Climate Policy after Copenhagen: The Role of Carbon Pricing

CLIMATE POLICY AFTER COPENHAGEN At the UN climate negotiations in Copenhagen, 117 heads of state concluded that low-em...

Author: Karsten Neuhoff

34 downloads 755 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

CLIMATE POLICY AFTER COPENHAGEN

At the UN climate negotiations in Copenhagen, 117 heads of state concluded that low-emissions development is necessary in order to combat climate change. However, they also understood that transition to a low-carbon economy requires the implementation of a portfolio of policies and programmes – a challenging endeavour for any nation. This book addresses the need for information about factors impacting on climate policy implementation, using as a case study one effort that is at the heart of attempts to create a low-carbon future: the European Union emissions-trading scheme (ETS). It explores the experience with the implementation of the ETS, including the role of vested interests, the impact of design details and opportunities to attract long-term investments. It also discusses how international climate co-operation can support the domestic implementation of low-carbon policies. This timely analysis of carbon pricing contains important lessons for all those concerned with the development of postCopenhagen climate policy. Karsten Neuhoff is Director of the Berlin ofﬁce of Climate Policy Initiative, a global research organisation whose mission is to assess, diagnose and support the efforts of nations to achieve low-carbon growth. He is also Research Director for Climate Policy Impact and Industry Response at the German Institute for Economic Research (DIW Berlin). He was previously an economist at the University of Cambridge leading climate policy and energy research projects, and worked with Climate Strategies on projects related to the European Union emissions-trading scheme.

Climate Policy after Copenhagen THE ROLE OF CARBON PRICING

Karsten Neuhoff

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Tokyo, Mexico City Cambridge University Press The Edinburgh Building, Cambridge cb2 8ru, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/ 9781107401419 © Karsten Neuhoff 2011 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2011 Printed in the United Kingdom at the University Press, Cambridge A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication data Neuhoff, Karsten. Climate policy after Copenhagen : the role of carbon pricing / Karsten Neuhoff. p. cm. isbn 978-1-107-00893-9 (hardback) 1. Emissions trading. 2. Climatic changes – International cooperation. 3. Environmental protection – International cooperation. I. Title. hc79.e5n447 2011 363.7380 746–dc22 2011004249 isbn 978-1-107-00893-9 Hardback isbn 978-1-107-40141-9 Paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

For Julia

Contents

List of ﬁgures List of tables List of boxes Acknowledgements List of abbreviations

page ix xii xiii xiv xv

1 Introduction

1

2 2.1 2.2 2.3 2.4 2.5

The role of a climate policy mix Putting a price on carbon The role of technology policy The role for targeted measures and regulation Managing distributional implications Conclusion

19 20 29 41 45 52

3 3.1 3.2 3.3 3.4 3.5 3.6

Implementing a carbon price: the example of cap and trade The SO2 trading programme in the USA The European Union emissions-trading scheme Setting the cap: too many cooks spoil the broth Distributing allowances: compensate or distort Sectoral coverage of a carbon-pricing scheme Conclusion

56 58 61 65 72 86 93

4 Shifting investment to low-carbon choices 4.1 The nature of uncertainty vii

97 98

viii

4.2 4.3 4.4 4.5 4.6 4.7

*

Contents

Response to uncertainty with taxes and cap-and-trade schemes Investment under uncertainty: contrasting different perspectives Addressing requirements of strategic investors Addressing requirements of project investors Addressing the needs of ﬁnancial investors Conclusion

100 105 109 115 125 129

5 Co-operation among developed countries: a role for carbon markets? 5.1 Using international co-operation to enhance domestic commitment 5.2 Transparent monitoring and reporting 5.3 Carbon-market-based international co-operation among developed countries 5.4 The economics of carbon-market-based co-operation mechanisms 5.5 The political economy of carbon-market-based instruments 5.6 A global carbon tax 5.7 Conclusion

148 151 155 158

6 6.1 6.2 6.3 6.4 6.5 6.6

162 164 173 178 181 185 198

A world of different carbon prices Screening for high carbon costs Do international cost differences matter? Dimensions of trade Corporate strategy: the longer-term view The industry value chain: leakage versus substitution effect Policy options to address leakage Conclusion

7 International support for low-carbon growth in developing countries 7.1 Framework for international co-operation 7.2 Financial needs for low-carbon development 7.3 The role of carbon markets to provide support for developing countries 7.4 Conclusion 8 Conclusion References Index

132 133 140 143

203 206 210 220 234 237 248 264

Figures

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.1 3.2 3.3

3.4 3.5 3.6

Relationship between the energy intensity of an economy and average energy prices page 22 Estimated price elasticities of demand for various commodities 24 EU-25 emissions attributed to steel and cement production in 2005, and expected emissions reduction 27 Stages of technology innovation and use 31 Installed wind power per year in MW 40 Global greenhouse gas abatement cost curve beyond business-as-usual (v. 2.1) to 2030 42 Power prices versus fuel/CO2 costs in Germany 46 Estimates of carbon tax impacts and redistribution policies 49 Three pillars of climate policy 53 Evaluation of prices under USA’s SO2 trading scheme 60 Structure of the European Union emissions-trading scheme 62 Price of EU emissions allowance for phases I and II (European Energy Exchange and European Climate Exchange) 66 Power sector illustration of distortions from freeallowance allocation 73 Comparison of new entrant allocation 79 Pyramid of distortions of the EU ETS 81

ix

x

3.7 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 5.1 5.2

5.3 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8

*

List of ﬁgures

Importance for sectors of differences in energy taxes, regulations and carbon pricing Generation share of different technologies in the UK: 1960–1997 Marginal damage and mitigation costs, and the impact of uncertainty (Weitzman framework) Mitigation-cost curve steeper than damage-cost curve Roles of different technologies over the next four decades using long-term emissions targets Renewables targets and the role of different renewable technologies Impact of carbon price projections for agents involved in investment decisions Ex ante and ex post costs of UK policies Components of a low-carbon perspective EU-25 emissions projections for 2008–2012 based on veriﬁed emissions in 2005 Put options on the price of carbon Market categorisation of risk determining ﬁnance structure, access and cost Extremes of setting climate policy targets Time frames used for the deﬁnition of policy targets and differentiation between input-based and output-based metrics Three main channels for linkage between countries Potential channels for leakage The economics of leakage along the production channel Value at stake for main manufacturing sectors v. UK trade intensity from outside EU at €20/tonne CO2 Value at stake for construction materials v. UK trade intensity from outside EU at €20/tonne CO2 Industrial activities with the highest cost increase from carbon pricing and their contribution to UK GDP Determining premium for domestic products and trade-related costs for imports Can local premium and trade costs (import costs) compensate for asymmetric carbon costs? Impact of carbon pricing on demand and trade ﬂows for EU

88 98 101 102 106 107 108 111 114 117 123 126 136

142 146 165 167 169 171 172 174 176 179

List of ﬁgures

6.9 6.10 6.11 6.12 6.13 7.1 7.2 7.3

7.4

*

xi

Illustration of value chain with potentials for efﬁciency improvements, substitution and leakage Value chain of concrete production Value chain of steel production using basic oxygen furnace process Policy options to address leakage concerns International co-operation could deﬁne limitations for the use of border adjustment Distribution of regional per capita greenhouse gas emissions in 1990 and 2007 Concept for actions that allow for a transition in individual sectors or technologies Economic potential for greenhouse gas emissions reduction in non-OECD, non-EIT countries relative to baseline emissions Special Report on Emissions Scenarios B2 European emissions trajectories with and without off-sets

181 182 184 186 197 204 207

223 225

Tables

3.1 4.1 6.1 7.1 7.2

Emissions-trading-scheme emissions cap as proposed by Member States and accepted by EU Commission Main determinants of investment choices across sectors Instruments to address leakage for production of different commodities Financial mechanisms to contribute to investment and operation Financial mechanisms to facilitate access to ﬁnance

xii

page 71 109 190 215 216

Boxes

2.1 4.1 4.2 5.1 6.1

Prices, cost-pass through: power-sector example Tax versus cap and trade Government-issued put options to guarantee a price ﬂoor Experience with policy indicators Leakage channels

xiii

page 46 101 123 142 165

Acknowledgements

I would like to thank David Newbery, for his continued insistence and ideas on ensuring a robust analytic analysis, and Michael Grubb, for the introduction to climate change policy and guidance on how to capture and communicate the complexities. For all the detailed input, constructive comments on drafts, and edits I would like to thank Alex Henney, Alexander Vasa, Andreia Meshreky, Anne Neumann, Anne Schopp, Angus Johnston, Andreas Tuerk, Arttu Makipaa, Axel Michaelowa, Cameron Hepburn, Chantal Morel, Chris Beauman, Dallas Burtraw, Damien Demailly, David Nelson, Denny Ellerman (also for data for Figure 3.1), Ferdinand Vieider, Gerri Ward, Heleen de Coninck, Hermann F. Amecke-Gen-Monnighoff, Ines Neubert, Jill Duggan, Jon Ward, Jonathan Mirrlees-Black, Kate Loveys, Kath Rowley, Michael Lührs, Michel Colombier, Misato Sato, Peter Wooders, Roland Ismer, Sam Frankhauser, Sarah Lester, Seabron Adamson, Terry Barker, Thilo Grau, Tim Laing, Umashankar Sreenivasamurthy, Victoria Roberts. I am grateful to Cambridge University, DIW Berlin, Climate Policy Initiative, Climate Strategies and the Economic and Social Research Council grant Towards a Sustainable Energy Economy (RES-152–25–1002) for funding the research for the book and for the underlying projects.

xiv

Abbreviations

AAU BAT BAU BOF CCGT CCS CDM CER CfD COMETR COP EAF EBRD ECX EEX EIT EPA EU ETS EUA GDP GEF GHG IEA

assigned amount unit best available technology business as usual basic oxygen furnace combined-cycle gas turbine carbon capture and sequestration clean development mechanism certiﬁed emission reduction contract for difference Competitiveness Effects of Environmental Tax Reforms Conference of the Parties electric arc furnace European Bank for Reconstruction and Development European Climate Exchange European Energy Exchange economies in transition Environmental Protection Agency European Union emissions-trading scheme European Union Allowance gross domestic product Global Environment Facility greenhouse gas International Energy Agency

xv

xvi

*

List of abbreviations

IFC International Finance Corporation IFI international ﬁnancial institution IISI International Iron and Steel Institute IMF International Monetary Fund IPCC Intergovernmental Panel on Climate Change IPR intellectual property rights JI joint implementation LPSA Local Public Service Agreement MIGA Multilateral Investment Guarantee Agency MOP Meeting of the Parties NAMA nationally appropriate mitigation action NAP national allocation plan ODA Overseas Development Agency OECD Organisation for Economic Co-operation and Development PRSP Poverty Reduction Strategy Paper RGGI Regional Greenhouse Gas Initiative SRES Special Report on Emissions Scenarios UNFCCC United Nations Framework Convention on Climate Change VAT value-added tax WB World Bank WCI Western Climate Initiative WTO World Trade Organization

one

Introduction

In the Copenhagen Accord at the UN climate negotiations in Copenhagen, 117 heads of state concluded that low-emissions development would be necessary in order to combat climate change. However, at the end of a two-year negotiation marathon, they could not agree on emissions targets. In the following months, eighty-three countries submitted proposals to the United Nations Framework Convention on Climate Change (UNFCCC) secretariat for nationally appropriate mitigation actions (NAMAs), thus supporting the Accord and gradually rebuilding momentum for international climate co-operation. One major challenge for the negotiations in Copenhagen was the shift of emphasis from marginal emissions reductions to low-carbon development. This is illustrated by the discussions on support mechanisms for climate policy in developing countries. The Kyoto Protocol deﬁned an international off-setting approach, the clean development mechanism (CDM): large emitters in developed countries can ﬁnance individual projects to reduce greenhouse gas emissions in developing countries if this is cheaper than reducing their own emissions (UNFCCC 1997). In this way, the cost of achieving the Kyoto emissions targets for the period 2008–2012 are 1

2

*

Introduction

reduced, but so are incentives to pursue more-ambitious mitigation actions in developed countries. As projects are directly supported within this international mechanism, domestic policy makers in developing countries have only limited involvement. The Copenhagen Accord of 2009 emphasises the role of lowemissions development strategies. It invites developing countries to submit descriptions of NAMAs they envisage using for the implementation of these strategies. The Accord outlines technology co-operation and ﬁnance mechanisms to provide international support for their implementation. This includes a commitment by developed countries to provide new and additional resources approaching $30 billion for the period 2010–2012 and ambitious ﬁnancing goals for 2020. The shift of emphasis from marginal emissions reduction to lowcarbon development was triggered by rapid evolution in the scientiﬁc/ economic landscape. More-ambitious and more-timely emissions reduction is necessary and is perceived to be more possible today than was anticipated at the time of the negotiations of the Kyoto Protocol: 1 Scientiﬁc evidence of the mechanisms of climate change has been strengthened, and the identiﬁcation of climatic trigger points has further increased concern. Regional climate-change impact assessments have raised additional concerns as many of the most vulnerable societies will be exposed to the greatest changes. Thus, the Intergovernmental Panel on Climate Change (IPCC) concluded that by 2050, global CO2 emissions must be 50–85 per cent below 2000 levels in order to limit global average temperature increases to 2.0–2.4 °C (IPCC 2007). 2 The urgency and scale at which global emissions have to decline to meet the 2 °C target have increased as developed countries have only pursued moderate efforts to mitigate greenhouse gas emissions during the last twenty years and emissions in several developing countries have increased rapidly in parallel with strong economic growth.

Introduction

*

3

3 Economic analysis has demonstrated that low-carbon development is viable. Simulations for the global economy with ambitious climate-change targets project a one-off gross domestic product (GDP) reduction in the order of 0–3 per cent. This is small compared with GDP growth projected for the modelling horizons. It requires technology innovation and the early shifting of investment from carbon-intensive to low-carbon and energy-efﬁcient infrastructure, buildings, industry and transport choices. The increasing emphasis on low-carbon development also shifts more emphasis on to the role of domestic policy frameworks. New technologies can enter the market only if regulatory structures provide for planning, standards and the necessary support infrastructure. Equally, investment will shift to low-carbon options only if investors feel comfortable with the implied risks. In particular, delays inherent in new sectors and technologies and uncertainty about future revenue streams are of concern. Policy choices on subsidies, taxes, tariffs and carbon-price schemes jointly determine such revenue streams and remain largely in the domain of national and regional policy makers. Any changes necessary to facilitate low-carbon development also need to be initiated by domestic policy makers. They can gather domestic political support by addressing the speciﬁc circumstances and concerns. This book describes the experience of the domestic implementation of climate policies through the example of carbon pricing, and discusses the implications for the design of international climate co-operation. Carbon pricing lends itself particularly well to such an analysis, as it can achieve multiple objectives. At the domestic level, it can shift production, consumption and investment to low-carbon choices. At the international level, it has been envisaged as a mechanism to facilitate commitment to mitigation targets among developed countries and to provide resources to ﬁnance mitigation action in developing countries. Moreover, both levels often interact – for example, in the industrial sector. However, as many industrial

4

*

Introduction

products are internationally traded, unilateral carbon pricing could arguably result in the relocation of production, and thus emissions, rather than in their reduction. Through the lens of carbon pricing, this book explores how international climate co-operation can be designed to support the domestic implementation of policies for low-carbon development. International co-operation creates a shared sense of action and responsibility, it can it provide an external framework of commitments to turn longer-term climate objectives into short-term policies and programmes and it can it provide support for mitigation and adaptation action in developing countries. Finally, international co-ordination is necessary to ensure that the various parts of the global effort combine to achieve global climate objectives. Chapter 2 explores the role of a climate policy mix. Carbon emissions from energy production and industrial processes are deeply entrenched in our economies. To mitigate the risk of catastrophic climate change, these emissions need to be reduced to a fraction of 2011 levels. The challenge is now to implement policy instruments to deliver the necessary emissions reduction. In the past, many of the climate policy discussions focused on marginal emissions reduction. To achieve this, economists recommend exposing producers and consumers to the environmental cost of carbon, thus creating incentives for efﬁciency improvements and for less-carbon-intensive production and consumption choices. The theoretical foundation for this approach is the ﬁrst fundamental theorem of welfare economics: the ‘invisible hand’ of the market will result in efﬁcient production and consumption decisions. This requires that a set of assumptions be satisﬁed, including market participants’ exposure to the costs of environmental externalities. Carbon taxes or emissions trading have therefore received much attention in public debate. The objectives of climate policy have shifted from delivering marginal emissions reduction to facilitating low-carbon development.

Introduction

*

5

This requires reassessment of whether all the assumptions required for the ﬁrst fundamental theorem of welfare economics are still satisﬁed – for example, whether internalising the cost of carbon alone delivers efﬁcient market outcomes. Innovation and learning-by-doing mean that the potential response is much more complicated and nonlinear – described by economists as non-convexities in cost functions. Interactions between actors and technologies create network effects that result in similar non-convexities for the beneﬁt functions. In addition, pre-existing infrastructure creates path dependency. All of these instances violate the assumptions of the welfare theorem and, therefore, pricing carbon is not sufﬁcient to deliver efﬁcient outcomes, but remains necessary. The Stern Review (2006) on climate change points to three sets of instruments that are necessary to facilitate a low-carbon transition: (i) putting a price on carbon; (ii) technology policy; and (iii) targeted regulation, with transparent and shared information and measures to engage individuals and companies in low-carbon opportunities. Countries are now using a combination of these policy instruments to tackle climate change. However, any policy intervention can distort economic incentives and create costs in implementation, enforcement and compliance. Policy intervention can have distributional implications and might interfere in private-sector decision processes. So, it is essential to assess carefully the speciﬁc circumstances of sector, country and industry structures when designing the policy mix for low-carbon transformation. Changes to policy frameworks, in particular if they directly affect relative prices of products and services, have distributional implications that shift costs and wealth between poor and rich and between rural and urban parts of society. As a result, implementation of individual climate-policy instruments can increase or reduce fuel poverty or inequality in a society. The more ambitious the objectives of climate-policy instruments, the more signiﬁcant could be the distributional impacts, which affect equity issues and political

6

*

Introduction

support. With careful analysis, it is possible to anticipate many of these impacts and to use complementing policy measures to balance the consequences, or to support individuals and companies during a transition. To illustrate the type of analysis that is necessary for any policy instrument, this book explores in more detail the role of carbon pricing in the economy. Carbon pricing increases the price of processes, products and services that are carbon intensive, thus creating incentives for the use and innovation of carbon-efﬁcient technologies, and inducing substitution of lower-carbon fuels, products and services, by industry and ﬁnal consumers. The price signal feeds into individual decisions that would be difﬁcult to target with regulation. Pricing also makes it proﬁtable to comply with carbonefﬁciency regulations, thus facilitating their implementation. Carbon prices can be delivered with a carbon tax or cap-and-trade schemes. As much as the theoretical features are essential, the real analytical and policy challenge revolves around the process of implementing the relevant policy instruments and the design of detailed provisions. Chapter 3 discusses the implementation of carbon pricing, using the example of cap-and-trade schemes. Early experience from the trading schemes for SO2 and NOX in the USA, and the subsequent European Union emissions-trading scheme (EU ETS) for CO2 allowances, are discussed. Cap-and-trade schemes were acceptable to stakeholders, aided co-ordination across countries, and delivered an emissions price. The chapter examines experience from the EU scheme, focusing on setting the cap and allocation of allowances under the EU ETS. Free-allowance allocation to emitters was the bargain chip offered to gain industry support for the scheme. But it exposed ministries tasked with the allowance allocation to intensive lobbying. In the pilot phase, lobbying for more allocation also inﬂated the cap. This was subsequently addressed as Kyoto and EU targets were ﬁxed prior to the allocation discussion.

Introduction

*

7

Repeated free allocation also creates various perverse incentives that undermine the economic efﬁciency of the scheme. Therefore, the European Commission proposed, and the EU legislative agreed (European Commission 2009a), to require the auctioning of most allowances in the EU scheme post 2012. After initial difﬁculties, support for the EU ETS is now shared across governments, industry and political groups. However, the experience of the ﬁrst years also offers several insights relevant for design of climate-change policy. The implementation of a scheme that allocates allowances with an annual value of about €40 billion is not trivial. Free allocation to emitters can also have undesired distributional impacts. In most markets, emitters will pass on carbon costs to product prices and thus to consumers. As a result, emitters proﬁt from the free allocation, while consumers bear the costs. Schemes to compensate households for the distributional implications of carbon pricing deserve careful consideration, to ensure equity and political support. The negative public perception of large emitters beneﬁting from free-allowance allocation was the second reason for the move towards large-scale auctioning in Europe after 2012. Policy instruments will ultimately be deemed successful, and should therefore be evaluated, only by the results they deliver. Most reduction in emissions is expected to be achieved through changes of investment choices and shifting ﬁnance to lower-carbon options. Chapter 4 assesses the delivery of investment responses. Lowcarbon development requires diffusion of existing and new technologies, infrastructure and business models. This requires clear, credible and long-term policies, as conﬁrmed in Copenhagen by a statement of 186 investment institutions representing assets of $13 trillion. International discussions focus on emissions targets, initially for developed countries. The expectation is that clear commitment to such targets also allows private actors to anticipate future opportunities for low-carbon processes, products and services, as well as

8

*

Introduction

constraints for carbon-intensive investment choices. Obviously, emissions targets for countries must be translated into policy instruments if they are to affect investment choices of private actors. Cap-and-trade schemes provide the opportunity directly to translate emissions targets into economic incentives for private actors. This can limit the uncertainty of policy design and implementation and thus strengthen the low-carbon investment framework. The ﬁrst example of this approach – the EU ETS – has succeeded in focusing the attention of carbon-intensive industries on exposure to carbon costs and on low-carbon opportunities. When evaluating individual projects, many investors remain concerned that the carbon-price signal is not sufﬁciently robust, and that carbon prices might drop in response to economic and political developments. It is therefore argued that the risk of extremely low carbon prices must be reduced, if low-carbon projects are to be facilitated. It was suggested that, for example, carbon taxes would be more predictable in the short term, and could thus increase investment certainty for investors operating with short investment horizons. In the long term, however, carbon taxes are more difﬁcult to predict, as they are subject to continued political negotiations. Even where carbon taxes are ﬁxed in the long term, this might increase rather than reduce exposure for low-carbon investors, who are competing in a world with uncertain fuel, commodity and technology costs. Financial investors, including pension funds, are another group of actors that can be supported with climate policy instruments. They need to shift investment from carbon-intensive and -exposed activities to low-carbon options, so as to ensure the long-term viability of their investment portfolio. The sector is only starting to explore the carbon risk inherent in some incumbent companies, and needs better tools to asses the impact of carbon risk on future performance. In addition, all too frequently, the focus is on the risks of low-carbon investment, owing to the involvement of new technologies, business models and partnerships.

Introduction

*

9

The discussion illustrates that the diversity of participants in our economies might respond differently to a policy instrument that delivers a carbon price. Hybrid cap-and-trade schemes, combining a price ﬂoor with an emissions cap, could target the needs of heterogeneous groups of investors. The evolving focus of climate policy, from marginal reduction in emissions towards low-carbon transformation of economies, is also reﬂected in discussions about low-carbon investment frameworks. Investment in new technologies, production processes, products and services can succeed only with appropriate infrastructure, institutional setting and social acceptance. Government support is often required for research, development, early deployment and adjustments to regulatory frameworks, as well as administrative standards and procedures. The co-ordination of all these activities requires a shared vision of a country’s low-carbon development trajectory. For this reason, the international discussions leading up to Copenhagen increasingly emphasised the importance of low-carbon development plans for developed and developing countries. The plans characterise industrial and technological development, energy use and emissions across different sectors of the economy. They can therefore ensure that initial mitigation efforts are consistent with long-term objectives (for example, understanding the implications of efﬁciency improvements of coal power stations, relative to other mitigation options). They can also ensure consistency of mitigation strategies across sectors – for example, by testing whether available biomass resources are consistent with their anticipated use in steel, cement, transport, heating, industry and power sectors, and by assessing whether electricity use in transport, industry and for heat pumps is consistent with anticipated generation and network structure. In addition, the plans allow national governments to prioritise actions according to long-term relevance and lead times, and to signal a consistent overall strategy to facilitate private-sector investment.

10

*

Introduction

Low-carbon development plans are therefore essential, not as a bureaucratic instrument, but as a process to create a shared vision and platform to discuss and initiate the appropriate policy actions and thus facilitate low-carbon investments. These actions might well vary across countries that differ in their social preferences, industry structures, ﬁnance sectors and institutional settings. Countries might diverge in their emphasis on short-term insurance of robust carbon prices, mid-term emissions targets translated into emissions-trading schemes, technology-support schemes and institutional design choices and administrative standards. The discussion of domestic policy frameworks to facilitate a lowcarbon development points to the various opportunities for international climate co-operation, discussed in Chapter 5. An adequate response to climate change requires action on a global scale. Otherwise it is impossible to achieve the earlier-mentioned 2 °C target and reduce emissions by at least 50–85 per cent below 2000 levels. Thus, the main driver for global climate co-operation will remain the engagement of all nations in supporting emissions reduction with their domestic efforts. Sometimes the need to act on a global scale, to achieve the 2 °C target, is interpreted to mean that individuals and countries have no responsibility to act individually in the absence of adequate action on a global scale. This is wrong – every tonne of carbon emitted accelerates climate change and increases the risks and costs for society. The chapter discusses in more detail why domestic mitigation action does not require the justiﬁcation of international co-ordination. If international agreements are not essential to facilitate action by individuals and nations, they can nevertheless be important in enhancing the effectiveness of domestic action in developed countries and in supporting low-carbon development in developing countries. First, global co-operation can enhance the level of understanding of climate-change impact and options for tackling it, bringing together experts, policy makers and industry actors. Second, global

Introduction

*

11

co-operation can increase the level of action by building conﬁdence that action will be effective. It also contributes to the sense of responsibility for a country’s own emissions and awareness of the damage emissions can inﬂict on domestic and foreign populations and thus enhance action to mitigate that impact. Third, international co-operation can contribute to meaningful reporting on actions that allows for rapid international learning concerning best-practice policymaking. Transparent reporting also facilitates the measurement of the performance of policy instruments while enhancing the accountability of policy makers and improving transparency for private-sector investors. International co-operation can, furthermore, provide a platform for commitments regarding emissions targets and for speciﬁc actions. Such commitments provide time frames and quantitative reference levels that can subsequently help to overcome domestic political barriers. External commitments can also enhance the credibility of longer-term strategies for private-sector investors, reducing capital costs and enhancing the scale of low-carbon investment. This leads to the question of how international carbon markets can support co-operation and domestic de-carbonisation efforts. Such markets can allow for trade at a national level, or, as more frequently observed, at the level of individual installations. This allows traders to identify the least-cost emissions-reduction opportunities in a bigger market, and could therefore offer the beneﬁt of reducing costs of climate policy. A joint scheme also has some political attractions: it might reﬂect increasing commitment by participating countries and could create momentum to drive implementation through any political adversity. Yet, for an effective joint scheme several risks have to be managed. First, if two countries have a joint scheme but negotiate future emissions targets separately, industry in the more-ambitious country will end up buying allowances issued by the less-ambitious country. This creates incentives to negotiate less-ambitious targets at any future negotiations. Second, emissions reduction requires a multitude of

12

*

Introduction

domestic policies, such as information provision, drafting of performance standards and suitable regulatory frameworks for new technologies. The responsibility of governments for these policies is clearly deﬁned, if targets are deﬁned for the same jurisdiction over which governments have responsibility. This allows governments to measure and manage policy success more effectively and this would be more difﬁcult in the case of joint schemes that extend beyond the boundary of their jurisdiction. Third, with a joint scheme, an individual country’s more-ambitious target will have only a marginal impact in the scarcity of a global scheme, and thus have a less-signiﬁcant impact on technology development and transition. By contrast, if domestic support in a country requests a more-ambitious climate policy, this directly translates into a tighter cap for a regional scheme, and results in higher carbon prices. The region will beneﬁt from accelerated low-carbon innovation and transformation. As the new low-carbon processes, products and policies diffuse to other regions, they contribute to accelerated global de-carbonisation Rather than designing a joint emissions-trading scheme, countries could initially develop individual schemes that are subsequently linked. As of summer 2010, separate emissions-trading schemes are evolving in Australia, New Zealand and several states of the USA, while EU countries have implemented a joint trading scheme. Several approaches are available that could result in direct and indirect linking of these schemes. It is for policy makers to decide whether to pursue early integration (for example, as proposed in 2015 by the European Commission) or to delay such linking until 2020. As an alternative to international emissions-trading approaches, some analysts have proposed a ‘global carbon tax’. Such a tax would be implemented on the national level and revenues could remain in domestic budgets. International co-ordination would determine the level, or a minimum level, for such a tax. This could create a joint momentum for action, offer opportunities for outside commitment to enhance the long-term credibility of a tax for investors and avoid

Introduction

*

13

leakage concerns arising from different carbon price levels. It might, however, be challenging to agree on ambitious tax levels among countries that have different preferences for the role of carbon pricing in the policy mix. Still, for smaller countries, it might be attractive to align their domestic carbon tax with some carbon price emerging in other regions, eventually leading to some agreement on a minimum carbon price level, to be delivered with policy instruments such as tax or cap and trade. The discussion on global carbon markets and carbon taxes suggests that we are facing a world of differentiated carbon prices (Chapter 6). The existence of asymmetric prices frequently raises concerns about carbon leakage: higher carbon prices might induce some industries to shift production or investment to countries with low or no carbon pricing. Rather than reducing emissions, climate policy would result in a relocation of emissions to another jurisdiction, a phenomenon typically described as ‘carbon leakage’. The direct environmental impact would probably be negligible unless signiﬁcant transport emissions are implied or a more-carbonintensive fuel is used, as the new production location is likely to have carbon intensity similar to or lower than old facilities. However, there are three indirect effects that are quite disconcerting. First, if emissions that are initially covered under a cap are relocated to another jurisdiction without a cap, this creates space under the cap for additional emissions, resulting in a net emissions increase. Second, as the relocated production facility will not face the carbon price, prices for carbonintensive products will not increase and thus incentives for innovation and substitution to low-carbon alternatives will be suppressed. Finally, given the coincidence of negative social, economic and environmental impacts associated with such relocation, policy makers are likely to address any potential risk of leakage by subsidies – free allowance allocation or direct ﬁnancial support – thus further undermining incentives for de-carbonisation and potentially creating administrative barriers for any change to low-carbon processes, products and services.

14

*

Introduction

However, it is important to note that cost increases from carbon pricing, relative to other cost components, are trivial for all but 1–2 per cent of economic activity. As the cost increase is concentrated in a narrow set of economic activities – in our analysis, twenty-four sub-sectors – it is possible to carry out a sector-speciﬁc analysis to assess whether the increase is really substantial. Owing to transportation costs, product differentiation and sunk-investment costs, there is no concern about leakage in several of these sub-sectors. Thus, only a few sub-sectors, such as basic steel and cement production, are likely to require targeted measures to address leakage concerns. These measures can differ across sub-sectors and can include conditional free-allowance allocation, direct ﬁnancial subsidies and border adjustments. All mechanisms have some economic, environmental or political risks attached, and require a careful assessment of the speciﬁc circumstances of a sector. If these measures are implemented as components of an internationally coordinated approach, some of these negative effects can be reduced. Close international co-operation will therefore be essential to ensure that any response to leakage protects the environmental effectiveness of carbon pricing. The sector-speciﬁc analysis of leakage concerns leads to the conclusion that it is possible for countries to pursue ambitious emissions targets and make use of the full carbon price as part of their policy mix. Thus, leading by example, they can help to accelerate technology development and diffusion, and contribute to international experience of low-carbon policy frameworks. Action by individual countries, however, is not a substitute for an international agreement that reﬂects common but differentiated responsibility for climate mitigation and thus creates the opportunity – and, where necessary, the support – for all countries to contribute to emissions mitigation action. Many developing countries have limited resources or capacities to pursue mitigation and adaptation actions and in addition face other pressing priorities, such as poverty alleviation, education and health.

Introduction

*

15

Accordingly, Chapter 7 explores how to provide international support for low-carbon growth in developing countries. In recent years, the CDM has been the major mechanism for international support of low-carbon action in developing countries. It is a project-based off-set mechanism that allows industrialised countries to acquire emissions-reduction credits generated through projects in developing countries. It has the double aim of generating low-cost emissions reduction and promoting sustainable development in the countries hosting projects. With ﬁnancial ﬂows directed towards developing countries, it has provided tangible evidence of the importance attributed to climate policy in Europe and Japan. It has succeeded in supporting initial projects and has been effective in creating awareness, interest, and expertise in both private and public sectors. The CDM off-setting mechanism has also put a price on carbon. However, it acts as a subsidy for low-carbon options, as opposed to providing a disincentive to choose carbon-intensive options. As each tonne of avoided carbon emissions receives the same CDM price, the mechanism creates proﬁts for actors in energy- and carbon-intensive sectors in developing countries. This can contribute to a lock-in of energy- and carbon-intensive activities, rather than facilitating a shifting towards low-carbon choices. The mechanism exempliﬁes the paradigm of delivering marginal emissions reduction through global mechanisms, largely circumventing the involvement of domestic policy makers and institutions. It even creates incentives for policy makers in developing countries not to implement effective domestic policy frameworks for low-carbon investments. Illustrating this dilemma, CDM support for new wind projects in China has been declined because of a domestic feed-in tariff. Effective domestic policy frameworks are essential to facilitate lowcarbon development in all countries. Hence, in 2009, discussions on international co-operation on climate policy increasingly focused

16

*

Introduction

on a new approach. (See, for example, UNFCCC (2009c) of the ad hoc working group on long-term co-operative action, discussed at Copenhagen.) This is comprised four distinct components: * Low-carbon development plans were initiated by South Africa and have subsequently been developed also by India, Brazil, China and Mexico. They outline the intended economic, energy and emissions trajectories for their respective countries and identify trigger points for policy interventions that can be expressed as NAMAs. * Nationally appropriate mitigation actions (NAMAs) are a set of projects, programmes and policies to shift a domestic sector or technology onto a low-carbon development trajectory. A NAMA links a set of actions that need to be pursued in parallel for a successful transition. Designing one NAMA for each transition in a sector or activity allows for their independent design and implementation, thus reducing overall complexity and creating early experiences and success stories. In response to the Copenhagen Accord, countries have submitted NAMAs to the UNFCCC secretariat. * Developed countries have committed to providing new and additional resources approaching $30 billion in the period 2010–2012 to support mitigation and adaptation measures in developing countries. In the context of meaningful mitigation actions and transparency on implementation, developed countries commit to a goal of jointly mobilising $100 billion a year by 2020 to address the needs of developing countries (UNFCCC 2009a). Linking the provision of this support to individual NAMAs creates an additional driver to enhance the scale, scope and/or speed of domestic implementation of the actions required for success. Linking the support to continued NAMA implementation enhances the stability of regulatory and policy frameworks. For example, a feed-in tariff is more likely to be stable if international support contributes to the incremental cost or provides guarantees for the tariff. This attracts domestic and international manufacturing and investment. Additional mechanisms are discussed to enhance

Introduction

*

17

technology co-operation, technical assistance and capacity-building programmes tailored to the state of development and diffusion of speciﬁc technologies, and according to sector and country needs. * Expanding monitoring and reporting, beyond greenhouse gas emissions, to cover lead indicators that capture the effect of policies and programmes in developed and developing countries, can enhance the implementation of an action or policy, facilitate international learning and create transparency to support privatesector investment and innovation. This requires detailed quantitative and qualitative evidence. It reﬂects the experience of industry and of other sectors, and points to the need to link outcome measures (such as changes in greenhouse gas emissions) to a combination of input, process and output indicators. The chapter explores how different ﬁnancial mechanisms and carbon markets can match the needs of actors and sectors effectively to support the implementation of NAMAs. Grants, loans, credit guarantees or equity funding can support public and private actors in dealing with the risks of new technologies and policy frameworks and create opportunities to acquire new skills and develop business models. International support for individual NAMAs can facilitate their implementation and enhance their long-term credibility. An effectively designed ﬁnancial mechanism is therefore an essential catalyst to shift large volumes of private-sector investment to lowcarbon technologies. The choice of ﬁnancial instruments needs to reﬂect institutional capacity and available resources. Experience of bilateral and multilateral co-operation for speciﬁc ﬁnancial instruments can inform the choice of institutions for their provision. The resource base of multilateral institutions can be strengthened with revenue from carbon pricing on international aviation and shipping. Commitment to hypothecation of domestic carbon revenues can create the public funds necessary for bilateral co-operation. If all support provided across all instruments were measured in grant-equivalent terms,

18

*

Introduction

developed countries’ contributions could be evaluated against their commitments. One conceptual challenge remains the transition from a CDM to a broader support framework for low-carbon development. A clear and shared vision of such a transition is necessary to avoid CDM interest groups delaying new approaches that are not in their speciﬁc interest while failing to attract investment for their own projects during a time of policy uncertainty. Well handled, the expertise and structures that were developed to explore, pursue, monitor and ﬁnance low-carbon projects under the CDM can be redirected towards the design and implementation of NAMAs and the pursuit of low-carbon investment in the new framework. In concluding, Chapter 8 summarises how national and international climate policy rapidly evolved in recent years. From a static analysis, focused on delivering marginal emissions reduction, attention has developed to facilitating a low-carbon transition. This requires an extension of policy instruments, away from simple carbon pricing and towards policy packages tailored to the speciﬁc needs of countries and sectors. The importance of low-carbon transformation as part of an adequate response to climate change is reﬂected in international climate negotiations. The focus has expanded from a top-down approach of targetsetting to trigger domestic action to a bottom-up approach that explores how domestic policy design and implementation can be supported through international co-operation.

two

The role of a climate policy mix

Economic models have demonstrated that economic growth is possible in scenarios with both high carbon emissions and ambitious emissions reduction to limit temperature increases. Common scenarios (such as reviewed in IPCC 2007) predict that stringent climate policies will reduce GDP in 2030 by no more than 3 per cent relative to business as usual scenarios, with many models suggesting an even smaller reduction.1 To put this number into perspective, 3 per cent was the average of global annual GDP growth over the last thirty years. High estimates for mitigation costs therefore imply that the global GDP projected for the year 2030 is reached one year later, in 2031. The Stern Review (2006) has identiﬁed three sets of policy interventions that are necessary to shift the economy onto a low-carbon trajectory and facilitate continued economic development: 1 Putting a price on carbon provides the economic incentive for individuals and corporations to participate in the transformation. By providing a signal to these players as to the value of various carbon-reducing options, the carbon price allows market players to make investment, production and consumption decisions in a way that minimises the impact on their economic well-being. 19

20

*

Climate policy mix

2 Tailored technology support encourages the development of new carbon-reducing technologies and can enhance performance and cost characteristics of existing technologies where the prospect of a carbon price alone would not be enough to motivate spending on research, development and deployment, or would not create sufﬁcient demand for new technologies. 3 Many efﬁciency-improvement opportunities exist that appear to be proﬁtable but are not pursued. This can be attributed to a variety of barriers, which differ widely across technologies, sectors and countries. Tailored programmes and policies can unlock some of these energy-efﬁciency opportunities. The experience with energyefﬁciency opportunities can be applied more broadly to low-carbon technologies. They will equally require a shift from existing technologies and associated institutions and behavioural patterns. In many instances, this shift will be inhibited not by costs but by a variety of factors that can be addressed with speciﬁc policies and programmes. Changes to the regulatory framework result in shifts of resources and opportunities between groups of population. For example, a price on carbon will increase energy prices and might disproportionately affect poor consumers. Section 2.4 explores distributional impacts and discusses how policy can balance negative impacts to ensure the social and political viability of climate policy.

2.1 Putting a price on carbon Pigou (1920) realised that individuals and companies may overuse common resources and advocated their taxation. Users pay a price corresponding to the damage others suffer and will thus internalise the external effects in their decision process. Coase (1960) argued for the deﬁnition and allocation of property rights to the scarce resource, and showed that this can also result in an efﬁcient use of the resource. Our atmosphere is just such a common resource, and science shows

Putting a price on carbon

*

21

that the volume of greenhouse gases emitted into the atmosphere will create serious environmental repercussions: the greenhouse effect. The scientiﬁc evidence has provided the basis for a global political consensus of the G8 economies that CO2 emissions must be reduced to a fraction of today’s level (BBC 2009). Some economists argue for a thorough assessment of the future damage that will result from today’s CO2 emissions, so as to set the price of carbon equal to the damage anticipated from each unit of CO2 emissions. Thus, private actors will only emit CO2 if their private beneﬁt exceeds the cost of future damage. Other analysts question the ability for science to quantify and politics to agree on such a price of carbon and ask whether the payments from emitters will beneﬁt the individuals who will suffer the damage. Hence policy makers usually deﬁne a target in terms of acceptable temperature increases from the greenhouse effect and aim to translate this into emissions targets compatible with temperature targets. In both cases, governments need policy instruments to encourage careful use of our scarce common resource – atmosphere – and in both cases putting a price on carbon is seen to be an important policy instrument. Anderson and Ekins (2009) review the impacts of the environmental tax reforms in Denmark, Finland, Germany, the Netherlands, Slovenia, Sweden and the UK, and ﬁnd evidence for emissions reduction from taxes in all seven countries. Finland and Sweden have had the largest environmental tax increases, and show the highest emissions response. Bruvoll and Larsen (2004) analyse the contribution of carbon taxes to reductions in emissions intensity in Norway during the 1990s. The impact on emissions is smaller than projected in many simulation studies (see European Commission 2007), as many energy-intensive users are exempt. Enevoldsen (2005) ﬁnds that carbon, SO2 and energy taxes together reduced CO2 emissions in Denmark by 9–11 per cent and in the Netherlands by 1–2 per cent over the period 1992–2000. The ﬁrst evaluations of the largest application of carbon pricing, the European Union emissions-trading scheme, suggest that it had reduced the emissions of installations covered by the scheme by about

22

*

Climate policy mix

1,400 Denmark

Average energy price $/t oil equivalents

1,200

Japan Norway

1,000

Austria Italy Luxembourg Sweden Switzerland Portugal Finland France 600 New Zealand Spain Netherlands United States of America Greece Hungary Australia 400 Korea Turkey 800 Germany

Mexico 200

United Kingdom

Canada

Czech Republic Slovakia Poland

Belgium 0 0.0

0.1

0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Average energy intensity (kg oil equivalent/$1995 GDP)

1.0

Figure 2.1 Relationship between the energy intensity of an economy and average energy prices. Source: From Newbery 2003. Copyright United Nations 2003. Reprinted with permission of the United Nations.

2.5–5 per cent in 2005 (Ellerman, Convery and de Perthuis. 2010). However, data are still scarce and therefore estimation uncertainties are large, particularly for the evaluation of long-term impacts. To consider other evidence, we turn to the fact that most carbon emissions result from energy consumption. Data on the link between energy prices and energy consumption offer some insights into the effect that can be expected from carbon pricing. Figure 2.1 depicts the average energy prices and the energy required to produce one unit of GDP. It shows that countries with higher energy prices deliver more GDP per unit of energy input. For example, Japanese companies and households face twice the energy prices of their counterparts in the USA and they deliver twice the GDP with one unit of energy.

Putting a price on carbon

*

23

Such cross-country comparison shows neither to what extent high energy prices in a country have contributed to the energy efﬁciency of the economy, nor to what extent countries with energy-efﬁcient economies levy higher energy taxes. Analysis of the performance of countries over time can answer this question, but needs to take into account the differences between short-term and long-term responsiveness to energy prices. Production and consumption patterns are adjusted in shorter time frames. Infrastructure and manufacturing design require investment and thus only respond in the longer term. Reviewing the performance of industrial sectors in twenty-six OECD countries from 1990 to 2005, Steinbuks and Neuhoff (2010) estimate that, with a 1 per cent increase in energy prices, in the shorter term energy consumption per unit of production reduces by 0.26 per cent – to 1 per cent (though there are differences across sectors), while the longer-term investment response results in a reduction of energy demand of between 0.03 per cent and 0.9 per cent. This is in line with estimates for long-run elasticities for demand of individual fuels, as summarised in Figure 2.2. This relationship would suggest that our economies can deliver the ambitious reductions in energy consumption and thus reduction in emissions required to stabilise global temperatures without jeopardising economic performance. Why not translate the link between high energy prices and high economic output per unit of energy into a link between high carbon prices and high GDP per unit of carbon? Our economies could grow while reducing emissions to sustainable levels. Does this representation provide the full picture? Let us assume OECD countries aimed to reduce their CO2 emissions by 90 per cent by 2050. This could be achieved if energy demand were reduced by a factor of 10 (ignoring for a moment economic growth). If we assume the demand responsiveness of energy to energy prices of −1, as in Figure 2.1, then this would require a tenfold increase in energy prices. However, many renewable energy technologies are viable at prices below this level, and would provide viable substitutes for

24

*

Climate policy mix

Elasticity (absolute values)

3 2.5 2 1.5 1 0.5

O

il G as C C oal em Re en P sid l t en El asti tia ect cs l e ric le ity ct C ric op it pe Pe y r( t r do W ol m es W he tic oo at al d p ly re ulp fin e C d) N opp ew e A sp r lu rin m t in St Ir ium ee on I lm o El nte ate re ec gra ri tri te al ca d s rc ste fu el rn ac e ste el

0

Figure 2.2 Estimated price elasticities of demand for various commodities. Source: Based on work with Anne Neumann and Misato Sato summarising the following studies: Bernstein and Grifﬁn 2005; Bohi 1981; Bohi and Zimmerman 1984; Brons, Nijkamp, Pels and Rietveld 2006; Chas-Amil and Buongiorno 2000; Considine 1991; Dennerlein 1990; Espey 1998; Filippini 1999; Fuss 1977; Graham and Glaister 2002; Hanly, Dargay and Goodwin 2002; Hekman 1978; Jans and Rosenbaum 1997; Jones 1996; Karlson 1983; La Cour and H. P. Mollgaard 2002; Lafferty et al. 2001; Lord and Farr 2003; MacKinnon and Olewiler 1980; Malach 1957; Mannaerts 2000; Reinaud 2004; Roller and Steen 2005; Ryan 2005; Schaefer 1979; Spierer 1988; Winters 1995.

carbon-intensive energy. Once substitutes for energy- or carbonintensive products are possible, and available on a sufﬁcient scale, carbon emissions can decline signiﬁcantly faster than projected, based on demand elasticities. Appropriate representation of substitution opportunities is therefore necessary, both in economic models and in the design of policy instruments.

Substitution of input fuels and carbon-intensive commodities The substitution of carbon-intensive input factors with low-carbon alternatives has been a major contributor to reducing CO2 emissions in the initial years of EU ETS. In the electricity sector, burning

Putting a price on carbon

*

25

natural gas instead of coal for power generation can reduce carbon emissions by about 50 per cent per unit of electricity produced. Moving towards renewable energy sources can virtually eliminate emissions during the operation of plants. In some instances, the use of, or investment in, conventional technology can be prohibited by regulation. For example, the European IPCC Directive (European Council 1996) effectively prohibited the use of mercury as a catalyst in chlorine production after 2007 (Concorde East 2004). Policies can also seek to reduce carbon emissions without totally banning them. Regulation can deliver this objective in many instances – for example, by requiring energy-efﬁciency standards for refrigerators. In the power sector, it is difﬁcult to design direct regulation that would achieve this objective on its own. The amount of electricity produced from carbon-intensive coal power stations will decline and will probably be replaced by renewable technologies. But many renewable technologies depend on wind, sun or water conditions, and their output is therefore difﬁcult to predict. Until sufﬁcient storage technologies are deployed, this might require the continued availability of some conventional power stations to meet power demand for a few critical hours per year. How could a government determine ex ante the number of ‘critical hours’ for which a power station should be allowed to operate, especially since this ﬁgure will depend on factors such as the amount of renewable generation, the wider generation mix and the climatic condition in a given year? Pricing carbon can make fossil-fuel-based power production sufﬁciently expensive to avoid its use for base-load power provision or to constrain the use even further. Carbon pricing might thus create the appropriate incentives to shift carbon-intensive generation technologies towards operating only at times of peak demand. Commodities often progress through many production steps before the ﬁnal product or service is delivered to consumers. Some substitution is possible at all steps, but it is difﬁcult to prescribe with regulation.

26

*

Climate policy mix

We refer to the consecutive production steps as the value chain, and will illustrate the substitution opportunities using the example of clinker, the main component of cement. Clinker is produced by heating limestone, which undergoes a chemical transformation, releasing carbon. Although carbon emissions can be reduced by using renewable energy sources for heating, the majority of the emissions are due to the chemical transformation and cannot be avoided. After milling, clinker is mixed with other substances to make cement. Emissions can be reduced via the substitution of some of the clinker with other materials suitable for cement production (Walker and Richardson 2006). Cement is used to create concrete structures. With more careful planning and execution, building structures can be made less carbon intensive by changing the balance between material costs and additional labour costs. Architects and engineers can choose between various materials such as concrete, steel, wood, stone and glass, and so strike a balance between cost-effective material choice and design criteria. If concrete, steel and glass prices reﬂect the price of carbon, the choice of inputs will shift. Finally, in many developed countries, investors face a choice between refurbishing existing buildings, which is labour intensive, and replacing old buildings with new ones, which is more material intensive. If the price of carbon is reﬂected in the prices of materials, it creates an incentive to refurbish rather than replace buildings. This example illustrates the different substitution opportunities that exist in the value chain. Several studies aim to quantify these substitution effects with more formal economic methods. Assuming producers pass the full carbon price on to products, it is possible to estimate the price increase of a product or service. For example, if the price of the commodity increases by 10 per cent, then a price elasticity of −0.5 implies that demand will fall by 5 per cent, while with a price elasticity of −1 the demand will decrease by 10 per cent. One of the challenges for economists is the estimation of such price elasticities. This is illustrated by the differing values of price elasticities estimated by different studies (Figure 2.2).

Steel

CO2 emissions

27

Efficiency improvements

*

Demand reduction

CO2 emissions (no carbon price)

CO2 emissions

Efficiency improvements

Demand reduction

180 160 140 120 100 80 60 40 20 0

CO2 emissions (no carbon price)

M.t CO2/year

Putting a price on carbon

Cement

Figure 2.3 EU-25 emissions attributed to steel and cement production in 2005, and expected emissions reduction (Grey area gives conﬁdence interval 10–90%)2

Estimates not only differ across different commodities, as expected, but also different estimates for the same commodity result in large variations of elasticities. The grey bars in Figure 2.2 indicate the results from our own estimation, also illustrating the uncertainty of the estimated parameters. Given these uncertainties in the input parameters, it is still difﬁcult to predict the reductions to demand that will be induced across different sectors by carbon pricing. This is a further reason to use the market mechanism, as this does not require governments to prescribe the optimal demand reductions for different commodities, but allows private actors to make their preferred choices. To illustrate this effect, we use average values of demand elasticities (Figure 2.2) and IPCC estimates for efﬁciency-improvement potentials to estimate the emissions reduction that a €20/tonne CO2 carbon price could create for steel and cement, through demand reduction and efﬁciency improvement (Figure 2.3). The results demonstrate the level of uncertainty associated with current estimates, and also point to the need both to pursue efﬁciency

28

*

Climate policy mix

improvements in production and to use the carbon-price signal to encourage substitution by lower-carbon commodities. Substitution opportunities involving lower-carbon inputs and services exist along the value chain. The carbon-price signal provides the incentives for companies and consumers to choose the suitable lower-carbon substitutes.

Efﬁciency improvements A carbon price increases the price of fossil energy and therefore creates incentives to use fossil energy more efﬁciently. It becomes more attractive to insulate or retroﬁt a building to make it more energy-efﬁcient, to buy a more efﬁcient car or to replace an energy-intensive electric engine with a carbon price, because the energy savings translate into bigger cost savings and can therefore justify higher expenditures on energy efﬁciency. Thus, carbon prices can contribute to improving energy efﬁciency and to reducing CO2 emissions. In many instances, the carbon price alone is unlikely to unlock the full energy efﬁciency potential, as the actors investing in energy efﬁciency may not beneﬁt from the fuel savings, or because limited information, other priorities and inertia of infrastructure and decision processes delay the response to carbon prices. In such instances, complementary regulatory instruments, as discussed in section 2.3, are desirable. However, without carbon pricing, regulations that are prescribing or subsidising the use of energy-efﬁcient or low-carbon technologies, might have a reduced impact owing to the ‘rebound effect’ (Sorrell 2007). If the increased efﬁciency from the insulation of houses reduces fuel consumption and heating costs, households could increase room temperatures or might stop turning down the heat when the house is unoccupied. Thus, the envisaged energy-demand reduction from efﬁciency measures can be partially off-set by a ‘rebound effect’. Carbon pricing compensates for the reduced fuel costs and avoids the rebound effect.

Technology policy

*

29

Summary Pricing carbon creates incentives for efﬁciency improvements and for substitution by low-carbon inputs, products and services. The existing capital stock will take some time to adjust to the impact of climate policies. Credible climate policy and carbon pricing will be important to ensure that during this period investment decisions anticipate the future carbon costs (Chapter 4). For many private, public and corporate decisions, carbon costs play only a marginal role and therefore do not capture the attention of decision makers. Also, investors in commercial projects and houses often do not bear the carbon costs of their future operation. For these reasons, complementary policies play an important role in shaping the transition to a lowcarbon economy.

2.2 The role of technology policy The large-scale reduction in emissions that is necessary to limit global temperature increases requires a combination of improvements in energy efﬁciency and large shares of low-carbon energy sources (Hoffert et al. 1998; Edenhofer et al. 2006) and, in particular, sources of energy without CO2 emissions (Caldeira, Jain and Hoffert 2003). In other words, resolving climate change requires the widespread use of a broad portfolio of climate-friendly technologies. What are the implications for technology policy? Traditional technology policy was primarily focused on enhancing the level of innovative activity or favouring a particular national industry – for example, through tax breaks for research and development or export subsidies. With the public focus on low-carbon development, it is also of interest to explore whether technology policy can be more targeted so as to accelerate innovation that contributes to climate objectives. The ﬁrst requirement for targeted technology policy is that governments have the ability to deﬁne such

30

*

Climate policy mix

climate objectives. For the energy system, it makes sense to divide such objectives into supply-side (energy carriers and storage) and demand-side technologies. On the supply side, technology innovation is unlikely to result in a discovery of totally new sources of energy. Thus, the objective to increase quality, efﬁciency and scale, while reducing costs, can be clearly deﬁned for wind, solar power, biofuels and marine energy technologies (wave and tidal stream). For carbon capture and storage as well as for nuclear power, additional tests are necessary to ensure storage satisﬁes long-term climate and safety requirements. For energy carriers, innovation will have to explore multiple pathways to ﬁnd the technologies that might ultimately be most suitable, for example, for long-distance renewable energy transport. They include different types of high-voltage technologies and possibly superconductivity for electricity transmission, as well as hydrogen and various synthetic fuels. Storage requirements will increase with higher penetration of intermittent renewable energy sources. The roles of hydro storage, innovative battery technologies, hydrogen and biomass/synthetic fuels will depend on how their performance develops with innovation. As the technology provides direct services to users, it is more difﬁcult to project and deﬁne requirements. While governments can clearly specify the needs for electricity and provide a tailored feed-in tariff, they will struggle to deﬁne the future communications interests of the general public. In this case, it is therefore more important to provide general support for innovation than to support speciﬁc solutions. Economics and engineering literature provide policy makers and negotiators with a rich source of information on technology and mechanisms to stimulate technology (see Bazilian et al. 2008 for a review). The literature characterises innovation systems with different linear and non-linear frameworks. Four stages of technology innovation are typically depicted in one linear trajectory: research

31

Country A

*

I. Research, development and demonstration

II. Deployment and diffusion

Country B

Technology policy

I. Research, development and demonstration

II. Deployment and diffusion

III. Creating long-term prospects for low -carbon technologies

IV. Using synergies from international co-operation

Figure 2.4 Stages of technology innovation and use

and development (R&D), demonstration, deployment and diffusion. More-detailed analysis shows that innovation is not a step by step process. Rather, it develops in a ‘chaotic’ way and depends on interactions between a variety of actors and coincidences (Lundvall 1992). The complex descriptions are, however, difﬁcult to operationalise. Figure 2.4 adds a second country to the linear structure, indicating the international nature of technology innovation, and thus also the opportunities that international co-operation offers (Cust, Grant Iliev and Neuhoff 2008). Four points of potential public policy intervention are depicted in the ﬁgure and structure the discussion in the remainder of this section.

I. Research, development and demonstration In the R&D phase, technologies face barriers related to proof-ofconcept and basic technological feasibility. The successful development of low-carbon products, processes and services furthermore depends on the innovative capacity of public and private bodies in a country, their access to international experience and resources and their funding and incentives. In the demonstration phase, new barriers emerge, relating to ramping up from lab to full scale. Upon the ﬁrst implementation of a technology in the ‘real world’,

32

*

Climate policy mix

the broader viability of a product, service or process is explored and economic, social and institutional barriers can emerge that a technology will encounter further along in the innovation process. Demonstration projects offer the opportunity to build an initial network of companies and people that can deliver the different components of the technology. Technology analysts and policy makers generally agree that the economic system does not enable private-sector investors to capture the full beneﬁt of their innovation. Despite the legal protection afforded by patents and intellectual property rights (IPR), Margolis and Kammen (1999) estimate that private returns on R&D across various sectors are between 20 and 30 per cent, while social rates of return are around 50 per cent. In sectors where markets are inﬂuenced by regulation, additional risk exists that governments will adjust regulation after successful innovations in such a way that investors will not capture the full beneﬁt (Grubb 2004a). Energy producers are not exposed to the full environmental costs of carbon and on the contrary continue to receive subsidies in many countries. This further reduces the beneﬁt for investors of low-carbon and energy-efﬁcient technologies and associated innovations. As private-sector investors do not capture the full beneﬁt of innovation, they also have less incentive to pursue innovation. To compensate for this, many countries use tax incentive schemes to encourage increased private-sector R&D efforts. The UK, for example, allows business to deduct 130–170 per cent of expenditure on R&D from income when determining corporate tax levels (HMRC 2010). Despite this, ﬁnancing resources for technologies for mitigation and adaptation make up only a small share (probably less than 3.5 per cent) of the resources devoted globally to all technology development and transfer, with most (probably over 60 per cent) of the resources provided by business (UNFCCC 2009b). Hence ‘several studies have concluded that funding for climate-related energy R&D should be increased two- to ten-fold’ (IEA and OECD 2009).

Technology policy

*

33

To increase the level of support and to ensure targeted research is pursued it might thus be equally important to scale up publicly funded R&D programmes. While in theoretical models public R&D programmes seem to allow for the most targeted support, in practice they face several challenges. They typically involve signiﬁcant volatility – for example, caused when political processes delay or change programmes in unpredictable ways, thus creating funding gaps that can undermine the ability to retain experts. Furthermore, the quantiﬁcation of overall research needs and their direct beneﬁts is difﬁcult, and thus public programmes often lose out if research with long-term beneﬁts competes for budget with shorter-term, more tangible public policy objectives. Finally, without established methodologies to quantify the need for speciﬁc research budgets, sharing of research funding between technology options is biased towards continuation of existing patterns, reﬂecting the interests of dominant stakeholders. This might explain why 50 per cent of European public funding for energy continues to be allocated to nuclear fusion and ﬁssion (IEA 2007).3 Intellectual property rights create opportunities for companies to receive return on R&D investment, thus incentivising future R&D investment. In the renewable energy ﬁeld, a wide set of companies own patents for individual technology components (Lee, Iliev and Preston 2009). This raises concerns that patent owners could at some point inhibit or slow down technology transfer and technology diffusion rates through exorbitant licensing rates, high information and negotiation costs for providing technologies and even by deliberately strategically blocking the use of technologies and creating barriers to the diffusion of innovation (Alic, Mowery and Rubin 2003). As any change to the intellectual property rights regime might impact on rents or reallocate rents, proposals to adjust the regime face ﬁerce opposition. This raises the question as to how other sectors succeed in advancing and diffusing complex new technologies on a global scale (Iliev and Neuhoff 2009).

34

*

Climate policy mix

Co-operative IPR arrangements seem to play an essential role for innovation and diffusion of high-tech products, and can include co-operative standard bodies, cross-licensing agreements and/or patent pools. Co-operative standards allow innovative companies to contribute to individual components of a supply chain, using common standards to interface with other components of the technology. For mobile telephones, drastic cost reductions have been possible through value-chain diversiﬁcation, specialisation of different actors in speciﬁc value-chain niches, as well as greater customisation. For low-income countries, process learning and adaptation to local conditions are of central importance, typically building upon existing technology developed elsewhere (World Bank 2008). Patent pools or standards bodies can contribute to the adaptation of low-carbon energy technologies to suit the climatic conditions and capabilities of domestic manufacturers and maintenance engineers in many developing countries. Multinationals often face high costs of operating in small markets or do not have the suitable capacity, and therefore third parties are needed to adopt and diffuse the technology. With the multitude of patents covering many components of any technology, new entrants risk infringing patents held by incumbent companies, resulting in delays and the signature of licensing agreements on unfavourable terms. Patent pools or standards bodies avoid such risks while offering IP owners the prospect of some revenues. This suggests that where leading companies in a sector initiate IP sharing agreements or public access to industry standards, this also addresses the basic IP requirement to facilitate technology adoption and diffusion in developing countries. However, where such an initiative is not present, this will probably not only disadvantage global technology development but might also be of particular concern for developing countries. Public intervention to facilitate co-operative IP arrangements might offer beneﬁts under such circumstances.

Technology policy

*

35

Co-operative IP arrangements do not necessarily prevent IP disputes, but offer processes to address these conﬂicts that are often quicker and cheaper. However, much of the success of this conﬂict resolution effort depends on the interests of powerful members in ﬁnding a solution – for example, avoiding a trade conﬂict between the EU and the USA. Will such conﬂict-resolution strategies work if (i) the ‘co-operation’ is not entirely self-motivated, but rather imposed by public policy intervention; and if (ii) one of the conﬂict parties is a smaller developing country, or a company active in such a country? In such instances, it is likely that a stronger role of governments will be necessary to ensure a rapid and cheap conﬂict resolution. In the case of developing countries, additional incentives might be required to adapt a technology to country-speciﬁc circumstances. Private parties that contribute to the adaptation of a technology cannot retain ownership of the technology – for example, with patents (World Bank 2008). This undermines the very incentives to pursue such adaptation. Possible solutions include a move towards applied R&D agencies focusing on outreach, testing, marketing, commercialisation and dissemination activities (Sagar, Bremner and Grubb 2008).

II. Deployment and diffusion As a technology reaches maturity, deployment and diffusion become important. Technologies in the deployment phase are not yet commercial because of cost barriers and therefore require advance investment. Can private actors shoulder the advance investment and recover it through future market revenue? In the mobile telephone and car industries, new products offer additional services that allow companies to charge a premium that ﬁnances innovation. In the pharmaceutical industry, patents are effective in protecting IPR. However, in many other sectors patents can be circumvented – for example, through small alterations to engineering designs – and thus patents offer only short-term protection or mainly serve as signalling devices for investors.

36

*

Climate policy mix

For technologies where the beneﬁts of innovation are less easy to appropriate, targeted support schemes are important. For example, energy-generation technologies offer a largely undifferentiated product consisting of engineering components, which can be redesigned to circumvent patents. In addition, much of the cost reduction is expected to result from mass production. This involves the expertise of, and innovation by, many suppliers of equipment (Jacquier-Roux and Bourgeois 2002). It is difﬁcult to envisage how a group of companies could agree on making large-scale investments in developing new technologies when the exact nature of the ﬁnal technology is not clear. Technology-speciﬁc policy is therefore required (Sandén and Azar 2005). Strategic deployment programmes create lead markets for technologies that are otherwise not yet competitive in a broader market. Governments use, for example, feed-in tariffs or tradable certiﬁcate schemes to increase the price paid for power generated with renewable energy sources. Thus, manufacturers of renewable technology can operate proﬁtably by selling their products in a viable market. The premium paid for renewable energy constitutes an investment of a society in a new energy technology. Governments that engage in technology-speciﬁc support can use the opportunity to require from private companies transparent and public reporting about the technologies and their performance. This can accelerate technology learning, and provides a basis for ongoing management of the strategic deployment programme. Government also needs to retain sufﬁcient institutional independence so as to be able to abandon support programmes should the technology not satisfy expectations. Technology-speciﬁc assessments are necessary for the appropriate timing of strategic deployment programmes (Nemet and Baker 2009). Premature deployment support could create wasteful investment or even misdirect research focus on shortterm technology solutions with limited long-term potential. Delayed deployment support can delay availability and large-scale use of potentially crucial low-carbon technologies.

Technology policy

*

37

The history of wind energy illustrates the beneﬁts of strategic deployment programmes as market-based approaches that provide continued incentives for technology improvement. Research-led attempts in Germany, the USA and other countries in the early 1980s, which focused on building multi-megawatt wind turbines as demonstration projects, failed both on engineering and cost grounds (Norberg-Bohm 2000; see also Bergek and Jacobson 2003). Meanwhile, in Denmark, the adoption of an energy plan in 1981 outlining a target of 10 per cent wind contribution by 2000 created private-sector conﬁdence in the future of the industry (Karnøe 1990). Private and subsequently public initiatives supported a wide range of R&D in small and large wind turbines (Jensen 2004). Through application experience, turbine manufacturers learned how to address design challenges, and turbine sizes gradually increased (Grubb and Vigotti 1997). Denmark provided approximately $1.4 billion in subsidies between 1993 and 2001, creating successful technology and a vibrant industry: annual revenues of Danish wind companies by 2001 were $2.7 billion, the vast majority of which came from its dominant position in export markets (Foxon 2003). In the diffusion phase, a technology that has become cost competitive evolves from a ‘new approach’ to become standard. For this, the regulatory and institutional framework must be further simpliﬁed, to make the use of the technology effortless. It also requires that expertise be available across society for the operation and maintenance of the technology. In order to increase political and public support for a transition to a new energy technology, some element of local manufacturing is necessary, and might require further development of skills, expertise and industrial capacity.

III. Creating long-term prospects for low-carbon technologies The more promising are future markets for low-carbon products, the more resources investors and corporations will dedicate to developing and improving them in order to capture a large share of this future

38

*

Climate policy mix

market. Thus, long-term policy frameworks can increase investor conﬁdence in the future demand for a low-carbon technology. For example, a government credibly committed to an emissions target is more likely to implement policies that make use of low-carbon and energy-efﬁciency technologies to achieve the target. A credible policy framework increases the conﬁdence of private actors that a strategic deployment programme will be continued until the technology either becomes commercially viable or is proven unsuitable. Thus, companies supplying a product (e.g., off-shore wind turbines) are more likely to use their best engineering teams. Otherwise they would be reluctant to take risks in an uncertain market. The more stable the strategic deployment programme, the more effective it can be in delivering innovation – unexpected demand increases and reductions both limit innovation. Unexpected demand increases typically result in scarcity prices and large rents for manufacturers, creating incentives to maximise current output at the expense of experiments to improve manufacturing performance (Neuhoff, Lossen et al. 2007). Demand reductions undermine investor conﬁdence in the new technology, again limiting resources available for innovation. Carbonpricing schemes create ﬁnancial beneﬁts that do not hinge on the implementation of technology-speciﬁc support schemes. Thus, they reduce risks from delays of technology-speciﬁc support schemes in the political process – for example, prompted by incumbent companies concerned about competition from new technologies. Carbon pricing works through three channels. First, it reduces the cost gap that companies or public subsidies experience and must cover while the costs of the new technology exceed the prices of conventional technologies. Second, with carbon pricing, a new technology will be cost competitive against conventional technologies sooner, thus reducing the time over which learning investment occurs. Finally, the proﬁtability of a new technology will be higher in the future as it competes against a more expensive (because more carbon-intensive) conventional technology.

Technology policy

*

39

International co-operation can offer commitment opportunities to increase the credibility of such long-term policy frameworks. For example, the European Renewables Directive creates time frames, targets and incentives for national governments to implement policies that result in renewables deployment (European Commission 2009b). Also, private actors are increasingly pursuing innovative activities that have prospects for a global market. This can also guide the design of technology policy – for example, by using the global scale to accelerate accumulation of knowledge and enhance stability of frameworks for investors in innovation.

IV. Using synergies from international co-operation International markets reduce the dependence of technology producers on demand from individual countries. Particularly for technologies that require support from strategic deployment programmes and are thus dependent on national policy frameworks, the opportunity to sell to several countries reduces the volatility of demand. This is illustrated in Figure 2.5, using the example of wind turbines. Domestic policy regimes in individual countries have changed radically over time and so has local demand. At one time wind turbine producers would have struggled to build up production capacity and the necessary expertise for technology innovation on the basis of the volatile demand of any one of the markets. However, the aggregation of demand in the global environment has resulted in a relatively stable and relatively steady increase in demand for wind turbines on the global level since the early 1990s. Thus, access to several national schemes can give technology companies the conﬁdence to invest in innovation, as a multi-country market is less exposed to the policy and regulatory uncertainties of individual governments. While national markets are often dominated by a limited number of technology companies, international markets increase the level of competition. This makes prices more cost reﬂective – that is, more informative and lower. Thus, international markets support governments that require robust information for the design of strategic

40

*

Climate policy mix

24,000

France Italy UK Germany India Spain

22,000 20,000 18,000 16,000 14,000 12,000

China

10,000 8,000

Denmark

6,000 4,000

USA

2,000

06

04

02

00

98

96

94

92

90

88

86

84

82

08 20

20

20

20

20

19

19

19

19

19

19

19

19

19

19

80

0

Figure 2.5 Installed wind power per year in MW. Source: Based on IEA, GWEC and Worldwatch Institute

deployment policies, and they reduce costs for tax-payers and consumers. The potential of providing a successful technology to serve a global market also increases the incentive to invest in innovation at both corporate and national levels. However, the focus on a global market should not preclude opportunities to explore different technological pathways. Technology innovation is inherently uncertain, and thus it will be globally beneﬁcial if various approaches to using a speciﬁc renewable energy source are pursued in parallel, accelerating the accumulation of experience. Synergies from global market integration need to be balanced by the opportunities of pursuing different technological solutions in different countries. International co-operation can accelerate the accumulation of experience. It can create opportunities for cost and knowledge sharing and facilitate co-ordination on standards for compatibility of technology components and frameworks for licensing. The existence of global markets or the accessibility of technologies is not sufﬁcient to promote the adoption of a technology in all countries. While adoption of more efﬁcient technologies offers large development

Targeted measures and regulation

*

41

co-beneﬁts, weak internal diffusion persists across most developing countries. The capacity of a country to adopt and operate technologies in local industry is a deciding factor. It is often impeded by a lack of local skill and expertise (World Bank 2008). A combination of policy to stimulate demand for low-carbon and energy-efﬁciency technologies can be paired with technical co-operation and assistance to support the development of local manufacturing (Bazilian et al. 2008).

Conclusion Innovation in low-carbon and energy-efﬁciency technologies is an essential component of low-carbon development. For many technologies, clear innovation objectives can be deﬁned, thus also opening the opportunity to pursue tailored technology policy. Robust frameworks exist to guide the tailored design of policy instruments, including provision of direct research support and the design of strategic deployment programmes. However, for products and services that aim to serve speciﬁc consumer demands, it might be less suitable to have governments deﬁning technology policy objectives and preferable to have more generic innovation-support schemes. In this situation, guidance on targeting innovation towards low-carbon options will emerge from climate policy frameworks and from policies such as carbon pricing that prompt low-carbon attributes for new products and services. For the achievement of low-carbon objectives, it is essential that suitable support frameworks be in place to ensure the diffusion of innovative low-carbon technologies to developing countries as well.

2.3 The role for targeted measures and regulation The most commonly used framework for assessing the opportunity to reduce carbon emissions in our economies is marginal abatement cost curves. Figure 2.6 depicts the potential reductions in emissions that can be delivered with different technical measures. The vertical

Gas plant CCS retrofit

Abatement cost per tonne CO2e 80 60 40 20

Reduced slash and burn agriculture conversion Reduced pastureland conversion

Lighting – switch incandescent to LED (residential) Appliances electronics Motor systems efficiency First-generation biofuels

Iron and steel CCS new build Coal CCS new build Coal CCS retrofit

Grassland management Organic soils restoration

Cars full hybrid

0 –20 –40 –60 –80 –100 –120 –140 –160 –180 –200

5

10

15 Geothermal

Rice management Small hydro Waste recycling Efficiency improvements other industry Landfill gas electricity generation Clinker substitution by fly ash Building efficiency new build Insulation retrofit (residential) Tillage and residue management Cropland nutrient management Cars plug-in hybrid Retrofit residential heating, ventilating and air conditioning Second-generation biofuels Appliances, residential

20

35 38 Abatement potential Gt CO2e per year Concentrated solar power Reduced intensive agriculture conversion High-penetration wind Solar photovoltaic Low-penetration wind Degraded forest reforestation Pastureland afforestation Degraded land restoration Nuclear 25

30

Figure 2.6 Global greenhouse gas abatement cost curve beyond business-as-usual (v. 2.1) to 2030. Note: The curve presents an estimate of the maximum potential of all technical greenhouse gas abatement measures below €80 per tonne CO2e if each lever was pursued aggressively. It is not a forecast of what role different abatement measures and technologies will play. Source: McKinsey and Company 2010.

Targeted measures and regulation

*

43

axis shows the estimated costs of reducing one unit of CO2 for each measure. The measures towards the left side of the ﬁgure, such as insulation of buildings, have ‘negative abatement’ costs: the energy savings that can be achieved exceed the investment costs for improved buildings insulation or efﬁcient lighting systems. This is reﬂected in the conclusions of an EU green paper on energy efﬁciency, which reported that at least 20 per cent of energy could be saved in a cost-effective manner (European Commission 2005). An extensive set of economic, sociological and policy studies have examined why these cost-effective opportunities to save energy and carbon emissions have not been realised. The main barriers that have been identiﬁed include inadequate information and interest, a lack of access to low-carbon products and insufﬁcient credit to ﬁnance the advance investment in energy efﬁciency and insulation (Hassett and Metcalf 1993). Institutional misalignments also create barriers between investors and beneﬁciaries. These are exempliﬁed by the tenant–landlord relationship: the landlord must pay for insulation but only the tenant beneﬁts from lower heating bills. Governments are increasingly implementing policies to address these market failures and realise low-cost opportunities to reduce energy consumption and carbon emissions (DeCanio 1998). Many measures are cost effective even without a carbon price and could be implemented once various barriers are removed. This leads to the question of whether there is a role for carbon pricing in promoting the use of low-carbon technologies. This chapter suggests there are four beneﬁts. First, carbon prices increase the cost of production with traditional carbon-intensive technologies. This creates market opportunities for producers that use low-carbon technologies and do not therefore face such a cost increase. For example, carbon capture and sequestration (CCS) is more expensive than other conventional approaches. It requires plants to satisfy additional design criteria and to be equipped with compression facilities, and the captured carbon must be

44

*

Climate policy mix

transported to geological storage sites. Thus, CCS will be utilised only where conventional technologies (which produce carbon emissions) face higher costs from carbon pricing. In theory, the prohibition of carbon-intensive technologies could also provide incentives for investment in low-carbon technologies. However, it is unlikely such regulation could micromanage all carbon- and energy-related activities efﬁciently. Regulation will therefore be a complement to, rather than a substitute for, carbon pricing. Second, carbon pricing can be designed to have additional impacts on decision-making processes. Often, company managers or private consumers are comfortable with existing technologies and processes, and are reluctant to change procedures and habits, or they disregard existing technologies. The design, implementation and visibility of emissions trading, with its reporting requirements and compliance obligations, can contribute to focusing a management’s attention on implementing energy-efﬁciency or direct carbon-emissions-reduction measures, accelerating the low-carbon transition. Third, it is often argued that governments should prescribe measures to improve energy efﬁciency and emissions reduction rather than use carbon prices. However, this would in many instances result in regulations that require consumers and companies to make choices that are not cost effective for the individual. Hence, enforcement of the regulation would be difﬁcult. Carbon pricing reﬂects the environmental externality in electricity and fuel prices, and thus makes compliance with energy-efﬁciency regulation cost effective for individuals, reducing the difﬁculty of enforcing compliance. Fourth, regulations prescribing or subsidising the use of energyefﬁcient or low-carbon technologies face the rebound effect. As improved energy efﬁciency reduces energy costs, it also reduces energyrelated expenditure. Some of the saved expenditure might be used to acquire additional energy services, such as longer heating periods or travel to further destinations. Combining regulation to support energyefﬁciency measures with carbon pricing can ensure that the overall

Managing distributional implications

*

45

expenditure on energy services stays constant and can thus avoid the rebound effect. In summary, carbon pricing improves the cost effectiveness of lowcarbon and energy-efﬁciency technologies and widens the application of these technologies. Many energy-efﬁciency measures are not implemented despite their cost effectiveness, which suggests a need for complementary measures to overcome the barriers that restrict their use.

2.4 Managing distributional implications The most widely discussed and most directly observable distributional impacts of climate policy are product price increases that result from carbon pricing. As a result, carbon pricing can have equity implications, which need to be carefully considered. This section ﬁrst explores how and when the carbon-price signal is transmitted from producers to consumers. Then the implications for producers and consumers are discussed separately.

Who bears the cost of low-carbon policies? Carbon prices create additional costs for industrial producers as well as domestic consumers. The ﬁrst question to answer is: who bears the costs? In most instances, companies can pass industry-wide carboncost increases through to product prices, as intended by carbon pricing. Final consumers will pay higher prices and bear the costs (see the power example in Box 2.1). Consumers, however, are not a homogeneous group, and there is some concern that the poor will face disproportionate cost increases. Carbon-pricing policies create additional revenue streams from carbon taxes or auctions of carbon allowances, and some of these revenues could be used to compensate consumers for the higher costs. One starting point for this discussion is, whether ﬁrms increase the price of their products for the price of carbon allowances used during production – for example, are costs passed through? The European

46

*

Climate policy mix

experience suggests that in liberalised power markets all carbon costs are passed to power prices (see Sijm, Neuhoff and Chen 2006; Matthes 2007). If utilities receive allowances at no cost, they will beneﬁt from the introduction of an emissions-trading scheme. In the absence of liberalisation, the regulatory regime matters – for example, under costbased regulation, utilities can pass on carbon prices if they incur real costs, but not if they receive emissions allowances at no cost. In the non-power sector, the discussion on the extent of cost passthrough is still ongoing. This is because it is too early for robust empirical evidence from the EU ETS to resolve the debate (for analysis of the b o x 2 . 1 Prices, cost-pass through: power-sector example Various technologies can be used to generate electric power. The amount of CO2 they emit per MW h of electricity generated differs. For example, a combined-cycle gas turbine produces about 0.48 tons of CO2 per MW h of electricity, while a typical existing coal power station emits about 0.95 tons of CO2 per MW h. Therefore, their production costs are also affected by varying carbon prices. A CO2 price of €20/ tonne CO2 increases the generation costs for a gas plant by €9.6/MW h and for the coal plant by €19/MW h. 100 90 80

Marginal costs hard coal power plant

60

Future contract for power (year + 1, base)

/ EUA

/ MW h

70

50 40

Future contract for CO2 allowances (year + 1, until 09/2005 Spot)

30 20

Future contract hard coal (year + 1)

10 0 2003

2004

2005

2006

2007

2008

2009

2010

Figure 2.7 Power prices versus fuel/CO2 costs in Germany (year ahead 2003– 2010). Source: Matthes 2010b.

Managing distributional implications

*

47

b o x 2 . 1 continued The price of electricity in competitive markets is set by the most expensive unit that is required to meet electricity demand. For many continental European markets the marginal units in most hours are coal power stations. Figure 2.7 illustrates how in Germany the price of the one-year forward contract for electricity can be explained by the combination of coal and carbon prices. In countries where gas turbines set the marginal power price, such as the UK and the Netherlands, the power price increased roughly in line with the carbon cost of combinedcycle gas turbines (CCGT). Experience from liberalised markets in Europe conﬁrms that carbon costs are passed through to power prices, despite free allocation of allowances (Sijm, Neuhoff and Chen 2006). This is because free-allowance allocation is a ﬁnancial transfer, and allowances can be sold proﬁtably if not used. Power companies will only use allowances to produce electricity if it is more proﬁtable than selling the allowances. Eventually, power consumers complained about the windfall proﬁts amassed by power companies that received allowances at no cost and subsequently charged higher power prices. These complaints triggered a move towards full auctioning of allowances for the power sector post-2012. In liberalised power markets, all power generators receive the price of the marginal unit. Therefore, low-carbon technologies, such as hydro or nuclear, beneﬁt from the electricity price increase. Higher prices, however, are borne by ﬁnal consumers. Governments will have to pursue policies to compensate poor households for higher electricity costs induced by higher carbon prices. The incentives to reduce energy consumption can be retained if initial units of electricity are provided at lower cost (so called live-line tariff), energy efﬁciency measures are subsidised and direct transfers are used as a compensation measure.

cement sector, see Walker and Richardson 2006 and for industryreported pass-through, see McKinsey and Company and Ecofys 2006). A longer observation period is required because pricing decisions are less transparent, and the overall carbon price impact is smaller. For products such as cement, prices are often set in annual contracting rounds, thus delaying the adjustment of product prices to changes in the carbon price.

48

*

Climate policy mix

While such delays complicate the analysis of carbon price pass-through, delays of one or two years do not inﬂuence who bears the costs in the long term. This leaves two main reasons that could prevent industry producers from passing through the full carbon price. First, innovation and substitution of lower-carbon alternatives: companies cannot pass on the full carbon cost to their product if competitors succeed in producing the same product less carbon-intensively. Also, if carbon pricing increases the costs of carbon-intensive commodities, consumers will shift to lower-carbon products and services. This falling demand, in turn, results in somewhat lower prices until producers have adjusted production capacity. This is a desirable effect of market-based approaches to reduce the proﬁtability of inefﬁcient or undesired products and creates an incentive for innovation and efﬁcient investment. Second, international competition with companies that face lower carbon prices or no carbon prices, or can reduce the price pass-through of carbon prices; obviously, companies can pass on input cost changes if all their competitors face similar cost increases. This has been demonstrated with input factors such as oil and commodity prices or exchange rates (Walker 2006; Hourcade et al. 2007). But what happens if some of the competitors do not face similar cost increases because they are producing in countries that do not impose carbon prices? Chapter 6 analyses industrial activities in more detail, and points to speciﬁc products that face signiﬁcant cost increases from carbon prices. Where these products are actively traded between regions with and without carbon prices, companies might not pass on the full carbon price to product prices.

Implications for consumers What will be the implications of carbon pricing for consumers? Studies typically ﬁnd that poorer consumers spend a larger proportion of their income on fuel- and energy-intensive products than richer consumers (Baker and Koehler 1998). Therefore, cost increases relative to income are higher for poorer consumers than for richer consumers (Cornwall and Creedy 1996; Labandeira and Labeaga 1999;

Managing distributional implications

Expenditure (–) and rebate/recycling (+) relative to household income (percentage)

1.5

*

49

US auction/lump-sum rebate (Congressional Budget Office 2000)

1 Lump-sum recycling (Parry 2004)

0.5 0 –0.5 –1 –1.5

US free distribution/decrease in corporate taxes (Congressional Budget Office 2000)

Estimates of impact of European carbon tax payments on households

–2 –2.5

Poorest

Second quintile

Middle income quintile

Third quintile

Richest

Figure 2.8 Estimates of carbon tax impacts and of redistribution policies for households (original studies scaled to carbon price of €20/tonne CO2). Source: Smith 1992; Congressional Budget Ofﬁce 2000; Parry 2004.

Smith 1992; Symons, Proops and Gay 1994; Symons, Speck and Proops 2002; Wier et al. 2005). Figure 2.8 shows the cost increase faced by different consumer segments from a carbon price of €20/tonne CO2 for heating and transport fuels. While the estimates differ across European countries (grey area), the cost increase relative to income is consistently biggest for the low income consumer segments.4 What really matters is not the increase in energy expenditure but the impact after government decides how to redistribute the additional revenue to compensate consumers or industry for their cost increase. * The Congressional Budget Ofﬁce (2000) rebate curve studied the impact of distributing the revenue from carbon pricing equally among all citizens. In this case, poor consumers will beneﬁt from the introduction of carbon pricing because they use the least energy and energy-intensive products.

50

*

Climate policy mix

By contrast, if government revenues from carbon pricing are used to reduce corporate taxation, the overall distributional implications are regressive, as suggested by the corporate tax reduction curve (Congressional Budget Ofﬁce 2000). Rich consumers own a larger share of companies and thus beneﬁt from the introduction of a scheme that recycles revenue through the reduction of corporate taxes. Poor consumers bear the cost increases of carbon-intensive products and services without compensation, and are thus worse off. * The analysis by Parry (2004) shows that, with care, it is possible to balance the use of revenues from carbon pricing in order to avoid any distributional impacts. Consumer-level data still mask differences within consumer segments, such as different transport and housing patterns in urban and rural environments. More generally, consumers choose different products and services, and will therefore be affected differently by carbon pricing. Compensation based on the average level of cost increases faced by the consumers in a segment will result in some consumers beneﬁting from the introduction of carbon prices, while others do not. Compensation schemes would have to be linked to individual activities to ensure all consumers were compensated for their speciﬁc energy-cost increase. This would require that consumers who spend more of their money on transport be compensated for their disproportionate cost increase under carbon pricing. Such compensation, however, would reward transport usage and undermine the incentive that carbon pricing is meant to provide: to substitute high-carbon consumption choices with lowercarbon products and services. Thus, the different levels of carbon emissions produced by consumers with similar income levels might become the biggest political challenge for the implementation of ambitious carbon prices. It is easier to balance differential impacts of carbon pricing on rich and poor consumer segments with the revenue created under carbon pricing policies.5 The experience of petrol taxes provides evidence of the overall viability of carbon pricing approaches. In order to reduce fuel *

Managing distributional implications

*

51

import dependency, increase revenues, cover road costs and reduce environmental impacts, European countries have continued to increase petrol taxes to a level that by now equals €280 and €330/ tonne CO2 in Germany and the UK, respectively (Matthes 2010a).6 The low-carbon transition will require a shift of economic activity from carbon- and energy-intensive products and services to lower-carbon alternatives. This creates opportunities for new companies to enter the ﬁeld or expand their activities, and thus undermines the competitive advantage of companies that are established in the carbon-intensive world. Where management of incumbent companies is too hesitant to embrace the low-carbon transition and to shift corporate strategy to a low-carbon business model and technologies, jobs, business relationships and the value represented by the company to its investors are all at risk. As with the introduction of information technologies, governments play an important role in providing appropriate training and education so as to ensure continued employment and the availability of necessary skills. These will be required either within incumbent companies that successfully manage the transition or within new companies that make use of new market opportunities. Governments can also provide the regulatory framework and support companies in the low-carbon transition. Government support will therefore be essential to prevent negative impacts for individual employees and to foster the necessary skills for a low-carbon transition. One challenge for governments in the low-carbon transition will be to prevent continued support for companies that fail to pursue the low-carbon transition. Such support is expensive, prevents opportunities for employees to shift to long-term viable activities that are compatible with climate constraints and rewards management and investors that have demonstrated inability to anticipate and respond to climate and market requirements. Continued support would undermine the effectiveness of any market mechanism and governance structure to select and reward good decision makers, and thus puts future economic success of companies and countries at risk.

52

*

Climate policy mix

However, if companies are owned by domestic shareholders and pension funds, then governments will struggle to impose stringent regulations where this would risk proﬁtability of the companies. Analytic frameworks to assess whether business strategies are compatible with climate objectives will allow pension funds to shift their investment portfolio to less-risky companies. Thus negative impact pensions can be avoided, and governments can pursue stringent regulation.

Summary The low-carbon transition will require careful management to avoid adverse distributional impacts. This involves training and education, where necessary, to facilitate the shift from carbon-intensive to lowcarbon employment opportunities, and transparent private- and public-sector reporting to allow pension funds and investors to reduce exposure to companies that fail to develop business models compatible with a low-carbon world. Implementation of carbon pricing creates rent transfers between different households. These need to be carefully assessed and addressed to avoid social inequality and maintain support for climate policy.

2.5 Conclusion A multitude of economic models have attempted to simulate the impacts of climate policy on economic growth. The GDP in a simulation with stringent climate policy is typically close to the GDP simulated in a model without such policy instruments. How is it that our economies can be characterised by different future trajectories with similar growth potentials? Economists used to explain this with network effects and non-convexities of technology. The network effect can be illustrated using the example of transport networks. A city might beneﬁt from a well-integrated road network for car-based transport or from a well-integrated public transport system. In this example, one might therefore expect a prosperous

Conclusion

*

53

Regulation.information

Technology policy

Pricing carbon

Climate policy

Balancing distributional impacts

Figure 2.9 Three pillars of climate policy

future with either transport system. If, however, two separate systems are not integrated and only cover parts of the town, then transport services are less satisfactory and potentially more expensive, undermining economic development. Non-convexities of technologies can be illustrated using the learning-by-doing phenomenon. The value that a new technology can provide to society increases with the experience of the users, operators and manufacturers, resulting in higher-quality, more-tailored solutions and lower production costs. Trajectories involving new technologies therefore tend to be more expensive in the initial years, but might offer equal or higher growth potential in the medium term. For economic policy instruments, this implies that it is no longer sufﬁcient to assess their short-term impact (for example, on economic growth). In addition, it is necessary to assess whether they move the economy onto a desired economic development trajectory. Thus, four dimensions of policy instruments need to be considered in parallel, as summarised in Figure 2.9. First, carbon pricing creates incentives for the use of more energyefﬁcient and lower-carbon technologies, allows producers and

54

*

Climate policy mix

consumers to substitute away from carbon-intensive products, and creates some incentives for innovation in low-carbon technologies. Second, regulation, information provision, institutional set-up and other policies address non-market barriers for increased use of energyefﬁcient and low-carbon technologies and services and accelerate the low-carbon transition. Third, technology policy can range from direct R&D support to strategic deployment programmes for new low-carbon and energy-efﬁcient technologies. Harvesting the synergies between the policies will allow for effective and low-cost emissions reduction. Sustained policies will also increase the credibility of the overall decarbonisation strategy. The emphasis given to the different pillars and policies implemented will depend on national circumstances. Finally, political support and long-term viability of policies will depend on the careful balancing of distributional impact – supporting people in their transition to employment based on low-carbon products and services, reducing exposure to investment in businesses that do not have a low-carbon business strategy and using some revenues from carbon-pricing schemes to address inequalities.

Notes 1 To achieve 550 parts per million of CO2, three of four models reported from comparison exercises of the Energy Modelling Forum (EMF) (see special issue Energy Economics, 26(4) (2004)) calculate around 1% GDP losses. Nine of eleven models of the Innovation Modelling Comparison Project (IMCP) calculate GDP losses of 1% or below to achieve 550 parts per million. Five of eleven IMCP models achieve 450 parts per million at less than 1% GDP loss (Edenhofer et al. 2006). Fischer and Morgenstern (2006) conclude that actors with perfect foresight generate lower abatement costs. 2 Demand elasticity for steel is assumed to be in the range 0.1–1.6 and for cement in the range 0.3–2.0. Carbon-efﬁciency improvements at €20/tonne CO2 are expected in the range of 5–20% for steel and 15–30% for cement. Product prices are €350 and €76/tonne for basic steel and cement, respectively. 3 The other half of energy funding is split between fossil fuels (13%), renewable energy (29%) and power conversion, transmission and storage (8%).

Conclusion

*

55

4 Rather than current income, Metcalf (1999) and Poterba (1991) also compare cost increases to lifetime measures, or total expenditure (Wier et al. 2005). 5 One approach sometimes discussed is to grant personal carbon credits to every consumer. This lump-sum transfer can ensure that poor consumers are not disadvantaged under carbon pricing. This can also offer additional beneﬁts, by improving information and raising awareness on carbon emissions, but risks high transaction costs. However, so far, complexity and high transaction costs have prevented further pursuit of the approach. 6 Based on the assumption of a 0.65 and 0.77 €/litre tax in Germany and the UK, respectively.

three

Implementing a carbon price: the example of cap and trade

Governments can put a price on carbon either by imposing a tax on carbon emissions (following Pigou 1920) or by using cap-and-trade schemes (following Coase 1960). Both concepts are simple in theory and have been discussed for decades. Ellerman et al. (2010) review the history of emissions trading and refer to Crocker 1966 as the ﬁrst explicit expression of an emissions market. CO2 tax schemes were set up in Sweden in 1991 and subsequently in Denmark, Finland, the Netherlands and Norway, while cap-and-trade schemes for SO2 and NOX were established in the USA during the same period. Cap-and-trade schemes have four basic components. (i) Governments set a cap on the total volume of emissions of a pollutant and create the corresponding volume of allowances. (ii) These allowances are distributed for nothing or sold to companies and individuals. (iii) The allowances can then be freely traded. This creates, in principle, economic efﬁciency. Companies that would face high costs to reduce their emissions will buy allowances from companies with lower costs, thus reducing the total cost of emissions reduction. (iv) Emissions are monitored and reported and, at the end of the accounting period, companies have to surrender to government allowances proportional to the volume of their emissions and can bank remaining allowances to the following year. 56

Cap and trade

*

57

Companies covered by a carbon cap-and-trade scheme face, in principle, the full price of carbon emissions. This is obvious where the company has to buy an allowance from another company or in an auction directly from the government. Even, however, where companies receive allowances at no cost, the principle applies. The company that received a free allowance at the beginning of a trading period has the option of either selling this allowance in the market or using it to cover emissions associated with its production. When using the allowance, the company will forgo the revenue from selling it. This is the opportunity cost of using allowances that are allocated at no cost. Section 3.4 discusses how different allocation provisions can distort this principle and can reduce the incentive to improve carbon efﬁciency of production or undermine the substitution effect. The alternative approach of a carbon tax requires companies and consumers to pay a tax proportional to the volume of carbon emissions associated with speciﬁed activities. Thus, in principle, carbon taxes have the same effect as cap-and-trade schemes, where allowances are sold by governments. Here the term ‘carbon pricing’ is used to indicate analyses that apply equally to both economic instruments. Differences to cap-and-trade schemes will be pointed out, in particular with respect to their political economy (this chapter), investment decisions (Chapter 4) and international links (Chapter 5). The term ‘emissions trading’ is often used in Europe as a synonym for cap-and-trade programmes, but it also describes schemes without an absolute cap for the total emissions volume, such as voluntary trading schemes in the USA.1 Carbon off-setting programmes in developing countries that deliver emissions reduction are also covered by the term emissions trading. Developing countries have not capped their emissions, and so far have not implemented cap-andtrade schemes. They participate in emissions trading via the clean development mechanism (CDM). Under CDM, certiﬁed projects in developing countries can sell credits from emissions reduction to developed countries that accept these credits within their cap-and-trade

58

*

Cap and trade

schemes. Thus, linkages created by emissions trading can put a price on carbon even in countries that have not capped their emissions. The example also illustrates that emissions trading does not always create a substitution effect. Producers in developing countries do not pay for carbon-intensive production; instead, they are paid for investments to reduce emissions. Thus, their production costs and competitive product prices do not increase to reﬂect the carbon price. This chapter ﬁrst outlines the fundamentals of cap-and-trade schemes using the example of the US SO2 emission scheme. The positive experience with the SO2 schemes triggered the discussion about the beneﬁts of cap-and-trade schemes over carbon taxes in domestic climate policy. Section 3.2 describes the background and main features of the EU ETS. Drawing on this experience, section 3.3 discusses lessons that emerge about setting a cap in such a scheme. Section 3.4 describes the lessons learned from distortions and inefﬁciencies associated with different methods of allowance allocation. Section 3.5 explores the rationale for deciding on the sectors covered by a scheme and section 3.6 concludes.

3.1 The SO2 trading programme in the USA Sulphur dioxide (SO2) emissions emerged in the 1950s in a number of industrial countries as a major environmental concern, with strong impacts on human health. In response, Japan legislated direct emissions controls for SO2 in 1962 that required the use of ﬂue gas desulphurisation for coal plants from 1968 (Law Concerning Controls on the Emission of Smoke and Soot). From 1973, a levy on sulphur oxide (SOX) emissions was imposed that gradually increased to a level of ¥300/tonne SO2 by 1987. The legislative measures at the national level were complemented with agreements between government and industry at regional level. In the USA, SO2 emissions were ﬁrst regulated by the 1970 amendments to the 1963 Clean Air Act, which required the use of ﬂue gas desulphurisation for new coal power plants (called the New Source Performance Standard). Similar requirements for existing plants

SO2 trading programme in the USA

*

59

could not be implemented, and as a result it was often more proﬁtable to continue the operation of old power stations than to build new plants that had to comply with the more stringent standards required for upgrades or new investments. This created a disincentive for the replacement or upgrading of old plants. Cap and trade was ﬁrst implemented in the US Clean Air Act of 1990 for SO2 emissions from large emitters. The scheme created a cap for total SO2 emissions that allowed the government to compensate existing plants with some free-allowance allocation in order to gain political support. Phase I ran from 1995 to 1999, with phase II beginning in 2000. The total cap was divided into allowances, each allowance being the equivalent of one ton of SO2 emissions (Ellerman et al. 2000). These allowances were distributed by US states to emitters proportionally to their historic fuel input multiplied by a benchmark emissions rate. A small share of the allowances was retained and subsequently auctioned on an annual basis. Owners of allowances can use them to cover their emissions or trade the allowances. Trading gives emitters the ﬂexibility to choose between (i) reducing emissions to the volume of their allocation; (ii) reducing emissions below the volume of their allocation by investing in desulphurisation, or closing down and selling excess allowances, or banking them for future use under more stringent caps; or (iii) continuing to operate the installations at a high emissions level and buying allowances to cover the extra emissions from other market participants. As companies traded more actively and the government auctioned allowances, the market for SO2 allowances became more liquid and the price more informative (Figure 3.1). The ﬂexibility, together with a liquid market, allowed companies to optimise across plants even where they are owned by different companies. Thus, the least-cost operation and investment choices can be pursued to comply with the emissions target. An important component of the cap-and-trade scheme is monitoring, reporting and veriﬁcation. At the end of a ﬁnancial year each

60

1,800 1,600 Price (nominal $/t)

1,400

*

Emission exchange Fieldston EPA auction

Cap and trade

Early sales Cantor Fitzgerald

1,200 1,000 800 600 400 200

Ja

Ja

nu a

ry nu 19 ar 92 Ja y 1 nu 99 3 a Ja ry 1 nu 99 a 4 Ja ry 1 nu 99 5 a Ja ry 1 nu 9 a 96 Ja ry 1 nu 99 a 7 Ja ry 1 nu 99 a 8 Ja ry 1 nu 9 ar 99 Ja y 2 nu 00 a 0 Ja ry 2 nu 00 a 1 Ja ry 2 nu 00 ar 2 Ja y 2 nu 00 3 a Ja ry 2 nu 00 ar 4 Ja y 2 nu 00 5 a Ja ry 2 nu 0 ar 06 y2 00 7

0

Figure 3.1 Evaluation of prices under USA’s SO2 trading scheme

emitter has to surrender one SO2 allowance for every ton of SO2 emitted (Ellerman et al. 2000). If the emitter does not surrender the necessary volume of SO2 allowances there is a penalty of $3,000 per excess ton of SO2 emitted.2 The SO2 cap-and-trade scheme in the USA is generally regarded as a success (Carlson et al. 2000). The design choice and use of freeallowance allocation created the political alliances that allowed for the implementation; the ﬂexibility offered by the cap-and-trade approach delivered the emissions reduction necessary to achieve the agreed targets. There was a lively debate about the extent to which experience of the SO2 trading scheme could be directly translated to CO2 cap-andtrade schemes because there were some market differences between the schemes. The SO2 cap-and-trade scheme was implemented against the background of pre-existing regulation, which ensured availability of emissions and technology data. Furthermore, the technology required for SO2 emissions reduction was already widely used for new installations. This simpliﬁed the discussions on the SO2 emissions caps, as it was clear that they could be achieved and it was possible to

EU emissions-trading scheme

*

61

estimate the maximum cost industry would incur. The main objective of the scheme was therefore cost minimisation and securing political support for the implementation. As the value of SO2 allowances is only a fraction of the value of carbon allowances, the unconditional allocation of allowances to emitters was politically less contentious.3 The positive feedback from the US cap-and-trade programmes, both for SO2 and for NOX, was one of the starting points for discussions of the EU ETS.

3.2 The European Union emissions-trading scheme In 1992, the European Commission proposed a Europe-wide carbon tax (European Commission 1992). But industry opposition, German concerns that carbon taxes would beneﬁt nuclear power and the reluctance of some Member States to support a common European tax policy prevented progress of this policy. After the European negotiators accommodated US requests for an inclusion of emissions trading in the Kyoto Protocol (UNFCCC 1997), the European Commission reconsidered the greenhouse gas management strategy and shifted from a tax-centred approach to the creation of carbon markets (Ellerman, Convery and de Perthuis 2010). An additional push for a uniﬁed European approach to climate policy was the joint opposition of the EU Member States to the US withdrawal from the Kyoto process. After senior politicians across Europe had criticised the USA’s behaviour they had to demonstrate domestic action themselves. European Member States had announced in the negotiations of the Kyoto protocol that they would deliver emissions reduction of 8 per cent by 2008–2012. By 2002, the increasing scepticism about the ability of governments to negotiate and enforce stringent voluntary agreements was conﬁrmed,4 and there was increasing dissatisfaction about the level of ambition and compliance with these voluntary agreements.5 Also, in the UK, support for the national emissions-trading scheme

Kyoto budget

Cap and trade

State D

*

State C Kyoto budget

Kyoto budget

State B

State A Kyoto budget

62

Annex 1 trade (assigned units) …

Allocation

Installations opting out before start of trading period

EU ETS trading period

Project credits • clean development mechanism • joint implementation

Banking (not pilot period) Next trading period

Figure 3.2 Structure of the European Union emissions-trading scheme

declined. The UK government had negotiated intensity-based and sometimes absolute emissions targets for individual sectors and companies and signed climate-change agreements. Companies could bid to reduce emissions beyond this baseline, and then trade reduction in emissions relative to the baseline. However, it turned out that the stringency of the climate-change agreements was lax and the price for emissions reduction certiﬁcates fell below €6/tonne CO2 (Smith and Swierzbinski 2007; Morgenstern and Pizer 2007). The UK trading scheme and a similar approach in Denmark created the risk of a multitude of schemes that would not be consistent with the ideal of a single European market, which is an overall objective of European integration. This gave additional urgency to a harmonised European approach. In October 2003, the European Parliament and the Council of the European Union passed Directive 2003/87/EC, which required all EU Member States to implement a common cap-and-trade scheme by 2005. Figure 3.2 illustrates the main features of the EU ETS. EU Member States have under the Kyoto Protocol an overall emissions-reduction target that was distributed among Member States in a subsequent burden sharing agreement, which sets a total carbon budget to cover

EU emissions-trading scheme

*

63

national carbon emissions. The EU ETS only covers 10,800 powerand energy-intensive industrial installations. Therefore, each Member State has to decide what fraction of their carbon budget they want to make available for these ‘covered’ sector (grey area), and how much of the budget they retain for the emissions from transport and other domestic and smaller industrial activities. This split is ﬁxed within each Member State and therefore the cap for the covered sectors is set before the trading period starts. At the same time, the covered installations and therefore the demand are determined. Both are speciﬁed in national allocation plans that are proposed by Member States, approved by the European Commission, and then enacted by the national parliaments. Thus, regulatory uncertainty that could otherwise result from later government intervention is minimised. The EU ETS was designed to function independently from the international context. This ensured that the scheme could be implemented even though the Kyoto Protocol had not been ratiﬁed and was not in force at the time the Directive was passed. It also ensures the continuation of the scheme post-2012 irrespective of the international situation. The Kyoto Protocol allows for some reduction in emissions in developing countries to be credited against emissions in developed countries. Therefore the ‘Linking’ Directive allows companies to implement projects that reduce carbon emissions in other countries that signed the Kyoto Protocol. Reduction in emissions is only counted if it would not be realised without the ﬁnancial support from the carbon credits (‘additionality criterion’). The processes differ for projects implemented in countries that have accepted binding targets under the Kyoto Protocol (so-called joint implementation (JI) projects)6 and to signatories to the Kyoto Protocol that have no binding targets (so-called clean development mechanism (CDM) projects). To ensure that countries not only buy project credits to meet their emissions-reduction targets but also pursue policies to reduce their

64

*

Cap and trade

domestic emissions, the Kyoto Protocol states that reduction in emissions under JI and CDM has to be supplementary to domestic emissions reduction. This has been translated into EU legislation by limiting the amount of JI and CDM project credits each installation can use to cover its emissions. While the limits differ across countries, their European average is 13.8 per cent (Vasa 2010). Some developed countries that signed the Kyoto Protocol, like Belarus, Ukraine and Russia, have emissions budgets (so-called assigned amount units (AAU)) to cover their emissions that far exceed their expected emissions in the period 2008–2012. This is a result of the unexpected emissions reduction from the economic downturn and subsequent economic transition of the former Soviet Union and associated countries. Under the Kyoto Protocol, the successor states are allowed to sell assigned units that are not required to cover domestic emissions (often labelled ‘hot air’) to other developed countries. Their use is controversial, because emissions have decreased due to nonclimate-related factors. Also, because their supply exceeds demand after the USA did not sign the Kyoto Protocol, it is unclear whether their trade creates scarcity and thus incentives for emissions reduction in the selling countries (Grubb 2004b). To insulate the EU ETS from uncertainties associated with AAU trading, the direct use of such assigned amount units was not permitted in the scheme. Two indirect channels for trade of assigned units to inﬂuence the volume of allowances in the EU ETS are illustrated in Figure 3.2. First, Member States can buy assigned amount units and increase the share of allowances they devote to EU ETS in their national allocation plans. This is, however, only possible before the beginning of a trading period (for example, in 2007 for the period 2008–2012). Consequently, it does not create uncertainties during the operation of the scheme. Second, many Member States have invested in CDM and JI projects to obtain project credits to cover some of the emissions in the non-EU ETS sector. Member States can instead decide throughout the trading period to buy AAUs to cover these emissions – for example, from

Setting the cap

*

65

Russia – and to sell their CDM and JI project credits to installations for use within the EU ETS scheme. Finally, banking of allowances can inﬂuence the supply–demand balance within one trading period. Emitters and, more broadly, any party with an address in the EU that has registered to participate in the trading scheme can bank an allowance for an unlimited period. It is expected that in the case of excess supply in the market such banking will induce entities to save allowances and thus ensure continued scarcity of allowances and positive prices. However, as the ﬁrst trading period (2005–2007) was designed as a pilot phase, banking of allowances from that period towards the second trading period (2008–2012) was not permitted. Thus the integrity of the second trading period was projected from potential difﬁculties in the pilot phase and it was ensured that the volume of allowances in the second trading period would not exceed the Kyoto target.

3.3 Setting the cap: too many cooks spoil the broth Two aspects of EU ETS are here discussed in detail: ﬁrst, cap setting, and then the allocation of allowances under the cap. In the European policy process both steps were closely linked. In the pilot phase I, Member States determined their national allocation plans, specifying the emitters covered by the scheme and the volume of allowances allocated to individual emitters. The plans also determined the volume of allowances retained for new plants that will be commissioned during the trading period and the volume of allowances to be auctioned during the trading period. Thus, the national allocation plans determined the volume of allowances issued by any one country, and the sum of all national allocation plans set the cap for EU ETS. The balance between the overall cap and the demand for these allowances by emitters covered by the scheme then determines the market price for allowances. If governments set lenient caps, allowance

66

Gas price increase

30

Financial crisis triggering economic recession and emission decline

25 20 15

February 2010

October 2009

June 2009

February 2009

June 2008

October 2008

February 2008

June 2006

October 2006

February 2006

October 2005

June 2005

February 2005

October 2004

-

October 2007

5

June 2007

No banking of phase II European Union phase I European Allowances Union Allowances phase I European Union Allowances

10

February 2007

Carbon price ( /t CO2)

Cap and trade

Identification of new mitigation options Release verified emission data

40 35

*

Figure 3.3 Price of EU emissions allowance for phases I and II (European Energy Exchange and European Climate Exchange)

prices are low. Stringent caps push up prices so as to induce additional measures to reduce emissions. Figure 3.3 shows the prices at which allowances for the pilot period 2005–2007 were traded. By 2005, forward contracts for allowances in the second phase (2008–2012) were trading, and are depicted in parallel in the ﬁgure. Throughout 2004 several Member States submitted national allocation plans with lenient caps. As a result the price at which forward contracts for allowances were traded dropped to about €8/tonne CO2. The main reason for this over-allocation was that allowances were allocated at no cost. Such free allocation increases the opportunity and motivation for lobbying. Furthermore, governments had limited and inaccurate information on emissions at the level of individual emitters and on abatement opportunities of different industries. In the pilot period (2005–2007), the European Commission did not have sufﬁcient leverage to request extensive cuts to the volume of

Setting the cap

*

67

allowances issued and allocated by Member States. The EU Directive on Emissions Trading only required ‘the total quantity of allowances allocated . . . [to] be consistent with assessments of actual and projected progress towards fulﬁlling the Member States’ contributions to the Burden Sharing Agreement among EU Member States’. Member States argued that their overall national allocation plans for the period 2005–2007 were generous because emissions reduction in the short term was harder to achieve and progress would speed up after 2007. State aid rules were the main instruments the European Commission used to reduce the volume of allowances allocated by Member States, and thus the overall cap. If allocation provisions implied more allocation of allowances to installations than their expected emissions, this was deemed an unwarranted state subsidy. In a draft French plan and subsequently in the submitted Polish plan the European Commission requested a reduction of allowance allocation to individual emitters. The Commission also requested cuts to the caps in Italy, as the implied reduction in the level of emissions was far too small relative to any reasonable trajectory towards the Kyoto commitment. These cuts did contribute to an increase of the allowance price in spring 2005. Fuel prices had a strong inﬂuence on the allowance price. In the ﬁrst half of 2005, market participants anticipated that a shift from electricity generation by coal power stations to gas power stations would be required to deliver emissions reduction. At the same time, natural gas prices increased. Thus, the carbon price at which it was economical to replace production of coal power stations by gas power stations also increased. Hence, market participants traded CO2 allowances at higher prices. As the natural gas prices continued to rise, the carbon prices required to switch would have exceeded €30/tonne CO2, but at this stage market participants investigated what other abatement options were available. For example, in the German power sector,

68

*

Cap and trade

a shift from lignite to hard coal generation was viable once allowance prices exceeded €20/tonne CO2. Observing this mitigation option, traders anticipated what would sufﬁce to deliver the reduction target, and the allowance price dropped to the corresponding level. Also, the cold late winter in 2005 followed by a hot summer inﬂuenced perceptions, and there may have been a structural reason for the increase: namely, the power companies in EU 15 (old Member States) were generally short of allowances and wanted to buy, but some of the industrial companies who were long were not interested in selling. This illustrates how uncertainties about the fundamental drivers for allowance prices resulted in creative explanations for price drivers and price volatility – which is typical for learning by market participants. In April 2006, the publication of veriﬁed emissions data for 2005 triggered a massive price drop. The total volume of allowances allocated to emitters exceeded emissions in the year, conﬁrming early concerns of a lack of scarcity due to the chosen allocation methodology. In fact, the veriﬁed emissions data was leaked a few days before its ofﬁcial publication, which created proﬁtable opportunities for some market participants with early information, and resulted in a price drop before the ofﬁcial publication date. This episode demonstrated that it is important for governments carefully to manage commercially sensitive data. Despite the conﬁrmation of the large surplus in the market by the veriﬁed emissions data, the allowance price stayed at €15/tonne CO2. Various explanations circulated, including delays to the operation of registries in some EU countries, small emitters not selling surplus allowances and ﬁnancial institutions with an open position in the power markets buying allowances to support the carbon price and thus also the electric power price. But, by December 2006, most of these perceptions seemed to have been disproved or vanished, and the allowance price had dropped to a few euros. Thus, the pilot period de facto ended in early 2007 because of over-allocation.

Setting the cap

*

69

It is often argued that if banking between the pilot period and second period (2008–2012) had been permitted, prices would have been less volatile and would not have collapsed (Ellerman and Joskow 2008). This would have allowed traders to arbitrage away the price difference between the ﬁrst and second periods. But the separate pilot period might have been crucial as part of the ﬁrst-ever application in Europe to protect the integrity of the scheme. The pilot period delivered both important results and lessons. On the results side it delivered a working system of monitoring and veriﬁcation of emissions and a trading environment with all the necessary components. Measuring the results in terms of directly avoided emissions is difﬁcult because eighteen months of signiﬁcant carbon prices is too short a sample period. In addition, peaking gas prices in this period would probably have resulted in additional emissions from coal power stations replacing gas in the absence of EU ETS. Ellerman, Convery and de Perthuis (2010) estimate that between 2.5 per cent and 5 per cent of emissions were saved in the period 2005/2006 owing to the EU ETS, but acknowledge large uncertainties in particular with regard to the counterfactual. The most important result of the pilot phase, however, was that the scheme alerted industrial and power companies to the issue of carbon prices. Nowadays carbon impacts are evaluated for all projects and are an inherent part of management decisions in most sectors. One important lesson from the pilot phase was the revelation of excess allocation of allowances, which can be attributed to four main factors: (i) poor data quality for historic emissions required for the basis of the initial allocation; (ii) the intrinsic optimism of industry about production growth and thus emissions increases; (iii) ﬂexibility granted to individual installations to exclude a year of low production from the base period; and (iv) insufﬁcient attention given to the impact of excess supply for other Member States. All Member States were lobbied by their domestic industries to increase their allocation. Such additional allocation created no direct

70

*

Cap and trade

costs for a Member State because there was no binding national cap for the pilot phase pre-Kyoto, and therefore national governments did not have to compensate extra allocation to the trading sector with additional mitigation efforts in other sectors. The negative impact of more allowances issued in one country was a reduction of overall scarcity in the European allowance market below the politically desired level, which is unlikely to receive much attention in domestic decision processes.7 The experience of over-allocation in the pilot phase threatened to be repeated when Member States developed their national allocation plans for phase II (2008–2012). A combination of imperfect data availability at the beginning of the allocation process and continued strong lobbying pressure resulted in proposed national allocation plans that envisaged allocation volumes that were incompatible with the Kyoto commitment and with a scarce allowance price. This time, however, the European Commission was in a stronger position to request changes to these plans, for three reasons. First, the low allowances prices from phase I demonstrated that cap and trade with lax caps cannot deliver scarcity prices. Second, climate change had moved up the political agenda and the European Commission had ensured that the administration received the necessary political support in negotiations with Member States. Third, the Kyoto targets provided clear criteria for the assessment of national allocation plans by the European Commission. The sum of allowances a Member State allocated to installations covered by EU ETS and the expected emissions from other sectors could not exceed the national targets.8 This formal and political power allowed the Commission on 26 November 2006 to establish clear quantitative criteria that national allocation plans had to satisfy. In its decision on the ﬁrst ten national allocation plans it required cuts of 63 million tonnes (Mt) CO2. It eventually reduced the allocation from 2,325 MtCO2 requested to 2,081 MtCO2, a reduction of 244 MtCO2,

Setting the cap

*

71

t a b l e 3 . 1 Emissions-trading-scheme emissions cap as proposed by Member States and accepted by EU Commission Proposed cap (million tonnes CO2) Germany 482.0 Greece 75.5 Slovakia 41.3 Sweden 25.2 UK 246.2

Allowed cap (million tonnes CO2) 453.1 69.1 30.9 22.8 246.2

representing a 10.5 per cent cut in proposed allocations and a 3.2 per cent reduction compared with 2005 veriﬁed emissions. Table 3.1 provides examples. Various countries threatened to sue the Commission on the decision – in particular the opposition of the German industry ministry was of concern. In the ﬁrst half of 2007, Germany was hosting both the G8 and EU Presidencies, and it would have been difﬁcult to lead international negotiations on long-term climate policies while undermining the credibility of the European ﬂagship policy instrument. Hence, the German government eventually accepted the decision of the Commission, while other Member States continued their law suits (van Zeben 2009). The excessive allocation by Member States points to an intrinsic difﬁculty of decentralising the cap setting to Member States. Any state faces the incentive to increase the cap, to satisfy either demands for free allocation by domestic industry or bigger auction revenues by its treasury, while the costs in terms of insufﬁciently scarce allowance markets are shared by all Member States. Therefore, the amendment to the EU Directive on Emissions Trading, passed as part of the Climate Package in December 2008, deﬁnes one common cap and a harmonised allocation methodology across Europe for installations receiving free allowances while the majority of allowances are to be auctioned.

72

*

Cap and trade

3.4 Distributing allowances: compensate or distort Economic textbooks usually state that the method of allocation does not affect the economic efﬁciency of cap-and-trade schemes because trading allows market participants to ﬁnd the least-cost emissionsreduction opportunities. The implicit assumption is that the allocation is based on one, ﬁxed, historic baseline. US cap-and-trade schemes for SO2 and most of the NOX programmes followed this model. Allowances were typically allocated for more than a decade using historic production volumes; the allocation then remained ﬁxed irrespective of subsequent operation, investment or even closure of the plant. The EU ETS, however, is characterised by initially short allocation periods, of ﬁrst three and then ﬁve years. Experience from these periods suggests that allocation decisions are based on recent information regarding individual emitters. Hence, the allocation of allowances is not, as in the textbook, a one-off transfer. Instead, future allocations are contingent on today’s operation, investment and closure decisions. Hence, owners of plants form expectations about how their current behaviour will inﬂuence future allocation decisions of governments. Expectation about future free-allowance allocation can thus distort today’s decisions. Repeated free-allowance allocation can create perverse economic incentives that reduce the efﬁciency of the emissions-trading scheme. (See Harrison and Radov 2002; Neuhoff, Rogge et al. 2006; Matthes, Graichen and Repenning 2005; Ellerman 2006.) Figure 3.4 uses three different power-generation technologies and the option to improve energy efﬁciency on the demand side to illustrate the efﬁciency of different allowance-allocation methods. Starting from a situation without emissions trading, it illustrates that it is cheaper to continue the operation of older coal power stations than to build and operate a new and more-efﬁcient coal power station. At typical gas prices, the production cost of gas-generated power is

Distributing allowances

*

73

Emission-based (E)

Energy efficiency

Fuel-specific benchmark (FB)

Gas

Uniform benchmark (UB)

New coal

Auction/grandfathering

Old coal

E FB UB

Investment/barrier costs

Production cost

CO2 cost Production cost

E FB UB

Production cost CO2 cost

Cost

E FB UB

CO2

Value of free allocation

Efficient production Choice of the best input Appropriate use of output

Figure 3.4 Power sector illustration of distortions from free-allowance allocation

again higher. Overall, cheap power limits the interest of companies and consumers in pursuing energy-efﬁciency measures. The carbon costs are highest for old, inefﬁcient, coal power stations while no carbon costs are incurred where energy efﬁciency measures are pursued. Thus, carbon pricing creates incentives for a substitution to more-efﬁcient power stations, less-carbon-intensive fuels (e.g., gas) and energy efﬁciency. However, as will now be discussed, repeated free-allowance allocation reduces some or all of these incentives.

Distortions from grandfathering with a moving baseline Where allowances are grandfathered using a moving baseline, the distortions from free-allowance allocation to existing facilities are strongest. Assume a plant received free allowances for the period 2005–2007 to match the average annual emissions in the period 1998–2002, and

74

*

Cap and trade

the same allocation methodology was repeated in the allocation for the period 2008–2012 matching average annual emissions in the period 2000–2004. The managers of the plant will use the historic precedent as the best indicator for future allocation, and assume that allowance allocation for the period 2012–2020 will match average annual emissions in the ﬁve-year period 2005–2009. Thus, reducing emissions by one unit in 2009 will reduce free allocation by one-ﬁfth of a unit in each of the years 2013–2020. Assuming a constant allowance value of €25/tonne CO2 and a discount rate of about 10 per cent, the reduced value of future free-allowance allocation is €18/tonne CO2 for every unit of reduced emissions. Therefore, the plant operator will only implement emissions reduction up to a cost of €7/tonne CO2. This effect is illustrated in Figure 3.4. Allocating allowances based on emissions in the previous period reduces the incentives for efﬁciency improvement in power generation, fuel switching and more efﬁcient use of electricity. It is called the ‘strong early action problem’ – lowering emissions too early reduces future allocation. If the EU ETS continues to use a grandfathering approach with a moving baseline, some emissions reduction could thus be postponed indeﬁnitely.

Distortions from benchmark allocation with a moving baseline Benchmarks offer an improved methodology of free allowance allocation. The volume of allowances is not linked to the actual emissions of a plant, but to the production of the plant. The production is then multiplied by a benchmark factor to determine the volume of freeallowance allocation. For example, the electricity production of a power station is measured in some base year and then multiplied by the carbon emissions the best available power station produces per unit of electricity production to determine the volume of free-allowance allocation. The beneﬁt of this approach is that the operator of the power station has an incentive to improve the efﬁciency of the

Distributing allowances

*

75

plant, and since this will not alter the volume of future allowance allocations, the owner keeps the beneﬁt from the sale of allowances. The discussion of benchmarks is frequently confused, because benchmarks can be deﬁned in various ways. First, the benchmark can either be directly applied to the output of a plant or can be related to the maximum throughput (capacity) that is multiplied by an industry-wide utilisation factor to determine the average output. Second, the measurement of the output or installed capacity can be based either on some historic baseline or on recent (possibly even current) data. From the perspective of economic incentives a benchmark will not create any distortions if it is based on a ﬁxed historic baseline. In practice, however, the base period for a benchmark is likely to shift towards recent observations with each new determination of freeallowance allocation. Sometimes it is argued that the reference time should be close to current production, but this creates incentives that can inﬂuence strategic and operational decisions of companies. Some of these incentives might be intended to shield speciﬁc commodities from the carbon-price signal if this price signal is not imposed globally, but it seems difﬁcult to avoid severe ‘unintended’ consequences of this approach. Figure 3.4 illustrates the effect of output-based benchmarks. If all generation technologies receive the same amount of free allowance allocation per unit of power generated (uniform benchmark), then their effective generation costs will all be reduced by the same margin. The ‘only’ inefﬁciency that appears in the theoretical model is that the power price does not reﬂect the carbon cost, which reduces the incentive to use electricity more efﬁciently. In practice, the benchmarks for free-allowance allocation have been higher for coal than for gas plants, while renewable and nuclear plants do not receive any allocation – they are fuel speciﬁc. This methodology creates the biggest subsidies for the carbonintensive fuels and consequently reduces economic incentives for

76

*

Cap and trade

efﬁcient fuel choices. Supporters of such fuel-differentiated benchmarks argue that in their calculations the fuel-speciﬁc benchmarks did not create any distortions. But such simulations are very sensitive to fuel price assumptions, and the supposedly well-designed schemes have perverse results if oil, gas or coal prices differ from the assumptions made by proponents when modelling their implications. In addition to these rather obvious distortions created by benchmarks, their use can have more subtle and possibly even more serious implications. First, if the benchmarks succeed in shielding some sectors from the carbon price, then these sectors will continue to invest in high-carbon assets and increase their emissions (Harrison and Radov 2002). This will push up the carbon price that is required to deliver overall emissions reduction, and increase the overall cost of climate policy (see also Böhringer and Lange 2005). Second, benchmarks that work effectively when used to collect data on industry performance and best practice may be vulnerable to gaming by companies if they are used to allocate allowances valued at hundreds of millions of euros. To limit gaming, benchmarks have to be very narrowly deﬁned and carefully administered, but this precision reduces the ﬂexibility offered by a market-based instrument and undermines the incentives for innovation in the production process and product speciﬁcations.9 Third, the different allocation decisions that emerged across European countries in the ﬁrst two phases of emissions trading suggest that deﬁnition and scope of benchmarks are driven by the political power of incumbent companies as much as by economic rationale. Free allowance allocation relative to auctioning offers incumbent companies an opportunity to lobby government for increased allocation of allowances, thus limiting the market opportunities and increasing the risk for new technologies and innovative companies.

Distortions from closure provisions Closure provisions can create incentives to keep in operation old and often inefﬁcient plants that might otherwise be shut down and

Distributing allowances

*

77

replaced. Most national allocation plans do not allow emitters to retain allowances beyond the period for which they are operating or operational. This creates a ﬁnancial beneﬁt for continued existence and thus can delay closure decisions, which creates an economic cost, as it results in a deviation from the least-cost operation. Consequently, the cost to move to a low-carbon economy is increased and the overall competitiveness of an economy is reduced.10 Although it would in theory be possible to avoid closure provisions in a national allocation plan, it is impossible for a government to make commitments about the speciﬁc allocation methodology for the next trading period. The allocation will result from the negotiations of the next allocation plan. It is difﬁcult to envisage any government allocating allowances to the previous owners of a plant that no longer exists. First, handing out valuable public resources without any direct tangible beneﬁt would be difﬁcult to explain to the public. Second, the previous owners have little weight in the bargaining process, whereas other emitters can threaten closure, with ensuing loss of jobs, innovation potential and future growth. Thus, it is in practice not possible to guarantee that an installation that closes will continue to receive free allowances in the national allocation plan for the next trading period. Where production is highly carbon-intensive and exposed to strong international competition, unilateral implementation of carbon pricing could result in relocation of production into areas with weaker, or no, carbon-price regulation. If such relocation results in the closure of plants, then the relocation could be delayed if closure provisions created incentives to encourage continued operation of plants. However, as these provisions will have to be closely linked to the production volume, they are likely to create similar distortions as discussed in the previous two sections.

Distortions from new-entrant allocation All national allocation plans retain some allowances in what is called a ‘new-entrant reserve’. This is probably one of the largest differences

78

*

Cap and trade

to previous SO2 and NOX cap-and-trade schemes. There are no such provisions in the US SO2 cap-and-trade schemes (Ellerman et al. 2000). Under the NOX budget programmes states allocate allowances within their Environmental Protection Agency (EPA)-set budget. Again, most states chose straightforward grandfathering for incumbents, while several states adopted updating, new-entrant and closure provisions. Their distorting impact on investment and closure decisions is smaller, as their value per power plant is a fraction of the value of carbon emissions (Martin, Joskow and Ellerman 2007). The initial justiﬁcation for the free allocation to new emitters in the European scheme was threefold. First, to ensure a fair treatment of all emitters, new emitters also need to receive free allowance allocation. In the absence of a free allocation to new emitters it would have been difﬁcult to justify the high levels of free-allowance allocation to existing facilities. Second, to avoid the risk that new emitters would not be able to buy sufﬁcient amounts of allowances in the market (i.e., free allocation to new entrants was used in order to postpone the need for auctions). If an auctioning system is not in operation, then new entrants would have to buy allowances from other emitters. It was argued that these emitters might exercise market power or might not engage in trade, thus constraining entry. Third, to compensate for some of the distortions resulting from closure provisions (as closure provisions create an incentive to retain old power stations in operation, new emitters also have to be subsidised in order to ensure companies replace inefﬁcient old plants). Although these are some tenable reasons to explain the role of new-entrant reserves, they also create several concerns. As was already discussed in the section on benchmarks, fuel-speciﬁc allocation can distort the investment and operation towards higher carbon technologies. Where benchmarks are applied in the power sector, only plants that use fossil fuels receive free allowances. Frequently, the benchmarks differ between gas and coal. Figure 3.5 illustrates that coal stations often receive more free allowances than gas-powered

Distributing allowances

*

79

1.2 Model coal power station, 6,000h, 43% efficiency

Mt. allowances/year

1.0 0.8

Model combined-cycle gas turbine (gas), 6,000h, 55% efficiency

0.6 0.4

Hard coal plant

0.2

Natural gas plant U K

nd s rla

y N et he

Ita l

d an Ire l

ar y H un g

n Sp ai

G er m an

y

0.0

Figure 3.5 Comparison of new entrant allocation. Source: Based on Neuhoff, Ferrario et al. 2006.

stations, and gas-powered stations often receive more than non-fossil power stations. This distorts investment decisions towards carbonintensive fuel choices, which increases future carbon emissions. In Germany, Spain, Hungary, Italy and the Netherlands coal generators received a much larger volume of free-allowance allocation in the period 2008–2012 than gas plants. Only the UK and Ireland provide the same allocation to both. These differences in allocation volumes incentivise new, carbon-intensive plants in countries with more generous new-entrant allocation. Owing to the exceptionally high level of subsidies under some proposed phase II national allocation plans, and the distorting effects of allocation decisions, the construction of coal power stations would have been more proﬁtable under the EU ETS than without an emissions-trading scheme (Åhman and Holmgren 2006; Matthes et al. 2006). The German national allocation plan to the Commission was an extreme case. It not only provided the highest allocation for new coal generation in general, but the draft Allocation Law also contained a provision allowing an even higher free allocation for new lignite-ﬁred installations. In addition, the proposed national allocation plan suggested the continuation of free fuel-speciﬁc allocation for fourteen

80

*

Cap and trade

years, which resulted in a surge of demand for coal power stations to be commissioned by 2012 and correspondingly high prices for contracts for the construction of the power stations, showing that new entrant allocation can increase costs of power generation. The proposal was not accepted by the European Commission. If today’s governments subsidise conventional power stations with ad hoc free-allowance allocation, then it is likely that future governments will do the same. But with ad hoc subsidies it is impossible to predict the future power prices that determine the revenue streams of today’s investments. This is essential in order for power plants and investors to have predictable revenue streams to ﬁnance their projects. Thus, new-entrant allocation undermines the basic principles of a liberalised power market and increases overall investment risk and costs. In 2008, the European Parliament and European Council followed a proposal of the European Commission to stop any further freeallowance allocation to the power sector post 2012, with some potential exemptions for new Member States. Thus, many of the distortions have been addressed and most coal-power projects cancelled.

Summary of distortions We can summarise the distortions described above in the following pyramid of distortions. Figure 3.6 illustrates that moving up the (stylised) pyramid eliminates distortions. It is important to note that free-allowance allocation for the period 2008–2012 does not create major distortions after its implementation. There are two notable exceptions. First, closure provisions create incentives to keep emitting installations operational until 2012. Second, newentrant allocations distort incentives for investment decisions. The main distortions from national allocation plans relate to their inﬂuence on expectations about the allocation post-2012. In contrast to this, expectations about the allocation and price levels post-2012 continue to inﬂuence investment and operational

Distributing allowances

Impacts

Distortions Allowance allocation method

*

81

More expenditure on Increase Less-energyextending plant life relative plant efficiency to new build operation investment Distortion Distortion Shields Reduces Discourages biased biased output (and incentives for plant towards towards consumption) energyclosure higher-emitting higher-emitting efficiency from plants plants average investments carbon cost

Auction

Benchmarking

Updating from previous periods

capacity only

X

capacity by fuel/ plant type*

X

X

output only

Y

output by fuel/ plant type*

X

X

X

X

emissions

X

X

X

X

X

X

Note: X indicates a direct distortion arising from the allocation rule. Y indicates indirect distortions if allocation is not purely proportional to output/emissions. * Differentiating by plant type adds additional distortions compared with purely fuel-based.

Figure 3.6 Pyramid of distortions of the EU ETS. Source: Neuhoff et al. 2006.

decisions. As illustrated above, even with a shift to capacity-related benchmarks post-2012, signiﬁcant distortions for investment and closure decisions remain for industrial installations. Only a commitment to full auctioning, as in the power sector, can eliminate all distortions. As the cost of buying CO2 allowances can to a large extent be passed on to consumers in the power and most other sectors, it is difﬁcult to justify continued free-allowance allocation. As the experience of the ﬁrst two trading periods illustrates, some initial free allocation is politically convenient to gain industry support, and might also be justiﬁed in terms of compensating industry for loss of value of high-carbon assets in a low-carbon world. Nonetheless, a fundamental difﬁculty of such compensation is that it implies governments should cover losses that a company incurs if it fails to make the right strategic decisions. The ﬁrst report of the IPCC was published on behalf of all major governments in 1992, and outlined

82

*

Cap and trade

the threats from climate change. Subsequent reports, and international negotiations, have underlined the validity of the analysis. Most of the assets that will be ‘compensated’ with free allocation are either fully depreciated or have only subsequently been built or acquired by their current owner. The free-allowance allocation will therefore reward managers for their failure to anticipate carbon policy despite publicly available information on the topic. No one would consider compensating an oil company for an oil ﬁeld that produces less oil than expected or a shoe producer if the trends of fashion change. Such compensation would create the wrong incentives for management and provide the wrong selection criteria for managers and successful companies. This suggests that trading schemes should also avoid compensating companies for losses in assets that forward-looking management could have anticipated and prevented with appropriate investment choices. There is a long history of government regulation to address environmental concerns, and it has been clear since the early 1990s that government regulation will be required to reduce anthropogenic greenhouse gas emissions. This shows that political economy drivers to compensate inefﬁcient carbon-intensive industries in order to gain their support have to be balanced with economic incentive schemes that reward innovative and forward-thinking companies and managers so as to deliver both environmental and wider economic objectives. The only argument that remains in use to support the possible continued use of free-allowance allocation relates to international competitiveness and leakage. As it is currently not clear to what extent other countries will implement stringent carbon-pricing schemes that expose producers to the full environmental costs from their emissions, there is a concern that carbon-intensive producers might relocate their production. Chapter 6 explores these concerns in more detail, and argues that for a very limited set of installations in some sectors the option to allocate allowances at no cost post-2012 should be retained for this case, and be explored in international

Distributing allowances

*

83

discussions together with other instruments that can be used to address leakage concerns.

Selling and auctioning allowances The European Commission package of January 2008 proposes to use auctions as the main allocation mechanism. This is a signiﬁcant shift from the ﬁrst two phases – the Directive on Emissions Trading required that in phases I and II of EU ETS at least 95 per cent and 90 per cent respectively of allowances are allocated at no cost. While in phase I almost all allowances were allocated at no cost, in phase II, Germany (9 per cent), the UK (7 per cent), the Netherlands (4 per cent) and Lithuania (3 per cent) reserved some allowances from free allocation, with Hungary, Austria and Ireland reserving less than 3 per cent of allowances. The German government had initially envisaged providing full free-allowance allocation, but then decided to use auctions to the extent permitted, largely in response to a public debate about the windfall proﬁts utilities made under free allocation. The change in Germany was too late for revisions in several Nordic countries, where initial ambitions for auctioning were hampered by the negative attitude in Germany. If allowances are no longer allocated at no cost, governments can either directly sell allowances to individual market participants or auction them. Direct sales are less transparent, creating the risk of market uncertainty and politically guided allocation. Thus, auctions have become the preferred approach, to create a clear and transparent interface between government and the private sector. The use of government auctions has developed a strong track record in government procurement and in sales of government bonds. The auctions of mobile phone licences have received signiﬁcant public attention. Auctions are run in many economic sectors, notably in gas and electricity markets on a daily basis.11 The question that has to be answered at the outset of any auction design is: what are the objectives of the auction? In the European

84

*

Cap and trade

context ﬁve main objectives were discussed. First, simplicity and transparency can enhance communication, participation and acceptance of the auction and the overall emissions-trading scheme. Second, low transaction costs, moderate information requirements, cash-ﬂow implications and low-price risk facilitate wide participation, including by small players. Third, the market-clearing price in the auction should reﬂect the value of allowances in the market. Fourth, the design should minimise the problems arising from collusion and abuse of market dominance or market power. And, ﬁfth, the design should help to maintain a liquid secondary market for emissions allowances. Two criteria that are applied in other auction contexts are less likely to be core objectives in the case of CO2 allowance auctions. Revenue maximisation should not be pursued in this case on the back of small emitters or players for whom trading is not a core activity, because these players are less informed and consequently will be disadvantaged. Efﬁciency of auctions relates to the question of whether the players who value the auctioned good at the higher price will buy the good in the auction. As carbon allowances are freely traded in secondary markets, it is not of concern in this case. The discussions in the UK and the wider European context have shown that the main objectives for an auction design are compatible with each other, thus avoiding difﬁcult trade-offs and negotiations with stakeholders who have different interests.12 A very simple uniform-price auction design can achieve all the objectives required. Such an auction can be repeated on a weekly, monthly or quarterly basis, which limits the credit and cash requirements for auction participants and avoids the risk that a buyer who acquires all allowances in an auction can subsequently dominate the market. If concerns remain, then provisions from government bond auctions can be replicated that limit the maximum share of allowances that can be bought by one auction participant and its afﬁliates.

Distributing allowances

*

85

A remaining question for the auction design in Europe, North America and Australia is what share of the allowances should be sold in forward auctions? For example, power companies sell electricity in forward contracts several years ahead of time, and then also sign contracts to cover their fuel purchases in the same time frame. If power companies anticipate receiving allowances at no cost, such hedging is not required. If the fraction of allowances allocated at no cost does not cover the forward power sales, then ‘forward auctions’ for allowances covering this volume of forward sales must be discussed. It is interesting to observe how the principles of auction design are reﬂected at the level of detailed implementation. For example, in the UK initially a group led by the environment ministry with industry stakeholders developed a shared perspective and recommendations for the objectives and implementation of the auction. The Debt Management Ofﬁce was then entrusted with a proposal for the detailed implementation. A subsequently published consultation document proposes to limit access to the competitive auction to selected intermediaries, most likely large banks (DEFRA 2007). This allows the Debt Management Ofﬁce to pass the responsibility for money-laundering checks to these intermediaries and replicates a model already applied for bond auctions in the UK. As such it might be a viable solution to the money-laundering requirement. It is, however, not clear how an auction design involving intermediaries impacts open access, simplicity and transparency, particularly if other European countries had followed this approach. The German government initially commissioned the KfW Bank, a government-owned development bank, to sell CO2 allowances on European spot markets for emissions allowances like European Energy Exchange (EEX) and European Climate Exchange (ECX). The objective for the bank was to achieve an average sales price equal to the volume-averaged sales price across the year and across

86

*

Cap and trade

European trading platforms. This provided government with some experience about spot markets, but also bridged the time until the government could run a tender for a third party to operate an auction dedicated to primary CO2 allowances. The tender was won by EEX. Since January 2010 EEX has been operating weekly auctions for 0.3 million emissions certiﬁcates and 0.57 million derivative contracts (emissions certiﬁcates for the following ﬁnancial years, to be paid at delivery). The auction was well received by the market and is typically oversubscribed by a factor of 6, and closely matches price results from daily spot and contract markets (+ 0.2 per cent and + 0.02 per cent for spot and derivative prices, respectively).13 For the period post-2012 the European Commission and the Member States are exploring what standards have to be satisﬁed and what level of harmonisation will be required across the auctions pursued by Member States. Commercial platforms that have been commissioned by individual Member States to pursue primary auctions on their behalf can also deliver this service for multiple Member States. Thus, gradual convergence or full integration of the auctions is possible. The European experience might have contributed to an accelerated move to auctions in other regions. For example, the Regional Greenhouse Gas Initiative (RGGI) covering emissions in ten northeastern and mid-Atlantic states of the USA held its ﬁrst auction in September 2008 and the ﬁrst compliance period started in January 2009.14 The states supply their allowances to a common auctioning platform and receive the auction revenue.

3.5 Sectoral coverage of a carbon-pricing scheme Carbon-pricing schemes that are currently implemented in Europe and proposed elsewhere differ in the range of sectors they cover. Whereas the European scheme covers power and industrial installations bigger

Sectoral coverage

*

87

than 20 MW (thermal), the US legislative proposals, for example, American Power Act (proposed by Senators John Kerry and Joe Lieberman in May 2010), had a broader coverage and also included transportation and heating fuels. The proposed Australian scheme extended coverage to the forestry sector. The philosophy underlying many of the discussions is that eventually coverage should be expanded to the whole economy. This raises two questions. First, should carbon pricing be applied to all sectors simultaneously or is there merit in using different price levels and policy measures across the sectors through a transition period? Second, if cap and trade is used as an instrument to deliver the carbon-price signal, should it be applied to a speciﬁc sector or cover all sectors? This section discusses the criteria that might explain why results depend on the starting point of countries and may differ across countries. Carbon pricing is implemented in a world of pre-existing taxation and regulation. Figure 3.7 illustrates the relative importance of energy taxes, regulation and carbon pricing for different sectors. Fuels to heat buildings are frequently subject to lower tax rates. Buildings are instead subject to increasingly stringent regulation for insulation or energy efﬁciency. Petrol taxes in the transport sector are relatively high – for example, in Europe, they are a multiple of even the highest prices observed under the emissions-trading scheme. In contrast, energyintensive industries have frequently been exempt from energy-related taxation and from regulatory energy-efﬁciency measures.

Beneﬁt of including a sector into the scheme Against the background of existing energy taxation, the case to include a sector in a carbon-pricing scheme depends on the role that the carbon price is expected to play in driving substitution, investment choices and technology innovation. The level of substitution delivered from carbon pricing will differ across sectors. The incentives will be higher for carbon-intensive products (after all, the purpose is to move demand from high- to

88

*

Cap and trade

Energy taxes

Transport

ice

pr

on Other rb ca s industry igh ost sh c re tion i u eq sac fect -R ran nd ef ebou pacts r -T s e s dres l im + Ad ributiona t s i D -

Energy-intensive industry

Carbon pricing

Buildings

Regulation

Figure 3.7 Importance for sectors of differences in energy taxes, regulations and carbon pricing

low-carbon choices). This can best be illustrated with two products with similar demand elasticities. Assume the manufacturing of the ﬁrst product results in twice the carbon emissions of the manufacturing of the second. A carbon price increase will raise the price of the ﬁrst product twice as much as the price of the second product. As a result of the bigger price increase, the demand and emission reduction for the ﬁrst product will be twice as great. This suggests that it is important to deliver the full carbon-price signal to sectors and activities that are carbon intensive. In addition, this will also create the strongest incentives for innovation of substitutes for the high-carbon products and services. The same argument dictates that the demand response of the carbon-price signal will be lower in sectors with high energy taxation – the cost increase relative to pre-existing costs will be smaller, as will be the demand response. Thus, implementing the

Sectoral coverage

*

89

carbon price will be most effective and should have priority in sectors that face low energy taxation or even receive energy subsidies, like energy-intensive industries. With regard to incentives for innovation, the carbon-price signal plays, in principle, the same role in rewarding low-carbon technologies across sectors. However, as discussed in Chapter 2, sectors differ in the extent to which the carbon-price signal will be the main driver for innovation. Thus, delivering the carbon-price signal is more important in industrial sectors with complex production processes, as governments have fewer opportunities to target support for innovation in such environments than in sectors like renewable energy, where strategic-deployment programmes can be used to drive innovation. From the perspective of the overall scheme, volatility of the allowance price could increase if large sectors that have low responsiveness to carbon prices are included. For example, if road transport were included in the cap-and-trade scheme, any changes in transport emissions would alter the scarcity levels of the trading scheme. With high pre-existing fuel taxes road transport is less responsive to the carbon-price signal. Thus, other sectors covered by the capand-trade scheme will provide most of the response to changes in emissions of the transport sector. This might create unexpected price changes and increase volatility of the carbon price. However, it could also result in a reduction of volatility as an increase in the number of sectors covered by the cap-and-trade scheme reduces the inﬂuence of any one sector (along a similar line in the international context, Aldy, Baron and Tubiana (2004) argue, ‘with more countries participating, emissions allowance prices would be subject to less uncertainty and variability’). A careful quantiﬁcation is required to evaluate the suitable sectoral scope for a cap-and-trade scheme.

Challenges of including sectors into the scheme The beneﬁts which can potentially be delivered by inclusion of a sector into a carbon-pricing scheme have to be weighed against the

90

*

Cap and trade

(political) costs created by the redistribution of rents, concerns about international competitiveness and transaction costs of the scheme. Politically, it is more difﬁcult to deliver a carbon price to sectors where it redistributes rents with signiﬁcant equity implications. This complicates the application of a carbon-price signal to sectors like domestic heating in countries where energy use for domestic heating is high. If, however, carbon prices gradually increase, as assumed in many projections, then it might be all the more difﬁcult to include these sectors in the future as they would then face an abrupt change from no carbon pricing to a high carbon price. Allocation of free allowances to domestic users, as envisaged in US legislative proposals, or using auction revenues to compensate households for cost increases, might be a more effective way to address the equity concerns than excluding a sector (see Paul, Burtraw and Palmer 2010). There is some concern that sectors producing internationally traded or tradable carbon-intensive products will redirect investment or shift production to countries with lower carbon prices. It is sometimes suggested that such sectors are excluded or shielded from the full carbon-price signal. However, these are the sectors where a functioning carbon-price signal would have the biggest impact on reducing demand and carbon emissions. After all, they are carbon intensive and often face low levels of energy taxation. Chapter 6 assesses which sectors would be at risk from leakage due to carbon price increases rather than demand reduction, and presents the different policy options to avoid such leakage. Transaction costs are frequently cited as a constraint on the inclusion of small emitters into a cap-and-trade scheme. In the EU ETS the threshold for ‘small’ emitters is set at 20 MW thermal power. The continued debate as to whether to increase or decrease this threshold suggests that the threshold is roughly correct, balancing the transaction costs of including smaller emitters in the EU emissions-trading scheme against beneﬁts. One way to reduce transaction costs is to

Sectoral coverage

*

91

apply cap and trade not at the level of individual emitters but further upstream. Carbon emissions from energy consumption are directly proportional to the consumption of fuel. Therefore, it is sufﬁcient to measure fuel consumption, which can easily be done at the level of reﬁning and imports. In an upstream scheme, reﬁneries or oil importers could be required to obtain allowances to cover carbon emissions that will result from the later fuel use. Despite the simplicity that upstream schemes offer, they have not been applied for large emitters because they are not suitable for freeallowance allocation that can create political support for the initial implementation, and they do not create monitoring, reporting and direct accountability at the emitter level, and thus forgo the direct involvement of emitters to encourage changes of operational and investment decisions.

Timing of inclusion From the perspective of individual sectors an early inclusion into the cap-and-trade scheme might have the beneﬁt that the sector is exposed to the trading scheme when the carbon-allowance price is still moderate. As governments pursue more-stringent climate policy, the sector will then adapt and innovate with gradually increasing carbon prices. This argument hinges very much on the level of conﬁdence people have in the potential of new technologies. If the analysts who argue that carbon capture and sequestration and other technology options can eventually deliver the necessary de-carbonisation at prices between €30 and €60/tonne CO2 are right, then there is little increase to be expected from carbon prices that have already reached €30/tonne CO2 in the pilot phase of the EU ETS. If future carbon price increases are not substantial, then the argument that early inclusion facilitates gradual adjustment to higher carbon prices is of little relevance.

92

*

Cap and trade

Governments’ responsibility for sectors in a carbon-pricing scheme Does inclusion of a sector into a carbon-pricing scheme relieve government from responsibility for emissions of that sector? If carbon pricing is understood to be only one component of a policy mix in which complementing policies are required, then governments need to retain overall responsibility for emissions. The management literature can help to ﬁnd the right balance – after all, it is a typical management challenge: how best to delegate a task while retaining overall responsibility for the delivery of the result. Indeed, the same act of balancing is required in climate policy as in management. If, on the one hand, a government takes its responsibility for overall emissions reduction as a mandate to micromanage the economy in order to deliver every ton of emissions reduction with a targeted policy, then the carbon-price signal, economic efﬁciency and incentives for innovation offered by the market-based approach are lost. If, on the other hand, a government is too hands-off, then market participants will not be able to deliver expected emissions reduction because, for example, diffusion of new technologies is not supported by appropriate institutional and regulatory frameworks. But, if the scheme does not deliver the necessary emissions reduction then government will eventually intervene and adopt different policies. Thus, a totally hands-off approach might undermine the credibility of a scheme as much as overly interventionist policies. A good carbon policy involves ﬁnding a balance between delegating to market-based instruments, providing regulatory and institutional frameworks and enforcing targeted technology policy. The right balance may well differ across sectors and countries, but will only be created by a government that accepts ultimate responsibility for the total emissions reduction. The discussion above points to various key criteria that have to be evaluated when deciding on the application of a carbon-price signal

Conclusion

*

93

to sectors, and in particular when deciding on the timing of inclusion into a cap-and-trade scheme. The differences observed across different countries illustrate both the different emphasis given to the criteria, and the different economic and political situation of the jurisdictions.

3.6 Conclusion Carbon taxes and cap-and-trade mechanisms are two instruments that are, in principle, well suited to price carbon. The discussion in this section was guided by the experience from cap-and-trade schemes. They have been successfully implemented across countries and US states for CO2, SO2 or NOX. They offer the opportunity to integrate the environmental regulation across several jurisdictions and, for example, impose a common carbon price across EU Member States. The European experience points to the importance of clearly separating the cap setting from the allocation to individual emitters. This simpliﬁes the process of choosing an appropriate cap that reﬂects emissions-reduction objectives and economy-wide mitigation opportunities. Initial free allocation of some allowances was implemented to gain industry support and compensate for value loss of high-carbon assets in a low-carbon world, but increasingly the focus is on using free allowances to support domestic consumers with direct transfers or in their pursuit of energy-efﬁciency measures (for calculations in the US context, see Burtraw, Palmer and Kahn 2005). If free allocation to emitters turns into a repeated feature, then the expectations of market participants about future free allocation will distort the carbon-price signal. This results in inefﬁcient operation and investment choices. In an existing scheme it is important to phase out free allowance allocation and move to auctioning of allowances, owing to continued distortions of free allocation. With careful design, cap and trade can

94

*

Cap and trade

thus be a viable means to deliver a carbon-price signal. An emissionstrading scheme where all allowances are auctioned creates revenue streams similar to a carbon tax. However, the political economy of the initial implementation of both instruments and the coordination of the price level across jurisdictions remain important differences. Also, the determination of the carbon price differs between both approaches. The carbon tax is set in a political process and can be revisited annually. The carbon price in an emissions-trading scheme follows from a political decision on the cap; linkages with other schemes and the volume of off-sets allowed for in a trading scheme that is often ﬁxed for many years at a time (see Chapter 5). This has distinct implications for price-formation processes for investment decisions (see Chapter 4). In the ﬁnal section the question was raised whether carbonpricing schemes should be applied simultaneously to all sectors of an economy. The ﬁrst instinct of economists and traders is to argue for an integrated scheme with one carbon price that allows the market to select the least-cost-mitigation options. However, other considerations are relevant. Across sectors the levels of demand response and innovation triggered by carbon prices differ. The equity implications of costs faced by different consumer segments also vary. In addition, pre-existing tax schemes, such as fuel taxes in the transport sector, are often at a far higher level than current carbon prices, which would not materially affect demand at the low level of current carbon prices. The smaller size of some emitters would create signiﬁcant transaction costs, and would probably require an upstream approach that makes distributors or importers of fossil fuels responsible for the emissions. The diverse set of factors to be considered suggests that the timing for the inclusion of different sectors might depend on the speciﬁc circumstances of a country. This might explain the differences in coverage of emissions-trading schemes discussed and implemented in different regions of the world.

Conclusion

*

95

Notes 1 www.chicagoclimatex.com/index.jsf. 2 In 1990, the penalty was $2,000 and escalates at the rate of inﬂation (Solomon 1995). 3 About 10 million tonnes of emissions are covered by the trading scheme, with prices moving between $270 and $850/short ton. Thus, the value of allowances, largely distributed at no cost, is in the order of $2.7–$8.5 billion. In contrast, the EU ETS covers about 2 billion tonnes of CO2 emissions, which at a price of €20/ tonnes CO2 is valued at €40 billion. 4 Japanese policy for industrial sectors seems to constitute the only exception. While based on agreements that are called ‘voluntary’, the efforts of industry are monitored, reported and discussed on an annual basis. In sectors that failed to meet their emissions targets, as was the case for steel and power in 2008, the sectoral targets were broken down to a company level. Companies were required to buy international off-sets (CDM credits) to cover their excess emissions. 5 In March 1996, the Federation of German Industry had signed sector targets with the German Environment and Industry Ministry for emissions reduction by 2005 relative to 1990 levels. The UNFCCC Report on the In-depth Review of the Second National Communication of Germany, dated 24 August 1999, noted ‘that in nearly all cases the divergence between the commitment and BAU [business as usual] forecast is small’ while ‘a similar pattern emerges for all of the agreements studied with a signiﬁcant part of their targets having already been achieved by 1996’ (UNFCCC 1999). 6 The Joint Implementation Supervisory Committee (JISC), under the authority and guidance of the Conference/Meeting of the Parties (COP/MOP), inter alia, supervises the veriﬁcation procedure deﬁned in paras 30–45 of the JI guidelines http://ji. unfccc.int/Sup_Committee/index.html. To avoid double accounting, the project host country has to reduce its cap by the amount of JI credits that have been granted. The beneﬁt for host countries is not emissions reduction, but co-beneﬁts like technology transfer, improvement of energy infrastructure and investment ﬂows. 7 The existing rules create a ‘prisoner’s dilemma where each individual Member State recognizes the collective interest to set restrictive caps for optimal reduction of emissions in the EU, but also has an interest to maximize the national cap’. European Commission 2008a. 8 Where Member States envisaged investing in CDM projects or buying assigned amount units from countries like Russia, the Commission required evidence that these expenditures were reﬂected in national budgets. 9 See Colombier and Neuhoff 2008 for a discussion of the distortions induced by benchmarks and Walker and Richardson 2006 for a discussion of options to reduce carbon intensity in the example of cement.

96

*

Cap and trade

10 To address this concern in the power sector, some national allocation plans contain transfer provisions. The free allowances that an existing plant A would have received can be transferred to a new plant B. Thus, the disincentive to close plant A has been reduced from the high level of free-allowance allocation of plant A to the lower level of forgone free-allowance allocation of the moreefﬁcient new entrant B. With the use of benchmarks for the allocation to plants, such transfer provisions lost their relevance. 11 A discussion of experiences from auctions across different sectors can be found, for example, in Klemperer 2002 or Jensen 2004. 12 See report on workshop in January 2008 with policy makers, academics and industry stakeholders: ‘Auctions for CO2 Allowances: A Straw Man Proposal.’ www.eprg.group.cam.ac.uk. 13 Presentation by Robert Seehawer of EEX, 24 June 2010. www.eex.com. 14 www.rggi.org.

four

Shifting investment to low-carbon choices

Most carbon-emissions reduction is expected from investment and reinvestment choices in power, transport, housing and industry. Hence, one of the main purposes of the carbon-price signal is to drive investment choices. These choices, especially in the infrastructure sector, are typically associated with long time frames over which returns are expected to ﬁnance the initial investment. This does not automatically imply, as frequently argued, that infrastructure and technology choice cannot evolve rapidly. Figure 4.1 illustrates the rapid investment in combined-cycle gas turbines for power generation which occurred after the liberalisation of the UK gas and electricity markets in the early 1990s. Within half a decade, the new technology captured the biggest market share of UK power generation. This illustrates that market environments can drive rapid change if the appropriate framework creates sufﬁcient certainty and incentives for investors. The European Union emissions-trading scheme has put carbon prices on the agenda with executives of emitting companies. The cost of carbon is relevant to the investment decisions of 73 per cent of energy-intensive industries in Europe.1 The Stern Review (2006) on the economics of climate change highlights some concerns about the impact of uncertainty on low-carbon investment choices: 97

98

*

Low-carbon choices

Electricity production (GW h)

200

Gas

150

Coal 100 Nuclear 50

0 1960

Oil 1980

2000

Figure 4.1 Generation share of different technologies in the UK: 1960–1997 (Fuel consumption for power generation, transformed to output using 1998 average efﬁciencies)2

In order to inﬂuence behaviour and investment decisions, investors and consumers must believe that the carbon price will be maintained into the future. [. . .] If there is a lack of conﬁdence that climate change policies will persist, then businesses may not factor a carbon price into their decision-making. But establishing credibility takes time. [. . .] In this transitional period, while the credibility of policy is still being established and the international framework is taking shape, it is critical that governments consider how to avoid the risks of locking into a high-carbon infrastructure, including considering whether any additional measures may be justiﬁed to reduce the risks.

4.1 The nature of uncertainty Dealing with market uncertainty is nothing new for investors in energyrelated markets. The impact of the additional uncertainty relating to carbon prices and allocation methodologies may at ﬁrst appear to be limited. There are, however, two aspects that exacerbate uncertainty. First, cap-and-trade and other climate policies are subject to regulatory uncertainty regarding the level of stringency and the rules

The nature of uncertainty

*

99

under which carbon markets and other policy instruments are implemented. By its very nature, regulatory uncertainty is driven by soft factors relating to future decisions of policy makers. These are difﬁcult to quantify, and therefore it is difﬁcult to attribute probabilities to different future scenarios for inclusion in analytic models used for investment analysis. Second, where a trading framework has been clearly established – for example, within the EU ETS after the national allocation plans have been decided upon – price formation is subject to market forces, reﬂecting uncertainties about technologies, demand and availability of input factors. While price uncertainty is typical of many markets, carbon-price uncertainty has some special features. There are no natural lower bounds for carbon prices or expectations of reversion to the mean in the long term. This is in contrast to most commodities, where marginal production costs set natural price ﬂoors. Also, the lack of a long price history implies that, in the initial years of a cap-and-trade scheme, it is difﬁcult to extrapolate future prices based on past experience. This practice is used in other markets to inform management and ﬁnancing decisions. The inherent uncertainty of future prices, not only of carbon, but also of fuels, commodities and technologies, leads many companies to pursue longer-term strategic decisions based on quantities. They explore how big will the market be and what share of this market can a form capture? The emphasis that actors attribute to carbon prices, to other regulatory instruments and to longer-term quantitative frameworks (e.g., emissions targets) varies across sectors and countries. Section 4.2 explores the complementarities of prices and quantities from an economic perspective. It may not be possible to establish generalised solutions, as section 4.3 illustrates how investment and ﬁnancing approaches can differ between actors. Section 4.4 discusses how the needs of investors with long-term perspectives can be addressed, while section 4.5 looks at policy instruments to support investment decisions by players with

100

*

Low-carbon choices

shorter-term perspectives. Finally, needs of ﬁnancial investors are explored in section 4.6, and section 4.7 concludes.

4.2 Response to uncertainty with taxes and cap-and-trade schemes Both climate impacts and mitigation opportunities are uncertain. This has implications for the choice of policy instruments (Newell and Pizer 2003). Focusing only on the direct economic impacts, and ignoring aspects of political economy, while both taxes and cap-andtrade schemes deliver the same outcome in the absence of uncertainty, their performance differs signiﬁcantly given uncertainty. The economic debate about cap and trade versus taxation dates back to a paper by Martin Weitzman (1974). He analysed policy instruments if costs and beneﬁts are uncertain. In this case a tax will keep the carbon price constant, and cap and trade keeps the emissions volume constant while the other parameter adjusts. Weitzman’s analysis implies that if the cost of reducing an additional unit of carbon emission is independent of the emissions level, and damage caused by emissions grows drastically with the emissions level, then cap and trade is the preferred policy instrument under uncertainty. In contrast, if the cost of reducing carbon emissions increases drastically with the mitigation effort and the damage caused by a unit of carbon emissions is independent of the emissions level, then taxation is the preferred policy instrument under uncertainty. Which of these scenarios describes the situation of climate change better? Looking at time frames might explain the different perceptions, as is illustrated in Box 4.1. In the short term, taxation is the preferred policy instrument, while in the long term, cap and trade is more suitable. These two policy choices are, however, inconsistent. A long-term quantity announcement will only be credible if it is enforced in the future. This suggests that policy makers must use a

Taxes and cap-and-trade schemes

*

101

b o x 4 . 1 Tax versus cap and trade Figure 4.2 uses the Weitzman framework to assess whether policy makers should set prices (taxes) or quantities (emissions trading). It shows that costs of additional mitigation efforts increase with higher levels of mitigation. At the same time, the damage from an extra ton of CO2 falls if mitigation results in lower concentrations of CO2. In a world without uncertainty, the carbon tax level can be set at the intercept of mitigation costs and damage – and the optimal level of mitigation is implemented. Alternatively, the cap of an emissionstrading scheme could be set to match the optimal level of emissions reduction. In this very simple world without uncertainty both schemes deliver the same result. With uncertainty the mitigation costs could be higher than anticipated (dashed line in Figure 4.2). The new optimal price and quantity of mitigation is given by the intercept between the dashed mitigationcost curve and the damage-cost curve. But this optimal point cannot be reached if governments have to choose and implement a policy instrument before all information is available. If the government uses emissions trading and sets a cap, then unexpectedly high emissions costs result in greater mitigation efforts than would be optimal. The dark grey area in the graph illustrates the

ost

ge c

a Dam

Cap

Risks insufficient mitigation with tax Risks excess mitigation with cap st

/t CO2

tion co

Mitiga

Tax

Emission reduction

Figure 4.2 Marginal damage and mitigation costs, and the impact of uncertainty (Weitzman framework)

102

*

Low-carbon choices

b o x 4 . 1 continued

/t CO2

Mit

igat

Cap

ion

cos t

emissions-reduction measures pursued which are more expensive than the damage avoided. If, alternatively, the government decides to implement carbon taxes, then less emissions reduction will be implemented than optimal. The light-grey area in the graph illustrates the damage caused by these extra emissions. Comparing both policy instruments we ﬁnd that in Figure 4.2 the welfare losses from underestimating the mitigation costs are higher if taxes are used than if emissions trading is used. The result can be generalised. If the mitigation-cost curve is ﬂat relative to the damage-cost curve, then emissions trading is the preferred policy instrument under uncertainty. Figure 4.3 demonstrates that if the mitigation-cost curve is steeper than the damage-cost curve, then the ordering of the policy instruments is inverted. One would expect that empirical evidence could be used to answer the question of whether mitigation- or damage-cost curves are steeper – but good data, including on elasticities, are scarce, and opinions still differ. Regarding the damage-cost curve, if temperatures roughly increase logarithmically with the level of atmospheric CO2 concentrations, and damages increase exponentially with temperatures, then damage is a linear function of CO2 concentration and the marginal damage-cost

Dama

Risks insufficient mitigation with tax Risks excess mitigation with cap

ge co

st

Tax

Emission reduction

Figure 4.3 Mitigation-cost curve steeper than damage-cost curve

Taxes and cap-and-trade schemes

*

103

b o x 4 . 1 continued curve is ﬂat (similar to Figure 4.3). This would argue for the use of carbon taxes set at the level of marginal damage (Hope and Newbery 2008). The question is how to calculate this damage (Tol 2003). Economists still discuss the appropriate discounting and risk factors (Guo et al. 2006), scientists discuss the impact on different countries and few people dare to put cost ﬁgures to impacts like large-scale droughts inducing starvation and migration. The conﬁdence in predictions and quantiﬁcation of climate-change impacts outside of the ranges experienced by our societies is still limited. National and international policy discussions therefore have deﬁned a temperature increase that our societies do not want to exceed, as the impact would be perceived to be unmanageable (Stern Review 2006). Nationally, and internationally, the objective has become to limit global temperature increases to 2° C. Climate models can in turn translate this political objective to CO2 emissions – they must be 50–85 per cent below 2000 levels to limit global average temperature increases to 2.0–2.4 °C (IPCC 2007). This approach has an implicit assumption that damages increase disproportionally once the threshold is crossed, reﬂecting the risk of large-scale disasters. The steep damagecost curve suggests a need for the use of targets and caps. Perspectives not only differ with regard to the damage-cost curves but also about the mitigation-cost curve. Many mitigation opportunities require new investment or refurbishment of existing buildings and infrastructure. This takes time and implies that in the very short term the mitigation potential is limited, and thus the mitigation-cost curve is steep. As the available time frame increases, the set of possible mitigation options expands and the mitigation-cost curve becomes ﬂatter. Some studies (see, e.g., Figure 4.4) predict for the long term (e.g., 2030), a ﬂat mitigation-cost curve across various measures that are available once the initial cheap options are realised.

policy instrument today if they want investors to believe that they will also use it in the future. If policy makers choose a quantity-based cap-and-trade approach, this can facilitate long-term investment decisions. The future carbon

104

*

Low-carbon choices

price will respond to new information about costs of technologies and fuels, and ensure appropriate remuneration of low-carbon strategies and investments. If policy makers choose a price-based approach, like taxation, then the stable price signal can support shorter-term investment choices. However, it is not yet clear what carbon-price level will be required to make certain technology options economically viable. Therefore, the use of carbon taxes means inherent uncertainties about long-term mitigation costs translate into uncertainties about future emissions levels, which creates another credibility issue. Investors will only have conﬁdence in a policy instrument that delivers against policy objectives. Otherwise, investors suspect policy makers might change the instrument, and they will wait to commit their resources until the changes have been implemented. Hoel and Karp (2001) and Newell and Pizer (2003) were among the ﬁrst in a long list of authors to model the relative merits of carbon taxes and emissions caps. They conclude that, with uncertainty, taxes are preferable to quotas. This is because with unexpected economic and emissions growth, the carbon price would rapidly increase, imposing high costs on their model economy. However, they assume technologies are exogenously given rather than that they emerge with private-sector investment and thus limit the ability of an economy to be ﬂexible in adjusting its carbon intensity. Considering the potential for technology innovation increases the ability of economies to respond to carbon constraints, and also shifts the trade-off between price- and quantity-based instruments towards quantity-based emission-trading schemes (Weber and Neuhoff 2010). To discuss these policy choices in more detail, we need to move from the perspective of ‘general’ investment decisions to a better understanding of the impacts on individual investment choices in different sectors.

Investment under uncertainty

*

105

4.3 Investment under uncertainty: contrasting different perspectives Companies across sectors differ in their response to the uncertainty for historic, institutional, organisational and technological reasons. Based on interviews with managers and analysts, this section offers ‘strawman’ representations of the main drivers for strategic decisions and project choices. The simpliﬁcation inherent in such representations implies that all the differences that exist within groups might be larger than the differences between groups. However, grouping allows for an easy identiﬁcation of perspectives, and hence the focus is on oil majors, technology companies, utilities, banks and project developers. Oil majors undertake investments with long horizons against internally developed scenarios of the global market and geopolitical evolution. The stringency of current climate policies is an important political signal, as this is an indicator of the credibility of future targets. Oil majors infer from long-term targets and perspectives the role of different technologies or the implied long-term carbon price. In the initial years, carbon prices are unlikely to have a strong impact on oil demand, owing to inelastic demand and already high taxes. Transport-sector-speciﬁc policies are likely to be more relevant. But, for the longer term, targets are the basis for estimates of future market opportunities and strategic decisions. Current spot and forward prices for oil and carbon are less important for the long-term investment decisions of oil majors, and more relevant for risk and uncertainty analysis to determine and manage shorter-term exposure to upside and downside risks. Investment projects are then benchmarked against projects from other business units. Projects are also assessed against possible projects undertaken by competing companies, in order to identify competitive advantages and anticipate proﬁtability. Technology developers and manufacturers are always eager to move their new technologies forward. Yet, to obtain funding they

106

*

Low-carbon choices

ual

s us

ss a sine

2,000

Early role of energy efficiency

1,500

Long-term role for renewables

1,000

Energy efficiency

Renewables

500

Nuclear Gas Oil Solid fuels

Phase out conventional coal 2040

2030

2020

1995 2000 2005 2010

1990

0

2050

Mio t oil equivalent

Bu

Figure 4.4 Roles of different technologies over the next four decades using long-term emissions targets3

must demonstrate credible scenarios for the role of their technology to third parties. They cannot use the approach of oil majors to deduce the future market share of their technology from future emissions targets because (i) internally developed scenarios illustrating the role of a certain technology are not credible enough to convince third parties to provide ﬁnancing and (ii) many pathways lead to long-term emissions targets. The time when individual technologies start to make signiﬁcant contributions to the energy mix will vary signiﬁcantly between the various pathways. This complicates planning and creates additional risk for technology investors, who typically do not have the ﬁnancial endurance to wait for markets to evolve. The results from the real-options simulations show that adoption of new technologies depends signiﬁcantly on investors’ view of uncertainty (Reedman, Graham and Coombes 2006). Figure 4.5 illustrates that explicit renewables targets, in this case deﬁned in the National Renewable Energy Action Plan 2010, can outline the technology

Investment under uncertainty

Long-term position of technology dependent on performance

300 Renewable power production (TW h Germany)

*

107

Geothermal Wind offshore

250 200

Prospect for technology demand

Wind onshore

150 Photovoltaic

100

Biomass

50 0 2005

Hydro 2010

2015

2020

2025

Figure 4.5 Renewables targets and the role of different renewable technologies

shares in the coming years and thus provide some reassurance to investors that government policies will be in place to address technical and administrative barriers for the deployment of renewables. These can contribute to conﬁdence that there will be a market for successful technologies in the time frame required by investors. Utility companies have long experience of how regulatory and policy choices determine investment outcomes. The differing market shares of nuclear energy across countries illustrate that such policy preferences are difﬁcult to explain using simple economic reasoning. Utilities are therefore mainly guided by current policy frameworks, such as the EU ETS, when assessing investment choices (Reinaud 2003). Current prices, forward prices and existing policies are dominant drivers for investment choices, and only credible commitments to changes of these policies will affect decisions.4 In the absence of any such strong guidance, some utility companies might continue with traditional investment approaches, focusing mainly on diversiﬁcation between coal and gas.5 Only the implementation of full auctioning of allowances to the power sector leads utilities to abandon coal projects. Banks provide debt to ﬁnance investments across different sectors. They must implement internal control mechanisms to ensure

108

*

Low-carbon choices

individual business units do not take excessive risks and are reluctant to engage in speculation about future evolution of markets and policies. Thus, policies have to be simple, transparent and credible. Banks would much rather use data on historic performance of technologies and sectors to assess investment risks (see Figure 4.6). Table 4.1 summarises the inputs that drive investment and strategy choices in different sectors. Oil majors and technology companies have a strong focus on the possible future role of their fuel/technology and their company in the relevant market. Given the large share of oil in global energy supply, oil majors can translate emissions targets into the impact for oil. By contrast, the initially small share of renewable technologies in global energy supply means that such calculations are more speculative for investors in new energy technologies, and additional guidance on the role envisaged for these technologies is frequently sought. Large utility companies have traditionally based their decisions on the national regulatory environment, best reﬂected in current and

Illustrative

Price CO2 allowances

Downside risk for high-carbon assets main obstacle for coal investments Short and not representative history

Difficult-to-quantify downside risk for low-carbon investments

Today Everyone looks at price history

2020

2050

Banks and utilities look at forward prices

Figure 4.6 Impact of carbon price projections for agents involved in investment decisions

Strategic investors

*

109

t a b l e 4 . 1 Main determinants of investment choices across sectors

Oil majors

Technology companies

Historic prices

Current price

Determines operation

Short-term production

Future prices

Difﬁcult to predict

Difﬁcult to predict

Future market share of fuel/ technology

Main driver for strategic choices

Main driver for strategic choices

Large utility companies Replication of successful strategies Perceived as best guide for future Valuable where available Increasingly used in scenarios

Banks, project investment Required to calculate volatility Main input for base case Value longterm contracts

future prices, and, in the case of EU ETS, the level and methodology of free-allowance allocation. With increasing emphasis put by policy makers and the public on the need to de-carbonise our economies, many European utilities have developed a quantitative understanding of the implications for the generation mix. For banks and project investors, carbon and energy costs are often a minor part of the decision process and are best dealt with using standard metrics based on historic volatilities and current and forward prices. The challenge for lowcarbon policy will be to address these different requirements so as to ensure that low-carbon investment is jointly pursued by the various market participants.

4.4 Addressing requirements of strategic investors Any investor wants to have reasonable certainty about future policy evolution. However, in a world with scientiﬁc, technological, economic and political uncertainty, policies that promise certainty for a long-term horizon are not credible. Policy design needs to balance

110

*

Low-carbon choices

regulatory certainty with ﬂexibility to respond to uncertain future impacts, which can be achieved by moving from general long-term objectives towards more speciﬁc mid-term targets and effective shortterm policies. In such a framework, robust policies implemented today provide the credibility for longer-term targets, and these targets provide guidance for strategic investors and technology companies. The general principle of moving from broader long-term objectives towards more tangible mid-term targets and short-term policies is uncontroversial. The deﬁnition of speciﬁc time frames and targets is typically more challenging. When setting long-term emissions targets, policy makers must balance the risks and costs of climate impacts against the feasibility and costs of emissions reduction.6 With increasing robustness of climate modelling, the majority of scientists are warning against temperature increases above 2 °C. The economic modelling of the costs to reduce emissions towards an emissions trajectory compatible with a 2 °C temperature increase is still uncertain, as reﬂected by the variability of results across modelling teams. The Stern Review (2006) reported costs in the range of 1 per cent loss of GDP. However, the stabilisation scenario of 500–550 parts per million CO2 equivalent of the Stern Review creates, according to the best estimate of IPPC 2007, average temperature increases between 2.4 °C and 2.8 °C. To limit temperature increases to 2 °C, CO2 concentrations have to be lower – 450 parts per million CO2 equivalent, or less. Fewer models have assessed the corresponding scenarios, but all of these predict GDP reductions of less than 3.3 per cent by 2030 (IPCC 2007) to achieve the necessary emissions reductions. As emissions reduction is translated into shorter-term targets for countries and sectors, the risk perception is shifted. Any one sector or actor considers the impact of individual or sectoral decisions on global climate change to be small, and is mainly concerned with the risk to the proﬁtability of business. Listening carefully to industry concerns, policy makers will therefore be reluctant to pursue any

Strategic investors

*

111

policy that creates any signiﬁcant risk that a speciﬁc sector might face difﬁculties, and they will prefer to implement a lenient policy, if required, to avoid this risk. In addition, a number of studies (e.g., Harrington Morgenstern and Nelson 2000 and Hammitt 2000) have found a continuous discrepancy between higher ex ante cost estimates and lower-cost options observed in ex post assessments (Pizer and Kopp 2005, argue that the current evidence is not conclusive). To illustrate, Figure 4.7 shows the comparison of ex ante and ex post evaluations of UK environmental air quality regulations. Both businesses and government overestimated costs of implementing measures to comply with the regulation. The comparison with ex post evaluations shows that costs can fall thanks to innovative ideas and optimisation of technology during mass production, based on learning by doing. The discrepancy between the risk aversion of different actors involved in bottom-up policy processes and top-down target-setting comes to light when mid-term targets are set and legislation is created – for example, in cap-and-trade schemes. To set effective and

Ex ante assessment Ex post assessment

6,000

Costs (£m)

5,000 4,000 3,000 2,000 1,000 0 Lead-free petrol

Euro I petrol cars*

2000 fuel standards

2005 fuel in Flue gas de- Low NOx 2000/1 sulphurisation burners

Figure 4.7 Ex ante and ex post costs of UK policies. Source: Based on AEA Technology Environment 2004.

112

*

Low-carbon choices

credible targets, this discrepancy must be resolved, which requires analysis and shared perspectives on how individual sectors could evolve from their current situation towards a low-carbon future. The ﬁrst important aspect of aligning the perspectives is the ability to pool risks across sectors. Any one sector faces many uncertainties as to whether technology improvements will allow for energy-efﬁciency improvements and emissions reduction, and will therefore be reluctant to accept stringent targets for the sector. Cap-and-trade schemes, which cover several sectors, involve many different possibilities for technology and process improvements, and can therefore set more ambitious targets. This suggests an advantage of cap-and-trade schemes over sector-speciﬁc agreements, but this advantage will only be realised if the cap setting is clearly separated from the allowance-allocation process in the scheme, as discussed in Chapter 3. The second aspect that needs to be considered when aligning perspectives is the time frame over which targets are set. The European debate of 2008 has converged on the 2020 horizon, while many proposals discussed in the US Senate outline targets for 2030. A 2020 time frame makes the results more tangible for current investors but leaves open the question about the level of ambition beyond 2020. This has the beneﬁt that it allows for decisions on the speciﬁc trajectory beyond 2020 in response to the ongoing discussion among policy makers and populations as they are gradually acknowledging scientiﬁc evidence on climate change. Shorter time frames of target-settings allow a sequential commitment to more ambitious emissions-reduction targets and policies. Time frames for 2030 are further away from bottom-up policy processes and thus allow for more ambitious targets. To be tangible for investors, the 2030 targets require a clear trajectories or speciﬁc milestones. Discussions about these shorter-term trajectory or milestones then reintroduce the challenge to align bottom-up and topdown approaches. If targets are not aligned with the environmental

Strategic investors

*

113

requirements, the expectation that they might later be tightened creates some uncertainty. Transparent processes, which ensure that targets are only tightened and not relaxed over time, are required to provide a stable framework for low-carbon investment.7 It is easy and popular for politicians and governments to declare long-term targets, but by themselves these targets are not very informative. Political parties and governments must demonstrate their commitment to these targets by investing political capital in shorterterm policies to support the long-term targets. They can do so by implementing low-carbon policies even when they face opposition from inﬂuential political lobby groups. Such government commitment gives private-sector investors conﬁdence that future governments are also likely to pursue stringent policies. This in turn attracts investment in technology development and low-carbon choices (Helm Hepburn and Mash 2003).

The evolving policy framework in Europe Figure 4.8 summarises some components of European climate policy. Member States and the European Commission want to avoid emissions trajectories that result in more than a 2 °C temperature increase. This requires that European emissions be reduced by 80–95 per cent, relative to 1990 levels, by 2050. These long-term objectives have been translated into more tangible targets. In March 2007, the European Council, composed of heads of European states, approved in the European Council the necessary mid-term targets. In January 2008, they were translated by the European Commission into a set of draft directives, the Climate Package, which was then passed by the European Parliament and Council in December 2008. With these directives, EU Member States commit to achieving a reduction of at least 20 per cent in the emissions of greenhouse gases by 2020, compared to 1990 levels, with the objective of a 30 per cent reduction by 2020, subject to international action.

114

*

Low-carbon choices

/year

ual 1%

ss as us

ed Assum

n gr Energy-efficiency improvements

emissio

20% renewable target 2020 (up from 5% share in 2005)

1

2010

30% emission target

3

20% emission target

4

Kyoto target

CO2 emissions in Gt/year (EU 27 countries)

5

2

usine owth b

2020

80–

95%

redu

ctio

2030

n tr

ajec

2040

tory

(lin

ear)

2050

Figure 4.8 Components of a low-carbon perspective

In addition, 20 per cent of ﬁnal energy consumption is to be met from renewable energy sources. For strategic investors, these targets inform investment choices between technologies with different carbon intensities. For the development and initial deployment of technologies such as solar photovoltaic, marine energy technologies or carbon capture and sequestration, the 20 per cent renewable target is more important. Governments subsequently outlined in National Renewable Energy Action Plans the trajectories for different renewable technologies up to 2020, and described speciﬁc actions to facilitate grid access, planning permits and to address ﬁnance. The experience with the pilot phase of the EU ETS suggests that mid-term targets need to be translated into milestones or emissions trajectories. In the absence of either, some governments submitted national allocation plans for the period 2005–2007 that were not on a linear trajectory towards the 2008–2012 emissions-reduction targets laid out by the Kyoto Protocol and the EU burden-sharing agreements. The governments argued that they could delay action providing they implement policies in the future rapidly to reduce emissions.

Project investors

*

115

The EU Directive on Emissions Trading from December 2008 therefore prescribes an annual reduction of 1.7 per cent of the EU ETS cap from 2013 to 2020, also to be applied in the subsequent years. A clearly deﬁned trajectory for emissions reduction and technology development is required to ensure that national governments pursue low-carbon policies in the short term. These policies will reward lowcarbon investments today and create market conﬁdence that new low-carbon investment will beneﬁt under future low-carbon policies. This situation can be compared to credit markets: a borrower must meet his ongoing commitments to reassure banks that he will pay back new loans in the future. Just as credit history is a major criterion for banks to decide on private loans, likewise the history of lowcarbon policies is a major criterion for investors. To summarise, investors with long-term perspectives on the role of technologies and market shares require mid-term emissions-reduction targets. These targets must be credible. Credibility ﬂows from a consistent set of long-term objectives, mid-term targets for emissions and low-carbon technology development and shorter-term milestones or trajectories. However, the basis for all credibility remains the political commitment. Private-sector investors will carefully measure this commitment by observing policy makers – do they invest political capital and implement climate policies even where they face opposition from important stakeholders?

4.5 Addressing requirements of project investors If carbon is not the core component of a project, investors deciding on projects and banks providing ﬁnancial assistance rarely engage in long-term policy analysis of carbon policies. Instead, they use historic and current ﬁnancial indicators, and take current market design and regulation as the best guess for what might be in place in the future. This situation constitutes a challenge for de-carbonisation policy with few historic carbon prices, volatile current prices and an

116

*

Low-carbon choices

evolving future policy framework. Of particular concern for lowcarbon investments is the risk of low or zero carbon prices. As it is difﬁcult to predict the probability of such events, investors and banks err on the safe side and assume higher probabilities. As a result, an uncertain carbon price might be highly discounted in investment and ﬁnancing decisions. In the following sections, we will discuss different options to increase conﬁdence in carbon prices until the price is ﬁrmly established, so as to facilitate low-carbon investment.

Length of implementation periods Increasing the length of periods over which cap-and-trade schemes will be in force is frequently discussed as one approach to increasing price stability. An argument in favour of long-term periods is that when the trading periods are too short the scarcity of allowances in the system, and thus the allowance price, will be too sensitive to climatic conditions, such as a cold winter, or short-term economic cycles, with a temporary increase of economic activity and consequent emissions. The ﬁnancial crisis triggered a large reduction of industrial output. The EU industrial production index fell in 2008 by 5 per cent and in 2009 by 18 per cent relative to 2007 (based on Eurostat data). As a result, emissions from industry and, to a smaller extent, power, declined. EU ETS emissions in 2008 were 3 per cent, and in 2009 14 per cent, below median projections in Figure 4.9. As EU ETS allows for banking and is in place beyond 2020, these impacts can be averaged out over many years and thus have less of an impact on the allowance price. It stayed near €15/tonne CO2. Longer trading periods will also mean there is more time for investors to respond to high allowance prices with investments that reduce emissions and, eventually, prices. Figure 4.9 illustrates the results of several power and non-power sector models that were combined to project EU-25 emissions for the

Project investors

*

117

2,500 88

Proposed national allocation plan II Final national allocation plan II

125

Mt. CO2/year

2,000 Other adjustments Adjustments for opt-in in phase II

1,500

national allocation plan II + (joint imlementation/clean development mechanism range)

1,000

Maximum projection

500

60% projections

Verified emissions

20% projections Minimum projection

0 2005

2008

2009

2010

2011

2012

Figure 4.9 EU-25 emissions projections for 2008–2012. Source: Neuhoff et al. 2006.

period 2008–2012, assuming €20/tonne CO2 prices.8 The results were calibrated to the veriﬁed emissions from the emitters in 2005. The uncertainties inherent in the projection reﬂect different assumptions about economic growth, fuel prices and energy-efﬁciency improvements. Also, as the length of implementation periods is increased, so does the likelihood that constraints on emissions will have to be tightened again during the trading period. Any political discussions on possible tightening will be reﬂected in the prices, and thus increase volatility. Not only the ﬁnal cap, but also the methodology of allowance allocation inﬂuences investment decisions.9 In some cases, it might be easier to make a long-term commitment to the methodology of allowance allocation (e.g., full auctioning) even for a time period for which no explicit emissions cap can be determined. Thus, a stable framework outlining future allocation decisions can facilitate investment decisions for time horizons for which ﬁnal caps are not yet determined (Åhman et al. 2007).

Banking and borrowing Allowing for banking is often suggested as a means for increasing price stability. Below, we discuss lessons from the US experience, the ﬁrst EU trading periods and implications for future policy design.

118

*

Low-carbon choices

The US SO2 cap-and-trade schemes also allocated far more allowances for the ﬁrst years than were required to cover emissions, but the allowances were deﬁned as bankable: companies could keep – or bank – unused allowances for future use, which many did because the cap was expected to be tighter. As a result, the allowance price did not drop to zero despite excess supply (Ellerman et al. 2000). The extent to which policy design should expand the role of banking, increasing the allowance supply so as to create some excess allowances that will be banked, is inﬂuenced by various factors.10 In the US SO2 cap-and-trade scheme it was possible for policy makers to offer a long-term horizon. The need for emissions reduction was well established, and the technology to tackle SO2 was widely available and used. Thus, banking between periods occurred in a stable regulatory environment. Although 31 per cent of allowances were banked in phase I of the US Acid Rain Program, the overall value of these allowances was still limited to a few billion dollars. That situation differed markedly from that of the pilot phase of the EU ETS, which did not allow for banking of allowances into the second period. In 2006, the excess supply of allowances had already become apparent, and by 2007 the price of allowances dropped to virtually zero. Banking had been rejected so as to prevent the spillover of possible difﬁculties in the pilot phase into phase II. This ensured that the EU ETS would make its contribution towards the Kyoto emissions targets, which are measured over the period 2008– 2012. Indeed, despite the crash of the price of allowances for the pilot phase, the forward price of allowances for phase II has remained robust. For subsequent phases, the EU Directive allows for banking. It is still being debated to what extent the design of the cap should rely on banking to create scarcity. It is undisputed that banking of CO2 allowances between trading periods can avoid price drops at the end of a trading period. But, should a cap-and-trade scheme be designed in a way that anticipates that a signiﬁcant share of

Project investors

*

119

allowances will be banked to the next trading period and therefore rely on banking as the primary mechanism to determine the allowance price? It is sometimes argued that this type of banking between periods can increase price stability (Newell, Pizer and Zhang 2005). This approach also offers the beneﬁt of avoiding price spikes if market participants struggle to reduce emissions to the overall cap level. The approach does, however, create two challenges. First, if there is an explicit recognition that initial caps are loose and higher than the desired level, they do not help market participants and policy makers to coordinate activities that are required for a transition to more carbon-efﬁcient buildings, transport infrastructure or supply chains. The beneﬁts associated with emissions-trading schemes – to deliver a carbon price that supports the shared vision of an emissions-reduction trajectory – are reduced. Second, if signiﬁcant shares of allowances in carbon-trading schemes were banked, their value could end up in the region of hundreds of billions of dollars by the year 2020. This is a multiple of the value of the SO2 allowances that were banked by utilities. Careful analysis is required to determine whether private-sector entities would be prepared to hold allowances of such a value, without charging a corresponding risk premium. A large risk premium on banking CO2 allowances would depress the current carbon price and delay low-carbon investment decisions. It is sometimes suggested that emitters should be allowed to borrow allowances from future periods. Emitters who cannot cover their emissions with current allowances would be allowed to surrender the necessary allowances in future periods. However, because of the various risks associated with such borrowing, it has so far not been permitted. First, emitters might default or gradually increase their allowance debt, and would demand collateral or guarantees to avoid gaming opportunities. Second, it is difﬁcult to anticipate how many allowances will be borrowed, therefore allowance scarcity and prices are more difﬁcult to anticipate. Finally, governments might not be

120

*

Low-carbon choices

able to make up for the excess emissions of emitters who ‘borrowed’ allowances from future periods by increased emissions reduction in sectors not covered by the cap-and-trade scheme, and would therefore fail to deliver against domestic commitments or their internationally agreed emissions-reduction target. If governments themselves could ‘borrow’ from future periods, they might set more-lenient targets for future periods to make up for their excess emissions.

Active government intervention Investment decisions and access to ﬁnance for low-carbon projects can be signiﬁcantly affected if investors are concerned that carbon prices might fall to low levels. Policies addressing this risk can thus be part of a strategy to support low-carbon investment. Their design can build on experience of government intervention to provide export credit guarantees or strategic oil reserves. Commodity-price stabilisation has been popular and continues to be applied, for example, in agriculture to reduce risks for small farmers (Newbery and Stiglitz 1981). However, it has frequently failed because (i) storage costs are high and (ii) coordination among multiple countries is difﬁcult. Some authors have explored the potential role for an independent carbon committee or bank to guard the stability of carbon prices (Helm, Hepburn and Mash 2003). One idea is to equip such a committee with tools and powers similar to those available to independent central banks for their management of currencies – for example, to avoid inﬂation. Politicians have an incentive to print money so as to increase public spending and increase economic activity. However, the beneﬁt of printing money is of short duration because it is quickly followed by increasing inﬂationary pressure. Independent central banks can delay government intervention by a few months and thus eliminate the short-term beneﬁts of printing money. However, this example does not lend itself to climate policy, because the time frames of carbon markets are different. Think of

Project investors

*

121

politicians who want to respond to the interests of incumbent companies in carbon-intensive sectors and reduce the carbon price. From their perspective, it might be worthwhile to spend some months convincing the carbon bank to issue extra allowances, because the allowance price would fall in expectation of this outcome at the beginning of the effort. The negative impact on future low-carbon investment would only be felt in years, not months. Hence, the institutional independence of such ‘banks’ would be unlikely to offer sufﬁcient protection over the relevant time frames to guarantee a robust carbon price. An institution that can make discretionary interventions in the carbon market is unlikely to be credible enough to create and hold investor conﬁdence in a robust carbon price.

Setting a reserve price in auctions Governments can announce a reserve price for allowance auctions. Assuming that a sufﬁcient fraction of allowances is to be auctioned, the reserve price will translate into a price ﬂoor for carbon allowances, because some of the allowances from the auctions will be required to satisfy demand. Consequently, all trades will be at or above the reserve price level. A price ﬂoor implemented via a reserve price can create conﬁdence in market participants for the duration of the trading period, and help to control price volatility. Market participants might also interpret a reserve price in one trading period as the most likely reserve price in subsequent trading periods. Implementation of such a reserve price requires co-operation between countries that have a joint cap-and-trade scheme (e.g., EU Member States). If only a fraction of the allowances envisaged for sale are auctioned because demand is limited at the reserve price, then, in principle, every country would like to be the ﬁrst to sell its allowances. Individual countries might try to pre-empt one another in selling their allowances before the other countries (Hepburn et al. 2006).

122

*

Low-carbon choices

This issue has been addressed for the EU ETS auctions post 2012. All countries will use one common commercial agent to auction allowances on their behalf according to a speciﬁed protocol. If the reserve price was binding, then the agent would return unsold allowances to the respective governments proportional to the total volume of allowances each country intended to auction. Excessive inﬂows of CDM or JI credits must be avoided if there is a risk that these could be available at sufﬁciently large volumes at prices below the reserve price. Policies like the supplementarity condition, requiring that part of the emissions reduction must be delivered domestically and limit inﬂows of allowances, can address this issue.

Put option contracts Governments can issue put options on future allowances and distribute them widely among market participants. Option contracts are commercial agreements between governments and private buyers, and are as such well protected by property rights. Thus, they can allow governments credibly to commit to a minimum level of stringency of future carbon policy and facilitate investment in low-carbon technologies. Governments would sell the options to market participants, who could subsequently trade the options. At the time of the expiry date, if the allowance price was below the strike price, the owner of an option could return it alongside a carbon allowance to the government (physical clearing). In return, the owner would receive the deﬁned strike price. If the allowance price was above the strike price of the put option, the put option would have no value. Box 4.2 illustrates how the mechanism works. An alternative approach for governments would be to issue a contract for difference (CfD) with investors on the future carbon price. The holder of such a contract would be entitled to receive the strike price stated in the contract less the actual price implicit in any carbon instrument that applies to fossil generation. If the carbon price exceeds

Project investors

*

123

b o x 4 . 2 Government-issued put options to guarantee a price ﬂoor Figure 4.10 illustrates the government-issued put options scheme with a simple example. The government promises to maintain the price of carbon emissions at or above €15. In order to do that, the government issues one hundred million put options on European Union Allowances (EUAs) with a strike price of €15 and duration of ﬁve years. If the price of EUAs falls to €10 at the end of the ﬁve-year period, and assuming inﬁnitely elastic supply of allowances at that price, option holders would buy an EUA on the market for €10 and sell it, using the put option, for a price of €15 to government. The government would then have to spend a maximum of €1.5 billion (100 million × €15). This would, however, be counted against the initial sales revenue from allowances and option sales. Anticipating the need to reimburse option holders, the government will ex ante sell fewer allowances. In addition, there will be a second effect: holders of the put option will, through their purchase of the allowances, drive up the carbon price, thus automatically stabilising prices. If the amount of option contracts issued is sufﬁciently large, the scheme effectively creates a ﬂoor of €15 for the allowance price. The policy objective is satisﬁed without triggering a ﬁnancial penalty for the government – creating a ‘win-win’ situation. This shows that put options provide investment security from three perspectives (Ismer and Neuhoff 2006). First, investors in projects that are at risk from carbon-price uncertainty can limit their downside risk.

Option price/value

Strike price

Option price CO2 price

Payment if CO2 price is low at expiry date

Figure 4.10 Put options on the price of carbon

124

*

Low-carbon choices

b o x 4 . 2 continued Second, if the authority has issued many put options, then market participants will buy allowances from the market and return these with the option to government should the allowance price fall below the strike price. This will, in turn, reduce the volume of allowances in the market, increase scarcity and push the allowance price up. With sufﬁcient put options in the market, the allowance price will not fall below the strike price of the option. Third, an authority that issued too many put options is aware of the ﬁnancial liability it would incur, and so it will pursue prudent carbon policies and adjust its issuance to avoid the large ﬁnancial liability that would be triggered if the carbon price fell below the strike price of the options. The maximum amount of the authority’s liability equals the strike price multiplied by the number of put options handed in; the proﬁt made by the third party per put option would then be the strike price minus the allowance price. In particular, the put options allow investors in abatement technology and in renewable energy to hedge against the risk that lower carbon prices reduce production costs of high-carbon products and services.

the strike price, the holder of the contract would be required to pay the difference. Thus, CfDs can create signiﬁcant ﬁnancial exposure and counterparty risk if the holder of the contract does not beneﬁt from the high carbon price – for example, if its production capacity is down. The main motivation for governments to engage with ﬁnancial instruments that allow for hedging of carbon-price volatility is to ensure that political risk is borne by the government not the investors (Grubb and Newbery 2007).

Price cap Unexpected events can create emissions levels that exceed the national target. Cap-and-trade schemes respond to such events by price increases that trigger additional mitigation efforts in the

Financial investors

*

125

remaining years of a trading period and by reducing the volume of allowances banked into the next trading period. It is sometimes debated whether additional ﬂexibility is required to respond to unexpected emissions increases that increase the price of allowances. Pizer (2002) argues for hybrid schemes combining a trading approach with a price cap. To implement price caps, governments could, for example, make additional allowances directly available to the market once the allowance price exceeded a price cap (Jacoby and Ellerman 2004). But if such a price cap were set too low, then this could undermine the credibility of future emissions targets. The price cap might prevent future price increases that are necessary to reward investors in low-carbon technologies and projects and thus reduce incentives for investment in innovation. Thus, low price caps might result in higher emissions and, ironically, higher average allowance prices in the long term. It is most difﬁcult to determine the appropriate emissions cap during the introduction of a cap-and-trade scheme, and yet this is possibly when a price cap is most needed. Again, to repeat the previous point, legislation would have clearly to specify the expiry of the price cap after two or three years, so as to ensure that the allowance price can create the necessary incentives for low-carbon investment. The CDM was sometimes envisaged as an alternative opportunity to avoid unacceptably high allowance prices. But, unless the shortfall is small or limited to a small share of the buyers, the time it takes for additional projects to be implemented and to deliver certiﬁed emissions reductions might limit the ability of the CDM to provide ﬂexibility. Allowances banked by market participants might offer more ﬂexibility to respond to short-term emissions variances.

4.6 Addressing the needs of ﬁnancial investors Decisions on project-level investment and on corporate strategy are strongly inﬂuenced by the ﬁnancial sector, which determines the

126

*

Low-carbon choices

availability and cost of ﬁnancing through bonds and new equity: in listed companies, management that is judged ‘underperforming’ by the ﬁnancial sector risks being replaced. Figure 4.11 illustrates three types of ﬁnance sources and their relative importance for different actors. First, early-stage technology investment is inherently risky, and is thus principally funded with venture capital, business angels or as part of on-balance-sheet investment by companies themselves. The main driver for such investment is expectations about future market opportunities. Hence, expectations about policy frameworks implementing emissions reduction up to 2020 and beyond to create demand for lowcarbon products are crucial to facilitating such investment. Public ﬁnancial support can contribute to the incremental costs if appropriation of beneﬁts of technology innovation is limited. Public ﬁnance can also provide opportunities for new entrants where incumbent companies are not exploring the technology space outside of their own expertise. Second, low-carbon strategies involve investment in low-carbon technologies. Their risk declines with their larger-scale application – for example, as part of deployment programmes for renewable-energy technologies. The remaining risk exposure determines the type of

Bonds fixed income

Equity investment

Venture capital, etc. Innovation

Low-carbon strategies Carbon-ignorant strategies Volume of funding accessible 5–8%

Capital costs

10–15%

25%

Figure 4.11 Market categorisation of risk determining ﬁnance structure, access and cost

Financial investors

*

127

ﬁnancing available. Equity investment is required to cover technology, policy or market risk. Loans are typically the cheapest ﬁnance and are available at low interest rates, but they require a high certainty that the money can be paid back. For example, onshore wind supported with guaranteed feed-in tariffs only requires 20–30 per cent equity and can use loans to provide the remaining ﬁnance. Lessmature technologies, less-established companies or less-stable policy frameworks require signiﬁcantly larger equity investments and are thus more expensive to ﬁnance. If low-carbon technologies can be clearly identiﬁed, tailored support schemes can be deﬁned and can provide long-term price guarantees insulating the investor from market and policy risk and, where necessary, increase returns. For other technologies, governments are less likely to be in a suitable position to identify and support individual process improvements, such as in chemical or reﬁning industries. In such instances, a robust carbon-price signal is likely to be more important. Third, companies pursuing ‘carbon-ignorant’ strategies typically act with limited consideration of future carbon constraints. Yet, such carbon-ignorant companies are still considered to be low-risk choices in investment portfolios, as they are using established technologies and existing business models and business relationships. Shifting ﬁnance from carbon-intensive to low-carbon choices will therefore require improved information on risks associated with carbon-ignorant business models, so as both to encourage an improvement of business models by incumbent companies and, where this fails, to ensure that capital is shifted to less-risky and more-sustainable investment opportunities. There is a variety of actors aiming to contribute to this process. They include private-sector initiatives like the Carbon Disclosure Project and ﬁnancial institutions issuing products like a low-carbon index tracker, the Low Carbon 100 Europe Index and the S&P/IFCI Carbon Efﬁcient Index. Credit-rating agencies are also starting to tackle this ignorance. This is essential, because pension funds and

128

*

Low-carbon choices

other investors seeking ﬁxed income from 15–20-year bonds require transparency about the risks they are facing. One early identiﬁcation of the carbon-exposed business model was the downgrading of the credit rating for Drax, the largest coal-ﬁred power station in the UK. This was Standard & Poor’s response to ‘Drax’s rising business risk because of its focus on coal-based generation, which is subject to increasingly stringent regulatory and environmental requirements’ (Standard & Poor’s 2009). The level and relevance of analysis and reporting on carbon-risk exposure is gradually improving. Initial evaluations compare the direct and indirect carbon emissions of a company against its ﬁnancial performance, thereby identifying carbon-intensive businesses. Subsequent work considers whether companies that produce carbonintensive products can pass on the carbon price to their consumers (Standard & Poor’s 2010). This information could decrease concerns about carbon-intensive companies, if they can demonstrate that they can increase product prices without losing competitiveness. It also allows for intra-sector comparisons. Companies will have a competitive disadvantage and increased risk exposure if they are more carbon-intensive than their peers in the same sector. Future analysis will have to take into account that deep de-carbonisation of our economies will be delivered through changing and optimising production processes and product and service portfolios. Carbon intensity, availability of low-carbon substitutes and innovation of low-carbon alternatives, not to mention evolving consumer preferences, create new market opportunities. However, they also imply declining demand for many carbon-intensive products. Declining markets not only reduce production volumes, but can result in long periods of excess capacity and therefore low margins. Such product and sectoral shifts associated with a deep decarbonisation have yet to be understood and considered by investors and rating agencies in their evaluation of the business models and strategy of companies.

Conclusion

*

129

Further analysis is required to support funds and rating agencies in their evaluation of carbon-risk exposure for equity and bond investments. This could both facilitate access to capital for low-carbon choices and use the inﬂuence that the shareholders have on corporate decision making to support co-operations in their design and pursuit of low-carbon strategies. It also raises questions about how policy instruments, regulatory frameworks and accounting and taxation rules can expose strategies that fail to address carbon considerations – for example, by removing implicit and explicit risk guarantees for carbon-intensive business strategies.

4.7 Conclusion It is generally agreed that large-scale emissions reduction will hinge on investment in low-carbon projects, infrastructure and technology. Private-sector companies are responsible for the majority of investment choices and so policy instruments must inﬂuence their decisionmaking processes. Credible mid- and long-term targets will guide corporate strategy and strategic investments, and thus facilitate development and deployment of new technologies. The commitment of governments to future targets will be judged by the policies they implement today. The perceived consistency of future targets and credibility of current policies depend on government and industry stakeholders developing a shared vision of the transition towards a low-carbon economy. Predicting the future carbon price that will be part of this vision is, however, difﬁcult – and perhaps not even necessary. After all, the future carbon price will have to respond to fuel and commodity prices and reﬂect the costs of low-carbon technologies. Thus, it is not commitment to a speciﬁc future carbon price that is required, but the conﬁdence that the future carbon price will adjust to a level consistent with the emissions target. Cap-and-trade schemes can offer a mechanism to delivers this carbon price.

130

*

Low-carbon choices

In the short term, simple and transparent policy instruments that reduce the risk of low carbon prices can facilitate investments in low-carbon projects. They can also inﬂuence investment in carbon-intensive projects, rather than allowing internal and external supporters of such projects to use the uncertainty of carbon prices as an excuse to continue with carbon-intensive projects. Carbon taxes would offer the necessary simplicity. Cap-and-trade schemes can also deliver price ﬂoors – for example, when combined during the initial years with components such as reserve prices in allowance auctions or government-issued put options on future carbon prices. This chapter has continually referred to the role of national governments. Even though companies act in global markets and climate policy is on the international agenda, national governments still have a considerable role to play. This raises the question of whether national governments can credibly implement low-carbon investment frameworks while the international regime is still evolving. The next chapter discusses the possible evolution of national and supra-national initiatives towards a global effort.

Notes 1 Survey by Point Carbon (2007). www.pointcarbon.com. 2 Based on BERR Energy Statistics 1960–1997, fuel consumption for power generation, transformed to output using 1998 average efﬁciencies. 3 Historic data projections for 20% domestic emissions reductions by 2020 and 80% by 2050. CCS fraction 2% in 2020, efﬁciency 85% in 2020 and 80% in 2050 (European Commission 2006). 4 A survey among utilities at the end of 2005 and the start of 2006 suggested that business required ‘courage’ to make investments where the return is dependent on there being a carbon price in 2013. Hamilton and Kenber 2006. 5 Yang and Blyth (2007) illustrate how the real option of waiting for clarity on climate policy instruments can delay investment in low-carbon power-generation technologies. 6 Pershing and Tudela (2004) discuss the appropriate formulation of such targets (e.g., as global-temperature increase, as CO2 concentration or as CO2 emissions).

Conclusion

7

8 9

10

*

131

Deﬁning targets that are most sensitive to human activity (e.g., annual emissions) avoids uncertainty that results if the impact of human activity on more indirect systems must be analysed (e.g., CO2 concentration or global temperature). Reducing uncertainties then reduces ‘opportunities for discord and delay’. Policies used to deliver the target levels must allow for this ﬂexibility. For example, cap-and-trade schemes should not commit to allocating large shares of free allowances over long time frames, as this would limit the level to which the cap could be tightened. In addition, if large shares of allowances are allocated at no cost and the carbon price rises, then the wealth transfer to the recipients also increases. This might not be politically sustainable and could result in changes to the scheme. The two new Member States Bulgaria and Romania were not yet included in the analysis. Ralf Glienke surveyed ﬁfty-ﬁve German companies in the spring of 2005. Thirtysix listed uncertainty about future allocation as important or very important for their emissions-trading strategy (Fraunhofer Institute, Karlsruhe (ISI): private communication). Schemes that heavily rely on banking of allowances might create a bias in favour of rich organisations and countries. Their existing wealth increases their ability to bank and their willingness to take the risk of uncertain future returns of such banking in order to reap future beneﬁts.

fi v e

Co-operation among developed countries: a role for carbon markets?

Limiting growth in global average temperatures to 2 °C above historic levels can only be achieved if greenhouse gas emissions are drastically reduced across all major emitting countries. It requires joint action across developed and, as discussed in Chapter 7, developing countries. The need for joint action frequently prompts the question whether countries can and should pursue climate policy prior to an international agreement, or beyond the level of ambition thus agreed. Section 5.1 explores how international co-operation can enhance the level of national ambition by creating a sense of responsibility for own emissions and building trust and a sense of joint effort. Thus, international co-operation can encourage and support ﬁrst movers and serve as a commitment device to facilitate effective policy implementation. International co-operation can provide early and tangible results by encouraging and supporting transparent monitoring and reporting (section 5.2). With better information, governments can implement policies and programmes more effectively. It also facilitates rapid international learning on policy success and failure on how to achieve low-carbon development. Of particular interest is the global carbon market. Section 5.3 outlines the design options, including a description and evaluation 132

Domestic commitment

*

133

of trading at government or installation levels and the design options for such trading. Linking national carbon markets is usually described and analysed from the perspective of static efﬁciency gains. However, with increasing focus on low-carbon development, the incentives this can create for innovation and diffusion of low-carbon technology and infrastructure need to be considered (section 5.4). As economic schemes create proﬁts and can transfer rents, the design choices on linking markets can have an impact on subsequent policy responses (section 5.5). While this book primarily uses emissions trading as an example for policy instruments, section 5.6 reviews the options for a global carbon tax and illustrates the parallels with emissions trading, before concluding the chapter with a short summary.

5.1 Using international co-operation to enhance domestic commitment Support nations in developing sense of responsibility for their actions The need for mitigation policies to be pursued on a global scale in order to limit global temperature increase to 2 °C is often taken to imply that there is little reason for individual nations to pursue ambitious mitigation policies, either in advance of international co-operation or above the level of ambition set by international co-operation. However, if the global community fails to deliver a low-carbon transformation of sufﬁcient scale and scope to achieve the 2 °C target, temperatures will increase faster, increasing the likelihood that ice shields and permafrost will melt and that weather patterns and vegetation will change. Every additional tonne of greenhouse gas emissions increases both the risk and impact of climate change. Most societies and ethical systems require that individuals and countries do not damage and harm others.1 From this follows an obligation

134

*

Co-operation among developed countries

for individuals and countries to reduce domestic greenhouse gas emissions that damage and harm others independent of the action of others. Thus, international co-operation is not a requirement or prerequisite for domestic climate-change policy. Perhaps the abolition of the slave trade and, later, of slavery itself, could serve as an example. They were abolished not in a global action, but country by country. And, as in the case of climate change, the transition had to overcome opposition from incumbent stakeholders.2 As countries begin pursuing ambitious domestic mitigation policies, taking responsibility for the impact of their own emissions, they discredit the arguments used by other governments for inaction or action at low levels of ambition. Thus, inactive governments will increasingly struggle to maintain the integrity of their position in interactions with domestic and international constituencies. As more countries take responsibility, the social and political pressure will increase on those countries not pursuing effective mitigation action. Returning to the example of the abolition of slavery, it would be economic suicide for any country today publicly to embrace slavery; it would result in the culpable groups in the country being excluded from the international community To contribute to the process of societies accepting their responsibilities, international co-operation can create a platform for policy makers and broader society jointly to discuss climate policy. Seeing potential damage wrought by climate change, even in foreign countries, can bring home to governments how people suffer from the damage caused by climate change.

The ‘free rider’ issue In simpliﬁed economic models, optimal government policy maximises the welfare of a country. This is modelled as the costs of pursuing mitigation action and the utility derived from the environment. With increasing mitigation action, more beneﬁts can be derived from the environment. The optimal level of mitigation action is reached when

Domestic commitment

*

135

one extra unit of mitigation action equals the beneﬁt from this action to the country. If the county can credibly calculate and is exposed to, let us say, 5 per cent of the global damage of climate change, then 95 per cent of the beneﬁt of reduced climate-change impact falls on other countries. The model suggests that a government should not consider the impact of mitigation on other countries, but only pursue mitigation action on a level justiﬁed by the beneﬁt it creates for its own domestic population. In our case, this would be only 5 per cent of the level considered by a government that takes full responsibility for damage imposed by national emissions. Concerned that other countries might ‘free ride’ on domestic emissions reduction, some stakeholders argue that only limited mitigation efforts should be pursued unilaterally. They argue that moreambitious mitigation efforts should only be pursued as part of an international agreement. If a country agrees to increase its mitigation effort, then all other countries would also increase their mitigation effort. Thus, the country not only beneﬁts from the direct impact of its mitigation (5 per cent) but also from the mitigation pursued by the other countries (95 per cent). As a result, an agreement can be reached that is welfare optimal, from a global perspective. However, countries differ in their perception of the level of urgency of the need to address climate change and thus also in the urgency with which they support mitigation policy (Figure 5.1).3 If all countries could come together and jointly agree the level of global mitigation effort, then a global agreement would be constrained by the level of support for mitigation effort expressed by the least ambitious country. If the level of action were increased, the least ambitious country would not increase its effort, thus also reducing the incentive for other countries to increase mitigation action. Thus, the likely outcome would be the pursuit of mitigation policy at the lowest common level of support. This approach to assessing climate change policy from a game theory perspective has been fertile ground for the analysis of coalition

136

*

Co-operation among developed countries

Ambition in line with domestic support Ambition set at lowest common denominator

70 60 50 40

India

Philippines

Brazil

South

Egypt

Turkey

Kenya

Mexico

Nigeria

Indonesia

Italy

Chile

USA

Developed countries

Russia

France

UK

Spain

Canada

Germany

30 Australia

Level of domestic support (percentage of population which state that they support carbon tax)

Survey result for level of domestic support

Developing countries

Figure 5.1 Extremes of setting climate policy targets (22,182 total participants). Source: BBC 2007.

formation, stability and the use of side-payments and punishment strategies. What seems to be too often ignored is that economics is a social science and thus inherently normative. The way in which economists talk about an issue can inﬂuence the way governments will or can talk about the issue. In the economic models, damage is codiﬁed as negative beneﬁt. Thus ‘reducing damage to other countries’ is formulated as ‘providing beneﬁt to other countries’. This in turn is the basis for the statement that other countries ‘free ride’ on our emissions reduction. Translating the abstract concept of ‘free-riding’ into concrete terms would sound inﬂammatory (e.g., ‘Europe should ignore the impact of European emissions on rainfall and drought patterns in Africa, unless Africa in turn reduces emissions’). The legal treatment of pollutants gives some indication of the moral values encoded in our legal systems, which also seem to consider the damage that our actions impose upon people in other countries. The multinational businessTraﬁgura paid $210 million in an out-of-court settlement to

Domestic commitment

*

137

the government and people of Ivory Coast and was in 2010 found guilty by a Dutch court for toxic waste ‘exports’ to Ivory Coast in 2006. Structuring international negotiations on climate co-operation according to the concern that other countries ‘free ride’ on emissions reduction has only limited potential, as it can only support an agreement at the lowest common denominator of ambition of countries. The European Union attempted to link a shift from 20 per cent to 30 per cent emissions reduction to the effort agreed by other countries in an attempt to increase the incentive for other countries to pursue mitigation action (see Ashton and Wang 2004),4 but seems to have failed to prompt international response in Copenhagen. As the approach does not seem to be very useful, and does not reﬂect the moral values of our societies, other approaches to framing and focusing national and international discussions about climate change are needed.

Building trust and sense of joint effort International co-operation contributes to an enhanced sense of responsibility for and motivation to reduce emissions, and facilitates low-carbon development. It can also contribute to a system that builds conﬁdence and ramps up ambition over time. Does this require an agreement on a joint level of ambition? A common level of ambition, agreed in a global forum, would probably be close to the level of ambition of the least ambitious country. It would be difﬁcult to agree and to implement a level of stringency that exceeds the level of domestic support in the least ambitious country. This was illustrated by the experience of the US delegation to Kyoto. The US delegation, with the support of the US Administration, negotiated an international agreement with fairly ambitious targets, but subsequently failed to gain the support of the domestic policy makers necessary for ratiﬁcation by Congress. Thus, if countries pursued mitigation actions according to their perceived sense of urgency, to match the scale of global damage, the outcome would be more ambitious than a global agreement based on a

138

*

Co-operation among developed countries

common level of ambition. International co-operation can create the basis for understanding the speciﬁc circumstances of a country so as to provide suitable support and encouragement. This will be informed by internationally joined assessments of whether humanity is undertaking an adequate response to climate change and will achieve the objective of limiting global temperature increases to the 2 °C target.

Encouraging ﬁrst movers International co-operation can encourage countries to accelerate their low-carbon development efforts. Countries invest large sums, often even signiﬁcant relative to GDP, to host international events such as World Cups, the Olympics or the World-Expo. National pride might equally be an incentive for taking leadership on low-carbon development, pride enhanced by international publicity identifying successful countries. Early movers on the various dimensions of low-carbon development also provide the framework for technology-oriented companies to develop and explore low-carbon products and services, thus gaining the beneﬁts of attracting and growing new industries. This in turn attracts capital, creates job opportunities and fosters experience with new technologies that will in turn help future growth. An early shift to investment in low-carbon infrastructure also reduces the risk that countries will invest in infrastructure and building projects that are not compatible with future energy-efﬁciency requirements and would thus not provide acceptable returns on the investment in the longer term. Early movers thus facilitate the early deployment and diffusion of low-carbon technologies and allow for learning on policy design, to fast-track effective and efﬁcient instruments.

Using co-operation mechanisms as a commitment device Climate-change mitigation is a long-term challenge, and thus requires a combination of long-term objectives translated into short-term actions. International co-operation can provide opportunities to enhance

Domestic commitment

*

139

commitment to long-term objectives, and can also provide a platform to assess whether short-term responses are aligned with long-term targets. The design of phase II of the EU ETS provides a helpful illustration. In 2007, the European Commission rejected several national allocation plans from Member States, as the plans did not comply with the obligations of the Member States under the Kyoto Protocol and subsequent EU burden-sharing agreement (see Chapter 3). In contrast, owing to the absence of such targets for the EU ETS pilot phase in 2005–2007, the Commission had no leverage to be more stringent in the evaluation of national allocation plans submitted by Member States, which resulted in excessive allocation of allowances and an inﬂated emissions cap. Just as important as the quantitative targets was the timeline implied by the Kyoto Protocol for the implementation of the emissions-trading framework. This prevented various stakeholders from delaying the implementation of EU ETS. Many climate-change mitigation efforts require signiﬁcant investment in infrastructure, in more energy-efﬁcient technology and in innovation and deployment of renewable power generation. Investors are concerned that investment choices pursued in light of the expectation of an ambitious climate policy framework might turn out to be less proﬁtable if a new government were later to weaken the policy objectives.5 Strong dependence upon regulatory frameworks might undermine the ability of governments to attract investment in low-carbon transformation. Could international processes be used to create external commitments, and thus enhance the credibility of domestic policy frameworks, so as to make them more effective? 6 Credibility of international commitments This raises the question about the credibility of international commitments. Henkin (1979: 47) explores the historic performance of international commitments and summarises: ‘It is probably the case that almost all nations observe almost all principles of international law and almost all of their obligations almost all of the time.’ Yet scholars

140

*

Co-operation among developed countries

heatedly discuss the extent to which commitments, which require signiﬁcant changes in government policy (as would potentially be the case in the climate domain) can achieve full compliance (Guzman 2008). Barrett (2003) questions whether it is possible to design a selfenforcing climate treaty to ensure full climate co-operation (a key point is the monitoring and veriﬁcation of the fulﬁlment of commitments made), and hence explores fallback options. International co-operation can enhance the level of domestic commitments to climate policy and provide time frames to drive domestic policy implementation and improve inter-temporal credibility of policymaking. The robustness of commitments over time and across changing constellations of national governments is likely to depend as much on the level of support they receive domestically and internationally as on the speciﬁc institutional and legal framing. Thus, the next section explores how transparent monitoring and reporting can both increase the resources dedicated to such agreements, thereby increasing the political costs of non-compliance, and facilitate international learning about effective policy implementation, thereby reducing the costs of compliance.

5.2 Transparent monitoring and reporting In the private sector, it is common practice to monitor and report the performance of business units, production processes and relevant markets. Problems can therefore be identiﬁed and resolved quickly, ideas can be tested and best practice established, and management has a basis for strategic choices. Equally, governments require robust information to make good choices on design and implementation and enforcement of policies. Transparency of the monitoring and reporting process is essential to ensure information can be scrutinised by a broad set of independent public and private actors. While it is virtually impossible for any third party to test whether aggregate ﬁgures provided on a national level are

Transparent monitoring and reporting

*

141

plausible, let alone accurate, inhabitants of a city, or companies in a sector, are in better positions to judge the accuracy of regional and sectoral disaggregated information. Thus, transparency creates trust in the robustness of the information, and makes the information useful – for example, for investors, who need to understand the political framework and ﬁnd robust evidence of its implementation and impacts, in order to evaluate and pursue low-carbon investment choices. However, the accuracy and scope of data available decreased in the decade prior to the global ﬁnancial crisis in 2008. As government regulation was increasingly criticised and constrained in globalising markets, the effort in data collection and management declined in many instances. The failure of the banking system in 2008 and the difﬁculties of identifying and formulating an adequate response to the economic crisis illustrate the potential costs for societies if governments cannot monitor and respond to systematic risks. International co-operation can contribute to increasing the scope and quality of information available to policy makers and the broader public. It can ensure high-level political support for initiatives to improve data availability, even by heads of governments, thus helping overcome inertia at the level of individual government agencies and providing a time frame for action. This could counter incentives for individual public and private actors to delay the introduction of transparent monitoring and reporting, where this could identify problems that expose mistakes or require action. If multiple governments jointly increase information requirements, private-sector companies are equally affected across jurisdictions, thereby tackling the already weak argument that the costs of monitoring and reporting would create competitive distortions. International co-operation on monitoring and reporting can contribute to harmonisation on indicators so as to facilitate international comparison of policies and programmes. Thus, co-operation not only facilitates the diagnosis and improvement of domestic policy frameworks, but also supports other countries in their implementation of de-carbonisation strategies. For this monitoring needs to

142

*

Co-operation among developed countries

cover more than sector-level greenhouse gas emissions because there are long delays in the response of emissions levels to policy actions, and thus while emissions reporting is crucial, it is not sufﬁcient for timely policy analysis and response. b o x 5 . 1 Experience with policy indicators Indicators to measure performance of policies and programmes are widely applied. In Poverty Reduction Strategy Papers (PRSPs), the least-developed countries negotiate with the World Bank policy improvements for a three-year horizon. If they deliver policy objectives, some debts are relieved (Coudouel, Hentschel and Wodon 2002). In Local Public Service Agreements (LPSAs) local authorities discuss with the UK central government areas of policy improvement, and receive funding if they deliver against policy metrics in a three-year horizon (DTLR 2001). Figure 5.2 also includes the Government Performance Results Act targets for central administration in the USA (US Senate 1993), the accession process of new Member States to the European Union and the Millennium Development Goals scheme (Black and White 2004). Policy target applied to Input

Time frame (years)

1

3

Government Performance and Results Act

Output

time Poverty Reduction Strategy Paper

Eur

ope

10

Response

Public Service Agreement

an

Un

ion

Acc

ess

ion

(UN) Millennium Development Goals

15

Figure 5.2 Time frames used for the deﬁnition of policy targets and differentiation between input-based and output-based metrics. Source: Lester and Neuhoff 2009.

International co-operation

*

143

b o x 5 . 1 continued The horizontal axis shows that in most cases it is not only the ﬁnal output but the successful implementation of policies or delivery of intermediate outputs which are measured. The vertical axis depicts time frames over which the policy targets are deﬁned. With shorter time frames, output-based metrics cannot be used, and intermediate output-based metrics are used. Several reasons can contribute to the use of intermediate outputbased metrics: * they allow for shorter time frames in line with political time-horizons (elections) * learning from early experiences can be used to improve implementation * it is difﬁcult to predict outcomes from transformational change and to attribute these to speciﬁc policies. Experience suggests that successful metrics are appropriate, relevant, selective and simpliﬁed, capturing cross-cutting outcomes.

5.3 Carbon-market-based international co-operation among developed countries The Kyoto Protocol and its provision for emissions trading envisaged an important role for international carbon markets in tackling global climate change. The Protocol deﬁned a framework for the trade in AAUs among developed countries. A country that is in compliance with its commitments under the Protocol, and anticipates that its emissions will be lower than the emissions cap that it has negotiated, can sell AAUs to a country that might desire to use these AAUs to cover domestic emissions that exceed its emissions target under the Kyoto Protocol. Trade in AAUs among countries creates a ﬂexible compliance mechanism for countries participating in the Kyoto Protocol. Reducing domestic emissions reduces the costs of buying AAUs and eventually can create revenue from the sale of surplus units. Thus,

144

*

Co-operation among developed countries

countries are incentivised to pursue emissions reduction, even if the national target has been achieved. Likewise, even where a country assumes that it will not be possible to achieve the national target, incentives remain to pursue emissions reduction.

International trade: on the country or installation levels? Countries can either trade AAUs directly with other countries or create an emissions-trading system that covers domestic installations and allows these installations to engage in international emissions trading. What are the relative merits of each approach? The effectiveness of any international mechanism hinges on countries’ commitment to their national emissions targets and compliance with the requirement to acquire AAUs or other equivalent units – for example, certiﬁed emission reductions (CERs) – for excess emissions. If trade is pursued at government level, this might require large governmental transfers to other countries. The reluctance of the USA to cover its emissions-reduction gap by buying AAUs from Russia, the country with the largest excess of AAUs, might well have been one of the reasons that the USA did not sign up to the Protocol. It seems to be more acceptable, at least in Europe and the USA, for private installations to buy greenhouse gas allowances in international markets. From this perspective, installation-based trading might provide a more robust compliance mechanism. In addition, international emissions trading might be more robust if pursued at installation rather than at government level, because it creates groups of investors and producers who support the continuity of carbon price for their investment returns and market opportunities, thereby creating political momentum towards a continuation of the scheme. Direct engagement by installations in international carbon trading can also beneﬁt the transparency of the carbon price: more actors involved in trading and interested in hedging future positions enhance

International co-operation

*

145

liquidity of the market. In contrast, AAU trade among countries is less frequent and might reﬂect political preferences for certain countries or price discounts or premiums for additional attributes. For example, countries with surplus AAUs negotiate deals including ‘greening AAUs’ to reassure buyers of the environmental integrity of the trade. However, even the larger number of actors involved in installation-based international trade might not be able to deliver a competitive price. Governments can choose to constrain (or tax) installations on national territory and thus impact on global markets. While installation-level carbon trading does ensure a carbon price for private actors, government-level AAU trade can also contribute to establishing a price for carbon. Once a country faces costs or, with each unit of domestic emissions, forgoes revenue from international trade in AAUs, it is more likely that the government will consider these costs when deciding on domestic programmes and policies, including policies to impose comparable carbon prices on domestic emitters. One advantage of government-level emissions trading is that governments beneﬁt from domestic emissions reduction through the sales of AAUs or reduced costs of buying AAUs, and thus have incentives to pursue complementary policies and programmes to support additional emissions reduction. With installation-based emissions trading, governments supporting low-carbon frameworks reduce the need for domestic installations to import allowances (or increase the volume of exported allowances). The increased proﬁtability of domestic companies, however, only partially feeds back into national budgets. Thus, it is difﬁcult to make general statements on whether installation- or government-level international emissions trading is preferable. Europe’s answer to this problem encompasses both aspects. In their national allocation plans, EU Member States dedicate some of their AAUs to the EU ETS. Installations covered by the scheme

146

*

Co-operation among developed countries

can then participate in international carbon markets as deﬁned by the European Linking Directive (Directive 2004/101/EC of the European Parliament and Council). However, installations cannot buy AAUs. Thus, both markets are clearly separated. National governments retain responsibility for emissions not covered by the EU ETS, and can trade AAUs with any party to the Kyoto Protocol that has a deﬁned emissions target under the Protocol. Figure 5.3 illustrates the parallel markets for trade of AAUs among countries and potential opportunities for trade in allowances at installation level. The remainder of this section explores ﬁrst the role of direct then of indirect linking of emissions-trading schemes.

Direct linking of cap-and-trade schemes A multitude of different cap-and-trade schemes is being developed and discussed across the EU, the USA and some of its states, Australia, New Zealand and Canada. If two of these schemes are linked directly, market participants could buy allowances in either scheme to cover their emissions. If we assume the Australian scheme and EU ETS were directly linked, and the Australian carbon price Country A

Country B

National budget

Trade of assigned units

Cap-and-trade scheme

via Indi pro rect jec lin tc k red its

Direct linkage of schemes

National budget

Cap-and-trade scheme

k lin its ct cred e r t di In rojec p via Credits from clean development mechanism/joint implementation projects

Figure 5.3 Three main channels for linkage between countries

International co-operation

*

147

was lower than the European price, a European emitter could buy ‘cheaper’ allowances in Australia to cover its emissions. This would reduce the allowances available for Australian emitters, and one additional emissions reduction must be implemented in Australia, albeit at lower cost than in the EU, and supported by the ﬁnancial transfer from the EU to Australia. Emitters and traders will arbitrage both markets in such a way that in the end the carbon allowance price in Australia will equal the EU price. High-cost emissions reduction in the EU is avoided and replaced by lower-cost emissions reduction in Australia. In the short term, this allows the collective emissionsreduction goal to be achieved at a lower cost than if each country met its own target solely through domestic reductions.

Indirect linking The Kyoto Protocol also deﬁned the CDM and JI mechanisms to allow for trading of emissions reduction on the project level (Articles 12 and 3 of UNFCCC 1997). As several emerging cap-and-trade schemes allow their emitters to use project credits to cover their emissions, emitters in different schemes can compete for these project credits. As of mid-2010, almost all demand for project credits originates in the EU and Japan. Yet, if there were demand from different countries, credits would go to the trading scheme or country with the highest carbon prices or willingness to pay, thus reducing scarcity and causing the carbon price to decline in that scheme. Over time, this link could lead to an equalisation of global carbon prices and willingness to pay in countries pursuing government purchases, provided the global CDM and JI markets become sufﬁciently large and the constraints on countries’ use of CDM credits in their domestic schemes (supplementarity criteria) are not binding. One concern arising with regard to such indirect linking relates to the criteria and deﬁnition of projects qualiﬁed to create CDM credits. All project credits to be used under the Kyoto Protocol

148

*

Co-operation among developed countries

must be certiﬁed by the CDM executive board. However, the USA has not ratiﬁed the Kyoto Protocol and is therefore not bound by the decisions of the CDM executive board. What would happen if it were to allow a broader set of projects to qualify and produce project credits? US emitters who initially acquired CDM credits, perhaps in anticipation of a national US carbon market, would then sell these credits and acquire projects that might be cheaper to realise, given that the requirements imposed upon them would be less stringent. A potentially different deﬁnition under the US cap-and-trade scheme could thus increase the number of credits available for other schemes, creating volatility and reducing carbon prices. Ironically, the very existence of other schemes creates a political environment that makes such a change of deﬁnition possible. If other schemes did not exist, emitters owning ‘expensive’ CDM projects would oppose a relaxation of the certiﬁcation standard to prevent devaluation of their projects. This illustrates the importance of international coordination for the deﬁnition of off-sets with common eligibility/accounting/reporting standards. The European decision to preclude the use of CDM credits from hydroﬂuorocarbon and adipic/nitric N2O projects post 2012 suggests that the governance structure of the CDM executive board constrains the ability to tighten standards and leads regions to set unilateral criteria.

5.4 The economics of carbon-market-based co-operation mechanisms In many aspects, the economic case for international carbon markets parallels the economic case for an extension of cap-andtrade schemes to cover additional economic sectors. In particular, this section discusses the static efﬁciency gains from trading in a larger scheme and raises questions about the dynamic efﬁciency of an early integration of national cap-and-trade schemes.

Co-operation mechanisms

*

149

Static efﬁciency gains from large trading schemes A larger trading scheme ensures that proportionately more investment and operational decisions are based upon a common carbon price. By contrast, if two regions were covered by separate cap-and-trade schemes, equilibrium carbon prices might be higher in one of the regions. Companies and consumers are incentivised to make investment, operational and consumption choices viable at the high carbon price. This is inefﬁcient from a static perspective, if at the same time lower-cost mitigation options in the other region are not pursued. Larger emissions-trading schemes include additional market participants, who are also more inclined to trade, because of the more-robust price and greater liquidity: making it easier to ﬁnd counterparties with matching requirements on time frame and structure of trades. Prices will vary less in response to changes to emissions at company level or in relation to regional economic activities (Bell and Drexhage 2005). As more countries link their schemes, the effect on prices of political decisions in any one country will be reduced and the scheme will be less exposed to political uncertainty (Aldy, Baron and Tubiana 2004). The larger the market, the smaller the market share of individual actors, thereby also reducing concerns about the possible exercise of market power. Over time, unexpected economic growth, surprising weather conditions or, as in the case of Japan, the long-lasting outages of several nuclear power stations might produce additional emissions and thus more allowance scarcity in a country. A joint carbon market has the beneﬁt of additional ﬂexibility to respond to such unforeseen situations. These unexpected shocks create ﬁnancial transfers, but they are likely to be acceptable where all parties beneﬁt from such ﬂexibility.

Dynamic efﬁciency gains of separate schemes if some countries pursue more-ambitious policies If some countries implement more-ambitious climate policies, this will accelerate innovation and the development of low-carbon

150

*

Co-operation among developed countries

societies. Companies will focus their management, research and ﬁnancial resources on low-carbon solutions if an ambitious policy framework offers the credible prospect of a future market. Being an early mover in low-carbon policies might also offer economic beneﬁts, such as a reduction of import dependency on fossil fuels (Edenhofer and Lessmann 2005). The example of some ambitious countries can, in turn: (i) enhance the political standing of groups in other countries that pursue socially responsible actions in their country; (ii) facilitate the adoption of more-ambitious objectives that have proven to be feasible; and (iii) accelerate emissions reduction in other countries, drawing on the behavioural examples, technologies and policy frameworks developed in early-mover countries (see Grubb, Köhler and Anderson 2002; Rosendahl 2004; Sijm et al. 2004). More-ambitious countries could continue to use a linked capand-trade scheme and expose their industry and consumers to the same carbon price. However, if carbon prices are seen as an important component of the policy package, more-ambitious countries might also set tighter caps for their emissions-trading schemes and thus create higher carbon prices. In a linked scheme the tighter cap of the more-ambitious country would have an impact upon all regions, with three implications. First, the carbon-price increase would be more moderate and thus the incentive for innovation reduced. Second, the tighter cap would reduce the allowance sales of the country and increase (net) imports by installations, so linking partners would beneﬁt. Third, some linking partners might object to the higher carbon price (see the discrepancies on monetary-policies priorities between European countries, dependent upon speciﬁc national circumstances). In sum, with linked emissions-trading schemes it is more difﬁcult and less rewarding for an individual country to increase its level of ambition on mitigation policy.

Political economy

*

151

5.5 The political economy of carbon-market-based instruments Alignment of cap-and-trade schemes with political responsibility National governments must pursue a wide range of policies on climate change (including the provision of information, regulation, market design and supporting infrastructure development) in order to facilitate energy- and carbon-efﬁcient investment, operation and consumption choices. The role that will be attributed to individual policy instruments is likely to differ across countries. Studies suggest that many cost-effective energy-efﬁciency and emissions-reduction opportunities exist (see Figure 2.5). One reason why this latent potential has not been addressed could be the reluctance of governments clearly to deﬁne and commit to policies of sufﬁcient scale and stringency to capture these potentials. This would require metrics to measure, and targets to deﬁne the success of, policies. A quantitative framework facilitates the provision of information to manage and improve policies and to give their implementation sufﬁcient priority. Clearly deﬁned national targets and milestones are therefore important parts of climate policy in order to facilitate implementation, benchmarking and execution of non-carbon-price policies by governments. National targets and trajectories are receiving growing emphasis in the EU climate package. EU Member States are also increasingly using national emissions-reduction targets to guide domestic policy. For example, the Climate Change Act 2008 in the UK prescribed a trajectory and an independent climate-change committee to audit not only whether the UK government is meeting past targets but also whether policies implemented by the government can be expected to deliver the required emissions reduction. National targets can be particularly effective if the territory for which the emissions target is deﬁned coincides with the geographical coverage of policies and programmes. However, this is not always

152

*

Co-operation among developed countries

possible and trade-off needs to be considered. If an emissions target is only deﬁned at a European level, so as to be aligned with the cap of the EU ETS, then performance of individual countries might not be measured and national governments might not take responsibility for mitigation action, creating the risk that national policies in support of the EU ETS will be inefﬁcient and watered down. Alternatively, if targets for transport emissions are allocated at the city level, with no responsibility remaining at national or European level, there is limited motivation for national- and European-level legislation on fuel efﬁciency and carbon pricing to support local action on urban planning and public transport schemes. Thus, neither a central nor a decentralised deﬁnition of targets and allocations to deliver against these targets is effective on its own. Responsibility for emissions reduction will have to be shared across different administrative levels. This is a common procedure in the ﬁnancial budgeting of organisations and countries. The central authority has responsibility for balancing the budget, then it splits that budget and allocates it to individual departments, which again are responsible for balancing their internal budget and are likely to make further budgetary breakdowns. This discussion illustrates the trade-offs involved in deﬁning the territorial coverage of emissions targets and in allocating political responsibility for policy implementation. Clarity on roles and responsibility is essential.

Implications of joined-up schemes for future climate negotiations Cap-and-trade schemes can only be linked if the countries have committed to ﬁrm emissions caps. In Europe, national emissions caps have been set in an interactive way, ﬁrst for 2008–2012, and, as of summer 2010, it is discussed whether to increase the level of emissions reduction for the period 2013–2020 from 20 per cent to 30 per cent. While deciding about targets in the EU context is a complex

Political economy

*

153

matter, the question arises: how would negotiations be held if emissions-trading schemes were more widely linked? The answer depends upon the basis for the international negotiations on emissions targets. A variety of approaches has been discussed. First, a country’s recent emissions can serve as a reference point for negotiations about the level of efforts to reduce emissions. Second, equal emissions budgets can be deﬁned per capita, combined with trading schemes, to allow countries with higher emissions levels to acquire some of the allowances from countries with lower emissions while they are reducing their own emissions levels. This is reﬁned in a third option, ‘contraction and convergence’, which envisages a gradual move from historic emissions levels towards equal per capita emissions budgets, for instance by 2050 (GCI 1996). As long as the focus of international discussions remains on the effort necessary to reduce emissions, there is a major drawback to joining up emissions-trading schemes of countries that are negotiating separately: it reduces the incentive for each country to aim for ambitious emissions-reduction targets. Under a joint cap-and-trade scheme, industry in a country or region with more-ambitious caps will buy allowances issued in other countries. This creates transfers to the less-ambitious country. Negotiation results will be judged not only by the level of ambition of the ‘global deal’ but also by the anticipated volume of transfers to competitors. Linking emissionstrading schemes of countries that are not jointly negotiating could thus reduce the level of ambition that can be achieved in climate negotiations.

The dynamics of linking emissions-trading schemes Linking is a politically attractive option, offering the opportunity to mirror political links in economic instruments, perhaps even using the instrument as a centrepiece for international efforts to build a global and comprehensive greenhouse gas mitigation regime (Philibert and Reinaud 2004). Usually a direct linking of schemes is

154

*

Co-operation among developed countries

envisaged – for example, allowing emitters of the Western Climate Initiative (WCI) in the USA to cover their emissions with EU allowances and vice versa (ICAP 2007). Such linking effectively integrates the cap-and-trade schemes of two jurisdictions. However, linking of schemes also creates a risk. Assume that one country abandons a linked emissions-trading scheme – for example, because of costs of acquiring allowances in international carbon markets. This risk is not unique to linked trading schemes, but with such linking the implications are more far reaching. Concerns about scheme abandonment by policy makers would not only jeopardise investment decisions in one country but also have implications for the linked countries. Should a scheme be abandoned, a broken link to the previously importing country would reduce demand for allowances in the remaining countries and result in lower allowance prices. Linking markets also creates the risk that individual countries will only consider the impact of their actions upon domestic industry and consumers, and ignore the impact of their decisions in linked countries. This was illustrated in Chapter 3, with the experience of the design of national allocation plans under the EU ETS. Member States were happy to allocate too many allowances to their domestic industry, and were less concerned about the impact of domestic overallocation on the overall scheme. Scarcity in the EU scheme required the resolute intervention of the European Commission in demanding that Member States implement cutback in the allocation of allowances under the national allocation plans. This suggests that schemes can only be linked where ‘free-riding’ on the scarcity created by other countries can be avoided. Credible commitment to emissions targets, which create binding caps in the participating countries, is thus a prerequisite. Experience from monetary policy might offer a guide to linking cap-and-trade schemes. Many European countries have ‘linked’ their currencies by adopting the euro. This raised concerns that individual

Global carbon tax

*

155

countries would ‘free ride’ on the monetary stability provided by others. Therefore, all participating countries agreed on criteria for their ﬁscal policies (Maastricht Treaty: European Union 1992). Enforcement of the criteria has proven difﬁcult and has only been achieved in a broad sense, but so far sufﬁciently well to ensure a robust currency, even during the ﬁnancial crisis. It is difﬁcult to say whether the Maastricht criteria and bilateral negotiations would have been enough on their own to deliver this outcome, because euro zone members’ EU membership also provided authority and leverage to encourage compliance. Linking emissions-trading schemes is a process, and even where strong links are initially not desired, it can be important to ensure that the design of national policies is ready for future linking. The short-term political ambition of national and regional actors to participate in some international emissions-trading framework might remain a means of ensuring early compatibility between schemes. Even where it does not result in the direct and immediate linking of schemes, it will avoid the need for future harmonisation. Such harmonisation will be difﬁcult once schemes are in place, because changes will create winners and losers and the potential for political opposition. Thus, early co-operation can create the opportunity for future linked emissions-trading schemes as a core component of a long-term solution to allocating global carbon budgets. Linked schemes could then provide a mechanism that would allow developed countries to support emissions-reduction measures in developing countries and thus ﬁnd the least-cost mitigation solution.7 The implied transfers might be a component of co-operation between developed and developing countries (see Chapter 7).

5.6 A global carbon tax Given the concerns about a global carbon market, some authors have expressed a preference for a global carbon tax (Nordhaus 2006). Although the tax would be implemented and collected at the national

156

*

Co-operation among developed countries

level, one might envisage an international process to determine the coverage and the level – or at least minimum level – for a carbon tax. Such an approach would not require regions such as the EU to abandon existing cap-and-trade schemes. It would sufﬁce if the scheme’s coverage, stringency (its minimum price level) and the exposure to real costs (auctioning) created an economic effect comparable to an agreed minimum carbon tax level. International co-operation over a common or minimum carbon tax level might deliver three objectives. First, it could contribute to the creation of political momentum for domestic implementation, in which countries would perceive their action as part of a global development. Thus, international co-operation could also provide time frames to drive domestic political processes and offer outside commitment to enhance their effectiveness. This requires careful balancing of domestic sensitivities, which for historic reasons seem to be particularly strong on taxation. The reassurance that tax revenues remain under national control might be essential to the political acceptability of international cooperation on carbon taxes. A process that engages all countries in the design of (and thus ensures ownership of) an international agreement on taxation will be essential. This poses the risk that a harmonised (minimum) tax level will be guided by the domestic politics of the least supportive country, and might therefore not be particularly high. This mirrors the challenge for agreeing a common level of emissions reduction among countries. The advantage of a harmonised international tax is that the tax level is not affected if a country joins or leaves the group. In the case of emissions trading, the net trade volume of the country joining or leaving alters the scarcity price for the remaining group. Second, an international commitment to a (minimum) carbon tax could enhance the continuity and credibility of a carbon tax. This would make the tax more effective in guiding investment decisions in energy-efﬁciency and low-carbon technologies. Some level of continuity is already guaranteed where governments use carbon taxes to

Global carbon tax

*

157

replace other revenues and thus depend on the revenue stream generated by the tax. Many de-carbonisation scenarios envisage a signiﬁcant increase of carbon prices over time, but the example of the abandoned fuel duty escalator in the UK illustrates that governments struggle credibly to commit to such trajectories. International co-operation might thus be valuable if it could enhance the commitment to a carbon tax trajectory. The third motivation for an internationally harmonised carbon tax level is that this avoids competitive distortions that might otherwise result if carbon prices differed signiﬁcantly across countries. Again, the challenge of engaging all countries and agreeing ambitious levels of prices is equivalent to the difﬁculty of achieving this outcome with joined-up emissions-trading schemes. In the past, countries have been protective about their right to determine their own policies and measures, and in particular taxation is a sensitive topic. Hence, the successful implementation of a carbon tax might require domestic initiative. Ultimately, much of the discussion on whether linked-up cap-and-trade approaches or globally agreed (minimum) level of carbon tax/carbon price is preferable depends on the effectiveness of the policy instruments in delivering low-carbon investment and innovation. This might be to a large extent determined by aspects of their domestic implementation (see Chapter 4). Finally, it is essential to compare not the theoretical models of policy instruments but the design and stringency of a policy instrument implemented after it ‘survived’ the political implementation process. For emissions trading, the generous free-allowance allocation can create perverse incentives but offer the opportunity to accommodate stakeholder interests using free-allowance allocation while retaining the stringency of the cap. In the case of carbon tax schemes, the political process might result either in a reduced tax level or provide for exemptions or rebates – for example, for energy intensive users.

158

*

Co-operation among developed countries

5.7 Conclusion International co-operation can increase the ambitiousness of domestic climate policies. It can contribute to a positive dynamic to overcome opposition by individual stakeholders and can accelerate the process of acknowledging responsibility for the damage and harm carbon emissions impose on others, thereby encouraging emissions reductions to a sustainable level. International processes and mechanisms provide opportunities for nations to commit to time frames and time scales of future actions, thus supporting implementation of domestic climate policies. Outside commitment also enhances the credibility of domestic policies and makes them more effective in attracting investment in low-carbon choices. Transparent monitoring and reporting provide robust information to allow effective policy design and implementation and reduce uncertainties that inhibit or delay private-sector investment in low carbon transformation. It can also facilitate international learning about successful policies and technologies, thus contributing to global mitigation. International co-operation can create high-level sign-up and outside commitment for effective and timely reporting, and thus overcome inertia and internal opposition by actors who beneﬁt from delayed action or ineffective response. Harmonisation of some key indicators might also facilitate international comparison, to identify best practice. Carbon markets are frequently talked about as a mechanism for international co-operation. Deﬁned at country level, international trade of allowances can create incentives to achieve and exceed emissions targets. The carbon price that emerges from such countrylevel trading would inform and support domestic mitigation policies, including carbon pricing. Moving international emissions trading to the installation level multiplies the level of actors and delivers a lesspolitically guided carbon price, albeit one obviously still guided by the

Conclusion

*

159

expectations about politically determined stringency of targets. It would also avoid the political challenge of direct government transfers to competing countries that are faster in delivering emissions reductions, replacing it with less-contentious and thus more-robust transfers between private actors in participating countries. Such a global carbon market could evolve through linking of national capand-trade schemes or it could be based upon indirect linkages through off-set markets. Linking carbon markets can combine the political momentum of joint action with the beneﬁt of harmonised carbon price levels that avoid competitive distortions and risks of carbon leakage (see Chapter 6). While static economic analysis shows the efﬁciency improvements that larger markets offer by capturing least-cost mitigation opportunities, dynamic economic analysis points to the forgone opportunity for accelerated technology development and exploration of de-carbonisation strategies where individual countries are willing to take more ambitious action than the average. The main challenge of joint carbon markets relates to their impact upon international negotiations on emissions targets. For international emissions trading to work, such targets are essential. But if markets are linked, a country that commits to more-ambitious targets will end up seeing its installations buying allowances from governments in other, possibly rival, countries at the expense of domestic sales revenue for allowances. Linking thus reduces the incentive for nations to set more-ambitious short-term emissions targets unless they negotiate jointly, as in the case of the EU. International co-operation on a harmonised (minimum) price level of carbon taxes presents an alternative to international carbon markets. These can also create political momentum, offer some outside commitment to support domestic implementation and address concerns about leakage associated with different levels of carbon prices. They are, however, more prescriptive about the speciﬁc policy instrument (carbon price) to be used by individual nations than international co-operation, which focuses on the formulation of

160

*

Co-operation among developed countries

emissions targets, and they fail to deliver prices above the level agreed to by the country least interested in an ambitious carbon price. This discussion suggests that carbon pricing might be applied at different levels across countries and sectors during the ﬁrst few years of implementation. Thus, both the political process of pursuing more ambitious targets and the evolution of societies and technologies towards creating a less-carbon-intensive world might be accelerated. However, this raises questions about whether such a vision is compatible with a world with close trade links across countries and continents. Will carbon-price differentials result in the relocation of production, rather than carbon-efﬁciency improvements?

Notes 1 Note that this is different from altruism. 2 There are several striking parallels here: (i) the importance of accurate, publicly available information about the impact of slavery on lives in raising public awareness while there were initial attempts by slavery’s defenders to distort the truth; (ii) the role of discounting the value of slave lives; (iii) gradually moreambitious action against slavery – for example, reﬂected in increasing legal penalties, from £100/head in 1807 to the death penalty for traders in 1827; and (iv) the role of compensation for former slave owners, which had reached £20 million in 1838. For a summary on abolition, see www.history.ac.uk/ihr/Focus/ Slavery/articles/walvin.html. 3 Further factors would have to be considered, including whether the survey is representative, or to what extent a government position can deviate from public opinion. 4 The European Council (EU heads of state) decided in May 2007 – subsequently implemented as EU Directive 2009/29/EC and Decision 406/2009/EC – that European countries would reduce greenhouse gas emissions by 20% relative to 1990 levels by 2020, rising to a 30% reduction relative to 1990 levels ‘provided that other developed countries commit themselves to comparable emissions reductions and economically more advanced developing countries commit themselves to contributing adequately according to their responsibilities and capabilities’ (European Commission 2008b). 5 Marsiliani and Renström (2000) discuss a similar situation for environmental policy.

Conclusion

*

161

6 Bodansky (2004) argues that commitment ‘provides a signal to the market that helps drive changes in private behaviour’. 7 Webster, Paltsev and Reilly (2006) investigate the value of international emissions trading and argue that the calculated beneﬁts stem largely from the burdenredistribution effect. The negative impact of emissions trading on the balance of payments could outweigh the beneﬁts from hedging and identifying least-cost abatement options.

six

A world of different carbon prices

The previous chapter argued that countries and regions might pursue climate policies at different levels of ambition. This would probably include different carbon prices for a transition period, perhaps up to 2020. Now we will assess the implications of different carbon prices in countries connected by international trade. There are important opportunities and difﬁculties to consider. The opportunities lie in the ﬁrst-mover advantages afforded to countries that support the development of low-carbon products and technologies. Just as Denmark’s wind technologies have been established in world markets, new market leaders will emerge for other products. But there is also some concern, that regions with high carbon price levels could risk inducing industry to alter investment, production and closure-decisions about plants, and thus move carbon-intensive production towards countries with lower or no carbon prices. This would have three implications that might have to be addressed by the design of the policy instrument and complementary measures. * Shifting of production in response to a more-stringent emissions cap and associated policy to a country without a stringent cap creates carbon leakage. The freed-up allowances in the ﬁrst country will be used by other sectors that reduce their efforts in emissions 162

A world of different carbon prices

*

163

reduction. Whenever some of the shifted production is not covered by a stringent national cap it will increase emissions in the new country and thus contribute to a global emissions increase. * Companies may limit passing through carbon prices to product prices either to protect current market shares, and hence delay relocation, or in response to output-based allocation of allowances. This dampens carbon-price signals for carbon-intensive commodities and services, reducing the economic incentive to substitute towards lower-carbon alternatives. * Countries that pursue a more-stringent carbon policy might put jobs and tax revenue at risk from relocation. Where such concerns are substantiated for speciﬁc sectors or effectively communicated as a big concern in the political process, they might limit the interest of governments to pursue ambitious climate policies. As all three effects have the same cause and typically occur in parallel, we will for simplicity group them using the term ‘leakage concerns’. Two additional channels for leakage not discussed here are illustrated in Box 6.1. First, carbon leakage can occur where stringent climate policy in some countries reduces fossil fuel demand and thus fossil fuel prices. This could induce demand and thus emissions increases in other countries, thus partially off-setting the emissions reduction. Second, lowcarbon technologies developed under a stringent climate policy regime can be applied also in other regions and reduce emissions in these regions, thus increasing the effectiveness of unilateral climate policy. While vividly discussed in theoretical terms and in policy environments, leakage is of very limited concern and not observed so far owing to moderate carbon prices, in the order of €20/tonne CO2 combined with free allocation linked to the activity level of industry. Low-carbon development is likely to require more stringency in the carbon-pricing scheme, raising the questions whether and where leakage could be issues. Leakage concerns do not equally apply to all economic activities. They only concern speciﬁc carbon-intensive products. For this

164

*

A world of different carbon prices

purpose, section 6.1 introduces a carbon cost screen to assess the industry sub-sectors for which carbon pricing signiﬁcantly increases production costs. Cost differentials are not, however, the only determinants of production and trade; other aspects that impact trade ﬂow and industry’s location decisions include the ability of companies to pass through costs, the barriers and drivers for international trade, including customer relationships, product differentiation, transportation costs and risks in investing abroad such as exchange rate volatility. These are discussed in section 6.2. Section 6.3 discusses the strategic objectives of corporate decisions that may extend beyond short-term proﬁt maximisation. Section 6.4 introduces a new dimension. It draws attention to the fact that carbon emissions tend to be focused on the manufacturing of upstream products (e.g., clinker for cement) that are then used as inputs for subsequent production steps (e.g., cement and concrete). A value-chain analysis illustrates the role of carbon pricing in various stages of the value chain. Building on the ﬁndings from the analysis that leakage is only of concern for a very small set of products, section 6.5 asks what policies could be used to address leakage concerns for them. It discusses implementation of effective carbon pricing for internationally traded products in a world with differing carbon prices.

6.1 Screening for high carbon costs From the perspective of the total economy, the potential cost increase from CO2 pricing corresponds to the volume of emissions. The EU ETS covers about 2 billion tonnes of carbon emissions annually. Thus, at an allowance price of around €20/tonne CO2 the costs incurred if all allowances are sold in auctions is €40 billion per annum. This cost is small relative to the total value added across EU Member States – with a GDP of €11 trillion (2005). Carbon pricing thus only increases average costs by 0.4 per cent, a cost change that is swamped by all the other cost differentials between countries.

Screening for high carbon costs

*

165

b o x 6 . 1 Leakage channels The Intergovernmental Panel on Climate Change (IPCC) deﬁnes leakage broadly as ‘the emission increase abroad caused by unilateral climate policy measures at home’ (IPCC 2007) and identiﬁes three channels (Figure 6.1). Macroeconomic models estimate that, if unmitigated, they could together increase emissions by 0.05 to 0.2 ton CO2 in regions not covered by climate policy for every ton of CO2 emissions reduced in countries with climate policy (IPCC 2007). Fossil fuel channel (Felder and Rutherford 1993). If demand for oil decreases owing to climate policy, then this could reduce oil prices. In response, demand for oil, and therefore emissions, could increase in other regions. If demand for coal decreases owing to climate policy then this has limited impact on mid-term coal prices owing to large coal resources. Uncertainty about coal demand driven by climate policy could limit investment in coal extraction and trigger coal price spikes that reduce coal demand. Demand for natural gas is initially expected to increase with climate policy owing to its low carbon content. Resulting price increases could reduce energy demand and emissions in other regions, but could also drive other regions towards more coal use, thus increasing emissions.

Climate policy

Fossil fuel channel • Oil (+) • Coal (0) • Gas (?)

Country A (with cap)

(?/+)

Technology (–) Country B

channel (+)

Production channel Direct emissions reduction

(?/+) Potential leakage

Figure 6.1 Potential channels for leakage

166

*

A world of different carbon prices

b o x 6 . 1 continued Potential leakage from the fossil fuel channel usually has the biggest leakage impact in macroeconomic models (Sijm, Kuik et al. 2004 and Barnett, Dessai and Webber 2004). The quantiﬁcation requires strongassumptions – for example, about strategic and institutional responses of oil-exporting countries to changing oil demand (Burniaux and Martins 2000). The potential impact is hidden, and almost impossible to address with policy instruments. This might explain why it receives less attention in policy discussions. Global co-operation on climate policy is the most suitable policy response and already pursued for several other reasons. Technology channel (Grubb, Chauis and Ha-Duong 1995). Low-carbon and energy-efﬁciency technologies developed and commercialised in countries with ambitious climate policies are likely to diffuse to other countries and contribute to emissions reduction in these countries. Governments are discussing how to support this process with international technology co-operation. Production channel. Carbon-intensive production might be relocated to countries with lower carbon prices. This would result in emissions being relocated rather than reduced. Relocation of production in response to climate policy in one country might not necessarily increase global carbon emissions. Emissions could decline because of improved efﬁciency of the new facility or a reduction of the overall transport volume for inputs and products. This ignores, however, an important aspect of the institutional setup. Figure 6.2 illustrates that reduced net exports results in increased demand from other countries. This increased demand might be partially satisﬁed by other countries with stringent emissions caps. In this case it would not alter global emissions. The additional demand will, however, at least partially, be met through increased production in countries without a binding absolute cap. In these countries the additional production results in additional emissions. As long as some countries remain without a stringent absolute emissions cap, most relocation of production will increase global emissions.

Screening for high carbon costs

*

167

b o x 6 . 1 continued

Tighter cap

Country A (with cap)

Country B (with cap)

Commodity Reduced net-export

Increased demand

Increased demand

Direct emissions reduction Country C (no/intensity cap)

Leakage

Figure 6.2 The economics of leakage along the production channel

Furthermore, if all the carbon allowances were auctioned by respective governments the money would be returned to the economy via tax reductions or support for low-carbon technologies, thus virtually eliminating a cost impact on the European economy. In line with this intuition, econometric analysis of the impact of environmental taxes imposed in some European countries over the last two decades shows no negative impact of carbon taxes on industrial activity (Andersen and Ekins 2009). Similarly, as discussed in a different context in Chapter 2, countries with high energy tax levels are economically successful. This again suggests that asymmetric high carbon prices are unlikely to create

168

*

A world of different carbon prices

competitiveness concerns for the overall economy. However, most of the carbon tax and energy tax schemes provide for some exemptions for very carbon-intensive products. Therefore, the speciﬁc effect on these products has to be analysed separately – after all, the emissionstrading schemes aim to target in particular these carbon-intensive products and so should include them. As industrial activities are not homogeneous, carbon costs have a disproportionate impact on speciﬁc industrial activities. These are the activities that might be relocated, and where emissions might leak to other parts of the world rather than being reduced in response to a carbon-price signal (see the related discussion on the ‘pollution haven hypothesis’, for example: Copeland and Taylor 1994). To identify which activities might be signiﬁcantly affected, we ﬁrst analyse the potential cost impact of carbon pricing. Then the international trade of associated products is evaluated as it is a major determinant in the ability to pass through carbon prices to product prices. A simpliﬁed metric is used to characterise the trade exposure of manufacturing activities which allows the screening of different industrial activities. Figure 6.3 gives an overview of the ‘potential value at stake’ across the main industrial sectors in the UK with potentially signiﬁcant carbon cost impacts. Although the analysis is based on UK data it is more widely applicable because the UK has a mix of industrial activities that is similar to the European average (as conﬁrmed by a comparison of the share of different manufacturing activities in the UK against the European average: Hourcade et al. 2007). To allow a comparison across sectors, cost increase from the carbon pricing is compared to the value added of a sector. The gross value added is the sum of wages, return and depreciation on capital, taxes and proﬁts in a sector. The ratio between cost increase and value added gives the metric used in the remaining part of the section – potential value at stake. This is a measure of how much of the value added created in a sector would be lost if the sector faced the full carbon cost but did not pass it through to the product price.

Screening for high carbon costs

Potential net /maximum value at stake

Net value at stake (100% free allocation; only electricity price impact)

Basic metals (including iron and steel)

15

Pulp, paper, printing and publishing Fabricated metal

10

Cement, lime and plaster Food and tobacco

5

0 10

Coke oven, refined petroleum and nuclear fuel

Plastic and rubber Non-ferrous metals

Wood

0

169

Maximun value at stake (no free allocation)

Electricity

20

*

Glass and ceramics

Chemicals Transport equipment

Electrical and optical equipment Textiles Machinery

20 30 40 UK trade intensity from non-EU countries (percentage)

50

Figure 6.3 Value at stake for main manufacturing sectors v. UK trade intensity from outside EU at €20/tonne CO2. Source: Based on Hourcade, Demailly, Neuhoff and Sato 2007.

The lower end of the bars for the different sectors show the ‘indirect’ impact on the costs incurred from the rise in electricity prices. In the UK power system, combined-cycle gas turbines are usually the marginal generation units, with emissions slightly below 0.5 tonnes CO2 per MWh electricity produced. The electricity producers thus have to submit 0.5 allowances per MWh electricity and will only produce if the revenue from selling electricity exceeds the value of these allowances. Assuming carbon prices of €20/tonne CO2 the power price thus rises by about €10/MWh.1 Carbon pricing also creates costs for direct emissions from energy use by industrial processes. The length of bars in Figure 6.3 depicts the range of cost increase if industries bought all their CO2 allowances. Thus, the top end of the bars shows the total potential value at stake for a sector if all allowances have to be bought and if the sector is not able to pass on carbon costs to product prices. If, however, allowances are allocated at no cost, then sectors will not face cost increases from

170

*

A world of different carbon prices

direct emissions and the potential value at stake only relates to the electricity price increase depicted by the lower end of the bar. Producers insulated from trade with regions with low or zero carbon prices are able to pass through the (opportunity) cost increase from carbon to product prices. For example, electricity producers in liberalised markets proﬁt from the scheme with free-allowance allocation. For products that are actively traded in global markets it is more difﬁcult to predict the outcome. Companies could increase prices to reﬂect (opportunity) costs of carbon, which protects proﬁt margins in the case of auctioning and may increase proﬁts in the case of free-allowance allocation. This strategy may, however, risk a loss of market share from cheaper imports gradually replacing domestic production. Alternatively, companies can maintain prices and lose proﬁt margins if allowances are auctioned. The reality is likely to be somewhere in between depending on the speciﬁc characteristics of a sector, which will be discussed in more detail in the next section. The x-axis in Figure 6.3 depicts the level of international trade intensity for the products of a sector, which is also referred to as ‘trade exposure’. International trade exposure (E) is obtained by dividing the trade volume (T) by the market size (S). The total trade volume is calculated as the sum of exports and imports; the total market size equals the sum of domestic demand and exports or equivalently the sum of domestic production and imports.2 Trade intensity is an imperfect indicator of the ability of sectors to pass on carbon costs to product prices. This is because international trade exposure is a dynamic parameter that depends on, and can change with, the industry structure regionally and internationally. For example, international trade exposure of electricity is zero in the UK because there are no power lines to countries outside the EU ETS (and this is unlikely to change in the near future). International trade exposure is currently also very low for cement, but this could obviously change if there were a sufﬁciently strong ﬁnancial incentive. Thus, current trade intensity is

Screening for high carbon costs

Potential net and maximum value at stake (percentage)

35

30

Manufacture of lime (SIC 26.52); gross value added £26 million (average 1997–1999) MVAS 26%

25

Manufacture of cement (SIC 26.51); gross value added £409 million (2004)

*

171

Total sector gross value added in 2004 £3,359 million

20

15

Manufacture of plaster products for construction purposes

10

Manufacture of ready-mixed concrete

5

Man. of other articles of concrete, plaster and cement

Manufacture of other non-metallic mineral products Production of abrasive products Cutting, shaping and finishing of stone

0 0

20

40 60 80 Non-EU trade intensity (percentage)

100

Figure 6.4 Value at stake for construction materials v. UK trade intensity from outside EU at €20/tonne CO2. Source: Based on Hourcade, Demailly, Neuhoff and Sato 2007.

only an initial static indicator and does not provide a comprehensive assessment of potential leakage for a sector that could be dynamic. Figure 6.3 depicts a two-digit sector representation according to Standard Industry Classiﬁcation codes. This represents an aggregation of many different activities that are merged under the sector classiﬁcations. This can obscure speciﬁc activities that might be particularly affected, and may raise concerns for activities that individually have low values at stake. For a more detailed analysis, we therefore move from the two-digit to the four-digit sector representation. To illustrate this disaggregation, Figure 6.4 gives the different fourdigit sector activities that form the two-digit sector category construction materials. It shows that only manufacturers of cement and lime have high values at stake owing to carbon prices. The total value added of these two

172

*

A world of different carbon prices

60

Casting of iron Preparation of yarn

Lime Allocation-dependent (direct) CO2 costs/ gross value added

40

Copper Other textile weaving Flat glass Other inorganic Household paper Veneer sheets basic chemicals Non-wovens Retreading/ Industrial gases rebuilding tyres Coke oven Fertilisers and Nitrogen Rubber tyres and Malt tubes manufacturing Starches and starch Hollow glass products Finishing of textiles Pulp and Refined petroleum paper

20

10 4 2 0 0.0

Basic iron and steel

30

Aluminium

Electricity (indirect) CO2 costs/ gross value added

Cement

Potential maximum value at stake and net value at stake (percentage)

50

0.2

0.4

0.6 UK GDP (percentage)

0.8

1.0

Figure 6.5 Industrial activities with the highest cost increase from carbon pricing and their contribution to UK GDP (assumed carbon price increase €20/tonne CO2, electricity price increase €10/MWh). Source: Based on Hourcade, Demailly, Neuhoff and Sato 2007.

activities is €0.6 billion, compared with more complex activities like manufacturing concrete products for the construction process (€1.8 billion) or the overall sector with a value added of €5 billion. Figure 6.5 summarises the analysis of 164 industrial activities of the economy. It depicts the twenty-four sectors with the highest value at stake from electricity cost increases and direct carbon emissions. Activities are sorted by their total value at stake, using the same metric on the y-axis as in Figure 6.3 and Figure 6.4. The horizontal axis depicts the contribution of the sectors to the GDP of the UK. The contribution of each sector to carbon emissions is represented in the graph by the area covered with the rectangle that represents the activity. The area is the product of the value created on the x-axis times the emissions intensity per unit of value created on the y-axis. The twenty-four sectors with the highest value at stake contribute about 1 per cent of the UK GDP, and 13 per cent of UK carbon emissions.

International cost differences

*

173

Lime, cement, basic iron and steel and reﬁneries are the sectors with the highest value at stake from direct carbon emissions. They are characterised by energy- and hence pollution-intensive production, relatively labour intensive, and contribute most to carbon emissions among the activities that exhibit leakage concerns. Aluminium, fertilisers and some inorganic chemicals, including chlorine production, have the highest value at stake from electricity price increases. A high potential value at stake does not, however, necessarily lead to leakage. For example, chlorine is a very hazardous substance and consequently can only be transported at high cost between countries. The next section will explore in more detail the potential leakage for these industrial activities. The reassuring aspect from the ﬁrst step of the analysis is that the value at stake in most industrial activities is relatively low, and for the UK leakage is of little concern across 99 per cent of GDP. This result is not unique to the UK – even in economies with a bigger focus on manufacturing, like Germany, less than 2 per cent of GDP is associated with activities that face signiﬁcant cost increases with carbon prices (Graichen et al. 2008).

6.2 Do international cost differences matter? Dimensions of trade In this section we ﬁrst discuss factors beyond production costs that determine trade ﬂows. Then production costs are differentiated between variable and ﬁxed costs, better to understand the implications of high up-front investment costs for the production of commodities.

Components of import costs It is frequently argued that in a competitive global market the producer with the lowest costs will capture the market and will replace production of all other producers. If government regulation increases

174

*

A world of different carbon prices

Premium for local goods

Trade cost

Home cost

• Customer trust • Tailored product attributes • Quick response to new specifications • Political weight of labour impact

Variable cost

• Transport cost • Port facilities • Storage • Interruption risk • Tariff risk • Exchange rate risk Carbon costs

Foreign cost

Import costs

Fixed cost

Figure 6.6 Determining premium for domestic products and trade-related costs for imports

the costs of production in one country, this creates a competitive disadvantage, and as a result the production will be shifted abroad. In contrast to such clear statements, for several commodities production has been maintained and there has even been additional investment observed in regions that exhibit signiﬁcantly higher fuel costs, wages or taxes. Figure 6.6 depicts factors that allow local producers to charge a premium for their goods and various traderelated costs that might explain this phenomenon. The premium that local products can achieve over imported goods can be explained by several factors. Customers have pre-existing relationships with local producers; proximity helps in building and maintaining trust in business relationships, which is particularly important in complex production processes like car manufacturing. The cost of a basic input such as steel is low relative to the risk for the production should delivery or speciﬁcations fail when consumer requirements for product speciﬁcations are changing over time. Local producers can respond faster than importers, who face delays in communication and longer transport chains. Finally, the level of political inﬂuence and attention of companies grows with the number

International cost differences

*

175

of employees included in the supply chain. Locally sourced inputs can also increase public support during difﬁcult times and have an advantage when seeking investment, R&D support and planning approvals from government. Also, the trade-related costs are the sum of many factors. The biggest component is generally the cost of transport, which can vary signiﬁcantly with the price of oil and the demand for shipping bulk commodities. It might rise if climate policy is applied to shipping and imposes costs for carbon emissions. Trade costs also include the cost of developing tailored port facilities, potentially increased risk of interruption of foreign production and additional storage costs to reduce their impact. Export and import tariffs and other trade barriers add to the costs of foreign products.

The interaction between import costs and carbon costs The combination of the trade-related costs and the premium that can be charged on domestic goods will be termed ‘import costs’ in this section. The interaction of import costs with carbon prices is discussed below in three steps: carbon prices are assumed to be the only difference in global production costs, other cost differences are subsequently considered and ﬁnally the inﬂuence of volatile global demand and prices is factored in. (i) Commodity prices and proﬁt margins have always been unpredictably volatile and therefore capital-intensive commodity industries became accustomed to recovering their ﬁxed costs in years of high margins and operating at low margins in other years. This averaged long-term view on ﬁxed costs is the basis of the initial discussion. The left side of Figure 6.7 illustrates that if cost structures are identical and carbon costs faced by home producers are less than import costs then home producers can pass on the full carbon costs to product prices without any trade impacts.

176

Import costs exceed carbon cost

*

A world of different carbon prices

Import costs less than carbon cost

Carbon cost exceed fixed + import cost Carbon cost

Import costs

Home cost

Foreign cost

Fixed cost

No concern

Variable cost Concern about reinvestment/ new investment

Carbon cost jeopardises operation

Figure 6.7 Can local premium and trade costs (import costs) compensate for asymmetric carbon costs?

The central part of the ﬁgure illustrates that if carbon costs exceed import costs home producers can decide to set the domestic price level at the import price level, which is determined by the sum of foreign production costs and import costs. Although it is proﬁtable to continue to produce, the company may not recover all the costs of its initial investment and will not invest in the future. Alternatively, producers can increase their price level above the foreign price level, including import costs, at the risk of gradually losing market share. The right side of Figure 6.7 shows that with carbon costs exceeding the sum of import costs and ﬁxed production costs it is no longer proﬁtable to continue operating. An investor may decide to build new production facilities abroad and replace home production. In this case the new production facilities abroad have lower total costs than the variable costs of the home producer. (ii) In reality, cost structures differ across countries. For example, Brazil has local access to cheap iron ore and coal for steel production and wages and tax levels are lower than in developed

International cost differences

*

177

countries. This advantage is often compensated where developed countries offer better infrastructure and a robust regulatory regime and thus have lower capital costs for investment. In the case of Brazilian steel, the excellent resource situation paired with improving institutional settings was expected to offer a signiﬁcant advantage over production (for example in Europe). Such differences in the cost structure could induce a gradual shift of production towards Brazil. Additional carbon costs if only imposed on European producers could accelerate this shift. Equally, the already intended relocation of a production facility could be blamed on carbon pricing. It is inherently difﬁcult to disentangle pre-existing cost differences and industrial trends from the impact of carbon pricing. (iii) What happens to a carbon price differential at times of low global demand and margins, when commodity prices can drop almost to variable costs? Like foreign producers, domestic producers will not be able to recover much of their ﬁxed costs. Domestic producers have to compete with their variable costs, including carbon prices, against the foreign variable costs and import costs. In this case, if carbon costs exceed import costs, then home production would run at a loss and short-term proﬁt maximisation would require stopping production. At times of global excess supply, the price buffer created by ﬁxed-cost recovery is absent, and in this case short-term responses to carbonprice differentials can be more immediate. In summary, import costs comprising both trade-related costs and premiums for domestic products can offer some protection for domestic production from the impact of carbon costs. For existing facilities, ﬁxed sunk investment costs are a further important factor in delaying leakage. The ‘protection’ will, however, be lower where there are preexisting cost differentials. The imminence of leakage might increase at times when excess production capacities squeeze margins and drive prices towards short-term marginal production costs.

178

*

A world of different carbon prices

6.3 Corporate strategy: the longer-term view Companies and their managers do not only focus on short-term proﬁt optimisation, but also pursue other objectives that are reﬂected in longer-term strategies. An important objective for many companies is their market share, which is often an important consideration for the reputation and remuneration of management. Protection of market share can reduce leakage concerns. Companies beneﬁt from the ﬂexibility to adjust their production volume to evolving demand. They may continue production at a site even where it is not proﬁtable in the short term in order to retain the option for future production, and hence the ﬂexibility to respond to future demand and customer requests. Furthermore, the costs of closing production facilities can be signiﬁcant, including compensation for employees, clean-up of sites and loss of political goodwill and management reputation. The production of a carbon-intensive commodity is frequently only one of the activities of a company. There may be links with the production of other less-carbon-intensive products and the commodity might be an input for subsequent production processes operated by the company or part of a range of products offered by a company. Companies are likely to be reluctant to close their own production facilities and buy from companies with whom they compete in other markets, and might prefer to pay a premium to continue production with their own facilities. The impact of carbon-price differentials on corporate decisions is strongly inﬂuenced by product attributes and the strategic perspectives of management. The challenge is that multiple storylines can be used to explain strategic decisions of management in the past and to predict future strategic choices. This makes it difﬁcult to test whether a storyline that is presented by a company in the political process reﬂects the corporate strategy or aims to inﬂuence policies in their

Corporate strategy

80 60 40 20 0 20

0

Import

100

179

Higher demand elasticity Higher trade elasticity

No trade distortions

Exports European Union demand

Base case

120 Domestic production

Basic oxygen furnace steel (mt.)

No carbon price

*

20 20 20 50 /t CO2 assuming full cost pass through Lost exports from foreign carbon pricing

Figure 6.8 Impact of carbon pricing on demand and trade ﬂows for EU3. Source: Based on Demailly and Quirion 2006.

favour. This complicates the use of such storylines as robust evidence for political processes. Some quantiﬁcation of how trade ﬂows are inﬂuenced by price differentials is, in principle, possible. Such estimations are calibrated using historic responses of trade ﬂows to price differentials and reﬂect the various dimensions of import costs (Gallaway McDaniel and Rivera 2003). It is sometimes argued that the rapid globalisation of producers in the steel, cement and chemical industries reduces some of the import costs and does therefore result in stronger responses of trade ﬂows to price changes. However, at the same time, uncertainties are emerging that might counter this effect – for example, increasing transport costs and other policy responses to climate change (Kuik and Gerlagh 2003). Figure 6.8 depicts estimates for the impact of carbon-price differentials in the case of steel. Historic trade elasticities are used to estimate the impact of the cost differentials on import and export ﬂows (Demailly and Quirion 2008). The ﬁrst column shows the current production and trade volumes of steel based on the carbon-intensive

180

*

A world of different carbon prices

blast-oxygen furnace process. The second column predicts the midterm impact of a €20/tonne CO2 carbon price on European production, which falls because both exports and demand decrease while imports increase. The third column uses higher demand and trade elasticities so as to illustrate the sensitivity of results to assumptions on input parameters. Sometimes the impact of carbon pricing in facilitating a shift from carbon-intensive to lower-carbon commodities is confused with the undesired impact of carbon leakage. To contrast both effects, the last two columns depict the modelling outcome assuming carbon pricing is applied at the global scale. The higher product prices feed fully through to demand and result in a reduction of demand. Thus, domestic production and imports decline. As foreign demand declines, exports decline (but to a lesser extent than in the case of unilateral implementation of carbon pricing). Production activities with the highest values at stake are characterised by large upfront investment costs. They therefore require management to take a long-term strategic view that reﬂects objectives like market-share protection and retaining the real option value of physical production facilities. This suggests that leakage concerns are very much a question of time frames. Private-sector expectations about future carbon-price differentials, trade costs and premiums for locally produced products will determine the level of leakage in the speciﬁc sub-sectors with high potential value at stake. But privatesector expectations are difﬁcult to quantify and objectively to discuss where policy makers have limited, and possibly biased, information about the situation of a company or sector. Discussions about leakage concerns should ideally be based on metrics that can be quantiﬁed in an unbiased manner. This would help avoid accusations that environmental leakage concerns are used as a smokescreen to pursue industrial protectionism. In our search for a better understanding of leakage we therefore return to fundamental economic analysis.

The industry value chain

*

181

6.4 The industry value chain: leakage versus substitution effect Industrial products are often linked: outputs of one industrial activity are traded, consumed and become inputs for other production activities. For example, clinker is used to make cement, which in turn is the basis for concrete products that become parts of buildings, which then offer housing and commercial services. Figure 6.9 illustrates this value chain. As discussed in Chapter 2, carbon pricing creates incentives for efﬁciency improvements in clinker production and drives substitution effects in the subsequent stages of the value chain. Where leakage is a concern, clinker producers will either refrain from reﬂecting carbon prices in clinker prices or clinker will be imported. Consequently, users of clinker will not face the full carbonprice signal and cement producers will have a lower incentive to make CO2 performance improvements, such as reducing the clinker content of cement. In the next stage of the value chain the lower concrete price delays some substitution of concrete products with other materials or through more careful use in the building sector.

Substitution

Lower clinker content

Leakage

Leaner structures

Clinker

Other building materials

Efficiency

Clinker imports

Cement

Concrete

Building

Cement imports

Figure 6.9 Illustration of value chain with potentials for efﬁciency improvements, substitution and leakage

182

*

A world of different carbon prices

Illustrative for UK

Cost increase relative to value added ( 20/t CO2 percentage

60 50

Cement

Clinker

70

Concrete products (concrete products for construction; mixed concrete, etc.)

Total cost increase

40 30

Cost increase passed on from first production stage (clinker)

20

Cost increase from higher electricity prices

10 0 0

500

1,000 1,500 2,000 2,500 Cumulative gross value added ( m)

3,000

3,500

Figure 6.10 Value chain of concrete production

We quantify the potential value at stake along the value chain in order better to understand the risk of leakage from forgone substitution effects. In line with the high carbon intensity of the clinker production, the cost increase for this production stage is above 60 per cent relative to the value added (Figure 6.10). The subsequent mixing of cement from clinker only requires electricity for milling and, as a result, is not energy or carbon intensive. If, however, the production of clinker and cement are assessed as a joint production activity, then the cost increase relative to value added (potential value at stake) again exceeds 30 per cent. Moving further down the value chain the carbon intensity of the various types of concrete production is relatively low. Even if production of concrete is jointly assessed with clinker and cement production, the potential value at stake is below 10 per cent. Given the transport costs for concrete parts, it is unlikely that this would result in any relocation of production. The analysis conﬁrms that with increasing aggregation across industrial activities the potential value at stake falls. This might

The industry value chain

*

183

hide potentially exposed activities. Even where production activities are currently integrated, new ﬁnancial incentives created by carbon pricing might result in separation of two activities. Currently, most clinker and cement is produced by the same companies at one location. If clinker production faces the full carbon costs it might well be relocated to areas with lower or no carbon prices. The analysis also indicates how far down the value chain leakage concerns can persist. The potential value at stake for clinker production indicates strong leakage concerns. If it were possible to prevent such leakage, perhaps by administrative measures that require on-site production of clinker (but would be difﬁcult to justify on World Trade Organization grounds), then cement producers could continue to produce their own clinker despite the higher carbon price. However, the potential value at stake of the joint clinker and cement production would still be rather high, and leakage concerns might persist at this level. Any concerns vanish once cement is integrated into concrete structures – at this stage the potential value at stake falls below 10 per cent and would be dwarfed by the transport costs for concrete structures. The situation for steel production is in many ways similar. Figure 6.11 illustrates the consecutive steps involved in steel production. In the basic oxygen furnace iron ore is reduced to semi-ﬁnished steel. This is subsequently hot rolled and further reﬁned into speciﬁc iron and steel products. Most carbon emissions result from the ﬁrst stage and cost increase relative to value added is highest at this stage. Again, the potential value at stake falls as costs are spread across subsequent production steps. Even if the cost increase from the carbon-intensive basic oxygen furnace is spread across all the production steps, the cost increase for ﬁnished products relative to value added is still well above 10 per cent. Thus, the situation differs from the cement value chain. The potential value at stake remains high but the international transport costs are comparatively low. This suggests a need to address leakage

184

*

A world of different carbon prices

Illustrative for UK 50

Cost increase relative to value added ( 20/t CO2 percentage

Semi finished

Hot rolled

Iron and steel

40

Total cost increase from CO2 pricing

30

Cost increase from passed-on CO2 pricing of first production stage only

20

Total cost increase from higher electricity prices

10

0 0

500

1,000 1,500 2,000 2,500 Cumulative gross value added ( m)

3,000

Figure 6.11 Value chain of steel production using basic oxygen furnace process4

concerns for the basic steel component of ﬁnished steel. Again this raises the question of whether parts of the value chain can be relocated individually. As the production of semi-ﬁnished steel is carbon intensive, it is the candidate for such relocation. In principle, few countries might be interested in hosting energy-intensive and environmentally unfriendly basic oxygen furnaces. However, one potential candidate would be Brazil, as it has local access to good iron ore and coal resources. Relocation would offer the additional beneﬁt of reducing transport demand for coal and iron ore. On the other hand, integrated steel works combine the basic oxygen furnace with hot rolling and can avoid energy consumption in repeated heating cycles. This integration reduces the incentive for relocation. Vice versa, if relocation incentives from high carbon costs are strong, both the plants might in theory be relocated. The value chain illustrates that carbon-intensive production constitutes an even smaller share of economic activity than suggested

Policy options to address leakage

*

185

by the four-digit standard industry classiﬁcation. This increases the environmental concern associated with leakage – emissions might migrate outside the geographical area covered by an emissions cap more easily as they are associated with less economic activity and therefore also investment cost. The analysis also conﬁrms the importance of the full carbon-price signal. The more complex the production process the more difﬁcult would it be for governments to administer the efﬁcient use of energy- and carbon-intensive materials at different stages of the value chain. A full carbon-price signal creates incentives for carbon-efﬁcient operational, investment and innovation decisions along the different stages of the value chain.

6.5 Policy options to address leakage The preceding discussion illustrated that leakage is only of concern for a few deﬁned sub-sectors and only if signiﬁcant carbon-price differentials are expected to be maintained for many years (e.g., up to 2020). Some of the aspects that will have to be considered in the evaluation of a speciﬁc sector are, at least currently, only discussed qualitatively. It might be difﬁcult to ﬁnd robust quantitative descriptions of management strategies, product differentiation or expected capacity expansion. However, once a concern has been identiﬁed for a speciﬁc sector it is desirable to return to a quantitative approach when tailoring the scale of any instrument to address leakage. Figure 6.12 uses a simpliﬁed illustration of the impact of asymmetric carbon pricing. The cost of buying allowances in an auction increases the production costs in one region (dark bar on top), which could result in relocation of production to the region with lower production costs (right column of each pair). Three basic options are available to prevent such asymmetric cost impacts, and thus to address leakage concern. On the left side of Figure 6.12, free-allowance allocation or direct ﬁnancial subsidies (state aid) compensate for the carbon cost increase. They limit or

186

*

A world of different carbon prices

CO2 costs

Cost

Cost

Cost

Price level

Allocation conditional on operation and state aid

Export taxes

Government-led sectoral agreement

Border adjustment

Inherent challenge • Little substitution to lowcarbon products/services • Distorts investment • Bureaucratic constraints for innovation • Risk of lock-in

• Has to be aligned with international climate engagement • Requires at least informal international co-operation

• Requires strong policies of developing countries • Risk of low common denominator

Figure 6.12 Policy options to address leakage concerns

prevent the cost increase, and all producers compete at the cost level similar to a world without carbon pricing. In the middle of Figure 6.12, export taxes implemented by countries with lower carbon prices, or some form of border adjustment for higher carbon prices, could adjust carbon cost differential if products are traded between countries with different carbon prices, and so recreate a level playing ﬁeld. The right side of Figure 6.12 illustrates how government-led sectoral agreements could create the same carbon price for all competing companies. This is similar to a global carbon price, but it might be only focused on speciﬁc sectors. All three options ensure all companies face comparable carbon costs and thus create a level playing ﬁeld with regard to the carbon price. Which instrument is most suitable to address leakage depends on the speciﬁc sector. To discuss the possible options, the characterisation of leakage concerns introduced in Figure 6.12 is reﬁned. Concerns about reinvestment and new investment are separately

Policy options to address leakage

*

187

assessed and the case where operation is jeopardised is differentiated so as separately to assess concern about reduced production and about closure of plants. New investment. For products with increasing production volume in a scenario with asymmetric climate policies, the plant investment could be shifted to regions with lower carbon prices in a scenario of asymmetric carbon prices. This new investment decision is probably most sensitive to carbon-price differentials. However, with stringent climate policy few carbon-intensive commodities will be required at larger volume and little new investment is required for their production in developed countries. Reinvestment. Even with constant or declining production volumes, plants require ongoing investment, maintenance and upgrading to match evolving product and environmental requirements. If these investments are not pursued, then it is likely that production of a plant will decline, perhaps to eventual closure. This is of concern where reinvestment requirements are high relative to the cost of new (greenﬁeld) investments and where there are limited technical links with related manufacturing activities. Reduced production with potential closure. If carbon costs of production are large relative to annual ﬁxed costs of an installation, then companies gradually reduce production volumes where they face increasing import volumes. Closure of plant. If carbon costs of production and ﬁxed annual costs are large, then producers typically face the decision of full production or closure of an installation rather than gradual adjustment of output. All four cases could result in some movement of carbon-intensive production from installations covered by an emissions cap to other regions. Thus, emissions reduction induced by the carbon-pricing scheme would not be genuine reductions but would be off-set by leakage of emissions. We will subsequently refer to these four categories as leakage channels.

188

*

A world of different carbon prices

State aid or free-allowance allocation to subsidise carbon-intensive production Governments can shield carbon-intensive production from the full carbon price. Free-allowance allocation can be used to subsidise production or investment in sectors with direct carbon emissions (Böhringer and Lange 2005). Direct public subsidies can also support investments and reinvestments in sectors that are facing either high costs from carbon emissions or signiﬁcant cost increases from electricity price increases. In Europe, such subsidies by Member States are called state aid, and are regulated across Europe so as to limit distortions of competition among Member States. The typical approval process by the Commission for State Aid granted by Member States offers the opportunity for a less-political decision process that could allow for better targeting subsidies than free-allowance allocation to installations with demonstrable leakage concern. For steel production, the ﬁxed operating costs are high. Therefore producers have to decide whether or not to operate an installation. They may not have the choice of gradually reduced output. Thus, free-allowance allocation can be linked to total production capacity with some benchmark value. Allocation has to be conditional on continued operation to create the necessary incentives but does not require a link to the precise production volume. The implementation of similar provisions in the ﬁrst two national allocation plans suggests that companies can choose the timing of closure of installations so as to retain continued allocation for almost two years after closure. This reduces the incentive for continued operation of plants. Alternatively, state aid could be used to support reinvestment decisions of companies. Further analysis is required to assess whether the reinvestment volume that limits the amount of state aid that can be allocated is sufﬁciently large to have a material impact on addressing leakage, and to what extent such reinvestment can be separated from operational costs that would not be covered by state aid.

Policy options to address leakage

*

189

The situation differs for clinker, the carbon-intensive input material for cement. For clinker the relative costs of CO2 are high. Therefore, allocation based on a capacity benchmark does not balance incentives. Free allocation proportional to some recent production volume of clinker would be required to address leakage concerns in this sector with domestic subsidies. This approach will, however, dampen the product price increase of clinker and throughout the cement value chain. Hence it dampens the incentives to reduce the clinker content of cement, the amount of cement in concrete and the amount of concrete in buildings. Thus, free-allowance allocation proportional to recent production volumes forgoes much of the desired incentive for substitution towards lower carbon materials for construction. The allocation of allowances could be made proportional to the output of cement production rather than output of clinker production in order to retain the incentive to reduce the amount of clinker in cement. This would not address emissions leakage as cement producers can import the carbon-intensive intermediate product (Demailly and Quirion 2006). Constraints to prevent this might be difﬁcult to justify under World Trade Organization (WTO) rules.5 The allocation of allowances based on benchmarks requires a precise deﬁnition of the production process and qualifying products. If, for example, the speciﬁcations of clinker are not clearly deﬁned, it could create incentives for producers to add clay to clinker production. This would increase the volume of clinker and thus the amount of free allowances that are allocated proportional to the volume of clinker produced. If, instead, the processes and product are very narrowly deﬁned, then the ﬂexibility of operation, investment and exploration of substitutes is reduced, thus increasing overall costs of emissions reduction. Finally, steam reformers are part of large chemical installations. Existing steam reformers are therefore likely to continue their operation as part of the overall facility. State aid could support new investment and large-scale reinvestment, should a detailed analysis demonstrate that they are at risk of relocation in a world of asymmetric carbon prices.

190

*

A world of different carbon prices

t a b l e 6 . 1 Instruments to address leakage for production of different commodities Benchmark on capacity Steel production (basic oxygen furnace) Clinker (for cement) Steam reformers (chemicals)

Benchmark on production

X

State aid ?

X ?

X

Table 6.1 summarises which approach seems most suitable for different sectors. All three policies have unintended negative side effects. They create administrative processes that link allocation of subsidies to carbon emissions of a plant. This will undoubtedly create an early action problem – where agents expect their actions today can allow them to capture future beneﬁts from public subsidy rather than from emissions reduction. This will distract and possibly distort investment and operational choices (Ellerman 2006; Matthes, Graichen and Repenning 2005). Subsidies to carbon-intensive production also reduce product prices and thus the economic incentive to shift towards lower carbon production technologies and/or product substitution. To compensate for this, the carbon price increases and additional mitigation efforts are pursued in other sectors. This deviation from the ﬁrst-best distribution of mitigation efforts increases the costs of climate policy. From an international perspective, the continued use of subsidies, and particularly the use of free-allowance allocation, might lead to a lock-in to inefﬁcient policies. Countries decide sequentially on their allocation plans for allowances or exemptions for carbon taxes. Once all countries have implemented such provisions, extensive coordination is required to phase out subsidies. Perhaps early international co-operation can ensure ‘sunset clauses’ are in place to facilitate the move to an efﬁcient carbon-pricing scheme in the long run.

Policy options to address leakage

*

191

Subsidies can address leakage, but at a high cost. They limit incentives for emissions reduction and innovation in the sector where they are applied. Therefore, the cost of delivering emissions reduction has to be borne by other sectors of the economy. It reduces the incentive for industry to develop low-cost options for emissions reduction that can be replicated in developed countries. The debates at state and federal levels in the USA, Australia and Japan illustrate that allocation and leakage are an intrinsic challenge for the implementation of effective carbon-pricing schemes. The ﬁrst response of industry representatives is to argue for free allowance allocation to address leakage concerns. From the perspective of company owners, any free-allowance allocation guarantees proﬁts. From an environmental perspective, free-allowance allocation has to be conditioned on operational and investment choices of companies to address leakage concerns. Different responses can be observed that are pursued to address leakage concerns. The Regional Greenhouse Gas Initiative has implemented a cap-and-trade scheme for the power sector in New England. Most of the participating states that have decided on the allocation methodology envisage 100 per cent auctioning of allowances. Leakage and competitiveness distortions, mainly relative to neighbouring states, will be monitored. In the discussions on carbon pricing by the Western Climate Initiative, leakage relative to neighbouring states is again an important topic. Both free-allowance allocation and border adjustment are discussed as measures to address concerns. The various proposals for a USA-wide emissions-trading scheme envisage free-allowance allocation and border provisions to address leakage concerns. Australian and New Zealand industries are voicing concern about competitiveness and leakage in the discussion of the design of these schemes. Free-allowance allocation to exposed sectors is again proposed. The EU ETS Directive provides criteria for the European Commission to identify sectors that are subject to leakage concerns. For these sectors

192

*

A world of different carbon prices

state aid, sectoral agreements, the use of export taxes and continued free-allowance allocation based on benchmarks can be used to address leakage concerns. This evolution raises one signiﬁcant concern. Different countries might implement domestic measures to address leakage, in particular free-allowance allocation and state aid. Any big country pursuing such a strategy will set a precedent that might be followed by other countries. Thus, national and regional emissions-trading schemes might be using inefﬁcient means to address (in some instances, even unwarranted) leakage concerns. It will be difﬁcult for other countries subsequently to implement more-efﬁcient schemes, or for any country individually to shift to a more-efﬁcient design. This illustrates the importance that has to be attributed to the decision on instruments to address leakage, so as to avoid lock-in to an inefﬁcient design choice.

Sectoral agreements to address leakage Sectoral agreements have been discussed at both industry and government levels. Co-operation among companies in the steel, cement and aluminium industries, and some initiatives in the power sector, typically involve voluntary agreements on sharing of best practice and collection of information for this purpose (Philibert 2004, IISI 2007). Such sectoral co-operation can play an important part in accelerating the response of industry to climate change (Baron et al. 2007). For sectoral co-operation at the government level in the Copenhagen Accord, developing countries were invited to describe NAMAs and to indicate their needs for international support (see Chapter 7). Many of the actions are deﬁned at the sector level, again illustrating the importance of sectoral co-operation. This raises the questions whether co-operation at the sectoral level can also address leakage concerns and how this would interact with the other objectives of sectoral co-operation. Sectoral agreements among companies cannot address leakage. Sometimes it is suggested by industry that trading schemes could be

Policy options to address leakage

*

193

developed that reward installations in developing countries for lowering emissions and create incentives for voluntary participation. However, this would imply that old, inefﬁcient plants in developed countries would fund investment in new and efﬁcient plants in developing countries. It is, however, unlikely that plants in developed countries would agree to subsidise their competitors. Even if big companies cover most of the emissions in a sector, the experience of the Indian steel sector illustrated the ability of smaller companies to grow rapidly – thus undermining any agreement among big companies (Sreenivasamurthy 2009). The national government is required for the implementation of a consistent policy that covers all installations. One could envisage governments wanting to focus their negotiations on a speciﬁc sector – for example, to demonstrate the viability of carbon pricing. What could be the shape of a successful sectoral agreement for a speciﬁc sector that allows for a full carbon price and avoids leakage concerns? Such an agreement would require a commitment of all participants to impose the full carbon price (i.e., prohibiting subsidies for domestic industry, for example, or using state aid or free-allowance allocation). Two options can be envisaged. The carbon price could be delivered through a domestic policy instrument of the participating countries (domestic carbon tax or cap-andtrade scheme). This might involve some agreement on the minimum price level that is imposed. Developing countries would probably request some level of international support in exchange for the implementation of ambitious climate policy. It would require extensive negotiation and political effort for international discussion among governments and their national champions to agree on more than a token minimum price. Alternatively, a trading scheme could be implemented for all installations covered by the sector in participating countries. It would imply that the sector would no longer be covered by the national trading

194

*

A world of different carbon prices

scheme. Such an approach would also not allow for smoothing of uncertainty about production volumes and technology innovation that is possible with trading schemes covering several sectors.6 Most proposals for government-led sectoral agreements envisage determining a benchmark emissions rate. Companies with emissions per unit of product that exceed this benchmark have to buy allowances to cover the additional emissions. If an installation produces fewer carbon emissions per unit of production than the benchmark, then it receives allowances for this difference and can sell those allowances. This creates the desired incentives for efﬁciency improvements in production. However, like free-allowance allocation based on benchmarks, the approach creates administrative constraints that restrict ﬂexibility for operation and investment. Also, the approach reduces the production costs of efﬁcient installations and only creates carbon costs for the inefﬁciency of installations, not for the full carbon externality. As a result, it will not drive consumers to explore substitutes or use carbon-intensive commodities more efﬁciently. If the policy does not result in a price increase for the carbonintensive commodity, then it does not create any leakage concern. But if there is no leakage concern, then no international approach is necessary to pursue the policy. Therefore, again the sectoral approach can equally be pursued with the objective of co-operation to enhance emissions reduction, and be relieved from the poisoned pill to negotiate among industry a solution that avoids leakage. In summary, we can say that concerns about leakage only arise in carbon-pricing schemes where producers pay the full price of carbon. Sectoral agreements are only required to address leakage if there is the ambition to impose the full carbon price as part of the agreement. However, there is little indication that this is the ambition of any of the schemes. Therefore, the objective of avoiding leakage has been gradually withdrawn from the discussions of sectoral approaches. This will increase their effectiveness in engaging a wider set of countries in pursuing climate policy.

Policy options to address leakage

*

195

Border adjustments The economics of border adjustment are simple. If leakage can be induced by domestic producers facing higher carbon prices, then leakage can be avoided when imports and exports are adjusted for the carbon price difference. Thus, the full carbon-price signal remains intact and creates incentives for innovation in new production processes, products and services, and supports the substitution towards lower carbon options (Demailly and Quirion 2006). The idea is already widely applied in schemes of value-added tax (VAT): for a car sold in Germany the sales price includes the VAT that was accrued over the various production steps. A private resident of Switzerland who buys a car in Germany initially bears the German VAT but gets a full refund when exporting it from Germany. The Swiss customs ofﬁce will levy VAT at the Swiss level when the car is imported. Thus, all cars competing for consumers in Germany include the German VAT in their sales price. Where they compete for Swiss consumers, the sales price includes VAT at the Swiss level. Thus, competition is not distorted despite the differing levels of VAT across countries. For the implementation of border adjustments, governments can choose whether the adjustment is done in allowances or in money. In the ﬁrst case, importers have to acquire allowances in the market or in auctions to cover the emissions associated with the production of their goods at the adjustment level, while exporters are compensated with allowances. Alternatively, the adjustment rate can be multiplied by the market price for carbon allowances to determine the import levy or export refund. The adjustment would be limited to a small number of speciﬁc, carbon-intensive commodities. This adjustment process could probably be pursued based on existing customs law and its product categories, and would therefore not require signiﬁcant additional administrative procedures or costs for governments or the private sector.

196

*

A world of different carbon prices

Then most difﬁcult question would be how far down the supply chain a border adjustment is applied. For example, in cement production the carbon-intensive commodity is clinker. Adjustments are applied to clinker at the level of carbon intensity of producing clinker with the best available technology. As a result, clinker costs and prices increase. This increases the costs of cement production and might result in some relocation unless border adjustment is also applied to the clinker content of cement. It is, however, not necessary to apply adjustments to products further down the supply chain. The cost increase for concrete products owing to higher clinker and cement prices are low relative to transport and other trade costs. Careful design and implementation is the key to WTO compatibility of border adjustment (Zhang 1998; Ismer and Neuhoff 2007). For this, the scheme may not differentiate between like products by foreign and domestic producers without due justiﬁcation. This requirement is met when charges levied at the border for imports or reimbursed for exports do not exceed the carbon costs of producing with best available technology (BAT). Also, border adjustment can only be applied to the extent that installations pay for their allowances. Border adjustment is not possible to the extent installations receive free allowances or state aid. The politics of border adjustment are more challenging. Developing countries have experienced a long history of border provisions with adverse impact on their economic development. This situation was not simpliﬁed by various proposals to use border measures as a stick to enforce participation in climate policy.7 Therefore, the clear anchoring in the general rules of the WTO is important to prevent policies that are initially targeted to address leakage concerns for a speciﬁc commodity to extend in scale or scope. This can involve international co-operation that clearly limits the scale and scope of border adjustment on carbon prices. Indeed, rather than creating barriers between countries, border adjustment for carbon-price differentials could support international

Policy options to address leakage

*

197

co-operation on climate policy – for example, by using net revenues to support climate policies in developing countries. More importantly, border adjustments allow countries to implement carbonpricing schemes with higher carbon prices so as to increase their de-carbonisation effort that is beneﬁcial for all countries. The political sensitivities associated with border adjustments require that they are discussed and implemented in close international co-operation. This creates trust and shared understanding among all parties about the objectives and limitations of border adjustment. Any country can engage in these discussions open-mindedly, because it retains the alternative options of state aid or using free-allowance allocation for exposed sectors. International co-operation could result in a formal or informal agreement to limit the use of border adjustment (see Figure 6.13). To be acceptable for developing countries, and WTO compatible, border adjustments have to be restricted to a narrow set of carbonintensive commodities. They should be limited to early stages of the value chain and to a scale that does not exceed carbon intensity of BAT. To avoid unfair subsidies, they also have to be limited to sectors that do not receive free allowances or state aid. The objectives that countries aim to pursue using border measures would also be limited to the implementation of effective domestic carbon pricing by adjustment of all trade at BAT. This would prevent Objective Design

Create effective domestic carbon price

Create incentives for foreign producers to improve efficiency

Create incentives for other countries to join deal

Assume best available technology for adjustment with all countries

Yes

No, because all are treated equal

No, because they are not discriminated against

Assume average carbon intensity and allow importers to demonstrate they are better

Yes

Probably not, if production from old/inefficient plant finds domestic users

No, because they are not discriminated against

Probably not, if production from old/inefficient plant finds domestic users

Only clear discrimination (e.g. adjustment > cost for best available technology can create incentives for third parties)

Apply adjustment only towards countries that do not participate in global deal

Probably not, if participating trade partners use regulation not carbon price

Figure 6.13 International co-operation could deﬁne limitations for the use of border adjustment

198

*

A world of different carbon prices

adjustment at a level exceeding BAT for importers that do not demonstrate efﬁcient production. This has been proposed to incentivise emissions reduction abroad. But it is difﬁcult to achieve because developing countries could export products from efﬁcient new plants and use other output for domestic use. The outcome could be additional complexity, creating non-tariff barriers. The limitation would ﬁnally prevent the use of border measures to punish non-participation in international climate efforts. Not only is the tariff level likely to be too low to be effective, but the approach also risks creating domestic opposition to any further climate action. It also precludes the use of border measures between participating countries that will be necessary if they use signiﬁcantly different policy mixes with different emphases on carbon prices. Border adjustment is politically contentious, but might well be implemented effectively – if pursued in an international framework that engages all countries. Developed countries would beneﬁt from effective carbon pricing, and may require international support to overcome opposition from incumbent companies wanting to retain free allowances. Developing countries would also beneﬁt if developed countries initiate leadership and provide examples of low-carbon transition in industry. International co-operation should ﬁnally ensure that some of the increased auction revenue is used to support climate policies in developing countries.

6.6 Conclusion The preceding chapters argued that if the level of domestic support exceeds the internationally agreed commitment level, countries could aim to pursue more-ambitious climate policies. To have an effect on investment decisions these domestic policies require commitment to mid-term targets that will probably involve higher carbon prices than in other countries or regions.

Conclusion

*

199

Would such differences in carbon prices would cause leakage by prompting companies to relocate production in countries with lower carbon prices? In these circumstances carbon pricing would shift rather than reduce emissions. As leakage is a concern for internationally traded commodities, this would imply that prices of the affected commodities would not increase to reﬂect carbon prices. Thus, leakage would also undermine the substitution effect. With relocation of industrial production, countries with high carbon prices may lose some jobs and tax revenue, which would undermine political support for the carbon-pricing scheme and the interest of other countries in implementing ambitious carbon-pricing schemes. Sector-speciﬁc analysis shows that leakage is not an economy-wide problem. It is only of concern for particular sub-sectors. For example, in the UK, carbon pricing results in non-trivial cost increases for only twenty-four sub-sectors, and these sectors represent only 1.1 per cent of the GDP (and less than 2 per cent in Germany). Whether leakage concerns are material for these sectors depends on sector-speciﬁc characteristics – for example, trade intensity, the origin of input factors, the ability of producers to address tailored needs of local consumers and the ratio between ﬁxed and variable costs as much as the expected capacity expansion. If carbon-price differentials are expected to persist over long time periods, then they might contribute to investment decisions for relocation in these sectors. The approach that is most prominent in cap-and-trade schemes is to use free-allowance allocation to compensate the sectors with strong leakage concerns about carbon cost increases. Effectively to address leakage, such allocation has to be conditional on continued operation, perhaps even conditional on speciﬁc production volumes and product types. This creates bureaucratic constraints and perverse incentives which limit the effectiveness of cap and trade to increase carbon efﬁciency, foster innovation and drive substitution towards lower-carbon commodities. Even so, from an environmental and economic perspective free-allowance allocation is the least desirable

200

*

A world of different carbon prices

approach to address leakage concerns, it is an established procedure, and therefore relatively easy to implement. It protects proﬁts and also undermines the substitution effect that could reduce future market volumes for high-carbon products, and thus receives signiﬁcant lobby support from incumbents in carbon-intensive sectors. In the European discussion, the use of state aid – explicit subsidy for investment and reinvestment choices – receives increasing attention as a means to address leakage concerns. As state aid rules are designed and enforced at the European level, but aid is granted usually at national level, this approach might offer an opportunity to move towards a less-politicised and more-technical decision process on the sectors affected and the level of support required. It is too early to judge whether the approach will succeed, and whether it could be replicated in other institutional settings. Government-led sectoral agreements have been proposed as alternative approaches to address leakage concerns while retaining environmental effectiveness. They could focus on delivering a similar carbon price in all relevant countries for speciﬁc sectors. This would create a level playing ﬁeld and avoid leakage. However, they would probably be complex to negotiate and to implement and might initially bear on opportunities that emerge from combining public- and private-sector expertise and initiative – for example, by sharing best practice. This will accelerate international co-operation on climate policy. Different approaches to border adjustment could compensate imports or exports of individual commodities for the production-cost differential directly associated with different carbon-price levels. Border adjustments are economically efﬁcient, and can be designed so as to be compatible with WTO rules. However, if pursued unilaterally, they risks repercussions for international co-operation on climate policy. The political sensitivities associated with trade-related measures require that border adjustments are only pursued in an international context that ensures trust and shared understanding of the purpose of the

Conclusion

*

201

measure and limits scale and scope to speciﬁc commodities where leakage concerns are clearly demonstrated. As these debates evolve in parallel in many countries, one big concern is that all schemes might ‘lock-in’ to a second-best solution. As some countries start to use free-allowance allocation or direct subsidies, others will follow. It might subsequently be difﬁcult to ﬁnd a way to alter these designs. Given the creation of vested interests, future improvements would be particularly difﬁcult where time frames for allocation decisions differ between countries. Thus, initial measures to address leakage concerns might seriously undermine the effectiveness of cap-and-trade schemes. ‘Sunset provisions’ might improve the situation by conditioning free allocation or carbon tax exemption to ongoing leakage concerns. The common challenge all countries face suggests beneﬁts from co-operation on analysis and policy design.

Notes 1 Higher carbon prices or more-carbon-intensive electricity production scale the value at stake in Figure 6.3 linearly – 20% higher prices increase the value at stake by 20%. 2 E = T/S = (exports + imports)/(demand + exports) = (exports + imports)/(production + imports). 3 Base case with demand elasticity of –0.6 and Armington trade elasticity of 5. Carbon intensity of BAT assumed to be 1.81 t CO2/tonne steel. Assumption that steel producers pass carbon price to product price. Increased demand elasticity −1 and Armington elasticity 10. 4 A second major steel production process is based on the electric arc furnace (EAF). While carbon pricing has a larger impact on the electricity costs of this process, the total cost increase is smaller because of the reduced process emissions. This can be easily explained – the main input is steel scrap. EAF steel production can be classiﬁed as steel recycling that is limited by availability of scrap steel. 5 Such requirements might be perceived as a trade barrier under rules of the WTO. See discussions on labelling or efﬁciency standards to support the use of lowcarbon products (Charnovitz 2004). 6 With a trading scheme for one sector, all installations would probably use similar technology and, more importantly, would depend on a similar technology

202

*

A world of different carbon prices

innovation to reduce emissions. Success of any one innovation is always uncertain, and therefore the sectoral target would have to be set rather leniently to ensure viability of the trading scheme even if innovation is delayed. With sectoral trading, the main beneﬁt of emissions-trading schemes is thus lost, combining uncertainties of innovation and growth rates of many sectors to reduce the uncertainty about the aggregate emissions-reduction opportunities. Obviously, the sectoral scheme could be linked to an outside carbon market; often the CDM market is mentioned in this context. But if there is an expectation that such a market will provide a credible price signal that is shared across sectors then this raises the question why there would be the need for a speciﬁc sectoral trading scheme. 7 Some proposals aim to compensate for average carbon intensities or to differentiate based on the climate policy implemented by the trade partner. This would, however, discriminate against some foreign producers. Proponents argue that their approaches could be exempt from stringent WTO requirements if they are presented as a component of an international environmental agreement. While this is in theory possible, it is uncertain how a WTO panel would rule. The approach would therefore not offer the certainty required for investment choices. Also, if carbon prices continue to differ across regions, the leakage might not necessarily occur along the lines of signatures of the international environmental agreement but along the lines of carbon-price differentials. This illustrates that border adjustment is an economic not a political instrument and should therefore also be implemented within the boundaries of economic rationale deﬁned by general WTO rules.

seven

International support for low-carbon growth in developing countries

During the Kyoto negotiations in 1997, only developed countries agreed to emissions-reduction targets.1 This reﬂected the principle of common but differentiated responsibility (Article 3(1), UNFCCC 1997). At the time, developed countries accounted for the majority of greenhouse gas emissions. Since then, many developing countries have exhibited impressive economic performance and a few countries have become richer in terms of GDP per capita than developed countries. As the economic development has largely followed the model set by Europe and the USA, early investment has focused on infrastructure built on cement, steel and manufacturing industries, and powered largely by fossil fuels, resulting in signiﬁcant emissions. Figure 7.1 illustrates how this has shifted the balance of emissions over the period 1990–2007. Counties such as the USA and Canada continue to exhibit extraordinarily high per capita CO2 emissions (y-axis) and emissions grew in line with population by 20 per cent. Emissions by economies in transition fell, with the structural changes after the fall of the Berlin Wall and efﬁciency improvements in industry, and are now at the level of per capita emissions of Western European countries in 1990. Western Europe 203

Low-carbon growth in developing countries

Developed country (2007) Developing country (2007) 1990 value

Latin America, the Caribbean

Population 2007: 6.5 billion East Asia

USA, Canada Japan, Australia, NewZealand Economics in Transition Middle East European Institute of Technology annex II, Morocco, Turkey Other Developing Countries

CO2t/capita 20 18 16 14 12 10 8 6 4 2 0

Africa

*

South Asia

204

Figure 7.1 Distribution of regional per capita greenhouse gas emissions in 1990 and 2007

achieved a 7 per cent per capita emissions reduction since 1990. With strong economic growth, per capita CO2 emissions in both China and India increased by 128 per cent. In addition, the populations of Asian and African countries have been growing. In aggregate, developing countries constituted 51 per cent of global CO2 emissions in 2007 (32 per cent in 1990). For the stabilisation of global temperature increases at 2 °C, it is no longer not sufﬁcient for developed countries to reduce their greenhouse gas emissions: in the mid-term, emissions from developing countries also have to peak and ultimately to decline.2 This is often seen as a challenge by countries experiencing strong economic

Low-carbon growth in developing countries

*

205

growth, raising the question as to how developing countries can pursue low-carbon development. To answer this question, this chapter largely draws on two international research projects with partners in India, Brazil, China, South Africa and Ghana.3 Workshops and country studies explored what would be necessary to make an individual sector or activity energyefﬁcient or low-carbon. The level of local interest in, and initiative to pursue, such a transformation was impressive. The analysis showed that a set of actions must be undertaken in parallel to facilitate a shift to low-carbon development of a sector or technology, often combining incentive schemes and adjustments to regulatory and administrative frameworks with training and capacity building. Only local stakeholders, from government, academia and often from industry and ﬁnance, have the knowledge and political support to identify, initiate and implement these actions. However, many developing countries have other pressing priorities and limited resources constraining their ability to ﬁnance measures to pursue emissions reduction. Therefore, international support can enhance the scale, scope and/or speed of local implementation, if tailored to these speciﬁc actions and designed so they can be accessed in the context of domestic initiatives. As the starting point for a low-carbon transformation can only be national actors, section 7.1 explores the domestic and international policy framework to integrate domestic action with international support. A variety of support mechanisms can be envisaged to facilitate domestic action, often categorised as technical assistance, capacity building, technology co-operation and ﬁnancial support. Section 7.2 focuses on mechanisms that can structure international ﬁnancial support for low carbon growth in developing countries, and criteria to select mechanisms that best meet the needs. Section 7.3 revisits the discussion on international carbon markets as a mechanism to facilitate ﬁnancial transfers and establish a carbon price. Section 7.4 concludes.

206

*

Low-carbon growth in developing countries

7.1 Framework for international co-operation Low-carbon development depends on appropriate domestic policies and programmes. International co-operation can only provide support to enhance the scale, scope and/or speed of their implementation. Mechanisms for international co-operation can therefore not be discussed without explicit reference to frameworks and programmes for domestic policy implementation. International negotiators had for a long time been reluctant to prescribe what domestic policies should be, preferring to negotiate targets for emissions outcomes and leaving policy detail to be determined at a domestic level. However, in the absence of a common vocabulary to talk about domestic climate policy, it is difﬁcult to deﬁne effective mechanisms to provide international support. Therefore, perhaps the most important achievement of the negotiations leading up to Copenhagen was the development of a new conceptual framework for international co-operation. This was reﬂected in the draft negotiation text published by the Chair of the ad hoc working group on long-term co-operative action under the Convention in the ﬁrst week of the Copenhagen negotiations (draft text from 11 December 2009: UNFCCC 2009a). Prior to publication, such a draft was discussed with parties involved in the negotiations to ensure it captured and reﬂected their perspectives. Discrepancies in negotiation positions were indicated by multiple wordings suggested for individual parts of the text. The text discusses four components of national and international climate policy in different levels of detail. Figure 7.2 outlines these components. (1) Low-carbon development strategies. South Africa was one of the ﬁrst countries to develop a long-term mitigation strategy outlining the intended economic, energy and emissions trajectory for the country (Winkler 2007), followed by Mexico and South Korea. This idea was later replicated in developed

Framework for international co-operation

3. International mechanisms

Capacity building

*

207

Technology co-operation

1. Low-carbon development strategy

2. Nationally Appropriate Mitigation Action

4. Reporting

Domestic International International verified

Figure 7.2 Concept for actions that allow for a transition in individual sectors or technologies

countries. In 2008, the UK Climate Change Commission was tasked with outlining low-carbon transformation needs and an emissions trajectory compatible with long-term mitigation targets deﬁned by the climate-change bill. (2) Nationally appropriate mitigation actions (NAMAs). The overall strategy is the basis for the identiﬁcation of trigger points for actions that allow for a low-carbon transformation. A NAMA can then be deﬁned as the set of policies and programmes necessary to facilitate the low-carbon transformation of a speciﬁc activity or sector of a country. Thus it can address the speciﬁc obstacles that could hinder or delay a shift to low-carbon services, technologies and infrastructure. As governments deﬁne and

208

*

Low-carbon growth in developing countries

implement NAMAs, they turn the strategy expressed in the low-carbon development plan into reality. Cross-sector policy instruments such as carbon-pricing schemes are an important component of one or several NAMAs. Given the political sensitivities associated with domestic policies and programmes, and given the country-speciﬁc circumstances, it is essential that NAMAs are developed and negotiated through the appropriate national policy processes. This was acknowledged in the Copenhagen Accord (para. 5), which led to an invitation to all countries, both developed and developing, to submit to the UNFCCC secretariat a list of NAMAs. Eightythree countries responded and submitted NAMAs they aim to implement. (3) International support mechanisms. Domestic initiative is essential for the success of any NAMA, and can be motivated by cobeneﬁts such as energy savings and reduced human health impacts. Discussion with stakeholders in the countries has also illustrated the difﬁculty of gathering sufﬁcient domestic support. International support could address regulatory, technical and ﬁnancial barriers for the implementation of NAMAs, and thus increase the scale, scope and/or speed of the implementation. The Copenhagen Accord therefore invited developing countries to indicate where they would require international support for the implementation of a NAMA. Previous submissions by developed and developing countries to the UNFCCC process have outlined a variety of international support mechanisms. They can play complementary roles in facilitating the adoption and diffusion of technologies. Depending on the speciﬁc needs of a country and sector, policies could require capacity-building measures, technical assistance, technology co-operation and/or ﬁnancial assistance. While capacity building and technical assistance are generally seen to be tailored to the needs of a sector and country, an ongoing

Framework for international co-operation

*

209

discussion is developing around the question of how best to determine the volume and structure of international ﬁnancial support to be provided for the implementation of NAMAs. The different options for ﬁnancial mechanisms are discussed in the remainder of this chapter. The Kyoto Protocol listed countries in two Annexes, creating only the categories of developing and developed countries. These categories no longer reﬂect the reality of national circumstances – many ‘developing’ countries now have economic performance and welfare well above the level of some ‘developed’ countries, and the capabilities of ‘developing’ countries vary enormously. However, it is politically difﬁcult to change these categorisations,4 not least because they are also the basis for coalitions formed by countries to negotiate in the international process. The challenge is therefore to build international support mechanisms that can match the scale and type of support provided to the speciﬁc capabilities and needs of each country, regardless of its formal categorisation. (4) Transparent monitoring and reporting. These create a set of indicators, also referred to as metrics or key performance indicators, and make them available to public and private decision makers. They facilitate the effective implementation of a policy, identify best practice from national and international learning, and create transparency and clarity to support private-sector investors and innovators. International co-operation can support domestic monitoring and reporting in both developed and developing countries. Co-operation can create timelines and frameworks to overcome inertia and translate high-level political support into effective monitoring and reporting. A limited level of harmonisation can improve and accelerate international learning from experience across countries. But national sensitivities are high, particularly if international veriﬁcation of autonomous domestic actions is discussed.

210

*

Low-carbon growth in developing countries

In the context of developing countries, three additional dimensions must be considered. First, some countries have limited resources and experience in monitoring. International support can enhance the quality of the monitoring and reporting which is being performed domestically and internationally. One example is the funding already provided for reporting in the National Communications under UNFCCC. Second, if international ﬁnancial support is provided to enhance the scale, scope and/or speed of the implementation of a NAMA in a country, some level of international veriﬁcation might be necessary. Without this, developed countries might relabel co-operation on non-climate-related aspects and include, for example, ﬁnancing for ‘energy-efﬁcient’ combat aeroplanes under their commitment to provide international support. Transparent reporting is the basis for an assessment to ensure ﬁnancial support is additional to existing and promised transfers under development assistance. Also, developing countries might not be forthcoming in reporting failure of individual components of a NAMA so as not to jeopardise ongoing ﬁnancial support, even if concealing the failure undermines the opportunity for international learning from the failure. Finally, developing countries have in the past resisted reporting requirements because of concerns that they create the basis for subsequent imposition of emissions reduction targets. The parallel development of reporting frameworks and of a broader set of climate policy instruments might reduce these concerns and clarify the role of information to support effective policy implementation and learning (Sippel and Neuhoff, 2009).

7.2 Financial needs for low-carbon development The design and implementation of an appropriate policy and regulatory framework for low-carbon transformation will involve a wide range of changes in socially sensitive sectors such as transportation

Financial needs

*

211

and buildings. They include changes to the national architecture of public policy, including taxation, subsidies and pricing policies in energy and transportation, as well as non-ﬁnancial mechanisms such as standards, procurement and urban planning. This transformation will only succeed if initiated and undertaken by domestic stakeholders. International ﬁnancial mechanisms can be used to contribute to incremental costs of low-carbon investments and to transition costs. By tailoring volume and structure of support to the speciﬁc sector and activities deﬁned in a NAMA, they can remove bottlenecks for the implementation of a NAMA and overcome resistance from actors who might otherwise have to bear the cost. Linking ﬁnancial support to the implementation of speciﬁc NAMAs allows governments in developed countries to demonstrate both the need for the support and the impact it can deliver. This can engage a broader public and is essential for acceptance and continuity of ﬁnancial support. International ﬁnancial offers the opportunity to enhance stability of regulatory frameworks. In many countries, frequent regulatory changes and limited enforcement of policies and programmes have reduced the credibility of domestic policies and programmes. The expectation of continued international ﬁnancial support for successful NAMA implementation can create incentives for domestic stakeholders to contribute to effective implementation and enforcement of the associated policies and programmes. This would increase the attractiveness to private-sector investors and allow domestic companies to develop and grow in pursuit of low-carbon strategies. International ﬁnance mechanisms can facilitate access to (low-cost) ﬁnance. The large-scale investment requirements of a low-carbon development can only be delivered if national and international private investment shifts from carbon-intensive to low-carbon activities and technologies. Most investment originates with institutions that are highly sensitive to risk–return ratios (e.g., sovereign wealth

212

*

Low-carbon growth in developing countries

funds, insurance companies, endowments, private banks) or face regulatory constraints on the risk exposure they can accept for their investment (e.g., pension funds). Additional risk guarantees and ﬁnancial incentives might be necessary to overcome these barriers, so as to create the initial interest and subsequently learning experience to facilitate a shift of private investment to low-carbon activities. International ﬁnancial support mechanisms can offer such guarantees directly to investors or can support domestic institutions in developing countries in overcoming initial risks. Every country will be, in principle, looking to receive transfers. However, the experience of many oil- and other resource-exporting countries points to the potential risks. In many instances, revenues from resource extraction have allowed bad governance practices to persist and contributed to an increase of inequality and consumption, preventing wider economic development. Transfer payments to support developing countries in ambitious climate policy must be carefully designed to match the needs of the country. The remainder of this section discusses how speciﬁc needs can guide the choice of the international support mechanism. Two basic categories for international ﬁnancial support mechanisms characterise the variety of public ﬁnancial mechanisms that are available and used by governments. First, governments provide direct grants to projects to cover incremental project costs or to create additional revenue streams – for example, through carbon credits. International mechanisms can also provide ﬁnancial support to countries as a contribution to the incremental costs that countries incur when implementing feed-in tariffs, supporting energy-efﬁcient buildings or introducing carbon pricing. Second, governments reduce ﬁnancing costs through the provision of preferential loans and equity, or through public credit guarantees that reduce the costs of commercial loans by eliminating country, currency, policy, technology or even project risk. The value of the international support provided with these instruments can be best accounted for as grant equivalent value. Risk guarantees or

Financial needs

*

213

preferential loans reduce the capital cost of a project and can thus replace an equivalent grant that would be necessary to make the project viable in the absence of the risk guarantee or preferential loan.

Contribution to investment and operation versus facilitating access to ﬁnance The choice of a suitable ﬁnancial support structure depends on activity, sector and national circumstances. Where constraints in capital access prevent low-carbon projects, the structure should be particularly geared towards providing access to ﬁnance through loans, credit and risk guarantees or equity ﬁnance. Facilitating access to ﬁnance is important for new technologies, as these face high risks because the intrinsic uncertainty about their reliability and future maintenance costs is aggravated by uncertainty about regulatory frameworks and infrastructure needs. Together, these factors weaken incentives for investors to provide large-scale ﬁnance to low-carbon projects in developing countries. This limits ﬁnancing sources to those private actors and funds that are prepared to bear higher risks in exchange for higher rates of return on employed capital. Credit guarantees can selectively remove some of the risk (e.g., currency, country, policy risk) and thus allow access to ﬁnance. Costs can therefore be reduced, and the necessary scale of low-carbon investment can be supported by institutions that are prepared to participate in ﬁnancing. This process can be initiated or complemented by direct provisions of loans. Facilitating access to ﬁnancing also allows projects to be undertaken on a commercial basis and thus contributes to the development of sustainable business models. Publicly initiated and ﬁnanced projects can only constitute a small share of the total volume of projects necessary to deliver low-carbon growth because the scale of investment exceeds the capacity of public budgets. Thus, it will be crucial to develop sustainable business models to deliver low-carbon and energy-efﬁcient technologies (using the same examples as above).

214

*

Low-carbon growth in developing countries

Provision of loans and credit guarantees to private actors can contribute to the development of business models and companies. However, incremental costs of some new technologies are signiﬁcant and can require additional support beyond grant-equivalent value of loans. In these cases, there may be insufﬁcient collateral or income streams to provide capital to cover incremental investment cost (Grant 2009). Grants can allow for local ownership, which is often seen to be essential for project success, and to initiate microﬁnance schemes (Gboney 2009). In addition, initial learning and transaction costs create barriers that can be overcome with regulatory design, technology co-operation and some additional costs for initial projects. In this case, direct grants can be simple, and create low transaction costs. They also provide support where beneﬁts are difﬁcult to appropriate by individual actors – for example, from technology improvements through learning-by-doing. More detail on the various mechanisms is available in a joint paper with Sam Fankhauser, Emmanuel Guerin, Jean Charles Hourcade, Helen Jackson, Ranjita Rajan and John Ward that is the basis for this section (Neuhoff, Fankhauser et al. 2010; see also Ward 2010; UNEP 2009). The remainder of this section will discuss the ﬁnancial mechanisms providing support during investment and operation (Table 7.1) and then ﬁnancial mechanisms to facilitate access to ﬁnance (Table 7.2).

Up-front support grants Up-front support is easy to implement and typically reduces transaction costs that are often challenging for small-scale projects. It also can facilitate capacity building – for example, as illustrated by a scheme of the European Investment Bank that pays venture capital funds’ initial management fees (European Investment Bank 2007). Up-front support is typically provided in the form of grant payments made by the public sector to help reduce the capital costs of a project or, more typically, to provide complementary institutional

t a b l e 7 . 1 Financial mechanisms to contribute to investment and operation Direct support Public ﬁnance mechanism

International to project

Indirect support

International to national

National to project

Up-front grant Global Environment Facility grants Other bilateral and multilateral direct - Standard technical funding initiatives assistance grants - ‘Smart’ grants

Overseas Development Agency

Investment support

Funding during operation

Grant linked to continuous delivery (ﬁnance + regulatory stability)

Incremental payment to renewable Removal of energy subsidies Carbon tax/cap-and-trade scheme

Off-set mechanisms (CDM) World Bank support

Provision of equity Asian Development Bank Clean - Private equity Energy private equity fund - Venture capital Europe Investment Bank / European Bank for Reconstruction and - Long-term investment Development Sovereign Wealth Funds

EIB/EBRD support for venture capital Carbon Trust and Transition fund set-up costs, and co-investment Economies venture capital in funds funds

t a b l e 7 . 2 Financial mechanisms to facilitate access to ﬁnance Direct support Public ﬁnance mechanism Provision of debt and equity - Loans (usually with governance conditions) - Credit lines - Equity (large projects, alongside foreign investors)

International to project International ﬁnancial institutions (e.g., European Bank for Reconstruction and Development, International Finance Corporation)

Indirect support International to national

National to project

International Monetary Fund and World Bank loans

Risk coverage Multilateral Investment Guarantee World Bank / International Finance Export credit Agency political risk insurance Corporation partial credit and agency - Full or partial guarantee - Policy to cover all or speciﬁc causes partial risk guarantees guarantees of non-performance - Other ﬁnancial products

Financial needs

*

217

support (technical assistance, capacity building, due diligence support etc.). To trigger low-carbon investments, grants are helpful when the capital costs of the low-carbon technology are greater than the costs of a fossil-based technology. This is the philosophy of the Global Environment Facility (GEF), whose mandate is to pay the ‘incremental cost’ of global environmental projects. However, the bulk of grants for supporting particular projects continues to be provided by national governments, often in the context of bilateral development assistance. Concerns about the risk of moral hazard problems involved in such grants have prompted some innovations. For instance, the European Bank for Reconstruction and Development (EBRD) supports energyefﬁciency projects in Eastern Europe, but only provides grants ex post, when projects are accredited as having delivered the identiﬁed improvements. This approach is designed to provide strong incentives for project success. Encouragingly, the local ﬁnancial sector also has developed instruments that cover the gap – for instance, short-term loans against a grant approval document. However, this up-front support provides no hedging against moral hazard and no incentive to maximise the operational performance of projects. This is the lesson from up-front tax credits to support wind projects, initially in California and later in India, which resulted in the underperformance of many projects owing to inappropriate locations, quality of turbines and of maintenance.

Support during operation Spreading support over the lifetime of low-carbon projects allows support to be linked to project performance, thus generating incentives for effective implementation, installation and operation. Operating support has the main beneﬁt of improved incentive properties – private-sector actors will aim to implement projects rapidly and choose suitable quality control and maintenance structures so as to ensure performance. In developed countries, the incremental costs created by renewable support schemes are typically

218

*

Low-carbon growth in developing countries

allocated to electricity users, so as to avoid subsidies for energy consumption. However, should such costs become signiﬁcant in developing countries, then international support could contribute directly towards these costs, or could support energy-efﬁciency measures to allow energy users to reduce their consumption and this exposure to energy prices. Operating support can be provided through feed-in tariffs, which provide long-term guarantees to buy renewable energy, often at abovemarket prices. Thus, additional revenues are provided to investors but are conditional upon project delivery, while the guaranteed price reduces investment risk and ﬁnancing costs. International support could provide grants to contribute towards these incremental costs. Carbon market mechanisms also aim to create support during operation. The different options are discussed in detail in section 7.3. The existing Kyoto clean development mechanism (CDM) allows for the sale of emissions reductions, delivered at project level in developing countries, to developed countries. However, administrative complexity has limited the regional and sectoral scope of its application, particularly for small-scale and complex projects. The uncertainty in demand for off-sets, and resulting price volatility, has led to signiﬁcant discounting of the value of off-sets in ﬁnancing decisions.

Facilitating access to ﬁnance Equity, loan or risk coverage are three additional options for international support to enhance the ability of domestic governments to implement a low-carbon investment framework. To clarify their relative merits for providing ﬁnance for low-carbon projects, a distinction can be drawn between mechanisms that transfer risk to the public sector and mechanisms under which the public sector shares in risk through the provision of capital. The main option for transferring risks to the public sector is insurance or guarantee products. They are logically suited to cover currency, country and policy risk, which are largely determined by

Financial needs

*

219

public policy decisions. This would encourage the simultaneous provision of credit-risk guarantees and the development of an attractive investment framework, with enhanced credibility of the overall domestic policy framework, possibly through contracts in which two countries declare themselves to be jointly and severally liable. The amount of compensation provided can be full or partial. They can also be provided only to creditors or to all providers of capital. Alternative government support schemes can be structured so that private insurance companies take the ﬁrst hit and governments back insurance companies – for example, to cover systematic risks. Such products can provide protection against certain speciﬁc events that cause non-performance – for example, political instability – or against general non-performance. Nevertheless, if risk guarantees are expanded to encompass the majority of potential risk components, it is more justiﬁed for public agencies to provide direct loans or equity, thus avoiding complexities and transaction costs. The other principal means for the public sector to improve access to ﬁnance is by direct provision of capital on terms that are advantageous compared to those that would be available in private capital markets. Typically, corporate and project ﬁnance will rely on a mix of credit and equity ﬁnance, balancing their speciﬁc risks (Myers 1984). Accordingly, governments can provide capital either as debt capital (loans) or equity capital, depending on the requirements of the project/enterprise and the risk aversion of the public investor. Debt does not dilute the shareholdings of existing owners; however, excessive leverage (or weak creditor supervision) may introduce moral hazard problems and, in case of underperformance, the leverage effect can be reversed into a trap. Equity ﬁnance dilutes shareholdings but has the advantage of providing new funders with access to information, as well as aligning incentives between fund providers and entrepreneurs.

220

*

Low-carbon growth in developing countries

In summary, loans or, potentially, equity contributions are preferable where comprehensive risk coverage is necessary. To limit moral hazard problems, risks that depend on actions of operators and investors should remain with private actors, focusing risk guarantees on currency, country and, possibly, policy risk components.

Summary The choice of a suitable support mechanism requires a balancing of the requirements and constraints of partner countries, private investors and international institutions, where they are involved. Thus, the mechanism and design might well be tailored to the speciﬁc NAMA that is supported and the institutions that are available in the countries involved. To facilitate early learning from the different experiences that this creates, transparent and timely monitoring and reporting are essential as they allow for an evaluation of the performance of different mechanisms and designs and a diagnosis of difﬁculties that need to be tackled. They also reduce uncertainty, facilitating private-sector investment, and are the basis for an engagement of a broader public, to ensure continued support.

7.3 The role of carbon markets to provide support for developing countries The most substantial experience with international carbon markets relates to the CDM introduced with the Kyoto Protocol (Michaelowa, Butzengeiger and Jung 2005). Companies investing in a low-carbon project in a developing country can register the project and, if accredited and veriﬁed, create certiﬁed emission reduction (CER) credits. CER credits are usually bought on long-term contracts by companies that are part of the EU ETS. Installations under the EU ETS can use CER credits up to a predeﬁned limit to cover their emissions, thus reducing their need to buy emissions allowances.

Support for developing countries

*

221

The international accreditation, monitoring and veriﬁcation of each project creates signiﬁcant transaction costs and thus restricts the applicability for small projects. In response, programmatic CDM were developed. They allow projects to be bundled and evaluated jointly to reduce transaction costs of small projects (Sterk and Wittneben 2005). Thus, project design and accreditation costs only accrue once, and with the help of statistical methods the evaluation can be limited to a sample of individual projects. Programmatic CDMs have thus been one improvement of the clean development mechanism. However, by August 2010, only three programmatic CDM projects have been registered, with expected cumulative emissions reduction of 174,000 tonnes by 2012 (UNEP Risoe 2010). As of August 2010, 2,306 CDM projects were registered, 40 per cent of the projects being located in China, 23 per cent in India, 8 per cent in Brazil, 5 per cent in Mexico and the remaining 25 per cent distributed across other developing countries. The CDM projects registered by August 2010 are expected to deliver more than 375 million tonnes of CO2 equivalent emissions reduction per year during the ﬁrst commitment period. Thus the expected volume of certiﬁed emissions reduction that will be available for the period 2008–2012 is 1.8 billion, with at least 1 billion CERs associated with projects in the pipeline (UNEP Risoe 2010). One strength of the CDM is the transparent governance structure for international decision making, linked to a clearly deﬁned metric. The CDM executive committee is responsible for accreditation of projects and monitoring of delivered emissions reduction. The mechanism creates incentives for private-sector agents to propose methodologies, initiate projects and negotiate planning and regulation within local and national administrations. This has created a dynamic that would have been difﬁcult to envisage without private-sector participation. Thus, the CDM has created a source of transfer that is not directly linked to the budgets of national governments in developed countries

222

*

Low-carbon growth in developing countries

which was assumed to be easier to agree in international political processes (as well as being better protected from short-term political volatility). The CDM established a global price for certiﬁed emissions reduction – a strong signal observed by private actors and policymakers in many regions of the world – and it has delivered projects across a range of countries and technologies. The CDM has been successful in supporting individual projects, establishing the use of some low-carbon technologies, creating local stakeholders for climate policy in developing countries and demonstrating a certain amount of commitment by developed countries to support developing countries (Castro and Michaelowa 2008). This raises the question as to what role the clean development mechanism will play in the future. This section considers several criteria for the evaluation of the CDM and alternative carbon-market-based mechanisms. The total costs of emissions reduction in developing countries, and thus the potential requirement to support the incremental component, will be signiﬁcant. Figure 7.3 approximates the volume of possible emissions reduction in developing countries for different carbon prices, based on the 2007 IPCC report. Assuming the total volume of emissions reduction could be realised, this would contribute annually about 8 billion tonnes at a price of $20/tonne CO2. If this price is paid for each tonne of emissions reduction, it would require annual payments of $160 billion. By contrast, the CDM volume in 2010 is approximately 0.35 billion tonnes per year.

Rents If each unit of emissions reduction is paid for at the same carbon price, the required transfer volume can be a multiple of the real costs incurred by low-carbon efforts. Figure 7.3 illustrates the differing costs per tonne of CO2 for different measures that reduce carbon emissions. If all emissions reduction received the same carbon price then developers of lower-cost projects, or the countries where such projects were

Support for developing countries

*

223

$/t CO2 <100

<50

Energy supply Transport Buildings (electrical) Buildings (other) Industry (electrical) Industry (other) Agriculture Forestry Waste

<20 <0 0

2 4 6 8 10 12 14 Gt CO2 -equivalent emission- reduction potential (2030)

Figure 7.3 Economic potential for greenhouse gas emissions reduction in nonOECD, non-EIT countries relative to baseline emissions, based on Special Report on Emissions Scenarios B2. Source: IPCC 2007.

hosted, could capture the rents between the carbon price and cost. On average, these rents are likely to be bigger than the additional project costs.5 While rents create economic incentives and can accelerate the de-carbonisation process, the scale of the rents involved could undermine political support for the mechanism in developed countries. As carbon markets in developed countries are competitive, and free entry by intermediaries is possible, much of the rent will ultimately be captured by companies or actors that provide the mitigation opportunity. Thus, companies in sectors such as steel, cement, chemicals or power that can offer opportunities for efﬁciency and carbon-efﬁciency improvements, will capture the rent. This will increase their activities and reduce product prices from these sectors and can result in higher demand for carbon-intensive products and services. China has imposed taxes on project credits, differentiated by project type and sustainability, in order to capture some of the rent and to use it, for example, for energy-efﬁciency programmes. The

224

*

Low-carbon growth in developing countries

tax rate for industrial gas projects is 65 per cent and for renewable projects 2 per cent (Zeppezauer 2009).

Domestic climate policy disincentives in developing countries The clean development mechanism creates an incentive for countries to delay implementation of domestic energy-efﬁciency and renewable policies. This is because low-carbon projects must satisfy the additionality criterion to generate CER credits. This requires that the project not be commercially viable without the sales revenue from CER credits. However, if there were domestic policies in place that would allow private actors to undertake low-carbon projects, such projects would not qualify as CDM projects, and the country would forgo international ﬁnancial inﬂows from the sale of CDM credits, and national stakeholders would no longer capture rents. To prevent this perverse outcome, the E+/E− rule was designed, specifying that all projects should be evaluated relative to the policy environment prior to the year 1997 (E+) or 2001 (E−), where E+ (E−) evaluations should be made disregarding more recent policies that have given competitive advantage to more- (less-) emissionintensive technologies. The rejection of ten Chinese wind projects by the CDM executive board meeting in 2010 was interpreted as a reverse application of this rule. The board suspected that China had reduced the feed-in tariffs for wind power so as to ensure wind projects would only be viable with CDM support and could thus qualify for the support (He and Morse 2010). Such inconsistent application of CDM rules poses a threat for investor conﬁdence for current and subsequent ﬁnancial mechanisms. The CDM is thus caught in a fundamental tension. Not taking policy developments into consideration might create excessive rents and undermine effectiveness and political acceptability in developed countries, but taking policy developments into consideration when assessing projects for CDM support could create disincentives for the pursuit of these policy developments in developing countries.

Support for developing countries

*

225

Reduced incentives for low-carbon transformation in developed countries The basic principle of off-setting mechanisms implies that lower-cost emissions reduction undertaken in developing countries replaces higher-cost emissions reduction in developed countries. It is frequently expected that carbon prices in developed countries would have to rise above current levels of €15–€20/tonne CO2 to support a low-carbon transformation. However, if such price levels are maintained for a few years then large amounts of off-sets would be generated and would depress the carbon price. The Kyoto Protocol has deﬁned a ‘supplementary criterion’ for international off-sets to ensure that off-sets complement and do not replace domestic mitigation action. In interpreting this criterion, EU countries have imposed limits on CDM and JI credits that can be used by installations in EU ETS. If this maximum amount is utilised then only limited emissions reduction has to be undertaken within Europe (Figure 7.4). EU emissions would exceed the path implied by a linear trajectory that brings Europe towards 80–95 per cent emissions reduction by 2050. 6

CO2 emissions in Gt/year

European 2020 targets (relative to 1990)

5

20% and full use of clean development mechanism/joint implementation 30% and full use of clean development mechanism/joint implementation 20% and no use of clean development mechanism/joint implementation

4

30% and no use of clean development mechanism/joint implementation

3 2 80–95% emissions reduction by 2050, relative to 1990

1 2010

2020

2030

2040

2050

Figure 7.4 European emissions trajectories with and without off-sets. Source: Based on Neuhoff and Vasa 2010.

226

*

Low-carbon growth in developing countries

The European Commission must re-evaluate the limits on the use of CDM credits if the EU changes emissions-reduction targets from 20 per cent to 30 per cent in the context of an effective international agreement. Combining a 30 per cent emissions-reduction target with a stop on all further use of CDM credits could make the EU emissions cap compatible with a linear trajectory towards 80–95 per cent emissions reduction by 2050. However, many large emitters (excluding power generators) oppose such adjustments, stating that the higher carbon prices that would result would jeopardise their economic viability and would result in the relocation of production activities and thus leakage of emissions. This illustrates the importance of the off-setting mechanisms for a larger set of emitters in developed countries. Attempts to change or replace the CDM to enhance the effectiveness of international ﬁnancial support might thus face opposition from unexpected groups of stakeholders in their attempt to reduce the stringency of the domestic emissions cap. Chapter 6 discusses empirical evidence on leakage concerns and explores alternative policy options to provide tailored support to individual sectors – while maintaining a stringent emissions cap. If off-setting mechanisms are to remain a component of international climate co-operation, some level of off-setting can be made compatible with stringent domestic mitigation efforts in developed countries by increasing the European emissions reduction target above 30 per cent. This would create headroom for the use of international credits.

Uncertainty Uncertainty about future carbon prices has been a challenge for investment in low-carbon projects under domestic cap-and-trade schemes (Chapter 4). International off-setting mechanisms are linked to the uncertain carbon price from national cap-and-trade schemes and face additional uncertainties. First, what constraints do domestic

Support for developing countries

*

227

cap-and-trade schemes impose on the use of off-setting credits? Second, which additional national cap-and-trade schemes or other mechanisms (e.g., voluntary agreements in Japan) will create demand for off-sets? Third, how will the different technologies perform in delivering emissions reduction? Landﬁll projects have delivered only 35 per cent of expected CERs, whereas N2O projects overachieved – with 123 per cent of project design document-expected CERs (Castro and Michaelowa 2008). Fourth, what additional project types will qualify as off-sets (CDM and JI credits)? A decision of the CDM executive board, qualifying supercritical coal-power stations in spring 2008 and altered in 2010, creating a potentially signiﬁcant increase in supply of potentially low-cost off-sets that could alter the future supply–demand balance. So far, off-sets from avoided deforestation do not qualify. Should this change in the future, then a large amount of additional off-sets could be provided to the market. As these uncertainties are related to political and regulatory decisions, they are particularly difﬁcult for market participants to assess. These uncertainties result in a signiﬁcant discount on the value provided by off-set credits for funding of low-carbon projects and the value off-sets represent for ﬁnancing institutions. This is particularly important as ﬁnancing institutions balance the income stream from off-sets against the risk–return characteristics of the underlying technology of the project. As discussed, innovative technologies have currently a higher degree of risk, so uncertainties about the future of off-sets and their volume create additional risks and complexities that negatively impact loan decisions for these technologies. Developing countries will require more certainty about future revenue streams from international support mechanisms if these mechanisms are to drive a low-carbon development strategy. Also, companies covered by cap-and-trade schemes in developed countries beneﬁt from more predictable future volumes of off-sets used in their domestic schemes to inform the decisions on mitigation efforts and investment to be pursued by installations covered by the schemes.

228

*

Low-carbon growth in developing countries

Concerns about the effectiveness of the CDM mechanism in supporting low-carbon transformations have triggered extensive discussions about alternative ﬁnancial mechanisms, as outlined in the previous section. Further developments of carbon-market-based instruments have also been discussed. As it is a broadly shared perspective that developing countries will not deﬁne absolute national emissions caps in the near future, the various approaches towards linking emissions-trading schemes cannot be used as a basis for co-operation. Hence, several approaches of carbon-markets-based international cooperation on the sector level are being discussed. Three examples are: * installation-based sectoral trading relative to benchmark performance; * sectoral trading by linking to a sector cap-and-trade scheme; * sectoral crediting or trading at the government level.

Installation-based sectoral trading relative to benchmark performance The concept envisages that installations in a deﬁned sector of a developing country can participate in an international carbon market. If an installation’s CO2 emissions per unit – for example, of steel or power – are below a benchmark level, it can sell the difference to installations in countries covered by a cap-and-trade scheme that need to cover their CO2 emissions.6 The approach can also comprise a requirement for all installations in the sector to participate; installations that are less efﬁcient than the benchmark have to buy allowances. The scheme creates incentives to improve the efﬁciency of production plants in developing countries, because more allowances can be sold if CO2 emissions per unit of production are reduced. In order for a developing country voluntarily to participate in such a scheme that includes all installations in a sector, the installations in the country must in aggregate be a net seller of allowances. All installations with emissions per unit of production below the benchmark

Support for developing countries

*

229

beneﬁt with every unit of production. Hence, the scheme creates a perverse incentive to increase the production of carbon-intensive products such as steel or power. This is the same effect that was described in Chapter 3 for some types of free-allowance allocation within cap-and-trade schemes. Furthermore, setting the benchmark is likely to combine a challenging combination of issues relating to data (availability), technology (production method, precise deﬁnition of basis and level of benchmark, coverage of on-site power) and politics (is the benchmark set at national or international levels and which international institution has the competence to negotiate technical aspects with global industrial players?). Higher benchmarks increase the ﬁnancial transfers to developing countries given a stringent target in developed countries. If the benchmark exceeds the performance of a plant, higher benchmarks also increase incentives to boost production at installations covered by the scheme. Both effects are in the interest of the developing country, but not of governments and industry in developed countries. The scheme creates a transfer of funds. For example, a steel plant covered by a cap-and-trade scheme in a developed country might buy allowances from a steel plant in a developing country that is able to sell CO2 allowances because its plant is more efﬁcient than the benchmark. Financial transfers to developing countries from potentially identical installations in developed countries create competitive distortions. They are unlikely to be accepted by installations and governments in developed countries, and hamper political support for the sectoral crediting approach in industrial sectors that are globally traded. Another difﬁculty relates to the deﬁnition of an installation. The EU ETS excludes small installations in order to limit administrative costs. These installations are covered by other domestic policy instruments, including energy, and potentially carbon, taxes, which are implemented by governments to deliver national emissions targets. A developing country without a policy objective deﬁned at national

230

*

Low-carbon growth in developing countries

level might not have such an instrument, and emissions-intensive companies might aim to escape the coverage of the scheme by decreasing the scale of individual installations.

Sectoral trading by linking to a sector cap-and-trade scheme Rather than deﬁning some type of benchmark for installations in a sector, an emissions target for an entire sector could be deﬁned. A national cap-and-trade scheme would thus be deﬁned at sector level. A sector cap would be established, with subsequent allocation to installations at no cost or through auctions. A generous cap allows the government to sell more allowances than required to cover emissions in the scheme. Linking the scheme to cap-and-trade schemes of other countries allows for an export of those allowances, creating net revenue for the country. If the majority of emissions covered by the linked national capand-trade schemes are from one speciﬁc sector, then any technology innovation that changes production efﬁciency or product demand will signiﬁcantly alter the overall supply–demand balance and scarcity prices of allowances. Hence, sectoral crediting or trading mechanisms must be integrated with broader emissions-trading schemes to increase predictability of future carbon prices. As for benchmark-based sector crediting approaches, one implementation challenge is the deﬁnition of sectoral boundaries. With a cap deﬁned for a sector, a country may aim to attribute emissions to other sectors so as to allow for more exports of allowances for the sector. To give an example, the attribution of activities to sectors already created under the EU ETS caused difﬁculties in the case of combined production, such as joint heat-and-power production or with regard to the treatment of waste gases (e.g., from steel production) and their use for power generation. The exclusion of small installations can also create opportunities for gaming. Distortions between sectors may constitute a second set of challenges. If we assume the power sector faces a signiﬁcant carbon price

Support for developing countries

*

231

created in a sector-trading scheme with auctioning of allowances, power prices would increase, while other fuel prices may stay constant. This might distort the efﬁciency of choice between gas, oil, coal or electric power as energy input, and could also create incentives for investment in decentralised power generation not covered by a scheme, potentially with inefﬁcient fossil fuel engines. The main challenge for the approach might, however, relate to the deﬁnition of a suitable emissions target for the sector. Future ‘business as usual’ emissions projections are inherently speculative. For example, the IEA projections for Chinese CO2 emissions in 2020 almost doubled between the World Energy Outlook 2002 and 2007 (IEA and OECD 2002; 2007). In order to ensure net exports of CO2 allowances, developing countries would probably aim to set emissions targets close to the higher end of projections. If mitigation turns out to be more successful than in such conservative scenarios, or to be aided by less-bullish economic growth rates, allowances corresponding to large shares of the speciﬁc sector would be available for sale in international markets. This inﬂow of credits would undermine the scarcity and domestic action in developed countries. Already the expectation of a potential large inﬂow could inhibit low-carbon investment by both the private and the public sectors. The concept of ‘no-lose targets’ aims to tackle this challenge. Developing countries can sell allowances if they reduce emissions below the target (and trajectory), but are not liable to buy allowances if they fail to deliver the country’s emissions target (Philibert 2000). Sector no-lose targets reduce the risk for developing countries and thus allow for an agreement on more ambitious targets (Bosi and Ellis 2005) and a reduction of the volume of allowances that will most likely be exported. However, the concept of no-lose targets is difﬁcult to apply if trade is implemented at the installation level. It would imply that CO2 allowances in the national trading scheme of a developing country would lose value as soon as aggregate emissions increased above the negotiated no-lose target. Such a carbon price

232

*

Low-carbon growth in developing countries

would do very little to support low-carbon investment, but would create ample opportunities for gaming. Hence, international no-lose targets can in effect only be combined with sectoral trading at government level.

Sectoral trading or crediting at government level Sectoral trading at government level is based on an internationally agreed emissions target for a speciﬁed sector. If the emissions volume exceeds the target level, the country is obliged to buy allowances from other countries. However, with no-lose emissions targets, no such obligation is imposed, and the scheme would be called a sectoral crediting mechanism. This allows also risk-adverse governments to sign up to relatively ambitious emissions-reduction targets. Shiell (2003) shows in a formal model that the resulting, more-stringent targets can represent a Pareto improvement. However, the disadvantage of the no-lose target is that as soon as it becomes apparent that the target cannot be delivered the incentive for emissions reduction is reduced to the level of discounted beneﬁts in future years if governments succeed in reducing emissions below the no-lose targets. Targets can be deﬁned in absolute terms or in emissions relative to some output metric of the sector (tonnes of cement or steel, MW h of electricity). If the country exhibits emissions in the sector below the target level, the difference can be sold by government in international markets. The revenue from these credits creates an incentive to implement effective sectoral policies to reduce carbon emissions (Baron and Ellis 2006). Several years may pass between the implementation of a policy package, private-sector investment response by public and private organisations and the impact on emissions of a country. This raises the question as to whether policy makers, particularly in a developing country, would pursue policies in which current policy makers and actors incur the costs, but beneﬁts in terms of ﬁnancial transfers only aid future governments, with a time lag of ﬁve to ten years.

Support for developing countries

*

233

To reduce time lags and thereby increase the beneﬁt to current policy makers and encourage continuation of the policy, international support can be linked to lead indicators (i.e., indicators that provide early responses to changes) reﬂecting effort, implementation and success of policies and programmes (see Box 5.1). Such lead indicators, or intermediate output metrics, could, for example, capture changes to the regulatory framework, changes of relative energy and carbon prices or responses of individual investors. However, with lead indicators, the direct link to quantiﬁed emissions reduction is lost. It also increases the emphasis on well-deﬁned lead indicators and effective monitoring to avoid gaming opportunities. Linking international support to lead indicators reduces ﬂexibility for governments to adjust their actions, as changes might require renegotiation of the agreed lead indicators or objectives. The advantage is that lead indicators are not directly linked to carbon emissions, creating the opportunity to negotiate the level of international transfers relative to some measure of incremental cost, thus reducing concerns of rent transfers and the need for the deﬁnition of baselines. With sectoral crediting at government level, the volume of transfers is linked to the volume of emissions reduction that is paid at the price of certiﬁed emissions reduction. This creates a challenge for both developed and developing countries: uncertainty of revenue ﬂows related both to the market-based price of certiﬁed emissions reduction and to the volume of these ﬂows. In addition, as all the emissions reduction is paid for at the same price, signiﬁcant volumes of rent transfers might be incurred as some emissions reduction is cheaper to implement than others. Arguably, setting a more ambitious target could reduce the value of rents transferred. But, at the same time, the likelihood of exceeding the target would increase. In the case of no-lose targets, incentives to implement policies to deliver emissions reduction would decline.

234

*

Low-carbon growth in developing countries

Given the potential scale of these ﬁnancial ﬂows, they might trigger credit import constraints in developed countries similar to current supplementarity limits for the use of CDM. This creates risks of lowcarbon prices for credits from sectoral crediting mechanisms. The scale of ﬂows will also have a signiﬁcant impact on budgets. Policy makers in developing countries will require some stability of ﬁnancial ﬂows so as to be able to fund projects or staff costs. These costs are often not directly related to the level of success of a policy or programme in delivering emissions reduction. Governments would have to sign long-term contracts to hedge some of the risks. Such contracts might well be signed with governments in developed countries as counterparty. In the process of negotiating such a contract, a sector crediting mechanism at government level might look increasingly similar to ﬁnancial support schemes as outlined in section 7.2.

7.4 Conclusion Climate policy objectives have changed, from marginal emissions reduction to low-carbon development as the framework for largescale emissions reduction. This is also reﬂected in the shifting focus from the CDM off-setting mechanism to various types of public ﬁnance mechanisms to support the implementation of NAMAs. The CDM has delivered several beneﬁts, including a set of active stakeholders across the globe exploring low-carbon opportunities and a sense of carbon as a tangible commodity. But it is not well suited as a major mechanism for the provision of large-scale support owing to the signiﬁcant rents that can often be captured as subsidies it provides to energy-intensive sectors. These reduce the incentive to move to lower-carbon industrial structures and fail to encourage lowcarbon policy frameworks. This raises the question as to whether future carbon-market-based mechanisms can effectively replace or complement the clean development mechanism. Sectoral trading approaches are frequently

Conclusion

*

235

discussed but might be difﬁcult to operationalise as a low-carbon investment framework. Sectoral crediting mechanisms at government level could combine some elements of carbon markets with effective incentives for domestic policy implementation, but the time lag between action and future reward in combination with uncertainties inherent in carbon markets remain issues. Therefore, the focus is increasingly moving to public ﬁnance mechanisms. These can aim to address multiple objectives. First, they can support countries where domestic resources are already stretched by the provision of basic needs, such as education, health and sometimes nutrition, let alone covering incremental costs of low-carbon choices. Second, they can tailor the ﬁnancial support to needs of actors who bear the costs, to enhance domestic support and facilitate implementation and contribute to regulatory stability of domestic policy frameworks. The choice of suitable ﬁnancial instruments depends on the requirements of actors and investors in the speciﬁc activity, sector and country. A variety of ﬁnancial mechanisms is available. They can provide ﬁnancial support in advance and during operation, or can reduce the exposure of private investors to policy-related uncertainties through the provision of equity, debt or risk coverage during an initial transition and learning period. Thus, they can unlock initial domestic and international private-sector investment in low-carbon choices. Financial mechanisms have been successfully implemented in national and international, and climate- and non-climate-related circumstances. Effectively to support low-carbon development they must be tailored to the fullest extent possible to the speciﬁc circumstances of the activity and country, while retaining a set of common rules and frameworks in order to limit the complexity for developing and developed countries using the mechanisms. Transparent monitoring and reporting can facilitate international learning on suitable design for international support mechanism and

236

*

Low-carbon growth in developing countries

domestic policy choices and thus improve their implementation. They also allow a broader public to engage with the international support structure, to ensure continued political support for the cooperation.

Notes 1 In this book I use the terms ‘developing countries’ and ‘developed countries’ for the sake of simplicity, and as they correspond to the classiﬁcations ‘Annex 1 countries’ and ‘non-Annex 1’ countries introduced in the Kyoto Protocol. 2 In countries like Brazil and Indonesia, the majority of emissions are related to deforestation but will be not addressed in this chapter. 3 The project was convened by Climate Strategies and includes contributions from Cambridge Intellectual Property Limited (CambridgeIP), Cambridge University, CIRED, ECN, IDDRI, IDS, IISI, Indian Institute of Technology Kanpur, International Institute of Economics and Management, LSE, Oxford University, Tsinghua University, the University of Cape Town and Universidade Federal Fluminense. Climate Policy, 9(5) (2009) reports on phase I of the project and results from phase II are available at www.eprg.group.cam.ac.uk/isdahome. 4 Winkler, Brouns and Kartha (2006) suggest approaches to differentiate among non-Annex I countries for future mitigation commitments. Michaelowa, Butzengeiger and Jung (2005) suggest that developing countries which exceed a threshold (deﬁned by GDP per capita and emissions) gradually become responsible for mitigation efforts. 5 This assumes the typical shape of marginal abatement cost curves is approximated with a quadratic function. 6 Samaniego and Figueres 2002 discuss the extension of CDM to sectoral CDM in order to cover a wider range of projects.

eight

Conclusion

Climate change is a big challenge for our generation and society. This has led many to believe that one big solution is required and that there can be one international agreement and preferably one big technology to solve it all. The negotiations in Copenhagen did not deliver the big solution, and I think that would have been an impossible task. What is this challenge? Limiting global temperature increases to 2 °C and reducing the risks associated with climate change demand large-scale reductions of carbon emissions. The reductions can only be achieved if all sectors of the economy are integrated into climate policy – to increase efﬁciency, ﬁnd substitutes for carbon-intensive products and services and access low-carbon energy sources. The objective is not only marginal reduction of carbon emissions but the low-carbon development of our societies. Low-carbon development thus has moved to the centre of climatechange policies: it became the backbone of discussions leading up to Copenhagen, was mentioned in the Copenhagen Accord, and was the basis for the subsequent submissions of proposals for NAMAs by eighty-three countries to the UN secretariat of the Framework Convention on Climate Change. This raises the question: what can countries to do develop in a low-carbon way? 237

238

*

Conclusion

Portfolio of climate policy instruments Many economic activities create signiﬁcant carbon emissions, which make it difﬁcult for governments to micromanage individual carbon policies across a wide variety of industrial, commercial, housing and transport activities. Carbon pricing allows governments to create incentives for companies and consumers to make carbon-efﬁcient choices across a diverse set of activities. Companies and consumers then have the ﬂexibility to ﬁnd the response that is most suitable for their speciﬁc circumstances. While initial evidence from the impact of the European Union emissions-trading scheme on carbon efﬁciency is encouraging, empirical evidence from energy usage in OECD countries is robust. Energy prices differ across countries owing to varying levels of natural resources as well as varying levels of tax imposed on energy consumption. Countries where energy prices are high require less energy per unit of GDP than countries with low energy prices. Another signiﬁcant aspect of climate policy is the development of low-carbon technologies. Since there are often various market failures associated with technology innovation, governments are generally expected to support research, development and, in some instances, strategic deployment of low-carbon technologies. This does not, however, imply that climate-change policy can be limited to the development and demonstration of low-carbon technologies. On the contrary, absolute emissions targets and trajectories, deﬁned by countries and supported through domestic policy frameworks, are necessary to attract private-sector interest in market opportunities for low-carbon technologies. The economic literature has identiﬁed a large set of barriers to the implementation of apparently viable energy-efﬁciency and low-carbon opportunities. Changes to regulatory and institutional frameworks are needed to remove these barriers. Targeted policies are also required to provide the necessary information and to raise consumer awareness about, and management attention to, the carbon implications of

Conclusion

*

239

decisions. Without these complementary measures, it is likely that carbon pricing would not be as effective as it could. Conversely, it would be difﬁcult to implement far-reaching regulation on carbon efﬁciency without a carbon-price signal. Consumers and companies might, for example, try to circumvent regulations and avoid carbon-efﬁciency investments if they do not beneﬁt from savings on carbon costs. Thus, carbon pricing and regulation are two complementary pillars of climate policy. Implementation of climate policy to facilitate low-carbon development is politically challenging, because the intent is actually to change economic activity – from carbon-intensive to low-carbon products and services. This is illustrated through the example of carbon pricing. The prices for carbon-intensive products and services have to rise if consumer and corporate choices are to change. Most of this price increase reﬂects additional costs of buying carbon allowances or paying carbon taxes. The use of the revenue by government can determine who wins and who loses from the introduction of carbon pricing. It will therefore be a politically sensitive question that requires carefully balanced compensation schemes. If major consumer and industry groups consider a scheme to be fair and equitable, this increases political support and regulatory stability.

Effective implementation of domestic policy instruments Economic models have succeeded in projecting low-carbon development, and policy analysts have collected large databases of the policy instruments that are available to encourage, incentivise or require government and private actors to make low-carbon choices. The challenge is in the choice of the suitable instruments, effective design and, most of all, implementation in the existing institutional and political setting. The European emissions-trading scheme is evidence that it is possible to implement an ambitious policy instrument. Yet that experience also highlighted difﬁculties (such as the distortions

240

*

Conclusion

from repeated free-allowance allocation) and how those difﬁculties have been addressed. A policy instrument is only effective if it impacts on decision makers. Of particular importance for low-carbon development are investment and innovation choices. Oil and technology companies have long-term perspectives on the market share of fuels, technologies and their products. Policy decisions can inﬂuence their expectations and thus investment decisions if they provide a credible longer-term policy framework. One criterion to determine the credibility of an emissions target is whether it is supported by a suitable mix of policy instruments. A cap-and-trade scheme can contribute to credibility if the carbon price for 2020 can adjust so as to ensure that the target will be met. In contrast to such long-term perspectives, investors in individual projects typically focus on the return over the initial ﬁve to ten years. Their investment decisions are normally based on existing legislation and regulations, and often take current carbon prices and their volatility as representative of future prices. For such investors, stable and predictable carbon prices reduce uncertainties and risk, thereby facilitating access to cheaper capital. The discrepancies between these two perspectives illustrate the challenge of policy design to balance different requirements, in this case the short- and long-term perspectives. Short-term emissionsreduction targets are typically based on assessments of available technologies and might lend themselves to carbon taxes. Long-term targets reﬂect the need to avoid extreme climate impacts and the expectations about technology improvements and innovation; they are likely to be more suitable for emissions trading. The implementation of emissions trading offers ﬂexibilities to bridge this gap. Investors with shorter-term perspectives can be protected from the risk of extremely low carbon prices during the initial years of a cap-and-trade scheme – for example, through a reserve price in auctions or government-issued put options. The credibility of future targets increases with a consistent and widely

Conclusion

*

241

shared vision of the trajectory towards de-carbonisation and implementation of the necessary policy instruments. A cap-and-trade scheme can thus play an important part in this policy mix, as it ensures the carbon price adjusts to technology cost and fuel prices, so as to deliver the target. The investment discussion opens two broader questions on lowcarbon ﬁnance. First, what is necessary to attract investors like pension funds to ﬁnance companies with low-carbon strategies and low-carbon projects? It is widely agreed that clear, credible and longterm policies are essential, but what does this imply for their detailed design, and is it necessary and effective to provide additional government support to enhance the credibility? Second, how can policies be designed, implemented and communicated so as to alert investors to the risks of carbon-ignorant or even carbon-dependent business strategies? Investors aware of the risks can support management in changing business strategies. Where this fails, risk-averse investors, like pension funds, can withdraw from companies with signiﬁcant carbon risks. Once pension funds have withdrawn from a company, the political weight of management declines, giving government a better handle on designing incentive schemes to encourage a shift to a low-carbon strategy.

How can international co-operation enhance domestic action? The global target of limiting temperature increases to 2 °C can only be achieved with ambitious mitigation actions by the majority of countries. The case for mitigation action by a country does not, however, depend on the behaviour of other countries. Countries are responsible for their own emissions, and can pursue actions to tackle these emissions together with other countries and, if necessary, in advance of international co-operation agreements or above the level of ambition agreed internationally.

242

*

Conclusion

This does not imply that international climate co-operation does not offer many beneﬁts. It contributes to an enhanced sense of responsibility for, and motivation to reduce, emissions. It can also provide support for countries in their effective implementation of polices for low-carbon development. For example, internationally available and transparent information facilitates rapid learning about the success of individual policy instruments, and thus aids policy design and implementation. Climate-change mitigation is a long-term challenge, and thus requires a combination of long-term objectives translated into shortterm actions. International co-operation can provide opportunities to enhance commitment to the long-term objectives, and can also provide a platform to assess whether short-term responses are aligned with long-term targets. The levels of awareness and public support for action to mitigate climate change differ across countries. This creates two basic choices. The ﬁrst is to agree in international negotiations on a common level of ambition – probably the lowest common level – and comparable emissions-reduction targets. The attempt to be more ambitious failed in the case of Kyoto when the US Congress did not ratify the protocol negotiated by their delegation. The second choice is for governments to commit to emissions-reduction targets at different levels of ambition, reﬂecting their own level of domestic awareness and support for climate policy. What are the implications for emissions trading of these two approaches? If countries pursue climate policy at similar levels of ambition, then the net ﬁnancial transfer between countries would be limited. Trade and ﬁnancial transfers would only increase if unexpected events, like a nuclear outage, required a shift to fossil-fuelbased generation, changing demand for emissions allowances in one country. Trade thus offers the ﬂexibility to respond to unforeseen events, and would not otherwise create large, potentially politically unpopular ﬁnancial transfers between countries.

Conclusion

*

243

If countries pursue climate policy at different levels of ambition reﬂected in targets with different stringency, companies in more ambitious countries would in a joint trading scheme buy allowances from governments or companies in less ambitious countries. It would result in potentially large ﬁnancial transfers. Anticipating such transfers, developed countries are reluctant to commit to more stringent targets than other developed countries. This would be a drawback of emissions trading among developed countries with different levels of ambition for their climate policy. However, a framework that allows countries to pursue an ambitious climate policy that includes high carbon prices is preferable to waiting for a global harmonised approach. It will accelerate the development of lower-carbon technologies, infrastructure and regulatory frameworks. Countries might beneﬁt from reduced dependency on imported fossil fuels, and their companies could become international ﬁrst movers for technology and services solutions. Once other countries also increase their level of ambition, their implementation of climate policies will be faster because it can build on existing experience of policy instruments and technologies. Eventually, perhaps by 2020, the levels of ambition might converge and allow for close linking of schemes. The discussion suggests that carbon pricing might be applied at different levels across countries and sectors during the initial years of climate policy. This raises the question whether carbon-price differentials would result in leakage of emissions, from the relocation of production to other countries rather than in emissions reduction. Only twenty-four sub-sectors demonstrate cost increases from direct carbon emissions or carbon-related electricity price increases that are signiﬁcant. Transport costs and risks, customer relationships, tailored products and many other factors further reduce the impact of carbon-price differentials for these sectors and suggest that leakage is of concern for only some of these sectors. For sectors where leakage concerns remain, governments can choose from several options to create a level playing ﬁeld. Allowances can be

244

*

Conclusion

allocated at no cost to emitters or direct subsidies can be provided via state aid. This approach can protect the proﬁts of companies. However, some companies might choose to close or relocate their production and sell the allowances that they received at no cost. Simple provision of subsidies does not address leakage concerns, and thus must be conditional on the continued operation of companies. However, this conditionality creates perverse incentives, and reduces the effectiveness of the scheme. Border adjustment could ensure that the carbon price is internalised in all products, irrespective of their origin, so as to make consumers responsible for their carbon emissions. Several different designs are discussed. One possibility is to require importers into a region to pay a tariff that corresponds to the carbon costs of producing the commodity (e.g., basic steel or clinker) with the best available technology. Thus, discrimination against foreign producers is avoided to comply with WTO rules. The political sensitivity of trade measures requires careful handling, to avoid undermining co-operative efforts across countries and to reﬂect the common but differentiated nature of climate responsibilities. An international process with a clear focus on solutions to leakage concerns would be necessary, to ensure transparency and to build trust. This could also ensure that any net revenue is used to support climate policy pursuit in developing countries.

How can international support enhance action in developing countries? Developing countries are part of any solution to climate change. With growing emissions levels and the increasing impact of climate change, there is both the need and the motivation to engage these countries in active climate policy. International support for low-carbon growth in developing countries can, as with co-operation among developed countries, contribute to a political momentum and provide outside processes and mechanisms to enhance the level and effectiveness of domestic action. In

Conclusion

*

245

addition, international support can address technical, technological and ﬁnancial constraints in developing countries. This section has focused on the role of ﬁnancial mechanisms, both because of their key role and to provide the context for a discussion of the role of carbon markets. Currently, the CDM represents the main channel for ﬁnancial transfers from developed countries to support the de-carbonisation efforts of developing countries. Companies in developed countries invest in projects in developing countries and the emissions reductions achieved are credited against the emissions of the companies in their home countries. CDM projects have been developed across a range of countries and technologies, and have established stakeholders and created expertise in developing countries. Expanding the scale and scope of CDM is often discussed. In principle, a common carbon price for CDM credits is economically efﬁcient, because it means the same price for every avoided tonne of carbon emissions and ensures the lowest-cost projects are selected. However, the approach also inﬂates the required funding stream because projects with lower costs still receive the full carbon price. This will probably limit political support and the volume of total emissions reduction achieved. This raises the question as to whether future carbon-market-based mechanisms can effectively replace or complement the CDM. Sectoral trading approaches are frequently discussed but might be difﬁcult to operationalise as a low-carbon investment framework. Sectoral crediting mechanisms at government level could combine some elements of carbon markets with effective incentives for domestic policy implementation, but the time lag between action and future reward, in combination with uncertainties inherent in carbon markets, remain issues. The focus is therefore increasingly moving to public ﬁnance mechanisms. These can aim to address multiple objectives. First, they can support countries in which domestic resources are already stretched

246

*

Conclusion

by the provision of basic needs such as education, health and sometimes nutrition, let alone covering incremental costs of low-carbon choices. Second, they can tailor the ﬁnancial support to the needs of actors who bear the costs, to enhance domestic support and facilitate implementation and contribute to regulatory stability of domestic policy frameworks. Financial mechanisms have been successfully implemented in national and international, and climate- and non-climate-related circumstances. Effectively to support low-carbon development, they must be tailored as far as possible to the speciﬁc circumstances of the activity and country, while retaining a set of common rules and frameworks for the design of ﬁnancial mechanisms, in order to limit the complexity for developing and developed countries using the mechanisms.

So does it add up to an adequate response to climate change? Scientists have clearly demonstrated what level of mitigation action is necessary to limit temperature increases. Economic models and bottom-up technology assessments have demonstrated that with global action it is possible to achieve the necessary emissions reduction. The challenge turns out to be the political willingness of countries to pursue the necessary actions and to design effective policies and programmes to encourage and incentivise private actors to make them real. Copenhagen got us one step closer to solving this challenge because it resolved the ‘free rider’ issue. Prior to Copenhagen, countries and actors delayed domestic action, arguing that unilateral domestic action would allow other countries to ‘free ride’ on unilateral mitigation efforts. The associated negotiation strategy failed, and the real free riders were exposed as domestic constituencies that were pointing the ﬁnger at other countries to excuse their own inaction.

Conclusion

*

247

Countries can now take full responsibility for their emissions and pursue the necessary mitigation action. The negotiations leading up to Copenhagen increasingly focused on low-carbon development – and this seems to have become the new paradigm. It offers a positive perspective; it provides governments with the tools to act; and it allows for a tailored response to climate change, building on domestic initiative and experience and working together to share experience and facilitate rapid learning. Various international support mechanisms can facilitate co-operation among developed and developing countries. Will many actions independently pursued by countries add up to an adequate and timely response? I think this is unlikely. In the end, we all tend to miss deadlines unless there is a good incentive in place (I have two hours left in which to submit this chapter). Thus, international co-operation will be essential to assess whether humanity is on track to tackle climate change and ﬁnd response strategies. International co-operation can build on domestic effort and can enhance the effectiveness of low-carbon development of all countries that wish to participate.

References

AEA Technology Environment (2004). An Evaluation of the Air Quality Strategy. London: DEFRA. Åhman, M., D. Burtraw, J. A. Kruger and Z. Lars (2007). ‘A Ten-Year Rule to Guide the Allocation of EU Emission Allowances’. Energy Policy, 35(3): 1718–1730. Åhman, M. and K. Holmgren (2006). Harmonising New Entrant Allocation in the Nordic Energy Sectors: Current Principles and Options for EU ETS Phase II. Stockholm: IVL (Swedish Environmental Research Institute). Aldy, J. E., R. Baron and L. Tubiana (2004). Addressing Cost: The Political Economy of Climate Change. Washington, DC: Pew Center on Global Climate Change. Alic, J. A., D. S. Mowery and E. S. Rubin (2003). US Technology and Innovation Policies: Lessons for Climate Change. Arlington, VA: Pew Center on Global Climate Change. November 2003. Available at www.pewclimate.org. Andersen, M. S. T. Barker, E. Christie, P. Ekins, J. FitzGerald, J. Jilkova, J. Junankar, M. Landesmann, H. Pollitt, R. Salmons, S. Scott and S. Speck (2007). Competitiveness Effects of Environmental Tax Reforms (COMETR). University of Aarhus, National Environmental Research Institute. Andersen, M. S. and P. Ekins (2009). Carbon-Energy Taxation: Lessons from Europe. Oxford University Press. Ashton, J. and X. Wang (2004). ‘Equity and Climate: In Principle and Practice’. In Pew Center on Global Climate Change, Beyond Kyoto: Advancing the International Effort against Climate Change. Arlington, VA: Pew Center on Global Climate Change: 61–84.

248

References

*

249

Baker, T. and J. Koehler (1998). ‘Equity and Ecotax Reform in the EU: Achieving a 10 Per Cent Reduction in CO2 Emissions Using Excise Duties’. Fiscal Studies, 19(4): 375–402. Barnett, J., S. Dessai and M. Webber (2004). ‘Will OPEC Lose from the Kyoto Protocol?’. Energy Policy, 32: 2077–2088. Baron, R. and J. Ellis (2006). Sectoral Crediting Mechanisms for Greenhouse Gas Mitigation: Institutional and Operational Issues. Paris: IEA/OECD. Baron, R., J. Reinaud, M. Genasci and C. Philibert (2007). Sectoral Approaches to Greenhouse Gas Mitigation: Exploring Issues for Heavy Industry. Paris: OECD/ IEA. Barrett, S. (2003). Environment and Statecraft. Oxford University Press. Bazilian, M., H. de Coninck, M. Radkac, S. Nakhooda, W. Boyd, I. McGill, A. Amin, F. von Malmborg, J. Uosukainen, R. Bradley and R. Bradley (2008). Considering Technology Within the UN Climate Change Negotiations. ECN-E–08-077. Available at www.ceem.unsw.edu.au/content/userDocs/ ECNtechnologyUN.pdf. BBC (2007). ‘Most Ready for “Green Sacriﬁces”’. Available at http://news.bbc. co.uk/1/hi/in_depth/7075759.stm. (2009). ‘G8 Leaders to Set Emissions Goals’. Available at http://news.bbc.co. uk/2/hi/science/nature/8135261.stm. (2010). ‘Traﬁgura Found Guilty of Exporting Toxic Waste’. Available at www.bbc.co.uk/news/world-africa-10735255. Bell, W. and J. Drexhage (2005). Climate Change and the International Carbon Market. Winnipeg: International Institute for Sustainable Development. Bergek, A. and S. Jacobson (2003). ‘The Emergence of a Growth Industry: A Comparative Analysis of the German, Dutch and Swedish Wind Turbine Industries’. In J. Metcalfe and U. Canter (eds), Change, Transformation, and Development. Heidelberg and New York: Physica-Verlag. Bernstein, M. A. and J. Grifﬁn (2005). Regional Differences in the Price Elasticity of Demand for Energy. Santa Monica, CA: RAND Corporation. Black, R. and H. White (2004). Targeting Development: Critical Perspectives on the Millennium Development Goals. London: Routledge. Bodansky, D. (2004). ‘Climate Commitments: Assessing the Options’. In Pew Center on Global Climate Change, Beyond Kyoto: Advancing the International Effort against Climate Change. Arlington, VA: Pew Center on Global Climate Change: 37–59. Bohi, D. R. (1981). Analyzing Demand Behaviour: A Study of Energy Elasticities. Baltimore, MD: Johns Hopkins Press. Bohi, D. R. and M. Zimmerman (1984). ‘An Update of Econometric Studies of Energy Demand’. Annual Review of Energy and the Environment, 9: 105–154.

250

*

References

Böhringer, C. and A. Lange (2005). ‘On the Design of Optimal Grandfathering Schemes for Emission Allowances’. European Economic Review, 49(8): 2041–2055. Bosi, M. and J. Ellis (2005). Exploring Options for ‘Sectoral Crediting Mechanism’. COM/ENV/EPOC/IEA/SLT. Paris: OECD. Brons, M., P. Nijkamp, E. Pels and P. Rietveld (2006). A Meta-analysis of the Price Elasticity of Gasoline Demand: A System of Equations Approach. Amsterdam: Tinbergen Institute Discussion Paper. Bruvoll, A. and B. M. Larsen (2004). ‘Greenhouse Gas Emissions in Norway: Do Carbon Taxes Work?’. Energy Policy, 32(4): 493–505. Burniaux, J. M. and J. O. Martins (2000). Carbon Emission Leakages: A General Equilibrium View. Paris: OECD Economics Department Working Papers. Burtraw, D., K. Palmer and D. Kahn (2005). Allocation of CO2 Emissions Allowances in the Regional Greenhouse Gas Cap-and-Trade Program. RFF Discussion Paper (05–25). Washington, DC: RFF. Caldeira, K., A. K. Jain and M. I. Hoffert (2003). ‘Climate Sensitivity Uncertainty and the Need for Energy Without CO2 Emission’. Science, 299(5615): 2052–2054. Carlson, C., D. Burtraw, M. Cropper and K. Palmer (2000). ‘SO2 Control by Electric Utilities: What Are the Gains from Trade?’. Journal of Political Economy, 108(6): 1292–1326. Castro, P. and A. Michaelowa (2008). Empirical Analysis of Performance of CDM Projects. Cambridge: Climate Strategies Report. Charnovitz, S. (2004). ‘Trade and Climate: Potential Conﬂicts and Synergies’. In Pew Center on Global Climate Change, Beyond Kyoto: Advancing the International Effort against Climate Change. Arlington, VA: Pew Center on Global Climate Change: 141–170. Chas-Amil, M. L. and J. Buongiorno (2000). ‘The Demand for Paper and Paperboard: Econometric Models for the European Union’. Applied Economics, 32: 987–999. Coase, R. (1960). ‘The Problem of Social Cost’. Journal of Law and Economics, 3(1): 3–44. Colombier, M. and K. Neuhoff (2008). ‘Sectoral Emission Agreements: Can They Address Leakage?’. Environmental Policy and Law, 38(3): 161–166. Concorde East (2004). Mercury Flows in Europe and the World: The Impact of Decommissioned Chlor-Alkali Plants. Report for the European Commission. Brussels: Directorate General for Environment. Congressional Budget Ofﬁce (2000). Who Gains and Who Pays under CarbonAllowance Trading? The Distributional Effects of Alternative Policy Designs. Washington, DC: CBO Publications.

References

*

251

Considine, T. (1991). ‘Economic and Technological Determinants of the Material Intensity of Use’. Land Economics, 67(1): 99–115. Copeland, B. R. and M. S. Taylor (1994). ‘North–South Trade and the Environment’. Quarterly Journal of Economics, 109(2): 755–787. Cornwall, A. and J. Creedy (1996). ‘Carbon Taxation, Prices and Inequality in Australia’. Fiscal Studies, 17(3): 21–38. Coudouel, A., J. Hentschel and Q. Wodon (2002). Poverty Measurement and Analysis: A Sourcebook for Poverty Reduction Strategies. Washington, DC: World Bank. Crocker, T. D. (1966). ‘The Structuring of Atmospheric Pollution Control Systems’. In H. Woloyin (ed.), The Economics of Air Pollution. New York: W. W. Norton: 61–86. Cust J., K. Grant, I. Iliev and K. Neuhoff (2008). International Co-operation for Innovation and Use of Low-carbon Energy Technology. Cambridge: Climate Strategies Working Paper. Available at www.climatestrategies. org. DeCanio, S. (1998). ‘The Efﬁciency Paradox: Bureaucratic and Organizational Barriers to Proﬁtable Energy-saving Investments’. Energy Policy, 26(5): 441–458. DEFRA (2007). ‘Consultation on Proposed UK Auction Design for Use during Phase II: EU Emissions Trading Scheme’. London: DEFRA. Demailly, D. and P. Quirion (2006). ‘CO2 Abatement, Competitiveness and Leakage in the European Cement Industry under the EU ETS: Grandfathering vs. Output-based Allocation’. Climate Policy, 6(1). (2008). ‘European Emission Trading Scheme and Competitiveness: A Case Study on the Iron and Steel Industry’. Energy Economics, 30(4): 2009–2027. Dennerlein, R. K.-H. (1990). Energieverbrauch privater Haushalte. Augsburg: Maro-Verlag. DTLR (2001). Local Public Service Agreements: New Challenges. London: Department for Transport, Local Government and the Regions. Edenhofer, O. and K. Lessmann (2005). ‘Climate Policy and Energy Security’. In F. Müller and A. Riechel (eds), Options for Future Climate Policy: Transatlantic Perspectives. Berlin: German Institute for International and Security Affairs: 111–123. Edenhofer, O., K. Lessmann, C. Kemfert, M. Grubb and J. Köhler (2006). ‘Induced Technological Change: Exploring its Implications for the Economics of Atmospheric Stabilization’. Energy Journal (Special Issue: Endogenous Technological Change and the Economics of Atmospheric Stabilization): 57–122.

252

*

References

Ellerman, D. (2006). ‘New Entrant and Closure Provisions: How Do They Distort?’. Cambridge, MA: Massachusetts Institute of Technology, Center for Energy and Environmental Policy Research. Ellerman D., J. F. Convery and C. de Perthuis (2010). Pricing Carbon: The European Union Emission Trading Scheme. Cambridge University Press. Ellerman, D. and P. L. Joskow (2008). The European Union’s Emissions Trading System in Perspective. Arlington, VA: Pew Center on Global Climate Change. Ellerman, D., P. L. Joskow, R. Schmalensee, J. P. Montero and E. M. Bailey (2000). Markets for Clean Air: The US Acid Rain Program. Cambridge University Press. Enevoldsen, M. (2005). The Theory of Environmental Agreements and Taxes: CO2 Policy Performance in Comparative Perspective. Northampton, MA: Edward Elgar. Enkvist, P., T. Nauclér and J. Rosander (2007). ‘A Cost Curve for Greenhouse Gas Reduction: A Global Study of the Size and Cost of Measures to Reduce Greenhouse Gas Emissions Yields Important Insights for Businesses and Policy Makers’. McKinsey Quarterly, 1. Espey, M. (1998). ‘Gasoline Demand Revisited: An International Meta-analysis of Elasticities’. Energy Economics, 20: 273–295. European Commission (1992). Proposal for a Council Directive Introducing a Tax on Carbon Dioxide Emissions and Energy. COM (92) 226 ﬁnal. Brussels: Commission of the European Communities. (2005). Doing More with Less: Green Paper on Energy Efﬁciency. Luxembourg: Ofﬁce for Ofﬁcial Publications of the European Communities. (2006). Energy and Transport in Figures 2006. Luxembourg: Ofﬁce for Ofﬁcial Publications of the European Communities. (2007). Imposing a Unilateral Carbon Constraint on European Energy-intensive Industries and its Impact on their International Competitiveness: Data and Analysis. Brussels: Directorate General for Economic and Financial Affairs of the European Commission. Available at http://ec.europa.eu/ economy_ﬁnance/publications/economicpapers_en.htm. (2008a) Commission Staff Working Document Accompanying Document to the Proposal for a Directive of the European Parliament and of the Council Amending Directive 2003/87/EC so as to Improve and Extend the EU Greenhouse Gas Emission Allowance Trading System. Brussels. 23 January 2008. (2008b). Proposal for a Decision of the European Parliament and the Council on the Effort of Member States to Reduce their Greenhouse Gas Emissions to Meet the Community’s Greenhouse Gas Emissions Reduction Commitments up to 2020. Presented by the Commission in January 2008.

References

*

253

(2009a). Directive 2009/29/EC of the European Parliament and of the Council of 23 April 2009 Amending Directive 2003/87/EC so as to Improve and Extend the Greenhouse Gas Emission Allowance Trading Scheme of the Community. Available at http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ: L:2009:140:0063:0087:en:PDF. (2009b). Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the Promotion of the Use of Energy from Renewable Sources and Amending and Subsequently Repealing Directives 2001/77/EC and 2003/ 30/EC. European Council (1996). Directive 96/61/EC: Concerning Integrated Pollution and Control. Brussels: IPPC Directive. European Energy Exchange (2010). EEX 2008 Figures. Leipzig: European Energy Exchange. European Investment Bank (2007). Evaluation of EIF Funding of Venture Capital Funds – EIB/ETF Mandate. Available at www.eib.org/attachments/ev/ ev_eif_funding_en.pdf. European Union (1992). Maastricht Treaty. European Union. Felder, S. and T. F. Rutherford (1993). ‘Unilateral CO2 Reductions and Carbon Leakage: The Consequences of International Trade in Oil and Basic Materials’. Journal of Environmental Economics and Management, 25: 162–176. Filippini, M. (1999). ‘Swiss Residential Demand for Electricity’. Applied Economic Letters, 6(8): 533–538. Fischer, C. and R. Morgenstern (2006). ‘Carbon Abatement Costs: Why the Wide Range of Estimates?’. Energy Journal, 27(2). Foxon, T. J. (2003). Inducing Innovation for a Low-carbon Future: Drivers, Barriers and Policies. London: The Carbon Trust. Fuss, M. A. (1977). ‘The Demand for Energy in Canadian Manufacturing’. Journal of Econometrics, 5. Gallaway, M., C. McDaniel and S. Rivera (2003). ‘Short-run and Long-run Industry-level Estimates of US Armington Elasticities’. North American Journal of Economics and Finance, 14(1): 49–68. Gboney, W. (2009). ‘Policy and Regulatory Framework for Renewable Energy and Energy Efﬁciency Development in Ghana’. Climate Policy, 9: 508–516. GCI (1996). Draft Proposals for a Climate Change Protocol Based on Contraction and Convergence: A Contribution to Framework Convention on Climate Change. Ad Hoc Group on the Berlin Mandate, 6 September 1996. Graham, D. and S. Glaister (2002). Review of Income and Price Elasticities of Demand for Road Trafﬁc. London: Centre for Transport Studies, Imperial College. Graichen, V., K. Schumacher, F. C. Matthes, L. Mohr, V. Duscha, J. Schleich and J. Diekmann, 2008. Impacts of the EU Emissions Trading Scheme on the

254

*

References

Industrial Competitiveness in Germany, Research Report 3707 41 501. UBAFB 001177. Dessau-Rosslau: Umweltbundesamt. Grant, K. (2009). ‘Concentrated Solar Power in South Africa’. Climate Policy, 9(5): 544–552. Grubb, M. (2004a). ‘Technology Innovation and Climate Change Policy: An Overview of Issue and Options’. Keio Economic Studies, 41(2): 103–132. (2004b) ‘The Economics of the Kyoto Protocol’. In A. D. Owen and N. Hanley (eds), The Economics of Climate Change. London: Routledge. Grubb, M., T. Chauis and M. Ha-Duong (1995). ‘The Economics of Changing Course: Implications of Adaptability and Inertia for Optimal Climate Policy’. Energy Policy, 23(4): 1–14. Grubb, M., J. Köhler and D. Anderson (2002). ‘Induced Technical Change in Energy and Environmental Modelling: Analytical Approaches and Policy Implications’. Annual Review of Energy and the Environment, 27: 271–308. Grubb, M. and D. Newbery (2007). Pricing Carbon for Electricity Generation: National and International Dimensions. EPRG Working Paper 07/22. Cambridge: Electricity Policy Research Group. Grubb, M. and R. Vigotti (1997). Renewable Energy Strategies for Europe: Electricity Systems and Primary Energy Sources. London: Royal Institute of International Affairs Energy and Environmental Programme. Guo, J., C. J. Hepburn, R. S. J. Tol and D. Anthoff (2006). ‘Discounting and the Social Cost of Carbon: A Closer Look at Uncertainty’. Environmental Science and Policy, 9: 205–216. Guzman, A. (2008). How International Law Works: A Rational Choice Theory. Oxford and New York: Oxford University Press. Hamilton, K. and M. Kenber (2006). Business Views on International Climate and Energy Policy. London: UK Business Council for Sustainable Energy. Hammitt, J. K. (2000). ‘Are the Costs of Proposed Environmental Regulations Overestimated? Evidence from the CFC Phaseout’. Environmental and Resource Economics, 16: 281–301. Hanly, M., J. Dargay and P. Goodwin (2002). Review of Income and Price Elasticities in the Demand for Road Trafﬁc. Report 2002/13. ESRC Transport Studies Unit, Centre for Transport Studies, University College London. Harrington, W., R. D. Morgenstern and P. Nelson (2000). ‘On the Accuracy of Regulatory Cost Estimates’. Journal of Policy Analysis and Management, 19(2): 297–322. Harrison, D. J. and D. B. Radov (2002). Evaluation of Alternative Initial Allocation Mechanisms in a European Union Greenhouse Gas Emissions Allowance Trading Scheme. National Economic Research Associates Report Prepared for DG Environment, European Commission.

References

*

255

Hassett, K. A. and G. E. Metcalf (1993). ‘Energy Conservation Investment: Do Consumers Discount the Future Correctly?’. Energy Policy, 21(6): 710–716. He, G. and R. K. Morse (2010). Making Carbon Offsets Work in the Developing World: Lessons from the Chinese Wind Controversy. Working Paper 90. Stanford Program on Energy and Sustainable Development. Hekman, J. S. (1978). ‘An Analysis of the Changing Location of Iron and Steel Production in the Twentieth Century’. American Economic Review, 68(1): 123–133. Helm, D., C. Hepburn and R. Mash (2003). ‘Credible Carbon Policy’. Oxford Review of Economic Policy, 19(3): 438–450. Henkin, L. (1979). How Nations Behave. 2nd edn. New York: Columbia University Press. Hepburn, C., M. Grubb, K. Neuhoff, F. Matthes and M. Tse (2006). ‘Auctioning of EU ETS Phase II Allowances: How and Why?’. Climate Policy (Special Issue: Emissions Allocation, Incentives and Industrial Competitiveness under the EU Emissions Trading Scheme), 6(1): 137–160. HMRC (2010). Research and Development (R&D) Relief for Corporation Tax. Available at www.hmrc.gov.uk. August 2010. Hoel, M. and L. Karp (2001). ‘Taxes and Quotas for a Stock Pollutant with Multiplicative Uncertainty’. Journal of Public Economics, 82: 91–114. Hoffert, M. I., K. Caldeira, A. K. Jain, E. F. Haites, L. D. D. Harvey, S. D. Potter, M. E. Schlesinger, S. H. Schneider, R. G. Watts, T. M. L. Wigley and D. J. Wuebbles (1998). ‘Energy Implications of Future Stabilization of Atmospheric CO2 Content’. Nature, 395(6705): 881–884. Hope, C. and D. Newbery (2008). ‘Calculating the Social Cost of Carbon’. In M. Grubb, T. Jamasb and M. Pollitt (eds), Delivering a Low Carbon Electricity System: Technologies, Economics and Policy. Cambridge University Press. Hourcade, J.-C., D. Demailly, K. Neuhoff and M. Sato (2007). Differentiation and Dynamics of EU ETS Industrial Competitiveness Impacts. Final Report. Cambridge: Climate Strategies. ICAP (2007). International Carbon Action Partnership. Berlin. www.icapcarbonaction.com. IEA (2007). R&D Budget 2007. Paris: IEA. IEA and OECD (2002). World Energy Outlook 2002. Paris: OECD/IEA. (2007). World Energy Outlook 2007: China and India Insights. Paris: OECD/IEA. (2009). World Energy Outlook 2009. Paris: OECD/IEA. IISI (2007). A Policy to Reduce Steel-related Greenhouse Gas Emissions. Brussels: International Iron and Steel Institute Policy Statement. Iliev, I. and K. Neuhoff (2009). Intellectual Property: Cross-licensing, Patent Pools and Cooperative Standards as a Channel for Climate Change Technology

256

*

References

Co-operation. Cambridge: Climate Strategies Paper September 2009. Available at www.climatestrategies.org/component/reports/category/43/ 216.html. IPCC (2007). Climate Change 2007: Mitigation of Climate Change. IPCC Fourth Assessment Report. Cambridge and New York: Cambridge University Press. Ismer, R. and K. Neuhoff (2006). Commitments through Financial Options: A Way to Facilitate Compliance with Climate Change Obligations. EPRG Working Paper: EPRG0625. Cambridge: Electricity Policy Research Group. (2007). ‘Border Tax Adjustments: A Feasible Way to Support Stringent Emission Trading’. European Journal of Law and Economics, 24: 137–164. Jacoby, H. D. and A. D. Ellerman (2004). ‘The Safety Valve and Climate Policy’. Energy Policy, 32: 481–491. Jacquier-Roux, V. and B. Bourgeois (2002). ‘New Networks of Technological Creation in Energy Industries: Reassessment of the Roles of Equipment Suppliers and Operators’. Technology Analysis and Strategic Management, 14(4): 399–417. Jans, I. and D. I. Rosenbaum (1997). ‘Multimarket Contact and Pricing: Evidence from the US Cement Industry’. International Journal of Industrial Organization, 15: 391–412. Jensen, S. G. (2004). ‘Reducing Costs of Emerging Renewable Energy Technologies: An Analysis of the Dynamic Development with Wind Power as Case Study’. International Journal of Energy Technology and Policy, 2(1/2): 179–202. Jones, C. T. (1996). ‘A Pooled Dynamic Analysis of Interfuel Substitution in Industrial Energy Demand by the G-7 Countries’. Applied Economics, 28: 815–821. Karlson, S. H. (1983). ‘Modeling Location and Production: An Application to US Fully Integrated Steel Plants’. Review of Economics and Statistics, 65(1): 41–50. Karnøe, P. (1990). ‘Technological Innovation and Industrial Organization in the Danish Wind Industry’. Entrepreneurship and Regional Development, 2(2): 105–124. Klemperer, P. (2002). ‘What Really Matters in Auction Design’. Journal of Economic Perspectives, 16(1): 169–189. Kuik, O. and R. Gerlagh (2003). ‘Trade Liberalization and Carbon Leakage’. Energy Journal, 24(3): 97–120. La Cour, L. and H. P. Mollgaard (2002). ‘Market Domination: Tests Applied to the Danish Cement Industry’. European Journal of Law and Economics, 14: 99–127.

References

*

257

Labandeira, X. and J. Labeaga (1999). ‘Combining Input-Output Analysis and Mirco-simulation to Assess the Effects of Carbon Taxation on Spanish Households’. Fiscal Studies, 20 (3): 305–320. Lafferty, R., D. Hunger, J. Ballard, G. Mahrenholz, D. Mead and D. Bandera (2001). Demand Responsiveness in Electricity Markets. Ofﬁce of Markets, Tariffs and Rates. Washington, DC: Federal Energy Regulatory Commission. Lee, B., I. Iliev and F. Preston (2009). Who Owns Our Low Carbon Future? London: Chatham House. Lester, S. and K. Neuhoff (2009). ‘Understanding the Roles of Policy Targets in National and International Governance’. Climate Policy, 9(5): 464–480. Lord, R. A. and R. K. Farr (2003). ‘Collusion and Financial Leverage: An Analysis of the Integrated Mill Steel Industry’. Financial Management, Spring: 127–148. Lundvall, B. (1992). ‘National Systems of Innovation: Towards a Theory of Innovation and Interactive Learning’. London and New York: Pinter Publishers. MacKinnon, J. G. and N. D. Olewiler (1980). ‘Disequilibrium Estimation of the Demand for Copper’. Bell Journal of Economics, 11(1): 197–211. Malach, V. W. (1957). ‘Elasticity of Demand for Canadian Exports’. Review of Economics and Statistics, 39(1): 23–30. Mannaerts, H. J. B. M. (2000). ‘STREAM: Substance Throughput Related to Economic Activity Model’, Research Memorandum 165. The Hague: Netherlands Bureau for Economic Policy Analysis. Margolis, R. M. and D. M. Kammen (1999). ‘Evidence of Under-investment in Energy R&D in the United States and the Impact of Federal Policy’. Energy Policy, 27: 575–584. Marsiliani, L. and T. Renström (2000). ‘Time Inconsistency in Environmental Policy: Tax Earmarking as a Commitment Solution’. Economic Journal, 462: 123–138. Martin, K., P. Joskow and D. Ellerman (2007). Time and Location Differentiated NOX Control in Competitive Electricity Markets Using Cap-and-Trade Mechanisms. Massachusetts Institute of Technology, Center for Energy and Environmental Policy Research: Working Paper-2007–004. Matthes, F. (2007). Zuteilungsgesetz 2008–2012 für den EU-Emissionshandel. Stellungnahme zu den Fragenkatalogen der Fraktionen CDU/CSU, SPD, FDP, DIE LINKE sowie BÜNDNIS 90/DIE GRÜNEN zur Anhörung des Ausschusses für Umwelt, Naturschutz und Reaktorsicherheit des Deutschen Bundestages. Berlin. (2010a). Der Instrumenten-Mix einer ambitionierten Klimapolitik im Spannungsfeld von Emissionshandel und anderen Instrumenten. Freiburg: Öko-Institut.

258

*

References

(2010b). Greenhouse Gas Emission Trading and Complementary Policies. Developing a Smart Mix for Ambitious Climate Policies, Report for German Environment Ministry, Berlin. Matthes, F., V. Graichen, R. O. Harthan and J. Repenning (2006). Auswirkung verschiedener Allokationsregeln auf Investitionen im Strommarkt. www.oeko.de. Matthes, F., V. Graichen and J. Repenning (2005). The Environmental Effectiveness and Economic Efﬁciency of the European Union Emissions Trading Scheme: Structural Aspects of Allocation. WWF. Freiburg: Öko-Institut. McKinsey and Company (2010). Version 2.1 of the Global Greenhouse Gas Abatement Cost Curve. McKinsey and Company and Ecofys (2006). Report on International Competitiveness. EU ETS Review. Metcalf, G. (1999). ‘A Distributional Analysis of Green Tax Reforms’. National Tax Journal, 52(4): 655–682. Michaelowa, A., S. Butzengeiger and M. Jung (2005). ‘Graduation and Deepening: An Ambitious Post-2012 Climate Policy Scenario’. International Environmental Agreements: Politics, Law and Economics, 5(1): 25–46. Morgenstern, R. and W. Pizer (2007). Reality Check: The Nature and Performance of Voluntary Environmental Programs in the United States, Europe, and Japan. Washington, DC. Resources for the Future Press. Myers, S. (1984). ‘The Capital Structure Puzzle’. Journal of Finance (July): 595–562. Nemet, G. F. and E. Baker (2009). ‘Demand Subsidies Versus R&D: Comparing the Uncertain Impacts of Policy on a Pre-commercial Low-carbon Energy Technology’. Energy Journal, 30(4): 49–80. Neuhoff, K., S. Fankhauser, E. Guerin, J. C. Hourcade, H. Jackson, R. Rajan and J. Ward (2010). Structuring International Financial Support for Climate Change Mitigation in Developing Countries. DIW Working Paper Series. SSRN eLibrary. Berlin: Deutsches Institut für Wirtschaftsforschung (DIW). Available at http://ssrn.com/paper=1596079. Neuhoff, K., F. Ferrario, M. Grubb, E. Gabel and K. Keats (2006). ‘Emission Projections 2008–2012 versus National Allocation Plans II’. Climate Policy, 6(4): 395–410. Neuhoff, K., K. Keats and M. Sato (2006). ‘Allocation, Incentives and Distortions: The Impact of EU ETS Emissions Allowance Allocations to the Electricity Sector’. Climate Policy, 6(1): 73–91. Neuhoff, K., J. Lossen, G. Nemet, M. Sato and K. Schumacher (2007). ‘The Role of the Supply Chain for Innovation: The Example of Photovoltaic Solar Cells’. EPRG Working Paper 07/32. Cambridge: Electricity Policy Research Group.

References

*

259

Neuhoff, K., K. Rogge, J. Schleich, J. Sijm, A. Tuerk, C. Kettner, N. Walker, M. Åhman, R. Betz, J. Cludius, F. Ferrario, K. Holmgren, G. Pal, M. Grubb and F. Matthes (2006). ‘Implications of Announced Phase 2 National Allocation Plans for the EU ETS’. Climate Policy, 6(5): 411–422. Neuhoff, K. and A. Vasa (2010). The Role of CDM Post-2012. Cambridge: Climate Strategies Working Paper. Newbery, D. M. G. (2003). ‘Sectoral Dimensions of Sustainable Development: Energy and Transport’. Economic Survey of Europe, 2: 73–93. Newbery, D. M. G. and J. E. Stiglitz (1981). The Theory of Commodity Price Stabilization: A Study in the Economics of Risk. Oxford: Clarendon Press. Newell, R. and W. Pizer (2003). ‘Regulating Stock Externalities Under Uncertainty’. Journal of Environmental Economics and Management, 45: 416–432. Newell, R., W. Pizer and J. Zhang (2005). ‘Managing Permit Markets to Stabilize Prices’. Environmental and Resource Economics, 31(2): 133–157. Norberg-Bohm, V. (2000). ‘Creating Incentives for Environmentally Enhancing Technological Change: Lessons from Thirty Years of US Energy Technology’. Technological Forecasting and Social Change, 65: 125–148. Nordhaus, W. D. (2006). ‘After Kyoto: Alternative Mechanisms to Control Global Warming’. American Economic Review, 96(2): 31–34. Parry, I. (2004). ‘Are Emissions Permits Regressive?’. Journal of Environmental Economics and Management, 47: 364–387. Paul, A., D. Burtraw and K. L. Palmer (2010). ‘Compensation for Electricity Consumers Under a US CO2 Emissions Cap’. In Vivek Ghosal (ed.), Reforming Rules and Regulations. Cambridge, MA: Massachusetts Institute of Technology. Pershing, J. and F. Tudela (2004). ‘A Long-term Target: Framing the Climate Effort’. In Pew Center on Global Climate Change, Beyond Kyoto: Advancing the International Effort against Climate Change. Arlington, VA: Pew Center on Global Climate Change: 11–36. Philibert, C. (2000). ‘How Emissions Trading Could Beneﬁt Developing Countries’. Energy Policy, 28: 947–956. (2004). Technology Innovation, Development and Diffusion. OECD and IEA Information Paper, COM/ENV/EPOC/IEA/SLT. Philibert, C. and J. Reinaud (2004). Emissions Trading: Taking Stock and Looking Forward. COM/ENV/EPOC/IEA/SLT, OECD/IEA. Pigou, A. C. (1920). The Economics of Welfare. London: Macmillan. Pizer, W. A. (2002). ‘Combining Price and Quantity Controls to Mitigate Global Climate Change’. Journal of Public Economics, 85(3): 409–434.

260

*

References

Pizer, W. A. and R. J. Kopp (2005). ‘Calculating the Costs of Environmental Regulation’. In K.-G. Maler and J. R. Vincent, Handbook of Environmental Economics. Vol. 3. Amsterdam: Elsevier: 1308–1344. Poterba, J. M. (1991). ‘Tax Policy to Combat Global Warming: On Designing a Carbon Tax’. In R. Dornbusch and J. M. Poterba (eds), Global Warming: Economic Policy Responses to Global Warming. Cambridge, MA: Massachusetts Institute of Technology Press. Reedman, L., P. Graham and P. Coombes (2006). ‘Using a Real-Options Approach to Model Technology Adoption Under Carbon Price Uncertainty: An Application to the Australian Electricity Generation Sector’. Economic Record, 82(1): 64–73. Reinaud, J. (2003). Emissions Trading and its Possible Impacts on Investment Decisions in the Power Sector. Paris: IEA Information Paper. (2004). Emissions Trading and its Possible Impacts: The Case of Energy-intensive Industries in the EU ETS. ET Workshop 5 October 2004. Roller, L.-H. and F. Steen (2005). ‘On the Workings of a Cartel: Evidence from the Norwegian Cement Industry’. American Economic Review, 96: 321–338. Rosendahl, K. (2004). ‘Cost-effective Environmental Policy: Implications of Induced Technological Change’. Journal of Environmental Economics and Management, 48(3): 1099–1121. Ryan, S. (2005). The Costs of Environmental Regulation in a Concentrated Industry. Cambridge, MA: Massachusetts Institute of Technology, Centre for Energy and Environmental Policy Research 05–010. Sagar A., C. Bremner and M. Grubb (2008). ‘Public–Private Roles and Partnership for Innovation and Technology Transfer’. In Climate Change: Technology Development and Technology Transfer. UN Department of Economics and Social Affairs. Available at www.un.org/esa/sustdev/sdissues/energy/op/beijing_hlccc_nov08/back_paper.pdf. Samaniego, J. and C. Figueres (2002). ‘Evolving to a Sector-based Clean Development Mechanism’. In K. Baumert, Building on the Kyoto Protocol: Options for Protecting the Climate. Washington, DC: World Resources Institute. Sandén, B. A. and C. Azar (2005). ‘Near-term Technology Policies for LongTerm Climate Targets: Economy Wide Versus Technology Speciﬁc Approaches’. Energy Policy, 33(12): 1557–1576. Schaefer, G. (1979). ‘The Demand for Newsprint: A Comment’. Canadian Journal of Economics/Revue canadienne d’Economique, 12(3): 518–522. Shiell, L. (2003). ‘Equity and Efﬁciency in International Markets for Pollution Permits’. Journal of Environmental Economics and Management, 46: 38–51.

References

*

261

Sijm, J. P. M., O. J. Kuik, M. Patel, V. Oikonomou, E. Worrell, P. Lako, E. Annevelink, G. J. Nabuurs and H. W. Elbersen (2004). Spillovers of Climate Policy: An Assessment of the Incidence of Carbon Leakage and Induced Technological Change due to CO2 Abatement Measures. Report 500036 002 (ECN report ECN-C–05–014). Bailrigg: ECN. Sijm, J. P. M., K. Neuhoff and Y. Chen (2006). ‘CO2 Cost Pass Through and Windfall Proﬁts in the Power Sector’. Climate Policy, 6(1): 49–72. Sippel, M. and K. Neuhoff (2009). ‘Lessons from Conditionality Provisions for South-North Co-operation on Climate Policy’. Climate Policy, 9(5): 481–494. Smeers, Y. (2007). Assessment of CO2 Regulations. IFRI Energy Breakfast Roundtable Brussels. Smith, S. (1992). ‘The Distributional Consequences of Taxes on Energy and the Carbon Content of Fuels’. European Economy, Special Edition 1 (1992). Smith, S. and J. Swierzbinski (2007). ‘Assessing the Performance of the UK Emissions Trading Scheme’. Environmental and Resource Economics, 37(1): 131–158. Solomon, B. (1995). ‘Global CO2 Emissions Trading: Early Lessons from the US Acid Rain Program’. Climatic Change, 30(1): 75–96. Sorrell, S. (2007). ‘The Rebound Effect: An Assessment of the Evidence for Economy-wide Energy Savings from Improved Energy Efﬁciency’. Report of UK Energy Research Centre. Spierer, C. (1988). Modélisation économétrique et perspectives à la long terme de la demande d’énergie en Suisse. Bern: Expertengruppe Energieszenarien, Schriftenreihe Nr. 17. Sreenivasamurthy, U. (2009). ‘Domestic Climate Policy for the Indian Steel Sector’. Climate Policy, 9(5): 517–528. Standard & Poor’s (2009). Drax Power Ltd. Issuer Rating Cut to ‘BB +’ On Weak UK Power Prices and Increasing Business Risk. Outlook Negative (Ratings Direct). (2010). European Companies Look to Carbon Markets and Reduction Measures to Manage their Emissions Exposure (Global Credit Portal). Steinbuks, J. and K. Neuhoff (2010). Operational and Investment Response to Energy Prices in the OECD Manufacturing Sector. Cambridge Working Paper in Economics 1015. Sterk, W. and B. Wittneben (2005). Addressing Opportunities and Challenges of a Sectoral Approach to the Clean Development Mechanism. JIKO Policy Paper 1/ 2005. Berlin: Wuppertal Institute for Climate, Environment and Energy. Stern Review (2006). The Economics of Climate Change. Cambridge: HM Treasury. Symons, E., J. Proops and G. Gay (1994). ‘Carbon Taxes, Consumer Demand and Carbon Dioxide Emissions: A Simulation Analysis for the UK’. Fiscal Studies, 15 (2): 19–43.

262

*

References

Symons, E., S. Speck and J. Proops (2002). ‘The Distributional Effects of Carbon and Energy Taxes: The Cases of France, Spain, Italy, Germany and the UK’. European Environment Agency, 12: 200–212. Tol, R. S. J. (2003). The Marginal Cost of Carbon Dioxide Emissions: An Assessment of the Uncertainties. Working Paper FNU 19. University of Hamburg, Centre for Marine and Climate Research. UNEP (2009), Catalysing Low-carbon Growth in Developing Economies: Public Finance Mechanisms to Scale Up Private Sector Investment in Climate Solutions. London: Vivid Economics. October. UNEP Risoe (2010). UNEP Risoe CDM/JI Pipeline Analysis and Database. Available at http://cdmpipeline.org/. UNFCCC (1997). Kyoto Protocol to the United Nations Framework Convention on Climate Change. Adopted by Consensus at the Third Session of the Conference of the Parties (COP3). (1999). Report on the In-depth Review of the Second National Communication of Germany. FCCC/IDR.2/DEU 24. (2009a). Chairs Proposed Draft Text on the Outcome of the ad hoc Working Group on Long-term Cooperative Action under the Convention. Available at http://unfccc.int/resource/docs/2009/cop15/eng/l07.pdf. (2009b). Recommendations on Future Financing Options for Enhancing the Development, Deployment, Diffusion and Transfer of Technologies under the Convention. United Nations Framework Convention on Climate Change (UNFCCC). Available at http://unfccc.int/resource/docs/2009/ sb/eng/02sum.pdf. (2009c). Negotiating Text. Ad Hoc Working Group on Long-term Co-operation Action under the Convention. Sixth Session. Bonn. 1.12 June 2009. Available at http://unfccc.int/resource/docs/2009/awglca6/eng/08.pdf. US Senate (1993). Government Performance and Result Act. Senate and House of Representatives of the United States. Washington, DC. van Zeben, J. A. (2009). ‘The European Emissions Trading Scheme Case Law’. Review of European Community and International Environmental Law, 18(2): 119–128. Vasa, A. (2010). Implementing CDM Limits in the EU ETS: A Law and Economics Approach. DIW Working Paper Series. SSRN eLibrary. Berlin: Deutsches Institut für Wirtschaftsforschung (DIW). Available at http://ssrn.com/ paper=1641062. Walker, N. (2006). Concrete Evidence? An Empirical Approach to Quantify the Impact of EU Emission Trading on Cement Industry Competitiveness. School of Geography, Planning and Environmental Policy, University College Dublin Working Paper.

References

*

263

Walker, N. and M. Richardson (2006). Developing National Standards for Durability, Performance and Environmental Sustainability of Concrete: An Irish Case Study Illustrating the Potential for ‘Win-Win’. University College Dublin Working Paper. Ward M. (2010). Engaging Private Sector Capital at Scale in Financing Low Carbon Infrastructure in Developing Countries. Wellington: GtripleC. www.gtriplec. co.nz/future-framework. Weber T. and K. Neuhoff (2010). ‘Carbon Markets and Technological Innovation’. Journal of Environmental Economics and Management, 60(2): 115–132. Webster, M., S. Paltsev and J. Reilly (2006). The Value of Emissions Trading. Massachusetts Institute of Technology Joint Program on the Science and Policy of Global Change. Weitzman, M. L. (1974). ‘Prices vs. Quantities’. Review of Economic Studies, 41 (4): 477–491. Wier, M., K. Birr-Pedersen, H. Jacobsen and J. Klok (2005). ‘Are CO2 Taxes Regressive? Evidence from the Danish Experience’. Ecological Economics, 52: 239–251. Winkler, H. (2007). Long-term Mitigation Scenarios: Technical Report. Prepared for Department of Environment Affairs and Tourism South Africa. Energy Research Centre. Winkler, H., B. Brouns and S. Kartha (2006). ‘Future Mitigation Commitments: Differentiating among Non-Annex I Countries’. Climate Policy, 5(5): 469–486. Winters, L. A. (1995). ‘Liberalizing European Steel Trade’. European Economic Review, 39: 611–621. World Bank (2006). Clean Energy and Development: Towards an Investment Framework. Washington, DC: World Bank. (2008). Global Economic Prospects 2008: Technology Diffusion in the Developing World. Washington, DC: World Bank. Yang, M. and W. Blyth (2007). Modeling Investment Risks and Uncertainties with Real Options Approach. Paris: IEA Working Paper Series LTO/2007/WP 01. Zeppezauer, C. (2009). China is the World’s Leading Host for Clean Development Mechanism Projects. Business Forum China. http://bfchina.eu. Zhang, Z. X. (1998). ‘Greenhouse Gas Emissions Trading and the World Trading System’. Journal of World Trade, 32(5): 219–239.

Index

Åhman, M. 79, 117 Aldy, J. E. 89, 149 Alic, J. A. 33 allowance allocation methods, efﬁciency of 72–73, 81–82 benchmark approach 74–75 deﬁnition of benchmark 75 fuel-differentiated benchmarks 75–76 with historic baseline 75 output-based benchmarks 75 serious implications of 76 closure provisions 76, 80 future commitments 77 relocation of production 77 grandfathering approach 73–74 international competitiveness and leakage 82 and investment decision 117 new-entrant allocation 77–78 investment decisions 80 problems with 78–79 allowances banking of, and price stability 117–119 problems with 119 borrowing from future 119–120 free allocation of, and carbon-intensive production 185, 188–192 and investment security 123–124 and leakage 191 and price caps 125

put option contracts 122, 123 see also auctions, carbon leakage, free-allowance allocation aluminium, values at stake owing to carbon prices 173 American Power Act 87 Andersen, M. S. 21, 167 Ashton, J. 137 assigned amount units (AAUs) 143–144 auctions, of allowances 83, 93–94 design of 84–85, 86 in Germany 85–86 in UK 85 objectives of 83–84 Azar, C. 36 Baker, E. 36 Baker, T. 48 banks, and low-carbon investment ﬁnance 107, 109 Barnett, J. 166 Baron, R. 89, 149, 192, 232 Barrett, S. 140 Bazilian, M. 30 BBC 21 Bell, W. 149 Bergek, A. 37 Birr-Pedersen, K. 49 Black, R. 142 Blyth, W. 130

264

Index

Bodansky, D. 161 Böhringer, C. 76, 188 Bosi, M. 231 Bourgeois, B. 36 Brazil CDMs in 221 NAMAs 16 Bremner, C. 35 Brouns, B. 236 Bruvoll, A. 21 Burniaux, J. M. 166 Burtraw, D. 60, 90, 93 Caldeira, K. 29 cap-and-trade schemes 6–7, 56–57, 58, 93–94 allowance allocation and leakage 199–200 banking of allowances 117–118 and price stability 118–119 and credibility 240, 241, 242 direct linking of 146–147, 153–154 economic advantages of co-operation 149 economic incentives for 8 implementation of more ambitious policies 150 implementation periods 116–117 indirect linking 147–148 monitoring, reporting and veriﬁcation 59 and political responsibility 151, 152 co-operation and commitment to caps 152–153 coverage of policies 151–152 importance of clear targets 151 price caps 125 sectoral risks 112 separation of cap setting from allocation 93 timing of inclusion 91 and uncertainty 100–103, 130 see also EU ETS; US SO2 emission scheme carbon banks 120–121 carbon capture and sequestration (CCS) 43 carbon costs 164 comparison across sectors 168–170, 171–173 impact on speciﬁc industrial activities 168

*

265

and internationally traded products 170–171 potential increase in 164 Carbon Disclosure Project 127 carbon leakage 13, 162, 163–164, 168, 173, 176, 177, 199, 243–244 border adjustments 195, 200–201 beneﬁts to developing countries 197, 198 design objectives of 197–198 implementation of 195–196 international co-operation 196–197 politics of 196–197, 198 WTO compatibility 196 and cost increases 14 fossil fuel channel 165–166 and impact of carbon pricing 180 and management time frames 180 options to prevent asymmetric cost impacts 185–187 closure of plant 187 new investment 187 reduced productivity 187 reinvestment 187 production channel 166 protection of market share 178 sectoral agreements 192 among ﬁrms 192–193 benchmark emissions rate 194 to deliver full carbon price 193–194 technology channel 166 see also carbon-pricing schemes; industry value chain carbon market 132, 158–159, 160 AAUs and EU ETS 145 and efﬁciency in international markets 149–150, 159 price competitiveness 145 trade in AAUs 143–144 at government level 144, 145 at installation level 144–145 carbon market mechanisms 218 carbon price as driver of investment choice 97 impact on oil demand 105 and investor conﬁdence 115–116 allowing for banking 117–118 government intervention 120–121

266

*

carbon price (cont.) implementation period of scheme 116–117 reserve price in auctions 121–122 uncertainty of 99–100 see also allowances; cap-and-trade schemes; carbon market; carbon tax carbon price differentials effect on corporate decisions 178–179, 199 effect on international trade 162–163 carbon-pricing schemes 57 advantages of 238 border adjustment 244 cost increases in UK 199 direct emissions costs 169 distribution of revenue 49–50 differences between consumers on similar incomes 50 differences within consumer segments 50 and domestic implementation of climate policies 3–4, 6, 14 economic incentives for 19, 21, 25, 26, 28 and efﬁciency improvement 28, 29, 53 estimation of demand response 26–27 use of market mechanism 27 government responsibility 92 impact on emissions 21–22 implications for poorer consumers 45–48 incentives for 73–74 initial application at different levels 243 international co-operation on low-carbon development 4 international competitiveness 90 and leakage 180 political costs 90 political sensitivity of 239 potential cost increase 164 and regulation 239 and risk reduction 38 role in promoting use of low-carbon technologies 43–45 sectoral coverage 86–87, 94 incentives for innovation 89 inclusion beneﬁts 87–89 inclusion of large sectors with low responsiveness to carbon prices 89

Index

Stern Review 5 targeted policies 238 transaction costs 90–91 transmission of costs to consumers 45–48 in power sector 47 see also cap-and-trade schemes; carbon costs; carbon leakage; differentiated carbon prices carbon tax 56, 57, 58 and investors 8 use over short term 130 carbon-intensive production, shifts in response to cap 162, 166, 199 Carlson, C. 60 Castro, P. 222, 227 cement production border adjustment 196 and carbon leakage 14 international trade exposure 170 values at stake owing to carbon prices 171, 173, 182–183 Certiﬁed Emission Reduction (CER) credits 220, 221, 227 additionality criterion 224 and auction reserve price of carbon 122 and indirect linking of cap-and-trade schemes 147 and US non-ratiﬁcation of Kyoto Protocol 147–148 Charnovitz, S. 201 Chauis, T. 166 China CDMs in 221 IEA emissions projections for 231 NAMAs 16 taxes on project credits 223 wind projects in 15 rejection of 224 chlorine production, values at stake owing to carbon prices 173 Clean Air Act 58 clean development mechanism (CDM) 15, 63, 218, 220–221, 228, 234–235 and developing countries 245 disadvantages of 15 evaluation of 222 delay of low-carbon projects in developing countries 224

Index

low-carbon transformation in developed countries 225–226 rents and demand for carbonintensive products 222–224 uncertainty effects 226–227 difﬁculties of assessment 227 expansion of 245 and high allowance prices 125 public ﬁnance mechanisms 245–246 transition to broader framework 18 transparent governance structure 221–222 see also international carbon markets Climate Change Act, and targets 151 Climate Change Agreements (UK) 62 Climate Change Commission (UK), longterm mitigation targets 207 climate change challenge of 237, 246–247 implementation of domestic policy instruments 239–240 and low-carbon development 237, 239, 247 prediction and quantiﬁcation of 103 climate policy commitment to long-term objectives 138–139 credibility of 139–140 game-theory perspective 135–136 impact on growth network effect 52–53 non-convexities of technologies 53 and investment 109, 115, 130 European targets 113–115 ex ante v. ex post assessments 111 pooling of risk across sectors 111–112 time frames and targets 110–111, 112–113 need for joint action 132, 133 uncertainty of international objectives 130 clinker production 181 benchmark allowances 189 substitution opportunities in value chain 26 value at stake along value chain 182–183 coal, demand for, and carbon leakage 165 Coase, R. 20, 56 Colombier, M. 95

*

267

combined-cycle gas turbines, investment in 97 commodity-price stabilisation 120 competition, in international markets and innovation technology 39 and pricing decisions 48 Concorde East 25 Congressional Budget Ofﬁce, rebate curve 49 contract for difference (CfD) 122 Convery, J. F. 22 Copeland, B. R. 168 Copenhagen Accord free rider issue 246 NAMAs 208 Cornwall, A. 48 corporate taxation, reduction from carbon-pricing revenues 50 Coudonel, A. 142 credit guarantees, and international support mechanisms 213 Creedy, J. 48 Crocker, T. D. 56 Cust, J. 31 damage cost curve 102–103 de Coninck, H. 30 DeCanio, S. 43 decision-making processes, impact of carbon pricing on 44 Demailly, D. 168, 179, 189, 195 Dessai, S. 166 developing countries business models and access to ﬁnance 213 co-operative arrangements 34 development of conceptual framework for international co-operation 206 low-carbon development strategies 206 NAMAs 16–17, 207–208 support mechanisms for climate policy 1, 2, 10, 15 and carbon pricing 3 transparent monitoring and reporting 209–210 emissions trading 57–58 importance of low-carbon development plans 9 increase in greenhouse gas emissions 203, 204

268

*

developing countries (cont.) installation-based sectoral trading 228–230 international research projects on low-carbon development 205 Kyoto Linking Directive 63–64 no-lose targets for sectoral trading 231–232, 233–234 technology innovation, need for support 34, 35, 40 see also carbon leakage; clean development mechanism; international support mechanisms differentiated carbon prices 13–14 domestic policy frameworks, role of 3, 9–10, 15 implications of changes to 5–6 and international co-operation 10–11 Drax, credit rating of 128 Drexhage, J. 149 E+/E− rule 224 early movers, in low-carbon development, advantages of 138, 162 early-stage technology investment 126 Ecofys 47 Edenhofer, O. 29, 54, 150 Ekins, P. 21, 167 electricity generation and CO2 emissions 46 compensation for poorer households 47 cost pass-through 47 price-setting 47 effect of increased carbon costs on 169–170 and international trade exposure 170 Ellerman, D. 22, 56, 59, 60, 61, 69, 72, 78, 118, 125, 190 Ellis, J. 231, 232 emission reduction global consensus on 21 shift of emphasis to low-carbon development 2–3 substitution opportunities in value chain 26, 28 and welfare theorem 4 emissions targets, and investment 7 see also EU ETS emissions trading 57

Index

employment, effect of carbon price on 163, 199 energy carriers, and technology innovation 30 energy consumption, link with prices cross-country comparisons 22–23 and economic performance 23–24 energy efﬁciency, EU green paper on 43 Energy Modelling Forum 54 energy supply, and technology innovation 30 energy-efﬁciency barriers 20 Enevoldsen, M. 21 environmental taxes, impact on industrial activity 167 exemptions to 168 EU Directive on Emissions Trading, amendment on cap setting 71, 115 EU emissions-trading scheme (EU ETS) 6–7, 8, 62–63, 238, 239 allocation periods and expectations 72, 80 allowance for banking 118 auctions 83 design of 84–85 objectives of 83–84 burden sharing agreement 62 carbon costs 164 carbon leakages 191 carbon pricing and emissions reduction 21 CERs 220 decentralisation of cap setting 71 deﬁnitions of sectoral boundaries 230 effect on emitting companies 97 independent functioning of 63 lessons from pilot period 69 excess allocation of allowances 69–70 Linking Directive 63–64, 148 limits on CDM and JI credits 225–226 market price for allowances 65–66 and fuel prices 67–68 veriﬁed emissions data and price drop 68–69 national allocation plans 65–70 phase II 70–71 transfer provision 96 opposition to 71 phase II as commitment device 139 proﬁtability of coal power station construction 79 reduction in volume of allowances 66–67

Index

state aid 200 targets for emissions reduction 114–115 threshold for small emitters 90 see also allowance allocation methods European Bank for Reconstruction and Development (EBRD), ex post grants 217 European Commission attempts to link international emissions reduction 137 carbon tax proposal 61 harmonised European approach to emissions reduction 62 European Council 25 European Investment Bank, up-front support grants 214 European Linking Directive 146 European Renewables Directive 39 export taxes, and carbon-cost differential 186 Fankhauser, S. 214 feed-in tariffs 218 Felder, S. 165 Ferrario, M. 79 fertilisers, values at stake owing to carbon prices 173 Figueres, C. 236 ﬁnancial investors, and climate-policy instruments 8 ﬁrst-mover advantage, international carbon prices 162 Fischer, C. 54 free-allowance allocation auctioning of 7 distributional impacts 7 and lobbying 6 free-riding, and international agreements 134–135, 136–137, 156 Gallaway, M. 179 Gboney, W. 214 Gerlagh, R. 179 Germany, emissions reduction in 95 and allowance auctions 85–86 and new-entrant allocation 79 global carbon tax 12, 155–156 avoidance of competitive distortions 157 continuity and credibility 156–157

*

269

design of 157 domestic initiative for 157 political momentum 156 Global Environment Facility 217 government-led agreements, on carbon price 186, 200 Government Performance Results Act (USA) 142 Graichen, V. 72, 173, 190 Grant, K. 31, 214 greenhouse gas emissions, shifting balance in 203 Grubb, M. 32, 37, 64, 124, 150, 166 Guerin, E. 214 Guo, J. 103 Guzman, A. 140 Hamilton, K. 130 Hammitt, J. K. 111 Harrington, W. 111 Harrison, D. J. 72, 76 Hassett, K. A. 43 He, G. 224 Helm, D. 113, 120 Henkin, L. 139 Hepburn, C. J. 103, 121 Heutschel, J. 142 HMRC, R&D expenditure 32 Hoel, M. 104 Hoffert, M. I. 29 Holmgren, K. 79 Hope, C. 103 Hourcade, J.-C. 48, 168, 214 Iliev, I. 33 import costs, components of 173–175 and carbon prices 175–176 international differences in 176 and volatile global demand 177 India CDMs in 221 NAMAs 16 industry value chain 181 and risk of leakage 182–185 innovation private-sector investors in 32, 35 public funding for 33 targeted support schemes 36 beneﬁts of 36, 37 see also wind energy

270

*

Innovation Modelling Comparison Project 54 intellectual property rights conﬂict resolution 35 co-operative arrangements 34 returns on R&D investment 33 Intergovernmental Panel on Climate Change (IPCC) 2 emissions-reduction targets 110 and free-allowance allocation 81 GDP reduction and climate policy 19 leakage deﬁnition 165 international carbon markets, co-operation in 234–235 installation-based sectoral trading 228–230 sector trading at government level 232–234 sectoral trading linked to cap-and-trade schemes 230–232 international climate co-operation 10–11, 247 advantages of 11, 132–133, 158–160, 241, 242 basis for negotiations on emissions target 153 need for joint negotiation 153 commitments to climate policy 140 common level of ambition 137–138 encouragement of ﬁrst movers 138, 150 free-riding 134–135, 136–137 levels of ambition and targets 242 implications for emissions trading 242–243 linking of separate schemes 12, 153–154, 159 importance of design for future 155 risks of 154 responsibility for own actions 133–134 risk management 11–12 see also cap-and-trade schemes; carbon market; monitoring and reporting international support mechanisms 208–209, 220, 235–236, 244–246 access to ﬁnance 213 basic categories for 212–213 credit and equity ﬁnance 219 direct grants 214 ﬁnancial needs for low-carbon development 210

Index

access to low-cost ﬁnance 211–212 implementation of speciﬁc NAMAs 211 stability of regulatory frameworks 211 insurance or guarantee products 218–219 moral hazard problems 217, 219–220 operating support 217–218 support for domestic monitoring and reporting 209 up-front support 214 international trade exposure 170 investment policy, and low-carbon development 240 access to ﬁnance 213 design challenge 240–241 and energy demand 23 international support mechanisms 211–212 long-term policy frameworks 38 and regulation 25 Ismer, R. 123, 196 Jackson, H. 214 Jacobson, S. 37 Jacoby, H. D. 125 Jacquier-Roux, V. 36 Jain, A. K. 29 Japan, emissions controls 58, 95 Jensen, S. G. 37, 96 Joint Implementation (JI) projects 63 and auction reserve price of carbon 122 and indirect linking of cap-and-trade schemes 147 Joint Implementation Supervisory Committee (JISC) 95 Joskow, P. 69, 78 Kammen, D. M. 32 Karnøe, P. 37 Karp, L. 104 Kenber, M. 130 Klemperer, P. 96 Köhler, J. 48, 150 Kopp, R. 111 Kuik, O. J. 150, 166, 179 Kyoto Protocol assigned amount units (AAUs) 64 trading of 64–65 banking of allowances 65 carbon off-sets 218

Index

categorisation of developed and developing countries 209 clean development mechanism 1–2 emissions trading 61 principle of common but differentiated responsibility 203 role of carbon markets 143 supplementary criteria for international off-sets 225 trading of emissions reduction 147 see also clean development mechanism Labandeira, X. 48 Labeaga, J. 48 Lange, A. 76, 188 Larsen, B. M. 21 Lee, B. 33 Lessmann, K. 29, 150 Lester, S. 143 lime, values at stake owing to carbon prices 171, 173 live-line tariff 47 loans to ﬁnance low-carbon projects 213 Local Public Service Agreements (LPSAs) 142 Lossen, J. 38 Low Carbon 100 Europe Index 127 low-carbon development, emphasis on 2–3, 9–10 evolution of climate policy 18 investment responses to 7–9 use of carbon pricing 4 and welfare theorem 4–5 low-carbon investment, uncertainty of 97–98 allowing for banking 117–118 and bank ﬁnance 107 credible targets 129 government’s role 120–121, 124, 130 information on risk 127–128 need for further analysis 128–129 by oil majors 105 over time 100–104, 112–113 price prediction 129 reserve price at auctions 121–122 sources of ﬁnance for 126–127 by technology developers and manufacturers 105–107 by utility companies 107

*

271

and utilities 107, 108 see also allowances; carbon price low-carbon technologies development of 240 and emissions targets 238 investment in 126–127 low-carbon transition effect on established companies 51 role of government 51, 54 and failing ﬁrms 51 regulation of shareholders 52 training and education 52 Lundvall, B. 31 Maastricht criteria 155 Margolis, R. M. 32 Marsiliani, L. 160 Martin, K. 78 Martins, J. O. 166 Matthes, F. 46, 51, 72, 79, 190 McKinsey and Company 47 Metcalf, G. 43, 55 Mexico CDMs in 221 long-term mitigation strategy in 206 NAMAs 16 Michaelowa, A. 220, 222, 227, 236 Millennium Development Goals 142 mitigation cost curve 101–102 monitoring and reporting international co-operation 141 harmonisation on indicators 141–142 transparency 140–141, 158, 209–210, 235 Morgan, M. G. 41 Morgenstern, R. 54, 62 Morse, R. K. 224 Mowery, D. S. 33 Myers, S. C. 219 national allocation plans, free-riding problems 154 National Renewable Energy Action Plan 106, 114 nationally appropriate mitigation actions (NAMAs) 1, 2, 16, 207–208, 220, 234 commitments by developed countries 16–17 ﬁnancial mechanisms for 17–18

272

*

nationally appropriate mitigation actions (cont.) international ﬁnancial support, need for veriﬁcation 210 monitoring and reporting of 17 see also international support mechanisms natural gas, demand for, and carbon leakage 165 Nemet, G. F. 36 Neuhoff, K. 23, 33, 38, 46, 47, 72, 79, 95, 104, 123, 143, 196, 210, 214 New Source Performance Standard 58 new technologies, importance of access to ﬁnance 213 Newbery, D. M. 103, 120, 124 Newell, R. 100, 104, 119 Norberg-Bohm, V. 37 Nordhaus, W. D. 155 oil, demand for, and carbon leakage 165, 166 oil majors, and low-carbon investment 105, 108 onshore wind, investment in 127 option contracts, see allowances Palmer, K. 93 Paltsev, S. 161 Parry, I. 50 patent pools 34 Paul, A. 90 Pershing, J. 130 personal carbon credits 55 petrol taxes 50 Philibert, C. 153, 192, 231 Pigou, A. C. 20, 56 Pizer, W. 62, 100, 104, 111, 125 policy indicators 142–143 pollution haven hypothesis 168 Poterba, J. M. 55 Poverty Reduction Strategy Papers (PRSPs) 142 poverty, and carbon pricing 45, 48–49 price elasticity, estimation of 26 product prices, effect of carbon price on 163 Proops, J. 49 Quirion, P. 179, 189, 195 Radov, D. B. 72, 76 Rajan, R. 214

Index

Reedman, L. 106 Regional Greenhouse Gas Initiative (RGGI) 86 and carbon leakages 191 regulation, and low-carbon transmission 54 design difﬁculties 25 enforcement of 44 rebound effect 28, 44 uncertainty, and climate policies 98 use in directing investment 25 regulatory framework, distributional impacts of change 20 Reinaud, J. 107, 153, 192 renewable energy prediction difﬁculties 25 substitution for carbon-intensive energy 24 and technology innovation 30 Renström, T. 160 Richardson, M. 26, 47, 95 Rogge, K. 72 Rosendahl, K. 150 Rutherford, T. F. 165 Sagar, A. 35 Samaniego, J. 236 Sandén, B. A. 36 Schumacher, K. 173 sectoral trading, at government level sectoral crediting mechanism 232, 233–234 time lags between implementation and emissions impact 232 link to lead indicators 233 sectoral trading, installation-based and benchmark performance 228–229 deﬁnition of installation 229 incentive to increase carbon-intensive production 229 transfer of funds 229 sectoral trading, link to cap-and-trade schemes 230 deﬁnition of sectoral boundaries 230 distortions between sectors 230 emissions target deﬁnition 231 no-lose targets 231–232 Shiell, L. 232 Sijm, J. P. M. 46, 47, 150, 166 Sippel, M. 210

Index

slavery, abolition of 134 Smeers, Y. 99 Smith, S. 48, 62 Sorrell, S. 28 South Africa long-term mitigation strategy in 206 NAMAs 16 South Korea, long-term mitigation strategy in 206 Soviet Union (former), emissions reduction in 64 Speck, S. 49 Sreenivasamurthy, U. 193 Standard & Poor’s/IFCI Carbon Efﬁcient Index 127–128 standards bodies 34 state aid, carbon-intensive production and leakages 185, 188–192 steam reformers, and state aid 189 steel production and carbon leakage 14 free-allowance allocation 188 impact of carbon price differentials 179–180 value chain 183–184 values at stake owing to carbon prices 173 Steinbuks, J. 23 Sterk, W. 221 Stern Review 5 carbon price uncertainty, and low-carbon investment choice 97–98 costs of emissions reductions 110 deﬁnition of manageable temperature increase 103 policy interventions for low-carbon economy 19 Stiglitz, J. E. 120 storage technologies 25, 30 strong early action problem 74 subsidies, to address negative effects of leakage 190–191 Swierzbinski, J. 62 Symons, E. 49 tax revenue, effect of carbon price on 163, 199 Taylor, M. S. 168 technology developers, and low-carbon investment 105–107, 108, 114

*

273

technology policy, role of 29, 41, 54 barriers to emission reduction 43 and investment 7–9 long-term frameworks, and investor conﬁdence 38 marginal abatement cost curves 41–43 and Stern Review 5 support for carbon-reducing technologies 20, 29, 30, 36 Tol, R. S. J. 103 trade-intensity, and carbon leakage 170–171, 199 Traﬁgura, toxic exports 136 transfer payments, risks of 212 transport costs as component of import costs 175 and value chain leakage concerns 183–184 Tudela, F. 130 UK emissions trading 61 auctions of allowances 85 uncertainty, in energy-related markets 98 and choice of policy instruments 100–103 UNFCCC, National Communications funding for reporting 210 US Acid Rain Program 118 US SO2 emissions scheme 58–59 allocation method 72 bankable allowances 118 and new entrants 78 success of 60–61 van Zeben, J. A. 71 Vasa, A. 64 VAT schemes, border adjustments 195 Vigotti, R. 37 Walker, N. 26, 47, 48, 95 Wang, X. 137 Ward, J. 214 Weber, T. 104 Webster, M. 161 Weitzman, M. 100 Western Climate Initiative (USA) 154 and carbon leakage 191 White, H. 142 Wier, M. 49

274

wind energy beneﬁts of strategic deployment programmes 37 global market demand beneﬁts 39 Winkler, H. 206, 236

*

Index

Wittneben, B. 221 Yang, M. 130 Zeppezauer, C. 224 Zhang, Z. X. 196