission outlines measures to reduce carbon dioxide emissions from transport. Commission press release, 31 st March 1998, Brussels. CEC Commission of the European Communities (1998m) CO 2 emissions from cars: New proposal from industry. Commission, press release, 3 rd June 1998, Brussels. CEC Commission of the European Communities (1998n) Commission proposes Directive on fuel-economy labelling. Porte Parole press release, IP/98/789. 4 th September 1998, Brussels. 74

Bibliography CEC Commission of the European Communities (1998o) Commitment by European Car Manufacturers to Reduce CO 2 Emissions from Passenger Cars Complies with EU Competition Rules. Commission press release, 6 th October 1998, Brussels. CEC Commission of the European Communities (1999a) Communication from the Commission to the Council and the European Parliament: Preparing for Implementation of the Kyoto Protocol, COM(1999)230final, 19 th May, Brussels. CEC Commission of the European Communities. (1999b) Commission Recommendation of 5 th February 1999 on the reduction of CO 2 emissions from passenger cars. 1999/125/EC, OJ L 40, p49, Brussels. CEC Commission of the European Communities (1999c) Implementing the Community Strategy to Reduce CO 2 from Passenger Cars: Progress Report on the Negotiations with JAMA and KAMA. Staff working paper SEC(1999)364, 3 rd March. Brussels. CEC Commission of the European Communities (1999d) Implementing the Community Strategy to Reduce CO 2 from Passenger Cars: State of Play of the Negotiations with Japanese and Korean Automobile Industries. Staff working paper SEC(1999)1004, 23 rd June. Brussels. CEC Commission of the European Communities. (2000a) Commission Recommendation of 20 th April 2000 on the reduction of CO 2 emissions from passenger cars. 2000/303/EC, OJ L 100, P55, Brussels. CEC Commission of the European Communities. (2000b) Commission Recommendation of 20 th April 2000 on the reduction of CO 2 emissions from passenger cars. 2000/304/EC, OJ L 100, P57, Brussels. CEC Commission of the European Communities (2000c) Decision 2000D1753 establishing a scheme to monitor the average specific emissions of carbon dioxide from new passenger cars OJ L202, 10.08.00, p 1, Brussels. CEC Commission of the European Communities (2000d) Auto-Oil II: A report form the services of the European Commission. Clarke, R. A. (1994) The Challenge of Being Green. Harvard Business Review, July-August 1994 Concawe (2000) EU commission call for evidence on ultra-low sulphur fuels: CONCAWE response 31.7.00. http://www.aeat-env.com/sulphur_review/sr-responses.html Council of the European Union (1989) Council Directive 89/458/EEC of 18 th July 1989 amending with regard to European emission standards for cars below 1,4 litres, Directive 70/220/EEC on the approximation of the laws of the Member States relating to measures to be taken against air pollution by emissions from motor vehicles. 89/458/EEC. Brussels: Official Journal of the European Communities. Council of the European Union (1991) Council Directive amending Directive 70/220/EEC on the approximation of the laws of the Member States relating to be taken against air pollution by emissions from motor vehicles 91/441/EEC, 26 th June 1991. Brussels: Official Journal of the European Communities. Council of the European Union (1992) Conclusions of the Presidency: Edinburgh European Council, 11-12 th December 1992. Council of the European Union (1993) Commission Directive 93/116/EC of 17 th December 1993 adapting to technical progress Council Directive 80/1268/EEC relating to the fuel consumption of motor vehicles. 93/116/EC. Brussels: Official Journal of the European Communities. Council of the European Union. (1996a) Council Conclusions 25 th June 1996. Brussels. 75

Bibliography Council of the European Union. (1996b) Council Conclusions in Response to COM(95)689 final. No. 8748/96, 2 nd July 1996. Brussels. Council of the European Union (1997a). Council Resolution on Environmental Agreements, agreed at the June Environmental Council meeting and adopted on 7 th October 1997. Council of the European Union. (1997b) Council Conclusions 16 th December 1997. Brussels. Council of the European Union. (1998a) Council Conclusions 23 rd March 1998. Brussels. Council of the European Union. (1998b) Council Conclusions 17 th June 1998. Brussels. Council of the European Union. (1998c) Council Conclusions in Response to COM(98)495. No. 11802/98, 7 th October 1998. Brussels. Council of the European Union (1998d) Directive 98/70/EC of the European Parliament and of the Council of 13 th October 1998 relating to the quality of petrol and diesel fuels and amending Council Directive 93/12/EEC, 98/70/EC. Brussels: Official Journal of the European Communities. Council of the European Union (2000) Decision 1753/2000/EC of the European Parliament and of the Council establishing a scheme to monitor the average specific emissions of CO2 from new passenger cars. Dargay, J.; Gately, D.. (1997) Vehicle ownership to 2015: implications for energy use and emissions. Energy Policy, Vol 25: (14-15): 1122-1123. UK: Elsevier Science Ltd.. DeCicco, J.; Ross, M.. (1993) An Updated Assessment of the Near-Term Potential for Improving Automobile Fuel Economy. Washington D.C., USA: American Council for an Energy-Efficient Economy De Cicco, J. M. (1993) Recommendations to the Interagency Climate Change Mitigation Group for Reducing Greenhouse Gas Emissions for Light Cars and Trucks. Washington DC: American Council for an Energy Efficient Economy. DETR (1998) UK Climate Change Programme: Consultation Paper. Norwich: HMSO DETR (1999) New Car Fuel Consumption and Emission Figures. Bristol: Vehicle Certification Agency. DETR (2000) UK initial information in response to the European Commission s call for evidence on the benefits of reducing sulphur in petrol and diesel to less than 50ppm 28.7.00 http://www.aeat-env.com/sulphur_review/sr-responses.html DG ENV website for sulphur review responses http://www.aeat-env.com/sulphur_review/sr-responses.html Dockery DW, Pope CA, Xu X, Spengler JD, Ware JH, Fay ME, Ferris BG, Speizer FE, (1993). An association between air pollution and mortality in six US cities. N Engl J Med 329(24): 1753-1759. Eamonn Bates Issue Tracker (January 2000) Environmental Agreements. Brussels: Eamonn Bates Environment. ECMT European Conference of Ministers of Transport, (1996) Monitoring of Specific Fuel Consumption and CO2 Emissions of New Cars. CEMT/CS/ENV/IND(96)4. Paris: European Conference of Ministers of Transport. ECMT European Conference of Ministers of Transport, (1997) CO2 emissions on transport. Paris: OECD Publications. EEA European Environment Agency (1997) Environmental Agreements: Environmental Effectiveness. Environmental Issues Series No 3.Copenhagen: EEA EEA European Environment Agency (2000) 76

Bibliography EEB European Environmental Bureau (19 th November 1996) EEB resolution on environmental agreements. Brussels: EEB EEB European Environmental Bureau (1998a) EU Commission has negotiated an agreement with car manufacturers which will exempt them from action according to Kyoto agreement. Brussels 6.5.98: EEB EEB European Environmental Bureau (1998b) EEB s reaction to ACEA s proposal on CO2 reductions from new passenger cars: Reject! Let the car industry not dictate the terms! Brussels 8.6.98: EEB EEB European Environmental Bureau (1998c) EU Commission should not accept disgraceful offer of the car industry on CO2 emissions: No treachery to Kyoto agreement! Brussels 28.6.98: EEB EEB website (http://www.eeb.org) ENDS Daily (13 th March 1997) ENDS Daily (16 th October 1997) EU Commission blasts car-makers CO 2 offer. ENDS Daily (18 th November 1997) EU car-makers stick to carbon dioxide offer ENDS Daily (19 th February 1998) EU car-makers to make new CO 2 offer. ENDS Daily (11 th March 1998) EU car firms to make new fuel efficiency offer. ENDS Daily (20 th April 1998) Fiat pledges to raise vehicle fuel efficiency. ENDS Daily (24 th July 2000) EU car firms making progress on CO2 pledge. ENDS Daily (6 th September 2000) France backtracks on fuel escalator. ENDS Daily (26 th September 2000) EU attempts to re-take the initiative on fuel. ENDS Daily (6 th October 2000) Carmakers need to do more on CO2 emissions. ENDS Daily (12 th October 2000) Sulphur-free fuel seen increasing CO2 emissions. Environment Watch: Western Europe (27 th October 1995) EU Commission prepares to tackle car CO 2 emissions. Environment Watch: Western Europe (4 th October 1996) French Carmakers in Fuel Economy Commitment. Environment Watch (15th September 2000) Voluntary agreement could pre-empt EU marine safety laws. European Parliament (1997a) Session documents: Report on the Communication from the Commission to the Council and the European Parliament on COM(95)689. PE221.181/fin, 25 th March. Brussels. European Parliament (1997b). Resolution on the Commission Communication to the Council and European Parliament on Environmental Agreement. A4-0224/97, July 1997. European Parliament (1998) Parliament s Resolution on Climate Change in the Run-Up to Buenos Aires (November 1998). Point 11. Brussels. Europe Environment (7 th January 1992) Motor Industry: ACEA wants 10% cut in CO 2 emissions and more effective recycling. No 378. Europe Environment (9 th July 1996) European Parliament Dissatisfied with CO 2 emissions from cars. No 281. Europe Environment (23 rd June 1998.) Climate Change: Motor Industry Commitment to Reduce CO 2 emissions. No 525. 77

Bibliography Europe Environment (4 th May 2000) Air Quality & Climate: Vehicle manufacturers call for sulphur free fuels. No 567. Europe Environment (1 st June 2000) Fuel Quality: Commission launches new round of consultations on improving fuels. No 569. EU website (http://europa.eu.int) Financial Times (5 th March 1996) Growth runs into a jam - can collaborative deals help solve the problem of too many makers and too little demand? Survey of World Motor Industry Financial Times, Europe (15 th January 1998) Survey of the world motor industry: Growth runs into a jam. Financial Times, Europe (4 th June 1999) EU petroleum industry losing sulphur battle. Fleming, David (1996), Stopping the Traffic, Country Life, vol 140, 19; 9 May, pp 62-65. Fleming, David (1997), Tradable Quotas: Using Information Technology to Cap National Carbon Emissions. European Environment, 7, 5, Sept-Oct. pp 139-148. Fleming, David (1998a), Domestic Tradable Quotas as an Instrument to Reduce Carbon Dioxide Emissions, European Commission, Proceedings, Workshop 1-2 July, EUR 18451. Fleming, David (1998b), Your Climate Needs You, Town & Country Planning, 67, 9, October, pp 302-304). Fleming, David (1999), Decoding a Message about the Market for Oil, European Environment, 9, 3, May-June. FT Automotive Quarterly Review (1999), Financial Times Business Ltd. London. Ghio A. J.; Samet J. M. (1999) Metals and air pollution particles, Air Pollution and Health p635-651. Holgate ST et al (Eds). San Diego: Academic press. Giaoutzi, M.; Damianides, L. (1990) Chapter 6: Greece in Transport Policy and the Environment: Six case studies edited by J. Barde and K. Button. London: Earthscan Publications Ltd. Glasbergen, P. (1998) Learning to Manage Energy by Voluntary Agreement: The Dutch Long-Term Agreements on Energy Efficiency Improvement. Green Management International, 22, Summer 1998. Global Commons Institute website http://www.gn.apc.org/gci/ Gouldson, A.; Murphy, J. (1998) Regulatory Realities. London: Earthscan Publications Ltd. Greene, D. L. (1997). Why C.A.F.E. worked. Oak Ridge National Laboratory (Tennessee), US Department of Energy. Greenpeace (August 1996) Why doesn t the car industry smile? London: Greenpeace Greenpeace website (http://www.greenpeace.org) Hare, W.L. (1997) Fossil fuels and climate protection: The Carbon Logic Greenpeace International Hawthorne, M.; Alabaster, T. (1999) Environmental Citizenship, Global Environmental Change 9: 25-43. Hockenstein, J.B.; Stavins, R. N.; Whitehead, B. W. (1997) Crafting the next generation of market-based tools. Environment, 39(4). Holman C. (1992) Car fuel consumption. Brussels: EEB. Horspool, M. (1998) European Union Law. London: Reed Elsevier (UK) Ltd. 78

Bibliography IPCC Intergovernmental Panel on Climate Change (1996) IPCC Working Group 1: Climate Change 1996-2 nd Assessment Report of the IPCC. Cambridge: University Press. Italian Delegation to MVEG (1991) Italian Delegation Proposals for Limiting CO 2 Emissions/Fuel consumption of Passenger Cars in Europe. 18 th July, 1999. J. D. Power website (http://www.jdpower.com) Jorgensen, P.; Mikkelsen, L. (11 th March 1998) Reduced CO 2 emissions from cars: Commission acknowledges new proposal from ACEA. Commission internal press release, IP/98/234. Brussels. Kageson, P (1998) Cycle-beating and the EU Test Cycle for Cars. European Federation for Transport and Environment. T&E 98/3. Brussels. Kageson, P (2000) The Drive for Less Fuel: Will the motor industry be able to honour its commitment to the European Union?. European Federation for Transport and Environment. T&E 00/1. Brussels. Kampfner, J. (16 th February 1998) Minister to warn carmakers on emissions. Financial Times: News (UK). Khalastchi, R; Ward, H.. (1998) New instruments for sustainability: an assessment of environmental agreements under Community Law. Journal of Environmental Law 10(2): 251-290. Oxford: Oxford University Press. Koopman, G. J. (1997) Long-term challenges for inland transport in the European Union: 1997-2010: Consequences for transport fuel economy and use. Energy Policy, 25(14-15):1151-1161. UK: Elsevier Science. Lovins, AB; Brylawski, M. M.; Cramer D. R.; Moore T. C. (1996) Hypercars: Materials, Manufacturing and Policy Implications. Colorado: Rocky Mount Institute. Marsh, G.; Hill, N.; Sully, J. (2000) Consultation on the need to reduce the sulphur content of petrol and diesel fuels below 50ppm: A Policy Maker s Summary, AEA Technology, November 2000. http://www.aeat-env.com/sulphur_review/sr-responses.html Marsh, P; Collett, P. (1986) Driving Passion: The Psychology of the Car. London: Jonathon Cape Ltd. Martin D.J., Michaelis L.A. (1992) Research and technology strategy to help overcome environmental problems in relation to transport, Commission of the European Communities, EUR-14713-EN, Brussels. Michaelis, L. A.. (1996) Policies and measures for Common Action - Sustainable Transport Policies - CO2 Emissions from Road Vehicles. Paris: OECD. Moore, T. C.; Lovins, A. B.. (1996) Vehicle design strategies to meet and exceed PNGV goals update to SAE Paper No. 951906, 3/96. Colorado: Rocky Mount Institute. Pallemaerts, M. (1999) The decline of law as an instrument of Community environmental policy. R.A.E.-L.E.A. 1999. Mys & Breesch. Nakamoto, M. (25 th January 1998) Japan carmakers renew their European offensive. Financial Times: World Trade News. Netherlands Delegation to MVEG (1991) Development of a Policy on the Reduction of CO 2 Emissions. 10 th January, 1999. OTA, US Office of Technology Assessment (1995) Advanced Automotive Technology: Visions of a super efficient Family Car. Washington DC: US Government Printing Office. Owen, D. (4 th August 1998) France re-assesses whether its love of cars is diesel-fired: New tax proposals may cause owners to examine the economy of running a diesel vehicle. Financial Times: World News, Europe. 79

Bibliography Phylipsen, G. J. M.; Blok, K.. (1996) Common European Policies and Measures for Greenhouse Gas Emission Limitation and Reduction. Department of Science, Technology and Society, Utrecht University. Plotkin, S. E.; Greene, D.. (1997) Prospects for improving the fuel economy of light-duty vehicles. Energy Policy, 25(14-15): 1179-1188. UK: Elsevier Science Ltd.. Pope CA III, Thus MJ, Namboodiri MM, Dockery DW, Evans JS, Speizer FE, Heath CW, (1995). Particulate air pollution as a predictor of mortality in a prospective study of US adults. Am JRepir Crit Care Med 151:669-674. Porter, M.; Van de Linde C.; (1995a) Green and Competitive: Ending the Stalemate. Harvard Business Review, September-October 1995. Porter, M.; Van de Linde C.; (1995b) Towards a new conception of the environment - competitiveness relationship. Journal of Economic Perspectives, 9(4): 97-131 Poulton, M. L. (1997) Fuel Efficient Car Technology. Southampton: Computational Mechanics Publications. RCEP Royal Commission on Environmental Pollution (1994) Transport and the Environment, 18 th Report. London: HMSO. Rehbinder, E. (1997) Environmental Agreements - a new instrument of environmental policy. Environmental Policy and Law, 27(4): 258-269. UK: Elsevier Science Ltd. Rennings, K.; Brockmann, K. L.; Bergmann, H. (1997) Voluntary agreements in environmental protection: Experiences in Germany and future perspectives. Business Strategy and the Environment, 6(5); 245-263 Schipper, L. (1995) Determinants of Automobile Use and Energy Consumption in OECD Countries. Annual Review of Energy and Environment, pp325-386. Segerson, K; Miceli, T.J. (1998) Voluntary environmental agreements: Good or bad news for environmental protection. Journal of Environmental Economics and Management, 36(2): 109-130 Simonian, (20 th August 1998) Clean Air cost will fall on industry as well as carmakers: The car industry has persuaded EU policymakers that the oilmen must share the cost of cutting pollution. Financial Times: World News. Simonian, H. (29 th June 1999) WW links come back to haunt Schroder. Financial Times: World News, Smith, M. (16 th June 1999) Row rages over waste of a car. Financial Times: Inside Track. Speizer, F. E. (1994) Air pollution and lung cancer, Epidemiology of Lung Cancer, J.M. Samet ed.; University of New Mexico School of Medicine and the New Mexico Registry, Ambuquerque, NM. The ENDS Report 207 (April 1992) Competing options for EC curbs on CO 2 emissions from cars. The ENDS Report 215 (December 1992) Deadlock over CO 2 controls on cars. The ENDS Report 216 (January 1993) Proposals for EC car emissions strategy beyond 2000. The ENDS Report 225 (October 1993) Commission given new ideas on CO 2 taxes on cars. The ENDS Report 252 (January 1996) Lengthy EC negotiations ahead on CO 2 strategy for cars. The ENDS Report 259 (August 1996) Greenpeace s fuel efficient car poses challenges to vehicle industry. 80

Bibliography The Federal Republic of Germany (2000) Comments of the Federal Republic of Germany on the updating of the EU Fuels Directive 98/70 regarding the sulphur content of fuels. 31.05.00 http://www.aeat-env.com/sulphur_review/sr-responses.html The Netherlands (2000) Consultation on the sulphur content of petrol and diesel fuels http://www.aeat-env.com/sulphur_review/sr-responses.html UK Delegation to MVEG (1991) Marketable permits: Variants for the Transport Sector. London: Department of Transport. Vleuget, J.; van Gent H.; Nijkamp. P (1990) Chapter 5: The Netherlands in Transport Policy and the Environment: Six case studies edited by J. Barde and K. Button. London: Earthscan Publications Ltd. Wade, J.; Holman, C.; Fergusson, M. (1994) Passenger car global warming potential: Current and projected levels in the UK. Energy policy 22(6): 509-522. UK: Butterworth-Heinemann Ltd.. Wells, P.; Nieuwenhius, P. (1997) The Death of Motoring. Chichester: John Wiley. What Car? (1999) What Car? Green Guide. July 1999. Waldegrave, UK: Haymarket Motoring Publications Ltd. Whitelegg, J. (1993) Transport for a sustainable future: The case for Europe. London: Belhaven Wilson R.; Spengler J. D. (eds) (1996) Particles in our Air. Concentrations and health effects. USA: Harvard University Press. Winward, J;; Schiellerup, P.; Boardman, B. (1998) Cool labels: the first three years of the European Energy label. Oxford, UK: Environmental Change Institute (ECI). 81

Appendix 6. ANNEX 1 The ACEA Commitment Figure 19 The ACEA Commitment ACEA Commitment on CO 2 emission reductions from new passenger cars in the framework of an environmental agreement between the European Commission and ACEA Introduction and Principles 1) This Commitment is based on an undertaking by ACEA itself and has the support of all car manufacturing companies: BMW, Fiat, Ford of Europe, GM Europe, Daimler-Benz, Porsche, PSA, Peugeot Citroen, Renault, Rolls-Royce, Volkswagen and Volvo, who have agreed to make every endeavour to contribute to the achievement of ACEA s goals. This Commitment demonstrates ACEA s support for significant reductions in CO 2 emissions in line with the European Union s undertakings under the United Nations Framework Convention on Climate Change following the Kyoto Conference. At the same time it aims at preserving the diversity of the product offerings of the European car manufacturers and at maintaining their competitiveness, as well as their financial performance and employment. 2) As long as its commitments (see below) are being honoured, ACEA is assuming that this Commitment provides complete and sufficient substitute for all new regulatory measures to limit fuel consumption or CO 2 emissions, and for any additional fiscal measures in the pursuit of the CO 2 objectives of this Commitment. Any fiscal measures, including their added value to this Commitment, will be taken into account in the monitoring procedure and their potential effects will be assessed in good faith. 3) The European automotive industry s CO 2 reduction commitments are very ambitious in the light of present and future technologies, and the industry is willing and prepared to commit substantial development efforts to implement the following commitments. 4) Together with the European Commission, ACEA will ensure that the Commitment is implemented in a manner which complies with applicable competition rules. ACEA COMMITMENTS 1) Some members of ACEA will introduce in the EU market, no later than 2000, models emitting 120g CO 2 /km or less, measured according to Directive 93/116/EC (see Technical Annex, Point 1 Measuring Procedure). 2) ACEA commits to achieve a target of 140gCO 2 /km by 2008, measured according to Directive 93/16/EC, on the average of the EU new car sales represented by ACEA classified as M1. This target will mainly be achieved by technological developments affecting different car characteristics and market changes linked to these developments. In particular, ACEA will aim at a high share - to the point of 90% - of new cars sold being equipped with CO 2 efficient direct injection gasoline and diesel technologies. Compliance with this target translates for the European automobile industry into an average CO 2 reduction of 25% for newly registered cars, compared to 1995. 3) In 2003, ACEA will review the potential for additional CO 2 reduction, with a view to moving further towards the Community s objective of 120gCO 2 /km by 2012 4) For 2003, ACEA considers an estimated target range of 165-170g of CO 2 /km to be appropriate. This translates into a reduction of 9-11% compared to the reference year 1995. (See Technical Annex, point 3: review in 2003/Estimated Target Range). 6) To assess compliance with these commitments, there will be a joint ACEA/Commission monitoring of all the relevant factors with regard to these commitments. 82

Appendix ACEA s commitments are based on the following: A) Availability of enabling fuels Given the outstanding importance of improved fuels for CO 2 reductions ACEA assumes the full market availability of fuels with a sufficient quality to enable the application of technologies needed for the industry to achieve its CO 2 commitments during the lifetime of this Commitment (s. Technical Annex, Point 2 Fuel Specifications). B) Distortion of competition In order to ensure a level playing field: - non-acea member car manufacturers will be committed to equivalent CO 2 reduction efforts for there sales in the EU, in line with the Council Conclusions of 25.6.96 - the Community will use its best efforts to continue to seek other car manufacturing countries, notably Japan, North America and Korea, will undertake equivalent car CO 2 reduction efforts, in line with the Kyoto Protocol spirit ensuring that the European automobile industry is not put at a competitive disadvantage in world markets by CO 2 reduction commitments in Europe. C) Promotion of car CO2-efficient technologies European car manufacturers have high expectations for certain technologies, in particular those associated with direct injected gasoline and diesel engines, which are two of the most promising routes to achieve the central commitment of 140gCO 2 /km in 2008. This commitment is based on the assumption of an unhampered diffusion of car 2 efficient technologies into the market via competition amongst ACEA members and other market participants which is expected to result in market mix changes. Therefore it is fundamental that any measures that might hamper the diffusion process of either of the CO 2 efficient technologies will be taken into consideration in the monitoring procedure. D) Acceptance of innovations The acceptance by the Commission of innovative concepts for vehicles replacing conventional cars in short haul traffic and of cars not producing fossil CO 2 as well as a share of cars using alternative fuels or propulsion systems as contributing factors to comply with the Commitment. MONITORING The joint ACEA/Commission monitoring procedure should cover: 1) The development of CO 2 emissions based on the collective achievement of reductions on the average EU fleet of new car sales represented by ACEA and according to the above commitments. 2) The development of the CO 2 emissions of non-acea car manufacturers for their sales in the EU. 3) Any developments regarding the underlying factors upon which ACEA s Commitment is based. 4) the impact on CO 2 emissions of new regulatory measures 5) the development of new breakthrough technologies (e.g. natural gas, hydrogen, fuel cells, electric drive), which might be available for production in the next decades, and the impact of the Community s 5 th R&D framework programme, which is expected to foster research in this area. 6) The development and promotion of other measures deemed to reduce fuel consumption i.e. telematics and optimisation of the infrastructure reducing congestion; driver education for fuel efficient behaviour; driver information on fuel efficiency. 7) the impacts on the financial performance, competitiveness and employment within the European automotive industry associated with this Commitment. The Commission s official reports on the monitoring results will not refer to individual companies achievements, to avoid competition being distorted. ACEA is willing to provide the necessary data to achieve the objectives of the monitoring. On the basis of the outcome of the monitoring, or if the impacts of this Commitment, on the European automotive industry, particularly its employment situation and its global competitive environment are detrimental, ACEA and the Commission will review the situation and make any necessary adjustments in good faith. 83

Appendix TECHNICAL ANNEX TO THE ACEA COMMITMENT ON CO2 EMISSION REDUCTION FROM NEW PASENGER CARS Point 1) Measuring Procedure ACEA s proposals have been established according to Directive 93/116/EC, which has been fully implemented as from 1.1.1997, and will be applicable for the coming years. The implementation of this new measuring procedure has led to an artificial average increase of 9% of the CO 2 emission figures, compared to the previously used directive, whereas the CO 2 emissions from cars in the real world have not changed. Point 2) Fuel Specifications Characteristics of the fuels are key factors in car CO 2 emission reductions: A) to achieve further emission reduction together with lowered CO 2 emissions the fuel efficient lean burn technology will be combined with special exhaust gas after-treatment devices capable to reduce NOx under lean burn conditions. But those systems are only working with fuels meeting specific requirements, in particular a low sulphur content; B) low sulphur fuels ease the NOx/CO 2 trade-off in favour of CO 2 emission reductions; C) low aromatics in gasoline and a high cetane number in diesel lead to CO 2 emission reduction too. ACEA acknowledges the outcome of the conciliation procedure between the Council and the European Parliament on 26.6.1998 and upholds its 140gCO 2 /km commitment by 2008. However, ACEA is expecting that fuels of the following better quality might be available in the market due to technical reasons, commercial competition as well as possible national policies: A) Some gasoline (e.g. Super-Plus, 98 octane as agreed in Germany) and some diesel plus with a maximum sulphur content of 30ppm are provided in 2000 on the whole EU market in a sufficient volume and geographical cover. B) In 2005 full availability of fuels on the whole EU market which satisfy the following: gasoline with a maximum sulphur content of 30ppm and of a maximum aromatic content of 30%; diesel with a maximum Any problems which might arise sulphur content of 30ppm and a cetane number of minimum 58. with respect to fuel quality will be considered in the monitoring procedure. Point 3) Review in 2003/Estimated Target Range ACEA is willing to contribute to a periodic monitoring of its commitments, jointly undertaken by ACEA and the Commission, which it sees as the main tool to examine the evolution during the period of the Commitment. This should include a joint Major Review in 2003, covering both ACEA and non-acea developments. This would incorporate he results of CO 2 emission reductions up to and including calendar year 2003, including comparison of that year s fleet average to the estimated target range. The reduction in CO 2 emissions will not be linear; the pace will notably depend on the timing of availability of the enabling fuels on the market as well as on the lead-times for new technologies and products and their market penetration. The reduction profile is therefore expected to be relatively slow initially and gather pace later. Given all the uncertainties and the lead-time necessary for introducing new technologies and models, ACEA considers and appropriate estimated target for 2003 to be within the range of 165-170gCO 2 /km. This is a reduction of 9-11% compared to the 1995 reference year. ACEA provides this estimated target range for 2003 on the following basis: A) it does not constitute a commitment of any sort by ACEA B) the provisions set out under Monitoring are fully implemented and any necessary adjustment to the 2008 commitment or the 2003 estimate are made in good faith C) in particular, fuels of sufficient quality are available - such that fuel issues so not constrain the application of technologies needed to improve fuel efficiency (see Point 2 above: Fuels specifications). 84