Green Industrial Policy

Home , Nuclear power in Germany, Schmieden

GREEN INDUSTRIAL POLICY: CONCEPT, POLICIES, COUNTRY EXPERIENCES COPYRIGHT © UN ENVIRONMENT, 2017 UN Environment gratefully acknowledges the financial support of Deutsche Gesellschaft für ISBN No: 978-92-807-3685-4 Internationale Zusammenarbeit (GIZ) GmbH for The report is published as part of the Partner- the layout and printing of this book. The publi- ship for Action on Green Economy (PAGE)—an cation was supported by the project “Enhancing initiative by the United Nations Environment low-carbon development by greening the econ- Programme (UN Environment), the Interna- omy in co-operation with the Partnership for tional Labour Organization (ILO), the United Action on Green Economy (PAGE)” funded by the Nations Development Programme (UNDP), the International Climate Initiative (IKI) of the Federal United Nations Industrial Development Organi- Ministry for the Environment, Nature Conserva- zation (UNIDO) and the United Nations Institute tion, Building and Nuclear Safety (BMUB). for Training and Research (UNITAR) in partner- Cover photo: Colourbox.com ship with the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE). PAGE also gratefully acknowledges the support of all its funding partners: This publication may be reproduced in whole or in part and in any form for educational or non-profit ◼◼ European Union purposes without special permission from the ◼◼ Federal Ministry for the Environment, Nature copyright holder, provided acknowledgement of Conservation, Building and Nuclear Safety, the source is made. PAGE would appreciate receiv- Germany ing a copy of any publication that uses this publi- ◼◼ Ministry for Foreign Affairs of Finland cation as a source. ◼◼ Norwegian Ministry of Climate and Environment No use of this publication may be made for resale ◼◼ Ministry of Environment, Republic of Korea or for any other commercial purpose whatsoever ◼◼ Government Offices of Sweden without prior permission in writing from PAGE. ◼◼ Swiss Confederation, State Secretariat for CITATION Economic Affairs (SECO)

Altenburg, T., & Assmann, C. (Eds.). (2017). Green Industrial Policy. Concept, Policies, Country Expe- riences. Geneva, Bonn: UN Environment; German Development Institute / Deutsches Institut für Entwicklungspolitk (DIE).

DISCLAIMER

This publication has been produced with the support of PAGE funding partners. The contents of this publication are the sole responsibility of PAGE and can in no way be taken to reflect the views of any government. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of PAGE concerning the legal status of any country, territory, city or area or of its authorities, or concerning delimitation of its frontiers or boundaries. Moreover, the views expressed do not necessarily represent the decision or the stated policy of PAGE, nor does citing of trade names or commercial processes constitute endorsement. UN Environment promotes environmentally sound practices globally and in its own activities. This publication is printed on 100% recycled paper, using vegetable-based inks and other eco-friendly practices. Our distribution policy aims to reduce UN Environment’s carbon footprint. GREEN INDUSTRIAL POLICY: CONCEPT, POLICIES, COUNTRY EXPERIENCES Green Industrial Policy - Concept, Policies, Country Experiences

ii ACKNOWLEDGEMENTS

This publication was produced by the United including Antoine Dechezleprêtre (LSE); Mathieu Nations Environment Programme (UN Environ- Glachant (MINES ParisTech); Paul Lanoie (HEC ment) in partnership with the German Develop- Montreal); Pepita Miquel-Florensa and Nicolas ment Institute (DIE) within the framework of the Treich (Toulouse School of Economics); Jérémy Partnership for Action on Green Economy (PAGE). Lucchetti (University of Geneva); Tareq Emtairah (Lund University); Wilfried Lütkenhorst (DIE); The book was conceptualized and implemented Wang Tong (China Automotive Technology and by the editors, Tilman Altenburg (DIE) and Claudia Research Center); Karsten Neuhoff (German Assmann (UN Environment), under the strategic Institute for Economic Research, DIW); Smeeta guidance and advice of Sheng Fulai and Steven Fokeer and Michele Clara (UNIDO); Ying Zhang, Stone from the Resources and Markets Branch at Sirini Withana and Bert Fabian (UN Environ- UN Environment. ment); Hubert Schmitz (Institute of Development Studies, Brighton); Mariano Laplane (Centro de Contributing authors for this publication were Gestão e Estudos Estratégicos, CGEE) and Rainer Tilman Altenburg (DIE), Stefan Ambec (Univer- Quitzow (Institute of Advanced Sustainability sity of Toulouse), Sandra Averous Monnery (UN Studies, IASS). Environment), Verena Balke (UN Environment), Richard Bridle (IISD), Liesbeth Casier (IISD), Aaron We would like to thank Robert Wilson for the Cosbey (IISD), Pedro da Motta Veiga (CINDES), design and layout of this publication and Junko Hans Eichel (Friedrich Ebert Foundation), Taira (UNITAR) and Beibei Gu (UN Environment) Michela Esposito (ILO), Steve Evans (University for additional assistance during the publication of Cambridge), Kaidong Feng (Peking University), process. Sincere thanks also go to Fatma Pandey, Alexander Haider (New School of Social Research), Rahila Somra and Desiree Leon for the adminis- René Kemp (United Nations University), Babette trative assistance with this publication. Never (DIE), Emilio Padilla (Universitat Autónoma de Barcelona), Anna Pegels (DIE), Sandra Polónia We are also grateful to the German Academic Rios (CINDES), Liazzat Rabbiosi (UN Environment), Exchange Service, the German National Academic Dani Rodrik (Harvard University), Daniel Samaan Foundation and the Mercator Foundation for their (ILO), Kai Schlegelmilch (Green Budget Germany), support through the the Carlo-Schmid-Programme. Willi Semmler (New School of Social Research), Qunhong Shen (Tsinghua University), Georgeta This publication was made possible with the Vidican Auktor (DIE) and Peter Wooders (IISD). support from the Partnership for Action on Green Economy (PAGE) and its funding partners, We would like to thank Catherine P. McMullen for as well as the Swedish Government. The layout her excellent and diligent language editing, which and printing of this publication was supported helped streamline the publication and create a by the Deutsche Gesellschaft für Internationale narrative throughout it. Invaluable support was Zusammenarbeit (GIZ) GmbH, with funding from provided to the editorial team by Gisele Müller, the German Federal Ministry for the Environ- Elena Antoni, and Verena Balke who commented ment, Nature Conservation, Building and Nuclear on drafts, reviewed chapters and helped pull the Safety (BMUB). publication together with persistence and relent- less efforts.

We are very grateful for comments and suggestions from a number of capable reviewers, Foreword

FOREWORD iii A green economy is increasingly accepted as a key driver in tackling climate change, poverty, pollution, health and any number of critical goals to improve life for this planet and its people. Bold words, but this report shows that it is within our grasp to turn them into action.

Restructuring industrial systems needs a diverse, cross-sector approach. By moving beyond traditional industrial policies to a framework that encompasses environmental and energy policies, the authors explore how this could accelerate structural transformation and enhance productivity. In fact, through research and case studies, it shows how green industrial policy frameworks can be a valuable tool for all economies.

For example, Morocco used to import 95 per cent of its energy via coal, gas and electricity in 2011. Now, around a third is from domestic renewable sources and the country is building Africa’s largest solar power plant, the Ouarzazate Solar Complex. The country plans to produce 15 per cent of electric capacity from solar power by 2020. As well as boosting Morocco’s renewable energy, the solar plant is also strengthening the local economy. two-wheelers and exporting some 5 million a year, People like the 35-year-old Azzedine, who works mainly to other Asian markets. as a driver, found jobs there. For the first time, he now earns a fixed salary every month and feels Many of the findings of this report are already lucky to be among the young people with a stable being put into practice around the world. Through job at the plant. However, it is not only big projects the United Nations Industrial Development Organ- that will bring about economic and environmental ization, the Partnership for Action on Green Econ- benefits. The demand for small independent elec- omy is helping governments to develop action tricity producers may also promote long-term jobs plans that include green industrial policy recom- and private sector development. mendations. For example, in Burkina Faso, Ghana, Peru, Senegal and China green industry assess- Likewise, in China, unbearable urban air pollution ments have been conducted or are under way. has led to health problems, which are in turn driving an increase in demand for electric mobility. Green industrial policies offer a practical way to It’s a demand being carefully nurtured through a shape inclusive, sustainable economies right basket of measures including research and devel- around the world. I sincerely hope that this report opment, technology sharing agreements with will raise awareness of the tools available to global investors, strategic public procurement, public and private decision-makers determined to purchase subsidies and city trials. Not only do build a better future for this planet and its people. emissions go down, but in 2014, China was already producing 85 per cent of the world’s electric

ERIK SOLHEIM Head of UN Environment Green Industrial Policy - Concept, Policies, Country Experiences

iv TABLE OF CONTENTS

ii ACKNOWLEDGEMENTS

iii FOREWORD Erik Solheim v LIST OF BOXES LIST OF TABLES vi LIST OF FIGURES vii ABBREVIATIONS ix EXECUTIVE SUMMARY Tilman Altenburg, Claudia Assmann PART 1: CONCEPTUAL FOUNDATIONS 1 Chapter 1 Green industrial policy: Accelerating structural change towards wealthy green economies Tilman Altenburg, Dani Rodrik PART 2: THE ECONOMIC AND SOCIAL CO-BENEFITS OF GREEN TRANSFORMATION 22 Chapter 2 What can developing countries gain from a green transformation? Emilio Padilla 38 Chapter 3 Gaining competitive advantage with green policy Stefan Ambec 50 Chapter 4 Enhancing job creation through the green transformation Michela Esposito, Alexander Haider, Willi Semmler, Daniel Samaan PART 3: ACCELERATING CHANGE 69 Chapter 5 In with the good, out with the bad: Phasing out polluting sectors as green industrial policy Aaron Cosbey, Peter Wooders, Richard Bridle, Liesbeth Casier 87 Chapter 6 Developing green technologies and phasing them in Babette Never, René Kemp 102 Chapter 7 Pricing environmental resources and pollutants and the competitiveness of national industries Kai Schlegelmilch, Hans Eichel, Anna Pegels 120 Chapter 8 Promoting circular economies Verena Balke, Steve Evans, Liazzat Rabbiosi, Sandra Averous Monnery 134 Chapter 9 Trade and investment law and green industrial policy Aaron Cosbey PART 4: COUNTRY EXPERIENCES 153 Chapter 10 Renewable energy as a trigger for industrial development in Morocco Georgeta Vidican Auktor 166 Chapter 11 Germany: The energy transition as a green industrial development agenda Anna Pegels 185 Chapter 12 Electric mobility and the quest for automobile industry upgrading in China Tilman Altenburg, Kaidong Feng, Qunhong Shen 199 Chapter 13 Ethanol policy in Brazil: A ‘green’ policy by accident? Pedro da Motta Veiga, Sandra Polónia Rios List Of Boxes / List of Tables

LIST OF BOXES v

73 Box 5.1 Jordan’s fossil fuel subsidy reform

74 Box 5.2 Fossil fuel subsidy reform in Morocco

76 Box 5.3 China’s drive for a greener economy

77 Box 5.4 Ontario’s coal phase out and renewables support

107 Box 7.1 Factors influencing foreign direct investment in Vietnam

108 Box 7.2 Germany’s tax exemptions for industry: Leaving efficiency potentials untapped

111 Box 7.3 Earmarking in Chile

114 Box 7.4 Raising acceptance for environmental fiscal reform through cash transfer programmes: The Indonesian Bantuan Langsung Tunai

201 Box 13.1 Proálcool era: mid 1970s to mid 1980s

203 Box 13.2 Deregulation and liberalization in the 1990s

206 Box 13.3 The energy and environmental balance of sugarcane ethanol

207 Box 13.4 Flex-fuel automobiles and international commodity prices from 2003 to 2008

209 Box 13.5 Back to the crisis: since 2008

LIST OF TABLES

8 Table 1.1 New green product and service opportunities for countries at different income levels

54 Table 4.1 Employment estimates of green jobs in the European Union

67 Table 4.2 Top ten HCS and LCS in the last available year

78 Table 5.1 Ontario’s electrical generating capacity (comparison 2010–2016)

98 Table 6.1 Overview of functions of front-runner desk for innovators and policy

147 Table 9.1 Summary: Legality of trade and investment measures

168 Table 11.1 Quantitative targets and status quo 2014 of the Energiewende

178 Table 11.2 German gross employment in renewable energies, overview of study results

190 Table 12.1 Main polices for the promotion of electric vehicles in China Green Industrial Policy - Concept, Policies, Country Experiences

vi LIST OF FIGURES

28 Figure 2.1 Key areas of regulation

55 Figure 4.1 Employment in the environmental sector, 2005–2013

56 Figure 4.2 Employment shares in EGSS in EU countries, 2014

59 Figure 4.3 Employment shares in GHG emitting industrial sectors by region

60 Figure 4.4 Share of CO2 emissions by industry and ILO region

72 Figure 5.1 Spectrum of types of disruptive green industrial policy

92 Figure 6.1 Long-term phase-in of building standards

113 Figure 7.1 Financing income tax cuts through environmental fiscal reform in South Africa: The revenue effects of various reform elements

122 Figure 8.1 Conceptual diagram of a circular economy

169 Figure 11.1 German greenhouse gas emissions 1990–2015, target 2020

170 Figure 11.2 European Energy Exchange electricity spot-market prices in Germany against renewable electricity generation

172 Figure 11.3 Research, development and deployment budget energy technologies, Germany

172 Figure 11.4 Relative patent shares in relevant energy technologies, Germany

173 Figure 11.5 World market shares wind energy converters

174 Figure 11.6 World market shares solar photovoltaic cells

176 Figure 11.7 Perception of Energiewende impacts on competitiveness

178 Figure 11.8 German gross employment in renewable energies, 2004–2013

179 Figure 11.9 Employment in renewable energy and coal industries, Germany, 2002–2015

180 Figure 11.10 Feed-in tariff components

205 Figure 13.1 Number of licensed vehicles per type of fuel, 1979–2015

210 Figure 13.2 Ethanol production in Brazil, 1980–2015 Abbreviations

ABBREVIATIONS vii ACEA European Automobile FAO Food and Agriculture Organization Manufacturers' Association of the United Nations ADB Asian Development Bank FINEP Brazilian Studies and Projects Financing Agency ADEREE National Agency for the Development of Renewable GATS General Agreement on Trade in Energy Sources and Energy Services Efficiency Morocco GATT General Agreement on Tariffs and ANP Brazilian National Agency of Trade Petroleum, Natural Gas, and GCF Green Climate Fund Biofuels GDP Gross Domestic Product ANPME Moroccan National Agency for the GEF Global Environment Facility Promotion of Small and Medium- GGGI Global Green Growth Institute sized Enterprises GHG Greenhouse gas BAFU Swiss Federal Office for the Environment GPA Agreement on Government Procurement BEE Indian Bureau of Energy Efficiency GW Gigawatt BLS U.S. Bureau of Labor Statistics HCS High-carbon sector BMF Austrian Federal Ministry of Finance IAA Brazilian Institute of Sugar and Alcohol BMU/ German Federal Ministry BMUB for the Environment, Nature ICMS Brazilian value added tax on sales Conservation, Building and and services Nuclear Safety ICSID International Centre for the BNDES Brazilian Development Bank Settlement of Investment Disputes BNEF Bloomberg New Energy Finance ICTSD International Centre for Trade and Sustainable Development BNSTP Bourse Nationale de Sous- traitance et de Partenariat IDR Indonesian Rupiah CDM Clean Development Mechanism IEA International Energy Agency CEEW Council on Energy, Environment IESO Ontario Independent Electricity and Water Systems Operator CGEE Brazilian Centre of Strategic IFC International Finance Corporation Management and Studies IISD International Institute for CIDE Brazilian fuel tax Sustainable Development COFINS Brazilian Social Security ILO International Labour Organization Financing Contribution IMF International Monetary Fund COMETR Competitiveness Effects of INSG International Nickel Study Group Environmental Tax Reforms IPCC Intergovernmental Panel on

CO2e Carbon dioxide equivalents Climate Change DENA German Energy Agency IRENA International Renewable Energy DIE German Development Institute Agency DIHK German Chamber of Industry and IRESEN Institut de Recherche en Energie Commerce Solaire et en Energies Nouvelles EGSS Environmental Goods and ISL International Synergies, Ltd. Services Sector ISO International Organization for EU European Union Standardization EZ Dutch Ministry of Economic LCS Low-carbon sector Affairs LDC Least developed countries Green Industrial Policy - Concept, Policies, Country Experiences

LCR Local content requirerment RCREEE Regional Center for Renewable viii LNP Liquefied natural gas Energy and Energy Efficiency LPG Liquid petroleum gas RREUSE Reuse and Recycling EU Social Enterprises LULUCF Land use, land-use change and forestry (greenhouse gas SCM Subsidies and Countervailing inventory sector) Measures MAD Moroccan Dirham SEEA System of Environmental- Economic Accounting MAP German Market Incentive Programme for Renewable TEDA Tianjin Economic-Technological Energies Development Area MASEN Moroccan Agency for Sustainable TISA Trade in Services Agreement Energy TRIMS Agreement on Trade-Related MCINET Moroccan Ministry of Industry, Investment Measures Trade and New Technology TRU Brazilian highways maintenance MEMEE Moroccan Ministry of Energy, tax Mining, Water and the UK United Kingdom Environment UN United Nations MENA Middle East and North Africa UNCTAD United Nations Conference on MME Brazilian Ministry of Mines and Trade and Development Energy UNDP United Nations Development MW Megawatt Programme NAFTA North American Free Trade UNEP UN Environment / United Nations Agreement Environment Programme NAPE German National Action Plan on UNFCCC United Nations Framework Energy Efficiency Convention on Climate Change NRDC Natural Resources Defense UNICA Brazilian Sugarcane Industry Council Association OECD Organisation for Economic UNIDO United Nations Industrial Cooperation and Development Development Organization OICA International Organization of VAT Value added tax Motor Vehicle Manufacturers VROM Dutch Ministry of Housing, Spatial OPA Ontario Power Authority Planning and the Environment PAGE Partnership for Action on Green WBCSD World Business Council for Economy Sustainable Development PAISS Brazilian Support Plan for WEF World Economic Forum Industrial and Technological WIEGO Women in Informal Employment: Innovation for the Sugar-Energetic Globalizing and Organizing and Sugar-Chemical Sectors WIOD World Input-Output Database PERG Moroccan Renewable Energy and Global Rural Electrification Project WITS World Integrated Trade Solution World Health Organization PIS Brazilian Profit Participation WHO Programme WTO World Trade Organization PROMASOL Development Programme of the WWII Second World War Moroccan market for solar water ZAR South African Rand heaters PROPER Indonesia’s Programme for Pollution Control, Evaluation, and Rating Executive Summary

EXECUTIVE SUMMARY ix TILMAN ALTENBURG AND CLAUDIA ASSMANN countries, and even in rich nations there are substantial clusters of poverty and unsolved prob- Humanity is confronted with profound and lems of human development. Therefore, we need mounting man-made environmental crises. The to tackle a dual challenge: To pursue economic United Nation’s Millennium Ecosystem Assess- development and wealth creation, particularly ment provided an alarming inventory of the solving the problems of deprived segments of degree of deterioration in many of the world’s societies, while keeping resource consumption ecosystems (MEA 2005). Global warming is now and pollution in accordance with Earth’s bioca- widely recognised as an immediate threat to pacity. Sharing prosperity more fairly is surely humanity. As the Intergovernmental Panel on one part of the agenda. The other part, which is Climate Change shows, only a few years are left to at the centre of this report, is to develop institu- radically decarbonise the world economy if disas- tional and technological solutions that enable us trous global warming is to be avoided (IPCC 2014). to decouple economic development and human Other environmental crises have so far received well-being from resource depletion and waste less public attention, but are also serious and production. The benefits are obvious. Tackling the potentially threatening the continuity of human environmental problems that are causing millions life on Earth. These include the loss of biodiver- of deaths and profound welfare losses today, and sity, depletion of water reserves, ocean acidifica- undermining the foundations for the development tion and reduction of soil fertility, among others of future generations, will pay off for all of us. (Rockström et al. 2009). Developing new institutions and technologies is Even in purely monetary terms–if we isolate a challenge for all countries. As the Global Foot- nature’s intrinsic value from the equation–the print Network has shown, almost all countries costs of environmentally unsustainable prac- that have achieved acceptable levels of human tices are enormous. The Lancet Commission on development–scoring 0.8 or higher on the Human Pollution and Health estimates welfare losses Development Indicators, which is UNDP’s thresh- due to environmental pollution at more than old for high human development–did so by over- US$ 4.6 trillion per year, or 6.2 per cent of global stepping the world’s biocapacity, whereas those GDP (Landrigan et al. 2017). The economic cost countries that stayed within the Earth’s limits of global warming has been estimated at more so far invariably failed to provide the conditions than US$ 1.2 trillion per year, reducing the world’s for a high level of human development (Global economic output by 1.6 per cent annually (DARA Footprint Network 2010; UNEP 2011). Put differ- and the Climate Vulnerable Forum 2012). Vari- ently, not a single country worldwide provides a ous other, yet less visible, environmental threats role model for achieving decent human develop- may cause loss and damages in similar orders of ment sustainably within Earth’s biocapacity. Just magnitude. The dramatic reduction of population to emulate the development pathways of today’s of bees and other insects so essential for pollinat- rich countries, assuming a linear development ing crops, and thereby securing global harvests, is trajectory along which countries gradually evolve just one example. from underdeveloped to developed, has never been a convincing proposition, given countries’ A key underlying reason of all these negative manifold and individual characteristics in terms trends is that the incentives that guide the way of history, culture and geography. Once we start people invest, produce and consume are not using sustainable development as a yardstick, it accounting for environmental costs. Those are becomes even more obvious that development is ‘externalities’ in the economic jargon. The need not about ‘catching up’ with today’s rich nations. to rethink our incentive systems is thus obvious and urgent. Incentives need to be inspired by the From this perspective, economic latecomers principle of sustainability. They must be designed to the globalizing world economy even have to ensure that environmental costs are internal- an advantage. They can build their cities, their ized, pollution is kept to a minimum, material manufacturing industries, their energy and consumption is reduced, and inputs are reused or transport systems and their institutions in new, recycled to the greatest possible extent. more sustainable ways that take their distinctive national characteristics into account. Surely, At the same time, there are social and economic those countries that industrialised early had challenges and aspirations. People want to live more time and better opportunities to accumu- decent lives. Poverty is still widespread in many late wealth and develop institutions that may now Green Industrial Policy - Concept, Policies, Country Experiences

help them cope with environmental challenges. Our report aims to provide guidance to policy- x However, latecomers are not as deeply locked into makers and practitioners as well as to contribute existing unsustainable infrastructures and centu- to the academic debate on green transformation ry-old institutional routines that often hamper strategies. It provides an up-to-date overview of change in many ways. the debate on green industrial policy, explores what countries can gain economically from In this report, we explore policy options for pursuing environmental integrity, and what policy managing structural change that accounts for options are available to accelerate the transfor- both the productivity and the environmental mation in ways that enhance well-being and challenges in a harmonised way. We use ‘green environmental sustainability together. Practical industrial policy’ as our key concept. The term examples are included in all chapters, and four ‘industrial policy’ encompasses sets of measures national examples of successful green structural that governments use to influence a country’s change are presented in detail, covering countries economic structure in the pursuit of a desired at very different levels of income and technolog- 1 objective. Until very recently, this desired objec- ical capacity. tive was first and foremost to enhance the productivity and competitiveness that in turn would The report has four parts. PART 1 discusses allow for economic growth and higher incomes. the conceptual foundations of green industrial However, as we have seen, looming environ- policy. Altenburg and Rodrik explain why look- mental catastrophes–as well as other alarming ing through the lens of industrial policy provides trends related to poverty, inequity, exclusion important insights for a green transformation. and conflict–force us to reconsider the kind of They summarize lessons learned from decades structural change we want. The 2030 Agenda of experimentation with, and research on, indus- for Sustainable Development, adopted by the UN trial policy and bring out key principles of smart General Assembly in 2015, reflects a more encom- policymaking that maximise the government’s passing perspective on the transformation of ability to overcome market failures while keeping societies that balances economic, social and envi- the inherent risks of misallocation and political ronmental objectives (UN 2015). capture to a minimum. Subsequently, the authors identify six extra challenges of green transforma- Our concept of green industrial policy starts from tions and explain the ways green industrial policy the assumption that we can learn a lot from several must go beyond the common practice of indus- decades of experimenting with policies aimed at trial policy in a business-as-usual setting. shaping economic structures in the pursuit of societal objectives. Many key principles of successful PART 2, The economic and social co-benefits of industrial policymaking can be derived, for exam- green transformation, shows that green industrial ple: the way entrepreneurial search processes can policy may bring a number of economic and social be channelled towards certain agreed societal co-benefits, in addition to environmental improve- objectives; how regulations, market-based instruments. Padilla argues why, despite existing trade- ments and financial incentives can be combined; offs between growth policies and environmental how public services are delivered most effectively; protection, the idea of growing first and cleaning and how mandatory and voluntary measures can up later is not a good approach for policymakers be coordinated to achieve the best result. Apply- and discusses what developing countries can gain ing these lessons to green industrial policy moves from a green transformation. He identifies twelve toward further specifics. These derive from the ways in which developing countries can reap need to harmonise the requirements of productivi- social and economic co-benefits of greening their ty-enhancing structural change with environmen- economies. These range from better conditions for tal objectives and to align national interests with human health, preservation of resources for future the protection of global commons. This has mani- growth, and avoidance of high switching costs in fold practical implications. For example, certain the future to immediate cost reductions through economic transformations need to be accelerated resource-efficient production and leveraging new to achieve results before ecosystems collapse and competitive advantages through environmental the original equilibrium cannot be restored. This goods and technologies. calls for more proactive policy guidance to phase out harmful technologies and policies and to use Ambec then focuses on the firm level, describing both ‘carrots and sticks’ to speed up the dissemina- how developing-country firms can gain competition of sustainable alternatives. tive advantage through green policies. He shows

1 There is no uniform and generally agreed definition of ‘industrial policy‘. For an overview see Warwick (2013). Executive Summary

that, although environmental protection often effects and economies of scale, from path-depend- comes at an additional cost to firms, it can also ent consumer behaviour and from political back- xi enhance their competitive advantages along four ing influenced by vested interest groups. Building channels: product differentiation through green on experiences from China, Germany, India and labels; development of new green products; produc- the Netherlands, the authors identify seven prin- tivity improvements that more than compensate ciples for the design of phase-in policies. for the costs of environmental protection; and knowledge spillovers in the innovation process. Schlegelmilch, Eichel and Pegels explore the rationale of environmental fiscal reforms and Esposito, Haider, Semmler and Samaan explore show how they need to be designed, particu- how a green transformation can create employ- larly in developing countries, to achieve the dual ment benefits. Green transformation necessar- purpose of protecting the environment and spur- ily affects labour markets, creating new jobs in ring competitiveness, industrial development and the environmental goods and services sector, jobs. Differential taxation signals environmental but also reducing employment opportunities in costs while leaving it to competitive market forces sectors that are deliberately phased out due to to find the best technological and organizational their polluting effects. Measuring the net effects, solutions. Environmental fiscal reforms enhance however, is difficult because environmental the competitiveness of clean industries and improvements are incrementally adopted across reduce competitive advantages of polluting indus- sectors throughout transition and it is therefore tries. Tax rates for enterprises exposed to inter- impossible to draw a clear line between envi- national competition thus need to be designed ronmentally sound and polluting sectors and carefully, and more efforts should be undertaken jobs. Also, it is difficult to attribute employment to harmonise environmental taxation internation- changes exclusively to environmental poli- ally. Revenues can be used for poverty reduction, cies. With all these limitations in mind, there is green infrastructure and other national priorities, evidence that what statistical authorities define and they can be channelled in a way that helps as the environmental goods and services sector to build reform alliances and overcome resistance. is increasing its employment share in most of the countries for which data exist. Balke, Evans, Averous Monnery and Rabbiosi show what is needed to shift from linear produc- PART 3, accelerating change, discusses some of tion systems, in which a large part of the material the key policies that help to implement the green inputs of production end up as waste, to circular transformation. Given the urgent need to reduce economies that reduce waste, reuse materials some environmental pressures, proactive poli- as much as possible and recycle the rest. Circu- cies are needed to accelerate the replacement of larity is thus a key principle to decouple produc- unsustainable products and practices with green tion from resource consumption and pollution. alternatives. Cosbey, Wooders, Bridle and Casier The authors discuss different circular economy provide an overview of policy options to phase out approaches and provide an overview of key policy environmentally harmful industries. Abandoning instruments. These range from eco-design guide- such industries is particularly challenging when lines that ensure convenience, longevity, repair- the level of invested capital is high and there are ability and recyclability and extend to the setup strong linkages throughout national economies of waste collection systems and the promotion of that create vested interests defending the status resource-saving business models, such as sharing quo. To manage the transition to clean alterna- platforms. The authors argue that such policies tives it is important to get public buy-in through need to be contextualized for different country consultations, combined with well-defined grad- conditions and they illustrate this with a series ual timelines for change and support measures of examples including eco-industrial parks and for those who are negatively affected. Societal nation-wide systemic solutions. acceptance is likely to be greater if the phase-out goes hand in hand with measures to develop envi- Cosbey then explores the ways in which green ronmentally sound substitutes. industrial policies might be restricted by international trade and investment law. Many indus- Never and Kemp show how the phase-in of green trial policy tools have trade-related aspects and alternatives can be accomplished. The challenge these are regulated through a range of multi- here is to cope with a variety of disincentives: the and bilateral agreements with strong enforce- new green alternatives typically need to develop ment mechanisms. For example, tariff policy is and become competitive in the face of established strongly constrained by bound tariff rate commit- technologies that benefit from existing network ments; subsidies and performance requirements Green Industrial Policy - Concept, Policies, Country Experiences

conditional on domestic content requirements and Da Motta Veiga and Polónia Rios assess the role of xii export performance are prohibited. Yet, many other policy in the emergence of the national bio-eth- green industrial policy options are not affected by anol industry. Sugar cane cultivation and the trade and investment law, such as feed-in-tariffs, industrial transformation of sugar into ethanol performance requirements for training of staff, have received strong backing since the 1970s science and education policies, funded demon- to replace gasoline consumption and decrease stration projects, and others. Procurement policies dependence on oil imports. More than 15 per cent can be used to source greener products, yet not of Brazil’s energy demand is covered by ethanol, to discriminate in favour of domestic suppliers if contributing to the country’s low-carbon footprint. governments are party to the WTO‘s plurilateral The authors argue that the alleged direct nega- Government Procurement Agreement. tive effects on Brazil's forest cover are largely unfounded, although some indirect effects may PART 4, country experiences, then examines the exist. High-yielding second generation sugar practical implementation of key policies in four cane may further reduce these effects. Yet until countries at different levels of income and tech- recently, the government has designed its etha- nological capacity: Morocco, China, Brazil and nol policy as a means to mitigate oil price fluctua- Germany. All examples address the simultane- tions rather than for environmental purposes, and ous challenge of fostering jobs and technological therefore cut support whenever oil prices were learning and creating competitive advantages in low. Also, despite some technological innovations, new industries while greening their economies. like the domestic development of new flex-fuel combustion engines, little has been undertaken Vidican Auktor shows how Morocco uses its to develop new industrial capabilities. favourable conditions for energy generation from solar radiation and wind to reduce its enormous Finally, Pegels discusses the German Energie- fuel import dependence, to create employment wende, the transition from coal and nuclear to and to trigger technological learning. In 2011, the renewable energy and enhanced energy efficiency, country was importing more than 95 per cent of from the perspective of economic co-benefits. Her its energy and its energy demand is expected to analysis shows that Germany has made consid- triple by 2030. Morocco’s government not only erable progress in the deployment of renewa- encourages foreign direct investment in solar and ble energies for electricity generation, whereas wind energy projects but also supports related other areas, such as energy efficiency, are lagging skills development and the emergence of domes- behind expectations. Regarding manufacturing tic supplier industries. Moreover, policymakers industries, Germany’s wind energy industry has foster both high-tech investments in concen- emerged as a new global leader. Other sectors trated solar power plants and low-tech rooftop have been less successful. The solar panel indus- solar thermal and photovoltaic projects to develop try boomed for some years but then experienced various segments of the labour market. many bankruptcies due to competition from low-cost imports. Employment in Germany’s envi- Altenburg, Feng and Shen provide an overview of ronmental goods and services sector reached an China’s policies to promote electric mobility with estimated 260,000 jobs as of 2013. Also, rising elec- the dual aim of curbing urban air pollution and tricity prices may have reduced the competitive- enhancing the competitiveness of its national ness of energy-intensive industries, a condition automobile industry. That industry is strategic that is difficult to quantify. for China’s technological upgrading, but it has not been able to reach the productivity levels of Overall, the report provides a comprehensive its international competitors so far. The shift to overview of the rationale for environmental trans- electric powertrains is therefore seen as an oppor- formation, of the synergies and potential trade- tunity to boost national competitiveness by leap- offs among social, economic and environmental frogging into a new generation of technologies. objectives and of different policy approaches and The government’s comprehensive support pack- experiences of practical implementation in a age is unrivalled by any other country. Progress wide range of country contexts. We hope it helps has been made in four areas of technology devel- analysts and practitioners to accelerate the green opment: modern cars and buses, low-speed cars, transformation in a way that harmonises soci- two-wheelers and battery manufacturing. etal objectives in the spirit of the 2030 Agenda for Sustainable Development. Executive Summary

REFERENCES xiii DARA and the Climate Vulnerable Forum. (2012). Rockström, J., Steffen, W., Noone, K., Persson, Å., Climate Vulnerability Monitor. 2nd edition. Madrid, Chapin, F.S., Lambin, E.F., Lenton, T.M., Schef- Geneva: A Guide to the Cold Calculus of a Hot fer, M., Folke, C., Schellnhuber, H.J., & Nykvist, Planet. B. (2009). A safe operating space for human- Global Footprint Network. (2010). The Ecological ity. Nature, 461(7263), 472–475. Wealth of Nations: Earth’s Biocapacity as a New UN (2015) Transforming Our World: The 2030 Agenda Framework for International Cooperation. for Sustainable Development. New York: United Intergovernmental Panel on Climate Change (IPCC) Nations General Assembly. (2014): Climate Change 2014: Mitigation of United Nations Environment Programme (UNEP) Climate Change. Contribution of Working Group (2011): Towards a Green Economy: Pathways to III to the Fifth Assessment Report of the Intergov- Sustainable Development and Poverty Erad- ernmental Panel on Climate Change [Edenhofer, ication. Nairobi: United Nation Environment O. et al. (eds.)]. Cambridge, United Kingdom and Programme. New York, NY: USA Cambridge University Press. Warwick, K. (2013). Beyond Industrial Policy: Emerg- Millennium Ecosystem Assessment. (MEA) (2005). ing Issues and New Trends. OECD Science, Tech- Ecosystems and Human Well-being: Synthesis. nology and Industry Policy Papers, No. 2. OECD Washington, DC: Island Press. Publishing. Landrigan, P.J., Fuller, R., Acosta, N.J., Adeyi, O., Arnold, R., Baldé, A.B., Bertollini, R., Bose-O’Reilly, S., Boufford, J.I., Breysse, P.N., & Chiles, T. (2017). The Lancet Commission on pollution and health. The Lancet. Green Industrial Policy - Concept, Policies, Country Experiences xiv PART 1: CONCEPTUAL FOUNDATIONS CHAPTER 1 1 GREEN INDUSTRIAL POLICY: ACCELERATING STRUCTURAL CHANGE TOWARDS WEALTHY GREEN ECONOMIES

Tilman Altenburg, Dani Rodrik Green Industrial Policy - Concept, Policies, Country Experiences

2 1. INTRODUCTION There are two major reasons for governments and countries with whom developing countries should societies to accelerate structural change in their catch up, ignoring the limits of our planet’s carry- economies and proactively shape the direction of ing capacity. Environmentalists have tended to the change. First, there is the challenge of creating put conservation first and downplay the challenge wealth. Structural change, that is, the reallocation of creating wealth for billions of people who aspire of capital and labour from low- to high-productiv- for a better material life. In this chapter we make ity activities, is a key driver of productivity growth an effort to bring these perspectives together and and higher incomes. This is particularly important suggest ways of balancing the inherent trade-offs. for developing countries where incomes are low and poverty is pervasive. According to the latest Industrial policy is our analytical angle. Industrial available estimates, 767 million people lived on policy refers to government actions to alter the less than US$ 1.90 a day, and 1.9 billion people in structure of an economy, encouraging resources the developing world still had less than US$ 3.10 to move into particular sectors that are perceived a day in 2013–a clear indication that the current as desirable for future development. Tradition- structural composition of national economies ally, industrial policy has focused on productiv- does not provide a sufficient number of produc- ity enhancement as the key mechanism that tive jobs (World Bank 2016). Second, economic would ensure rising returns to capital and labour development has so far been achieved at the cost and thus enable economic growth and prosper- of severe overexploitation of natural resources. ity. Increasingly, however, the goals of industrial Humanity is approaching various ecological policy have been broadened. In practice, industrial tipping points beyond which abrupt and irrevers- policy agencies undertake measures to influence ible environmental change at large geographical structural change such that regional disparities scales is likely to happen (Rockström et al. 2009). are reduced, labour-intensive industries or small Radically new techno-institutional systems are enterprises are encouraged and/or the econ- needed to decouple economic development and omy becomes environmentally more sustaina- human well-being from resource depletion and ble (Altenburg and Lütkenhorst 2015). The main waste production. While many of the required objective of this chapter is to show how industrial technologies are already available, the incen- policies can be designed to deal with the dual tives guiding resource allocation need to change challenge of shifting economic structures in a way profoundly to disrupt current unsustainable tech- that prosperity is increased while at the same time nological pathways and change some economic replacing environmentally unsustainable activi- subsystems entirely, such as those for energy ties with sustainable ones. The rationale for indus- provision and transport (IPCC 2014). trial policy rests on the idea that market prices are not always the best guide to allocating invest- This chapter explores the policy options for ments. We shall argue in the chapter that this idea managing structural change that accounts for applies with much greater force where climate both the productivity and the environmental chal- change and green technologies are concerned. lenges in a harmonised way.2 This is a challenge for all countries. Yet we put developing econo- A quick word about the term ‘industrial policy’: mies at the centre of our analysis because this We use this term because it has a well-recog- is where the need to accelerate wealth creation nised meaning and a long history. But the range of is greatest, and many stakeholders perceive this policies we shall cover goes much beyond indus- as incompatible with environmental conserva- try itself. In view of this, some terms that have tion. Governments typically put economic growth recently come into use, such as structural trans- above environmental objectives, arguing that part formation policies or productive development of the income generated can be used to clean up at policies, would perhaps have been more appropri- a later stage. Also, the policy discourse has often ate. We stick with the traditional term; though we been biased towards specific objectives: Indus- caution the reader that the kind of issues we cover trialists have mainly sought solutions modeled here concern entire economies and not just manu- after the successful cases of early industrialising facturing industries.

2 While this chapter puts the dual challenge of creating wealth and greening economies at the centre of its analysis, industrial policy should in fact be assessed against a wider range of societal objectives, as laid down in the 2030 Agenda for Sustainable Development (UN 2015). Increasing employment opportunities for youth, reducing regional disparities or supporting women’s entrepreneurship are all legitimate goals. Ultimately, industrial policy has a norma- tive content that depends on what societies define as desirable future courses for social and economic development (Altenburg and Lütkenhorst 2015). Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

We proceed in three steps. Section 2 analyses the policy. In section 3 we extract the main lessons dual challenge in greater detail. It first summa- from various decades of controversial debate on 3 rizes what we know about the link between industrial policy and bring out three key princi- structural change and wealth creation; it then ples of smart policymaking that maximise the addresses the need to decouple human develop- governments’ ability to overcome market failures ment from non-renewable resource consumption while keeping the inherent risks of misallocation and pollution, and it shows how this translates and political capture to the minimum. Section 4 into a structural transformation of economies, takes this debate one step further, exploring the including those of developing countries. The extra challenges of a green transformation and in section ends with a discussion of synergies and what ways green industrial policy must go beyond trade-offs between the two objectives. Sections the common practice of industrial policy in a 3 and 4 then deal with the design of industrial business-as-usual setting. Section 5 concludes.

2. THE DUAL CHALLENGE: CREATING WEALTH FOR A GROWING POPULATION WHILE STAYING WITHIN PLANETARY BOUNDARIES

2.1. CREATING WEALTH: THE NEED FOR Manufacturing plays an important role in the STRUCTURAL CHANGE process of diversification. The manufacturing sector is particularly well-suited for serial Productivity growth is a precondition for improv- production allowing for enormous economies ing living standards and maintaining competitive- of scale. On average, it supplies highly produc- ness in the globalized economy. Low total-factor tive and well-paid jobs; it is where most private productivity is one of the main reasons for persis- sector research and development and commer- tent poverty in developing countries. Low income cial innovation take place and where most royal- and lower-middle income countries in particu- ties are generated. It generates demand for jobs lar need to boost productivity growth to reduce in upstream and downstream activities from poverty. This implies the pursuit of more produc- mining to distribution as well as production-ori- tive ways of doing business within each existing ented engineering, information technology sector as well as to accelerate the structural trans- and financial services. Due to this innovative- formation across sectors, reallocating resources ness, manufacturing creates particularly large from low productivity activities in agriculture, knowledge spillovers which enhance productiv- small trade and simple services to new activi- ity in non-manufacturing activities (Cohen and ties that are knowledge-intensive and exploit the Zysman 1987; Helper et al. 2012). Also, productiv- advantages of inter-firm specialisation. ity convergence–that is, sectors that are further away from the technological frontier increase Countries get richer as they diversify their pools productivity faster than the more advanced ones– of knowledge and create institutions that facili- appears to be especially rapid in manufacturing tate continuous recombination of this knowledge (Rodrik 2013). Last but not least, most manufac- for the improvement of existing or creation of turing goods are easily tradable and can there- new and better goods and services. Empirically, fore be exported to world markets almost without the link between increasing diversification of demand restrictions;4 this allows countries to production and employment and rising incomes reap economies of scale even when their internal is very clear, at least at early stages of develop- market is constrained by low purchasing power ment (Imbs and Wacziarg 2003). Except for some and small population size. Historically, there is natural resource exporting countries, the coun- a clear correlation between phases of economic tries that achieve the highest incomes are the growth and expansion of the manufacturing ones that are able to combine diversified knowl- sector (Rodrik 2006). Looking at the post-WWII edge pools in ways that are difficult to emulate by performance of developing countries, the most others (Hausmann et al 2014).3

3 This dynamic is captured in the Economic Complexity Index developed by Hausmann and others. In their Atlas of Economic Complexity, the authors show how strong the correlation is when natural resource exporters are excluded (Hausmann et al. 2014). 4 By comparison, “an expansion of non-tradables is self-limiting, as the domestic terms of trade eventually turns against non-tradables, choking off further investment and growth“ (Rodrik 2006). Green Industrial Policy - Concept, Policies, Country Experiences

impressive growth stories were based on export- computerization in the next few decades. Apply- 4 led growth in manufactures, particularly in East ing the same methodology to developing coun- Asia (Stiglitz and Yusuf 2001; Commission on tries, the World Development Report 2016 finds Growth and Development 2008). even higher automation potentials–two thirds of today’s jobs in developing countries could be However, the share of manufacturing value added lost to automation, for instance–but assumes that in GDP tends to have a historical maximum, automation will proceed more slowly due to time beyond which it starts to decline. This transi- lags in technology adoption (World Bank 2016:219). tion towards post-industrial economies happens due to three factors. First, technological progress In the past, reallocating workers from low produc- in manufacturing reduces demand for workers tivity agriculture to export-oriented light manu- and shifts employment to services, where the facturing activities was a powerful driver of potential for automation is not as big. Second, industrialisation and productivity growth. East as incomes rise, demand shifts away from food Asian economies especially benefited from this and manufactures to increasingly differentiated shift, from Korea and Taiwan in the 1970s to more services. Third, manufacturing industries become recent industrialisation experiences in China, more and more knowledge-intensive and there- Vietnam and Cambodia. Progress in labour-saving fore create demand for specialised production-ori- technologies, however, is now likely to radically ented services in areas such as engineering, reduce the opportunities for boosting produc- information technology and finance. It should be tivity through the attraction of investment in noted that the boundaries between manufactur- labour-intensive export industries. At the same ing and services are increasingly blurred and the time, globalization creates new opportunities for interdependency of manufacturing and services industrial development as it facilitates access to increases. This is reflected in an increasing share hitherto inaccessible technologies and markets. of value added from services embodied in manu- However, it can also accelerate premature dein- facturing products (OECD 2015). dustrialisation as it puts newly emerging small industries in direct competition with highly The problem of today’s developing economies is competitive global corporations that have accu- that, with the exception of some East Asian coun- mulated knowledge and network externalities tries, manufacturing value added and employ- over decades and, on top of that, exploit the econ- ment tend to stagnate at very low levels. Most omies of scale associated with globalized markets. developing countries are moving from agricul- Even in the latecomer countries’ own domestic ture or mining as their main economic drivers to markets, imports often stifle local industry devel- services without going through a proper process opment. While a number of highly competitive of industrial development, a process known as firms and regional clusters in developing- coun premature deindustrialisation (Rodrik 2016). In tries have been able to reap the opportunities of Latin America, manufacturing industry`s contri- global markets, such successes have been the bution to GDP and employment has peaked exception rather than the rule. This explains why early at a much lower level than one would have since the 1950s, “very few [countries] have become expected from the patterns of today’s industri- high-income economies. Most developing coun- alised countries and is now shrinking. In Africa, tries have become caught in what has been called manufacturing industries are stagnating at a low a middle-income trap, characterized by a sharp level (Diao et al. 2016). The same study finds that deceleration in growth and in the pace of produc- labour productivity is stagnant or even declining tivity increases” (Agénor et al. 2012). in the modern sectors. Given the importance of manufacturing and modern services as drivers The few successful upgraders include oil export- of diversification and productivity growth these ers and Eastern European countries benefiting trends cast doubts on Latin America’s and Africa’s from EU accession, as well as East Asian coun- prospects for future economic growth and welfare. tries. The latter in particular placed emphasis on technological learning and capacity building, Two factors are particularly important to under- especially in manufacturing. They had insti- stand these trends: labour-saving technological tutions in place to manage structural change, progress and globalization (Rodrik 2014). New providing coordination for the emergence of new technologies are reducing demand significantly economic activities, nurturing entrepreneurship for routine labour activities in manufacturing and investing in education and skills develop- and services (Brynjolfsson and McAfee 2014). ment to ensure that human capital adapts to For the US, Frey and Osborne (2013) calculate changes in the productive structure (Amsden that about 47 per cent of jobs are susceptible to 1989; Wade 1990). The lesson from their success Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

is clear: the creation of wealthy economies is International Panel on Climate Change predicts strongly correlated with the ability to manage that if we continue to manage our economies in 5 structural change in a way that enhances produc- the same way, global mean surface temperature tivity in a socially inclusive way. will increase by 3.7°C to 4.8°C by 2100 compared to the average for 1850 to 1900 (IPCC 2014). Melting 2.2. TAKING ECOLOGICAL BOUNDARIES of polar ice and thawing of permafrost soils are INTO ACCOUNT: A GAME-CHANGER FOR two dangerous accelerators of global environmen- STRUCTURAL CHANGE tal change. But there are other big threats to the Earth System calling for urgent action including The global economy is on an unsustainable path. loss of biodiversity, ozone depletion, ocean acid- Since the industrial revolution, the world econ- ification, water shortage, soil degradation, accu- omy has grown at the expense of the environment. mulation of nitrogen in aquatic ecosystems and Natural resources have been exploited without the accumulation of chemical waste and plastics allowing stocks to regenerate, pollutants have (Rockström et al. 2009; WBGU 2014). accumulated in the biosphere, ecosystems have These fundamental threats to humanity need been degraded severely and biodiversity has been to be taken into account when thinking about lost at an alarming rate. Already in the early 2000s, further growth and structural change of econ- the Millennium Ecosystems Assessment, a UN-led omies. The way economic transactions are global assessment of the Earth’s ecosystems, currently organized largely ignores the social cost concluded that about 60 per cent of the ecosys- of resource depletion and pollution. Natural capi- tem services examined had been degraded or tal embodied in fertile soils, fresh water, clean air were used in ways that cannot be sustained (MEA and productive ecosystems is being wasted. This 2005). Similarly, UNEP (2011) summarises a series waste undermines the basis for future economic of reports showing severe overexploitation of fish development and jeopardizes the progress made stocks, increasing water scarcity, decreasing soil on social welfare (Fay et al. 2015). Therefore, we quality and unsustainable rates of deforestation. need to recognise environmental sustainability as Through product and process innovations, fundamental to the production process. resource efficiency is increasing worldwide. Put In essence, human well-being and economic differently, fewer natural resources are needed to progress need to be decoupled from non-renew- produce the same unit of output; but this increase able resource consumption and emissions (UNEP in efficiency has been quite modest, with the 2011). To make economic development sustaina- effect that GDP growth globally has more than ble, resource efficiency needs to increase at least outweighed the efficiency gains (Jackson 2016; at the same rate as economic output. The largest Wiedmann et al. 2015). This led to a situation challenge is how to achieve the steep decline in where “global material extraction more than GHG emissions needed to keep global temper- doubled in the past 30 years, from around 36 ature rise well below 2°C. To achieve this, global billion tonnes in 1980 to almost 85 billion tonnes carbon intensity would have to be reduced by 6.3 in 2013, an overall growth of 132 per cent” (Vienna per cent every year to 2100, much faster than the University of Economics and Business 2016). modest annual decline of 1.3 per cent achieved Environmental contamination also increased. between 2000 and 2014 (PwC 2015). In the case of anthropogenic greenhouse gases, emissions rose from 33 to 49 Gt CO e per year 2 So far, none of the major economies has achieved from 1980 to 2010 (IPCC 2014). Due to continued this. However, ‘absolute decoupling’ is not impos- growth of the global population and increased per sible. Enormous resource efficiency jumps are capita consumption, particularly since the turn technologically feasible: with the shift to renew- of the century, “anthropogenic pressures on the able energy, the use of smart information and Earth System have reached a scale where abrupt communication technology systems, the use of global environmental change can no longer be energy-saving technologies and, last but not least, excluded” (Rockström et al. 2009). Research on changes in consumer behaviour. To accelerate the environmental systems highlights the existence required technological and business model inno- of tipping points at which environmental change vations, however, economic incentives need to accelerates due to self-reinforcing mechanisms be set very differently. Above all, environmental and systems are unable to restore their previous costs need to be much better reflected in prices, equilibrium (Lenton et al. 2008). regulations must be tightened and subsidies for fossil fuels and other unsustainable goods and Global warming is the most pronounced threat to practices need to be phased out. human development and the environment. The Green Industrial Policy - Concept, Policies, Country Experiences

Doing so will invariably have deep and system- if catastrophic climate change is to be avoided 6 wide implications, comparable to those observed (McGlade and Ekins 2015). Fossil fuel reserves as during the first industrial revolution or the rise well as assets that depend on transforming and of information technology (Pérez 2002). It will trading fossil fuel, such as refineries, power plants change the way we farm our land and manufac- and petrol distribution networks, may therefore ture goods, where we source our energy, how we be overvalued. Rapid technological progress in transport things and how we build our infrastruc- low-carbon technologies and/or more ambitious ture and design our cities. Among the various decarbonisation policies may force the holders environmental challenges, mitigating climate of carbon assets to adjust their values, which in change will have arguably the deepest implica- turn may cause a carbon bubble shock with deep tions for structural change because it affects the repercussions for banks, pension funds and insur- energy and transport sectors that so far have ance companies (Weyzig et al. 2014). The Econo- fuelled economic development, literally. Accord- mist Intelligence Unit estimates that within the ing to the Intergovernmental Panel on Climate global stock of manageable assets the value at risk

Change, global annual CO2 emissions will need to due to climate change ranges from US$ 4.2 trillion be reduced 42 to 57 per cent by 2050, relative to to US$ 43 trillion between now and the end of the 2010, and 73 to 107 per cent by 2100 (IPCC 2014). century (EIU 2015). The Financial Stability Board To achieve such levels of decarbonisation, major recognises such asset stranding related to climate systemic changes are indispensable: Electric- change to be a relevant risk to the global financial ity generation needs to shift fully from fossil to system and therefore put a reporting system in renewable sources; as power generation is decar- place, the Task Force on Climate-related Financial bonised, transport, heating and other energy using Disclosure (TCFD 2017). In fact, some institutional sectors need to be electrified, including road traf- investors have started to withdraw from carbon fic; and resource efficiency needs to be increased assets (Schwartz 2015). Hence there are market radically across all industries, including the shift mechanisms at work that accelerate the struc- to circular economies where waste is reduced, tural change towards a low-carbon economy. reused or recycled (Fay et al. 2015). While mining and power supply industries are Some of these changes are already in full swing, most affected, structural change in other indus- others yet to come. Global energy systems–and tries is following. Regulators in all main automo- all the related manufacturing and service activi- tive markets including the European Union, USA, ties related to power generation, transmission and Japan and China, have defined roadmaps for

storage–are already undergoing a fast and radi- reducing average CO2 emission levels of new cars. cal change. Renewable energy technologies have Within a few years, these levels can no longer be been adopted widely around the world. Electricity achieved by efficiency gains in fuel-driven cars from hydro, geothermal and certain biomasses alone, forcing manufacturers to incorporate elec- can now compete with fossil fuel-based electric- tric and hybrid cars into their product range and ity, as do wind and solar power in good locations, to rapidly increase their share in overall sales. The and further cost reductions are expected. While private sector is in fact responding. Electric vehi- 15 years ago renewable energy power installa- cle deployment has recently taken off with expo- tions played a negligible role in global electricity nential growth rates, albeit from a low basis. In generation, “the world now adds more renewable 2016, the global stock of electric cars exceeded two power capacity annually than it adds in net new million, up from a few hundred ten years earlier capacity from all fossil fuels combined” (REN21 (OECD and IEA 2017). With rapidly falling battery 2016; REN21 2017). prices and increasing battery performance, electric cars will soon be fully competitive with fuel-driven Firms continuing to invest in unsustainable tech- cars (Altenburg et al. 2017, this volume). Early nologies run the risk of having to write off major movers such as Tesla and Toyota are taking market investments. According to McGlade and Ekins shares from established carmakers that have been (2015), climate research suggests that “to have at slower to adapt. Similar changes can be observed least a 50 per cent chance of keeping warming in other product categories: reflected in growing below 2°C throughout the twenty-first century, the markets for organic food, biodegradable packaging cumulative carbon emissions between 2011 and and renewable building materials, for example. 2050 need to be limited to around 1,100 gigatonnes of carbon dioxide.” As a consequence, about one Not only products will change, but also production third of oil reserves, one half of gas reserves processes and business models. Circular econ- and at least 80 per cent of known coal reserves omy models are being developed to minimise cannot be burnt and need to be kept in the ground material and energy flows through industrial Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

systems and make sure residuals of one produc- degradation undermines the ecological foundation process are used as input for another (Ellen tions for economic growth and human well-be- 7 MacArthur Foundation 2012). In energy systems, ing, most obviously in countries that depend new technologies enable the development of on economic activities in agriculture, forestry decentralized mini grids where customers can or fisheries. Second, pollution and waste typi- flexibly respond to price signals, supplying power cally reflect inefficiencies in production, and or reducing demand when the price on the grid resource-saving techniques tend to amortize is high and consuming power when it is low very quickly even without consideration of posi- (Nathaney et al. 2016). Worldwide, new business tive externalities. Third, sticking to traditional models are flourishing, models based on sharing products and processes as the worlds’ domi- rather than owning assets, in most cases facil- nant economic actors shift to greener goods and itated via online marketplaces. These include production techniques will drive a wedge between sharing of cars, accommodation or taxi services. local and global practices. This makes it more Last but not least, the certification and accredi- difficult to compete in the future, considering that tation industry is receiving a boost, as economic trade and investment treaties increasingly regu- actors are increasingly obliged to prove that their late environmental issues and that lead firms in production processes comply with various envi- global value chains impose progressively higher ronmental requirements. environmental standards. Fourth, countries should avoid getting locked into unsustainable In sum, the recognition of ecological system infrastructure and business practices because boundaries has already become a game-changer the costs of switching in the future will likely be for economic development. Incentive systems disproportionally high. Therefore, today’s invest- are changing–still too slow from an environmen- ments in high-carbon energy infrastructure may tal perspective–and a lot of experimentation is turn into financial burdens soon, as renewable happening in terms of new products and processes. energy becomes cheaper and commitments to decarbonise become binding and costly. Devel- 2.3. CHANGE WILL AFFECT oping countries are in an advantageous position DEVELOPING COUNTRIES in the sense that most of their energy and urban infrastructure is yet to be built, so they can avoid How relevant are these changes for developing costly misdirected investments in unsustaina- countries, taking into account that “many are ble infrastructure. Fifth, many new green tech- unable to keep up with the investments to satisfy nologies come with co-benefits. For example, the basic needs of their citizens, let alone the effi- investing in clean air greatly improves health cient cities, roads, housing, schools, and health conditions and reduces health-related expendi- systems they aspire to create” (Fay et al. 2015:2)? tures; and communities can be electrified at lower In fact, many people in developing countries, cost when new technologies make it easier to use including government officials, regard environ- local sources of renewable energy at small scales. mental protection a luxury their countries should Sixth, green industrial policies drive innovation. deal with at later stages of development, once the While new-to-the-world types of innovation will more pressing problems of human development mostly likely be developed in a relatively small have been solved. number of countries with strong national innovation systems, certain innovations may also Still, even when governments put their own be developed in poorer countries and drive local national social and economic objectives first, productivity growth and job creation. Table 1.1 there are strong arguments for not delaying the provides an illustrative overview of new green transition to a green economy (Padilla 2017 and product and service opportunities, differentiating Ambec 2017, this volume). First, environmental between countries by level of income. Green Industrial Policy - Concept, Policies, Country Experiences

Table 1.1: New green product and service opportunities for countries at different income levels 8 Higher middle and high income countries Low and lower-middle income countries

New Renewable energy technologies including Low- and medium tech, low cost products products high-tech components of solar photovoltaics, such as solar water heaters, solar water concentrated solar power, wind turbines and pumps, solar driers; drip irrigation systems; geothermal technologies; energy storage rainwater harvesting technologies; LPG, LNG or technologies including fuel cells and lithium- ethanol cook stoves; LNG-based three-wheeler ion batteries; electric vehicles; new lightweight taxis. materials; bioplastics; carbon capture and Inputs for global green production for which storage technologies; high performance factor endowments exist: such as lithium, rare building façades. earths, cellulosic ethanol.

New Design and operation of smart grids, closed- Simple low-cost services such as for operation services cycle eco-industrial parks, intelligent transport and maintenance of decentralized and mini systems, advanced energy management electric grid solutions; labour-intensive waste systems, electronic road pricing, tracking recycling; low-carbon livestock management; and tracing systems for environmental management of rapid transit systems. performance along value chains. Labour-intensive tasks in emerging green global value chains, such as assembly of solar panels or lithium-ion cells.

Source: Adapted from ClimateTechWiki (n.d.).

In fact, awareness of the need for green industrial fleet emissions, to restrict access of high polluting policy is clearly increasing among developing vehicles in inner cities or to oblige taxi fleets to country stakeholders. Governments of countries run on ethanol or compressed natural gas. at very different income levels have enacted green economy strategies, ranging from Ethio- 2.4. THE DUAL CHALLENGE pia, Rwanda, Cambodia and Vietnam to Mexico and China. Growing recognition is also reflected Governments around the world are thus by the 197 countries party to the 2015 Paris Agree- confronted with a dual challenge: to accelerate ment under the United Nations Framework structural change towards higher productivity Convention on Climate Change, thereby commit- in a way that is socially inclusive and to align ting to limit the increase in the global average economic development with the carrying capac- temperature to well below 2°C above pre-indus- ity of our planet. Recognizing the need to harmo- trial levels, to pursue efforts to limit the temper- nise both objectives and to make industrial policy ature increase to 1.5°C above pre-industrial levels environmentally sound is essential. In this regard, and to formulate and communicate long-term the unanimous global agreement on the 2030 low greenhouse gas emission development strat- Agenda for Sustainable Development has been a egies. So far, the existing commitments, known major achievement (UN 2015). as Nationally Determined Contributions, are not sufficiently ambitious to stay below the envis- However, not a single country has been able, so aged threshold levels (Climate Action Tracker far, to enhance the welfare of its citizens without 2015), and we may expect a gap between political increasingly depleting its resource base. In this declaration and actual implementation; but the regard, governments in search of a welfare-en- trend towards greener technologies and incentive hancing sustainable economy are entering systems is unlikely to be reversed. Especially in uncharted territory. There are obvious trade-offs the field of energy generation, renewable energy between the welfare and environmental sustaina- technologies are becoming much more cost-ef- bility objectives, at least in the short and medium ficient in many more locations. Developing and term. Ostensibly, internalizing environmental emerging economies now account for about half costs that have been externalized in the past of global renewable energy investments and the increases the apparent cost of production and market for renewables-based mini-grids is boom- reduces cost-competitiveness if competitors ing (REN21 2017). With regard to air pollution, do not have to bear these costs. Moreover, green many developing countries and municipalities investments have opportunity costs: governments have taken drastic measures to regulate trans- need to find a good balance between the necessary port. These include measures to reduce allowable investments in environmental improvements and Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

other outlays, such as those for health, education to be able to design green industrial policies and infrastructure. But there are multiple syner- while Maximising gains and minimizing costs. 9 gies as well. The search for green technologies This will be challenging, given that optimal solu- will create many opportunities for economic tions depend on situation-specific factors–from development, health benefits, improved efficien- resource endowments and techno-institutional cies and better living conditions. Governments capabilities to the distribution of power between must understand these synergies and trade-offs the polluting incumbents and green newcomers.

3. THE ROLE OF INDUSTRIAL POLICY Unfettered market-based allocation of resources interdependent investments, subsidizing early is unlikely to foster structural change in a socially entrepreneurial search processes, promot- optimal manner, one that allows for high producing cleaner industries and facilitating market tivity, broad-based societal inclusion, generalized transparency and information flows. Generally wealth and increasing environmental sustaina- speaking, industrial policy aims to reinforce or bility. In many instances, markets do not send out counteract the allocative effects of markets with the right price signals. This is the case when an the objective of restructuring economies towards initial investment that would have triggered many a better societal outcome (Rodrik 2004). It should knowledge spillovers is not carried out because be noted that industrial policy is about promot- the investment in itself does not immediately pay ing desirable structural change in general and not off in a way the individual financier could appro- limited to industry or even to manufacturing. priate the profits stemming from all the resulting secondary technological developments and Here, an important qualification is in order to market opportunities. In other words, the social avoid a common misinterpretation of industrial or public returns are larger than the individual or policy. Proponents of modern industrial policy do private returns, which is a very common phenom- not think policymakers are better than entrepre- enon given the cumulative character of tech- neurs in anticipating market opportunities. They nological development. Market-based resource know the market mechanism is a smart institu- allocation also has its limitations in the presence tional arrangement. In many regards, markets of coordination failures. These occur when the reflect what people want and how much they are viability of a new business depends on simultane- willing to pay for alternative options. Also, markets ous investments in related fields, with the effect encourage the creativity of individuals who take that no firm risks an investment unless someone personal risks in the pursuit of profits. Ideally, guarantees the necessary complementary invest- competition among firms with different business ments. Similarly, markets do not facilitate the concepts will reward efficient entrepreneurs and socially optimal level of entrepreneurial ‘discov- force less efficient ones out of the market. It is this ery’ of cost: When an investor undertakes a new process of entry, innovation and exit in a compet- activity, he or she discovers the underlying costs, itive environment that drives productivity growth as well as the benefits (Hausmann and Rodrik and determines where firms, regions, or countries 2003). However, this knowledge, especially the have relative advantages. The role of industrial likelihood of high returns, quickly becomes public policy is not to replace this creative process with and other investors will copy the business model. top-down bureaucratic planning, but to embed it This is good for the overall economy, but bad for within broader social welfare processes to improve the pioneering investor who bore the full risk of the outcomes for society at large. failure but sees his innovation profits dissipate. Beyond externalities and coordination failures, Put differently, the social value of discovering the societies tend to have different preferences, many cost of a new activity exceeds the private gains. of which cannot be fully expressed in market This is another important market failure. Similar prices–often because they imply ethical consid- market failures occur, for example, when markets erations or touch upon entrenched societal values. do not reflect the full environmental costs of an For example, people have different preferences investment or when market actors lack relevant when it comes to attaching economic values to information. cultural norms or to biodiversity. This affects This is where industrial policy enters the scene. how people strike their personal balance between It has important roles in encouraging industries economic opportunities and the related risks of, with potential knowledge spillovers, coordinating for instance, genetic engineering, global warming or nuclear energy. People also differ with regard Green Industrial Policy - Concept, Policies, Country Experiences

to the degree of social inequality or employment government intervention is necessary and 10 insecurity they are willing to accept. Also, they what instruments are best suited to overcome have different views on where, how much and market failures depends on the gap between with which measures the state should interfere what self-organized private actors would to regulate such issues. Against this background, achieve and the optimal outcomes in the public industrial policy is about facilitating stakeholder interest. This is likely to be very context-spe- dialogues on the direction of structural change, cific and to change over time. Industrial policy moderating different viewpoints, finding compro- should thus be conceived as a collaborative mises and creating consensus on broadly defined process of discovery in which public and development pathways. As well, it is about adapt- private actors closely interact and continuously ing regulatory frameworks and incentive schemes negotiate and adapt their contributions to the in such a way that creative entrepreneurial development of the respective industry. search processes are encouraged and channelled 2. Discipline. Such embeddedness obviously towards the achievement of agreed goals. This entails risks of collusion and capture by private again implies strategic collaboration between the interests. To minimise these risks, govern- private sector and governments to jointly identify ments need to maintain full autonomy in deci- barriers that need public-private coordination for sion-making and be able to use disciplining their removal. devices against abuse (Evans 1995). Govern- ments need to draw a clear line between collab- In essence, industrial policy aims to complement oration in the public interest and favouritism. the market mechanism. Given the pervasiveness This presupposes clearly defined objectives of market imperfections and the legitimacy of that are broken down into measurable perfor- investment criteria that go beyond microeconomic mance indicators. Furthermore, it requires efficiency, the question is not whether to apply monitoring and evaluation routines to contin- industrial policy or not, but how to do it (Rodrik uously check the performance of firms and 5 2004). support programmes against existing bench- marks. Governments need to have the inde- Critics point to various ways in which indus- pendence to adjust or even withdraw incentive trial policy is frequently being abused by interest packages without falling prey to lobbyists. groups. Industry lobbyists demand specific subsi- Unbundling the roles of policy formulation, dies and orchestrate resistance when subsidies funding, implementation and evaluation can shall be withdrawn. As there are usually substan- be helpful to insulate such performance-based tial information asymmetries between the lobby- systems against political interference. Putting ists and the public sector it is often easy to develop implementation out to tender, encouraging a storyline justifying subsidies. Similarly, poli- competition among service providers and ticians may claim and allocate funds to satisfy monitoring their performance through inde- electorates and protect firms in their jurisdictions pendent agencies further enhances effective- rather than to maximise public welfare on the ness. Clear and transparent rules as well as basis of scientific evidence. Keeping such political conditionality and sunset clauses are also help- capture to a minimum is indeed a major challenge. ful to keep rent-seeking behaviour in check. To cope with this challenge, three basic principles 3. Accountability. Policymakers and implement- of effective industrial policymaking should be ing agencies should be held accountable for applied when designing and implementing indus- their industrial policies. This can be achieved trial policy (Rodrik 2014; Altenburg et al. 2008): using various reporting requirements and obligations to disclosure as well as more general 1. Embeddedness. Policymakers need to maintain democratic checks and balances by central close relationships with the private sector and auditing authorities, political parties, inde- other stakeholders to get a deep understand- pendent courts and a free press. Accounta- ing of how specific economic sectors func- bility serves not only to prevent corruption, tion, what the business rationale of relevant favouritism and other forms of collusive private actors is and where bottlenecks exist behaviour but also helps to legitimize appro- that hold back improvements. To what extent priate industrial policies.

5 Countries that managed to close the technological and income gap vis-à-vis more advanced economies invariably employed a range of carrots and sticks to protect and nurture their national industries. Empirical evidence shows this for the early catching-up experiences of Germany, the United States and Japan as well as for the more recent post- World War II examples, from the early “Asian Tigers” of Korea and Taiwan to the current emergence of China. None of the countries that strictly followed the Washington Consensus, in contrast, has achieved comparable success in terms of technological upgrading, economic growth, and poverty reduction (Rodrik 2005; Chang 2009). Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

4. GREEN INDUSTRIAL POLICY: HOW IT IS DIFFERENT 11 Industrial policy is about anticipating relevant How is green industrial policy different from long-term trends of technology and market industrial policies that do not systematically development and providing incentives to adapt integrate the perspective of environmental the structure of a national economy in such a constraints? In many ways, steering investment way that it can take advantage of the change. towards a green economy is not that different As climate change mitigation and other ecologi- from steering them towards conventional indus- cal challenges increasingly influence the future trial policy objectives, such as higher value added direction of economic development, environmen- and enhanced productivity. As Schwarzer (2013:vi) tal considerations need to become a key part of puts it, “…green industries are essentially infant industrial policymaking. This is what green indus- industries, with all the characteristics of conven- trial policy is about (World Bank 2012; Hallegatte et tional infant industries and subject to the same al. 2013; Lütkenhorst et al. 2014; Pegels 2014). opportunities and challenges of promoting them.” Various information and coordination failures The boundaries between green industrial policy call for facilitation. Policymakers as well as entre- and environmental policy are not clear-cut. Envi- preneurs take decisions without knowing what ronmental policies aim at protecting and sustain- the future will look like; therefore, policies carry ably using our natural environment. Intentionally risks of misallocation and political capture, which or not, some of these policies drive structural need to be kept to a minimum. Also the available change. For example, carbon prices shift invest- instruments are very similar, including infor- ments from fossil fuels to renewable energy; mation and coordination platforms, regulations, ambitious automobile emissions standards accel- standards and labels, differential taxes and credit erate the substitution of traditional fuel with subsidies. And the three basic principles of effec- electric vehicles that in turn requires different tive policymaking apply. Finally, as in conven- types of supplier industries; environmental fiscal tional industrial policy, shifting to new green reforms that tax environmental consumption industries requires public support and therefore instead of labour may reduce the international needs to find ways for dealing with the ‘losers’ competitiveness of resource-intensive indus- and smoothing the adaptation of firms and work- tries while making labour-intensive activities force (Fay et al. 2015). This is why the researchers more competitive. Other environmental policies and practitioners concerned with green transfor- mainly induce process innovations and thereby mation can learn many lessons from the concep- have only little effect on structural change, such tual discussions about industrial policy and its as new pollution control technology upgrades in successes and failures of implementation. existing industries. In this volume, environmental and energy policies that deliberately push struc- Yet, green industrial policy is also different. It goes tural change into the desired direction are consid- beyond the traditional notion of industrial poli- ered part of green industrial policy. In addition, it cies in at least six important ways (Altenburg and encompasses policies that enhance the national Pegels 2012; Lütkenhorst et al. 2014): benefits of the green transformation in terms of higher incomes and better employment oppor- 1. the focus on environmental externalities tunities. Hence, we define green industrial policy as an additional and particularly damaging as including any government measure aimed to market failure accelerate the structural transformation towards 2. a clear predictable distinction between ‘good’ a low-carbon, resource-efficient economy in ways and ‘bad’ technologies, based on their envi- that also enable productivity enhancements in ronmental impacts, and therefore systematic the economy. steering of investment behaviour in a socially agreed direction Green Industrial Policy - Concept, Policies, Country Experiences

3. the urgency to achieve structural change within claim that ambitious caps would threaten their 12 a short period of time, to preclude the risk of international competitiveness in order to keep the catastrophic environmental tipping points cap high and the price of use rights low; and they 4. enhanced uncertainty due to long time hori- ask for assignment of free use rights. As govern- zons of some transformations as well as ments do not want to harm their national indus- dependence on policy changes tries, cap-and-trade systems so far have often 5. additional policy interfaces and therefore the failed to set ambitious caps (Helm 2010; IPCC 2014). need for particularly encompassing policy Environmental taxes, in contrast, do not guaran- coordination tee an upper limit to resource use or pollution 6. a motivation to manage global commons, because industry’s readiness to pay taxes defines such as the atmosphere and oceans, for long how much they will reduce that resource use or term sustainability, which may not always be pollution. But taxes have several advantages. As aligned with immediate national interests. they are directly set by a government authority, the additional cost for firms is more predictable. In what follows, we will address each of these Also, taxes create a double dividend as they not defining features one by one. only reduce environmental impacts but also raise revenues for the government. These can be used 4.1. THE IMPORTANCE OF to reduce other taxes or increase government ENVIRONMENTAL EXTERNALITIES spending, both of which help to build societal support for environmental tax reforms. Finally, The most obvious specificity of green indus- taxes are easier to implement than cap-and-trade trial policy is that it aims to correct the failure systems, making them particularly attractive for of markets to reflect the social costs of envi- developing countries (Schlegelmilch et al. 2017, ronmentally harmful production. For compa- this volume). nies investing in green technologies, the private return lies significantly below the social return, As market instruments that encourage entre- resulting in underinvestment. Quoted in The preneurial search and cost-effective allocation, Guardian Newspaper, regarding his report for the both cap-and-trade and environmental taxes are UK government, Stern tells us “Climate change is increasingly being applied internationally. For a result of the greatest market failure the world several reasons, however, pricing environmental has seen” (Benjamin 2007; Stern 2007). Hence the goods is not sufficient (Fay et al. 2015). One limi- theoretical case for applying industrial policies tation is that there may be other market failures to accelerate and upscale investments in green hampering green transformations–for example technologies is even stronger than it would be for those related to incomplete information, lack other technologies. of coordination or inadequate appropriability of research and development investments. Another To close the gap between private and social limitation consists in ethical concerns about pric- returns the first-best solution would be to price ing. Not everyone would agree with the basic idea the use of environmental goods, such as water or that everything nature provides can be expressed clean air. This has a big advantage: market actors in monetary values. These critics hold, for exam- can use their ingenuity to find the most cost-ef- ple, that the preferences of future generations fective way to consume less of these goods. There cannot be fully reflected in market prices. In are basically two ways that governments can addition, first-best policy instruments may not be attach prices to environmental goods: cap-and- available for political or administrative reasons. trade systems and environmental taxes on resource consumption or emissions. In cap-and- Hence policy mixes are usually required that trade systems, governments define an upper limit combine market-based instruments, regulations, for the use of a resource or emissions and then capacity building, subsidies and other compo- distribute or auction use rights among economic nents in various ways. The right combination actors, which can then be traded. This encour- depends on country conditions, such as what ages all participants to explore and implement degree of policy complexity can be handled the most cost-effective solutions6. Defining the and how well the government is insulated from cap and allocating use rights, however, is not easy lobbying pressure. Also, governments need for political reasons. Polluter lobbies typically to anticipate the trade-offs between pricing

6 For the case of climate change mitigation, Cramton et al. (2017) convincingly argue why a global carbon price would be much more effective than the current practice of individual pledges with weak review mechanisms. Rather than depending on altruism it would create a reciprocal common commitment, whereby “each country would commit to placing charges on carbon emissions sufficient to match an agreed global price formula.” Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

environmental goods and competitiveness. competing in the market place. In contrast, the Polluting industries will intentionally face higher logic of green transformations implies that where 13 costs and may therefore lose any advantage over environmentally sustainable solutions compete competitors from other jurisdictions where the with harmful ones it is in the public interest to same industries are not taxed. At the same time, accelerate the substitution rather than waiting pushing industries early on to develop clean tech- for markets to reward commercially superior nologies may result in early mover advantages if alternatives. Never and Kemp (2017, this volume) other jurisdictions impose similar conditions with discuss how standards can be used to accelerate a time lag (Porter and van der Linde 1995; Ambec the diffusion of green technologies. 2017, this volume). Third, an important part of green industrial policy is to proactively phase out harmful technologies. 4.2. SYSTEMATICALLY STEERING In some cases, such as when substances that INVESTMENT BEHAVIOUR deplete the ozone layer or greenhouse gases are concerned, it is not enough to promote the devel- The overarching objective of bringing the econ- opment and deployment of sustainable alterna- omy back into a ‘safe operating space for human- tives. Green industrial policy defines road maps ity’ necessarily gives structural change a direction and sets incentives to phase unsustainable tech- (Rockström et al. 2009). First and foremost, tradi- nologies out. Cosbey at al. (2017, this volume) tional industrial policy aims at enhancing produc- show how this can be done in practice. tivity growth and incomes but, in most cases, leaves it to market forces to find the most lucrative Fourth, while conventional industrial policy rarely technologies and business models. Green industries to affect consumer behaviour, influenc- trial policy, by comparison, is driven by scientific ing purchase decisions is an important element evidence of environmental threats. This implies of green industrial policy. Mandatory labelling a much clearer picture of which technologies programmes may help to make markets transpar- and business models are good or bad. Underlying ent and enable consumers to distinguish products green industrial policy is the Pigouvian idea of with different environmental effects. Educational steering investment behaviour systematically and programmes can encourage people to reuse and permanently towards what governments conceive recycle things. However, it should be noted that as environmentally sustainable (Spratt 2013:12). consumers do not respond perfectly to price signals. Even when new products exist that are This leads us to four peculiarities of green indus- better in many ways and cheaper, many consum- trial policy when it comes to issues of technol- ers stick to the bad old alternatives because they ogy choice and promotion: First, there needs to do not understand the situation well, because be agreement on which technologies are good their neighbours have not changed or simply for a sustainable future. This is far from trivial out of force of habit. Green industrial policy can because alternative solutions may exist, all with use a wide range of options to encourage green some trade-offs that imply difficult value judg- consumption and shift markets using advertise- ments. For example, from a decarbonisation ments, nudges and green default options, among perspective, biofuels are desirable substitutes for other schemes. A rapidly growing literature fossil fuels, but their commercial production may shows how insights from behavioural science can lead to monocultures, loss of biodiversity, higher be used to influence consumers in a pro-environ- food prices and increased pressure on unutilized mental way (Sunstein and Reisch 2013; Price 2014). vulnerable land. In a similar vein, nuclear energy and large-scale carbon capture and storage are advocated by some as necessary elements of 4.3. URGENCY TO ACT FAST AND decarbonisation strategies but rejected by others UPSCALE EXPERIMENTATION for their inherent risk of large-scale contamination. What is desirable thus depends on value Some economic activities have strong impacts on judgments, and political deals are needed to specific ecosystems or even on the entire Earth define what merits support. System. These ecosystems have a certain capacity to react to disturbances and return to their Second, there is a case for subsidizing deploy- previous equilibrium state; but thresholds exist ment of clean technologies, even beyond the beyond which such return is no longer possible point where they break even with harmful tech- and systems may collapse, in some cases with nologies. Traditional industrial policy would potentially catastrophic effects for life on Earth foster technologies only at their infant stage (Lenton et al. 2008). It is thus truly vital to avoid and withdraw support as soon as they start such tipping points. Green Industrial Policy - Concept, Policies, Country Experiences

So humankind is approaching, or even trans- be overestimated. Also, it should be noted that 14 gressing, various thresholds at the global level there are ways to limit the risks of misallocation where irreversible and catastrophic change may and share them with private investors, through happen (Rockström et al. 2009). To stay within a competitive bidding processes and other means. safe operating space, quantum leaps in resource efficiency are needed that in turn require- radi 4.4. DEALING WITH ENHANCED UNCERTAINTY cally different technologies and business models in various fields. Among the various thresholds, Many of the objectives of a green transfor- global warming is the one for which the most mation cannot be achieved in the short term. sophisticated models exist to assess: how much Decision-makers need to define long-term carbon can still be emitted to keep warming targets—such as the European Commission’s below tolerable limits, by when the world econ- target to cut the EU’s greenhouse gas emissions omy needs to become carbon-neutral, how much to 80 per cent below 1990 levels by 2050—and then the transition would cost and how much the cost define intermediate milestones and derive tech- would increase if action got delayed. While such nology and policy road maps for achieving the calculations necessarily have methodological target (European Commission 2011). limitations, they all concur in their assessment that the necessary technology switch needs to With longer time horizons, uncertainty increases. happen within the next one or two decades if Three types of uncertainty compound here. First, global warming beyond 2°C is to be avoided. Also, there is uncertainty about technologies and it is widely agreed that delayed implementation of markets. These are always difficult to predict, but mitigation measures will make it much more diffi- predictions become even more uncertain when cult and costly, if not impossible, to reach given systemic change is envisaged that stretches climate targets. Costs increase–due either to out over various decades. Moreover, considering greater environmental damages if targets are not that so far not a single country has succeeded met or to the greater stringency of the necessary in systematically decoupling economic welfare mitigation measures if the original target is main- and growth from resource consumption, there tained–and opportunities to act early on low-cost are no role models for a green economy. Govern- mitigation measures are missed (Executive Office ments of developing countries are no longer well of the President of the United States 2014). advised to emulate technologies and institutions from rich economies but rather need to find their This provides a very strong rationale for green own pathways. industrial policy that is ambitious and leads to results quickly. More mission-oriented innova- Second, there is policy uncertainty. Green indus- tion programmes are probably needed to facili- trial policy is strongly driven by politically defined tate big coordinated investments and accelerate objectives, rather than by new technologies and the development of critical key technologies, such market opportunities, which makes it essential as for energy storage (Foray et al. 2012). In addi- to have predictable and stable long-term policy tion, sunset clauses and compensation schemes frameworks in place. Political factors–such as will be needed to phase out harmful technologies, the level of ambition of policies to phase out coal as well as guarantees and subsidies to acceler- or fuel-driven automobiles, the political will to ate the dissemination of clean substitutes. Many implement carbon taxes or the willingness and of these policies involve risks of misallocation. ability to sustain preferential tariffs for renewable Governments cannot know beforehand whether a energy– strongly affect the profitability of invest- certain public investment in a new technology or ments (Karp and Stevenson 2012). At the same business model will pay off in the future. However, time, these political factors are often contested this is not an argument against such invest- and they change when new administrations take ments. If there are good reasons to assume that office or public pressure mounts for or against experimentation creates knowledge spillovers certain measures. to society that are large compared to the private Third, there is uncertainty about ecosystem return on investment, supporting such exper- dynamics. Policy frameworks need to respond to imentation makes sense. While this holds for environmental changes that are difficult to predict industrial policy in general, it is particularly rele- because the effects of environmental distur- vant for policy areas where solutions are needed bances are non-linear. If disturbances are minor urgently and quickly, because any delay leads to and time-bound, systems tend to return to their an escalation of costs. In the presence of tipping previous equilibrium state; but thresholds exist points in the Earth System, industrial policy beyond which systems may collapse (Scheffer support for sustainable technologies can hardly Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

2009). Natural science research is thus needed specialised firms make parallel investments in to understand the inherent economic risks in batteries and charging infrastructure, and new 15 ecosystems’ dynamics and inform policymakers technical standards are developed as well as and investors. credible policy road maps that signal the phase- out of fuel engines. Well-managed coordination Overall, uncertainty tends to be larger in green processes with strong political backing are needed transformations than in ordinary market-driven to bring such change about. transformations. Governments thus have a particularly important role in reducing uncer- While the failure of markets to bring about simulta- tainties and related investment risks. They can neous large-scale investments in complementary do so by drafting roadmaps to augment inves- fields is one of the key reasons for adopting indus- tors’ confidence in long-term policy targets. The trial policy in general, such coordination failure is EU’s energy targets and its reduction targets particularly problematic when change is system- for automotive fleet emissions are cases in wide and transformative, as it is the case in the point. Governments can also provide guaran- greening of economies. As economic subsystems tees. Renewable energy laws in many countries tend to be interlocked, transformative change in combine guaranteed purchases from independ- one subsystem tends to have repercussions on ent power producers with guaranteed minimum others. When energy systems shift from fossil to prices, known as feed-in tariffs. Also, they can renewable biofuel, it has unintended consequences anchor certain long-term targets in international on land and food prices; if dams are built to create treaties. All these measures help to lock polices hydropower, it affects water supply for agriculture; in, shielding them from political cycles and if agriculture shifts to organic, markets change for increasing investment security. producers of fertilisers and agrochemicals.

On the other hand, given technological uncer- Hence, new policy interfaces become relevant to tainties, policy frameworks also need a certain understand the interdependencies and optimise degree of flexibility to respond to changing the outcomes for all sectors of a society, especially circumstances, such as new environmental risk civil society. Political decision-makers need to take assessments, emerging technological options or various stakeholder interests into account when changing prices. The challenge is thus to find a designing policies. This, again, is not a unique good balance between providing ‘directionality’ feature of green industrial policy, but it is a particu- and encouraging entrepreneurial experimentation larly relevant element of it, because stakeholders (Mazzucato 2013). lobbying for the preservation of jobs provided by polluting industries are often better organized 4.5. ADDITIONAL POLICY INTERFACES AND than environmental groups or social activists, for NEED FOR POLICY COORDINATION instance. The political feasibility of green economic reform therefore often depends on compromises Often the green transformation goes far beyond among these interests, even though some compro- the replacement of individual technologies. What mises may reduce the policies’ effectiveness in is pursued is a transformative change of entire terms of environmental performance. Hence poli- production systems, such as the energy system or cies need to be co-designed by participants with the transport system. This requires simultaneous strictly environmental interests and those help- changes on several fronts including the developing to maximise social welfare, as well as those ment of various interdependent technologies and supporting economic or industrial interests. business models and the related adjustments of regulations and support systems. Such systemic 4.6. THE MOTIVATION TO PROTECT NOT change is unlikely to proceed smoothly without a ONLY NATIONAL INTERESTS BUT ALSO proactively coordinating public agency. For exam- GLOBAL COMMONS ple, no company would dare to invest in offshore wind parks unless other investors ensure the The agenda of green industrial policy is partly synchronized establishment of a grid that allows driven by international agreements, such as to bring electricity to the shore and further on to the Paris Agreement, where governments have the main centers of demand. That, in turn, requires committed to decarbonise their economies; the complex plan approval procedures involving Montreal Protocol on the protection of the ozone affected communities as well as regulatory provi- layer; and other treaties concerning fisheries sions for electricity transmission through vari- management, marine and air pollution control, ous systems. Likewise, carmakers are unlikely to proper management of hazardous materials or shift from fuel-driven to electric cars unless other genetic diversity of crops. All these agreements Green Industrial Policy - Concept, Policies, Country Experiences

have a differential effect on industries: restrict- market leaders and creating thousands of manu- 16 ing the expansion of resource-using industries facturing jobs. But after a few years, Chinese low and/or forcing them to use product substitutes cost competitors started to crowd out the German or to develop different technologies and busi- industry, and many German solar panel manu- ness models, for example. They all serve to solve facturers went bankrupt (Lütkenhorst and Pegels collective action problems at an international, 2014). This led observers to criticise German solar often global, scale. policy as one that created a market but failed to build an industry (Paris Tech Review 2012). This is another important difference vis-à-vis From a global perspective, however, Germany’s traditional industrial policy. Normally, govern- support for the industry enabled the first large- ments will try to implement the required policies scale photovoltaic module production that in turn in such a way that they enhance the productiv- brought unit prices down–by 80 per cent in the ity, competitiveness and employment potential of period 2009–2015 (IRENA 2016). The supply trig- its domestic industries. Thus “…the benefits vest gered worldwide deployment of this green tech- almost exclusively in the implementing country, nology. In terms of national industry, China was and the costs are borne by foreign producers–a the main beneficiary (Pegels 2017, this volume). traditional mercantilist outcome” (Cosbey 2017, this volume). When it comes to green industrial More recently, a similar story is happening with policy, there are likely to be positive externalities electric vehicles, where China is the pioneer that for the global environment and for other coun- accelerates the global diffusion of greener tech- tries. As pioneering countries develop solutions nologies. Here, the government heavily subsidises for environmental pressures, their industries the shift from internal combustion to electric will allow other countries to solve their domestic engines, thereby making China the lead market problems at a lower cost (Fankhauser et al. 2012). where new models are developed, tested and These spillover effects have important implica- rolled out in mass production (Altenburg et al. tions for the way we judge public subsidies. While 2017, this volume). Given that China is the world’s in normal conditions, industrial policy should not largest automobile market and served largely by enter into a competition on the basis of subsi- multinational carmakers, China’s industrial policy dies; a ‘subsidy race’ among nation states can be is accelerating the cost degression of electric cars a good thing as it accelerates the development and batteries to the benefit of the rest of the world. and global dissemination of green technologies (Rodrik 2014:471). Currently, such positive spillovers from national policies to global green technology diffusion have Hence, policymakers need to ponder the effects mainly happened unintentionally. To accelerate on domestic industry and global commons. the worldwide diffusion of technological solutions Ideally, green industrial policy improves both in for managing global commons, more international tandem, but outcomes differ in practice. Germa- technology cooperation is needed. This requires ny’s solar policy was temporarily seen as a expansion of mechanisms that fund international successful industrial policy in the traditional, research and development, knowledge sharing national sense. A fairly high guaranteed feed-in and technical assistance to developing countries, tariff contributed to the diffusion of solar panels such as the Global Environmental Facility and the in the German market, and local companies UNFCCC’s Technology Mechanism. reaped early mover advantages, becoming world

5. CONCLUSIONS Governments, those of developing countries in competitiveness and more and better jobs. This particular, are facing a dual challenge: they need chapter shed some light on the policy options. to advance structural change towards higher productivity while at the same time decoupling In the long term, there is no trade-off among human well-being and economic progress from social, economic and environmental objectives: resource consumption and emissions. This there is no human development or economic implies the need to better integrate industrial success on an uninhabitable planet. In the short and environmental policies–rethinking the term, however, there are trade-offs to be consid- former from an environmental perspective and ered. For example, pricing environmental goods exploring how the latter can contribute to greater puts an additional, but appropriate, burden on producers who have been able to externalize Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

these costs in the past and may jeopardize their When it comes to the urgent need to decou- competitiveness and employment capacity. At ple human development from non-renewable 17 the same time, moving to greener economies resource consumption and pollution, market holds many benefits even from a purely economic signals alone are even less effective. First, envi- perspective, as greater resource efficiency lowers ronmental costs are not sufficiently reflected costs, early movers may develop new markets, in market prices, and second, the green trans- asset-stranding is avoided and new job opportu- formation requires system-wide changes–such nities open. as radical redesign of the predominant energy systems–that cannot take place without well-co- Governments need to understand the oppor- ordinated interventions to deal with multiple tunities and pitfalls to minimise the costs and coordination failures. As a consequence, green maximise the gains, striking a fine balance industrial policy is in many aspects similar to between environmental objectives and competi- traditional industrial policy, but it has to come to tiveness, industrial development and job agendas. grips with additional layers of complexity: The It should be noted that this is a classic political need to avoid negative environmental externali- transformation project rather than a techno- ties requires specific policy instruments, such as cratic exercise. First, because finding the right cap-and-trade systems and environmental taxes. balance of competing objectives and decid- The urgency to phase out polluting technologies ing among various alternative techno-institu- and replace them with green substitutes within tional pathways implies value judgments; and short time frames calls for more comprehensive second, because transformations always create and aggressive research and development and winners and losers. Governments need to create technology diffusion programmes. The neces- consensus on the direction of change and facil- sary restructuring of entire economic sub-sys- itate compromises among stakeholders. This in tems presupposes long-term strategies spanning turn presupposes a thorough understanding of several decades that offer clearly defined interim interest groups and their power resources. When targets and, where necessary, credible long- some elements of the transformation encounter term subsidy schemes and financial guarantees. strong resistance from interest groups, govern- Furthermore, particularly comprehensive policy ments need to identify less contested no-regret coordination and consensus-building mecha- options, such as resource efficiency programmes nisms are needed to manage radical systemic with quick economic returns or pollution control changes at the interface of industrial and envi- programmes with benefits for large parts of the ronmental policies. Thus as a rule, green indus- society, to propel the transformation forward. trial policy is more ambitious than most industrial Fortunately, there are already strong forces driv- policies of the past. This increases the risks of ing the green transformation, such as the decreas- misallocation and political capture. We argue, ing price of renewable energy generation, an however, that the long-term costs of not taking, increasing number of environmentally conscious or of delaying, action are much larger than the citizens ready to wield their voices and their risks of losing part of the industrial policy funds buying power, lead firms in global value chains to non-performing programmes and that there pushing for greener supplies, international trea- are proven policy design principles that greatly ties demanding greener standards and institu- reduce the risks of ineffectiveness and capture. tional investors pulling out of carbon assets. Countries are likely to reap multiple benefits when Accelerating structural change always requires they take a proactive stance and accelerate their a proactive public sector. The case for industrial green transformation in a way that combines policies is theoretically very strong and backed by social, economic and environmental objectives evidence. Public policies have a role in supporting and when they design their policies according to research and development, subsidizing entrepre- the three basic principles of effective industrial neurial cost discovery, coordinating complemen- policymaking. Well-designed green industrial tary investments that need to be undertaken policies are crucial not only for bringing economic simultaneously and facilitating information development back into the safe operating space sharing and technological learning. In these for humanity; they can also serve as an invest- areas, social benefits tend to be much higher than ment programme for long-term productivity gains. returns to private investors, so that markets alone cannot provide socially optimal solutions. Green Industrial Policy - Concept, Policies, Country Experiences

18 REFERENCES Agénor, P.-R., Canuto, O., & Jelenic, M. (2012). Avoiding Commission on Growth and Development. (2008). Middle-income Growth Traps (Economic Premise The Growth Report: Strategies for Sustained No. 9). Washington, DC. Growth and Inclusive Development. Washington Altenburg, T., & Lütkenhorst, W. (2015). Industrial D.C.: The International Bank for Reconstruction Policy in Developing Countries: Failing Markets, and Development/The World Bank. Weak States. Cheltenham: Edward Elgar Publish- Cosbey, A. (2017). Trade and Investment Law and ing Ltd. Green Industrial Policy. In Altenburg, T., & Altenburg, T., & Pegels, A. (2012). Sustainability-ori- Assmann, C. (Eds.). (2017). Green Industrial ented innovation systems: Managing the green Policy. Concept, Policies, Country Experiences transformation. Innovation and development, 2(1), (pp. 134–151). Geneva, Bonn: UN Environment; 5–22. German Development Institute / Deutsches Insti- tut für Entwicklungspolitk (DIE). Altenburg, T., Feng, K., & Shen, Q. (2017). Electric mobility and the quest for automobile industry Cosbey, A., Wooders, P., Bridle, R., & Casier, L. (2017). upgrading in China. In Altenburg, T., & Assmann, In with the good, out with the bad: Phasing out C. (Eds.). (2017). Green Industrial Policy. Concept, polluting sectors as green industrial policy. In Policies, Country Experiences (pp. 185–198). Altenburg, T., & Assmann, C. (Eds.). (2017). Green Geneva, Bonn: UN Environment; German Industrial Policy. Concept, Policies, Country Expe- Development Institute / Deutsches Institut für riences (pp. 69–86). Geneva, Bonn: UN Environ- Entwicklungspolitk (DIE). ment; German Development Institute / Deutsches Institut für Entwicklungspolitk (DIE). Altenburg, T., Rosendahl, C., Stamm, A., & Drachenfels, C. von. (2008). Industrial policy: a key element of Cramton, P., MacKay, D. J., Ockenfels, A., & Stoft, S. the social and ecological market economy. In (2017). Global Carbon Pricing: The Path to Climate Deutsche Gesellschaft für Technische Zusam- Cooperation. MIT Press. menarbeit (Ed.), The social and ecological markt Diao, X., Harttgen, K., & McMillan, M. (2017). The economy: a model for Asian development? Changing Structure of Africa’s Economies (World (pp. 134–153). Eschborn: GTZ. Bank Policy Research Working Paper No. 7958). Ambec, S. (2017). Gaining competitive advantage Washington D.C. with green industrial policy. In Altenburg, T., & Economist Intelligence Unit (EIU). (2015). The cost Assmann, C. (Eds.). (2017). Green Industrial Policy. of inaction: Recognising the value at risk from Concept, Policies, Country Experiences (pp. climate change. London: Economist Intelligence 38–49). Geneva, Bonn: UN Environment; German Unit. Development Institute / Deutsches Institut für Ellen MacArthur Foundation. (2012). Towards the Entwicklungspolitk (DIE). circular economy. Isle of Wight: Ellen MacArthur Amsden, A. H. (1989). Asia’s next giant: South Korea Foundation. and late industrialization. New York: Oxford European Commission (2011): A Roadmap for moving University Press. to a competitive low carbon economy in 2050. Benjamin, A. (2007). Stern: Climate change a’market Communication from the Commission to the failure’. The Guardian, 29. Retreived from www. European Parliament, the Council, the Euro- theguardian.com/environment/2007/nov/29/ pean Economic and Social Committee and the climatechange.carbonemissions Committee of the Regions. Brussels: EC. Brynjolfsson, E., & McAfee, A. (2016). The second Evans, P. B. (1995). Embedded autonomy: States and machine age: Work, progress, and prosperity in a industrial transformation. Princeton paperbacks. time of brilliant technologies. New York, London: Princeton N.J.: Princeton University Press. W.W. Norton & Company. Executive Office of the President of the United Chang, H.-j. (2009). Industrial policy: Can we go States. (2014). The cost of delaying action to stem beyond an unproductive confrontation? Paper climate change. presented at the Annual World Bank Conference Fankhauser, S., Bowen, A., Calel, R., Dechezlepretre, on Development Economics, Seoul. A., Grover, D., Rydge, J., & Sato, M. (2013). Who will ClimateTechWiki. (n.d.). Technology information. A win the green race? In search of environmental Clean Technology Platform. Retrieved from www. competitiveness and innovation. Global Environ- climatetechwiki.org/technology-information mental Change. (23(5)), 902–913. Climate Action Tracker. (2015). 2.7°C is not enough – Fay, M., Hallegatte, S., Vogt-Schilb, A., Rozenberg, J., we can get lower. Climate Action Tracker Update. Narloch, U., & Kerr, T. M. (2015). Decarbonizing Cohen, S. S., & Zysman, J. (1987). Why manufacturing development: Three steps to a zero-carbon future. matters: The myth of the post-industrial econ- Climate change and development series. Wash- omy. California Management Review. (29.3), 9–26. ington D.C.: World Bank. Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies

Foray, D., Mowery, D. C., & Nelson, R. R. (2012). Public McGlade, C., & Ekins, P. (2015). The geographical R&D and social challenges: What lessons from distribution of fossil fuels unused when limiting 19 mission R&D programs?. Research Policy, 41(10), global warming to 2 °C. Nature. (527), 187–190. 1697–1702. Millennium Ecosystem Assessment (MEA). (2005): Frey, C. B., & Osborne, M. A. (2013). The future of Ecosystems and Human Well-being: Synthesis. employment: How susceptible are jobs to comput- Island Press, Washington, DC. erisation? Retrieved from www.oxfordmartin. Nathaney, P., Collison, K., Finan, R., Haydel, J., & Johal, ox.ac.uk/publications/view/1314 H. (2016). Energizing Policy Revolution for the Hallegatte, S., Fay, M., & Vogt-Schilb, A. (2013). Grid Revolution. (ICF White Paper No. September Green Industrial Policies: When and How. World 2016). Bank Open Knowledge Repository. Retrieved Never, B., & Kemp, R. (2017). Developing green tech- from https://openknowledge.worldbank.org/ nologies and phasing them in. In Altenburg, T., & handle/10986/16892 Assmann, C. (Eds.). (2017). Green Industrial Policy. Hausmann, R., Hidalgo, C. A., Bustos, S., Coscia, M., Concept, Policies, Country Experiences (pp. Simoes, A., & Yildirim, M. A. (2014). The Atlas of 87–101). Geneva, Bonn: UN Environment; German Economic Complexity. Cambridge: MIT Press. Development Institute / Deutsches Institut für Hausmann, R., & Rodrik, D. (2003). Economic develop- Entwicklungspolitk (DIE). ment as self-discovery. Journal of development OECD. (2015). Service-manufacturing linkages. In economics, 72(2), 603–633. OECD Science, Technology and Industry Score- Helm, D. (2010). Government failure, rent-seeking, board 2015 (pp. 202–203). OECD Publishing. and capture: The design of climate change policy. OECD, & IEA (2017). Global EV Outlook 2017. Two Oxford review of economic policy, 26(2), 182–196. million and counting. Paris: OECD. Helper, S., Krueger, T., & Wial, H. (2012). Why Does Paris Tech Review. (2012). The German Manufacturing Matter? Which Manufactur- solar energy crisis: Looking for the ing Matters?,. Brooking: Metropolitan Policy right incentive scheme. Retrieved from Programme. www.paristechreview.com/2012/04/13/ Imbs, J., & Wacziarg, R. (2003). Stages of Diversifica- german-solar-crisis/?media=print tion. American Economic Review, 93(1), 63–86. Pérez, C. (2002). Technological Revolutions and Intergovernmental Panel on Climate Change (IPCC). Financial Capital. The Dynamics of Bubbles and (2015). Climate change 2014: mitigation of climate Golden Ages. Cheltenham and Northampton: change (Vol. 3). Cambridge University Press. Edward Elgar Publishing. International Renewable Energy Agency (IRENA). Pegels, A. (Ed.). (2014). Green industrial policy in The power to change: Solar and wind cost reduc- emerging countries. Routledge studies in ecolog- tion potential to 2025. Bonn. ical economics: Vol. 34. London: Routledge Taylor & Francis Group. Jackson, T. (2016). Prosperity without growth: Foun- dations for the economy of tomorrow (Second Pegels, A. (2017). Germany: The energy transition edition). London: Routledge. as a green industrial development agenda. In Altenburg, T., & Assmann, C. (Eds.). (2017). Green Karp, L. S., & Stevenson, M. (2012). Green industrial Industrial Policy. Concept, Policies, Country Expe- policy: Trade and theory. Policy research working riences (pp. 166–184). Geneva, Bonn: UN Environ- paper: Vol. 6238. Washington, D.C.: World Bank. ment; German Development Institute / Deutsches Lenton, T. M., Held, H., Kriegler, E., Hall, J. W., Institut für Entwicklungspolitk (DIE). Lucht, W., Rahmstorf, S., & Schellnhuber, H. J. Porter, M. E., & van der Linde, C. (1995). Green and (2008). Tipping elements in the Earth’s climate Competitive: Ending the Stalemate. Harvard Busi- system. Proceedings of the national Academy of ness Review. (73), 120–134. Sciences, 105(6), 1786–1793. Price, M. K. (2013). Using field experiments to address Lütkenhorst, W., Altenburg, T., Pegels, A., & Vidican, environmental externalities and resource scar- G. (2014). Green industrial policy: Managing trans- city: Major lessons learned and new directions formation under uncertainty. Discussion paper for future research. Oxford review of economic / Deutsches Institut für Entwicklungspolitik: policy. (30(4)), 621–638. 28/2014. Bonn: Dt. Inst. für Entwicklungspolitik. PwC (2015). Conscious uncoupling? Low Carbon Lütkenhorst, W., & Pegels, A. (2014). Stable policies Economy Index 2015. October 2015. PwC LLP. – turbulent markets: Germany’s green industrial policy: The costs and benefits of promoting solar Renewable Energy Policy Network for the 21st PV and wind energy. IISD report. Winnipeg: Inter- century. (2016). Renewables 2016 Global Status national Institute for Sustainable Development. Report. Paris: REN21. Mazzucato, M. (2013). The entrepreneurial state: Renewable Energy Policy Network for the 21st Debunking public vs. private sector myths. century. (2017). Renewables 2017 Global Status London/New York: Anthem Press. Report. Paris: REN21. Green Industrial Policy - Concept, Policies, Country Experiences

Rockström, J., Steffen, W., Noone, K., Persson, Å., Stiglitz, J. E., & Yusuf, S. (2001). Rethinking the East 20 Chapin, F.S., Lambin, E.F., Lenton, T.M., Scheffer, Asian Miracle. Washington, D.C, New York: World M., Folke, C., Schellnhuber, H.J., Nykvist, B., Wit, C. Bank. A., Hughes, T., van der Leeuw, S., Rodhe, H., Sörlin, Sunstein, C. R., & Reisch, L. A. (2014). Automatically S., Snyder, P. K., Costanza, R., Svedin, U., Falken- green: Behavioral economics and environmental mark, M., Karlberg, L., Corell, R. W., Fabry, V. J., protection. Harvard Environmental Law Review. Hansen, J., Walker, B., Liverman, D., Richardson, K., (38), 127–158. Crutzen, P., & Foley, J. A. (2009). A safe operating Task Force on Climate-related Financial Disclosure space for humanity. Nature, 461(7263), 472–475. (TCFD). (2017). Final Report.: Recommendations Rodrik, D. (2004). Industrial Policy for the Twen- by the Task Force on Climate-related Financial ty-First Century. Discussion Papers / Centre for Disclosure,. Basel. Economic Policy Research: no. 4767. London: United Nations (UN). (2015). Transforming our World: Centre for Economic Policy Research. The 2030 Agenda for Sustainable Development. Rodrik, D. (2006). Industrial development: Stylized New York. facts and policies.: Draft chapter prepared for the United Nations Environment Programme (UNEP). U.N.-DESA publication Industrial Development for (2011). Towards a green economy: Pathways to the 21st Century. sustainable development and poverty eradication Rodrik, D. (2008). Growth Strategies. In P. Aghion & : a synthesis for policymakers. Nairobi: UNEP. S. Durlauf (Eds.), Handbooks in economics: Vol. Vienna University of Economics and Business (2016). 22. Handbook of economic growth (pp. 967–1014). Global material extraction by material category, Amsterdam: Elsevier. 1980–2013. Retrieved from www.materialflows. Rodrik, D. (2013). Unconditional convergence in net/materialflowsnet/trends/analyses-1980-2013/ manufacturing. The quarterly journal of econom- global-material-extraction-by-material-cate- ics. (128). gory-1980-2013/ Rodrik, D. (2014). Green Industrial Policy. Oxford Wade, R. (1990). Governing the market: Economic review of economic policy. (30(3)), 469–491. theory and the role of government in East Asian Rodrik, D. (2016). Premature deindustrialization. industrialization. Princeton, N.J.: Princeton Journal of Economic Growth. (21), 1–33. University Press. Scheffer, M. (2009). Critical transitions in nature and Weyzig, Francis, Kuepper, B., van Gelder, J. W., & van society. Princeton University Press. Tilburg, R. (2014). The Price of Doing Too Little Schlegelmilch, K., Eichel, H., & Pegels, A. (2017). Pric- Too Late. The impact of the carbon bubble on the ing environmental resources and pollutants and EU financial system: A report prepared for the the competitiveness of national industries. In Greens/EFA Group – European Parliament (Green Altenburg, T., & Assmann, C. (Eds.). (2017). Green New Deal Series). Industrial Policy. Concept, Policies, Country Expe- Wiedmann, T. O., Schandl, H., Lenzen, M., Moran, riences (pp. 102–119). Geneva, Bonn: UN Environ- D., Suh, S., West, J., & Kanemoto, K. (2015). The ment; German Development Institute / Deutsches material footprint of nations. Proceedings of Institut für Entwicklungspolitk (DIE). the National Academy of Sciences of the United Schwartz, J. (2015). Norway Will Divest From Coal States of America, 112(20), 6271–6276. in Push Against Climate Change. NY Times. Wissenschaftlicher Beirat der Bundesregierung 05/06/2015. Globale Umweltveränderungen (WBGU). (2014). Schwarzer, J. (2013). Industrial policy for a green Human Progress within Planetary Guardrails: A economy. IISD report. Winnipeg, Manitoba, Contribution to the SDG Debate (Policy Paper No. Beaconsfield, Quebec: International Institute for 8). Berlin. Sustainable Development; Canadian Electronic World Bank. (2012). Inclusive green growth: The path- Library. way to sustainable development. Washington Spratt, S. (2013). Environmental taxation and devel- D.C.: World Bank. opment: A scoping study. IDS working paper: Vol. World Bank. (2016). World Development Report 2016: 433. Brighton: IDS. Retrieved from www.ids.ac.uk/ Digital Dividends. Washington DC: World Bank. publication/environmental-taxation-and-development-a-scoping-study Stern, N. (2007). The economics of climate change: The Stern review. Cambridge: Cambridge Univ. Press. Chapter 1 Green Industrial Policy: Accelerating Structural Change Towards Wealthy Green Economies PART 2: 21 THE ECONOMIC AND SOCIAL CO-BENEFITS OF GREEN TRANSFORMATION Green Industrial Policy - Concept, Policies, Country Experiences 22 CHAPTER 2 WHAT CAN DEVELOPING COUNTRIES GAIN FROM A GREEN TRANSFORMATION?

Emilio Padilla Chapter 2 What Can Developing Countries Gain from a Green Transformation?

1. INTRODUCTION 23 This chapter reviews the reasons for developing The chapter then explores, in section 3, the differ- countries to green their economies and to resist ent gains that environmental and green industrial deferring action on environmental concerns for policies may produce when designed to transform later phases of development. It first examines the economies in developing countries. Significant relationship between economic growth and envi- synergies can build between some green poli- ronmental pressures. The findings do not support cies and social and economic gains in developing a defence of growing first and cleaning later or a countries, though there may also be trade-offs PART 2 PART justification of inaction based on differentiated that require careful consideration. responsibilities for environmental degradation. These two arguments are elaborated in subsections 2.1 and 2.2, respectively.

2. THE RELATIONSHIP BETWEEN ECONOMIC GROWTH AND ENVIRONMENTAL PRESSURES This section discusses two main arguments. income levels enable societies to reduce their First, a hypothesis based on the environmen- environmental impact. To explain this relation- tal Kuznets curve has frequently been used ship, Grossman and Krueger (1991) identify three to support that economic growth will gener- effects: scale, composition and technology. In this ate the solution to environmental problems, so scenario, the size of the economy, the scale effect, emphasis on environmental and green indus- would increase environmental pressures. At the trial policies can be addressed later in devel- same time, the environmental pressures would oping countries. Second, developing countries’ be reduced by the structural change that accom- lesser responsibility for historical environmen- panies economic growth, the composition effect, tal degradation, especially for global problems, is and by ongoing technological improvements, the used to suggest that poor countries should not technology effect. If the composition and technol- make efforts to green their economies. ogy effects progress in the same direction and are strong enough to compensate the scale effect, a 2.1. IS IT SMART TO GROW FIRST AND turning point could appear in the trend tracking CLEAN UP LATER? THE ENVIRONMENTAL environmental degradation and economic growth. KUZNETS CURVE-ARGUMENT Beyond that turning point, environmental degradation would decrease as the economy grows. The relationship between economic growth and Assumptions about structural change of an econ- environmental degradation is a very complex omy explain the compositional effect: Industriali- one. The debate on this relationship was domi- sation—the increased relevance of manufacturing nated for many years by the hypothesis—based on and extractive activities and the mechanisation an environmental version of the classic Kusnets of agriculture—involves more intensive exploita- curve—that emerged in the early 1990s with the tion of natural resources and subsequent gener- seminal study of Grossman and Krueger (1991) on ation of waste. However, there is a point where the potential environmental effects of the North the relevance of services starts to increase. The American Free Trade Agreement. That study was technology effect assumption is that technologies followed by a series of empirical articles analysing improve, in ways that bring environmental bene- the relationships on economic growth and envi- fits, with economic development. Supposedly, in ronmental consequences for different pollutants combination these two effects would reduce envi- and countries (Shafik and Bandyopadhyay 1992; ronmental pressure. Panayotou 1993). This argument has, however, several flaws. An SCALE EFFECT VERSUS COMPOSITION AND TECHNOLOGY EFFECTS increased share of services in total production does not necessarily entail a lower environmen- According to the environmental Kuznets curve tal impact. Some services demand inputs and hypothesis, economic growth increases envi- pull emissions and material resources from other ronmental pressures at early stages of develop- sectors. Such services include wholesale and ment; but, after a turning point, high per capita retail trade, public administration, and hotels and Green Industrial Policy - Concept, Policies, Country Experiences

restaurants, a sector strongly related to tourism EMPIRICAL EVIDENCE 24 activities (Alcántara and Padilla 2009; Piaggio The empirical evidence for the hypothesis is not et al. 2015). In addition, even though the propor- conclusive. While the environmental Kuznets tion of services may increase, this does not mean curve seems more relevant to local pollutants that the number of industries producing goods with clearly perceived consequences for health will decrease with income growth. Therefore, the and local environments, such as particulate composition effect may reduce the pressure per matter or sulphur dioxide, it seems less appli- unit of production but not the absolute environ- cable to global pollutants. There is empirical mental pressure (Roca et al. 2001). That is why, it evidence that different countries follow differ- may lead to a reduction in the intensity of resource ent paths in this relationship, so what is found to use or emission per unit of output, a relative decou- hold for a group of countries may not be valid for pling of environmental pressure from economic the path followed by individual country (Piaggio growth, but not to a reduction of total environmen- and Padilla 2012). Some authors also argue that tal degradation, which would be considered an the environmental Kuznets curve could appear absolute decoupling from economic growth. to improve in developed economies because they As for the assumptions about technological change, have exported their polluting activities to devel- any positive consequences for the environment oping countries (Suri and Chapman 1998). following economic growth will only happen with the appropriate policies in place. Without these, IMPORTANCE OF ENVIRONMENTAL POLICIES new technologies may often result in a higher rate Even in the case of evidence supporting the of environmental degradation: a salient example hypothesis, the level of pollution estimated to is how improvements in fish location technology produce the turning point is usually too high for resulted in devastation of fish populations. In addi- most countries to risk and would deliver unbear- tion, while technological improvement may lead able environmental degradation with irreversible to a more efficient use of resources in productive damages if developing countries followed the processes, sometimes they involve new produc- projected path (Stern et al. 1996). Many environ- tive processes presenting new environmental mental damages cannot be reversed and so the pressures, as has resulted from hydraulic fractur- previous level of environmental quality cannot ing, or fracking, of geologic formations to allow oil be recovered (Arrow et al. 1995). Examples of and gas extraction. Even in the case of efficiency these irreversible environmental losses include improvements, the final effect on the environment soil erosion, aquifer destruction, or extinction is uncertain, due to a tendency of consumers to of species or ecosystems. Other problems with use more of a resource when the services obtained unpredictable consequences include irreversible from it become less expensive through improved changes to the Earth’s climate system, with asso- efficiencies, called a rebound effect (Schipper and ciated impacts that will confound us for centuries Grubb 2000; Gillingham et al. 2013). at a minimum.

ENVIRONMENTAL QUALITY AS A LUXURY GOOD Therefore, the solution to environmental problems requires appropriate and prompt political Another interpretation of the environmental action because economic growth alone will not Kuznets curve hypothesis presents environmen- solve environmental problems. In fact, one of tal quality as a luxury good (Selden and Song the most relevant conclusions from the debate 1995; Martinez-Alier 1995). In other words, as on the environmental Kuznets curve is precisely incomes increase, demands for environmental the importance of resolute environmental poli- quality increase and environmental degradation cies in making economic growth compatible will be remedied. This explanation ignores that with sustainable development (Ekins 1997). poor populations, disproportionally composed Appropriate environmental policies can deter- of women, often directly depend on the quality mine less damaging paths for the industriali- of environmental resources to meet their basic sation of developing countries, paths than can needs (UN 1995). Another flaw of this approach benefit from lessons learned and avoid mistakes is that environmental quality is not a private experienced by countries that industrialised good that can be bought in the market. Instead, at an earlier stage (Dasgupta et al. 2002). Thus, it depends on environmental policies that are appropriate environmental policies can help less decided in the political arena (Roca and Padilla developed countries to pass over environmental 2003). Therefore, environmental quality strongly degradation phases. Just as some societies have depends on countries’ political systems and the skipped the phase of using landline telephones strength of their institutions. or desktop computers, smart policies will move Chapter 2 What Can Developing Countries Gain from a Green Transformation?

directly to the use of clean technologies and First, projections and experience suggest that efficient processes, without repeating pollut- many developing countries are most threatened 25 ing processes and mistakes of environmental by global environmental problems (IPCC 2014). degradation that industrialised countries experi- Therefore, they should have a strong interest in enced. These cleaner technologies and efficient the strength and enforcement of global environ- processes have already been researched, devel- mental agreements so the remedies to these prob- oped and demonstrated in more developed coun- lems can be applied. However, to succeed, these tries, so they are ready to be adopted and adapted global environmental agreements need devel- to developing countries’ needs. oping countries to participate. This can be well illustrated with the case of climate change: it is 2 PART 2.2. HISTORICAL RESPONSIBILITY FOR now evident that no international agreement can ENVIRONMENTAL DEGRADATION? be effective if emerging economies do not take part. This fact became explicit in the negotiations Inequalities in the contribution to environmen- that achieved the Paris Agreement, to which they tal degradation have served to suggest that less committed. The same evidence of commitment developed countries should grow first and leave to finding solutions applies for most global envi- environmental concerns for a more affluent ronmental problems, such as biodiversity loss future. These differences in the contribution to or ozone depletion. Global environmental agree- environmental degradation have been referred ments can also provide additional benefits for to as historical responsibility, or differentiated developing countries. responsibility, for the current degradation of natu- Second, local environmental pressures—such as ral resources and environmental services: devel- air and water pollution, soil erosion and resource oped countries industrialised through intense depletion—produce immediate repercussions on consumption of fossil fuels, minerals, forests, the welfare of developing countries’ citizens, due food and fibre and other natural resources. They to their higher dependence on natural resources also burdened the carrying capacity of the envi- for direct consumption and income generation and ronment by dumping waste into the air, water and their greater vulnerability to environmental risks. land. In contrast, developing nations have started Future development cannot be sustained by follow- only recently to industrialise. As well, given ing the same path of environmental degradation as this legacy, today’s poorer countries are under the one adopted by industrialised countries. strong pressure to reduce poverty and improve the livelihoods of their citizens. Therefore, some The next section analyses the opportunities that proponents of the concept of differentiated may be associated with green industrial policies responsibilities argue that developing countries and the actions oriented to green the economy of are not in the position to make economic sacri- developing countries. fices for environmental reasons—and that it would not be fair to ask them to. However, neither limited historical liability nor pressing economic needs are good reasons for developing countries to reject development and implementation of sound environmental policies. Green Industrial Policy - Concept, Policies, Country Experiences

26 3. THE BENEFITS THAT INTRODUCING GREEN INDUSTRIAL POLICIES AND TRANSFORMING TO A GREEN ECONOMY WILL DELIVER TO DEVELOPING COUNTRIES As subsection 2.1 demonstrated, the ‘grow first, (IPCC 2014). Lack of clean freshwater availability clean up later’ argument rests on weak empiri- and health problems associated with pollution cal grounds, has several conceptual flaws, and are clear examples of how environmental degra- ignores the risk of irreversible ecological damages dation imposes a high burden on the welfare of and dependence on natural resources, especially poor populations. According to the World Health by the poor. However, a concern of many develop- Organization (2016a), around 3 billion people in ing countries is that the cost involved to achieve low and middle-income countries cook and heat a green transformation is too high and cannot their homes using solid fuels of biomass and coal, be afforded by low-income countries with more highly inefficient sources that produce indoor pressing needs than greening their economies. pollution. As a consequence, 4.3 million people a However, this need not be the case. Instead, many year die prematurely (WHO 2016a).. This affects policy measures are affordable and deliver signifi- especially women and children, due to their cant co-benefits in association with transforming gender roles and household responsibilities, such to a green economy. This can be attributed to a as cooking and spending a lot of time indoors large part to market failures, such as environmen- (World Bank 2003). Measures that provide access tal or knowledge externalities, for example, or to to clean fuel and technologies could reduce this government failures, such as subsidies to environ- death toll at a moderate cost and improve the mentally harmful activities. These failures lead welfare and the economy of these low-income to poor natural resource management, industry populations. Estimates suggest that, in addition, inefficiencies, or unexploited opportunities in the outside air pollution caused 3 million premature development and diffusion of industries or tech- deaths worldwide in 2012, with 88 per cent occur- nologies. Correcting these problems pays off from ring in low and middle-income countries (WHO an economic point of view for consumers, but also 2016b). In general, measures to improve air qual- for industrial producers. Governments should ity, as well as the availability and quality of fresh explore the potential synergies between develop- water, will reduce negative health effects that ment and environmental protection that would could help to increase labour force productivity facilitate their particular transition to a sustaina- (OECD 2001). ble development path. The following subsections describe twelve good reasons for considering the 3.2. DETERIORATION OF ENVIRONMENTAL introduction of green industrial policies and trans- RESOURCES AND SERVICES forming to a green economy. UNDERMINES THE POTENTIAL FOR FUTURE GROWTH OF DEVELOPING 3.1. ENVIRONMENTAL IMPROVEMENT COUNTRIES BRINGS HIGH QUALITY ENVIRONMENTAL GOODS AND SERVICES AND BETTER Degrading the environment for the sake of HUMAN HEALTH increased economic growth is not a rational policy because economic growth is highly dependent on The first obvious benefit of policies promoting the availability of natural resources and on the a green transformation of the economy is to environment’s capacity to assimilate waste. This have better access to high quality environmen- is particularly true for developing countries, given tal goods and services and better conditions their citizens’ dependency on natural resources for human health. The welfare of low-income and services for direct consumption and income populations in developing countries is closely generation. Environmental degradation is already related to their access to, and the quality of, imposing high costs on several developing coun- environmental goods and services and they are tries: averaging at 8 per cent of GDP across a therefore the most seriously affected by envi- sample of countries that represented 40 per cent ronmental degradation (Broad 1994; Jehan and of the developing world’s population in a World Umana 2003; Bowen and Fankhauser 2011; World Bank analysis (World Bank 2012). An improve- Bank 2012). In addition, the lack of resources and ment in the management of natural resources and proper institutions limits these citizens’ capac- services can certainly increase their economic ity to react and adapt to environmental degra- productivity, such as when the stock of a renew- dation and makes them more vulnerable to it able resource has recovered, and can contribute Chapter 2 What Can Developing Countries Gain from a Green Transformation?

to improve both the environment and economic tradable quotas were implemented. These find- opportunities. Tackling market and government ings suggest that tradable quotas can be a benefi- 27 failures in natural resource management would cial element in the management of some fisheries improve environmental quality, worker productiv- while alternative or complementary measures ity and public welfare. would be required in others. Another example that seemed to deliver positive results is the Chil- 3.3. MEASURES TO AVOID THE OVERUSE ean Jack Mackerel Individual Transferable Quotas OF COMMON-POOL RESOURCES WILL System (Kroetz et al. 2016). Some authors suggest, however, that community rights-based manage-

INCREASE RESOURCE PRODUCTIVITY 2 PART ment that limits the harvest would be more Environmental goods can be common-pool successful in contributing to poverty alleviation in resources. These are limited resources that are the context of small-scale fisheries in developing shared, when there is rivalry over consump- countries (FAO 2005). tion and it is difficult to exclude users. These These are examples of measures that, besides resources frequently suffer from the open access their environmental benefit of resource sustain- problem: there are no clear owners and there is ability, pay for themselves in economic terms by no management system restricting the use of the solving the previous mismanagement of envi- resources (Ostrom 1990; Aguilera 1994). When the ronmental resources. Hence, they can be seen as resource is overexploited, the benefits accrue to no-cost measures, at least over the longer term. In those who overused the resources and the costs addition, achieving more stable and sustainable accrue to those who did not. Many low-income harvests could make the development of resource people around the world depend on common- processing industries more feasible. pool resources in the form of fisheries, forests or pastureland. Proper management of these common-pool resources would reverse their dete- 3.4. MEASURES TO INCREASE RESOURCE rioration and increase their economic productiv- EFFICIENCY OFTEN PAY FOR ity. According to the nature of the resource and THEMSELVES the institutional framework, different measures could combine to induce proper management. The Key improvements from the application of market enforcement of property rights or of clear social instruments and complementary measures norms is usually required for a solution. designed for green economic transformations are found in environmental efficiency gains. These In some cases the introduction of economic are realized as lower resource use and lower incentives can help to recognise the value of the pollution per unit of production that also may resource and its scarcity. An example of these produce significant short-term economic gains. incentives is the payment for environmental Improvements in production processes that save services, an approach that can reverse degra- resources can pay off the initial costs in a short dation of ecosystem services and help alleviate period of time. In some situations resources are poverty (Pagiola et al. 2005), if applied in an appro- simply wasted, as many economic processes priate manner. However, economic incentives are are inefficient due to market distortions, inertia, not always the solution: In some cases they may lack of proper information or bad management. conflict with existing values and social incen- In many industrial processes, existing technolo- tives for conservation, so any such incentives gies could reduce energy consumption and pollu- should be carefully adapted to each context. In tion. These savings could provide net economic the case of fisheries, the open access problem has benefits rapidly, besides having a positive envi- led to overfishing and the exhaustion of resources, ronmental impact. According to the World Bank with a strong decrease in the yield and associ- (2010) energy efficiency investments can produce ated jobs and income. An efficient way to manage a triple dividend: greater energy savings, fewer the problem of overfishing would be through the emissions and more jobs. Worldwide, every addi- distribution of tradable quotas–limiting total tional dollar spent on energy efficiency saves resource capture. This tool seems highly effec- more than twice that investment on the supply tive in preventing fisheries collapse and increas- side with higher savings in developing countries ing the productivity of the resource. Costello et (Bosseboeuf et al. 2007; World Bank 2010). Thus, al. (2008) and Heal and Schlenker (2008) found energy efficiency can present not only a ‘no-cost’ empirical evidence supporting this hypothesis measure in many cases but often constitutes a through the analysis of more than 11,000 fisheries. ‘negative-cost’ thus saving money over the mid to Chu (2009) found evidence of a positive effect in long-term (Figure 2.1). 12 of the 20 analysed fish stocks where individual Green Industrial Policy - Concept, Policies, Country Experiences

There are great potential savings to be gained savings is significant for developing countries 28 through energy efficiency in buildings. This is a because of their current lower efficiency levels particularly powerful point for developing coun- and early phase of industrialisation. According to tries, as their building standards are yet to be the IEA (2008), energy consumption in the indus- formulated and they will soon be building most of trial sector could be reduced by 20 to 25 per cent the infrastructure they will be depending on for with existing technologies and best practices the next few decades (World Bank 2010). In the contributing to growth. case of manufacturing, the potential for energy

Figure 2.1: Key areas of regulation

Abatement cost

€ per tCO2e 60

0 20 40 -30

-60 Abatement potential GtCO2e per year -90

-120

-150

Energy efficiency Long-term stable international Mechanism to drive regulation, e.g., technical system for power and industry selected key technologies standards for buildings down the learning curve and transportation Targeted systems for agriculture and deforestation linked to national development agendas

Source: Copyright (c) 2017 McKinsey & Company. All rights reserved. Reprinted by permission. In their report “Pathways to a low-carbon economy: Version 2 of the global greenhouse gas abatement cost curve”,

McKinsey and Company identify four key areas of regulation to reduce CO2 emissions at the lowest cost. The graph demonstrates that Policy area (1) Energy efficiency regulation will result in a net economic benefit or 'negative cost' for the economy: This is the case as energy efficiency regulation, such as standards or technical norms, can overcome market imperfections that currently prevent the private sector from realising energy efficiency measures that would result in a net profit (for which the cost savings of reducing energy consumption are higher than the initial investment in energy efficiency measures) (McKinsey & Company 2009).

Various measures in the power sector could pay 3.5. ONCE LOCKED IN TO RESOURCE- for themselves, of which one is the utilisation of INTENSE AND POLLUTING ENERGY off-grid electricity systems. Rural areas in low-in- AND TO UNSUSTAINABLE URBAN come countries have scattered demand and low INFRASTRUCTURE DEVELOPMENT, consumption levels that challenge the logic of COUNTRIES WILL FIND IT DIFFICULT providing grid electricity. Off-grid solar power TO SWITCH TO GREEN PATHWAYS IN systems can provide modern energy services THE FUTURE at lower costs than diesel generators, while reducing the noise and fumes (IFC 2007; Bowen A way to improve environmental and economic and Fankhauser 2011). Very small scale units efficiency is through better urban planning, avoid- can replace commonly used kerosene lanterns ing city sprawl models, because more dense and improving indoor air quality and reducing fire compact city layouts reduce energy consumption hazards (REN21 2016). and pollution, especially in transport (Rickwood et al. 2008; OECD 2012a). Poor countries are still developing most of their infrastructure and are rapidly urbanising. Besides the tragedy of irreversible environmental damages, the economic Chapter 2 What Can Developing Countries Gain from a Green Transformation?

costs of greening the economy at a later stage of Fund, fossil fuel and electricity subsidies reached development may be too high due to infrastruc- US$ 4.9 trillion in 2013 and 5.3 trillion in 2011, or 29 ture, sectoral and technological lock-in. The costs 6.5 per cent of that year’s global GDP (Coady et al. of replacing long-lived fossil-fuelled machin- 2017). Other types of support that lead to the over- ery and infrastructure or of reversing inefficient use of natural resources include mineral extrac- urbanisation models may be too high. Present tion, water irrigation and agriculture. Dobbs et technological and infrastructure decisions may al. (2011) and the World Bank (2012) estimate the lock countries into a high-carbon economy that total annual economic support to natural resource can involve much greater costs in a future world overexploitation to be between US$ 1 trillion and of decarbonisation (World Bank 2012). The risk of US$ 1.2 trillion annually. Eliminating fossil fuel 2 PART stranded assets, relinquished because of sudden or subsidies alone would have reduced fossil fuel premature write-downs, devaluations or conver- air pollution deaths by 55 per cent, raised revenue sion to liabilities, as well as divestment pressures, by 4 per cent, and increased social welfare by 2.2 highlight these financial pressures and point per cent of global GDP in 2013, while at the same towards the benefits of a low-carbon transition time cutting global carbon emissions by 21 per (Baron and Fischer 2015; Caldecott et al. 2016). cent (Coady et al. 2017). Some estimates suggest that three quarters of global energy subsidies are Focusing on economic growth, following the same provided by the governments of developing coun- path as developed countries and greening later tries (Coady et al. 2017). would be much more costly than following a path now to transform to a green economy. Develop- In addition to their direct damage, energy subsi- ing countries have the advantage, from the envi- dies lock economies into inefficient technologies ronmental point of view, that so much of their that will be more difficult and costly to replace in infrastructure is yet to be built, so they can use the future. This possibility makes perverse incen- current knowledge of best practice to minimise tives even more toxic for the futures of developing environmental and economic costs in their devel- countries, where infrastructure is to a great extent opment paths. They can take advantage of current still to be built, despite the immediate attraction knowledge to avoid carbon intensive development they offer to quell potential social unrest. While and to adopt cheaper, cleaner, and more efficient the poor are credited as the source of such unrest, production options. This would also avoid the the actual benefits of these perverse incentives high costs of replacement as these carbon-in- most often accrue to the elites in developing tense technologies and processes become obso- countries, perpetuating inequities (Clements et al. lete, likely long before they break down. 2014). Examination of evidence gathered between 2005 and 2009 in twenty developing countries 3.6. REDUCING INEFFICIENT AND HARMFUL found that the richest 20 per cent of households INCENTIVES CAN IMPROVE THE capture an average of six times more in fuel ENVIRONMENT AND PROVIDE POSITIVE subsidies than the poorest 20 per cent (Arze del ECONOMIC IMPACTS Granado et al. 2012). The environmental and economic benefits of In some cases governments introduce inefficient, subsidy reform have been widely discussed or perverse, incentives that encourage or accel- (Coady et al. 2017). However, with a restructur- erate environmental deterioration. A direct way ing of benefits, there will always be winners and to improve the environment while enhancing losers and sometimes any change can incite polit- economic efficiency is by eliminating the incen- ical and social resistance. Historically, and more tives to overexploit resources or to produce and recently through the networks of social media, consume polluting processes. These perverse this resistance arises from misleading informa- incentives are highly inefficient as they distort tion about the value of the environmental damage market values away from the actual values of and the need for subsidies. The most responsi- environmental resources and the material costs ble remedy is better information presenting the of environmental degradation. short and long term benefits of the reforms, and These incentives often take the form of subsi- explaining the damages imposed by perverse dies and are maintained because of the pressure incentives especially in terms of inequity and of special interest groups or the fear that elimi- injustice. To reinforce the message, reforms nating them would lead to social unrest. As an should also entail compensation measures target- example, world fisheries subsidies were estimated ing the groups most affected, especially lower and at around US$ 35 billion in 2009 (Sumaila et al. middle-income populations, so that the reduction 2016). According to the International Monetary of subsidies is not seen as an unfair policy (Jakob Green Industrial Policy - Concept, Policies, Country Experiences

et al. 2015). In return, newly available public reve- volumes of environmental goods. Many opportu- 30 nues could be used to finance programmes aimed nities are linked with the export of green goods to increase political acceptance of these reforms, and services, including tapping into a growing including information campaigns, social cushion- market of environmentally aware customers in ing or additional measures to facilitate the transi- high-income countries. Countries should also tion to a greener economy (Clements et al. 2014; bear in mind that environmental awareness and Jakob et al. 2015). international policy stringency grow over time, thus increasing demand for green products and 3.7. DEVELOPING COUNTRIES MAY services. Not investing in that growth market LEVERAGE NEW COMPETITIVE might therefore be associated with high oppor- ADVANTAGES IN ENVIRONMENTAL tunity costs. On the risk side, there may be the GOODS AND TECHNOLOGIES BY possibility of losing part of domestic demand, at ADAPTING/PRODUCING GREEN least short-term, if this reorientation of produc- PRODUCTS tion involves changes in cost. Several emerging economies and developing 3.8. DEVELOPING COUNTRIES CAN countries have been very successful in developing SIGNIFICANTLY PROFIT FROM THE new exporting sectors based on green products. ADAPTATION AND DEPLOYMENT Specific industrial policies have been successfully OF ALREADY PROVEN GREEN implemented in China, Brazil, Ethiopia, Indonesia, TECHNOLOGIES Tunisia, Mexico or Morocco to develop new sectors with green aspects (OECD 2012b). For the most Some emerging economies can successfully part, developing countries cannot afford the risk compete in green innovation sectors. However, at the cutting edge of new technologies. However, lower-income countries may not be in the they can adapt the best available technologies to position to invest in cutting-edge green their circumstances and promote the development innovation research and development due of green sectors with the appropriate support, to their lack of resources, human capital and leading to increased exports and jobs. Green technological capacity. They may rather focus industrial policies should focus on untapped but on facilitating the diffusion and adaptation clear comparative advantages, and many devel- of existing green technologies, to make their oping countries have very rich natural endow- own industry more competitive and to reduce ments that could inspire development of green environmental impacts. There is much to be industries, such as solar energy in North Africa gained in promoting the adoption of green (Vidican Auktor 2017, this volume). While high-in- innovation and technologies. To accelerate this come countries have traditionally pioneered green process of adoption, adaptation, and deployment, innovations, developing countries are catching up, international openness is a relevant factor particularly East and South Asia as represented (World Bank 2012). This requires trade and by China and India (Dutz and Sharma 2012; World investment policies that encourage the import Bank 2012). In 2010, green goods and services and deployment of foreign green technologies. constituted 3.4 per cent of exports from develop- Imports play an important role in transferring the ing countries compared to 6 per cent from high-in- green technology embodied in green products. come regions (Dutz and Sharma 2012). The current Foreign direct investment also plays a relevant production structure of developing countries role in technology transfer (Moran 2015). To already indicates that they could catch up grad- facilitate the adoption of green technologies it ually: One indicator for this is the share of ‘green will be necessary to build local receptivity in the and close-to-green products’ that amounted to form of capacity development. To maximise the about 13 per cent of developing countries’ exports co-benefits obtained in the adoption of developed in 2010, compared to about 15 per cent in high-in- green technologies, capacity development should come countries. Compared to the export percent- be customised to local needs and environments ages for green goods and services in developing (World Bank 2012). Low-income countries should countries and high-income countries, the shares also focus on innovations that can be adapted to for green and close-to-green product exports are meet the needs of poor consumers at very low much more alike in the two country groups (Dutz costs per unit (World Bank 2012). These include and Sharma 2012). formal or informal innovations that seek to provide more product with less resources for This indicates substantial opportunity for devel- more people (Prahalad and Mashelkar 2010; World oping countries in regard to increasing export Bank 2012). A good example is the wide adoption Chapter 2 What Can Developing Countries Gain from a Green Transformation?

of off-grid solar electricity enabled and made 3.11. NOT PARTICIPATING IN INTERNATIONAL affordable through business models based on ENVIRONMENTAL AGREEMENTS 31 responsive community-oriented customer service, AND FOLLOWING THE HIGH- mobile phone credit accounts, units that can be POLLUTION DEVELOPMENT PATH de-activated remotely upon default and profit COULD UNDERMINE THE LONG- gains as the customer pool increases in volume TERM POTENTIAL FOR EXPORTS (REN21 2016). AND INDUSTRIAL DEVELOPMENT OF DEVELOPING COUNTRIES

3.9. GREEN ENERGY POLICIES INCREASE More stringent environmental policies in the 2 PART ENERGY SECURITY, REDUCING THE international context may have a series of VULNERABILITY OF ECONOMIES TO OIL consequences for developing countries. Some PRICE SHOCKS AND DEPENDENCE ON governments are concerned that enforcing envi- FUEL IMPORTS ronmental policies in their countries could affect the external competitiveness of their industry. In According to Awerbuch and Sauter (2006), oil some cases, pressure from industry lobbies have price increases and ongoing instabilities depress led to introduction of exemptions to resource economic growth by raising inflation and unem- or pollution-intensive sectors, which introduce ployment and devaluing financial and other distortions and clearly reduce the efficiency and assets. The deployment of renewables could help effectiveness of environmental policies (Ekins countries to avoid costly economic losses by and Speck 1999; Baranzini et al. 2017). An option displacing fossil fuels. to deal with competitiveness concerns that is Increased efficiencies and renewable energy gaining weight in the academic and policy debate sources provide additional co-benefits by increas- on international environmental agreements is to ing energy security and reducing the vulnerabil- introduce trade sanctions on those countries not ity of economies to oil price shocks as well as the adopting similar environmental measures. These dependence on fossil fuel imports. could take the form of compensatory taxes, such as border tax adjustments. In case these measures were generalised, this could induce exporting 3.10. POSITIVE EMPLOYMENT EFFECTS DUE developing countries to adopt similar environ- TO GREEN SECTORS AND TECHNOLOGY mental standards to avoid sanctions. However, INSTALLATION AND SERVICING imposing these kinds of sanctions on developing MAY PRODUCE MORE DECENT JOBS countries is hardly justifiable on ethical grounds if AND PROVIDE A POSSIBLE DOUBLE there are not compensatory—or environmentally DIVIDEND motivated—transfers from developed countries to Green sectors and technologies are likely to be developing ones. more labour-intensive in the short term and As more stringent environmental policies, green then to provide more jobs in total over the long industrial processes and market preferences for term, so the change in focus from conventional to green production and products become wide- renewable energy sources could also create jobs spread in industrialised economies, sticking to (Fankhauser et al. 2008). Job creation could also polluting economies may reduce opportunities increase through public investment that develops for that developing country to export to a devel- green production and technologies. There may oped country market. If more strict environmental also be a double dividend when changing from standards are adopted in developed nations, these taxing labour to taxing environmental damages– are likely to be imposed on imported products as an environmental benefit plus an increase in well, with the same effect on exporting opportu- labour demand resulting from environmental nities of developing countries that do not make fiscal reforms that make labour cheaper relative adequate efforts to green their industries. to energy (Pearce 1991; Schlegelmilch et al. 2017, this volume). In the specific case of international climate agreements, countries have incentives to free ride, as The change to a greener economy may also climate policies can be considered a public good. involve transition costs in terms of employment That is, free riders will benefit from the reduced when the induced change in relative prices impacts of climate change, while not making any temporarily leads to a situation in which pollut- effort to reduce their own emissions. To solve this ing industries and jobs are phased out while new problem, an increasingly popular idea is that there green alternatives emerge incrementally (World should be mechanisms to avoid the free riding Bank 2012; UNIDO and GGGI 2015). Green Industrial Policy - Concept, Policies, Country Experiences

of those countries not participating in climate emissions, particularly from agricultural and 32 agreements. One example of this is the proposal deforestation activities and other land use change made by Nordhaus (2015) of a ‘climate club’ in (Bowen and Fankhauser 2011). Efficiencies alone which those countries participating in the global could encourage developing countries to minimise agreement could impose trade sanctions to those the global costs of mitigation. Flexible mecha- not participating, in such a way that the incen- nisms, such as the clean development mechanism, tives to free ride disappear. However, as stated by allow developing countries to obtain funds from Nordhaus (2015) low-income countries cannot be richer countries for projects that reduce emis- expected to match commitments of rich countries, sions, mainly in the energy and manufacturing so they should only assume the obligations of club sectors, or as payments for reduced emissions membership once they pass to the middle-income from deforestation and forest degradation. These group. Another proposal in climate policy discus- schemes help achieve global reductions at lower sions is that these sanctions could take the form costs, because they are implemented in develop- of carbon border tax adjustments, in such a way ing countries, and thus make sense for countries that imports from free riding countries could be with binding obligations. They also involve addi- charged according to their carbon content as if tional financial, technological transfer and capac- they were subject to a similar policy (Lockwood ity development support required for successful and Whalley 2010) or similar measures, such as implementation that can also facilitate the making them buy emission permits, oriented to transformation of the energy and manufacturing ‘level the carbon playing field’ (Houser 2008). sectors of the host developing countries. These mechanisms can provide incentives for devel- Various studies analyse the design of carbon-moti- oping countries to participate in climate agree- vated border tax adjustments and how they could ments, and motivate developing countries to be made consistent with WTO rules (Cottier et al. expedite the green transformation of their indus- 2009; Mattoo et al. 2009; Jakob et al. 2014; Rocchi tries. International carbon trading could initiate et al. 2015). These measures could eliminate the countries to follow a low-carbon development incentives to free ride and are defended on the path and provide gains to those performing well grounds of avoiding negative competitive effects (Bowen and Fankhauser 2011). Global climate of carbon pricing in energy and emission-inten- agreements must also include climate-finance for sive industries. These measures therefore help developing countries to help their mitigation and to level the playing field for green and low GHG adaptation efforts and to build their resilience to products and provide an incentive for countries to climate change. The Paris Agreement indicated invest in their green industrial capacity as part of the non-binding plan will provide US$ 100 billion global value chains. per year for the period 2020 to 2025 to developing countries for decarbonisation, including technol- In this context, not participating in those agree- ogy transfers, and for adaptation measures. This ments and following a high-carbon development amount will be increased after this period. The path could undermine the long-term potential for fund, and its eventual increase, argues in favour of exports and industrial development of developing decarbonisation measures and in environmentally countries. Early adoption of green industrial poli- sound adaptation and capacity building in devel- cies in developing countries’ economies and an oping countries. early green transformation could provide countries with an early mover advantage as regards A successful example of the implementation low-carbon industrial capacity and the produc- of a global agreement for another global envi- tion of environmental goods and services. It can ronmental problem is the Montreal Protocol on also reduce costs of transition at a later point and Substances that Deplete the Ozone Layer. This avoid costs associated with trade-related policy protocol was signed in 1987 and ratified by 197 measures of importing countries, such as carbon countries, most of them developing countries. tax equalisation or similar measures. Together with the original Vienna Convention for the Protection of the Ozone Layer, they became the 3.12. INTERNATIONAL ENVIRONMENTAL first UN treaties to achieve universal ratification AGREEMENTS CAN ALSO INVOLVE (UNEP 2014; 2015). The success of the Montreal FINANCING OPPORTUNITIES AND Protocol, which induced a transformation in the TECHNOLOGY TRANSFERS FOR manufacturing processes of the involved indus- DEVELOPING COUNTRIES tries, has much to do with the special treatment that it gives to developing countries: They were Developing countries have significant opportu- given longer periods for meeting reduction nities for low cost reductions of greenhouse gas targets and provided with means to do so, with Chapter 2 What Can Developing Countries Gain from a Green Transformation?

the establishment of the Multilateral Fund for was created to promote sustainable development the Implementation of the Montreal Protocol in and the protection of the global environment by 33 1991 (Brander 2013). This fund provided financial providing funds for the development of projects assistance to developing countries for meeting with environmental benefits, and has been indis- their mitigation targets through projects oriented pensable to facilitate the participation of develop- to replace polluting technologies. The Multilateral ing countries in various international agreements. Fund provided a strong incentive for developing countries to join the Protocol and implement Finally, the Green Climate Fund was established measures that reduced the use of ozone-depleting in 2010 by 194 countries party to the United substances. Other factors contributed to success- Nations Framework Convention on Climate 2 PART ful implementation of the Montreal Protocol, Change (UNFCCC) as a financial mechanism of including the availability of affordable alternatives the UNFCCC supporting the global response to and the existence of trade sanctions that made climate change. Its resources are allocated to possible punishments for non-adherence credible. low-emission and climate-resilient projects and programmes in developing countries. Developed Another important finance tool that has provided countries formally agreed to jointly mobilise incentives to developing countries to partic- US$ 100 billion per year by 2020 (GCF 2016). To ipate in various international environmental receive any of these funds, and the development agreements is the Global Environment Facility opportunities they offer, developing countries (GEF 2016). The fund is now a partnership of 183 must be parties to the relevant multilateral envi- countries, multilateral implementing agencies, ronmental agreements. organizations and the private sector. The fund

4. CONCLUSIONS A review of the arguments that are usually made governance of institutions, with the capacity to against the application of environmental and avoid rent-seeking and political capture by vested green industrial policies in developing countries interests. These measures include feed-in tariffs shows that these neglect important information. for renewables, tax breaks for innovative firms or Neither the belief that economic growth alone will green public procurement. Public sector efforts automatically lead to the best possible future nor face huge risks of failure from interest group pres- the differentiated responsibilities between indus- sure, rent-seeking behaviours or imperfect infor- trialised and developing countries for current mation, problems that lead to outcomes favouring environmental pressures offer persuasive argu- specific groups rather than society as a whole ments for rejecting the adoption of environmental (Pegels 2014; Rodrik 2014). and green industrial policies. In addition, besides the environmental improvement that can be Policymakers should also be able to eliminate obtained through these policies, there are a series support programmes once they are no longer of potential co-benefits for developing countries justified or accomplishing their objectives (World in terms of welfare enhancement and social and Bank 2012). Some policies that promote promis- economic gains. Policies should be designed in an ing green industries and technologies could fail if integrated manner, taking account of a country’s these problems are not addressed adequately and specific circumstances, to maximise gains and appropriately. In addition, several green indus- co-benefits and manage potential risks. trial policies require short-term investments that provide benefits only over the long term. The If some green policies are win-win options, why lack of financial institutions able to support such are they not developed at a more rapid pace? First, investments hinders their implementation in various government policy and market failures low-income countries. International finance and require appropriate policy counter measures to institutional support may be required to facilitate neutralize them (World Bank 2012; Lütkenhorst the success of green transformation processes in et al. 2014). Second, implementing the appro- low-income countries. priate measures is particularly challenging in developing countries. Developing countries face Each country should assess the opportunities of several economic and institutional limitations applying green policies in its particular context, that can impede success. Some green industrial balancing the expected benefits against the policies that are specifically designed to support potential risks. Moreover, each country should particular industries or technologies require good choose its own path of green transformation. This Green Industrial Policy - Concept, Policies, Country Experiences

involves customising choices about sequencing that are readily observable. Each developing 34 and prioritizing measures that yield the highest country should balance the costs of early action short-term benefits, such as energy efficiency against the costs of lock-in and should follow a improvements or a proper management of natu- greening path appropriate to its specific needs. ral resources. These customised choices could also lead to fewer or no regrets, such as efficient Finally, the international context can play an infrastructure construction that forestalls later important role in the definition of environmen- replacement costs. Justifications for and potential tal policies and the development paths of devel- benefits of the policies should be clearly identified oping countries. A context of more stringent for all stakeholders. Putting a price on pollution environmental policies in developed countries helps to reduce environmental problems and can produce a variety of results for developing pressures, but it is not enough. Each instrument countries. For developing countries to succeed in and mechanism should be combined with differ- following a clean development path, it is impor- ent measures according to the particular context. tant that developed countries support them by The potential costs of promoting innovative but providing the appropriate financial mechanisms. unproven technologies may be great, so green Some mitigation measures will require interna- industrial policies are less risky when they focus tional finance to facilitate implementation. Help on technologies that have been tested elsewhere from the international community to facilitate and adapted to the local situation or when they provision of proper financial instruments, insti- complement untapped comparative advantages tutional support, or technology transfers may be crucial to convert potential gains into actual ones. Chapter 2 What Can Developing Countries Gain from a Green Transformation?

REFERENCES 35 Aguilera-Klink, F. (1994). Some notes on the misuse Costello, C., Gaines, S. D., & Lynham, J. (2008). Can of classic writings in economics on the subject Catch Shares Prevent Fisheries Collapse? Science, of common property: Journal of Southern African 321(5896), 1678–1681. Studies. Ecological Economics, 9 (3), 221–228. Cottier, T., Nartova, O., & Bigdeli, S. Z. (2009). Inter- Alcántara, V., & Padilla, E. (2009). Input–output national Trade Regulation and the Mitigation of subsystems and pollution: An application to Climate Change: World Trade Forum. Cambridge:

the service sector and CO2 emissions in Spain. Cambridge University Press. Ecological Economics, 68(3), 905–914. Dasgupta, S., Laplante, B., Wang, H., & Wheeler, D. 2 PART Arrow, K., Bolin, B., Costanza, R., Dasgupta, P., Folke, (2002). Confronting the Environmental Kuznets C., Holling, C. S., Jansson, B. O., Levin, S., & Mäler, Curve. Journal of Economic Perspectives, 16(1), K. (1995). Economic growth, carrying capacity, 147–168. and the environment: Ecological Applications. Dobbs, R., Oppenheim, J., Thompson, F., Brinkman, Science, 268(5210), 520–521. M., & Zornes, M. (2011). Resource revolution: Meet- Arze del Granado, F. J., Coady, D., & Gillingham, R. ing the world’s energy, materials, food, and water (2012). The Unequal Benefits of Fuel Subsidies: needs. London: McKinsey & Company. A Review of Evidence for Developing Countries. Dutz, M. A., & Sharma, S. (2012). Green Growth, Tech- World development, 40(11), 2234–2248. nology and Innovation (Policy research working Awerbuch, S., & Sauter, R. (2006). Exploiting the oil– papers No. 5932). Washington, D.C. GDP effect to support renewables deployment: Ekins, P. (1997). The Kuznets curve for the envi- Ecology Law Quarterly. Energy Policy, 34(17), ronment and economic growth. Examining 2805–2819. the evidence. Environment and Planning, 29, Baranzini, A., van den Bergh, J. C. J. M., Carattini, S., 805–830. Howarth, R. B., Padilla, E., & Roca, J. (2017). Carbon Ekins, P., & Speck, S. (1999). Environmental Taxa- pricing in climate policy: Seven reasons, comple- tion in Practice. Environmental and Resource mentary instruments, and political economy Economics, 13(4), 369–396. considerations. Wiley Interdisciplinary Reviews: Fankhauser, S., Sehlleier, F., & Stern, N. (2008). Climate Change, 8(4), 1–17. Climate change, innovation and jobs. Climate Baron, R. & Fischer, D. (2015). Divestment and Policy, 8(4), 421–429. Stranded Assets in the Low-carbon Transition. Food and Agriculture Organization of the United Paris. Nations (FAO). (2005). Increasing the contribution Bosseboeuf, D., Lapillonne, B., Eichhammer, W., & of small-scale fisheries to poverty alleviation Boonekamp, P. (2007). Evaluation of Energy Effi- and food security. FAO technical guidelines for ciency in the EU-15: Indicators and Policies (No. responsible fisheries: Vol. 10. Rome: FAO. 1/2). Paris. Gillingham, K., Kotchen, M. J., Rapson, D. S., & Wagner, Bowen, A., & Fankhauser, S. (2011). Low-Carbon Devel- G. (2013). Energy policy: The rebound effect is opment for the Least Developed Countries. World overplayed. Nature, 493 (7433), 475–476. Economics, 2, 145–162. Global Environment Facility (GEF). (2016). What is Brander, E. (2013). The Montreal protocol: a model for the GEF. Retrieved from Global Environment future multilateral environmental agreements. Facility (GEF) website: www.thegef.org/gef/ Public Policy and Governance Review. whatisgef Broad, R. (1994). The poor and the environment: Green Climate Fund (GCF). (2016). Green Climate Friends or foes? World development. (6), 811–822. Fund. Global context. Retrieved from www.green- Caldecott, B., Harnett, E., Cojoianu, T., Kok, I., & climate.fund/about-gcf/global-context Pfeiffer, A. (2016). Stranded Assets. A Climate Risk Grossman, G., & Krueger, A. (1991). Environmental Challenge. Impacts of a North American Free Trade Agree- Chu, C. (2009). Thirty years later: The global growth ment. Working Paper: Vol. 3914. Cambridge, MA: of ITQs and their influence on stock status National Bureau of Economic Research. in marine fisheries. Fish and Fisheries, 10(2), Heal, G., & Schlenker, W. (2008). Economics: Sustaina- 217–230. ble fisheries.Nature, 455(7216), 1044–1045. Clements, B., Coady, D., Fabrizio, S., Gupta, S., & Shang, Houser, T. (2008). Leveling the carbon playing field: B. (2014). Energy subsidies: How large are they International competition and US climate policy and how can they be reformed? Economics of design. Washington, D.C.: Peterson Institute Energy & Environmental Policy, 3(1). for International Economics; World Resources Coady, D., Parry, I., Sears, L., & Shang, B. (2017). How Institute. Large Are Global Fossil Fuel Subsidies? World development, 91, 11–27. Green Industrial Policy - Concept, Policies, Country Experiences

Intergovernmental Panel on Climate Change (IPCC). OECD. (2012b). Green Growth and Developing Coun- 36 (2014). Climate Change 2014: Impacts, Adaptation, tries – Consultation Draft. Paris: OECD Publishing. and Vulnerability, Working Group II Contribution OECD. (2001). Human health and the environment. In to the Fifth Assessment Report of the Intergov- OECD (Ed.), OECD Environmental Outlook. Paris: ernmental Panel on Climate Change. Cambridge: OECD. Cambridge University Press. Ostrom, E. (1990). Governing the commons: The International Energy Agency (IEA). (2008). Energy evolution of institutions for collective action. technology perspectives: Scenarios & strategies Cambridge, England, New York: Cambridge to 2050: in support of the G8 plan of action. Paris: University Press. OECD. Pagiola, S., Arcenas, A., & Platais, G. (2005). Can International Finance Corporation (IFC). Selling Solar. Payments for Environmental Services Help Lessons from More than a Decade of IFC’s Experi- Reduce Poverty? An Exploration of the Issues and ence. Washington, D.C. the Evidence to Date from Latin America. World Jakob, M., Chen, C., Fuss, S., Marxen, A., & Edenhofer, development, 33(2), 237–253. O. (2015). Development incentives for fossil fuel Panayotou, T. (1993). Empirical Tests and Policy Anal- subsidy reform. Nature Climate Change, 5(8), ysis of Environmental Degradation at Different 709–712. Stages of Economic Development (Working Paper Jakob, M., Steckel, J. C., & Edenhofer, O. (2014). No. WP238). Geneva. Consumption- Versus Production-Based Emis- Pearce, D. (1991). The Role of Carbon Taxes in Adjust- sion Policies: 297–318. Annual Review of ing to Global Warming: 938–948. The Economic Resource Economics, 6(1), 297–318. Journal, 101(407), 938. Jehan, S., & Umana, A. (2003). The environment-pov- Pegels, A. (2014). Why we need a green industrial erty Nexus. Development Policy Journal, 3, 53–70. policy. In A. Pegels (Ed.), Routledge studies in Kroetz, K., Sanchirico, J. N., Peña-Torres, J., & Corderi ecological economics: Vol. 34. Green Industrial Novoa, D. (2016). Evaluation of the Chilean Jack Policy in Emerging Countries (pp. 1–9). London: Mackerel ITQ system. IDB working paper series: Routledge. no. IDB-WP-672. Washington, D.C.: Inter-Ameri- Piaggio, M., Alcántara, V., & Padilla, E. (2015). The can Development Bank (IDB). materiality of the immaterial: 1–10. Ecological Lütkenhorst, W., Altenburg, T., Pegels, A., & Vidican, Economics, 110, 1–10. G. (2014). Green industrial policy: Managing trans- Piaggio, M., & Padilla, E. (2012). CO2 emissions and formation under uncertainty. Discussion paper economic activity: Heterogeneity across coun- / Deutsches Institut für Entwicklungspolitik: tries and non-stationary series. Energy Policy, 46, 28/2014. Bonn: Dt. Inst. für Entwicklungspolitik. 370–381. Martínez-Alier, J. (1995). The environment as a REN21. (2016). Renewables 2016 Global Status luxury good or “too poor to be green”? Ecological Report. Paris. Retrieved from: www.ren21.net/ Economics, 13(1), 1–10. GSR-2016-Report-Full-report-E Mattoo, A., He, J., Subramanian, A., & van der Rickwood, P., Glazebrook, G., & Searle, G. (2008). Urban Mensbrugghe, D. (2009). Reconciling Climate Structure and Energy—A Review. Urban Policy Change And Trade Policy. Washington, D.C.: The and Research, 26(1), 57–81. World Bank. Roca, J., & Padilla, E. (2003). Emisiones atmosféricas McKinsey & Company. (2009). Pathways to a y crecimiento económico en España: La curva Low-Carbon Economy. Version 2 of the Global de Kuznets ambiental y el protocolo de Kyoto. Greenhouse Gas Abatement Cost Curve. Economía Industrial, 351, 73–86. McKinsey & Company Roca, J., Padilla, E., Farré, M., & Galletto, V. (2001). Moran, T. H. (2015). The Role of Industrial Policy as Economic growth and atmospheric pollution in a Development Tool: New Evidence from the Spain: Discussing the environmental Kuznets Globalization of Trade-and-Investment (CGD curve hypothesis. Ecological Economics, 39(1), Policy Paper No. 071). Washington, D.C. 85–99. Nordhaus, W. (2015). Climate Clubs: Overcoming Rocchi, P., Arto, I., Roca, J., & Serrano, M. (2015). Free-riding in International Climate Policy. Carbon-Motivated Border Tax Adjustment: A American Economic Review, 105(4), 1339–1370. Proposal for the EU. SSRN Electronic Journal. OECD. (2012a). Compact City Policies. A Comparative Advance online publication. Assessment (OECD Green Growth Studies). Paris: Rodrik, D. (2014). Green industrial policy. Oxford OECD Publishing. Review of Economic Policy, 30(3), 469–491. Chapter 2 What Can Developing Countries Gain from a Green Transformation?

Schipper, L., & Grubb, M. (2000). On the rebound? United Nations Environment Programme. (2014). Feedback between energy intensities and energy Montreal Protocol Celebrates Landmark Achieve- 37 uses in IEA countries. Energy Policy, 28(6–7), ment with Universal Ratification of All Amend- 367–388. ments: Press Release, 09/12/2014. United Nations. Selden, T. M., & Song, D. (1995). Neoclassical Growth, (1995). The United Nations Fourth World Confer- the J Curve for Abatement, and the Inverted U ence on Women. Beijing, China - September 1995. Curve for Pollution. Journal of Environmental Action for Equality, Development and Peace. Economics and Management, 29(2), 162–168. United Nations Environment Programme. (2014). Shafik, N., & Bandyopadhyay, S. (1992). Economic Montreal Protocol Celebrates Landmark Achieve-

growth and environmental quality: Time-series ment with Universal Ratification of All Amend- 2 PART and cross-country evidence. Policy research ments: Press Release, 09/12/2014. Retrieved from working papers: WPS 904. Washington, D.C.: Office UNEP News Centre of the Vice President, Development Economics, United Nations Industrial Development Organization World Bank. (UNIDO) and The Global Green Growth Institute. Stern, D. I., Common, M. S., & Barbier, E. B. (1996). (2015). Global Green Growth. Clean Energy Indus- Economic growth and environmental degra- try Investments and Expanding Job Opportuni- dation: The environmental Kuznets curve and ties. Volume I: Overall Findings. Vienna, Seoul. sustainable development. World development, World Bank. (2003). Indoor Air Pollution: November 24(7), 1151–1160. 2003. Washington, D.C.: World Bank. Sumaila, U. R., Lam, V., Le Manach, F., Swartz, W., & World Bank. (2010). World Development Report 10. Pauly, D. (2016). Global fisheries subsidies: An Development and Climate Change. Washington, updated estimate. Marine Policy, 69, 189–193. D.C.: World Bank. Suri, V., & Chapman, D. (1998). Economic growth, trade World Bank. (2012). Inclusive green growth: The path- and energy: Implications for the environmen- way to sustainable development. Washington, tal Kuznets curve. Ecological Economics, 25(2), D.C.: World Bank. 195–208. World Health Organizaton (WHO). (2016a). Household United Nations. (1995). The United Nations Fourth air pollution and Health (Fact Sheet No. 292). World Conference on Women. Beijing, China - World Health Organizaton (WHO). (2016b). Ambient September 1995. Action for Equality, Development (outdoor) air quality and health (Fact Sheet No. and Peace. 313). United Nations Environment Programme. (2015). The Montreal Protocol on Substances that Deplete the Ozone Layer. Retrieved from: ozone.unep.org/en/ treaties-and-decisions/montreal-protocol-substances-deplete-ozone-layer Green Industrial Policy - Concept, Policies, Country Experiences 38 CHAPTER 3 GAINING COMPETITIVE ADVANTAGE WITH GREEN INDUSTRIAL POLICY

Stefan Ambec Chapter 3 Gaining Competitive Advantage with Green Industrial Policy

1. INTRODUCTION 39 Sustainability should be considered an inte- cost of environmental protection can lead to a gral part of the economic development process. competitive advantage for firms. This approach Economic growth is often associated with pollu- maintains that green policy can boost a firm’s tion and ecosystem degradation, conditions that competitiveness while improving environmen- threaten human health and economic activities tal quality. Four paths are distinguished by which that rely on natural resources. Despite this feed- green policy might enhance competitiveness. back between poor environmental quality and the First, green standards and certifications enable PART 2 PART economy, policymakers are generally reluctant firms to differentiate their products from those to impose more stringent environmental regula- of their competitors. Section 2 explains why this tions on firms. They fear that it might erode firms’ product differentiation strategy can be profitable competitiveness, slowing down the country’s and what the role of public policy can be. Second, economic development process. firms can make money by selling and adopting green technology. As explained in section 3, this A dominant opinion concerning environmental argument particularly applies to environmentally protection is that it adds cost to firms. Accord- sound technologies that are bought by developed ing to this opinion, reducing the pollution from countries due to energy mandates or financed by production processes requires that firms replace carbon offsetting projects in developing countries. dirty inputs with cleaner but costlier ones. This Third, investment in greener technology can lead may involve a switch to more expensive tech- to productivity improvements that more than nologies or investment in devices that reduce offset additional costs of environmental protec- or remedy polluting emissions, treat their waste, tion. This argument relies on the well-known recycle water, or improve energy efficiency. Porter Hypothesis reviewed in section 4. Section 5 All these strategies add new expenses to their examines how green policy can increase compet- balance sheets. We argue that such expenses itiveness due to knowledge spillovers in the inno- can nevertheless turn out to be profitable in the vation process. The conclusions present policy long run. recommendations for designing green growth policies that are likely to deliver competitive The aim of this chapter is to review economic advantages in international markets. arguments and to show that the additional

2. HIGHER MARKET SHARES BY DIFFERENTIATING PRODUCTS BASED ON ENVIRONMENTAL QUALITY

2.1. SUPPLYING GREENER PRODUCTS AS A Supplying greener products often entails addi- VERTICAL DIFFERENTIATION STRATEGY tional costs for the producer. By going beyond mandatory standards and thus increasing their Firms can reduce their negative impact on the costs, companies put themselves at a disadvan- environment by improving the environmental tage compared to competitors. Yet they might be quality of their products throughout the products’ able to recover this extra production cost through life cycles. Firms can use less polluting inputs such revenue if consumers agree to pay more for envi- as fewer pesticides or other harmful chemicals; ronmental quality. Supplying greener quality increase their share of renewable sources of energy, products can be viewed as a vertical differenti- such as wind and solar power; or adopt sustainable ation strategy. By differentiating their products production procedures such as replanting forests from those of competitors, firms move away from after harvesting or using fishing practices that head-on comparison with many similar, or seem- have less impact on ecosystems. They can also ingly identical, products. That differentiation facilitate the recycling of their product and choose may only be the environmental intensity of their fewer or less polluting packaging materials (Haned production methods. That approach takes advan- et al. 2015). Firms are sometimes forced to do so tage of the market power of a niche of consumers when governments set more stringent regulatory who are willing to pay more for greener prod- standards, but in many cases firms go beyond ucts. Ultimately, the price premium on greener minimal or mandatory standards on environmen- products can offset their higher production costs. tal quality. Such a successful vertical differentiation strategy Green Industrial Policy - Concept, Policies, Country Experiences

requires that consumers are able to identify prod- The final consumers are not the only ones to value 40 ucts with higher environmental quality, ideally the environmental attributes of products. Private through eco-labels. companies often take steps to green their supply chain. For instance, production plants involved 2.2. LABELLING AND CERTIFICATION in the ISO 14001 environmental management systems certification procedure commit to using Firms can signal the higher environmental qual- environmental performance criteria for select- ity of their products through certification and ing their suppliers.9 Governments increasingly labelling. Examples include the Forest Steward- support products and suppliers with better envi- ship Council certificate for wood produced from ronmental performance through green public sustainably managed forests and the Marine procurement standards and policies. The U.S. Stewardship Council certificate for seafood Federal Acquisition Regulations provide detailed harvested using sustainable fishing practices. rules governing procurement by all federal agen- Organic food labels issued by the U.S. Department cies. For instance, these rules specify that the U.S. of Agriculture or by the European Commission are Environmental Protection Agency has to prepare presently one of the most popular ways to differ- guidelines on the availability, sources, and entiate products in the food industry. This does potential uses of recovered materials and asso- matter, since the similitude of many agricultural ciated products, including solid waste manage- products means that label differentiation is often ment services. These guidelines require federal the only way to secure a competitive advantage. agencies themselves to develop and implement Organic food labels are well known by consumers affirmative procurement programmes for Envi- and often supported by public authorities. This ronmental Protection Agency designated prod- system can incur higher costs through manda- ucts (Kunzik 2003). tory certification procedures and lower yields for producers at least in the short run, as it restricts 2.3. CREDIBLE LABELLING AND the use of many inputs, such as pesticides, antibi- CERTIFICATION: A ROLE FOR otics, or some fertilisers.7 GREEN POLICY

Fair trade labels also have a green component For the vertical differentiation strategy to be insofar as they favour less intensive agricultural successful, the eco-label should convey credi- practices and organic standards. Environmen- ble and transparent information to consumers. tal criteria in fair trade production include the Credibility requires that consumers trust what a banning of certain harmful chemicals, reduction label means in terms of environmental protec- of the use of pesticides, and promotion of natural tion. The criteria for labelling and its implemen- biological methods to preserve soil and biodiver- tation should be advertised in an understandable sity such as crop rotation as well as a preference way for non-experts. Procedures for certification for less intensive small scale farming. Evidence should be immune to corruption and manipula- shows that consumers are paying higher prices tion by companies that would like to greenwash for labelled fair trade and organic products. For their products by obtaining a green logo without instance, in a survey of coffee producers in paying the cost of environmental protection. The Central America and Mexico, Méndez et al. (2010) label itself should be legible and easily identifia- found a significant positive correlation between ble for consumers. Public policy can help with average sales prices for coffee for both fair trade that matter in several respects. Public adminis- and organic labels. A study of 228 Nicaraguan trations can facilitate the definition and dissem- coffee producers evaluated an average price of ination of labelling criteria, support certification 8 US$ 0.84 per pound for fair trade coffee, US$ 0.63 by agencies and NGOs, ensure the traceability of for organic coffee and US$ 0.41 for conven- products along the supply chain, and simplify and tional coffee during the 2000–2001 coffee price harmonise the framing of labels to consumers. At crisis (Bacon 2005; Dragusanu et al. 2014). Hence, the country level, public authorities should make organic and fair trade certification can help labelling for international markets both feasible producers offset higher production costs. at a reasonable cost, and credible. Relying on the most respected and internationally known labels

7 For instance, Ramesh and others (2010) found that in India organic farming reduced yields by 9.2 per cent on average in the sample of certified farms they surveyed. The figures dropped by 20 per cent for rice and wheat and was up to 25 per cent lower than conventional farming for cotton. 8 Net of costs paid to the cooperative for transport, processing, certification, debt service, and export. 9 In a sample of 4,000 facilities in seven OECD countries, Johnstone and Lebonne (2007) found that 43 per cent assess their suppliers’ environmental performance. Chapter 3 Gaining Competitive Advantage with Green Industrial Policy

has many advantages. By facilitating the adop- That can also help attract foreign investment in tion of internationally recognised labels for green firms that produce eco-labelled products. Further- 41 and sustainable products, public administrations more, supporting internationally recognised labels can facilitate exports of eco-certified products to is an important criterion in the evaluation of the international markets where the labels are valued, WTO consistency of government-supported stand- especially higher income regions such as the EU. ards under WTO law (Cosbey 2017, this volume).

3. SELLING AND ADOPTING GREEN TECHNOLOGIES 2 PART

3.1. ENVIRONMENTAL POLICY TRIGGERS THE the wholesale market price. Many States in the DEMAND FOR GREEN TECHNOLOGIES US have opted for renewable portfolio standard programmes, which generally require a mini- Environmental regulations can also boost mum proportion of electricity demand to be met economic growth and economic activity for some by renewable sources. Those programmes are sectors by spurring demand: solving environ- usually implemented on the basis of renewable mental problems has become a business oppor- energy certificates issued for verified renewable tunity for companies supplying pollution-control generators. Those entities exceeding their renew- technologies. For instance, the implementation able energy obligations can earn money by selling of more stringent air quality standards in Europe certificates to other entities that fail to meet their and the US increased the demand for scrubbers obligations. Both types of support to renewables to filter SO and NOx emissions from coal power 2 have been very successful. They have boosted the plants. It also fostered innovation in scrubber and wind and solar power industries not only where coal combustion as more patents were issued they were implemented but also abroad. Massive following the application of the regulation (Popp investment in solar power in countries like 2006). Government intervention for environmen- Germany has not only given jobs to new compa- tal protection can therefore boost the growth of nies involved in installing photovoltaic panels, but eco-industries, supplying technologies that miti- has also contributed to making China the larg- gate or clean-up pollution. est solar photovoltaic cell producer in the world (Pegels 2017, this volume). Besides command-and-control regulation, such as more stringent standards, governments may also decide to apply market-based, or economic 3.2. PHOTOVOLTAICS IN CHINA instruments. Instead of prescribing a certain environmental standard to be achieved or technology From 2003 to 2009, the exponential growth of to be used, market-based instruments incentivise photovoltaic installations was concentrated in the adoption of clean technology or demand for a few developed countries, primarily Germany, pollution-control technologies. Subsidies are one Spain, Japan and the US. Yet, during this period, popular economic policy measure often issued the production of photovoltaic cells moved to by governments as a way to offset additional cost emerging economies, notably China. Starting with of higher environmental standards for compa- a 1.6 per cent market share for cell and module nies. They may therefore be used by governments assembly production in 2003, China became the to incentivise firms to adopt environmentally world leader with a 35 per cent market share in sound technologies that are cleaner but more 2007, just five years later, followed by the Euro- costly. This can provide business opportunities for pean Union, with 29 per cent market share (De sectors and technologies that may not be compet- La Tour et al. 2011). This shift of leadership from itive in the current economic environment or Europe to China has triggered much criticism their current state of technological development. in Europe’s policy debate as feed-in tariffs were supposed to give first mover advantages to Euro- Perhaps the most famous example of subsidies pean companies in the wind and solar power employed to advance environmental objectives, industries. They did indeed promote activities among others, is the feed-in tariff for renewa- that were not exposed to international competi- ble energy sources implemented in more than tion, such as photovoltaic installation. However, 50 countries including many developing coun- China took advantage of low-cost labour and tries. By issuing a feed-in tariff, public authorities relatively cheap energy to set up cell production commit to purchase electricity produced from and module assembly lines without much prior wind and solar power at a fixed price, well above experience in manufacturing cells. According to Green Industrial Policy - Concept, Policies, Country Experiences

De La Tour et al. (2011), China managed to obtain been concentrated in only a few emerging econ- 42 the technology through three channels. Impor- omy countries—mainly China, India and Brazil— tantly, China gained access to technology through for only a few specific technologies. Overall, the purchasing the manufacturing equipment in a CDM has nevertheless been very successful in competitive market. Furthermore, they obtained bringing foreign direct investment into develop- the required skills and knowledge by way of hiring ing countries. It has turned into a lively market highly skilled Chinese executives trained abroad with 180 transactions involving US$ 2.5 billion in in universities or in the photovoltaic industry. total in 2005 alone and up to US$ 6.5 billion more China attracted foreign direct investment under in 2008 (Lecoq and Ambrosi 2007; Kossoy and a policy that obliged foreign investors to accept Ambrosi 2010). joint ownership. Such joint ventures are likely to induce more knowledge spillovers between the 3.4. GAINING COMPETITIVE ADVANTAGE BY foreign investor and the local firm. From a coun- ISSUING CARBON CREDITS try on the receiving end of technological transfers, China has evolved into a major producer in the The projects financed by the CDM are diverse. photovoltaic solar industry. Carbon credits can come from increasing the energy efficiency of buildings, or recovering biogas 3.3. CLEAN DEVELOPMENT MECHANISM AND from agriculture and landfills, for example. World- CARBON OFFSET wide, a large share of CDM projects concerns electricity production through investments in Companies in developing countries have bene- renewable sources of energy such as hydro, wind, fited from another policy involving transfers from solar or biomass. developed countries: the Clean Development Mechanism (CDM) part of the Kyoto Protocol. The economic gains that host countries derive This policy allows greenhouse gas emitters that from CDM projects extend beyond the company are subject to emission reduction obligations to that implements a project for carbon credits. offset their own emissions by investing in emis- These gains sometimes spread to the supply sion-reducing projects in less developed econ- chain of green technologies, fostering production omies. The aim is to take advantage of cheaper in developing countries. For instance, most wind mitigation opportunities in developing countries, power projects implemented in India use equip- thus making climate change mitigation more ment produced by local manufacturers, mainly cost-effective. The CDM can be used by compa- Suzlon and Enercon India (Dechezleprêtre et al. nies, such as electricity producers or cement and 2009). Furthermore, many firms have used CDM glass manufacturers, involved in the European projects as collateral to obtain upfront financing Union Emission Trading Scheme to obtain new from financial partners. For instance, Lecocq and allowances. It has been a substantial stimulant: Ambrosi (2007) report the case of pig-iron produc- during the 2002 to 2008 second phase, around 1 ers in Brazil who obtained loans from a Dutch bank by transferring the carbon credits gain from billion tonnes of CO2 credits were bought in the Trading Scheme (European Commission 2017). replacing coal with lower GHG-emitting charcoal. More generally, many firms meet corporate social Future credits are valuable collateral because they responsibility obligations by offsetting their own are paid in strong currencies—US dollars, euros or carbon emissions through carbon credits granted yen—by investors with high credit ratings. by the CDM. Local companies have benefited from another The CDM creates carbon credits based on key component of many CDM projects: technol- projects certified by the UN Framework Conven- ogy transfers. In the sample of 3296 CDM projects tion on Climate Change provided that an emis- examined by Schmid (2012), 36 per cent of them sion reduction is ‘additional’ to any that would claimed to have a technological transfer dimension, have occurred in the most plausible alterna- accounting for 59 per cent of the total emission tive scenario to the implementation of the CDM reduction. Haites et al. (2006) found a rate of tech- project, or the business-as-usual scenario. The nical transfers of the same magnitude in a sample additionality criterion has been criticised for of 860 CDM projects: one-third involved technol- creating perverse incentives for developing coun- ogy transfers and accounted for two-thirds of the tries, as it rewards the absence of climate miti- annual emission reductions. In a closer look at CDM gation policies. Projects must also be accredited projects, Dechezleprêtre et al. (2008) distinguish by the host government, based on sustainable between two types of technology transfer: knowl- development criteria (Olsen and Fenhann 2008). edge transfer and equipment transfer. Knowledge One drawback of the CDM is that projects have transfer takes place when a local project benefits through the transfer of knowledge, information, Chapter 3 Gaining Competitive Advantage with Green Industrial Policy

technique or technical assistance from a foreign 3.5. GREEN GROWTH BY ATTRACTING partner. Equipment transfer describes the import INVESTMENT IN CARBON OFFSETTING 43 of devices such as wind turbines. In their sample of Given the amount of money available through 644 CDM projects, the researchers found that 43 per carbon credits, it might come as a surprise that cent of the projects involved technology transfers: most of the projects ended up in three coun- 9 per cent consisted of only equipment transfer, 15 tries: China, India and Brazil, which together per cent included only knowledge transfer, and the accounted for nearly 80 per cent of the credits remaining 19 per cent involved both equipment and in 2006 (Lecoq and Ambrosi 2007). Most Afri- knowledge transfer. can countries were left out. Partially, this can be 2 PART Dechezleprêtre’s team investigated the determi- explained by the administrative costs of accred- nants of technology transfers among CDM projects. itation. Under the Kyoto Protocol, countries Unsurprisingly, the size of the project and the have to set up designated national authorities to openness to trade for the host country positively participate in the CDM. The designated national influence the probability of technology transfers. authorities have to support projects by signing a According to Schmid (2012), higher tariffs on envi- letter of approval to issue carbon credits (Olsen ronmental goods and services impede the like- and Fenhann 2008). While this process is meant lihood of technological transfer in CDM projects. to respect countries' sovereignty, it also means Reducing tariffs on those environmental products that no international standard for accreditation might make technological transfer easier. One key exists: countries have established their own ingredient that might matter at the country level distinct procedures and criteria. Therefore, it is seems to be its technological absorptive capabil- easier for companies to focus on a few technolo- ity. Technological absorptive capability refers to a gies in a small number of countries. country’s ability to conduct research and to under- Governments can take action to attract CDM stand, implement, and adapt imported technolo- projects—and more generally attract investment gies (Popp 2011). It has to do with the workforce’s for carbon offsetting—by making accreditation technological literacy and skills, which are influ- easier. Procedures could be standardised to help enced by many factors that are controlled by public alleviate the administrative cost of setting up a authorities, such as education and infrastructure. CDM project (Schmid 2012). Governments can Overall, empirical studies have found no effect also design accreditation procedures to maxim- or an ambiguous role of technological absorptive ise the economic benefits from carbon offsetting capability on technological transfer (Dechezleprê- investment. For instance, they can include tech- tre et al. 2008; Schmid 2012; Murphy et al. 2013; nological transfers as a criterion for accreditation. Gandenberger et al. 2015). According to Deche- As Popp (2011) showed, this has been the case in zleprêtre’s 2008 study, technological absorptive South Korea, which requires that ‘‘environmentally capability10 has a positive influence on technology sound technologies and know-how shall be trans- transfers in the energy sector and the chemical ferred’’ through CDM projects in Korea. As a result, industry, but a strong negative effect in agriculture. 88 per cent of the emissions reductions from CDM According to the authors, the contrast seems to projects in South Korea come from projects that reflect an antagonistic effect that such capability involve technology transfer. Similarly, Chinese has on international technology transfers. On the guidelines for CDM project approval stipulate that one hand, it facilitates transfer when firms in the ‘‘CDM project activities should promote the trans- host country have skills to adopt a new technology. fer of environmentally sound technology to China’’ On the other hand, it increases the likelihood that (Haites et al. 2006). As discussed above, depending technology is already available locally, and there- on the sector, technology transfers can be made fore reduces the need for international transfers. more likely through a targeted increase in a coun- The former effect holds in the energy sector and try's absorptive capacity (Dechezleprêtre et al. the chemical industry where more newly devel- 2008; Popp 2011), as well as openness to trade, and oped technologies and advanced techniques are thus support a country’s economic development involved, while the latter effect seems predomi- and growth. nant in agriculture where local conditions influence the technological needs.

10 While this paper uses the term 'technology absorptive capacity', Dechezleprêtre’s study employs the term 'technological capabilities', which is constructed following an index of technological capabilities developed by Archibugi and Coco (2004). The index accounts for the following factors: the creation of technology (number of patents and number of scientific articles), the technological infrastructures (Internet penetration, telephone penetration and electricity consumption) and the development of human skills (percentage of tertiary science and engineering enrolment, mean years of schooling and literacy rate). Green Industrial Policy - Concept, Policies, Country Experiences

44 4. INCREASE A FIRM’S PRODUCTIVITY: THE PORTER HYPOTHESIS

4.1. THE PORTER HYPOTHESIS 4.2. EMPIRICAL EVIDENCE OF THE PORTER HYPOTHESIS More than 20 years ago, Michael Porter suggested that pollution was generally associated with a On the empirical side, Jaffe and Palmer (1997) pres- waste of resources, or with lost energy potential: ent three distinct variants of the Porter Hypoth- “Pollution is a manifestation of economic waste esis. In their framework, the ‘weak’ version of the and involves unnecessary or incomplete utili- hypothesis is that environmental regulation will sation of resources… Reducing pollution is often stimulate certain kinds of environmental innova- coincident with improving productivity with tions, although this does not mean that the inno- which resources are used” (Porter and van der vation is socially beneficial. The ‘narrow’ version Linde 1995). Based on this reasoning, Porter argues of the hypothesis asserts that flexible, market- “properly designed environmental regulations can based environmental policy instruments, such as trigger innovation that may partially or more than pollution charges or tradable permits, give firms a fully offset the costs of complying with them.” This greater incentive to innovate than do prescriptive has come to be known as the Porter Hypothesis. regulations such as technology-based standards. In other words, it is possible to reduce pollution Finally, the ‘strong’ version posits that properly emissions and production costs at the same time, designed regulation may induce innovation that resulting in ‘win-win’ situations. more than compensates for the cost of compliance and improves the financial situation of the The Porter Hypothesis is controversial: First, the firm. Many researchers have tested the different evidence initially provided to support it is based versions of the Porter Hypothesis empirically. Over- on a small number of company case studies, in all, the empirical literature provides evidence for which firms were able to reduce both their pollu- the weak version but not for the strong one. Most tion emissions and their production costs. As studies find a positive although sometimes muted such, it can hardly be transferred to the entirety relationship between, on the one hand, more strin- of firms. Second, economic theory would suggest gent environmental policies and innovation meas- that, if there are opportunities to reduce costs and ured by investment in research and development inefficiencies, companies should identify them and, on the other, new technologies or successful by themselves without the need for government patent applications. However, some studies find intervention (Palmer et al. 1995). However, many that the impact of environmental regulations on studies have proposed analytical justifications for productivity or business performance turns out to the Porter Hypothesis. It could be that the interests be negative in general (Lanoie et al. 2011; Ambec et of companies and their managers are not aligned, al. 2013). perhaps due to risk aversion, time-inconsistency, asymmetric information, or other circumstances. Other empirical investigations suggest that the Regulations force firms to adopt innovations that impact of regulation on productivity may depend on are profitable for the firm but not for its -manag its stringency: For instance, Berman and Bui (2001) ers. As Ambec and Barla (2006) argue, the Porter found that refineries located close to Los Angeles Hypothesis can be valid if a market failure exists in are significantly more productive than other US addition to the environmental externality. Exam- refineries, despite the more stringent air quality ples include knowledge spillovers or market power. regulation in the Los Angeles area. Similarly, Alpay For instance, Simpson and Bradford (1996) inves- et al. (2002) report that the productivity of the Mexi- tigate the impact on environmental regulation in can food-processing industry is increasing with a model with firms competing on international the pressure of environmental regulation, which markets. They show that more stringent environ- leads them to conclude that more stringent regu- mental regulation commits a domestic firm to an lations can positively affect productivity. It seems aggressive cost-reducing programme, thereby that, albeit the scientific evidence of a win-win situ- creating a first mover advantage. ation, as Porter suggests, is mixed, more stringent environmental regulation can be good for business by fostering innovation, which provides firms with competitive advantage. The open question for policymakers is how to design policies to obtain a causal chain as proposed in the Porter Hypothesis. Chapter 3 Gaining Competitive Advantage with Green Industrial Policy

4.3. PUBLIC POLICIES FOR PROFITABLE subsidised. To reduce potential negative effects GREEN INNOVATIONS on profits within an industry, pollution charges 45 can be earmarked within the taxed industry. The Porter favours stringent but flexible economic money collected can be redistributed directly to instruments such as pollution charges or trad- firms, depending on size, measured by output or able emission permits. These economic instru- work force. It could also be used to subsidise envi- ments are more likely to enhance innovation than ronmental research and development or the adop- command-and-control instruments, such as tech- tion of cleaner technologies. nological standards, as they give more freedom to firms with regards to the technology used to abate Other environmental policy mechanisms, like 2 PART pollution. In contrast, by imposing a given tech- cap-and-trade schemes may be difficult to estab- nology or certain inputs, a technological standard lish politically. One way to improve political provides fewer incentives to innovate. Simi- feasibility of such a scheme can be to give away larly, emission or performance standards do not a number of permits for free to an industry, at encourage firms to go beyond the standards set. least during the initial distribution phase. Under On the other hand, environmental policy instru- this approach, firms will still have incentives to ments such as pollution charges can provide an invest in pollution abatement technologies as incentive for firms to emit less as they can save long as they can make money by selling addi- money by reducing emissions further. Similarly, tional permits. Later on, fewer permits should in the presence of an emission permits trading be granted for free and more sold by auction, to system, firms can save money by purchasing ensure entry of new players into the industry. fewer permits on the market or even sell their own emission endowments by cutting emissions Other policy instruments can be effective in beyond their own emission rights. greening firms. Voluntary approaches can be a substitute for mandatory environmental The use of economic instruments appears to be a constraints in countries with weak institutions necessary condition for the Porter Hypothesis to (Never and Kemp 2017, this volume). Examples hold. This is known as the narrow Porter Hypoth- include information disclosure programmes or esis. Economic instruments may support compli- programmes to encourage ISO 14001 certification ance with environmental regulation by rewarding through training and advertising the benefits of champions and punishing laggards within the certification. Empirical evidence supports this. industry. One way to do so, is through a ‘feebate’, For instance, Powers et al. (2011) found that India’s a differentiated tax and subsidy scheme around Green Rating Programme caused large pulp and an emission target: Firms are taxed for pollution paper plants with the worst environmental perfor- emissions above the target, and subsidised for mance in their study to reduce emissions of emission units below it. Examples include the certain pollutants by 9 to 19 per cent. In the same so-called ‘eco-bonus-malus’ scheme implemented vein, Garcia et al. (2007) estimated that Indone- in France for vehicle CO2 emissions (d’Hault- sia’s PROPER programme reduced firms’ emis- foeuille et al. 2014). Cars that emit more than the sions by one-third. standard are taxed, while those emitting less are

5. INCREASE A SECTOR’S PRODUCTIVITY THROUGH KNOWLEDGE SPILLOVERS

5.1. KNOWLEDGE SPILLOVERS AS A technologies become public knowledge when MARKET FAILURE transferred to production processes, firms will often not get the full return on their research Green policies can create demand and may there- and development investments: Some, if not all, fore provide incentives to invest in research and of the knowledge embodied in the invention also development in this sector and may therefore becomes available to competitors who can fully provide incentives to invest in research and or partly copy or improve the new technology. development in this sector, fostering innovation These knowledge spillovers benefit the economy, in environmentally sound technologies. These and society as a whole, whenever new technol- may generate positive externalities among firms ogies are developed; yet they discourage indi- in the research and development process. As vidual firms from investing in new technologies. knowledge is by nature a public good, and new As a result, market forces underprovide research Green Industrial Policy - Concept, Policies, Country Experiences

and development investment. Public policies suggests two particularities of the innovation 46 that foster investment in environmentally sound process in emerging economies. First, emerging technologies should mitigate this market failure economy countries tend to specialise in adapting for the benefit of all and assure that an optimal green technologies to local conditions. Second, level of research and development investment is spillovers are likely to be greater within the coun- provided, including for innovative firms. try itself than abroad, which is an argument for increasing support to research and development in 5.2. KNOWLEDGE SPILLOVERS FOSTERING emerging economy countries, as most of the gains GREEN INNOVATION of innovation will benefit local constituents. Recent estimates suggest that knowledge spill- 5.3. POLICIES FOR GREEN INNOVATION overs do also have significant positive effects on green innovation. Dechezlepretre et al. (2014) As we have seen in section 4, there is now ample analyse knowledge spillovers in clean and dirty evidence that environmental regulation can stim- technologies, based on patent citations. This infor- ulate innovation in green technologies. Moreover, mation is part of the state-of-art patent applica- knowledge spillovers can lead to productivity-en- tions: Innovators applying for a patent are required hancing innovation reinforcing the positive effect to cite all previous innovations on which the new described by the Porter Hypothesis. Public policies innovation is based. A citation indicates that the that increase the demand for green technologies knowledge contained in the document has been do not only reduce pollution and the use of natural useful to develop the innovation. For this reason, resources, they also foster innovation and there- patent citation can be seen as a measure of knowl- fore growth. edge spillovers. The 2014 study by Dechezleprêtre et al. covers four technological fields: energy produc- Governments can promote innovation in green tion, automobiles, fuel and lighting. Clean patents technologies in several ways. First, firms should receive an average of 43 per cent more citations be rewarded for investing in research and devel- than dirty patents, and are cited by more prominent opment, which means protecting their inventions patents. These results suggest that public support with effective patents, thus mitigating the problem to research and development would be more effec- of market failure as outlined in 5.1. Public author- tive in boosting innovation and growth if they ities can support green innovations by granting targeted green technologies. patents more easily, reducing transaction costs for submitting new patents, and enforcing the The positive effect of policy supporting research property rights of patents. This means facilitat- and development is likely to hold primarily for ing patent monitoring and litigation through the developed countries where most of the innova- judicial system. Another measure can also be to tion occurs, including for green technologies. For facilitate technology transfers through licensing instance, Lanjouw and Mody (1996) found that the agreements. Also, firms can be rewarded for invest- US, Japan and Germany accounted for two-thirds ment in research and development by the provision of climate mitigative or adaptive innovation. Yet of subsidies or tax cuts. emerging economies increasingly produce a significant share of green innovations. In the sample they Second, since technological absorptive capacity analysed, Dechezleprêtre et al. (2011) found that is seen by some authors as a determinant of a 18.5 per cent of the climate-oriented innovations country’s ability to innovate, governments should patented from 2000 to 2005 originated in China, invest in education, technological training and South Korea, Russia or Brazil. It is likely that emerg- knowledge dissemination infrastructure, including economy countries will catch-up on green ing internet access. In emerging economies, these innovation, triggered by domestic green policies, by factors can help foster innovation in green technol- the demand for green technologies from developed ogies that are best suited to local conditions. In less countries and by investment in carbon offsetting. developed countries, this can facilitate the transfer of green but complex technologies in some sectors Another important component of the geography such as the energy sector for wind and solar power. of innovation is knowledge dissemination. Deche- As Vidican Auktor (2017, this volume) shows in her zleprêtre et al. (2011) estimated the export of climate case study on renewable energy in Morocco, the mitigative inventions by country. They found that country provides significant opportunities for the emerging economy countries tend to export less localisation of certain products and services and than developed countries: around 7 per cent for the creation of green jobs. Moreover, such invest- China or for Brazil, as compared to 42 per cent for ments may be able to further attract CDM and the United States or 56 per cent for Germany. This carbon offset projects. Chapter 3 Gaining Competitive Advantage with Green Industrial Policy

6. SUMMARY AND CONCLUSION FOR GREEN GROWTH 47 Being environmentally sound does not need to might not be effective in the future. In the last be detrimental to competiveness. A firm can years, the generous support for wind and solar deploy several strategies to reduce its negative power through subsidies has been cut in many effect on the environment, while at the same countries in Europe, with some countries like time securing a competitive advantage in inter- Germany now switching from a feed-in tariff national markets. It can invest in environmental system that had long been considered successful research and development, adopt cleaner tech- to a competitive bidding process. For a variety of PART 2 PART nologies, supply environmentally sound technol- reasons, CDM projects have become less attrac- ogies and enhance its product’s environmental tive in recent years. The volume of carbon-off- quality continually. Even if those strategies turn setting projects has fallen since 2012. In 2015, the out to be costly, they can be profitable. Investment market experienced an excess supply of projects. in cleaner technologies can lead to productivity The average price per ton of CO2 for all projects improvements in the long run, which will then was US$ 3.3 in 2015, with a record US$ 0.1 for the spread through knowledge spillovers. Carbon lowest valued projects (Hamrick and Goldstein finance, such as investment in CDM and other 2016). However, the 2015 Paris Agreement might carbon offsetting projects, now drives demand. reverse this trend. In Article 6, the Agreement Renewable energy mandates contribute to this launches a new mechanism that aims at financ- demand, through measures like feed-in tariffs ing carbon-offsetting projects in developing and renewable portfolio standards. In response, countries through Internationally Transferred firms that specialise in producing climate-ori- Mitigation Outcomes. Importantly, policy instru- ented technologies are expanding, particularly in ments need to be well suited to the particular emerging economies, through the installation of context in which they are employed and therefore solar photovoltaic panels, wind turbines and other will require readjustments over time. renewable energy sources. Finally, the growth of organic farming and fair trade has created new Second, policies implemented for enhancing opportunities in agriculture and the food industry profitable green growth should suit a country’s in many developing countries. respective level of development. Less developed countries should prioritise improving their Many public policies can help to secure a compet- technological absorptive capacity, simplifying itive advantage with green business strategies. and standardising the accreditation process for First, environmental policy instruments targeted carbon offsetting projects and building up a reli- at fostering green innovation should be flexible. able supply of green, certified products for export. This means implementing economic instruments Emerging economy countries can afford to subsi- such as refunded emission taxes or tradable dise investment in green technologies to support allowances rather than prescribing technologies their own industry. They should also strengthen through specific regulation. Second, industrial their intellectual property rights to attract tech- policy should make patenting and technological nological knowledge from foreign investors and transfer easier and more effective. It should also encourage its transfer. favour public support for environmental innovation to mitigate underinvestment due to knowl- Finally, it is worth mentioning that protecting edge spillovers. Third, technological absorption natural resources and reducing pollution enhance capacity should be improved by investing in societal well-being through several channels that education, technological training and infrastruc- can be indirectly beneficial to firms. Economic tures, such as communication, transportation and activities rely on ecosystem services provided by energy. Fourth, governments should work with forests, soils, rivers, lakes and oceans. Workers non-government organizations and international are in better health and thus more productive if organizations to facilitate environmental labelling they have access to clean water and air. All those with transparent criteria and a reliable traceabil- indirect effects should be included in an accurate ity of products throughout the supply chain. cost-benefit analysis of green policies.

To conclude, three issues deserve further mentioning: First, public policies that have been successful in the past in bringing green growth Green Industrial Policy - Concept, Policies, Country Experiences

48 REFERENCES Alpay, E., Kerkvliet, J., & Buccola, S. (2002). Produc- Dragusanu, R., Giovannucci, D., & Nunn, N. (2014). The tivity growth and environmental regulation in Economics of Fair Trade. Journal of economic Mexican and US food manufacturing. American perspectives, 28(3), 217–236. journal of agricultural economics, 84(4), 887–901. European Commission. (2017). The EU Emissions Ambec, S., & Barla, P. (2006). Can Environmental Trading System (EU ETS). Retrieved from ec.eu- Regulations be Good for Business? An Assess- ropa.eu/clima/policies/ets ment of the Porter Hypothesis. Energy Studies Gandenberger, C., Bodenheimer, M., Schleich, J., Review, 14(2), 42–62. Orzanna, R., & Macht, L. (2015). Factors driving Ambec, S., Cohen, M. A., Elgie, S., & Lanoie, P. (2013). international technology transfer: Empirical The Porter Hypothesis at 20: Can Environmental insights from a CDM project survey. Climate Regulation Enhance Innovation and Competitive- Policy, 16(8), 1065–1084. ness? Review of Environmental Economics and García, J. H., Afsah, S., & Sterner, T. (2009). Which Policy, 7(1), 2–22. Firms are More Sensitive to Public Disclosure Archibugi, D., & Coco, A. (2004). A new indicator of Schemes for Pollution Control? Evidence from technological capabilities for developed and Indonesia’s PROPER Program. Environmental and developing countries. World development, 32(4), Resource Economics, 42(2), 151–168. 629–654. Haites, E., Duan, M., & Seres, S. (2006). Technology Bacon, C. (2005). Confronting the Coffee Crisis: Can transfer by CDM projects. Climate Policy, 6(3), Fair Trade, Organic, and Specialty Coffees Reduce 327–344. Small-Scale Farmer Vulnerability in Northern Hamrick, K., & Goldstein, A. (2016). Raising ambi- Nicaragua? World development, 33(3), 497–511. tion. State of the voluntary carbon markets 2016. Berman, E., & Bui, L. T. M. (2001). Environmental Washington, D.C. Regulation and Productivity: Evidence from Oil Haned, N., Lanoie, P., Plouffe, S., & Vernier, M. F. (2015). Refineries. Review of Economics and Statistics, Profitability of Ecodesign: An Economic Analysis 83(3), 498–510. (Working Paper). Canada. Chabé-Ferret, S., & Subervie, J. (2013). How much Jaffe, A. B., & Palmer, K. (1997). Environmental Regu- green for the buck? Estimating additional and lation and Innovation: A Panel Data Study. Review windfall effects of French agro-environmental of Economics and Statistics, 79(4), 610–619. schemes by DID-matching. Journal of Environ- Johnstone, N., & Labonne, J. (2007). Environmental mental Economics and Management, 65(1), 12–27. policy, management and R&D. OECD Economic Cosbey, A. (2017). Trade and Investment Law and Studies, 2006(1), 169–203. Green Industrial Policy. In Altenburg, T., & Kossoy, A., & Ambrosi, P. (2010). State and trends of Assmann, C. (Eds.). (2017). Green Industrial the carbon market 2010. Washington, D.C. Policy. Concept, Policies, Country Experiences (pp. 134–151). Geneva, Bonn: UN Environment; Kunzik, P. (2003). National Procurement Regimes German Development Institute / Deutsches Insti- and the Scope for the Inclusion of Environmen- tut für Entwicklungspolitk (DIE). tal Factors in Public Procurement. In N. John- stone (Ed.), The environmental performance of Dechezleprêtre, A., Glachant, M., Haščič, I., Johnstone, public procurement. Issues of policy coherence N., & Ménière, Y. (2011). Invention and transfer of (pp. 193–220). Paris: Organisation for Economic climate change–mitigation technologies: a global Co-operation and Development. analysis. Review of environmental economics and policy, 5(1), 109–130. Lanjouw, J. O., & Mody, A. (1996). Innovation and the international diffusion of environmentally Dechezleprêtre, A., Märtin, R., & Mohnen, M. (2014). responsive technology. Research Policy, 25(4), Knowledge spillovers from clean and dirty 549–571. technologies. CEP discussion paper / Centre for Economic Performance: Vol. 1300. London: Centre Lanoie, P., Laurent-Lucchetti, J., Johnstone, N., & for Economic Performance. Ambec, S. (2011). Environmental Policy, Innova- tion and Performance: New Insights on the Porter De La Tour, A., Glachant, M., & Ménière, Y. (2011). Inno- Hypothesis. Journal of Economics & Manage- vation and international technology transfer: The ment Strategy, 20(3), 803–842. case of the Chinese photovoltaic industry. Energy Policy, 39(2), 761–770. Lecocq, F., & Ambrosi, P. (2007). Policy Monitor. The Clean Development Mechanism: History, d’Haultfœuille, X., Givord, P., & Boutin, X. (2014). The Status, and Prospects. Review of Environmental Environmental Effect of Green Taxation: The Economics and Policy, 1(1), 134–151. Case of the French Bonus/Malus. The Economic Journal, 124(578), F444-F480. Lins, C., Williamson, L. E., Leitner, S., & Teske, S. (2014). The first decade 2004–2014: 10 years of renewable energy progress. Chapter 3 Gaining Competitive Advantage with Green Industrial Policy

Méndez, V.E., Bacon, C.M., Olson, M., Petchers, S., Popp, D. (2006). International innovation and diffu- Herrador, D., Carranza, C., Trujillo, L., Guadarra- sion of air pollution control technologies: The 49 ma-Zugasti, C., Cordon, A., & Mendoza, A. (2010). effects of NOX and SO2 regulation in the US, Effects of Fair Trade and organic certifications on Japan, and Germany. Journal of Environmental small-scale coffee farmer households in Central Economics and Management, 51(1), 46–71. America and Mexico. Renewable Agriculture and Porter, M. E. (1991). America’s Green Strategy. Scien- Food Systems, 25(03), 236–251. tific American, 264(4), 168. Morrison, A., Raju, D., & Sinha, N. (2007). Gender Porter, M. E., & van der Linde, C. (1995). Toward a New Equality, Poverty And Economic Growth. New Conception of the Environment-Competitiveness

Delhi: The World Bank. Relationship. Journal of economic perspectives, 2 PART Murphy, K., Kirkman, G. A., Seres, S., & Haites, E. 9(4), 97–118. (2013). Technology transfer in the CDM: An Powers, N., Blackman, A., Lyon, T. P., & Narain, updated analysis. Climate Policy, 15(1), 127–145. U. (2011). Does Disclosure Reduce Pollution? Never, B., & Kemp, R. (2017). Developing green tech- Evidence from India’s Green Rating Project. nologies and phasing them in. In Altenburg, T., & Environmental and Resource Economics, 50(1), Assmann, C. (Eds.). (2017). Green Industrial Policy. 131–155. Concept, Policies, Country Experiences (pp. Ramesh, P., Panwar, N. R., Singh, A. B., Ramana, S., 87–101). Geneva, Bonn: UN Environment; German Yadav, S. K., Shrivastava, R., & Rao, A. S. (2010). Development Institute / Deutsches Institut für Status of organic farming in India. Current Entwicklungspolitk (DIE). Science, 98(9), 1190–1194. Olsen, K. H., & Fenhann, J. (2008). Sustainable Schmid, G. (2012). Technology transfer in the CDM: development benefits of clean development The role of host-country characteristics. Climate mechanism projects. A new methodology for Policy, 12(6), 722–740. sustainability assessment based on text analysis Simpson, R.D., & Bradford, I. R. L.I.I. (1996). Taxing of the project design documents submitted for Variable Cost: Environmental Regulation as validation. Energy Policy. (36), 2819–2830. Industrial Policy. Journal of Environmental Østergaard, C. R., Timmermans, B., & Kristinsson, K. Economics and Management, 30(3), 282–300. (2011). Does a different view create something Stefan, A., & Paul, L. (2008). Does It Pay to Be Green? A new? The effect of employee diversity on innova- Systematic Overview. Academy of Management tion. Research Policy, 40(3), 500–509. Perspectives, 22(4), 45–62. Palmer, K., Oates, W. E., & Portney, P. R. (1995). Tight- Vidican Auktor, G. (2017). Renewable energy as a ening environmental standards: The benefit-cost trigger for industrial development in Morocco. In or the no-cost paradigm? Journal of economic Altenburg, T., & Assmann, C. (Eds.). (2017). Green perspectives. (9(4)), 119–132. Industrial Policy. Concept, Policies, Country Expe- Partnership for Action on Green Economy (PAGE). riences (pp. 153–165). Geneva, Bonn: UN Environ- (2017). Green Industrial Policy and Trade: A Tool- ment; German Development Institute / Deutsches Box. Geneva. Institut für Entwicklungspolitk (DIE) Pegels, A. (2017). Germany: The energy transition Wara, M. (2007). Is the global carbon market working? as a green industrial development agenda. In A special collection. Nature, 445(7128), 595–596. Altenburg, T., & Assmann, C. (Eds.). (2017). Green Industrial Policy. Concept, Policies, Country Expe- riences (pp. 166–184). Geneva, Bonn: UN Environ- ment; German Development Institute (DIE Green Industrial Policy - Concept, Policies, Country Experiences 50 CHAPTER 4 ENHANCING JOB CREATION THROUGH GREEN TRANSFORMATION

Michela Esposito, Alexander Haider, Daniel Samaan, Willi Semmler Chapter 4 Enhancing Job Creation Through Green Transformation

1. INTRODUCTION 51 Over the last decades it has become clear that This chapter identifies and discusses opportu- economic growth has to be environmentally nities and challenges for labour markets and sustainable and socially inclusive. A production provides examples of policies and practices mode is needed that is less reliant on fossil fuels that can help achieve an inclusive and employ- and uses natural resources more efficiently (OECD ment-intensive green transition. Section 2 2014; UNEP 2011). To this end, economies will have provides a systematic overview of how green to go through a process of structural transforma- transformations may affect the quantity and qual- PART 2 PART tion. This process can be induced and guided by ity of employment. In section 3 we undertake an green industrial policies. Such structural change attempt to define green jobs. Section 4 presents towards an environmentally sustainable econ- empirical evidence on employment trends in omy will have, and already has, considerable those labour market segments that can be attrib- effects on labour markets and incomes (ILO 2011; uted relatively well to green sectors. Section 5 ILO and UNEP 2012). While green transformation introduces an overview of labour market and poses challenges for certain polluting sectors and social policies that can facilitate a green transi- related jobs, it also creates opportunities for new tion from a social point of view. The conclusion jobs in other industries and may open the way for summarizes the key findings of this chapter. more and better jobs in the future.

2. HOW GREEN TRANSFORMATIONS AFFECT EMPLOYMENT Industrial policies are commonly understood of natural resources such as heavy industries or as government policies that attempt to strategi- fisheries and sectors with heavy polluters such as cally strengthen the development and growth of fossil fuel-based energy production. These sectors certain economic activities and that very often, will see the largest adjustments in employment, but not necessarily, concern parts of the manu- while effects in other parts of the economy may facturing sector. If industrial policies achieve not be as intense. their goals, they lead to a structural transformation of the economy, or accelerate or decelerate Affected sectors will not necessarily be identical an already ongoing transformation. During this across countries. Improving or preserving envi- process, inevitably, new jobs and occupations ronmental quality covers a whole range of poten- emerge while others disappear. The net employ- tial issues, ranging from pollution of air and water, ment effect of such a transformation is the differ- global climate change, ocean degradation, waste ence between new jobs created and old jobs lost. production, depletion of non-renewable resources, unsustainable use of renewable resources, loss of Green industrial policies, specifically, target biodiversity, to degradation of soil (ILO and UNEP the improvement of environmental quality and 2012; ILC 2013). As countries face different environ- contribute to environmental sustainability (Alten- mental challenges, the requirements of structural burg and Rodrik 2017, this volume). Most envi- transformation vary, as do related employment ronmental policies contain features of industrial challenges and outcomes. For example, green- policies, as they provide incentives to accelerate house gas (GHG) emissions in Brazil arise largely the development of certain sectors, sub-sectors from agriculture and forestry, whereas in Germany and technologies and to phase out others. As a they mostly stem from electricity generation, result, sectors with strong effects on the environ- manufacturing and the transport sector. The ment, as well as workers and enterprises operat- German energy sector has traditionally used ing in these sectors, will be more affected than more non-renewable resources such as coal, oil others. The first step in understanding the effects and gas than Brazil, which traditionally relied on of green industrial policies on labour markets and hydropower; and per capita water consumption incomes is to identify the sectors that strongly for agricultural purposes is much higher in Brazil affect the environment. These include sectors than in Germany. These brief examples show that with activities directly aimed at the preserva- countries face very different challenges on their tion or restoration of the environment such as paths towards an environmentally sustainable for recycling, renewable energy, or eco-tour- future. Accordingly, workers and enterprises will ism; sectors strongly relying on the utilisation be affected differently. Green Industrial Policy - Concept, Policies, Country Experiences

When trying to measure the employment effects, Moreover, employment considerations should 52 various dimensions need to be taken into not only concern the number of jobs created or account. On the one hand, there are jobs that lost, but also their quality, including governance are newly created in greener sectors, industries and coordination mechanisms. This comprises or enterprises as well as those that emerge due a whole range of themes relating to the world to spillover effects in non-green sectors. Spill- of work. The International Labour Organization over jobs in other sectors may emerge through (ILO) has developed a Decent Work Agenda that supply chains into green sectors as indirect jobs, distinguishes four objectives that also address through increased capacity of non-green sectors quality of employment (ILO 2017): promoting jobs as improved competitiveness through higher by boosting investments, skills development, new resource productivity, or through additional job creation and sustainable livelihood; guaran- consumer demand resulting from newly created teeing rights at work; extending social protection incomes as induced effects. On the other hand, by ensuring equal and safe working conditions environmental policies may negatively affect for men and women, ensuring a balance between labour absorption in polluting industries. For working hours and rest and access to healthcare; example, successful development of a competitive and promoting social dialogue between employ- renewable energy sector would most likely reduce ees and employers. The promotion of “sustained, the number of jobs in the coal mining industry inclusive and sustainable economic growth, full and along its supply chain. Estimating the over- and productive employment and decent work for all impact of green industrial policies on the total all” has been included into Sustainable Develop- number of jobs in the economy is therefore diffi- ment Goal 8 as part of the United Nations’ 2030 cult and poses some methodological challenges. Agenda for Sustainable Development that aims to achieve economic, social and environmental While net employment effects of a transition sustainability by 2030 (UN 2015). are important and of high interest to policymakers, the gross effects—the total numbers of jobs Many aspects of the world of work are regulated created and lost in different segments of the by International Labour Standards issued by the economy—are also essential from an employment ILO. The adherence to International Labour Stand- policy perspective. The shift of jobs across enter- ards can help improve employment outcomes. prises, industries, and sectors may entail adjust- A green transformation is successful, from the ment costs for enterprises and workers. Additional perspective of green industrial policies and the government policies may become necessary to ILO’s Decent Work Agenda, if the transition leads support or even trigger these transitions. Shifting to less environmental damage by the economy, employment patterns can also affect occupations, for example lower total GHG emissions per unit defined as a category of jobs with main tasks and of gross domestic product, and to better employ- duties that are characterized by a high degree of ment outcomes, in terms of more jobs, better job similarity, and therefore have ramifications for the quality and better compliance with International development of skills and re-training as well as Labour Standards. adjustments of the education system (ILO 2012).

3. DEFINING GREEN JOBS There is no generally agreed definition of green environmental performance, may fall behind jobs. One challenge lies in defining when a sector, international standards over time and would no and accordingly an employment or occupation in longer be considered green. this sector, is green. Only a few economic sub-sectors, such as the production of wind turbines or Furthermore, it should be considered that a green solar panels, can be unambiguously classified as product or service can be produced through an green industries. In most activities, boundaries environmentally unsustainable production between green and non-green are blurred. When process such as a wind turbine built with inputs produced and transported in a CO intensive way; an auto part is incorporated in an electric vehicle 2 it may be considered a green product, but when and a green production process can produce a the same part is assembled into a conventional non-green product or service, such as produc- car it would not be considered a green product. tion of petrol-powered cars with reduced water An additional difficulty results from the fact that use. Hence, it is important to distinguish between technology evolves. An industry that is green environmental impacts of production processes today, in the sense that it has an above-average and outputs. Chapter 4 Enhancing Job Creation Through Green Transformation

The definition of green jobs used by ILO includes be carried out for sale or for own consumption. those in sectors that produce green goods The US Bureau of Labour Statistics uses a similar 53 and services as well as occupations in green definition. The European Commission’s statistics processes that are environmentally favourable agency, Eurostat, uses Environmental Goods and (ILO and UNEP 2012). In addition, the ILO green Services Sector (EGSS) categories (Eurostat 2009). jobs definition includes the constraint that they According to its definition, EGSS categories are must be decent jobs. The International Confer- goods and services produced for two general ence of Labour Statisticians (ICLS) proposed to purposes: environmental protection–preventing, consider criteria that define jobs that belong to reducing and eliminating pollution and any other the environmental sector as well as criteria that degradation of the environment; and resource 2 PART define decent jobs. The Conference proposal management–preserving and maintaining the is meant to be consistent with the standards stock of natural resources and hence safeguard- used in labour statistics, such as some sets of ing against depletion. decent work indicators from the ILO, and with the standards in environmental statistics, such The definition of green jobs to be used will as the System of Environmental-Economic depend on its purpose. For example, to deter- Accounting (SEEA), established by the United mine new skill requirements of new occupations Nations (United Nations Statistics Division 2003; and new jobs, a narrow definition of green jobs in UN 2012). Based on the latter, the ICLS defines environmentally related activities may be most environmental activities as activities with the helpful; while an analysis of the social dimen- aim of eliminating or diminishing pollution and sions of a green transition may require a broader deterioration of the environment or of using focus on employment effects throughout the natural resources more efficiently. These envi- economy. No matter which definition is applied, ronmental activities can be the main activity of green jobs are in any case only a subset of the an economic unit or an auxiliary one, and can overall employment effects that will arise during a green transition.

4. EMPLOYMENT EFFECTS OF GREEN TRANSFORMATIONS: WHAT DO WE KNOW? Given the blurred boundaries between green and occupations and skills requirements. Third, there non-green products and processes as well as are some estimates of the carbon-intensity of the heterogeneity of environmental challenges economic activities, which allow us to assess across countries, it is impossible to unambigu- which sectors are likely to experience the largest ously quantify the overall employment impacts of labour market changes. green transformation. However, some estimates exist for parts of the green transformation. A few 4.1. EVIDENCE FROM OECD COUNTRIES key studies have investigated the employment effects of green industries at national levels. Due Employment estimates of the environmental to different definitions used, these are not compa- sector have been conducted mostly within the rable, but taken together they provide a fairly good EU. First attempts to estimate its size were under- overview of trends. taken by the OECD and EU (OECD and Eurostat 1999). Ecotec (2001) was one of the first studies to This review uses three types of information: First, analyze the size of the sector within the EU for there is evidence from OECD countries, including the period 1994 to 1999. Over 2 million full-time aggregate statistics on the Environmental Goods equivalent green jobs existed in the EU in 1999, and Service Sector as well as publications that with approximately 1.5 million jobs in pollution shed light on specific sub-sectors and on impacts management and around 650,000 jobs in resource of certain green policies on labour markets in management. Employment growth rates in these these countries. Second, there is documentation activities were above economy-wide averages. from a selected number of developing countries The fraction of employment in the environmental showing the employment gains at initial stages sector over total paid employment was approxi- of transformation as green industries emerge. mately 1.3 per cent. It should be noted that none of these sources tracks job losses in polluting industries, spillover As the number of EU countries expanded, the effects into non-green industries, or changing Ecotec report was updated by Ernst & Young in Green Industrial Policy - Concept, Policies, Country Experiences

2006, showing that total employment increased EU-15 to 1.04 million jobs in the EU-25 over the 54 from 1.45 million jobs in the EU-15 of 1999 to same period (Ernst&Young 2006). A 2009 update 1.85 million in the EU-25 of 2004 in pollution analyzed green jobs through to EU-27, between management activities while resource manage- 2000 and 2008 (Ecorys 2009) (Table 4.1). ment activities increased from 0.6 million in the

Table 4.1: Employment estimates of green jobs in the European Union

Growth rate % Activity Employment (2000) Employment (2008) Average annual Waste management 844,766 1,466,673 7.1 Water supply 417,763 703,758 6.7 Wastewater management 253,554 302,958 2.3 Recycled materials 229,286 512,337 10.6 Others 129,313 193,854 5.2 Renewable energy 49,756 167,283 16.4 Air pollution 22,600 19,067 -2.1 Biodiversity 39,667 49,196 2.7 Soil & Groundwater 14,882 18,412 2.7 Noise & Vibration 4,176 7,565 7.7 Total 2,005,764 3,441,102 7.0 Source: Ecorys (2009)

As the UN established the SEEA framework, the energy from renewable sources, the production EU’s EGSS concept evolved to complement it. of wind and solar power stations and equipment, Eurostat shows that across the EU the share of and installations for heat and energy savings. EGSS in total value added grew by more than 50 Figure 4.1 shows the development of the EGSS in per cent between 2000 and 2012, while employ- selected EU countries and the EU28 between 2005 ment grew from 2.785 million to more than 4.1 and 2013. The growth rate of jobs has been mostly million in full-time equivalents during the same positive. According to Figure 4.2, the employ- period, which corresponds to an average annual ment shares of the EGSS for the latest available growth rate of about 3 per cent. The majority of year (2014) range from around 0.9 per cent of total these new jobs were created in energy resource employment in Ireland up to 4 per cent in Austria management, in particular the production of and Luxembourg. Chapter 4 Enhancing Job Creation Through Green Transformation

Figure 4.1: Employment in the environmental sector, 2005–2013 55 Panel A: Employment shares in EGSS in EU countries

4 PART 2 PART

EU28

ance

Poland

Austri

ithuania

Romani

Bulgaria

herlands

embour

German

Net

x Lu 2005 2013*

Note (*): For Austria, Germany, Luxembourg, Netherlands and Romania the shares are from 2012, for France and Lithuania, the shares are from 2011. For some countries the base year (2005) refers to 2007 or 2008. Thus, not comparable by definition.

Panel B: Annual employment growth in EGSS in EU countries, 2005–2013

Romania

Luxembourg

Germany

France

Bulgaria

Austria

EU28

-20 -10 010320 040

Source: ILO Research department based on Eurostat. Green Industrial Policy - Concept, Policies, Country Experiences

Figure 4.2: Employment shares in EGSS in EU countries, 2014 56

4.0

) 3.5

3.0

t share (% 2.5

2.0

Employme 1.5

1.0

0.5

nia

ance

Latvi

Irelan

Poland

Austri

Esto

enmark

Sweden

ithuania

Belg

Bulgaria

Slovenia

Republic

an Unio

Romani

embour

German

Kingdom

x Lu

Netherlatnd

Czech

Europe Country United

Until the function was defunded in 2013, green According to the projection, the green recovery job estimates were made in the United States programme would produce larger total employ- by the Department of Commerce and by the ment effects and generate higher household Department of Labor’s BLS using a case-by-case incomes than investing in alternative scenarios classification of green industries. The US Depart- such as the oil industry. ment of Commerce (2010) estimated the number of green jobs in the United States in the private Morgenstern et al. (2002) also focus on the US, sector in 2007 to be around 2.4 million, under a investigating the effects of environmental poli- broad definition, which amounted to 2 per cent cies for four industries: pulp and paper mills, of total private sector employment in that year. plastic manufacturers, petroleum refiners, and BLS (2012) estimated that around 3.4 million jobs iron and steel mills sectors. On average, a US$ in the United States could be classified as green 1 million increase in environmental spending jobs in 2011 with around 2.5 million of those in the would produce 2 to 3 additional jobs across the private sector. The BLS figures suggest that 2.3 four industries. Belova et al. (2013) revised those per cent of private sector jobs, and 4.2 per cent of estimates by increasing the number of industries public sector jobs, can be classified as green jobs included in the analysis. For the same four indus- in the US, but none of these numbers consider the tries, the authors find larger employment effects aspect of ILO’s Decent Work. Given the most up-to- than Morgenstern et al. (2002). For the other six date estimates, the biggest share in US green industries, results were positive and significant private sector employment was in manufacturing, only for the rolling and drawing industry. In followed by construction and professional, scien- particular, each additional US$ 1 million of envi- tific and technical services (BLS 2012). ronmental expenditure would produce 22 to 23 additional jobs in rolling and drawing. Finally, the For the United States, Pollin et al. (2008) estimate authors conclude that for 6 out of 10 industries, the projected employment effects of a proposed the model indicates 10 to 30 additional employ- 10-year green recovery programme. It would ees hired for US$ 1 million spent in abatement consist of ten broad implementation steps, rang- expenditures (Belova et al. 2013). ing from a cap-and-trade system to technology-specific programmes. The authors consider Kato et al. (2012) analyze the short-term effects six green investment areas: retrofitting build- of an environmental tax on structural change for ings, mass transit, smart grid, wind power, solar a set of nine high-income countries: Germany, power and advanced bio-fuels. They estimate Australia, France, Hungary, Japan, Korea, Sweden, that 935,200 direct jobs would result, along with UK and the US. They distinguish between high 586,000 indirect jobs and 496,000 induced jobs, by and low intensities of carbon-consuming sectors spending $100 billion in these investment areas. and evaluate three different policy options. Chapter 4 Enhancing Job Creation Through Green Transformation

The first option is a carbon tax levied on goods A number of studies document the evolution and produced by highly carbon-intensive sectors effects of ethanol production from sugarcane in 57 and a subsidy paid on goods produced by low Brazil (Smeets et al. 2006; Goldemberg et al. 2008; carbon-intensive sectors. The overall employ- Herreras Martinez et al. 2013; Motta Veiga and ment effect is estimated to be positive at around Polónia Rios 2017, in this volume). Goldemberg 0.5 per cent. Secondly, they examine the effects et al. (2008) estimate that for 300 million tons of a carbon tax levied on goods produced by high of sugarcane produced, 700,000 direct jobs are carbon-intensive sectors, without any subsidy in created, directly and mainly in agriculture, and return. In this case, the carbon tax would produce indirectly through supply chains for equipment, a decrease in employment of roughly 0.4 per chemical supplies, production and maintenance. 2 PART cent. The last policy option consists of a carbon The number of jobs per unit of energy produced tax levied on highly carbon-intensive sectors also exceeds the number of jobs required by the and a revenue-neutral wage subsidy paid to both oil industry for the same output. Concerning sectors. This policy would produce the greatest wages, the authors report that workers in the net gains in terms of employment and output. sugarcane industries in Sao Paulo State receive around 80 per cent higher wages than workers in UNEP (2008) uses a literature survey to estimate the production of other agricultural crops. Sugar- the employment effects of green transformation cane wages were even higher than 50 per cent and, in particular, energy efficiency improve- of those in services and 40 per cent of those in ments in six economic sectors: renewable energy, industry. Wages in the less-developed Northeast buildings, food and agriculture, basic industry Region remain lower than in Sao Paulo State, and recycling, transportation and forestry. In the although wages have been rising there, too. renewable energy sector, the net employment effect produced by green transformation ranges Employment in this sector is highly formalized: from 1.4 to 2.5 million jobs by 2020 in Europe, In sugarcane production, 72.9 per cent of the jobs while estimates for the US vary considerably. are formal, compared to only 40 per cent of other For instance, a study conducted by the Ameri- rural jobs in Brazil. In Sao Paulo State, formal can Solar Energy Society estimates that by 2030 employment in sugarcane production reached the US will have created 1.3 million direct and 7.9 93.8 per cent by 2005. In the Northeast Region the million indirect jobs in renewable energies under formal to informal employment ratio amounts a business-as-usual scenario in renewable energy to only 60.8 per cent, yet this is still higher than (UNEP 2008). According to IRENA (2017), renew- the average figure for agriculture. Since the early able energy employment was 1,163,000 in the 2000s, the number of jobs has been decreas- European Union and 777,000 in the United States ing in sugarcane and ethanol production due to in 2016. In the residential building sector, Wade, mechanisation (Smeets et al. 2006). As Herreras Wiltshire and Scrase (2000) found that energy-ef- Martínez et al. (2013) point out, mechanisation ficient investment programmes would produce has been deliberately promoted by policy deci- positive effects in nine European countries. In sions to decrease soil degradation and to reduce particular, for every EUR 1 million invested in air pollution due to sugarcane field burning used energy-efficiency programmes, 11.3 to 13.5 full- in manual harvesting (Motta Veiga and Polónia time equivalent direct jobs would be created. Rios 2017, in this volume). Still, the sector offers a considerable number of decent jobs in agriculture. 4.2. DOCUMENTATION FROM At the same time, skill levels have been increasing DEVELOPING COUNTRIES in the sugarcane industry since the early 2000s, while additional investment for job creation is For developing countries, the evidence is even relatively low in ethanol compared to the chemical more patchy. With regard to renewable energy and petrochemical industry (Smeets et al. 2006). excluding large hydropower, IRENA (2017) shows Herreras Martínez et al. (2013) forecast future major employment effects for China at 3,643,000 developments in biofuel production in Brazil’s jobs, Brazil at 876,000 jobs, India with 385,000 Northeast Region using input-output models. jobs and Bangladesh with 162,000 jobs in 2016. Different scenarios are simulated and increases in A general overview of a number of sources can productivity and employment, ranging from 12,500 provide more anecdotal evidence from a number up to 160,000 jobs until 2020, are estimated. of specific case studies in developing countries, ILO and UNDP (2011) assessed the green employ- describing both what already exists and projec- ment potential in Lebanon for five sectors: energy, tions for future green jobs. construction, waste management, agriculture and forestry. In the energy industry, the report Green Industrial Policy - Concept, Policies, Country Experiences

estimates that around 4,000 jobs would be created generation. It is in this sub-sector where green 58 by 2020 through the introduction of concentrated job gains are substantive and visible. Other green solar power, wind energy, photo-voltaic solar and occupations, such as those related to pollution by expansions of hydraulic power and solar water control and resource efficiency or circular econ- heater markets. Additional employment effects omy investments, may increase more slowly or would be due to the introduction of renewable may have remained below the radar given that energy systems in manufacturing. fewer statistics are available. To what extent green employment will increase in such areas Vidican Auktor (2017) reviews the development will of course depend on the level of ambition of of renewable energy based on solar, wind, and green industrial policies and related environmen- hydro energy in Morocco as a green industrial tal incentive schemes. policy. The earliest experience of Morocco with renewable energy stems from the Programme d’Électrification Rurale Globale (PERG) intro- 4.3. WHICH INDUSTRIES ARE LIKELY TO duced in 1996. The programme increased rural UNDERGO MAJOR EMPLOYMENT electrification to 98 per cent in 2010. Estimates TRANSFORMATIONS? suggest that 13,000 direct and indirect jobs were So far, evidence for job gains has been sought in created through the programme in 2006 (Vidi- narrowly defined employment categories, particu- can Auktor 2017, this volume). The Noor-Ouarza- larly renewable energy and other EGSS. Yet green zate solar power complex represents the biggest transformations go along with technological renewable energy project in Morocco and is change and shifts in relative prices that affect being built in phases with Noor I concentrated demand for labour and skill requirements in solar power being connected to the power grid virtually all industries. Effects may be direct and in 2016. It created approximately 1,800 jobs in the indirect, and the outcomes in terms of job quan- construction process and 250 permanent oper- tity and quality may be positive or negative, all of ations jobs. Noor II, III and IV concentrated solar which is difficult to predict. power are expected to go online in 2017 and 2018. The African Development Bank estimates that The magnitude of the employment challenge in the construction of Noor 2 creates 2,000 to 2,500 green transformation can be estimated by deter- jobs temporarily in the construction sector with mining the share of workforce in carbon-inten- 400 to 500 permanent operation jobs (IRENA sive sectors. People working in these sectors will 2016). Additional projects have been completed have to go through some transition. Either these in Tarfaya, where Africa’s largest wind farm was carbon-intensive sectors become greener through established. The wind farm went online in 2014 technological change such as an energy sector and 700 construction jobs were created during shift from fossil fuels to renewable energy or the the construction process. Additionally, 50 opera- respective sector will have to shrink. In the first tional jobs were established by the Tarfaya Wind case, workers in these sectors are likely to have Farm (IRENA 2016). Beyond renewable energy to adapt or modify their skills and working meth- generation, Siemens built a factory in Morocco ods, or to change enterprises. In the second case, producing rotor blades for onshore wind turbines. workers find fewer jobs in the sector and have to Siemens (2016) estimates that its new factory will move to other, greener sectors. The same applies create up to 700 jobs. to other unsustainable activities if environmental policies provide the respective signals, such as Lastly, Borel-Saladin (2013) focus on South Africa, agricultural subsectors that deplete soil fertility where the Green Economy Accord was signed and water or reduce biodiversity. in 2011 with the aim of creating at least 300,000 green jobs. Official estimates suggested that To estimate this challenge, exemplified for the 98,000 net jobs would be created in the short-run, decarbonisation challenge, this section utilizes and an additional 462,567 net jobs in the long-run. data from the World Input-Output Database The big driver of this increase would be natu- (WIOD). The database has been constructed by ral resource management. However, the lack of merging information from National Input-Out- skilled labour might impede growth. put tables and Socio-Economic Accounts for the period 1995 to 2011, along with Environmental Overall, this review of developing country docu- Accounts in the years 1995 to 2009. Using CO mentation reveals how difficult it is to estimate 2 intensity as a metric, it can be seen that most the employment impacts of green transition GHGs originate from just a few industries. Top and to compare development across countries, carbon-emitting sectors are similar in all 40 coun- with the notable exception of renewable energy tries for which data are available and include the Chapter 4 Enhancing Job Creation Through Green Transformation

energy sector, agriculture, transport, and heavy emissions of the HCS ranges from 79 per cent of manufacturing (Appendix). total GHG emissions in the Americas to 89 per 59 cent in Asia and the Pacific, using ILO regional As these sectors are responsible for more than 80 classification. Conversely, the contribution of per cent of GHG releases, the employment chal- low-carbon industries (LCS) to total GHG emis- lenges would mostly concern these top emitters, sions is very small, especially in Asia and the the high-carbon sector (HCS). The share of GHG Pacific, and always below 10 per cent.

Figure 4.3: Employment shares in GHG emitting industrial sectors by region PART 2 PART Europe and Central Asia Countries covered: Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, 22% Greece, Hungary, Ireland, Italy, Latvia, Lithuania, High-carbon sectors Luxembourg, Malta, Netherlands, Poland, Portugal, 41% Romania, Russian Federation, Slovakia, Slovenia, Low-carbon sectors Spain, Sweden and United Kingdom. Other sectors 37%

The Americas Countries covered: Brazil, Canada, Mexico and United States.

21% High-carbon sectors 49% Low-carbon sectors 30% Other sectors

Asia and the Pacific Countries covered: Australia, China, India, Indonesia, Japan, Republic of Korea and Taiwan. 12%

High-carbon sectors Low-carbon sectors 35% 53% Other sectors

Source: ILO Research Department based on the World Input-Output database (WIOD 2013).

Note: Last available year: 2009. Africa and Arab States are not included because of lacking data availability. High- carbon sectors include the top ten listed in the Appendix. At the same time, low-carbon sectors include the top ten listed in the Appendix. All the remaining sectors are classified as other sectors.

According to ILO (2011), the top fifteen carbon-in- carbon-intensive sectors are the largest providers tensive industries in high-income countries of employment in Asia and the Pacific, account- employ a relative small share of workers. The ing for 53 per cent of the workers in the sample largest share of the employed workforce in indus- (Figure 4.3). This discrepancy is due to the large trialised countries is not in the HCS. However, still employment shares in the agricultural, forestry about every fifth worker, 22 per cent, works in a and fishing sectors in developing and emerging carbon-intensive sector that will be, or is already, economies. The employment shares for these affected by green transformation. By contrast, three sectors can be between 30 and 80 per cent, Green Industrial Policy - Concept, Policies, Country Experiences

depending on the country, and all of these natu- carbon-intensity within highly aggregated natu- 60 ral-resource exploiting sectors are generally cate- ral-resource sectors: whether and to what extent gorized as carbon-intensive. It should be noted agriculture or forestry actually emits carbon, or that this classification only provides a rough even sequestrates it, depends on the way the approximation, given enormous differences of sector is operated.

Figure 4.4: Share of CO2 emissions by industry and ILO region

Europe and Central Asia Countries covered: Austria, Belgium, Bulgaria, Cyprus, Czech Republic, 4% Denmark, Estonia, Finland, France, Germany, 10% Greece, Hungary, Ireland, Italy, Latvia, Lithuania, High-carbon sectors Luxembourg, Malta, Netherlands, Poland, Portugal, Romania, Russian Federation, Slovakia, Slovenia, Low-carbon sectors Spain, Sweden and United Kingdom. Other sectors 86%

The Americas Countries covered: Brazil, Canada, Mexico and United States. 7%

14% High-carbon sectors Low-carbon sectors Other sectors 79%

Asia and the Pacific Countries covered: Australia, China, India, Indonesia, Japan, Republic 2% 9% of Korea and Taiwan.

High-carbon sectors Low-carbon sectors Other sectors 89%

Source: ILO Research Department based on the World Input-Output database (WIOD 2013).

Note: All charts refer to the year 2011 (last available data point). Africa and Arab States are not included because of lacking data availability. High-carbon sectors include the top ten listed in the Appendix. At the same time, low-carbon sectors include the top ten listed in the Appendix. All the remaining sectors are classified as other sectors.

Still, these estimates highlight the challenges and emerging economies, a much larger propor- faced by developing and emerging economies in tion of the work force will be affected. Policies terms of employment. In industrialised countries, therefore need to help workers to transition from mainly heavy industries and the energy sector the agriculture sector to productive, non-re- are going through the green transformation, with source-intensive 'decent work' in other sectors of relative small employment shares; in developing the economy. Chapter 4 Enhancing Job Creation Through Green Transformation

5. POLICY OPTIONS TO STRENGTHEN 61 LABOUR MARKET PERFORMANCE Labour market challenges can be better addressed 5.2. PROMOTING ENTREPRENEURSHIP when effective policies are in place. Most labour market policies, including active as well as Green transformation will not necessarily be passive labour market tools, can support the effi- driven by large, existing firms. It offers opportu- ciency and equity of economic transformations. nities for start-ups and small firms that develop

Green industrial policies should therefore have new ideas and green innovations. In this regard, 2 PART employment and social policy components. Good employment considerations should not only jobs and business success in emerging green concern employees but also the promotion of sectors are interdependent. On the one hand, self-employment and entrepreneurship. Green workers want to benefit from new and decent jobs; industrial policies should include a variety on the other hand, only if workers are well-trained, of measures for that purpose, for example to mobile enough and prepared to take on new chal- improve access to credit or support the formaliza- lenges can a green transformation be successful. tion of informal enterprises, which is particularly important in developing countries. 5.1. MOBILITY ASSISTANCE AND SKILL ADJUSTMENTS 5.3. REVENUE-NEUTRAL GREEN TAXES

The workforce needs to be allowed, and encour- Empirical evidence suggests that levying envi- aged, to successfully shift through a green ronmental taxes while reducing taxes on labour transition. The ability to manage transitions produces positive employment effects (Schlegel- successfully is included in the ILO work under the milch et al. 2017, this volume). This approach, term workforce mobility. Such mobility can take called revenue recycling, increases economic various forms. Workers need empowerment to efficiency by internalizing environmental costs. fulfill new assignments or take on new jobs within Considering the economic consequences in the same enterprise, for example in an electric isolation from social and health benefits, three utility company that shifts its operations from interacting effects must be managed wisely: the fossil fuel based energy to renewables. Workers primary cost of the introduction of new green may also have to change jobs, leaving the enter- taxes will reduce economic activity; revenue recy- prise but remaining in the same industry, or they cling will have a positive effect on the economy; may have to shift to different sectors. and levying new taxes on top of already existing taxes that interact with the existing taxes will To facilitate these transitions, dedicated enter- add additional costs (Goulder 2013). Thus, from a prises should be supported in establishing purely economic perspective, the potential of an training systems to ensure that employed and economic expansion depends on the strength of unemployed workers can upgrade and adjust their the revenue recycling effect, in relation to the sum skills or even be retrained as needed. Reform of of the first and third effects. education systems may also be necessary, especially to help young workers enter the labour The first effect has been discussed widely in the market. Public employment services can help context of high-income countries. The Canadian re-allocate job seekers to open positions. province of British Columbia introduced a revenue-neutral carbon tax in 2008 with reductions Geographical mobility should also be supported, in corporate and personal income taxes, and with especially when massive shifts of the industry lump-sum payments to low-income households. structure result from green industrial policies. Yamakazi (2015) reports that emission-intensive This may be the case, for example, when shifting and trade-exposed industries suffered from fall- from fossil to renewable energy: Fossil energy ing employment opportunities due to the new plants are usually located next to energy-intensive tax, while other sectors increased their employ- industries and cities, whereas renewable power ment, suggesting a movement of labour between plants are located where the respective power sectors. On an aggregated level, he concludes resources are abundant. Shifting energy systems that the tax reform had positive effects on will therefore affect spatial distribution, and work- employment. In Germany, green taxes have been ers will have to move physically (ILO 2011). recycled to cut pension contributions, thereby Green Industrial Policy - Concept, Policies, Country Experiences

lowering labour cost. Estimates suggest that the et al. (2017) suggest that revenues should be recy- 62 green tax reform created around 250,000 jobs in cled in ways that meet the respective national 2003 alone (EEA 2011). priorities and enhance political support for environmental tax reforms. Verifiable evidence for lower and middle-income countries is very limited. Findings from high income countries may not be representative of 5.4. DECENT WORK AND SOCIAL DIALOGUE developing countries because of differences in Social dialogue is a mechanism through which factor prices, endowments and labour market interested representative groups, such as trade characteristics (Bowen and Kuralbayeva 2015). unions and business associations, and the Importantly, the economies of many developing government aim to build consensus among major countries are characterized by large informal stakeholders. Effective dialogue can help resolve labour markets. Kuralbayeva (2013) focuses explic- crucial social and economic issues and improve itly on modeling labour market characteristics economic performance. Given that the transition of developing countries with an informal sector towards a greener economy will entail profound and rural-urban migration. She concludes that changes in production processes and technolo- revenue-neutral green tax policies can reduce gies, as well as reallocations of jobs, close coop- unemployment in the private sector: by using the eration between government and the social additional government income to reduce payroll partners will be central to the success of this taxes, the cost of employment in the formal transformation (ILO and UNEP 2012). sector decreases. At the same time, environmental taxes decrease the income of self-employment A greener economy does not automatically create in the informal sector. Thus, in the model, the tax high-quality, decent jobs. Job quality needs to be burden shifts towards the informal sector, which monitored and measures taken to ensure that then creates incentives to operate formally and labour legislation is applied and workers and reduces formal unemployment (Kuralbayeva 2013). employers can organize and make use of collective bargaining. In this regard, International Landa Rivera et al. (2016) assess the effect of Labour Standards provide both a legal and insti- carbon taxes and revenue recycling on the Mexi- tutional framework and practical guidance for can economy. They conclude that such reform work in a greener and more sustainable economy. would be beneficial for the economy and the envi- Similarly, while a green economy is very likely to ronment, yielding a double dividend: emissions be healthier and safer for workers and the public, would decrease, while GDP, employment, wages, caution is nonetheless needed to prevent possible but also inflation, would increase. Schlegelmilch new occupational hazards (ILO and UNEP 2012).

6 CONCLUSIONS A green structural transformation creates oppor- may happen in other sectors as well, and the tunities for more and better jobs, but it also poses narrowly defined EGSS may have spillover effects challenges for employment and incomes. To on other activities. While on the one hand new evaluate the impacts of green transformation on green jobs emerge, there will also be job losses in employment, various definitions of green jobs are polluting industries. Estimations have shown that in use. Many statistical agencies use a narrow between 20 to 50 per cent of workers worldwide definition covering only sectors involving - envi are currently still working in carbon-intensive ronmental goods and services. Looking at EGSS industries, where the need to change business only, studies for the EU, the United States and models, jobs and occupations is particularly other countries have demonstrated the poten- acute. Thus, for a full appraisal of employment tial for the environmental sector in job creation. effects, detailed subsectors and general interac- Annual green job growth rates vary, but have been tion effects between sectors need to be consid- as high as 37 per cent in the case of Germany. The ered. EGSS figures only capture a small part of the share in total employment ranges between 1 to 4 labour market change. per cent. Moreover, employment considerations should not However, when studying overall employment only focus on the number of jobs created or lost effects of green transformations, it does not but also the quality of jobs. The ILO Decent Work suffice to look at the emerging green sector in Agenda–encompassing job creation, rights at isolation. Greening sectors, occupations and jobs work, social protection and social dialogue–can Chapter 4 Enhancing Job Creation Through Green Transformation

offer guidance on how the green economy should provide 'decent work' and be socially inclusive. To 63 emphasize this agenda, ILO’s definition of green jobs limits the concept to those jobs that can also be considered as 'decent work'.

Labour market and social policies, in particular active and passive labour market policies, can and should support successful green transformation by enhancing the mobility and skills of workers. 2 PART Compliance with International Labour Standards in all sectors of the green economy can ensure commitment to job quality. Finally, well designed revenue-neutral green tax policies can have positive effects for the environment and employment. Green Industrial Policy - Concept, Policies, Country Experiences

64 REFERENCES Altenburg, T, & Rodrik, D. (2017) Green industrial Goldemberg, J., Coelho, S. T., & Guardabassi, P. (2008). policy: Accelerating structural change towards The sustainability of ethanol production from wealthy green economies. In Altenburg, T., & sugarcane. Energy policy, 36(6), 2086–2097. Assmann, C. (Eds.). (2017). Green Industrial Policy. Goulder, L. H. (2013). Climate change policy’s inter- Concept, Policies, Country Experiences (pp. 1–20). actions with the tax system. Energy Economics, Geneva, Bonn: UN Environment; German Devel- 40, S3-S11. opment Institute (DIE). Herreras Martínez, S., van Eijck, J., da Cunha, M. P., Belova, A., Gray, W. B., Linn, J., & Morgenstern, R. D. Guilhoto, J. J., Walter, A., & Faaij, A. (2013). Anal- (2013). Environmental regulation and indus- ysis of socio-economic impacts of sustainable try employment: a reassessment. Center for sugarcane–ethanol production by means of Economic Studies. Washington: U.S. Census inter-regional Input–Output analysis: Demon- Bureau. strated for Northeast Brazil. Renewable and Borel-Saladin, J. M., & Turok, I. N. (2013). The impact Sustainable Energy Reviews, 28, 290–316. of the green economy on jobs in South Africa. International Labour Organization (ILO). (2011). South African Journal of Science, 109(9–10), 01-04. Towards a Green Economy: the Social Dimension. Bowen, A., & Kuralbayeva, K. (2015). Looking for green Geneva: ILO. jobs: the impact of green growth on employment. International Labour Organization (ILO). (2012). Inter- Grantham Research Institute Working Policy national standard Classification of occupations Report. London: London School of Economics and ISCO-08. Geneva: ILO. Political Science. International Labour Organization (ILO). (2013). Brander, M. (2012). Greenhouse Gases, CO2, CO2e, and Sustainable development, decent work and green Carbon: What Do All These Terms Mean? Econo- jobs. Fifth Item on the Agenda, International metrica, White Papers. Labour Conference 2013, Geneva: ILO. Bureau of Labor Statistics. (2012). Employment in International Labour Organization (ILO). (2017). Green Goods and Services. Decent Work. Retrieved from www.ilo.org/global/ Da Motta Veiga, P., & Polónia Rios, S. (2017). Ethanol topics/decent-work/lang--en/index.htm Policy in Brazil: a green industrial policy by acci- International Labour Organization (ILO), & United dent? In Altenburg, T., & Assmann, C. (Eds.). (2017). Nations Development Programme (UNDP) (2011). Green Industrial Policy. Concept, Policies, Coun- Green jobs assessment in Lebanon. Synthesis try Experiences (pp. 199–217). Geneva, Bonn: UN Report, Geneva: ILO. Environment; German Development Institute International Labour Organization (ILO), & United (DIE). Nations Environment Programme (UNEP) (2012). Ecorys. (2009). Study on the competitiveness of the Working Towards Sustainable Development, EU eco-industry. Opportunities for Decent Work and Social Inclu- Ecotec. (2001). Analysis of the EU Eco-Industries, sion in a Green Economy. Geneva: ILO. their Employment and Export Potential. Intergovernmental Panel on Climate Change. (2007). Ernst & Young Environment and Sustainability Climate Change 2007: Synthesis Report. Contri- Services. (2006). Eco-industries, its size, employ- bution of Working Groups I, II and III to the Fourth ment, perspectives, and barriers to growth in an Assessment Report of the Intergovernmental enlarged EU. Panel on Climate Change, Geneva. Erumban, A. A., Gouma, R., Los, B., Stehrer, R., Temur- International Renewable Energy Agency (IRENA). shoev, U., Timmer, M., & de Vries, G. (2011). World (2016). Renewable Energy and Jobs. Annual input-output database (WIOD): construction, chal- Review 2016. Abu Dhabi: International Renewable lenges and applications. Paper presented at the Energy Agency. DIME Final Conference (Vol. 6, p. 8). International Renewable Energy Agency (IRENA). European Environment Agency (EEA). (2011). Envi- (2017). Renewable Energy and Jobs. Annual ronmental tax reform in Europe: implications Review 2017. Abu Dhabi: International Renewable for income distribution. Technical Report. Energy Agency. Eurostat. (2009). The environmental goods and services sector (EGSS), a data collection handbook. Chapter 4 Enhancing Job Creation Through Green Transformation

Kato, M., Mittnik, S., Samaan, D., & Semmler, W. (2013). Society and University of Campinas, Brazil. Employment and Output Effects of Climate Poli- Timmer, M. P., Dietzenbacher, E., Los, B., Stehrer, R., & 65 cies. In Lucas, B., & Semmler, W. (Eds.). (2013). Vries, G. J. (2015). An illustrated user guide to the The Oxford Handbook of the Macroeconomics world input–output database: the case of global of Global Warming (pp. 445–476). Oxford: Oxford automotive production. Review of International University Press. Economics, 23(3), 575–605. Kuralbayeva, K. (2013). Effects of carbon taxes in an United Nations (2012). System of Environmen- economy with large informal sector and rural-ur- tal-Economic Accounting. Geneva: Central ban migration. Grantham Research Institute Framework.

Working Paper No. 139. London: London School of 2 PART United Nations. (2015). Transforming our World: The Economics and Political Science. 2030 Agenda for Sustainable Development. Landa Rivera, G., Reynès, F., Cortes, I. I., Bellocq, F. X., United Nations Environment Programme (UNEP). & Grazi, F. (2016). Towards a low carbon growth in (2008). Green jobs: Towards decent work in a Mexico: Is a double dividend possible? A dynamic sustainable, low-carbon world. UNEP/ILO/IOE/ general equilibrium assessment. Energy Policy, ITUC. 96, 314–327. United Nations Environment Programme (UNEP). Morgenstern, R. D., Pizer, W. A., & Shih, J. S. (2002). (2011). Towards a Green Economy: Pathways to Jobs versus the environment: an industry-level Sustainable Development and Poverty Eradica- perspective. Journal of Environmental Econom- tion. Geneva. ics and Management, 43(3), 412–436. United Nations Statistics Division. (2003). System of Organisation for Economic Cooperation and Devel- Environmental Economic Accounting. opment (OECD). (2014). Towards Green Growth? Tracking Progress. Paris: OECD Green Growth United Nations Statistics Division. (2017). ISIC Rev.2. Studies. International Standard Industrial Classification of All Economic Activities, Rev.2. Retrieved from Organisation for Economic Cooperation and Devel- https://unstats.un.org/unsd/cr/registry/regcst. opment (OECD), & Eurostat. (1999). The Environ- asp?Cl=8 mental Goods & Services Industry - Manual for Data Collection and Analysis. Paris. Vidican Auktor, G. (2017). Renewable energy as a trigger for industrial development in Morocco. In Pollin, R., Garrett-Peltier, H., Heintz, J., & Scharber, Altenburg, T., & Assmann, C. (Eds.). (2017). Green H. (2008). Green recovery: A program to create Industrial Policy. Concept, Policies, Country Expe- good jobs & start building a low-carbon economy. riences (pp. 153–165). Geneva, Bonn: UN Environ- Amherst: Political Economy Research Institute, ment; German Development Institute (DIE). University of Massachusetts. Joanne Wade, J., Wiltshire, V. & Scrase, I. (2000). Schlegelmilch, K., Eichel, H., & Pegels, A. (2017). Pric- National and Local Employment Impacts of ing environmental resources and pollutants and Energy Efficiency Investment Programmes. the competitiveness of national industries. In London: Association for the Conservation of Altenburg, T., & Assmann, C. (Eds.). (2017). Green Energy. Industrial Policy. Concept, Policies, Country Expe- riences (pp. 102–119). Geneva, Bonn: UN Environ- World Input-Output Database (WIOD) (2017). WIOD. ment; German Development Institute (DIE). Retrieved from www.wiod.org/home Siemens. (2016). Siemens to Build Rotor Blade Yamazaki, A. (2016). It’s not a job killing policy - The Factory for Wind Turbines in Morocco. Press case of BC’s revenue neutral carbon tax. Sustain- release, 10/03/2016. Retrieved from www.siemens. able Prosperity Research Note. Retrieved from com/press/pool/de/pressemitteilungen/2016/ www.sustainableprosperity.ca/content/its-not- windpower-renewables/PR2016030214WPEN.pdf job-killing-policy-case-bcs-revenue-neutral-carbon-tax Smeets, E., Junginger, M., Faaij, A., Walter, A.C., & Dolzan, P. (2006). Sustainability of Brazilian bio-ethanol. Universiteit Utrecht Copernicus Institute, Department of Science, Technology and Green Industrial Policy - Concept, Policies, Country Experiences

66 APPENDIX This Appendix illustrates the methodology under- The following countries are included in the analy- lying the figures and tables presented in the chap- sis of the chapter: Australia, Austria, Belgium, Brazil, ter. Data on GHG and employment by sector were Bulgaria, Canada, China, Cyprus, Czech Repub- taken from the World Input-Output Database lic, Denmark, Estonia, Finland, France, Germany, (WIOD), 2013 release. The WIOD has been part of Greece, Hungary, India, Indonesia, Ireland, Italy, a project funded by the European Commission Japan, Korea, Latvia, Lithuania, Luxembourg, Malta, and officially launched in 2012. The core of the Mexico, Netherlands, Poland, Portugal, Romania, WIOD has been constructed by linking national Russia, Slovak Republic, Slovenia, Spain, Sweden, supply and using tables with statistics in interna- Taiwan, United Kingdom and USA. tional trade (Erumban et al. 2011). It also includes

socio-economic accounts, employment and envi- GHG EMISSIONS IN TERMS OF CO2 EQUIVALENTS ronmental indicators on resource use and emis- Methane and nitrous oxide are converted into sions (WIOD 2017). their CO2 equivalents in order to have a measure of the total GHG emissions produced by a given CO2 emissions are calculated in the form of CO2 equivalents and therefore include also other GHG industry in a particular country. CO2 equivalents emissions. The WIOD contains information on are estimated through the Brander (2012) formula, other emissions and resource utilisation such as where: energy use, water consumption, or material use. CO equivalents= tons of the gas x global Unfortunately, only GHG emissions and energy 2 warming potential11 of the gas. consumption are systematically available for most industries and across countries. Energy CO emissions are summed up with methane and consumption and GHG emissions are highly 2 nitrous oxide in CO2 equivalents to get the total correlated and lead to a similar ranking. Water GHG emissions. consumption and material use are only available for a limited number of industries. While EMISSION INTENSITIES the available information on water consump-

tion and material use is in line with the ranking Following ILO (2011), direct CO2 intensities are based on GHG emissions, there are too many estimated as ratio of GHGs measured in kilotons missing data points to establish a ranking on over the total output in millions of US dollars. A

resource-intensity. sector is classified as high-carbon (HCS) if its CO2 intensity is above the median or as low-carbon To conduct the analysis, information coming (LCS) if its CO2 intensity is below the median. from the national input-output tables is merged with data on social and economic accounts for Six countries are included in the analysis: France, the period 1995 to 2011, along with environmental Germany, United Kingdom, United States, China indicators for the years 1995 to 2009. In total, data and Brazil. The reference year is 2009. Results is available for 35 industrial sectors based on the show that the majority of industries is always ISIC rev.3 classification in 40 countries (United high or low-carbon intensity, independently of the Nations Statistics Division 2017). The database country taken into account. At the same time, its also includes information on emerging and devel- relative ranking is quite stable across countries. oping economies for which detailed sectorial This result is consistent with ILO (2011) and it information is typically difficult to find. In particu- points out that the level of technological progress lar, 33 countries are developed economies, 5 are in a given industry is similar across countries. upper-middle and 2 are lower-middle income The top ten high and low-carbon sectors accord-

countries. In addition, 13 out of the 40 countries ing to their CO2 intensity levels and their relative are not OECD members, while 12 countries are not frequency in the cross-country comparison are EU members. presented (Table 4.1).

11 The Global Warming Potential (GWP) indicates the amount of global warming a gas causes over a given period of time,

relative to CO2 (Brander, 2012). The period of reference is usually 100 years. The higher the GWP the most dangerous the gas with respect to CO2. Data on the GWP come from IPCC (2007). Chapter 4 Enhancing Job Creation Through Green Transformation

The authors also analyze the change over time in ranking of few industries changes over time. emission intensities. France is the reference coun- However, changes are quite limited: none of the 67 try. Three years are taken into account: 1995, 2003 top polluter industries shifts into the low-carbon and 2009. The analysis shows that the relative category from one year to the other.

Table 4.2: Top ten HCS and LCS in the last available year.

HCS LCS Agriculture, hunting, forestry and fishing Electrical and optical equipment PART 2 PART Air transport Financial intermediation Basic and fabricated metals Health and social work Other community, social and personal services Machinery Mining and quarrying Post and telecommunications Other non-metallic minerals Real estate activities Electricity, gas and water supply Renting of M & Eq. and other business activities Sale, maintenance and repair of motor vehicles and Water transport motorcycles; retail sale of fuel Inland transport Transport equipment Wholesale trade and commission trade, except of motor Coke, refined petroleum and nuclear fuel vehicles and motorcycles

Source: ILO Research Department based on WIOD.

Note: Last available year: 2009.

EMPLOYMENT SHARES

Employment shares are estimated using the average of employed persons by sector and area in a particular year. The estimates are, then, divided by the total employment in a particular region to get the share. Green Industrial Policy - Concept, Policies, Country Experiences 68 PART 3: ACCELERATING CHANGE Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy CHAPTER 5 69 IN WITH THE GOOD, OUT WITH THE BAD: PHASING OUT POLLUTING SECTORS AS GREEN INDUSTRIAL POLICY

Aaron Cosbey, Peter Wooders, Richard Bridle, Liesbeth Casier Green Industrial Policy - Concept, Policies, Country Experiences

70 1. INTRODUCTION Industrial policy, broadly cast, can be defined The case studies in this volume, and most of the as, “a set of policies that selectively favours the broader discussion around green industrial policy, development of certain industries over others” are concerned either with measures taken to (Schwarzer 2013). Green industrial policy, more enhance the business environment in which new specifically, is defined as (Altenburg and Rodrik domestic industries might flourish, or with more 2017, this volume): targeted assistance for particular sectors. In both cases, the green industrial policy mechanisms "comprising any government measure aimed to employed are positive measures, encouraging the accelerate the structural transformation towards development of specific sectors through support, a low-carbon, resource-efficient economy in ways incentives and complementary enabling measures. that also enable productivity enhancements in the economy". This chapter deals with industrial policy of a different type. It explores the ways in which Cosbey (2013) describes green industrial policy as governments might support green industrial supporting the development of domestic indus- development by primarily negative measures– tries that produce green goods, or goods that: that is, by measures that seek to diminish the role of certain sectors within the economy, phasing ◼◼ have better environmental performance in them out and thereby fostering space for green operation than their competitors including sectors to operate and compete. Lütkenhorst et electric vehicles, renewable electricity-general. (2014) and Altenburg and Pegels (2012) discuss ating equipment, energy-efficient appliances this sort of green industrial policy strategy as ◼◼ directly address environmental problems such ‘pathway disruption’: the dismantling of key pollut- as environmental remediation technologies ing sectors of the economy to allow the growth ◼◼ are produced in a way that is environmentally of greener alternatives. These authors argue the preferable to their competitors as in the case of importance of disruptive green industrial policy: it organic agriculture. targets sectors that are environmentally harmful, By far the majority of green industrial policy over and thus need to be phased out. the last decades–in China, India, Germany and The harmful sectors most in need of active disrup- Denmark, for example–target the development tion are locked in, and have the financial power of new low-carbon energy technologies, such as and political clout to frustrate urgently needed solar photovoltaics, wind turbines and power stor- restructuring. As well, these sectors merit atten- age (Buen 2006; Lewis and Wiser 2007; Wang et al. tion because they tend to be deeply embedded in 2010; Zhang and He 2013; Lütkenhorst and Pegels the social and economic fabric—the oil and gas 2014; Ganesan et al. 2014; Vidican Auktor 2017, industry and associated transportation and power this volume). This is understandable, given their sectors are good examples. As such, their eventual significant recent growth in demand and - enor dissolution implies significant transition costs that mous near-term potential for further growth (IEA should be anticipated and appropriately managed. 2015a; IEA 2015b; BNEF 2016). But green industrial This is especially important to avoid particularly policy goes beyond renewable energy to cover a negative effects on already disadvantaged groups wide range of environmental sectors such as such as the women and youth in a society. transportation technologies, especially electric and low-emission vehicles and batteries; low-GHG The following section defines what is meant by emission waste management operations; drought- ‘disruptive green industrial policy’, describing and salt-resistant plant varieties; water-saving some of the tools that can be used to achieve it technologies; and technologies for heating, cooling and offering a spectrum on which various green and energy conservation in buildings. industrial policy efforts might be focused. Section 3 offers several case studies of green industrial policy, aiming to draw from them lessons that can be applied more broadly. Section 4 then uses those lessons to generate policy guidance that should be useful to decision-makers. Section 5 concludes. Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy

2. DEFINING DISRUPTIVE GREEN INDUSTRIAL POLICY 71 What we will call disruptive green industrial such as land, grants of dedicated infrastructure policy is defined as: measures to restrict or phase or support of research and development efforts. out environmentally undesirable sectors in such a Incentives might also be regulatory, as when way as to provide opportunity for greener sectors specific firms are granted exclusive marketing to prosper. Like the other forms of green industrial rights or concessions. Removing these sorts of policy discussed within this volume, disruptive incentives to targeted firms or sectors has the green industrial policy seeks to promote green same effect as imposing taxes, charges, levies or industrial development within the implement- fees: it renders them less viable relative to greener ing jurisdiction. It is properly understood as a competitors, and moves the economy toward a complement to the more familiar positive sort of greener mix of economic activity. Removing fossil green industrial policy insofar as it facilitates the fuel subsidies, for example, makes renewable desired growth in green sectors by making room energy substitutes more competitive and guides for them to flourish. the economy toward consumption of less polluting final goods through price signals. Removing There are many tools that might be used in the tax advantages for energy- and material-intensive practice of disruptive green industrial policy, to sectors amounts to restructuring the economy in help phase out targeted sectors: favour of greener, high-value added firms.

Environmental taxes, charges, levies, fees: Any Mandated phase out: The most powerful set of sort of financial burden placed on firms and tools force a mandated phase out of the targeted sectors will decrease their competitiveness rela- sector or firms. This is demonstrated by the tive to other actors in the economy. If the burden mandated phase out of coal-fired electricity is aimed at correcting negative environmental generation in Ontario, Canada. Where, as in the externalities, then the effect is to encourage econ- Ontario case, the targeted sector is the domain of omy-wide structural change toward a greener government agencies, the phase out involves a state, and to enhance the economic viability policy decision to end the operation of the sector’s of competing green firms and sectors. As such, operations. Where private interests operate the disruptive green industrial policy bears some sector, a phase out will involve regulatory and resemblance to fiscal reform, when it seeks to legal bans or restrictions on sales or operation, as increase taxes on those things that society wants in the phase out of lead as a paint additive in most less of and reduce taxes on those things that soci- OECD countries. ety wants more of, preferably in a revenue-neutral fashion. In the case of green industrial policy, There is a spectrum of types of disruptive green society wants less pollution and resource waste industrial policy, with the differentiating factor and more employment and income (Stoianoff et al. being the level of effort devoted to promoting 2016; Hanna et al. 2012; UK Green Fiscal Commis- the sectors that will benefit from the diminished sion 2009; Schlegelmilch et al. 2017, this volume). role of the phased out activities. As noted above, disruptive green industrial policy is characterized Elimination of incentives: Another set of tools by the intent to not only reduce undesired types of removes incentives granted to sectors targeted for activity, but to also promote more desired types of phase out. These might be financial incentives or domestic economic activity. Figure 5.1 illustrates subsidies such as cash grants, low-interest loans, this spectrum. loan guarantees, limits on liability, price supports, purchase mandates, various types of tax breaks. They can also include free or low-cost inputs Green Industrial Policy - Concept, Policies, Country Experiences

Figure 5.1: Spectrum of Types of Disruptive Green Industrial Policy 72

Charges, or removing incentives, with revenue recycled to green sectors. Or bans/regulations with complementary policies to Pollution charges, taxes, aimed foster greener substitutes. But at greening a targeted sector, but no discrimination against foreign not at creating a new substitute producers. (See the Morocco case sector. study below.)

soft disruptive GIP hard disruptive GIP

Charges, removing incentives, Charges, or removing incentives, or bans/regulations to restrict with revenue recycled to green targeted sector, without comple- sectors. Or bans/regulations with mentary policies to foster the complementary policies to foster emerging substitutes. (See the greener substitutes. Discrimina- Jordan case study below.) tion aimed at fostering domestic capacity. (See the Ontario case study below.)

At the left end of the spectrum, representing soft also be specific: for example, renewable energy policy, exist policy options that diminish the role generators and the related input manufactur- of undesirable activities with the explicit under- ers and services. But these beneficiaries may standing that simply by doing so some green be domestic, or they may be foreign exporters sectors will be enabled. An example of this type of green goods. At the right end of the spectrum, of policy is a carbon tax, which by itself would corresponding to what we have called hard benefit a broad variety of green firms and sectors, disruptive green industrial policy, are policies but which attempts to neither identify particular that combine deliberate phasing out of target targets for phase out, nor chart the course of the sectors with deliberate cultivation of substitute transition in the direction of particular domestic domestic sectors. An example of this is the phase champions. Further across the spectrum would be out of coal-fired electricity generation in Ontario, the reduction or elimination of fossil fuel subsi- Canada and the concurrent promotion of domes- dies. These tend to be targeted at specific firms tic renewable energy sectors. and sectors, and the resulting beneficiaries will

3. EXAMPLES OF DISRUPTIVE GREEN INDUSTRIAL POLICY This section relates four examples of disruptive Canadian province of Ontario to phase out coal green industrial policy, aiming to both illustrate and promote renewables (Box 5.4). the types of approaches described above, and to draw lessons from specific cases that might 3.1. FOSSIL FUEL SUBSIDY REFORM be useful guidance in a more general context. It covers fossil fuel subsidy reform, including brief Fossil fuel subsidy reform has been on the inter- case studies of subsidy reforms in Jordan (Box 5.1) national agenda since the 2009 G20, when those and Morocco (Box 5.2), China’s reform of value countries committed to phasing out inefficient added tax refunds to polluting and resource-in- fossil fuel subsidies (G20 2009). A number of tensive sectors (Box 5.3), and policies in the similar declarations have followed, including Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy

Sustainable Development Goal 12, Target C, where While the precise figure is difficult to determine– the signatories committed to “rationalise ineffi- given the different ways of defining subsidies– 73 cient fossil-fuel subsidies that encourage wasteful there is no disagreement about the resulting consumption” (UN 2015). environmental impacts. Consumer subsidies artificially lower the price of fossil fuels, encourag- Fossil fuel subsidies can be broadly divided into ing over-consumption. This lowers incentives for two types: those delivered to consumers through conservation or for efficiency investments, as well artificially low retail prices and those delivered as reducing the viability of investment in substi- to producers of fossil fuels as grants, low-in- tutes such as renewable energy sources. That terest loans, preferential tax treatment or other viability is already impaired by the fact that fossil such measures. fuel power sector investments are locked in for many decades. Producer subsidies lower produc- While the World Trade Organization’s Agreement tion costs, enhance profits, and invite further on Subsidies and Countervailing Measures has investment in fossil fuel exploration and extrac- a legal definition of subsidies in general, more tion, as well as for the research and development useful here is the definition of fossil fuel subsi- that produces technologies that further lock in the dies offered by Kojima and Koplow (2015) “meas- fossil fuel power sector. ures that target fossil fuels, or fossil fuel-based electricity or heat, and cause one or more of the Eliminating fossil fuel subsidies would have a following effects: a reduction of net energy costs, palpable effect on global GHG emissions. Terton a reduction of energy production or distribution et al. (2015) note that 13 countries included fossil costs, and an increase in the revenues of energy fuel subsidy reform in their Paris Agreement suppliers”. commitment to Intended Nationally Determined Contributions. Burniaux and Chateau (2014) esti- There are a number of estimates for the magni- mate that eliminating fossil fuel subsidies would tude of global fossil fuel subsidies, ranging from cause an 8 per cent reduction in GHG emissions US$ 250 million to US$ 5 trillion, depending on by 2050. Schwanitz et al. (2014) estimate a 6.4 per the definition used (Kojima and Koplow 2015). The cent decrease by 2050. Merrill et al. (2015) model high end of those estimates comes from the Inter- subsidy removal in 20 countries and find an national Monetary Fund (2013a) based on a defini- average GHG emissions reduction of 11 per cent tion that includes external costs of fossil fuel use by 2020. It is worth noting that all these figures such as the social cost of carbon and health costs likely understate the true extent of effects, since of air pollution. they focus only on a demand response to price increases and do not consider the effects of producer subsidies or investment distortions.

Box 5.1: Jordan’s fossil fuel subsidy reform Jordan imports almost all of its natural gas and oil. In 2005, after several years of spiking international prices and controlled domestic prices, the annual cost of sheltering domestic consumers, power generators and industry had reached an unsustainable 5.8 per cent of GDP, or US$ 711 million (Bridle et al. 2014). Over an implementation period of seven years, the country gradually but completely eliminated its subsidy programmes, eventually allowing domestic prices to be objectively set monthly by a formula that reflects international prices.

This was the third time in 30 years the government had tried to reform its energy-pricing regime, with the first two ending unsuccessfully after massive public protests. Several elements helped ensure that this time the effort would succeed (Fattouh and El-Katiri 2012):

◼◼ The reform was announced well in advance and implemented over a long time frame. ◼◼ The government took great pains to educate Jordanians about the need for change and the unsustainability of the status quo. ◼◼ Public sector wages and pensions were increased, with higher increases for low-income earners, and private sector low-wage earners were separately compensated. Almost 60 per cent of the population benefited under these schemes. The poor received further food subsidies and social safety net programmes were enhanced. Green Industrial Policy - Concept, Policies, Country Experiences

74 ◼◼ Fossil fuel subsidy reform was pursued as part of a broader energy sector reform, wherein the electricity sector was rationalized, unbundled and made to charge end users the true costs of generation and distribution. Measures to mitigate the consequences of reform were only taken after it became obvious that they were creating substantial difficulties for households and industry, particularly for low-income households (Fattouh and El-Katiri 2012). Despite the government’s efforts to communicate the need for reform, the initial effects sparked massive protests.

Jordan did not attempt to phase in greener alternatives in parallel with its reform efforts. This meant that while consumption dropped in response to price increases—a significant environmental outcome—there was no growth in new green sectors, much less domestic green sectors.

While both CO2 emissions and emissions per unit of GDP dropped from 2005 to 2009, both have since steadily increased (Merrill et al. 2015).

Box 5.2: Fossil fuel subsidy reform in Morocco Morocco imports over 90 per cent of its primary energy sources (IEA 2014a). In 2000, the country suspended international indexing of consumer prices of gasoline, kerosene, diesel and liquid propane gas (LPG), in an effort to shelter consumers from climbing world prices. Gradual domestic price increases kept the fiscal cost of these subsidies low until 2009, but international price spikes thereafter led to an unsustainable subsidy expenditure of 6.6 per cent of GDP by 2012 (Merrill et al. 2015)

The government began the process of subsidy reform in 2013, but retained subsidies that it deemed to be sheltering the poor, such as those for electrical generating fuel and LPG. In 2014, it removed all fuel subsidies but those for LPG, though diesel subsidy reform is on a slow timetable (IEA 2014a). Electricity prices are still regulated to be lower than production costs.

The government learned from previous unsuccessful reform efforts, undertaking a massive public information campaign and increasing support for the poor, who were proportionately much more threatened by the price increases. A conditional cash transfer programme targeting poor rural households expanded from 80,000 families in 2009 to 466,000 families in 2014. A health insurance scheme for the poor increased its coverage from 5.1 million beneficiaries in mid-2013 to 8.4 million beneficiaries in early 2015 (Merrill et al. 2015).

At the same time, the 2009 National Energy Strategy included the objective of accelerating development of renewable energy, targeting a 42 per cent share of total mix for renewables by 2020, or 2 GW each of wind, solar and hydro capacity. Since then, the government has invested in research and development for renewables and attracted major investments, including the world’s largest concentrated solar power plant, the first phase of which came on line in 2016 (IRENA 2016b), and roughly 750 MW of new wind capacity (IRENA 2016a). In 2016, a consortium was awarded a concession to build 850 MW of wind power capacity over 5 projects. As part of the deal, Siemens—one of the consortium partners—will invest US$ 110 million in a domestic manufacturing facility for turbine blades, intended for both domestic use and export to Africa and Europe (Yaneva 2015). Morocco’s efforts to strengthen its renewable energy potential are further discussed in a dedicated case study in this book (Vidican Auktor 2017, this volume). Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy

These environmental outcomes—and in particu- investments in renewable energy support were lar the consequences for the competing renewa- made possible by the savings from fossil fuel 75 ble energy sector—are the basis of the argument subsidy reductions. that fossil fuel subsidy reform can be thought of as disruptive green industrial policy. Fossil Furthest to the right on the spectrum would be fuel subsidies crowd out investment in renewa- cases where fossil fuel subsidies were removed, ble energy in several ways, primarily by making renewable energy was promoted, and domes- its competitors cheaper (Bridle and Kitson 2014; tic firms in the renewable energy sector were Bazilian and Rentschler 2016). IEA (2014b) esti- fostered. That is, while fostering renewable energy mates that in the Middle East oil subsidies reduce investment may promote domestic capacity, that the costs of generating electricity from new result would be only one of many, and would by oil-fired plants to 30 per cent of what they other- no means be certain. Other possible results, for wise would be, and gas subsidies reduce those example, might be simply increased imports of costs to 45 per cent of the unsubsidised levels. renewable energy technologies. In the case of Elsobki et al. (2009) argue that while Egypt has Morocco, fostering domestic capacity was explic- excellent wind regimes, investment in wind power itly part of the strategy (Vidican Auktor 2017, this there is not viable because of subsidies granted volume). The tools used included spending on to gas-fired electricity producers. While many domestic research and development and the countries undertake fossil fuel subsidy reform requirement that access to lucrative wind power primarily in pursuit of fiscal savings, there can concessions was conditional on commitments to be a robust mix of motivations that may include investment in domestic manufacturing. the promotion of renewable energy, as in Morocco. It is noteworthy that both Morocco and Jordan The International Monetary Fund (2013b) notes had tried previously to reform their fossil fuel that promoting renewable energy was one of the subsidy regimes, only to have those efforts key objectives behind the reform of fossil fuel scuttled by widespread public opposition. In subsidies in Kenya. both cases this was ultimately overcome by a The two cases of Jordan and Morocco make two-pronged approach that included extensive it clear that fossil fuel subsidy reform can be public outreach and a redirection of the saved situated at various places along the spectrum funds into social programmes and transfers described in Figure 5.1, depending on the particu- designed to soften the price shock in particular lars of the effort. Jordan’s reform efforts evened for low-income consumers. the playing field for renewable energy investment in the country, but they are situated toward the soft 3.2. CHINA’S GREEN VALUE ADDED end of the spectrum, since there was no accom- TAX REFORM panying drive to foster those greener alternatives. Fossil fuel subsidy reform in and of itself, absent China has made great efforts to phase out ener- complementary policies aimed at simultaneously gy-intensive, resource-consuming and pollut- fostering domestic renewable energy manufacturing sectors by means of fiscal policy as part of a ing capacity, does not guarantee the emergence of broader push to restructure along green economy green substitutes. In the case of Jordan, the lack of lines. subsequent renewable energy investment can be China’s desire for economic restructuring serves interpreted as testimony to the multiple obstacles many objectives, including environmental but that such investment faces beyond price compe- also including economic. The eleventh five-year tition with conventional fuels, such as legal and plan covering 2006 to 2010–described by some as regulatory barriers, and financing. revolutionary and a turning point (Fan 2006)–had By contrast, Morocco engaged in dedicated efforts an admirable aim as one of its key elements. The to attract this sort of investment in parallel with policy planned to promote adjustments in indus- its reform efforts, with resulting growth in its trial and economic structure and transforming renewable energy sector. This situates the Moroc- economic growth from being driven by quanti- can case to the right of Jordan’s case on the spec- tative expansion of industry to being driven by trum. In the case of Morocco, the linkage between balanced development of the primary, secondary phase out and promotion was part of the broader and tertiary industries and by structural optimiza- vision, as enunciated in the 2009 National Energy tion and upgrading (Kai 2006). Strategy. But the linkage was also financial: the Green Industrial Policy - Concept, Policies, Country Experiences

76 Box 5.3: China’s drive for a greener economy China’s first national climate change policy, China’s National Climate Change Programme (Government of China 2007), set out as one of its objectives:

"to deepen institutional reform of foreign trade in controlling export of energy-intensive, pollution-intensive and resource-intensive products, so as to formulate an import and export structure favourable to promote a cleaner and optimal energy mix".

In a subsequent policy paper outlining the steps taken to implement the Programme, in a section entitled Adjusting the Economic Structure to Promote the Optimization and Upgrade of the Indus- trial Structure, the government (Government of China 2008) described its efforts as:

"Limiting the excessively rapid expansion of high energy intensive and emission intensive industries… By adjusting tax rebates for exports and customs duties, the government is working to restrain the export of high energy-intensive, pollution-intensive and resource-intensive products".

Set in the context of a broad programme of policies to address climate change, this effort was explicitly aimed at restructuring the national economy by discouraging polluting firms and sectors. It was accompanied by other efforts aimed at reducing inefficient production capacity that—in 2007 alone—led to the forced closure of 14.4 GW of small thermal electrical generating capacity, 46.6 million tons of iron-smelting capacity, 37.5 million tons of steelmaking capacity, 52 million tons of cement production capacity, as well as over 2000 small coal mines and heavily polluting papermaking plants, chemical plants, and printing and dyeing mills (Government of China 2008). These measures were accompanied by, among other things, a target to raise the proportion of renewable energy in primary consumption to 10 per cent by 2010 (Government of China 2008).

In 2010 the government cancelled export value added tax (VAT) refunds for a list of 406 energy-intensive products, effectively raising the cost of production and putting those sectors at a significant disadvantage relative to other export sectors (Wang et al. 2010). Up until that point a portion of the VAT paid domestically had been refundable at the point of export, a policy commonly used to promote exports.

In 2010 China cancelled value added tax refunds manufacturing capacity in solar and wind power to a number of energy-intensive products, as (Wang 2010, Zhang and He 2013). In the context of part of the broader drive for economic restruc- that sort of support, any disruption of polluting turing along more sustainable lines. While this sectors could be expected to have salutary effects policy was not accompanied by specific support on green domestic capacity, even absent explicitly for greener domestic economic capacity, the stated objectives to that effect. intent was clearly to foster growth in lower-impact sectors such as the services sector and, 3.3. ONTARIO’S COAL PHASE OUT coming as it did in the context of China’s climate change policy, in the various renewable energy At the hard end of the spectrum of green indus- sectors. As such it sits toward the hard end of the trial policy approaches is the phase out of unde- spectrum of green industrial policies shown in sirable sectors accompanied by explicit efforts to Figure 5.1, but not very far in that direction. support domestic capacity in the sectors’ greener substitutes. The case of a phase out of coal-fired This case, however, shows how problematic electricity generation in Ontario, Canada demon- such distinctions can be. It is true that there strates a policy that was inextricably linked with were no policies linked to the value added tax efforts to establish a renewable-energy manufac- reform that tried to promote substitute green turing sector in the province. sectors. At the same time, however, this policy came in the context of a myriad of government policies aiming to shape the Chinese economy along greener lines, including powerful support at the federal and local levels for domestic Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy

Box 5.4: Ontario’s coal phase out and renewables support 77 The independent power regulator in Ontario, Canada’s most industrialised province, described it as “the single largest greenhouse gas reduction measure in North America” (OPA 2007); the phase out of coal-fired electricity generation in that province saw the percentage of coal in the energy mix drop from 25 per cent in 2003 to zero in 2014.

The phase out had multiple objectives including protecting public health, reducing pollution and addressing climate change. Ultimately, though, a key element of the programme’s political feasibility was the accompanying suite of green industrial policy measures that held the promise of jobs from green industrial development in the province, measures designed to replace coal-fired power with renewable power generated from components manufactured in province (Harris et al. 2015). Those measures included:

◼◼ a generous feed-in tariff programme for renewable power (up to US$ 0.80/kWh) with 20-year contracts, conditional on high use of Ontario-made components of between 50 and 60 per cent after 2012 ◼◼ streamlining of the environmental approval processes for renewable energy projects ◼◼ net metering provisions to accelerate residential uptake of renewable energy generation ◼◼ the 2010 Green Energy Investment Agreement (Government of Ontario 2013), a partnership agreement between Ontario and a consortium led by Samsung, under which the consortium committed to investing US$ 7 billion in Ontario’s solar and wind manufacturing sector, and to the development, construction and operation of 2,500 MW of solar and wind generation projects in the province, the power from which would be purchased by Ontario at guaranteed preferential prices ◼◼ the Long-Term Energy Plan (Ontario Ministry of Energy 2010) that set a target of 10,700 MW of wind, solar and bioenergy capacity to be added by 2021.

The coal phase out was accompanied by a suite In terms of dissemination of renewable energy of green industrial policy measures designed generating capacity, the programme has been to promote renewable energy capacity in the successful and will easily exceed the targets depressed industrial heartland of Ontario. This announced at the programme’s outset. Signif- was to have the effect of both replacing the lost icant changes have come about in the generat- coal-fired generating capacity–25 per cent of the ing mix since the programme’s inception (Table generating mix in 2003–and developing a renew- 5.1). The government’s 2007 Long Term Energy able energy sector. Foremost among the green Plan set the goal of doubling renewable gener- industrial policy measures was a feed-in tariff ating capacity to 15,700 MW by 2025 (OPA 2007). regime for all types of renewables, with rates in By May 2016, the province already had grid-con- the initial scheme ranging from US$ 0.10/kWh nected renewable capacity of 13,030 MW, with for large-scale landfill gas to US$ 0.80/kWh for another 850 MW expected to be added over the residential-scale solar photovoltaic (OPA 2009). subsequent 18 months (IESO 2016). While many rates were later reduced, the latter is one of the highest premiums ever offered globally for any type of renewable energy.

Was the Ontario exercise in phase out green industrial policy successful? Given the multiple objectives, success would need to be assessed on a number of different levels. Green Industrial Policy - Concept, Policies, Country Experiences

Table 5.1: Ontario’s Electrical Generating Capacity (comparison 2010–2016) 78 Total Installed Capacity (MW)

Aug 2010 May 2016 % change

Nuclear 11,446 12,978 13%

Hydroelectric 7,903 8,432 7%

Coal 6,434 0 -100%

Oil / Gas 8,792 9,942 13%

Wind 1,084 3,823 253%

Biomass / Landfill Gas 122 495 306%

Solar 0 280 -

Total 35,781 35,950 0.5%

Source: IESO 2010, IESO 2016.

These figures understate the growing share of In terms of jobs created, the Association of Power renewables in Ontario since they do not include Producers of Ontario (APPrO 2015) estimates that embedded capacity–capacity to produce elec- between 2008 and 2014 wind power in Ontario tricity that does not go through the grid, such as created 88,842 jobs, if indirect effects are counted. rooftop solar panels and on-site wind turbines. The corresponding figure for solar power was In August 2010 embedded capacity in Ontario 91,972 jobs and the biofuels figure was estimated amounted to 700 MW (IESO 2010). In May 2016 at 21,378. These figures need to be viewed with there was 2,987 MW of renewable embedded caution, both because of the uncertain nature of capacity–a 427 per cent increase (IESO 2016). the multiplier used to derive indirect impacts and Adding embedded capacity in the figures, the because they are not net figures. They tell us how total share of renewables in the May 2016 mix many jobs were created in the renewable energy rises from 14.7 per cent to 24.2 per cent. However, sector, but not how many were lost in other renewables were not able to take up the entire sectors due to the direct effects of the phase out burden of replacing coal’s share of the generating and the indirect effects of any higher electricity mix; installed capacity of gas and nuclear have prices directly attributable to renewables. APPrO risen by much more than the percentage rise in (2015) argues this latter effect is small. total capacity. In terms of the programme’s success at fostering In terms of price impacts, a full analysis is beyond domestic capacity in the renewables sector–a key the scope of this assessment. But the consumer question–the available evidence is entirely anec- price of electricity did increase by an average dotal. The March 2016 large-scale auction awarded of 70 per cent between 2010 and 2016 (Ontario projects to a number of Canadian firms including Energy Board 2016) and several commenta- those with headquarters in Ontario, partnerships tors found that the over-generous feed-in tariff with Ontario firms, and Ontario-based subsidiaries contracts were a significant factor in that increase of multinationals (Blackwell 2016). This outcome (Auditor-General of Ontario 2015; Ontario Soci- is despite the fact that local content requirements ety of Professional Engineers 2016). Large-scale were dropped from the programme in 2013 as a renewable procurement in Ontario has now gone result of a challenge under World Trade Organi- to the reverse auction system successfully used zation rules. The Samsung consortium built the by other jurisdictions, whereby concessions for contracted four manufacturing facilities in Ontario, defined blocks of power are awarded to the lowest in partnership with other multinationals–such as qualified bidder. In a round of bids completed in CS Wind, Siemens, SMA Solar–and some Canadian March 2016, costs had fallen dramatically from firms. Canadian Solar, founded in Canada in 2001, what was initially paid out under the feed-in tariff was the world’s second largest solar company by regime: the average successful bid price for wind MW shipped in 2015, operating in 18 countries with power projects was US$ 0.066/kWh, and the equiv- over 8,000 employees (Canadian Solar 2016). In alent figure for solar was US$ 0.121/kWh (Clean- 2014, Northland Power, an Ontario company that Tech Canada 2016). started in forestry, successfully completed the biggest non-hydro renewable energy financing Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy

deal in history—a US$ 5.8 billion agreement to and further research is needed to assess the build a 600 MW offshore wind farm in the Neth- success of Ontario’s drive to build domestic capac- 79 erlands in collaboration with Siemens AG and two ity in renewable energy technologies. Dutch firms (McCarthy 2014). This sort of anecdotal evidence is indicative, but it is not conclusive, 4. OPTIONS FOR POLICYMAKERS In undertaking green industrial policy, policymak- transition has the desired results. In doing so, ers have a wealth of experiences for guidance on particularly if the measures employed are closer what works and what does not. As discussed in to the hard end of the spectrum, policymakers this volume, the experience of traditional indus- should respect good practice for green industrial trial policy holds many helpful lessons, even policy (Altenburg and Rodrik 2017, this volume). though green industrial policy is distinct in some respects, facing its own unique challenges If there is not an obvious successor sector, the and advantages (Altenburg and Rodrik 2017, this desire may be simply to remove undesirable activ- volume). Disruptive green industrial policy is simi- ities as part of a move to green the economy, in larly able to draw on decades of rich experience. the expectation that whatever activity flourishes But, as noted above, there are a few elements of in the newly opened economic space will likely this type of industrial policy that set it apart. In be less environmentally harmful than the phased this section we explore the questions that arise out activity. This strategy has the disadvantage about disruptive green industrial policy as distinct of potential unanticipated results, such as the from other forms of green industrial policy: rush to coal that followed Germany’s phase out of nuclear power (Pahle 2010). But a more fundamen- ◼◼ To what extent is it necessary to actively tal problem is public support: removing a sector phase in the good while phasing out the bad, that provides employment and contributes to by pursuing a hard disruptive green industrial conventional GDP is unpopular, no matter what policy approach? the environmental benefits. There is some polit- ◼◼ How do states choose those sectors they want ical value in being able to point to the promise of to phase out? This is akin to the traditional replacement green jobs, as in the case of Ontario. and difficult industrial policy challenge of picking winners. Finally, while the phase out of a competing ◼◼ How should states go about the potentially sector may offer powerful price signals for new painful process of phase out? green entrepreneurs, price signals alone are not likely to bring about the kind of fundamental and timely technological development, innovation and 4.1. MUST WE ALSO PHASE IN commercialisation needed to address the press- THE GOOD? ing environmental issues ahead (Hallegatte et al. 2013). Bazilian and Rentschler (2016) point to Can a phase out or restriction alone, without a number of non-price barriers, such as informa- an accompanying effort to foster replacement tion, capacity and financial constraints, that may sectors, constitute green industrial policy? If thwart results without complementary measures there is an obvious substitute sector—for example, such as infrastructure investments, training and renewables for fossil fuels, electric vehicles for capacity building, and support for energy effi- internal combustion engine vehicles–then a lack ciency and innovation. Lütkenhorst et al. (2014:15) of proactive measures to develop the substitute argue that: domestically may simply mean that the goods will be imported. There is nothing inherently "The introduction of environmentally sustaina- wrong with such a scenario; indeed, the prerequi- ble technologies at the requisite scale and speed sites for successful industrial policy are demand- involves the breaking up of entrenched devel- ing, and smaller and developing economies in opment paths…This massive challenge of soci- particular must consider whether it makes sense etal adjustment coupled with the exceeding to strive for domestic capacity in sectors where urgency of action thus requires more than simply the conditions may not be favourable (Altenburg getting the prices right…[T]he key task is to create 2011). The point is, though, once a country has incentives that push forward the development, made the difficult decision to engage in disruptive testing, deployment and upscaling of sustaina- green industrial policy, it should probably engage ble technologies. This in turn presupposes the in positive measures to ensure that the resulting Green Industrial Policy - Concept, Policies, Country Experiences

combination of smartly designed subsidies…with ◼◼ The status quo should see the sector for phase 80 incentives to encourage and steer R&D efforts in out as receiving significant fiscal outlays, the desired direction as well as investments into whether in terms of outright transfers or in dedicated infrastructure and multi-stakeholder terms of tax breaks. This would mean lower partnerships". opportunity cost to the phase out or, put another way, reduced fiscal outlays. As noted Ultimately, while disruptive policy is critically above, fossil fuel extraction and consumption important in freeing the space for desirable score well on this criterion. forms of economic activity, it is not sufficient to ◼◼ Candidates for phase out should have viable ensure that those reforms maximise their poten- substitutes that are significantly environmen- tial to actually foster those activities, particularly tally superior. given the urgency of the need for change. For ◼◼ Ideally, the substitute sector would be a bene- that, complementary policies are needed to help ficial economic driver: once the phase out of phase in the sectors that will succeed those being the targeted sector and the phase in of the phased out (Never and Kemp 2017, this volume). favoured substitute are complete, the economy will realign with stated national development 4.2. CHOOSING THE SECTORS TO PHASE OUT objectives. That is, the exercise of restructur- AND TO PHASE IN ing should be well grounded in the broader context of the development objectives of the One of the traditional criticisms of industrial implementing state, helping to move toward policy–and in particular hard policies–is that some desired end. The new sector might governments are not well placed to pick winners involve a higher degree of added value, for and losers (Pack and Saggi 2006). The losers in example, in a country that aims to move away this sense are those sectors that do not receive from extraction or production of commodities, support. Choosing who deserves support is indeed or might provide good prospects for skilled a difficult decision, ideally informed by - knowl employment and spin-off benefits. edge of latent comparative advantage–knowledge that may not even be discovered yet by the firms Even after policymakers choose the sectors to themselves (Rodrik 2004). target for phase out, as well as the sectors for phase in, they still need to decide to actively The choice of sectors to target for phase out, via pursue domestic capacity in the substitute disruptive green industrial policy, is much more sector. The alternative is to foster growth in that straightforward. While the choice in each coun- sector without discriminating between foreign try will depend critically on the context—resource and domestic proponents. This is a fundamental base, sectoral mix, sectoral environmental perfor- industrial policy choice, and it demands a great mance, industrial capabilities, fiscal and macroe- deal of information, vision and capacity—which conomic context, socio-economic profile, and so are not always available to the decision-makers. on—policymakers in each case can nonetheless be guided by a few common principles: Where policymakers do choose to promote domestic capacity in the substitute sector, then ◼◼ The sector for phase out should generate additional guidance is clearly in order beyond significant environmental damage, with poor what is offered above—that is, guidance on how opportunities for improvement. The first point to phase in a sector, involving both the choice of is obvious, but the second is critical. If there sector and the manner of implementation. There are viable prospects for reducing the envi- is an extensive literature on this topic (Grossman ronmental damage caused by a sector, then 1990; Rodrik 2004; Pack and Saggi 2006; Harrison producer states might choose to invest in and Rodriguez-Clare 2010; Altenburg 2011; Halle- developing that potential, avoiding the costs gatte et al. 2013; UNCTAD 2016). Never and Kemp inherent in any radical transformation and (2017, this volume) provide lessons for the phase perhaps benefiting from first mover status in in of green technologies. the new innovative and relatively green space. This decision may come down to a compari- In the course of choosing sectors to benefit, green son of the environmental consequences of the industrial policy faces a challenge that may potential substitute sector relative to those be less relevant to traditional industrial policy: that could realistically be achieved by green- given the objective to achieve environmental ing the polluting sector at similar cost. The improvement, it becomes necessary to add an classic example of this dynamic is clean coal environmental filter to the criteria used in picking versus renewables. winners. It could be argued that a proper conduct Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy

of industrial policy would already include such These are not challenges typically faced in the a criterion, but in any case it certainly needs to exercise of traditional industrial policy that seeks 81 be part of the choice of sectors to promote in to build up new sectors but not to directly disrupt the context of green industrial policy. All forms existing ones. But overcoming them will be para- of economic activity have some environmental mount to successful disruptive green industrial effects, so while substitute sectors may leave us policy. Public support for disruption is crucial, better off for having replaced target sectors, they especially if the phase out involves conflicts with themselves will also have environmental impacts, powerful vested interests. There are a number of and these need to be understood as a prerequi- key elements to achieving necessary buy-in: site to commitments to sectoral promotion. The classic example of this dynamic is the push to ◼◼ Public education about the nature and magni- promote bio-fuels as substitutes for gasoline and tude of the problem and the benefits of the diesel in transport; there is a growing consensus planned reform: that the environmental impacts of this particular People need to be convinced that the status quo is substitute sector can be significant (Delucchi 2006; unsustainable, undesirable, and that the plan will Runge and Senauer 2007; Searchinger et al. 2011). work. This was a key element in the fossil fuel subsidy reform processes of Jordan and Morocco, 4.3. HOW TO PHASE OUT? as well as many others (IMF 2013a; IMF 2013b; Vagliasindi 2012). Deliberately phasing out a sector is fraught with difficulties, particularly if it is entrenched and ◼◼ Compensation for vulnerable affected popula- provides a significant contribution to the national tions including under-represented groups in or regional economy. The case studies of Jordan society such as women and youth: and Morocco demonstrate that reform is difficult– The International Monetary Fund (2013b) in both cases it had been unsuccessfully tried describes the success of Indonesia at fossil fuel before–but possible. China has a command econ- subsidy reform in 2005—after multiple unsuccess- omy that makes it easier to phase out targeted ful attempts dating back to 1997—as due in large sectors, but even there, compliance at the local part to well-communicated social spending with level is a challenge. Ontario was fortunate in the resulting savings, including on unconditional that it has no domestic coal production, but the cash transfers and health insurance for the poor, restructuring was nonetheless a major effort, is and assistance in fuel switching away from subsi- still underway, and produces a case study in the dised kerosene for cooking. A number of coun- political science of green economic transition tries have also used a direct cash transfer system (Harris et al. 2015). The programme faced signif- to ensure that the most vulnerable would not be icant opposition on the grounds that it would negatively affected (IMF 2013b). When the primary increase electricity prices and stymie economic consequence is job loss, the response might be growth, and that it constituted inefficient envi- training, relocation assistance and a strong social ronmental policy (McKitrick 2013). The prom- safety net. The precise form of the compensation ise of new jobs from the promoted renewable scheme will vary with national circumstances, energy sectors was critical in garnering public but the principle of ‘low net harm’ is important, acceptance. and central to a just transition. Whether the shift takes the form of removing subsidies or levying The process of adjustment in the face of disrup- charges, a portion of the resulting saving/revenue tion, however urgent and necessary, should can be used to support this sort of compensation. respect the principles of the International Labour Organization’s principle of Just Transition (ILO ◼◼ The promise of prosperity: 2010). According to this principle, the need to respond to climate change is paramount and If hard green industrial policy is being pursued, urgent; however, the process of transition that as in Ontario, governments can hold out the pros- results needs to be planned with full under- pect of short-term pain from adjustment balanced standing of, and accommodation for, the massive off against long-term gain, typically in the form consequences for industries, jobs and workers. of jobs, from the advent of new green economic This is not only a moral stance, it is also prag- activity. The spending to support such measures matic: Without the consensus that derives from a achieves both the goal of cushioning adjust- just transition, necessary reforms will not be polit- ment and the goal of promoting green substitute ically viable. sectors. Green Industrial Policy - Concept, Policies, Country Experiences

Support from the affected business community ◽◽ Gradual reform: 82 may also be vital to the success of the programme. The more gradual the disruptive process, the It is important to identify who will lose in the less likelihood there is of stranded assets reform process, and to plan ways to minimise the among the affected sectors, reducing the oppo- loss as much as possible, consistent with achiev- sition. A balance is needed here, though, such ing the objectives of the programme. that the speed of the process is appropriate to the urgency of the environmental problems ◼◼ Impacts on the business community can be addressed. softened by: ◽◽ Consultation with the affected sectors well in ◽◽ Support for affected non-targeted firms: advance of the policies, to garner input relevant to their final shaping: Those firms from outside the target sector that are threatened might be helped by measures to This process may reveal ways to revise draft boost competitiveness, including via measures policies that reduce damage to increase energy efficiency and reduce waste. This support can be funded by a portion of the ◽◽ Transparent and timely announcement of savings or revenue generated by the phase policies and timelines, giving some measure out policies, whether they come in the form of of stability and predictability: subsidy removal or charges. This allows for investments to be made with some certainty about future conditions.

4. CONCLUSIONS Disruptive green industrial policy poses several Impact assessment of the effects on the environ- challenges that are not as salient for policymak- ment, the industry and the economy at large–as ers engaging in conventional green industrial well as social consequences for women and youth policy. It forces governments to pick specific and other marginalized populations and whether losers–firms and sectors that they wish to see or not gender gaps are further perpetuated by the restricted or phased out. Choosing those targets proposed policy–should occur at every step of the is a matter of assessing those sectors contrib- transition. If the transition is not well managed, uting most to environmental damage, for which the desired disruption can be stymied or signifi- there is an obvious viable green substitute. If cantly delayed by popular opposition. the target sector currently receives subsidies, so much the better: the shift can be financed in part Fortunately, successful models are available, by the savings. Ideally, substitute sectors align and some were discussed here. From them, it well with nationally enunciated development can be seen that it is crucial to get public buy-in, priorities and fit well with domestic capabilities. through extensive consultation and transparency combined with well-defined gradual timelines The most difficult aspect of disruptive green for change, and extensive efforts to ensure that industrial policy is orchestrating a just transi- there is support for those that might be nega- tion, as illustrated by the chequered experience tively affected. Private sector buy-in is also key. to date with one form, fossil fuel subsidy reform. Here again the process is important, with a need The sectors responsible for most environmental for consultation, transparency and well-planned damage tend to be those with significant capi- gradual efforts. Support can be offered to non-tar- tal investment and strong linkages throughout geted firms that may be negatively affected. For national economies in production and associated both the public and the business community, it supporting activities. The transition away from might help to be able to point to future growth such sectors, while necessary, will entail some created by new green domestic firms. level of stranded assets, obsolete infrastructure, direct and indirect job losses, and the attendant Not all disruptive green industrial policy will political battles against those negatively affected. actually make efforts to foster domestic expertise Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy

in the sectors that arise from the disruption. Each address a host of other obstacles to transition, country will need to identify appropriate poli- and for these a suite of complementary measures 83 cies on a spectrum that runs from soft efforts— may be needed. In fact, China implemented such restricting or phasing out sectors with no efforts policies, setting targets for renewable energy to steer the resulting emergence of green substi- penetration, providing finance, and investing tutes—to hard efforts: restricting or phasing out in education and research and development sectors with specific green substitutes in mind, to foster cleaner high value added production and fostering the growth of domestic players (Altenburg et al. 2017, this volume). in those substitute sectors. For some countries, particularly those without the requisite tech- The speed and scale of the transition needed to nical and fiscal capacity, hard disruptive green combat climate change, arguably humanity’s industrial policy may simply be infeasible. In this most pressing global environmental problem, respect, disruptive green industrial policy does demand a combination of phase out measures not differ from green industrial policy, or indeed and phase in measures where it is feasible, even from industrial policy. if our experience with industrial policy tells us that not all efforts to foster domestic capac- Where it is feasible, however, striving to phase ity will succeed. In that sense, disruptive green in substitutes for those sectors being targeted industrial policy should be seen as a crucially will offer greater potential to achieve the desired important complement to other sorts of green results of the transition. Simply getting the prices industrial policy efforts, disrupting the status right, as China did by revoking value added quo to create the space within which new green tax refunds to undesirable sectors, may fail to sectors can emerge. Green Industrial Policy - Concept, Policies, Country Experiences

84 REFERENCES Altenburg, T. (2011). Industrial Policy in Developing Burniaux, J.-M., & Chateau, J. (2014). Greenhouse Countries: Overview and lessons from Seven gases mitigation potential and economic effi- Country Cases (German Development Institute ciency of phasing-out fossil fuel subsidies. Inter- Discussion Paper No. 4/2011). Bonn. national economics, 140 (2014), 71–88. Altenburg, T., & Pegels, A. (2012). Sustainability-ori- Canadian Solar. (2016). Investor Presentation. First ented innovation systems: Managing the green Quarter 2016 Update. Retrieved from http://phx. transformation. Innovation and development, 2(1), corporate-ir.net/External.File?item=UGFyZW- 5–22. 50SUQ9NjM1NjMyfENoaWxkSUQ9MzM5NzQ2f- Altenburg, T, & Rodrik, D. (2017) Green industrial FR5cGU9MQ==&t=1 policy: Accelerating structural change towards CleanTech Canada (2016, March 20). IESO hands wealthy green economies. In Altenburg, T., down contracts for next wave of Ontario renewa- & Assmann, C2017). Green Industrial Policy. ble energy projects. CleanTech Canada. Retrieved Concept, Policies, Country Experiences (pp. from www.canadianmanufacturing.com/envi- 1–20). Geneva, Bonn: UN Environment; German ronment-and-safety/ieso-hands-contracts-next- Development Institute / Deutsches Institut für wave-ont-renewable-projects-164147/ Entwicklungspolitk (DIE). Cosbey, A. (2013). Green Industrial Policy and the Altenburg, T., Feng, K., & Shen, Q. (2017). Electric World Trading System (ENTWINED Issue brief mobility and the quest for automobile industry No. 17). Stockholm. Retrieved from www.iisd.org/ upgrading in China. In Altenburg, T., & Assmann, library/green-industrial-policy-and-world-trad- C. (Eds.). (2017). Green Industrial Policy. Concept, ing-system Policies, Country Experiences (pp. 185–198). Delucci, M. (2006). Life cycle Analyses of Biofuels. Geneva, Bonn: UN Environment; German University of California Davis. Development Institute / Deutsches Institut für Elsobki, M., Sherif, Y., & Wooders, P. (2009). Clean Entwicklungspolitk (DIE). Energy Investment in Developing Countries: Association of Power Producers of Ontario (APPrO). Wind power in Egypt. Winnipeg. (2015). The Value of Electricity to Ontario. Toronto. Fan, C. C. (2006). China‘s Eleventh Five-Year Plan Retrieved from Association of Power Produc- (2006–2010): From „getting rich first“ to „common ers of Ontario website: www.appro.org/index. prosperity“. Eurasian geography and economics, php?option=com_content&task=view&id=130&It 47(6), 708–723. emid=166 Fattouh, B., & El-Katiri, L. (2012). Energy subsidies in Auditor-General of Ontario. (2015). Annual Report the Arab world (Arab Human Development Report, 2015. Research Paper Series). Bazilian, M., & Rentschler, J. (2016). Reforming fossil G20. (2009). G20 Leaders’ Statement, the Pittsburgh fuel subsidies: Drivers, barriers and the state of Summit. Pittsburgh. progress. Climate Policy, 17(7), 891–914. Ganesan, K., Choudhury, P., Palakshappa, R., Jain, Blackwell, R. (2016, March 11). Ontario offers R., & Rajes, S. (2014). Assessing Green Indus- contracts to 11 firms to build new renew- trial Policy: The India experience. IISD report. able energy projects. The Globe and Mail. Winnipeg: International Institute for Sustainable Retrieved from https://beta.theglobeandmail. Development. com/report-on-business/industry-news/ energy-and-resources/ontario-taps-firms-to- Government of China. (2007). National Climate build-new-renewable-energy-projects/arti- Change Programme. Retrieved from www.china. cle29130899/?ref= www.theglobeandmail.com org.cn/english/environment/213624.htm Bloomberg New Energy Finance (BNEF). (2016). New Government of China. (2008). China‘s policies and Energy Outlook 2016. New York. actions for addressing climate change. Beijing, China: Information Office of the State Council of Bridle, R., & Kitson, L. (2014). The Impact of Fossil- the People’s Republic of China. Fuel Subsidies on Renewable Electricity Genera- tion. Geneva. Government of Ontario. (2013). Green Energy Invest- ment Agreement. Retrieved from www.ontla. Bridle, R., Kitson, L., & Wooders, P. (2014). Fossil-fuel on.ca/library/repository/mon/27006/323101.pdf subsidies: A barrier to renewable energy in five Middle East and North African countries. Geneva. Grossman, G. M. (1990). Promoting new industrial activities: A survey of recent arguments and Buen, J. (2006). Danish and Norwegian Wind Indus- evidence. OECD journal: economic studies. (14), try: The Relationship between Policy Instruments, 87–125. Innovation and Diffusion. Energy Policy. (34(18)), 3887–3897. Chapter 5 In With The Good, Out With The Bad: Phasing Out Polluting Sectors as Green Industrial Policy

Hallegatte, S., Fay, M., & Vogt-Schilb, A. (2013). Green Kai, M. (2006). The 11th Five-Year Plan: Targets, Paths Industrial Policies: When and How. and Policy Orientation: Address by the Minister of 85 Hanna, A., Schreiber, B., & Lehrer, E. (2012). Green the National Development and Reform Commis- Scissors 2012: Cutting Wasteful and Environmen- sion, March 23. Retrieved from www.gov.cn/ tally Harmful Spending. USA. english/2006-03/23/content_234832.htm Harris, M., Beck, M., & Gerasimchuk, I. (2015). The End Kojima, M., & Koplow, D. (2015). Fossil Fuel Subsidies: of Coal: Ontario’s coal phase-out. Winnipeg: Inter- Approaches and Valuation. national Institute for Sustainable Development. Lewis, J. I., & Wiser, R. H. (2007). Fostering a renewa- Harrison, A., & Rodríguez-Clare, A. (2012). Trade, ble energy technology industry: An international Foreign Investment, and Industrial Policy for comparison of wind industry policy support Developing Countries. In D. Rodrik & M. R. Rosen- mechanisms. Energy Policy, 35(3), 1844–1857. zweig (Eds.), Handbook in economics: Vol. 9,5. Lütkenhorst, W., Altenburg, T., Pegels, A., & Vidican, G. Handbook of development economics (2012 ed.). (2014). Green Industrial Policy: Managing Trans- Amsterdam: Elsevier. formation under Uncertainty (German Devel- Independent Electricity Systems Operator (IESO). opment Institute Discussion Paper No. 28/2014). (2014). Energy Output by Fuel Type. Retrieved Bonn. from www.ieso.ca/Pages/Power-Data/Supply. Lütkenhorst, W., & Pegels, A. (2014). Stable Policies, aspx Turbulent Markets–Germany‘s Green Industrial Independent Electricity Systems Operator (IESO). Policy: The costs and benefits of promoting solar (2010). 18 Month Outlook. An Assessment of the PV and wind energy. Winnipeg, Bonn. Reliability and Operability of the Ontario Electric- McCarthy, S. (2014). Northland seeks to expand wind ity System (December 2010 to May 2012). Toronto. farms. The Globe and Mail. 15/05/2014. Independent Electricity Systems Operator (IESO). McKitrick, R. (2013). Environmental and economic (2016). 18 Month Outlook. An Assessment of the consequences of Ontario‘s Green Energy Act. Reliability and Operability of the Ontario Electric- Ontario prosperity initiative. [Vancouver, B.C.]: ity System (July 2016 to December 2017). Toronto. Fraser Institute. Retrieved from www.ieso.ca/-/media/files/ieso/ Merrill, L., Bassi, A. M., Bridle, R., & Christensen, L. T. document-library/planning-forecasts/18-month- (2015). Tackling Fossil Fuel Subsidies and Climate outlook/18-month-outlook--2016jun.zip. Change: Levelling the Energy Playing Field. International Energy Agency (IEA). (2014). Morocco Denmark: Nordic Council of Ministers. Retrieved 2014. Paris. from http://search.ebscohost.com/login.aspx?di- International Energy Agency (IEA). (2014b). World rect=true&scope=site&db=nlebk&AN=1159272 Energy Outlook 2014. Paris: IEA. Never, B., & Kemp, R. (2017). Developing green tech- International Energy Agency (IEA). (2015a). Energy nologies and phasing them in. In Altenburg, T., & Technology Perspectives 2015. Paris. Assmann, C. (Eds.). (2017). Green Industrial Policy. Concept, Policies, Country Experiences (pp. International Energy Agency (IEA). (2015b). World 87–101). Geneva, Bonn: UN Environment; German Energy Outlook 2015. Paris. Development Institute / Deutsches Institut für International Labour Organization (ILO). (2010). Entwicklungspolitk (DIE). Climate change and labour: The need for a “just Ontario Energy Board. (2016). Historical elec- transition”. International Journal of Labour tricity rates. Retrieved from www.ontarioen- Research: 2(2). Geneva: International Labour ergyboard.ca/OEB/Consumers/Electricity/ Office. Electricity+Prices/Historical+Electricity+Price International Monetary Fund (IMF). (2013a). Case Ontario Ministry of Energy. (2010). Ontario’s Long- Studies on Energy Subsidy Reform – Lessons and Term Energy Plan. Building Our Clean Energy Implications. Future. Toronto. International Monetary Fund (IMF). (2013b). Energy Ontario Power Authority (OPA). (2007). Integrated Subsidy Reform: Lessons and Implications. Power System Plan I: Submitted to the Ontario Washington, DC. Energy Board August 29, 2007. Toronto. International Renewable Energy Agency (IRENA). Ontario Power Authority (OPA). (2009). FIT Rules 1.1. (2016a). Morocco Country Profile. Retrieved (Sept. 30). Toronto. from http://resourceirena.irena.org/gateway/ countrySearch/?countryCode=MAR Ontario Society of Professional Engineers. (2016). Ontario’s Energy Dilemma. Reducing Emissions International Renewable Energy Agency (IRENA). at an Affordable Cost (OSPE Research Report). (2016b). Renewable energy and Jobs 2016. Abu Toronto. Dhabi. Green Industrial Policy - Concept, Policies, Country Experiences

Pack, H., & Saggi, K. (2006). The Case For Industrial Terton, A., Gass, P., Merrill, L., Wagner, A., & Meyer, E. 86 Policy: A Critical Survey. Washington, D.C.: The (2015). Fiscal Instruments in INDCs: How Coun- World Bank. tries Are Looking to Fiscal Policies to Support Pahle, M. (2010). Germany’s dash for coal: Explor- INDC Implementation. Geneva. ing drivers and factors. Energy Policy, 38(7), UK Green Fiscal Comission. (2009). The case for 3431–3442. green fiscal reform: Final report of the UK Rodrik, D. (2004). Industrial Policy for the Twen- Green Fiscal Commission. London: Green Fiscal ty-First Century. Discussion Papers / Centre for Commission. Retrieved from www.greenfiscal- Economic Policy Research: no. 4767. London: commission.org.uk/images/uploads/GFC_Final- Centre for Economic Policy Research. Report.pdf Runge, C. F., & Senauer, B. (2007). How biofuels could United Nations Conference on Trade and Develop- starve the poor. Foreign affairs. (86), 41–53. ment (UNCTAD). (2016). Structural transformation for inclusive and sustained growth (Trade and Schlegelmilch, K., Eichel, H., & Pegels, A. (2017). Pric- development report). New York, Geneva. ing environmental resources and pollutants and the competitiveness of national industries. In Vagliasindi, M. (2012). Implementing energy subsidy Altenburg, T., & Assmann, C. (Eds.). (2017). Green reforms: Evidence from developing countries. Industrial Policy. Concept, Policies, Country Expe- Washington D.C.: World Bank. riences (pp. 102–119). Geneva, Bonn: UN Environ- Vidican Auktor, G. (2017). Renewable energy as a ment; German Development Institute / Deutsches trigger for industrial development in Morocco. In Institut für Entwicklungspolitk (DIE). Altenburg, T., & Assmann, C. (Eds.). (2017). Green Schwanitz, V. J., Piontek, F., Bertram, C., & Luderer, G. Industrial Policy. Concept, Policies, Country Expe- (2014). Long-term climate policy implications of riences (pp. 153–165). Geneva, Bonn: UN Environ- phasing out fossil fuel subsidies: Guiding Reform ment; German Development Institute / Deutsches of Energy Subsidies Long-Term. Energy Policy, 67, Institut für Entwicklungspolitk (DIE). 882–894. Wang, Q. (2010). Effective policies for renewable Schwarzer, J. (2013). Industrial Policy for a Green energy—the example of China‘s wind power— Economy. Winnipeg: International Institute for lessons for China‘s photovoltaic power. Renew- Sustainable Development. able and Sustainable Energy Reviews, 14(2), 702–712. Searchinger, T. R., Heimlich, R. A., Houghton, F., Dong, A., Elobeid, J., Fabiosa, S.,. . . Hayes, T.-H. Y. (2011). Wang, X., Li, J., & Zhang, Y. (2010). Can Export Tax be Use of U.S. Croplands for Biofuels Increases Genuine Climate Policy? An Analysis on China’s Greenhouse Gases through Emissions from Land- Export Tax and Export VAT Refund Rebate Poli- Use Change. Science. (319(5867)), 1238–1240. cies (Idées Pour Débat No. 08/2010). Paris. Stoianoff, N. P., Kreiser, L., Butcher, B., Milne, J. E., & Yaneva, M. (2015). Siemens to build EUR 100m Ashiabor, H. (Eds.). (2016). Green fiscal reform blade factory in Morocco. SeeNews. 29/12/2015. for a sustainable future: Reform, innovation and Retrieved from https://renewablesnow.com/ renewable energy. Critical issues in environ- news/siemens-to-build-eur-100m-blade-facto- mental taxation: volume 17. Cheltenham: Edward ryin-morocco-507214/ Elgar Publishing. Zhang, S., & He, Y. (2013). Analysis on the development and policy of solar PV power in China. Renewable and Sustainable Energy Reviews, 21, 393–401. Chapter 6 Developing Green Technologies And Phasing Them In CHAPTER 6 87 DEVELOPING GREEN TECHNOLOGIES AND PHASING THEM IN

Babette Never, René Kemp Green Industrial Policy - Concept, Policies, Country Experiences

12 88 1. INTRODUCTION The development and phase-in of green technol- failures may be more pronounced due to stronger ogies present complex challenges for industrial- information asymmetries and cost barriers. ised and developing countries alike. First, neither international nor domestic market prices reflect In this chapter, the goal is to identify concrete the environmental externalities that come with options that governments in developing countries conventional, non-green technology choices. can usefully and realistically do to phase in green Therefore, producers and consumers do not technologies. Factors for success and steps to receive a cost-based incentive from the market manage transitional change are outlined and it is to shift towards green technologies. Second, other argued that, given the various market failures and market failures such as inadequate information lock-in effects, a deliberate process for phasing on products and amortization periods, uncertainty in green technologies is required in most cases. about benefits, unclear incentives or behavioural Longer time-spans, mapped-out policy steps and barriers impede investments in green technolo- well-designed cooperation mechanisms among gies. As a consequence of those market failures, stakeholders need consideration. Uncertainty has product development and consumer preferences to be overcome. Furthermore, policies that ignore are focused on non-green technologies. Third, the heavily vested interests in existing technolo- green technologies are disadvantaged by the gies and systems are unlikely to work. The impor- continuous investment in and habitual use of tance of these various elements is illustrated by polluting technologies that reinforce the lock-in the cases discussed and aligns with innovation of socio-technical systems (Kemp 1994; Unruh system dynamics (Markard and Truffer 2008). 2000, 2002). They also align with political economy studies of industrial policy that emphasize the importance Lock-in means that existing rules, organizational of rent management and the need of a competent structures, sunk costs and actors’ interests as well bureaucracy able to deal with information asym- as knowledge and skill accumulation combine to metries (Rodrik 2014; Altenburg and Engelmeier stabilize an entrenched paradigm (Geels 2004; 2013; Schmitz et al. 2015). Kemp 1994). This is reinforced by regulating agencies that promote well-known and non-risky Further, successful phase-in experiences in technologies, by economies of scale of incumbent industrialised countries of Germany and the versus emerging products and by network effects Netherlands and in emerging economies of China such as the interdependency of technologies, and India are presented. The selected countries infrastructure and industries growing over time all have specific phase-in programmes for green (Unruh 2002). The development and implementa- technologies but differ with regard to finan- tion of alternative technologies and systems thus cial and institutional capacities; therefore, each face serious problems, in the form of opposition country developed different policy approaches. from current beneficiaries and of short-term costs This chapter outlines useful elements of a green weighed against undetermined long-term benefits. transition approach for developing countries, with special attention to difficulties they may These obstacles particularly challenge radi- encounter. Transition management literature cal systemic changes, those requiring shifts in offers useful starting points as it discusses a institutions and long-term development periods. variety of options for policymakers. The litera- The new techniques and technologies initially ture suggests that technology phase-in becomes produce low payoffs because they have not yet possible through experimentation with different benefited from the dynamic scale and learning policy mechanisms, through the development effects that result in cost reductions per unit of and protection of specific technological niches output and incremental improvements (Kemp or through a combination of both. Transition 1994). Moreover, they tend to be especially chal- management for developing countries thus offers lenging in developing countries where poli- more flexible options to address market failures, cymakers have limited financial, institutional lock-in and vested interests than traditional tech- and professional resources and less experience nology policy approaches that support a specific with state-driven technology development and pre-defined technology only with research and phase-in. While lock-in effects may be weak in development and other financial incentives. some developing countries where techno-institutional development is just beginning, market

12 A different version of this chapter has been published as 'Green transition, industrial policy, and economic development' (Kemp and Never 2017), which was based on the version of this chapter. Chapter 6 Developing Green Technologies and Phasing Them In

Phasing in green technologies in developing from information failures and behavioural barri- countries is less about discovering new techno- ers. Thus, both sectors require government inter- 89 logical niches and more about recognizing which vention for successful phase-in. existing opportunities coincide with development objectives and building on that advantage. The structure of the chapter is as follows: Section Few developing countries can financially nurture 2 explains the transition management approach industries from infancy to global maturity. For by summarizing the relevant literature. It gives a the purpose of this chapter, it is therefore useful brief overview of potentially relevant factors for to look at sectors with a fairly established set of managing transitions in developing countries. technologies available on the global market that Section 3 suggests which of these factors have can be adapted to local requirements. Renewa- been important in existing phase-in processes. ble energy, particularly solar photovoltaics, and Five policy examples are presented: energy energy efficiency were identified to be such efficiency in buildings in Germany, industrial sectors. Both are constrained by a variety of energy efficiency in China, energy efficiency in market failures, uncertainties and lock-in situa- India, solar photovoltaics in India, and a broader tions. Renewable energies have to be phased in transition management approach in the Nether- against a strong lock-in of fossil fuel-based elec- lands comprising renewable energy and energy tricity and heating systems in most countries, efficiency. In conclusion, section 4 specifies key and often prices are still not competitive with elements of a green transition approach for devel- conventional grid-based electricity. Energy effi- oping countries by discussing the lessons from ciency implementation suffers primarily from the five examples. non-transparent and inadequate pricing—but also

2. PHASING IN FROM A TRANSITION MANAGEMENT PERSPECTIVE The transition management literature offers many The development of strategic, long-term visions is starting points for overcoming market failure a necessary starting point for government-driven and lock-in and for dealing with uncertainty and phase-in processes. Participating stakehold- vested interests. However, operationalisation in ers need to change existing goals and principles, developing countries requires focus on particu- instead of waiting for others to challenge old goals lar factors. This section outlines prospective key or present new ideas. The re-clarification of visions elements for such an operationalisation by clarify- and lasting goals serves stakeholders’ long-term ing the goals, design principles and tools of transi- anticipation of policy developments and thus gives tion management. certainty to investors, innovators and consumers. Visioning exercises provide the desired result Transition management seeks to overcome a situ- of the transition—in this case, a renewable ener- ation of lock-in and market failure by co-producing gy-driven economy. Once these desired results are and coordinating policies step by step with clear agreed, determining the necessary interim goals targets and programmes for system innovation. and planning the policies needed to reach the Co-production of policies means support for both interim goals can start. top-down and bottom-up ideas and processes. The goal is to accelerate a structural transformation The design of interim steps and goals does not of the whole economy or of important sectors by require omniscience on the part of the govern- utilizing innovation opportunities and designing ment to pick winners, nor does it require that all policies that foster private investment into those mistakes must be avoided “What is needed instead opportunities. Transition management relies is a set of mechanisms that recognises errors and on the development of a long-term vision with revises policies accordingly” (Rodrik 2014: 472). interim transition goals and pathways by involv- Several tools for selecting innovation possibilities ing relevant actors and incorporating experimental and models for nurturing them exist: a flexible programmes and projects as well as mechanisms portfolio of options, an open search process with of policy learning (Loorbach 2010; Kemp et al. 2007). business and research, or a prior identification For developing countries, the complementary of user needs to help select priority sectors and expansion of institutional capacities and technol- technologies. It is crucial to enable policy learn- ogy-related skills is essential. ing while countering vested interests through Green Industrial Policy - Concept, Policies, Country Experiences

embedded autonomy (Evans 2012). Embedded The transition management approach has been 90 autonomy is the engagement of government criticised for its too-strong focus on technolog- representatives in business networks to obtain ical, rather than broader societal, innovation; relevant information and capabilities to formulate its emphasis on the policy design phase and its feasible economic goals without allowing political de-emphasis of political factors, as well as the capture. Political capture here means more than role of societal actors in the implementation corruption; powerful interest groups may attempt phase (Kemp et al. 2007). The disregard of polit- to divert policy measures away from government’s ical factors presents the most important caveat. intended goals to undermine the effectiveness of The approach presupposes that policymakers and those measures (Lütkenhorst et al. 2014). implementing agencies are technically capable and fully committed to increasing social welfare The operational part of transition manage- without being influenced by lobbying pressure, ment emphasizes two aspects: experimental patronage, populism or party politics. In the real programmes and projects and/or strategic niche world, such conditions are the exception rather management. Experimental programmes and than the norm. Systemic transitions result from projects can be used to test technological options, complex political bargaining processes among learn about user satisfaction, identify prob- competing interest groups, while policymakers lems with a range of technical issues and create and implementers often pursue self-interests networks for cooperation (Kemp et al. 2007). The that mingle with their official mandates. Lobby- role of governments is to make sure that inter- ists will try to influence the policymaking process esting experiments happen and that lessons to gain personal benefits, and politicians as well get evaluated and disseminated. For promising as bureaucrats will strive for power resources and technologies with possibilities for branching political support (Chang 1996). National politics and cost-reduction, they may engage in strategic are too often shaped by exclusive elite deals and niche management, in case the market smothers patronage, which strongly deviate from the ideal- any such developments. Strategic niche manage- type of a state with an efficient, meritocratic and ment involves the promotion of promising techno- accountable bureaucracy. logical experiments through financial incentives, networking and supportive policies for scaling These risks loom clearly in the management of up in a way that protects the experiments from systems transitions. Here, policymakers need to competitive pressure. Once these innovations create special economic incentives, policy rents, have reached full market maturity and begun to to steer the economy towards a more desirable diffuse widely, the supportive policies can be with- new state and withdraw incentives from unde- drawn (Kemp et al. 1998). To what extent strategic sirable activities such as polluting. Such rent niche management presents a feasible option for management presupposes particularly capa- developing countries depends on the function- ble and welfare-oriented governments that are ing and capacities of local innovation systems. immune to political capture (Khan 2004; Alten- In both experimental programmes and strate- burg and Engelmeier 2013). Therefore, the risks of gic niche management, starting small and scal- public or private interest groups distorting transi- ing up successful programmes through iterative tion policies should not be underestimated. processes supports the process from a nurturing phase to an actual launch. Transition management does not prescribe one particular path for phasing in green technologies. Policy learning is essential. The transition Instead it offers a framework for exploiting green management approach acknowledges that development opportunities in a forward-looking, a successful phase-in of green technologies adaptive way. Environmental protection benefits requires a long time-span with several cycles of can be combined with economic benefits in such adjusting policies until full implementation of a green transitions approach. The next sections the desired outcome is reached. Explicit policy analyse different successful cases to identify both learning rounds are planned to evaluate which common and unique elements of phasing in from interim goals have been achieved; whether the a transition management perspective that could process is dominated by certain actor groups and be particularly important for developing coun- vested interests; and how regulations, procedures tries. Exploring the political economy behind and goals need to be adjusted to match long-term each of these practical examples is beyond the and short-term concerns (Rotmans et al. 2001). scope of this chapter. Interestingly, however, the Ideally, this leads to gradual changes that mini- cases show that governments of industrialised mise social resistance. and emerging economies alike have been able to manage transitions effectively. Chapter 6 Developing Green Technologies and Phasing Them In

3. HOW TO INSTIGATE AND MANAGE THE 91 TRANSITION? PRACTICAL EXAMPLES

The following five examples of successful ◼◼ Builders, owners and tenants lack information phase-in programmes will all be introduced by and knowledge regarding a building’s energy a brief explanation why such a programme was performance. required, followed by a description of what has ◼◼ The variety of technologies and the meas- been done. Then the programme’s effects will urement challenges add to the complexity of be discussed in the context of industrial policy. delivering more efficient buildings. Finally, key elements of each approach will be summarized. The German, Chinese and Indian The German government’s approach to increasing cases are all sector-specific examples. The case energy efficiency in both old and new buildings study of the Netherlands showcases a more involves measures around three pillars: regulation, systemic approach on a comprehensive level, mainly command-and-control instruments; finan- encompassing the whole energy sector. For devel- cial incentives; and information to build trust. oping countries, the choice between a more selec- The three pillars were designed to complement tive or systemic approach depends on the state of and mutually strengthen each other (Schimschar the sector, such as market and readiness to inno- 2013; Schröder et al. 2011). Moreover, the different vate; number and interest of actors; technologies measures are linked to renewable energy objec- already available; the linkages between actors tives to form a coherent energy policy package. By and market segments and analysis of which way of these measures, by 2050 primary energy technologies can be locally adapted in line with consumption in buildings should be only 20 per development objectives. More developed market cent of consumption in 2008. segments with potential for strong linkages to Germany’s first thermal insulation regulation others, such as solar photovoltaic modules, seem was developed early in 1977, but political momen- to work best through a systemic approach. tum towards implementing mandatory standard setting only gained traction in the 1990s. 3.1. ENERGY EFFICIENCY IN BUILDINGS Currently, the energy saving ordinance, called IN GERMANY Energieeinsparverordnung (EnEV) and comprised of the German building code and the act on the In Germany, buildings account for around 35 per promotion of renewable energies in the heat cent of final energy consumption that includes sector, provides the regulatory framework. The heating, ventilation and warm water provision German energy saving ordinance was introduced (NAPE 2014). Roughly 75 per cent of residential in 2002, in line with that year’s European Union buildings are older than 30 years and require directive on the energy performance of buildings, improved heating and insulation systems. In and was revised and tightened in 2007, 2009 and spite of existing regulations, energy efficiency in 2013. The EnEV 2013 is the first regulation that buildings is difficult to achieve, especially in older outlines a stepwise tightening of construction law buildings that require refurbishment. The follow- to a carbon neutral standard by 2021 (El-Shagi et ing market barriers exist in industrialised and al. 2014). Standards are thus incrementally intro- developing countries alike, calling for governmen- duced and tightened over time. tal intervention (Liu et al. 2010): Financial incentives to comply with regulations ◼◼ When constructing a new building, future and voluntary mechanisms are provided in the monthly energy costs for cooling, lighting form of subsidy programmes for the refurbish- and heating make up a small share of the cost ment and construction of new buildings by the calculations. Energy subsidies may reinforce public bank, Kreditanstalt für Wiederaufbau this disincentive for energy efficiency invest- (KfW), and for heating with renewable ener- ments. Owners may be unwilling to modernise gies, Marktanreizprogramm zur Nutzung erneu- buildings because they do not want to write off erbarer Energien im Wärmemarkt (MAP). The investments. Both situations distort cost-ben- government’s 2004 CO2 Reduction Programme efit perceptions. includes two subsidised loan schemes: the KfW ◼◼ Building owners may have other interests programme for energy efficient refurbishment than tenants, creating split incentives among and the KfW programme for energy efficient stakeholders. new buildings. Grants and loans are staggered ◼◼ The building sector’s many stakeholders lack according to the degree of energy efficiency: coordination. Green Industrial Policy - Concept, Policies, Country Experiences

the more energy efficient the building is, the time to adjust. New standards are tested in pilot 92 higher the grant and the more favourable the projects to identify the optimal technology repayment conditions for loans. This approach available on the market and revised every few covers the whole range of technically feasible years. While the regulatory framework is being options on the market and it is revised regularly expanded the German Energy Agency, called to increase energy efficiency. Preferential loans Deutsche Energie-Agentur (DENA), and KfW with low interest rates for energy efficiency can promote model and experimental projects and be combined with loans for a new heating system commission research to accompany and evaluate based on renewable energies. running programmes. For example, DENA developed and tested the standards for Low-Energy Information-based programmes, such as the buildings on 400 individual projects (Figure 6.1). mandatory energy performance certificate for KfW subsequently adopted these standards to buildings and the targeted promotion of jobs and support an additional 5,000 prototype buildings expertise related to energy efficiency, comple- (Schröder et al. 2011: 35). ment the policy package. Various information and trust-building measures for homeowners, tenants, The German energy efficient building programme experts and the construction industry support has had positive effects on energy consumption, knowledge exchange and provide security for regional value chains and investment dynam- stakeholders in the market. ics in the sector. The incremental introduction and tightening of regulations accompanied by The incremental introduction of standards research and model projects have led to substan- presents a particularly interesting element in tial reductions in primary energy consumption the German three-pillar approach. It provides over time (Figure 6.1). the market with both clear signals and sufficient

Figure 6.1: Long-term phase-in of building standards

Primary energy requirement in kWh/m2a

300 Minimum requirements (WSVO/EnEV) 250

200 Solar buildings

150 Building practice

100 Low-energy buildings

50 Research 3-litre buildings (Model projects) Zero-heating energy buildings 0 Energy-plus buildings -50 1980 1985 1990 1995 2000 2005 2010 2015

Source: National Energy Efficiency Action Plan Germany (2014).

The incremental tightening of regulatory require- In 2011, the energy efficient refurbishment of ments pushes energy efficient building practices existing buildings affected regional value chain towards greater efficiency. The continuous devel- by EUR 14 billion and generated about 287,000 opment and testing of most efficient buildings full time jobs (Weiß et al. 2014). Since 2006, 3.7 in model projects, such as solar buildings and million dwellings could be refurbished or newly zero-heating buildings, create a pull-effect on the built in an energy efficient manner with an over- market. The actual building practice lies between all investment volume of EUR 182 billion. On aver- the respective regulation at the time and the most age, between 2006 and 2011 every public Euro efficient technologies possible, becoming more invested in KfW subsidy programmes generated energy efficient over time. EUR 10–12 of private investment (Weiß et al. 2014). Chapter 6 Developing Green Technologies and Phasing Them In

Additionally, a strong indirect effect of regulation (Zhou et al. 2010). Following the introduction of the on innovation can be witnessed. Investors in the Energy Conservation Law in 2001, the government 93 high-end housing market segment strongly react developed a comprehensive package of manda- to a narrowing technology advantage compared to tory and voluntary policies and measures aimed adopters in the low-quality market, even if their at advancing energy efficiency and energy saving. own performance is not threatened by regula- The Medium and Long Term Energy Conservation tion in the near future (El-Shagi et al. 2014). These Plan was published in 2004. In the 11th Five Year results confirm a weak form of the Porter Hypoth- Plan (2006–2010), command-and-control regula- esis on regulation and innovation–that environ- tions were combined with taxes and subsidies— mental regulation stimulates innovation and thus sticks, carrots, sermons and prohibitions (Yang et increases the competitiveness of firms and the al. 2015). economy (Ambec 2017, this volume). Sticks: The Differential Electricity Pricing Policy for Overall, the continuous revisions of regulation industries consists of four categories with differ- and incentives, as well as the combination of ent surcharges that increase with consumption. different types of measures in a policy package, More efficient enterprises thus pay less. Between explain the success of the German programme 2004 and 2010, these types of taxations were subse- thus far. A high degree of collaboration among quently increased. However, initial charges were actors in the policy process and the integration found to be inefficient in terms of increasing energy of experimental pilot projects with accompany- efficiency investments. This was due to heteroge- ing research that fostered feedback and learn- neous, counteracting local policies and fluctuating ing cycles, as well as the gradual tightening of selling prices and other production costs (Yang et requirements in all fields, supported the achieve- al. 2015). To phase-out inefficient enterprises, the ments greatly. Finally, energy efficiency meas- surcharge in this category was increased by a factor ures were explicitly coupled with the promotion of four over time. of renewable energy usage in buildings. Carrots: Companies can receive a reward of 250 Yuan per tonne of coal equivalent saved through 3.2. INDUSTRIAL ENERGY EFFICIENCY technical upgrading and engineering projects IN CHINA or the Ten Key Energy Saving Projects. A number of research and development support strategies A decade ago, the Chinese industry appeared to for different business sectors as well as financial be locked into an unsustainable, fossil fuel-based compensation for technical retrofitting and the path. Industrial energy consumption increased phasing-out of small and inefficient industrial significantly in China: from 34 per cent in 1990 plants exist (Yang et al. 2015). to approximately 70 per cent of total national energy consumption in 2009. This decreased to Sermons: The Top 1000 energy savings agreement 50 per cent in 2015, but 65 per cent of all electric- between key industry and government was volun- ity was still produced from coal in 2016 (China tary when first introduced in 2006 and gradually Energy Portal 2017). Since energy prices for indus- extended to provincial and local levels (Yang et al. try are negotiated regionally, the market does not 2015). In 2012, the programme became mandatory send a consistent countrywide signal to invest in and was expanded to the Top 10,000. It requires energy efficiency. Before the Chinese government companies to annually send their energy use statis- started its industrial energy efficiency programmes, tics to government and to meet national and inter- companies hardly invested in energy efficiency. national standards. The Top 10,000 programme and The central government does provide targets and its combination with energy management systems guidelines to local governments, but keeps policy helped to raise awareness among provincial author- goals broad enough to allow for local interest ities and top-level management. But the implemen- alignment and bundling of policies and incentives tation of it is impeded by a lack of understanding (Harrison and Kostka 2014). in non-Top 1000/10,000 companies, as well as a lack of funding and adoptable technical means (Gold- Since the 1980s, China had been monitoring energy berg et al 2011; Li et al. 2014). The proper adoption of use in industry. In 1995, the government published energy management systems–either the national a guideline for energy management in industry, standard or ISO 50001–can actually make it easier indicating to companies that future regulation was for companies to comply with regulations and to come. Concrete interest in energy management monitoring schemes. The electricity saved may systems in industry started to rise in the early then qualify the business for a different electricity 2000s after it became clear that rapidly increasing price category. energy demand had to be brought under control Green Industrial Policy - Concept, Policies, Country Experiences

Prohibitions: The 2006 Plan on Energy Conser- in a rather short time span, while also establish- 94 vation and Emissions Reductions sets targets for ing the topic firmly on the political agenda and the closure and phase-out of small and inefficient creating local jobs. The shift from voluntary to production facilities. Several provincial and local mandatory measures is based on an assessment governments, however, prioritise local economic of the actual capabilities of leading industry to development and jobs. They protect local factories reduce energy intensity. Strategies are adjusted from closure or let them operate unofficially, thus over time if they do not bring the desired results: opposing national policies (Yang et al. 2015). experimentation, policy learning and strategic space for local interest alignment characterize China’s approach to industrial energy efficiency did the phase-in process. Fiscal incentives are part of produce the envisioned reduction in energy inten- the broader energy efficiency policy package, but sity of 20 per cent by 2010, but progress since has are not yet interlinked with other mechanisms to been rather slow. The implementation of national maximise the effect. The existing political econ- energy policies relies on creative bundling of inter- omy challenges between central and local govern- ests, incentives and policies by local governments ance levels indicate that awareness, capabilities and administration to minimise opposition of local and interests of different actors require projected players (Harrison and Kostka 2014). The govern- considerations in the planning of green technol- ment is struggling to achieve structural change ogy phase-in. This is particularly important for that actually replaces less energy-efficient firms developing countries facing similar challenges. and technologies with more efficient ones across all sectors during the 12th 5-year period of 2011 to 2015 (Ke et al. 2012). Official documents state that 3.3. ENERGY EFFICIENT LABELLING OF 340 million tonnes of coal equivalent have been APPLIANCES IN INDIA saved under the Ten Key Projects programme until In India, residential energy demand accounts 2010.13 A gradual, cautious introduction of a carbon for 45 per cent of the country’s primary energy trading system is now envisioned to comple- consumption, 80 per cent of which can be ment existing measures. The diffusion of these ascribed to only five appliances: ceiling fans, TVs, measures to medium and small energy intensive lighting, refrigerators and air-conditioners. The companies presents a challenge in the next stage consumer appliance market grew at an annual of China’s phase-in process. average of 13 per cent between 2003 and 2013, Concerning industrial policy effects on the supply increasing overall energy demand (Jairaj et al. side, the market for energy efficiency consultants 2016). Inefficient appliances have a lower price on and energy service companies has been devel- the market and consumers often lack the infor- oping since 2006. From 2010 to 2015, this market mation that more energy efficient appliances save grew by 31.9 per cent annually, employing 654,000 more on their electricity bill so that the higher people in approximately 2,600 companies by purchase price can thus be amortized quickly. 2015 (IbisWorld 2015). In spite of these numbers, Manufacturers had no incentive to invest in the energy efficiency consultants and energy service development of more efficient, higher priced companies have been criticised for not working goods that they feared consumers would not buy effectively enough due to imperfect business (Chaudhary et al. 2012). These market failures models, asymmetric information, high transac- needed to be overcome to use India’s projected tion costs, lack of ability to build a relation of trust potential to save up to 240 TWh of electricity to customers and lack of skills especially in new through energy efficiency standards by 2030 energy industry services (Kostka and Shin 2013). (Letschert 2014). India has adopted a ‘market- Energy auditing capabilities vary greatly through- plus approach’ (Harrison and Kostka 2014). The out the country. While the market is developing rules of the market are defined by the state, but well, it is far from having reached maturity and the limited capacity of the state to ensure wide- use of its full potential across the whole country. spread implementation was acknowledged from the start, recognizing that different actor groups Overall, the current Chinese policy package and with diverging interests needed to be brought on path for phase-in are characterized by a short common ground (Harrison and Kostka 2014). period of voluntary agreements and incentives, followed by comprehensive mandatory require- In 2001, the government set up the Bureau of ments for large industry. This led to an impressive Energy Efficiency (BEE) with the mandate to achievement of energy consumption reductions develop an energy efficiency programme for

13 Ke and co-authors (2012) doubt this figure. Due to measurement and calculation challenges, it is difficult to ascribe effects to specific programmes. Chapter 6 Developing Green Technologies and Phasing Them In

household appliances and lighting, buildings and labelling, while 17 additional products were under industries, including standards and labelling. It the voluntary labelling scheme. In December 95 had been clear to stakeholders for several years 2012, the target of the standards and labelling that a standard and labelling programme for appli- programme had already been surpassed: 7,766 ances was due and it was introduced in 2006. In MW of new generation capacity addition were the planning and design phase, extensive consul- avoided, greatly surpassing the 3,000 MW esti- tations had been held with manufacturers, utilities, mated (BEE 2014). Market uptake for appliances standardization bodies and other stakeholders. with 4 and 5 star levels of efficiency has generally The programme consists of a star-rating scheme been slower than for medium-level appliances based on minimum energy performance standards (Jairaj et al. 2016). While the initial phase of the set by BEE and represented by a range from 1 to programme had been highly participatory with 5 stars. Labels were voluntary at first, then made a clear roadmap, the frequency of review and mandatory once labelled products reached 50 per the intention to ratchet up current standards are cent of overall sales. Facilities for testing the appli- not sufficiently clear to all manufacturers now. ances were developed to allow for independent Consumer awareness programmes have achieved testing and mandating government procurement growing brand recognition of the star label, but at the highest star ratings created market pull the share of inefficient devices is still significantly (Chaudhary et al. 2012). BEE plans to ratchet up the higher than the one for higher star-labelled ones. performance standards for the stars over time in The absence of a targeted communication plan an attempt to promote product innovation. limits the full outreach potential, for instance via civil society organizations (Jairaj et al. 2016). In the initial phase of the programme, standards These factors currently retard the building of a and labels were designed to encompass a majority broad societal acceptance of energy efficient of the market and to provide the clear message appliances across all of India. that standards would be ratcheted up. Extensive consumer outreach and information was provided The factors responsible for the success of the to build up the labels as a brand. Manufactur- programme thus far have been a detailed under- ers self-certified the products by adopting the standing of the local needs, the sector and institu- approved testing procedure and were thus respon- tional context as well as extensive consultations sible for the accuracy of the labels. Due to under- with a range of stakeholders (Chaudhary et staffing, BEE outsourced the process of collection, al. 2012). The idea of embedded autonomy and verification and processing of self-certificates to communication worked well in this case. This a consulting firm (Jairaj et al. 2016). BEE carries ensured the buy-in of all relevant actors, includ- out market assessments regularly to understand ing incumbents and innovators in the system. market penetration, but these are not publicly The gradual tightening of standards and labels available, making it difficult to judge how much in line with the roadmap initially provided to the diffusion of efficient products has advanced stakeholders was initially helpful, but needs to be in each product category (Khandari 2011). Since improved so that the process is clear to all rele- BEE has been functioning as both the designer vant stakeholders. of programmes and facilitator to state governments and businesses at all levels, the reliance on 3.4. SOLAR PHOTOVOLTAICS IN INDIA outsourcing services and the development of an energy efficiency consultants and energy services India has a high solar irradiation and sufficient market soon became inevitable. While this market amounts of open space in several of its states. is evolving, it is still largely located in Delhi, There is a vast potential for both on and off-grid Mumbai and Pune (Harrison and Kostka 2014). solar energy, offering both energy and socio-eco- Future challenges across India include the estab- nomic development opportunities. However, the lishment of trusted client-energy service relation- energy system has been locked into fossil fuels ships as well as increasing the attention of public with a strong local coal industry for a long time. agencies to electricity price and expenditures. The key dilemma was this: Converting solar Co-benefits are the key argument for energy effi- irradiation into electric energy was much more ciency in India, thus bundling interests of diverg- expensive than the well-established technology ing and possibly opposing stakeholders (Harrison of coal-fired power plants. It was clear that solar- and Kostka 2014). based electricity would become much cheaper once it was produced at scale and that burning By 2016, frost-free refrigerators, tubular fluores- coal caused severe health problems related to cent lamps, room air conditioners and distribu- air pollution, as well as global warming. However, tion transformers had progressed to mandatory the market did not provide any incentive for Green Industrial Policy - Concept, Policies, Country Experiences

developing solar projects as long as electricity annual rate of more than 300 per cent in the last 96 from coal was cheap and environmental external- five years. In 2014, approximately 125,000 people ities were not considered. Additionally, a lack of were employed in the solar photovoltaics sector regulatory certainty discouraged investments in in India, both on and off-grid. At the same time, the sector. the tariff that the government has to guarantee for new solar projects has been reduced to below The central government’s solar photovoltaics EUR 0.08/KWh on average in 2014 to 2015 (IRENA programme sought to change this. It offered a 2015)–not much more than the cost of a KWh from policy package to develop solar energy technol- coal-fired power plants. ogy into a viable and competitive alternative by shielding it from price competition with coal. As The Jawaharlal Nehru National Solar Mission was such, it can be understood as an example of stra- less successful with regard to another objective, tegic niche management, in the terms of transi- namely to boost domestic photovoltaic cell and tion management literature. module manufacturing capabilities. To achieve this objective, a local content requirement was An interest in solar policies began to arise in the included in the solar energy incentive package. states of Gujarat and Rajasthan in the early 2000s. There are two types of photovoltaic cells: crystal- In 2008, the Jawaharlal Nehru National Solar line silicon and thin film. Before the Solar Mission, Mission was announced as part of the National India already had a decent production of crystal- Action Plan on Climate Change. It provides a long- line silicon cells that was mainly export-oriented. term vision with clear targets to be achieved in The local content requirement made local sourc- each of its phases. The Solar Mission’s different ing mandatory for crystalline silicon cells but not phases include a comprehensive set of meas- for thin film cells, which were not produced locally ures: preferential feed-in tariffs, renewable energy at all. Project developers then tried to avoid the purchase obligations and certificates, tax incen- mandatory local sourcing by almost exclusively tives, preferential loan schemes, as well as local offering projects on the basis of thin film technol- content requirements to support the build-up of ogy (Johnson 2013). Therefore, a well-intentioned national manufacturing capabilities. In terms of local content regulation resulted in a bias towards targeted innovation support, government institu- foreign thin film photovoltaic manufacturers tions receive a 100 per cent, and the private sector rather than supporting the local crystalline silicon a 50 per cent, research and development subsidy manufacturing industry (CEEW and NRDC 2012). for advancements in solar energy (Sahoo 2016; Quitzow 2015). State-level incentive programmes In a nutshell, the solar policy was highly success- complement the federal legislation. ful in terms of phasing in solar power technologies from the beginning, but not necessarily for The most interesting element of the Solar Mission the promotion of local green industrial production. is the reverse bidding approach that brought the The policy design and implementation incorpo- required subsidies down at an unprecedented rate a good deal of experimentation and system- speed. The Government of India auctioned atic learning. This is noteworthy. With regard to licenses for solar power projects asking the the reverse bidding approach, the government bidders to hand in proposals below a reference started with a small test run that helped to identify price of EUR 0.27/KWh that the government would some loopholes in the bidding process that were guarantee for 25 years. In the first round, the eliminated in subsequent rounds. In fact, the spec- government auctioned very small projects of just ification of tenders was improved from one round 5 MW. Already here, investors’ engagement was to the next. Also, the government commissioned overwhelming, and the lowest bidders offered to a systematic mid-term review to an independent build such plants with a feed-in tariff of only EUR organization. The report praised that the solar 0.11/KWh. In subsequent rounds, the government mission “follows a phased approach that allows auctioned larger units and, from round to round, the government to modify guidelines and poli- received lower bids. This way, the government cies based on the experiences gained and lessons managed to trigger an enormous growth of the learned in earlier phases” (CEEW and NRDC 2012). solar energy market while at the same time bring- It also clearly identified the local content require- ing the price differential between conventional ments as dysfunctional, and these were suspended and solar energy down and greatly reducing the for the next phases. Overall, the Indian govern- fiscal cost of the solar feed-in tariff (Altenburg and ment’s sequential approach–with a clear roadmap, Engelmeier 2013). Since 2009, the start of the Solar intermediate targets, packaging of different meas- Mission, India’s installed solar capacity increased ures and experimental learning–has been hugely from virtually zero to 5,130 MW by January 2016. successful for phasing in solar photovoltaic energy. The solar energy market has been growing at an Chapter 6 Developing Green Technologies and Phasing Them In

3.5. TRANSITION MANAGEMENT IN Following the work of the transition platforms, THE NETHERLANDS 31 transition paths have been explored through 97 research and innovation experiments. These The Dutch transition management approach for included, for example, biomass for electricity, sustainable energy is a systemic green industrial clean fossil fuel, micro cogeneration and ener- policy approach. Its goal is to stimulate new busi- gy-producing agricultural greenhouses. The ness in sustainable energy and to foster a broader government acts as a process manager, dealing energy transition. To develop such a systemic with issues of collective orientation and inter- green industrial policy, the traditional divide departmental coordination. It also has a respon- between the Ministries of Economic Affairs and sibility to undertake strategic experiments and the Ministry of Environment had to be overcome. programmes for system innovation. This happened in a deliberative process that resulted in the co-production of the transition The transition approach went beyond technology management framework (Kemp and Rotmans support. The creation of capabilities, networks 2009). It envisioned the set-up of seven transition and institutions for transitional change through platforms, the development of a transition action agenda-setting, new partnerships and instru- plan and different transition paths explicitly ments was equally important, leading to actor entailing experimentation as well as institution, coordination across national, state and local capability and network building. levels and cross-sectoral policy coordination. The Interdepartmental Project Directorate Energy The transition approach officially started in Transition presents one group that had a central 2002 with the project implementation transition role in the process. It was set up in parallel to the management of the Ministry of Economic Affairs. transition platforms for this purpose. It plays an In 2004 to 2005, the energy transition process important role in taking initiatives, connecting gained speed through the establishment of four and strengthening initiatives, evaluating existing transition platforms: new gas, green resources, policy and acting upon the policy advice from the chain efficiency and sustainable mobility. In Regieorgaan14 and transition platforms, to stimu- these platforms individuals from the private and late interdepartmental coordination and to make the public sector, academia and civil society the overall transition approach more coherent come together to develop a common ambition for (VROM and EZ 2008, p. 10). particular areas, to develop pathways and to identify useful transition experiments. In 2006, two The front-runners desk, also created in 2004, additional platforms were established, sustaina- presents a second, particularly interesting group ble electricity supply and built environment. The (Table 6.1). It was set up to help companies solve transition path of energy producing greenhouses problems innovatively and to help policy to became a platform of its own in 2008. become more innovation-friendly. Problems varied from difficulties with getting financial support The suggestions of the transition platforms lead from government or private finance to problems to a transition action plan, containing the follow- with getting permits. Between January 2004 and ing goals: March 2006, 69 companies approached the desk to discuss problems. In 59 per cent of the cases, the ◼◼ minus 50 per cent CO2 in 2050 in a growing economy problems were solved thanks to the intervention of ◼◼ an increase in the rate of energy saving to the desk, in 12 per cent of the cases the companies 1.5–2 per cent a year could not be helped, and in the remaining 29 per ◼◼ the energy system getting progressively more cent of cases the desk was still dealing with the sustainable issue at the time of the evaluation. ◼◼ the creation of new business.

14 The Regieorgaan is a committee that is responsible for developing an overall vision for the Netherlands’ energy supply and to formulate a strategic agenda based on inputs from the transition platforms. Its members are the chairs of the seven transition platforms and four independent experts. Green Industrial Policy - Concept, Policies, Country Experiences

Table 6.1: Overview of functions of front-runner desk for innovators and policy 98 Functions for innovators Functions for policy

Make existing instruments more conducive for Obtain financial support from existing instruments innovation

Get into contact with relevant agencies and government Improve policy coordination between ministries and people within ministries

Stimulate case-sensitive implementation of existing Overcoming legal problems and problems with permits and new policy

Widen their network and strengthen the organizational Stimulate policy development in areas of the innovation set up of the innovation trajectory chain not well covered by policy

Business support and public relation help for the Be serviceable to business in a case-sensitive way successful market introduction

Source: Weterings (2006).

In 2011, the transition management was officially In summary, the transition management abandoned as part of an overhaul of innovation approach has many good elements, but also policy by a minister for Economic Affairs who suffered from serious problems. Among the posi- was pro-nuclear and pro-fossil fuels for economic tive lessons are the societal agreement on transi- reasons. Unofficially, however, it continued in the tion targets, the built-in evaluations and learning form of a green deal approach, support for a bio mechanisms, the phased approach and the use refinery plant and a EUR 50 million programme of public-private partnerships in implementa- for zero-energy buildings. For electric mobility, tion. One problem was that the liberalisation of an action plan with a budget of EUR 65 million the energy markets worked against the energy was created (Van der Beeson et al. 2014). In 2013, transition project, leading to policy inconsist- a sustainable energy covenant was signed by ency. A second problem was the large influence employer organizations, worker organizations, of incumbents from the gas and oil sector in the the financial sector, nongovernmental organi- platforms. The support of research and innova- zations and the national government in which tion in sustainable technologies has proved easier the parties agreed to invest and work towards in than the phase-out of non-sustainable technolo- energy efficiency and renewable energy. It is an gies. The politics of this make this a difficult task attempt to involve business and societal organi- for any country. zations in the energy transition.

The effects of transition management are difficult to pin down. The experiences with the front-runner desk are very positive and the business of electric mobility received a big boost. Between 2008 and 2013, the economic value of the e-mobility sector in the Netherlands showed a six-fold increase in gross value added: from EUR 20 million to 120 million by 2013. The number of jobs in this sector increased from 300 to 1,600 full- time equivalents. The effects of the covenant for sustainable energy are unclear as an evaluation is pending, but progress appears to be low, due to the weak instrumentation of the approach. Chapter 6 Developing Green Technologies and Phasing Them In

4. CONCLUSION: KEY ELEMENTS OF A GREEN 99 TRANSITION APPROACH FOR DEVELOPING COUNTRIES A green transition approach for developing coun- Fourth, a sequential approach works well in the tries explores options to manage the transition examples discussed here. This can take the form from the current to a green economy that reflect of gradual tightening of regulations and standards the complexity of societal transformations in or the testing in pilot projects before supporting a the local context. To agree on different paths and broader scale-up. to implement long-term objectives, intermediate steps as well as experimental public-private Fifth, explicitly include policy learning in the processes are useful. A green transition approach phase-in process to achieve socially and econom- highlights the importance of open-ended ically acceptable and successful implementation. collective learning processes, and it is actively These processes are complex and programmes concerned with capacity development at the level may turn out quite differently than planned. of government agencies, knowledge-diffusing Current market and technology developments institutes and business actors. Relying on embed- may also require adjustments over time. The ded autonomy for the formulation and adaptation Chinese and Indian cases have shown that allow- of policies presents a key to successful transi- ing some policy space for strategic bundling of tion management. While this chapter focused on different stakeholder interests at local levels may examples from solar photovoltaics and energy be an important part of this policy learning. efficiency in China, Germany, India and the Neth- Sixth, designing a policy package was helpful in erlands, other green technologies may be as attrac- many of the examples discussed here. This policy tive for a broader green transition approach in a package can include policy-push and market-pull developing country. Regardless of the technology policies as well as research and development, and policy field, seven key elements that increase institutional capacity, skill and job creation meas- the chances for successful phase-in over time, ures, such as building-up a consultancy and certi- based on government support, could be identified. fication industry. First, pro-active planning on the part of govern- Seventh, adequate implementation control mech- ment is required as an initial step. This includes anisms need to be put in place. The building and the development of a long-term vision and a clear financing of technology testing facilities and the roadmap with interim goals and steps. Choices evaluation of implementation schemes are as are informed by policy intelligence about barriers important as fostering consumer awareness for and the effectiveness of policy interventions. an energy efficiency label, for example. Second, this vision and roadmap need to be These seven elements are features of transition communicated early and clearly to investors, management, boiled down and pinpointed for a innovators and other stakeholders to identify green transition in developing countries. They technologies and innovations and to prepare the present no guarantee for the successful develop- producers and consumers. The inclusion of rele- ment and phase-in of green technologies in each vant stakeholders such as manufacturers, busi- and every case. Other factors such as foreign ness associations and standardization bodies direct investments or financial constraints may at an early development stage is advisable as affect phase-in processes. Coupling phase-in with embeddedness without political capture. phase-out by simultaneously reducing support Third, the selection of options for support and the for non-sustainable technologies and increasing forms of support should be conducted carefully, incentives and subsidies for green technologies with the help of independent experts. The chal- is also important (Cosbey at al. 2017, this volume). lenge is to avoid wasteful subsidies and political The green transition approach outlined here gives capture by lobby groups. Auctions, competition starting points for policymakers in developing among service providers and time-bound subsidy countries interested in possible roadmaps for schemes help making support cost-effective. phasing in green technologies. Green Industrial Policy - Concept, Policies, Country Experiences

100 REFERENCES Ambec, S. (2017). Gaining competitive advantage Harrison, T., & Kostka, G. (2014). Balancing Priorities, with green industrial policy. In Altenburg, T., & Aligning Interests. Comparative Political Studies, Assmann, C. (Eds.). (2017). Green Industrial Policy. 47(3), 450–480. Concept, Policies, Country Experiences (pp. IbisWorld. (2015). Energy Efficiency Consultants in 38–49). Geneva, Bonn: UN Environment; German China: Market Research Report. Retrieved from Development Institute / Deutsches Institut für www.ibisworld.com/industry/china/energy-effi- Entwicklungspolitk (DIE). ciency-consultants.html Altenburg, T., & Engelmeier, T. (2013). Boosting solar International Energy Agency (IEA). (2015). Energy investment with limited subsidies: Rent manage- balances per country. Retrieved from www. ment and policy learning in India. Energy Policy, iea.org/statistics/statisticssearch/report/?- 59, 866–874. year=2015&country=CHINA&product=Balances Bureau of Energy Efficiency (BEE). (2014). Energy International Renewable Energy Agency (IRENA). Efficiency in India – Challenges & Lessons. Pres- (2015). Renewable Energy and Jobs. Annual entation by Baskar Serma, In-session Technical Review 2015. Abu Dhabi. Expert Meeting on Energy Efficiency ADP, 13th Jairaj, B., Agarwal, A., Parthasarathy, T., & Martin, March 2014. Bonn. S. (2016). Strengthening Governance of India’s Chang, H.-J. (1996). The political economy of indus- Appliance Efficiency Standards and Labelling trial policy. Basingstoke: Palgrave Macmillan. Program.: Issue Brief. Chaudhary, A., Sagar, A. D., & Mathur, A. (2012). Inno- Johnson, O. (2013). Exploring the effectiveness vating for energy efficiency: A perspective from of local content requirements in promoting India. Innovation and Development, 2(1), 45–66. solar PV manufacturing in India. Discussion China Energy Portal. (2017). 2016 Detailed Electricity paper, Deutsches Institut für Entwicklungspoli- Statistics. Retrieved from chinaenergyportal.org/ tik: Vol. 2013, 11. Bonn: Deutsches Institute für en/2016-detailed-electricity-statistics/ Entwicklungspolitik. Cosbey, A., Wooders, P., Bridle, R., & Casier, L. (2017). Ke, J., Price, L., Ohshita, S., Fridley, D., Khanna, N. Z., In with the good, out with the bad: Phasing out Zhou, N., & Levine, M. (2012). China’s industrial polluting sectors as green industrial policy. In energy consumption trends and impacts of the Altenburg, T., & Assmann, C. (Eds.). (2017). Green Top-1000 Enterprises Energy-Saving Program Industrial Policy. Concept, Policies, Country Expe- and the Ten Key Energy-Saving Projects. Energy riences (pp. 69–86). Geneva, Bonn: UN Environ- Policy, 50, 562–569. ment; German Development Institute / Deutsches Kemp, R. (1994). Technology and the transition to Institut für Entwicklungspolitk (DIE) environmental sustainability. Futures, 26(10), Council on Energy, Environment and Water & Natu- 1023–1046. ral Resources Defense Council. (2012). Laying Kemp, R., & Never, B. (2017). Green transition, indus- the foundations for a bright future : assessing trial policy, and economic development. Oxford progress under phase 1 of India’s national solar Review of Economic Policy, 33(1), 66–84. mission: Interim Report. Delhi. Kemp, R., Rotmans, J., & Loorbach, D. (2007). Assess- El-Shagi, M., Michelsen, C., & Rosenschon, S. (2014). ing the Dutch Energy Transition Policy: How Does Regulation, Innovation and Technology Diffu- it Deal with Dilemmas of Managing Transitions? sion: Evidence from Building Energy Efficiency Journal of Environmental Policy & Planning, Standards in Germany. SSRN Electronic Journal. 9(3–4), 315–331. Advance online publication. Kemp, R., Schot, J., & Hoogma, R. (1998). Regime Evans, P. B. (2012). Embedded Autonomy: States and shifts to sustainability through processes of Industrial Transformation. Princeton paperbacks niche formation: The approach of strategic niche Embedded autonomy. Princeton: Princeton management. Technology Analysis & Strategic University Press. Management, 10(2), 175–198. Geels, F. W. (2004). From sectoral systems of innova- Kemp, R., Rotmans, J., & van Asselt, M. (2001). More tion to socio-technical systems. Research Policy, Evolution than Revolution. Transition Manage- 33(6–7), 897–920. ment in Public. Foresight. (3(1)), 15–31. Goldberg, A., Reinaud, J., & Taylor, R. (2011). Promotion Kemp, R., & Rotmans, J. (2009). Transitioning policy: Systems and Adoption of Energy Management co-production of a new strategic framework for Systems in Industry. Some International Lessons energy innovation policy in the Netherlands. Learned Relevant for China. Policy Sciences, 42(4), 303. Chapter 6 Developing Green Technologies and Phasing Them In

Khan, M. (2004). Strategies for State-Led Social Schimschar, S. (2013). Policy Instruments: The Case Transformation: Rent-Management, Technology of Germany. In F. Pacheco Torgal, M. Mistretta, A. 101 Acquisition and Long-Term Growth. In Asian Kaklauskas, C. G. Granqvist, & L. F. Cabeza (Eds.), Development Bank (Ed.), Which Institutions are Nearly Zero Energy Building Refurbishment Critical to Sustain Market Development, Indus- (pp. 15–60). London: Springer London. trialization and Long-Term Growth in Viet Nam? Schmitz, H., Johnson, O., & Altenburg, T. (2015). Rent Vietnam: Asian Development Bank (pp. 60–73). Management – The Heart of Green Industrial Vietnam: Asian Development Bank. Policy. New Political Economy, 20(6), 812–831. Khandari, R. (2011). Energy Fixing. Down to Earth. Schröder, M., Ekins, P., Power, A., Zulauf, M., & Lowe, Retrieved from www.downtoearth.org.in/ R. (2011). The KFW experience in the reduction coverage/energyfixing-33562 of energy use in and CO2 emissions from build- Kostka, G., & Shin, K. (2013). Energy conservation ings: operation, impacts and lessons for the UK. through energy service companies: Empirical London. analysis from China. Energy Policy, 52, 748–759. Unruh, G. C. (2000). Understanding carbon lock-in. Letschert, V. E. (2014). Energy efficiency appliance Energy Policy, 28(12), 817–830. standards: Where do we stand, how far can we Unruh, G. C. (2002). Escaping carbon lock-in. Energy go and how do we get there? An analysis across Policy, 30(4), 317–325. several economies. Berkeley. Van der Beesen, P., Munnix, S., & Reitsma, S. (2014). Li, Y., Wang, G., & Zhu, C. (2014). Research on the Verzilvering verdienpotentieel Elektrisch Vervoer Development of Energy Management System in Nederland [Cashing the revenue potential Standards System. of electric transport in the Netherlands]. Neth- Liu, F., Meyer, A. S., & Hogan, J. F. (2010). Mainstream- erlands Enterprise Agency / Rijksdienst voor ing Building Energy Efficiency Codes in Develop- Ondernemend Nederland. ing Countries. Washington, D.C.: The World Bank. VROM (Ministry of Housing, Spatial Planning and Loorbach, D. (2010). Transition Management for the Environment) & EZ (Ministry of Economic Sustainable Development: A Prescriptive, Affairs). (2008). Instellingsbesluit van het Regie- Complexity-Based Governance Framework. orgaan Energietransitie Nederland (Implementa- Governance, 23(1), 161–183. tion decision about the advisory and supervision Lütkenhorst, W., Altenburg, T., Pegels, A., & Vidican, group for the energy transition in the Nether- G. (2014). Green industrial policy: Managing trans- lands). Staatscourant, 25 February, no. 39. formation under uncertainty. Discussion paper, Weiß, J., Prahl, A., Neumann, A., Schröder, A., Bett- Deutsches Institut für Entwicklungspolitik: Vol. genhäuser, K., Hermelink, A., Manteuffel, B. von. 2014, 28. Bonn: DIE. (2014). Kommunale Wertschöpfungseffekte durch Markard, J., & Truffer, B. (2008). Technological inno- energetische Gebäudesanierung. Berlin. vation systems and the multi-level perspective: Weterings, R. (2006). Quick scan koplopersloket. Towards an integrated framework. Research Een evaluatie van werkwijze, output en Policy, 37(4), 596–615. effecten. Competentie Centrum Transities, The National Action Plan on Energy Efficiency (NAPE). Netherlands. (2014). National Action Plan on Energy Efficiency. Yang, M., Patiño-Echeverri, D., Yang, F., & Williams, Germany. E. (2015). Industrial energy efficiency in China: Quitzow, R. (2015). Assessing policy strategies for Achievements, challenges and opportunities. the promotion of environmental technologies: A Energy Strategy Reviews, 6, 20–29. review of India’s National Solar Mission. Research Zhou, N., Levine, M. D., & Price, L. (2010). Overview Policy, 44(1), 233–243. of current energy-efficiency policies in China. Rodrik, D. (2014). Green industrial policy. Oxford Energy Policy, 38(11), 6439–6452. Review of Economic Policy, 30(3), 469–491. Sahoo, S. K. (2016). Renewable and sustainable energy reviews solar photovoltaic energy progress in India: A review. Renewable and Sustainable Energy Reviews, 59, 927–939. Green Industrial Policy - Concept, Policies, Country Experiences 102 CHAPTER 7 PRICING ENVIRONMENTAL RESOURCES AND POLLUTANTS AND THE COMPETITIVENESS OF NATIONAL INDUSTRIES

Kai Schlegelmilch, Hans Eichel, Anna Pegels Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

1. INTRODUCTION 103 Pricing the use of environmental resources and supporting green technologies and keeping the the emission of pollutants is one of the best ways costs of adaptation manageable. to signal environmental costs while leaving it to competitive market forces to find the best tech- This chapter explores the various benefits and the nological and organizational solutions. The main rationale of environmental fiscal reform elements pricing instruments are cap-and-trade systems in more depth. It points out how the elements and environmental taxes. These market-based need to be designed, particularly in developing instruments use the main signal in a market countries, to achieve the dual purpose of protect- economy, the price, to indicate the costs of an ing the environment while spurring competi- investment or consumption decision. Leaving out tiveness, industrial development and jobs and that signal would be missing a major opportunity– offsetting potential negative effects on poor men if not the major opportunity–to shift a market and women. economy in the direction of a green economy that Section 2 of this chapter provides definitions and complies with long-term sustainability criteria. explains the rationale as well as design aspects of Environmental taxes are a very crucial element of environmental fiscal reforms. Section 3 discusses green industrial policy. Since these instruments competitiveness issues and presents options to are potentially so effective at redirecting invest- cushion possible negative effects on industries. ment, they may have strong negative effects on Section 4 focuses in on the use of revenues as a the competitiveness of polluting industries. On particularly important design feature. In section the positive side, they increase the competitive- 5, the environmental fiscal reform in Germany is ness of industries that manufacture sustaina- highlighted as a case of green industrial policy bly and offer efficient green technologies and success and a practical example of many of services. As policymakers proceed with green the environmental fiscal reform design options industrial policy design and implementation, it is discussed. Overall conclusions are drawn in important that they anticipate these effects, while section 6.

2. ENVIRONMENTAL FISCAL REFORMS

2.1. DEFINING ENVIRONMENTAL or pollution. The burden of taxes should fall more FISCAL REFORMS on ‘bads’ than ‘goods’ so that appropriate signals are given to consumers and producers” (EEA 2005: Environmental fiscal reform is a political process 84). Environmental tax reform is a subset of envi- aimed at putting a price on environmentally ronmental fiscal reform. harmful behaviour through the use of fiscal reform tools. These tools include a range of environmental taxes, fees and charges, as well 2.2. THE RATIONALE BEHIND as channelling the revenues toward particular ENVIRONMENTAL FISCAL REFORM targets. The Organisation for Economic Co-op- Before discussing effects and design principles, eration and Development (OECD) defines envi- it is important to understand what environ- ronmentally related taxes as “any compulsory, mental fiscal reform instruments are meant to unrequited payment to general government levied accomplish. They are useful to achieve a range of on tax bases deemed to be of particular environ- goals that can be broken down into three broad mental relevance” (OECD 2006: 26). Environmental categories: fiscal reform also covers the removal of supports or subsidies that are environmentally harmful, 1. Benefiting the environment such as tax exemptions for kerosene or subsidies 2. Raising fiscal revenues for coal. 3. Increasing fiscal efficiency and competitiveness Within environmental fiscal reform, the narrower concept of an environmental tax reform is, BENEFITING THE ENVIRONMENT according to the European Environment Agency, a “reform of the national tax system where there is a The most common rationale for environmental shift of the burden of taxation from conventional fiscal reform is its positive environmental impact. taxes, for example on labour, to environmentally This environmental benefit is often referred to damaging activities, such as resource extraction as the first dividend of environmental fiscal Green Industrial Policy - Concept, Policies, Country Experiences

reform (Goulder 1995). Increasing the price of general consumptive or environmental purposes. 104 environmentally harmful products or processes Spending can also finance research and develop- by taxing them discourages their use through a ment for technological innovations or fund infra- market mechanism. This type of government structure required for a green transition. Revenues intervention corrects a market failure if the envi- can furthermore be used to compensate vulnera- ronmental damage of a given action constitutes ble groups, individuals or companies for increased an externality. Negative externalities arise when- prices that can result from the implementation of ever the actions of one party harm another party environmental fiscal reform instruments. and the first party does not bear the full cost of their harmful actions (Gruber 2011). Market actors Practice has shown that the administrative costs receive a distorted price signal because the exter- involved in raising environmental taxes are often nalized costs are not included in the price paid by comparatively low. Environmental taxes, and the actor. Increasing the price of environmentally energy taxation in particular, have the advantage harmful products or processes, to the extent of of being relatively easy to administer and there- fully internalizing the external costs, incentivises fore may prove particularly attractive in countries actors to limit their harmful behaviour. The envi- where tax collection mechanisms are not yet well ronmental fiscal reform intervention thus gives developed. The underlying rationale is that energy market actors a correct, or at least a more accu- taxes, such as on petroleum products, can usually rate, price signal—one that better reflects the full be levied from a very limited number of actors– and thus true costs of their actions. importers, refineries and depots–and are therefore relatively simple to administer and enforce. Even Market actors have free choice whether to curb when energy taxes are collected at the points of their actions or to pay a price, which is very much sale, metering infrastructure is often already in in line with the philosophy of a market economy. place, such as at petrol pumps or electricity meters They can individually adapt their behaviour and (Fay et al. 2015). For example, in Germany the will choose or develop the least cost options to administrative costs of environmental tax reform avoid environmental damage. Theoretically, this are estimated at just 0.13 per cent of the revenues leads to a situation in which a given environmen- raised (OECD 2006). tal objective is achieved at minimal cost. INCREASING FISCAL EFFICIENCY This internalization of external costs is thus a AND COMPETITIVENESS very attractive concept. In practice, estimating the external cost of certain actions can be a complex Environmental fiscal reform can improve the process–even for something as simple as produc- efficiency of fiscal systems by removing distorting, using and disposing of a plastic bag. None- ing effects. Regular taxes, such as payroll taxes, theless, a variety of methods are used to estimate distort markets in a way that makes certain goods various costs imposed by pollution including and services artificially unattractive and creates a costs to health, to productivity, and to the envi- loss of economic efficiency, in this case for labour. ronment. Much progress has been made, allowing Environmental fiscal reform can ease the pres- us to fairly accurately assess the external costs sure on governments to tax labour by providing an in many areas, particularly in transport (Freeman alternative source of revenues, thus opening fiscal 2003; Schwermer 2012a, 2012b; Schwermer et al. space to reduce labour costs, to remove labour 2014; Parry 2014). market distortions and ultimately to increase employment. This is another dividend of environ- RAISING FISCAL REVENUES mental fiscal reform (OECD 2000).

Raising revenues for government spending is the When environmentally harmful behaviour second obvious benefit of environmental fiscal becomes more expensive, market actors search reform. Be it through taxes, fees and charges; for, and tend to find, ways to achieve their goals through the removal of environmentally harmful by changing their production and consumption subsidies; or through auctioning pollution allow- patterns. If revenues from energy taxes on fossil ances: environmental fiscal reform enables govern- fuels are used to reduce taxes on labour, labour ments to collect funds, thus easing the pressure to becomes more attractive relative to production collect funds from other areas. The funds can then factors that rely on the use of energy. Therefore, be spent in various ways–to balance the budget, energy tends to be substituted by labour. Ideally, reduce overall public debts, reduce other taxes, people’s knowledge and engineering capaci- reduce social security contributions or increase ties find innovative ways to use energy more spending. The funds can be channelled towards efficiently and to substitute fossil energy with Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

renewable energy sources. Several studies find compliance, polluters must own permits at least that when structural unemployment15 exists in equal to the quantity of pollution they emitted 105 an economy, environmental tax reform can boost during a given time period. Polluters can emit less employment and profits (Bovenberg and Van der than allowed by their permits, and sell the excess; Ploeg 1998; Holmlund and Kolm 1997; Schöb and or emit more than allowed, and buy permits from Koskela 1996). other participants. In effect, the buyer pays a charge for polluting and the seller gains a reward A high tax on the use of fossil fuels and on related for operating with reduced emissions. emissions makes investments in alternative forms of energy production more attractive, leading to In theory, emissions trading and carbon taxes innovation in these fields. According to the Porter have very similar features; but in practice, carbon Hypothesis, environmental regulation can help and particularly energy taxes are easier and businesses to overcome market failures in inno- quicker to implement. Most countries already vation, thereby allowing them to gain an advan- have tax systems in place, while they would have tage over their competitors from countries without to set up emissions trading schemes from scratch taxes that incentivise innovations (Porter and Van (Pegels 2016). Furthermore, trading schemes often der Linde 1995; Ambec 2017, this volume). Innova- turn out to be ineffective in practice. Since emis- tion, in turn, can lead to job creation in new and sion permit prices are frequently far too low to possibly politically favoured industries. If revenues influence investment decisions, the major objec- are recycled to fund environmental innovation, tive of the trading system is not achieved. Without lower tax rates may be needed to achieve envi- going into detail on the various reasons under- ronmental goals because switching to alternative lying this observation, one conclusion from the technologies becomes cheaper. European experience is that energy and carbon taxes should fill in, at least for the interim time 2.3. HOW ARE THESE RATIONALES until trading systems are delivering as intended. TRANSLATED INTO PRACTICE? This could be achieved using a carbon floor price that ensures at least a minimum price has to be Countries in Europe are at the forefront of design- paid for all emissions. ing and applying environmental fiscal reforms, The Eastern European countries that joined the having started in the late 1980s, intensifying their EU in 2004 have various long-established systems efforts in the 1990s, and further refining these of environmental fees and charges mainly reforms since then. In most cases, energy and addressing air and water pollution, but including carbon dioxide are at the core of environmen- several product taxes. Their approaches provide tal taxation, but they are often complemented useful and inspiring examples for less developed by taxes on transport, waste and land use (Vivid countries as they design and implement their Economics 2012). Most countries used these sustainable development (Schlegelmilch 1999). revenues to reduce labour costs and increase employment, since unemployment was the major When they joined the EU, many of these Eastern political concern at the time of their introduction. European economies were quite weak and had As well, most countries apply a revenue-neutral relatively low administrative capacities; however, approach, thus matching the new environmental they had established comprehensive systems tax with corresponding reductions in other areas. of environmental fees and charges. These fees and charges enabled funding for at least some In addition to environmental taxation, in 2005 environmental infrastructure. While far from European countries decided to introduce a carbon sufficient, they accomplished more than would emissions trading system. In an emissions trad- have been possible otherwise, since official state ing system, governments set an overall limit budgets provided hardly any funding for envi- on emissions and distribute permits or limited ronmental purposes. In effect, these fees and authorizations to emit up to the level of the over- charges applied the polluter pays principle and all limit. The governments may sell the permits; provided incentives to avoid pollution, although but in many existing schemes they grant permits the incentives were not strong enough to produce to regulated polluters according to specific crite- substantial effects (Schlegelmilch 1999). Coun- ria—for example, a baseline derived from each tries that are currently industrialising can explore polluter’s historical emissions. To demonstrate the options of following this model or of directly

15 Structural unemployment exists in an economy in which wages are slow to adjust and labour is relatively immobile between sectors. Changes in demand and in production technology can create structural imbalances in the labour market so that at a given wage, the supply of labour is higher than demand. Green Industrial Policy - Concept, Policies, Country Experiences

leapfrogging to the approach of comprehensive lessons learned from these island states provide 106 environmental fiscal reforms. constructive insights. In many of these countries, energy taxation is the most important element Experience from developing country peers also they use. Several countries direct these reve- can provide valuable guidance. A useful compar- nues to promotion of energy savings, efficiency ative study of several African and Indian Ocean and renewables. These countries face significant island states, for example, provides detailed challenges, including rising sea levels and storms insights into how developing countries can tran- worsened by climate change, while depending on sition from fossil fuels to renewable energies, fossil fuel imports. Moving to energy independ- including through the application of environmen- ence through renewable sources and greater effi- tal fiscal reform elements (Cottrell et al. 2015). The ciencies is a very attractive goal.

3. MANAGING EFFECTS ON COMPETITIVENESS From an environmental perspective, full inter- or are less energy-efficient, while also providing nalization of environmental costs would be the incentives to reduce pollution by the most effi- priority aim of reforms. In practice, however, cient means at their disposal. At the same time, competitiveness and social consequences need the taxes create a competitive advantage for careful consideration, and both aspects are firms with environmentally sound products and important factors in the design of an effective processes. As instruments of green industrial environmental fiscal reform, particularly in the policy, environmental taxes can have positive revenue-raising component of environmental effects on competitiveness–as demonstrated by taxes. Ideally, environmental fiscal reforms can numerous examples: have neutral or even positive effects on competitiveness at firm, sectoral and national levels; and, ◼◼ Environmental taxes can spur innovation in practice, many have been shown to produce when price changes trigger improvements in positive outcomes (Pegels 2016). However, nega- products and production processes (Görlach et tive effects on individual industries are possible. al. 2005). In the following subsections, the potential for ◼◼ Environmental taxes can spur economic positive versus negative effects on competitive- growth as demonstrated for six EU countries ness are discussed first. Then, various options are in a targeted research project report. All six presented that show the potential to reap positive countries achieved an increase in GDP of up and manage negative effects, particularly through to 0.5 per cent compared to a baseline with- the combination of tax exemptions with negoti- out the environmental tax reform measures, while CO emissions decreased. The report ated performance agreements, carbon border 2 adjustments, and international coordination on states, “As a general rule, the effects of the environmental pricing. The subsequent section 4 [environmental tax reform] will be positive discusses how effective design of revenue recy- on economic activity, depending on how the cling components can help environmental fiscal revenues from the environmental taxes are reforms avoid negative social consequences. recycled. However, it is likely that there will be transition costs, so the gains may not be immediate” (COMETR 2007: 41–42). 3.1. POSITIVE OR NEGATIVE EFFECTS? ◼◼ Companies may benefit from reducing costs by increased efficiency, but also from growing When considering how to prevent possible nega- markets for the environmentally sound prod- tive effects on competitiveness, it is important ucts they are either selling, applying, or both. to recall the rationale for environmental fiscal This has been shown for several companies in reform: the implementation of the polluter pays Germany (Knigge and Görlach 2005). principle and the internalization of external environmental damage costs. “By seeking to protect the environment, environmentally related taxation is by definition intended to distort production decisions and have a disproportionate impact on polluters” (OECD 2010: 144). Thus, the very objective of environmental fiscal and, more narrowly, tax reform is to create a competitive disadvantage for those companies that pollute Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

These benefits demonstrate the potential posi- ◼◼ Revenues can be used to counter negative tive effects for the industry. These examples of effects and to support investment in reduced 107 energy tax outcomes show that competitiveness energy use or installation of appropriate concerns usually relate only to a few energy-in- technologies. tensive sectors and are frequently exaggerated, for several reasons (Green Fiscal Commission 2010): Fears about energy price increases expressed by companies can prompt governments to over-com- ◼◼ Fluctuations in energy prices on global pensate industry. During discussions with indus- markets tend to be far more significant than try regarding the impact of such measures, the the effect of a tax on energy. regulator has less information than business—an ◼◼ Not all energy-intensive goods are highly information asymmetry that puts business at an traded internationally; in these cases, increased advantage in negotiations. In some cases, the costs can be passed on to the consumer. cheapest possible energy price is not a top priority ◼◼ An increase in energy prices will incentivise for business (Box 7.1). both energy efficiency measures and innovation, which may result in stable or even falling energy costs for firms over time.

Box 7.1: Factors influencing foreign direct investment in Vietnam A 2015 review of investor sentiment in Vietnam revealed that increased energy prices do not exert a significant influence on the investment decisions of those looking to make foreign direct investments. Instead they are concerned by a lack of skilled human resources and an unreliable electricity supply.

In general, private investment is affected far more by regulatory conditions and the political situation in the country in question. In Vietnam, six basic problems act as a major barrier to investment, according to representatives of government interviewed in the Vietnam Economic Times:

◼◼ No transparency ◼◼ No consistency ◼◼ No synchronization ◼◼ No stability ◼◼ No possibility, an issue which is presumably related to a lack of policy reform in the country ◼◼ No predictability. Neither environmental taxation nor energy prices were mentioned.

Source: Garg et al. (2015); Cottrell et al. (2016).

Although the OECD conducted several studies to negative effects on competitiveness. This could detect the risk of industry relocation as a result be discouraging for the existing energy-efficient, of environmental taxation, there is no substantial resource-light industries already contributing to evidence that it occurs (Arlinghaus 2015). Hence, the economic transformation that green indus- at least in the OECD context, the supposed trade- trial policy pursues; while it also removes incen- off between achieving environmental objectives tives for other industries to become more efficient. and maintaining competitiveness of energy-in- In the UK, for example, exemptions for industry tensive industries does not seem to exist. have resulted in energy-intensive sectors missing the low-hanging fruits of energy efficiency invest- When governments decide to apply protective ments (Cottrell et al. 2017). measures, such as reduced tax rates or even full exemptions, they may combine them with negotiated performance agreements, such as the introduction of an energy management system, and limit their duration. Yet, it is important to take into account that all those exemptions also have Green Industrial Policy - Concept, Policies, Country Experiences

3.2. REDUCED TAX RATES tax rates for industries, while other consumers 108 have to pay the full rates. Sometimes tax rates Reduced tax rates are the most widely applied are reduced for all industries, sometimes only for protective measure for industries and thus are those that are energy-intensive and face interna- a major element of many environmental fiscal tional competition. When used, tax reductions and reforms. Almost all countries provide reduced refunds must avoid overcompensation (Box 7.2).

Box 7.2: Germany’s tax exemptions for industry: Leaving efficiency potentials untapped In Germany, energy-intensive companies in the manufacturing sector receive a refund of 90 per cent of their energy and electricity taxes, even though the Federal Environment Agency has estimated that this subsidy almost completely neutralizes the environmental tax. These exemptions were due to be phased out in 2012, but were extended to 2020. Energy management systems have been made compulsory, as have annual energy savings of 1.3 per cent from 2012 to 2015 and 1.35 per cent for 2016. Annual energy savings between 1991 and 2009 have been well above the planned rates of 1.3 per cent and 1.35 per cent. Thus, these rules and associated subsidies are not compelling anything. Instead large potentials for efficiency and pollution reduction remain untapped in Germany’s energy-intensive industries.

Sources: Andersen and Ekins (2009); Roland Berger (2011); BMWi (2013); UBA (2016).

The United Kingdom has a different problem. The starches, pulp and paper and basic chemicals major concern regarding energy and carbon taxa- (Dröge et al. 2009). However, as mentioned above, tion in the UK is not industry, but the affordability the OECD could not find substantial evidence of of heating for private households. Badly insulated this behaviour (Arlinghaus 2015). houses and a relatively cool climate result in fuel poverty. Policymakers thus shy away from taxing Exemptions and reductions are also unattrac- heating fuels. Hence, the UK applied a climate tive for administrative and political reasons. change levy, which was imposed on industry Each exemption and special treatment of a tax only, not on private households. However, ener- increases its complexity, and therefore its admin- gy-intensive industries also managed to receive istrative costs, and opens the door to rent-seeking substantial reductions to protect their compet- behaviour. Accurately discriminating between itiveness. Consumption-dependent thresholds those who should be eligible for a tax exemption for the reductions led to an unequal distribution and those who are not, based on defined criteria, of the burden of the levy, favouring energy-inten- is complicated. Offering a lower tax rate to indus- sive and mostly large businesses over small and tries that qualify, according to specific indica- medium-sized enterprises. tors of how their international competitiveness is threatened, often compels firms to shift their These examples illustrate how tax reductions behaviour to fulfil these criteria, rather than to produce negative effects. First of all, when the increase efficiencies or reduce pollution. This environmental damage caused by an actor, such distorts markets in unpredictable ways. as energy consumption, is the basis for granting tax reductions, those who damage most receive If reductions are provided nonetheless, they the highest benefits. They will then be the last should at least be combined with compensa- to adapt their behaviour towards protecting the tory measures, such as negotiated agreements, environment. Only proper pricing will ensure concrete investment commitments or, prefera- and promote an efficient allocation of resources bly, mandatory energy management and auditing across sectors over the longer term. The argu- requirements. ment of carbon leakage–industries relocating to low-tax countries–applies only to a small group of industries that have both high energy costs and strong international competition. Among those sectors could be lime, cement, iron, steel, aluminium, refined petroleum, fertilisers and nitrogen, Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

3.3. LINKING REDUCED TAX RATES TO cent. The agreed baseline of 1987, which was NEGOTIATED AGREEMENTS later changed to 1990, allowed industry to benefit 109 from the emission reductions resulting from the Negotiated agreements with industry might German reunification and, more specifically, the also be a way to safeguard environmental action modernisation of Eastern German industry. Due when industries are exempt from environmental to these influences, emissions decreased by about taxes for competitiveness reasons. For example, 15 per cent between 1987 and 1993, thus signifi- industries and their associations can commit cantly reducing the level of ambition of indus- to target-oriented action, forestalling regulatory try’s commitment to decrease emissions in any legislation and other instruments such as envi- substantive sense (Böcher 1996: 193). ronmental taxation. Sometimes referred to as voluntary agreements, that term does not capture Different industrial sectors of Germany’s economy the true nature of most negotiated agreements are still benefiting from the energy and electric- since they usually result from the threat of bind- ity tax reductions in the order of EUR 7.6 billion ing requirements (Héritier and Lehmkuhl 2008). per year, according to the subsidy report of the Government (BMF 2015: 17). As well, industry Reliable evaluations of negotiated agreements are benefits from reduced social security contribu- still scarce, in contrast to other instruments, such tions. The European emissions trading scheme as taxes. This may be due to their relative novelty further supports avoiding cuts while complying in the policy arena. Complicating the analysis with its environmental agreement by providing further, often targets are not quantified or, if they the option of buying permits instead of pursu- are, important parameters such as the baseline ing actual emission reductions by each firm. are unclear. Hence, industry is supported substantially by the Very few negotiated approaches have been tried in combined policy package. combination with environmental taxation. Hence, there are not many conclusions to be drawn. Swit- 3.4. LINKING REDUCED TAX RATES TO zerland had negotiated an environmental agree- ENERGY MANAGEMENT REQUIREMENTS ment regarding heating emissions combined with the taxation of heating fuels. It was agreed When negotiated agreements are not considered and noted in the CO law that if emissions do not 2 binding enough, environmental tax reductions decrease according to fixed interim targets, the can also be linked to mandatory requirements, tax rate would be increased. After the emissions such as installing energy management systems. had not decreased sufficiently in 2014, the govern- They contribute to overcoming the potential ment increased the CO levy by almost 25 per 2 dilemma between providing direction and the cent in 2016 (BAFU 2016). Hence, in the end, this competitiveness of industries. Though tax rates agreement has only postponed implementation of and incentives are reduced, industry takes effective instruments and delayed effective emis- concrete action to reduce emissions. sion reductions and green investments.

Another example occurred in Germany, where The aim of energy management requirements is a negotiated environmental agreement was to support the identification and implementation connected to tax reductions for the manufac- of profitable energy saving potentials. They typi- turing industry. When the tax reductions were cally comprise two major elements: granted, there was no initial specification of ◼◼ The introduction of a comprehensive data substantial and concrete compensatory meas- monitoring system to establish how much and ures that industry should take–only a rather what kind of energy is used when and where vague mention of general industry contributions ◼◼ The realisation of concrete energy-saving to achieving emission reductions. Even this vague potential and thus possible measures, as formulation was not voluntary, but resulted from changes of behaviours and processes or of the threat of otherwise stricter energy taxation. investments. After the introduction of the tax reductions, the German government faced pertinent questions Such a requirement can be considered a no-re- from the European Commission about how to gret option, because it does not put a real burden ensure industry’s contribution to climate targets on the companies, apart from some adminis- despite reduced incentives. This led to the spec- trative costs. It supports companies in estab- ification of the non-binding target in the - nego lishing a data information system on energy tiated agreement to reduce emissions by 20 per flows and greenhouse gas emissions, increasing Green Industrial Policy - Concept, Policies, Country Experiences

transparency. On this basis it is much easier to by endangering their industries. In practice, this 110 identify concrete investment options for improv- would mean that products from countries with a ing energy efficiency and the use of renewa- lower level of environmental taxation would be ble energies. This combination of instruments, charged at the border by an environmental tax of carbon/energy taxation and energy management a level similar to the domestic one. systems, has been applied in the United Kingdom, for example, where positive experiences were A border carbon adjustment could also include noted (Government Digital Service n.d.). relieving exported products from the part of the energy taxation that exceeds the level encoun- Often this instrument combination can be differ- tered abroad. Hence, border carbon adjustments entiated into two further options: offer an effective means of ensuring the competitiveness of energy-intensive industries while still ◼◼ Establishing just the data information flow, incentivizing industry to take ambitious actions while still identifying options for investments towards climate change mitigation. and changes of behaviour in case future steps are anticipated as possible Practical examples include the tax on ozone-de- ◼◼ Exploiting these potential options by imple- pleting chemicals in the US and on primary menting measures in a certain time frame. construction materials in Denmark, called the gravel tax (Internal Revenue Service 2007; Chris- The former option basically comes without added tensen 2011; Wageningen University & Research costs, apart from the administrative efforts, but it 2016). However, border carbon adjustments have brings about benefits in the form of more trans- not been applied more broadly, since most coun- parency on energy and of cost-saving poten- tries shy away from the risk of triggering trade tials. Its application can therefore be generally wars. The targeted country or its industries might recommended. Several countries implemented consider a border carbon adjustment an unfair this option, among them Germany and Denmark measure, since it could be considered disguised (UNIDO 2015). The latter option may involve protectionism if not adequately designed, intro- substantial costs for the companies. Hence, duced, communicated and implemented. the potential design element of requiring the implementation of these identified investment options should be applied with more care. Here, 3.6. INTERNATIONAL COOPERATION AND the investment cycle is likely the most impor- COORDINATION OF CARBON PRICING tant criterion to be taken into account: when International cooperation and coordination is replacement investments are due anyways, a likely to be an effective way of avoiding competi- switch to more efficient machinery results in tive disadvantages. However, there are only a few lower marginal costs. Assuming that the energy practical cases where a minimum of cooperation management requirements take the invest- could be achieved. ment cycle into account and include minimum requirements for profitability, the second option One example is the agreement of September can be recommended. 2016 between the Canadian provinces Ontario and Quebec and extending their cooperation on 3.5. BORDER CARBON ADJUSTMENTS climate change with Mexico. They all agreed to work more closely on carbon tax policy (Clean- Border carbon adjustments offer a trade mecha- tech Canada 2016). Another example is the Euro- nism to counterbalance potentially higher costs pean Union: Since 2004, there has been a common due to domestic environmental taxation, thus framework for energy and carbon taxation in the levelling the playing field between countries EU that requires at least minimum energy tax with higher environmental taxation and those rates on all energy products in all EU Member with lower or none. The aim is to allow coun- States, while any country is free to exceed these tries to be frontrunners and not to punish them rates (EU 2004). Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

4. RECYCLING OF REVENUES 111

As stated earlier, environmental fiscal reforms do invites interest groups to lobby for certain reve- not only set incentives for structural change by nue streams to fund their interests. Finally, it may shifting the competitive positions of industries, prove problematic when earmarked revenues but they have the additional benefit of raising do not match the need of a particular spending government revenues. There is a longstanding project. This was the case for the UK Climate and ongoing debate about the best way to allo- Change Levy revenues that were earmarked to cate these revenues (OECD 2005; World Bank fund reduced national insurance contributions, 2005). However, since revenue raised through but did not raise sufficient revenues to match environmental fiscal reform is equivalent to all the reductions (Cottrell et al. 2013). Although other revenues, the actual issue is how govern- attractive from a communication point of view, ment revenues in general should be spent. It earmarking revenues can be unattractive from a should be clear that there is no straightforward fiscal point of view. Both politicians and finance answer. However, there are some common spend- ministries generally demand the freedom to allo- ing options. cate revenues according to current requirements (World Bank 2005). Probably the most sensible way to recycle environmental revenues is to evaluate them from a political and strategic perspective. One strat- Box 7.3: Earmarking in Chile egy worth considering is to identify the highest national political priority, which may not be Chile’s Centro de Investigation y Plan- related explicitly to the environment, and use ification del Medio Ambiente explored the environmental revenues toward this aim. A options to develop a sustainability fund second approach is to identify winners and losers for the mining sector. The proposed fund from the environmental fiscal reform and use the would support sustainability issues, serve revenues to mobilize powerful stakeholders and to diversify production, integrate mining seize opportunities to build coalitions. Political companies into the community and help and economic leaders who have the perseverance to conserve water and biodiversity. The to lead the process through the inevitable chal- public sector and mining companies would lenges are key to the success of environmental finance it voluntarily. However, it would fiscal reform. Approaching spending choices from be difficult to base the fund on taxes: Any this very practical point of view can be reasonable, change in the tax system that allocates given that spending decisions are fundamentally revenues to a specific purpose requires political in character. Thus, assigning environ- a presidential decree or a change in the mental revenues to policy priorities and strategic constitution because Chile’s constitution uses can be key factors for building reform coali- prohibits it. tions successfully. Source: Schlegelmilch and Joas (2015: 55f.). 4.1. EARMARKING—PROS AND CONS

Earmarking the revenues for particular purposes can be an option to demonstrate transparently 4.2. EARMARKING REVENUES FOR how the revenues are used, which facilitates GREEN INVESTMENTS communication. The advantages and disadvantages of four particular earmarking options invite One way to spend environmental revenues is to analysis: green investments, tax shifting, offset- earmark them for environmental spending. This ting potentially negative effects on competitive- may be intuitive, as taxpayers understand that ness and managing effects on poor people and price increases resulting from environmentally particularly vulnerable groups in society such as harmful behaviour are financing investments women and youth. in environmentally-related public goods. If the goal of an environmental fiscal reform is to reach The concept of earmarking requires some context. certain environmental targets, green spending First, the constitutions of many countries do can help to accelerate the process. If funds are not allow earmarking (Box 7.3). Also, the prac- used in this way, tax rates to achieve a defined tice can complicate budget processes because it environmental target can be lower. The targeted Green Industrial Policy - Concept, Policies, Country Experiences

EU research study produced a model demonstrat- lower distortive taxes such as taxes on labour 112 ing that emission reductions could be achieved, is an attractive option to maximise economic along with substantially lower tax rates, if 10 per welfare, but requires the identification of the most cent of revenues were invested in energy effi- distortive taxes. This is likely to vary according to ciency measures alone (COMETR 2007). country conditions.

Spending options can, for example, include subsi- One aspect to keep in mind, however, is the need dies for research and development on clean to balance market efficiency with social desir- technological solutions. They can also include ability. For example, Takeda (2007) finds that directing funds into green infrastructure like strong double dividend effects arise when reve- public transport, smart grids, buildings or renew- nues are recycled by lowering taxes on capital able energy systems. This way of pricing positive but not on labour or consumption. While this externalities is much in line with the spirit of may be economically correct, it could result in industrial policies and is often much appreciated socially undesirable outcomes, such as more by industry since it is an incentive rather than a inequity and injustice within the tax system, as punishment. Nonetheless, it must be noted that well as in political risks and destabilization. This connecting environmental revenues to outright might result, for instance, in a case where an subsidies for certain groups may encourage environmental tax was raised from a large group rent-seeking behaviour among the recipients of actors and the revenue used to lower taxes on (UNEP 2010). While strategic use of revenues capital, which would benefit only a small group can increase the likelihood of the fiscal reform’s that is likely well off already . continuation, without good management, recipients might grow dependent on payments. An example for tax shifting is the South African 2009 budget in which the South African Reve- nue Service proposed a tax reduction worth ZAR 4.3. ENVIRONMENTAL TAX SHIFT (South African Rand) 13,500 million to personal income taxes and another ZAR 1,000 million to Recycling revenues to lower other distorting business taxes (Speck 2010). Had the proposal taxes in the economy is a common use of envi- been implemented, most of the tax cuts would ronmental revenues. The revenues from Germa- have been compensated by increasing indirect ny’s environmental tax reform were used mostly taxes, mainly fuel and electricity taxes. These to lower pension payments that were increasing measures would have contributed to the progres- labour costs and thus contributing to unemploy- siveness of the tax system, thereby reducing ment. Recycling revenues to lower distorting inequity. The overall effect would have been taxes is often described as leading to one of the a loss of revenues worth ZAR 4,575 million dividends of environmental fiscal reform, since (Figure 7.1). Environmental fiscal reforms are not this step creates wealth by removing existing necessarily revenue neutral–they can also lead to distortions, such as tax-induced unemployment a net increase or decrease of revenues. (Jaeger 2012). Using environmental revenues to Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

Figure 7.1: Financing income tax cuts through environmental fiscal reform in South Africa: The revenue effects of various reform elements 113

Adjustment in monetary threshold

Adjust personal income structure

Industrial policy

Increase in general fuel levy

Electricity tax

Incandescent light bulb levy

Air passenger departure tax

Plastic bag levy

Diamond export levy

Increase in excise duties on tobacco and alcohol

Final Budget Effect

-15,000 -10,000 -5,000 0 5,000 -15,000 revenue effect in (in million ZAR)

Source: Adapted from Speck (2010).

4.4. OFFSETTING NEGATIVE IMPACTS 4.5. OFFSETTING NEGATIVE IMPACTS ON ON COMPETITIVENESS THE POOR

Some revenues could be earmarked for mitigating Protecting the poor is a prerequisite for any envi- or reducing potentially negative effects of environmental fiscal reform, particularly in developing ronmental taxes on the competitiveness of some countries. At the same time, it is important not to industries, by reducing other taxes such as corpo- undermine incentives to reduce energy and water rate or labour taxation or by providing targeted consumption (OECD 2005). Differentiated pricing grants to co-finance investments in energy effi- for different types of users can be an effective ciency. This can be a way to address competitive- option, as targeted cash transfers can be. Since ness concerns regarding firms in other countries this chapter focuses on environmental fiscal that do not face similar taxes. Firms that have reform as a green industrial policy, inclusive- already taken steps to reduce pollution should ness is not discussed in much detail. Nonethe- be preferred, rather than simply returning taxes less, a practical example of Indonesia’s Bantuan paid to polluter and pollution reducer alike (OECD Langsung Tunai is useful to demonstrate the 2005). Furthermore, it is also important that potential for cash transfer programmes in raising compensations are designed in a way that does a population’s acceptance level for environmental not encourage firms to produce more emissions fiscal reform (Box 7.4). (World Bank 2005). Revenue recycling might be an effective way to encourage industry acceptance of reforms. In China, for example, a combination of pollution taxes and support for pollution abatement expenditures has proven to be effective (Barde et al. 2009). Sweden applied a charge on nitrogen oxide emissions, but it recycled revenues back to the same sector, utilities, on the basis of each company’s electricity generation (OECD 2013). This way, it favoured clean and efficient companies while protecting competitiveness. Green Industrial Policy - Concept, Policies, Country Experiences

114 Box 7.4: Raising acceptance for environmental fiscal reform through cash transfer programmes: The Indonesian Bantuan Langsung Tunai In the aftermath of the Asian financial crisis in the late 1990s and early 2000s, Indonesia initiated attempts to reduce fossil fuel subsidies, which had neared 20 per cent of total domestic revenues. The savings were initially recycled through general spending, specifically health-care and education. Since the economic crisis and a regime of corruption, collusion and nepotism had already built social discontent, the subsidy reform steps triggered violent demonstrations. A hike in global oil prices challenged reform implementation even further. In 2005, the Indone- sian government nonetheless removed subsidies for industrial users and raised gasoline and kerosene prices by more than 150 per cent within one year. Despite the immense increase of fuel prices, opposition against the reform was relatively moderate. This can be explained by the simultaneous introduction of the compensation programme Bantuan Langsun Tunai. The programme was a targeted and well-communicated, unconditional cash transfer to compensate poor households. All households with monthly fuel expenditures below the threshold of IDR (Indonesian Rupiah) 175,000 received monthly payments of IDR 100,000 over six months. Twen- ty-eight per cent of all Indonesian households received those payments. Between 2005 and 2006, compensatory spending for the programme amounted to EUR 1.93 billion and made up more than 50 per cent of the saved spending from subsidy cuts in 2005. Nonetheless, the subsidy reform contributed substantively to reducing the state budget deficit.

Sources: Beaton and Lontoh (2010), Widjaja (2008), World Bank (2012).

The Indonesian case illustrates that while envi- use and other environmentally relevant activi- ronmental fiscal reforms maintain their economic ties. Hence, there is hardly any need to introduce and environmental appeal, policymakers need new taxes, since many of the existing taxes could to be prepared to recycle large parts of revenues simply be increased. These increases would alle- to those who cannot easily carry the burden viate the necessity for payments from the federal of higher prices. In another study, Schlegelm- to the local level. This, in turn, would free funds ilch (2011) analyzed conditions and suggested on the federal level, which could be used to lower concrete steps for Indonesia to continue envi- federal taxes, for example on income, or invest in ronmental fiscal reforms. The suggestions build targeted programmes to reduce poverty. mainly on the good tax bases at Indonesian local levels, which are often based on natural resource

5. GERMANY’S ENVIRONMENTAL FISCAL REFORM: A GREEN INDUSTRIAL POLICY SUCCESS STORY

The implementation of environmental fiscal reform ◼◼ Reduced tax rates were granted for energy-in- in Germany can be considered a green industrial tensive companies and the manufacturing policy success story. The effort was undertaken industry. against the background of high unemployment, the ◼◼ Reduced tax rates were granted for several top national priority at the time of implementation, environmentally-sound activities. and the need to reduce greenhouse gas emissions. ◼◼ Minor parts of the revenues were used for promoting renewable energies and energy effi- The reform was implemented between 1999 and ciency in the buildings sector. 2003 and consisted of the following main elements: ◼◼ Overall, the environmental fiscal reform was almost revenue neutral. ◼◼ Transport and heating fuel taxes were increased. ◼◼ An electricity tax was introduced in 1999 and As settled in the coalition agreement of 2002, the increased between 1999 and 2003. environmental fiscal reform was evaluated in ◼◼ Social security contributions were reduced. 2004 in a series of studies to assess if and how it should be continued. Model-based estimates of Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

its outcomes came to the conclusion that positive The major barrier, perceived negative effects on effects could be expected and a double dividend some industries, was overcome by reducing more 115 was likely to be achieved in comparison to a refer- distorting labour taxation such as social secu- ence scenario without the environmental fiscal rity contributions to the pensions fund, but also reform (Kohlhaas 2005; Bach et al. 2001). These by providing reduced rates for major industries. positive effects include: While opposition may have persisted in some parts of industry for strategic reasons, that is, to safe- ◼ Reduction of CO emissions by about 2–3 per ◼ 2 guard eligibility for exemptions, the prospect of a cent and a lasting reduction of fuel sales double dividend ensured a broad alliance in soci- ◼◼ Creation of up to 250,000 additional jobs, assum- ety including unions and parts of industry. Even ing that the employment gains would not be when world oil price surged and led to protests, the accompanied by higher wages German government stood firm since reducing ◼◼ A slightly positive effect on GDP. unemployment was its top priority at the time, and a major means was the reduction of labour costs. While causation can be difficult to prove, other Hence, linking revenue use to the national top studies found interesting correlations (BMF 2004: priority turned out to be very effective in overcom- 35–43; BMU 2004: 16–18; Schlegelmilch 2005: 10–12; ing barriers and maintaining the environmental Bundesverband CarSharing 2016): fiscal reform. ◼◼ After decades of decrease, the number of From a political point of view, it was also very passengers using public transport started important to rally the winners of the reform: Often increasing from 1999 and was about 7–8 per enterprises did not know they were benefiting, but cent higher in 2006. perceived themselves as losers before calculating ◼◼ The number of members of car sharing organ- their profits and finding out about the actual results. izations has been steadily growing since 1999, Some companies did not dare speak out in public sometimes with two-digit percentage numbers, in favour of the environmental fiscal reform to and has reached more than 1.2 million drivers avoid positioning themselves against the indus- in 2016. try mainstream, which was communicated by the ◼◼ The manufacturers of solar water heaters simi- large industry associations. The risk of opposing larly experienced strong growth—originating the mainstream was often perceived higher than not only in the higher heating taxes, but also in the potential benefits of further steps of an environ- a funding programme providing grants financed mental fiscal reform.16 by the environmental tax revenue. Several years after the start of the reform, the The sharply increasing crude oil world price at European Commission increased pressure on the period of analysis–from about US$ 18.00 per the German government to require a solid return barrel in 1998 to about US$ 39.00 in 2000–certainly in exchange for the state aid to energy-intensive played a role in causing some of these effects, and industries in the form of the reduced tax rates. the individual influences can be challenging to Therefore, the government made the tax reduc- isolate. But undoubtedly, the various efficiency tions subject to the introduction of an energy gains, cost-savings, tax reductions, market expan- management system. In 2011, a new government sions, and innovations stimulated by environ- formed by conservatives and liberals implemented mental fiscal reforms improved the competitive further important, though smaller, environmen- position of many German companies (Knigge and tal fiscal reform elements. A study by the Hein- Görlach 2005). However, industry associations rich-Böll-Foundation analysed and recommended mainly stressed negative effects. Meyer et al. (2001) several elements, of which the following were provide an enlightening analysis of the discourse implemented (Ludewig et al. 2010): in the wake of the tax reform implementation, when industry associations and well-organized ◼◼ Ticket fees on air transport were introduced, energy-intensive enterprises acted as strong and differentiated according to the length of flights. vocal vetoes. ◼◼ A tax on nuclear fuel was introduced to increase utilities’ share in total costs of nuclear power. Another reason for opposition may be the empha- ◼ The heavy goods vehicle toll, introduced in sis of early discussions and the first draft of ◼ 2005, was extended to cover roads other than the reform law on potential negative effects for motorways. broader parts of industries. Due to these discussions the draft was adapted to its current form.

16 Schlegelmilch (2005), personal communication. Green Industrial Policy - Concept, Policies, Country Experiences

◼◼ The industrial exemptions from the energy tax fiscal reform as such has reached a cross-party 116 were reduced. consensus since its introduction. This is a major ◼◼ A financial transaction tax was adopted. political success, not least since it ensures investment certainty. These amendments show that, as sensitive as individual reform elements may be, the environmental

6. CONCLUSIONS Environmental fiscal reforms are a core element competitiveness at the border. However, practical of green industrial policy. They contribute to the application is scarce. The few experiences gained relative profitability of green investments and from the case of the Danish primary construction they are easy to administer while having substan- materials/gravel tax and the US ozone-depleting tive revenue potential, particularly in the form of chemicals tax should be further evaluated and energy taxes. fed into an attempt by the European Commis- sion to establish a similar system (Internal Reve- Revenues can then be used for such national nue Service 2007, Christensen 2011, Wageningen priorities as poverty reduction, investments in University & Research 2016). In fact, the European health, green infrastructure and others. Another Union would be particularly suited to implement option is to promote green industries by using border carbon adjustments, since it is one of the revenues to foster research, development, and largest global markets, has excellent implemen- deployment of clean technologies—an example of tation capacities, and relatively strong political supporting positive effects instead of penalizing will to establish environmental protection meas- negative effects. From a political economy point ures. Carbon border adjustments would help to of view, revenue recycling can help to build broad overcome most concerns regarding weakened alliances and overcome political resistance. competitiveness of energy-intensive industries that may currently exist due to the relatively high Competitiveness aspects can be managed by level of energy prices in the EU. designing tax schemes accordingly. When environmental taxes are applied, enterprises exposed Linking reduced tax rates to requirements for to international competition are eligible for energy management is also hardly used, although reduced tax rates in most countries. However, it could be applied at short notice and eventually if potentials for energy saving and renewable benefit industry. It is in most cases a no-regret energy are high and efficiencies are low, it may option, because it does not put real burden on be reasonable to apply taxes without reduced industry, apart from adoption costs, but can mobi- rates. Industry will then be compelled to tap lize substantial savings. the efficiency potentials, thus reducing costs and increasing competitiveness. Increased effi- In overall conclusion, environmental fiscal reform ciency can, on the national level, contribute to an can affect the competitiveness of industry both improved trade balance and less dependence on positively or negatively. But even when effects are fossil or other polluting energy imports. negative, they tend to be small, and various design options exist to further minimise these effects or An ideal option to protect domestic industries compensate the affected industries. Many design would be to enhance international cooperation options have not yet been exploited, or even and coordination on the introduction of carbon explored, to their full potential. From a political and energy pricing. However, international coor- point of view, rallying the beneficiaries of reforms dination on this issue is not very developed yet. is just as important as negotiating with the losers. Many countries have introduced or are consid- Often there is a lack of awareness of the benefits ering some form of carbon pricing and recent they are reaping and, even when aware, stake- years have seen the emergence of a number of holders may not be encouraged to become more coordinating initiatives such as the Carbon Pric- outspoken about the positive effects they expe- ing Leadership Coalition (World Bank et al. 2016). rience. Understanding the nature of the bene- Nonetheless, there is still large variation in the fits and communicating the findings will help global prevalence and levels of carbon prices. to build coalitions for change, as will linking the Until global coordination is more developed, reform revenues to national priority aims, such as border carbon adjustments could be an option employment or poverty reduction. to protect domestic industries by levelling their Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

REFERENCES 117 Ambec, S. (2017). Gaining competitive advantage Bundesministerium für Wirtschaft und Energie with green industrial policy. In Altenburg, T., & (BMWi) (2013). Spitzenausgleich-Effizienzsystem- Assmann, C. (Eds.). (2017). Green Industrial Policy. verordnung (SpaEfV) – Ein Baustein im Rahmen Concept, Policies, Country Experiences (pp. der Neujustierung des sog. „Spitzenausgleichs“. 38–49). Geneva, Bonn: UN Environment; German Berlin. Development Institute / Deutsches Institut für Bundesverband CarSharing. (2016) CarShar- Entwicklungspolitk (DIE). ing-Jahresbilanz 2015: Wachstum und Konsolidi- Andersen, M. S., & Ekins, P. (2009). Carbon-energy erung im deutschen CarSharing-Markt. taxation: Lessons from Europe. Oxford: Oxford Christensen, T. (2011). Border adjustment regulations University Press. in the Danish Gravel tax. Arlinghaus, J. (2015). Impacts of Carbon Prices on Cleantech Canada. (2016). Ontario, Quebec working Indicators of Competitiveness. OECD Envi- with Mexico on carbon tax. Retrieved from www. ronment Working Papers: Vol. 87. Paris: OECD canadianbiomassmagazine.ca/news/ontario- Publishing. quebec-sign-carbon-tax-deal-with-mexico-5866 Bach, S., Bork, C., Kohlhaas, M., Lutz, C., Meyer, B., COMETR. (2007). Competitiveness Effects of Envi- Praetorius, B., & Welsch, H. (2001). Die ökologis- ronmental Tax Reforms: Report to the European che Steuerreform in Deutschland: Eine modell- Commission, DG Research and DG Taxation and gestützte Analyse ihrer Wirkungen auf Wirtschaft Customs Union. und Umwelt; mit 88 Tabellen. Heidelberg: Cottrell, J., Bridle, R., Yongqiang, Z., Jingli, S., Xuxuan, Physica-Verl. X., Beaton, C., Leopold, A., Meyer, E., Sharma, Barde, J.-P., Andersen, M. S., & Schlegelmilch, K. S., Cheng, H. (2013). Green Revenues for Green (2009). Introducing Environmentally Related Energy: Environmental fiscal reform for renew- Taxes: A Window of Opportunity for China. In able energy technology deployment in China. Economic Instruments for Energy Efficiency and Winnipeg. the Environment (pp. 257–304): CCICED. Cottrell, J., Fortier, F., & Schlegelmilch, K. (2015). Fossil Beaton, C., & Lontoh, L. (2010). Lessons Learned from Fuel to Renewable Energy: Comparator Study of Indonesia’s Attempts to Reform Fossil Fuel Subsi- Subsidy Reforms and Energy Transitions in Afri- dies. Winnipeg. can and Indian Ocean Island States. Incheon. Böcher, M. (1996). Konzepte für eine ökologische Cottrell, J., Schlegelmilch, K., Runkel, M., & Mahler, A. Steuerreform. Diskussionsprozeß und umweltpo- (2016). Environmental tax reform in developing, litische Interessenstrukturen: Diplomarbeit an der emerging and transition economies: Studies 93. Philipps-Universität Marburg. Hamburg. Bonn. Bovenberg, A. L., & van der Ploeg, F. (1998). Is unem- Cottrell, J., Damian, L., Runkel, M., Schlegelmilch, K., ployment good for the environment? Environ- & Zerzawy, F. (2017). Environmental Tax Reform in mental and Resource Economics, 12(2), 137–150. Asia and the Pacific. Bangkok. Bundesamt für Umwelt (BAFU). (2016). Reduktionziel Dröge S., van Asselt, H., Brewer, T., Grubb, M., Ismer, R., 2014 nicht erreicht: CO2-Abgabe auf Brennstoffe Kameyama, Y., Mehling, M., Monjon, S., Neuhoff wird 2016 erhöht. Retrieved from www.admin.ch/ K., Quirion, P., Schumacher, K., Mohr, L., Suwala, W., gov/de/start/dokumentation/medienmitteilun- Takamura, Y., Voituriez, T., Wang, X. (2009). Tack- gen.msg-id-58016.html ling leakage in a world of unequal carbon prices. Bundesministerium der Finanzen (BMF). (2004). European Environment Agency (EEA). (2005). Monatsbericht März: Fünf Jahre ökologische Market-based instruments for environmen- Steuerreform. Berlin. tal policy in Europe (EEA technical report). Bundesministerium der Finanzen (BMF). (2015). Copenhagen. Subsidy Report. Berlin. European Union. (2004). EU energy tax directive. Bundesministerium für Umwelt, Naturschutz und Fay, M., Hallegatte, S., Vogt-Schilb, A., Rozenberg, J., Reaktorsicherheit (BMU) (2004). Die ökologis- Narloch, U., & Kerr, T. (2015). Decarbonizing Devel- che Steuerreform: Einstieg, Fortführung und opment: Three Steps to a Zero-Carbon Future. Fortentwicklung zur Ökologischen Finanzre- Climate Change and Development. Washington, form (Stand: Februar 2004). Bundesministerium DC. für Umwelt, Naturschutz und Reaktorsicher- Freeman, A. M. (2003). The measurement of envi- heit: Art.-Nr. 2201. Bonn: Bundesministerium für ronmental and resource values: Theory and Umwelt, Naturschutz und Reaktorsicherheit, methods (2nd ed.). Washington, DC: Resources for Referat Öffentlichkeitsarbeit. the Future. Green Industrial Policy - Concept, Policies, Country Experiences

Garg, V., Bridle, R., & Clarke, K. (2015). Energy pricing, OECD. (2005). Environmental Fiscal Reform for 118 energy supply and FDI competitiveness in Viet Poverty Reduction. DAC Guidelines and Refer- Nam: An assessment of foreign investor senti- ence Series. Paris: OECD Publishing. ment. Winnipeg. OECD. (2006). The political economy of environmen- Görlach, B., Knigge, M., & Lückge, H. (2005). Wirkun- tally related taxes. Paris. gen der ökologischen Steuerreform auf Inno- OECD. (2010). Taxation, Innovation and the Environ- vation und Marktdurchdringung. UBA-FB: Vol. ment (Chinese version). 00,858,5. Berlin: Ecologic. OECD. (2013). The Swedish Tax on Nitrogen Oxide Goulder, L. H. (1995). Environmental taxation and the Emissions: Lessons in Environmental Policy double dividend: A reader’s guide. International Reform (OECD Environment Policy Papers). Tax and Public Finance, 2(2), 157–183. Parry, I. W. H. (2014). Designing Fiscal Policy to Government Digital Service. (n.d.). Environ- Address the External Costs of Energy (CESIFO mental taxes, reliefs and schemes for Working Paper No. 5128). businesses. Retrieved from www.gov.uk/ Pegels, A. (2016). Taxing Carbon as an Instrument of green-taxes-and-reliefs/climate-change-levy Green Industrial Policy in Developing Countries. Green Fiscal Commission. (2010). Competitiveness Bonn. and environmental tax reform (Briefing Paper No. Porter, M., & van der Linde, C. (1995). Green and 7). London. competitive: Ending the stalemate. Harvard Busi- rd Gruber, J. (2011). Public finance and public policy (3 ness Review. (73), 120–134. ed.). New York: Worth Publishers. Roland Berger. (2011). Effizienzsteigerung in stro- Héritier, A., & Lehmkuhl, D. (2008). The Shadow of mintensiven Industrien. Ausblick und Hand- Hierarchy and New Modes of Governance. Jour- lungsstrategien bis 2050. Munich. nal of Public Policy, 28(01), 1–17. Schlegelmilch, K. (1999). Green Budget Reform in Holmlund, B., & Kolm, A. S. (2000). Environmental tax Europe: countries at the forefront: with 20 figures reform in a small open economy with structural and 142 tables. Springer Science & Business unemployment. International Tax and Public Media. Finance, 7(3), 315–333. Schlegelmilch, K. (2005). Insights in Political Internal Revenue Service. (2007). Ozone Deplet- Processes on the Ecological Tax Reform from a ing Chemicals (ODC) Excise Tax Audit Ministerial Perspective. Munich. Techniques Guide. Schlegelmilch, K. (2011). Rapid Assessment on the Jaeger, W. (2012). The double dividend debate. In J. readiness of Indonesia towards an Environmen- E. Milne & M. S. Andersen (Eds.), Handbook of tal Fiscal Reform for greening the economy: Final research on environmental taxation (pp. 211–230). draft 15/12/2011. Munich. Cheltenham, UK: Edward Elgar Publishing. Schlegelmilch, K., & Joas, A. (2015). Fiscal Considera- Knigge, M., & Görlach, B. (2005). Auswirkungen der tions in the Design of Green Tax Reforms. Green ökologischen Steuerreform auf Unternehmen. Growth Knowledge Platform Research Commit- Quantifizierung der Effekte der ökologischen tee on Fiscal Instruments (Working Paper No. Steuerreform auf Umwelt, Beschäftigung und 03/2015). Innovation: Vol. 4. Berlin: Ecologic. Schöb, R., & Koskela, E. (1996). Alleviating Unem- Kohlhaas, M. (2005). Gesamtwirtschaftliche Effekte ployment: The Case for Green Tax Reforms (CES der ökologischen Steuerreform. Quantifizierung Working Paper No. 106). Munich. der Effekte der ökologischen Steuerreform auf Schwermer, S. (2012a). Economic Valuation of Envi- Umwelt, Beschäftigung und Innovation: Vol. 2. ronmental Damage Methodological Conven- Berlin: DIW, Abteilung Staat. tion 2.0 for estimates of environmental costs. Ludewig, D., Meyer, B., & Schlegelmilch, K. (2010). Dessau-Roßlau. Pricing Carbon and Cutting Energy Subsidies to Schwermer, S. (2012b). Methodological Convention reduce the financial deficit in Germany. Washing- 2.0 for Estimates of Environmental Costs: Annex ton, DC. A. Dessau-Roßlau. Meyer, T., Schicha, C., & Brosda, C. (2001). Diskurs-In- Schwermer, S., Preiss, P., & Müller, W. (2014). Method- szenierungen: Zur Struktur politischer Vermit- ological Convention 2.0 for Estimates of Environ- tlungsprozesse am Beispiel der “ökologischen mental Costs: Annex B. Dessau-Roßlau. Steuerreform”. Wiesbaden: Westdeutscher Verlag. Speck, S. (2010). Options for promoting Environmen- Milne, J. E., & Andersen, M. S. (Eds.). (2012). Handbook tal Fiscal Reform in EC Development Cooperation of research on environmental taxation (Paper- - Case Study South Africa. back edition). Cheltenham, UK: Edward Elgar Publishing. Takeda, S. (2007). The double dividend from carbon regulations in Japan. Journal of the Japanese OECD. (2000). Greening Tax Mixes in OECD Countries: and International Economies, 21(3), 336–364. A Preliminary Assessment. Paris. Chapter 7 Pricing Environmental Resources and Pollutants and The Competitiveness of National Industries

Umweltbundesamt (UBA). (2017?). Umweltschädli- che Subventionen in Deutschland (aktualisierte 119 Ausgabe 2016). Für Mensch & Umwelt / Umwelt- bundesamt. Dessau-Roßlau, Dessau-Roßlau: Umweltbundesamt, Fachgebiet I 1.4. United Nations Environment Programme (UNEP). (2010). Green Economy - Driving a Green Economy Through Public Finance and Fiscal Policy Reform. United Nations Industrial Development Organiza- tion (UNIDO). (2015). What are the steps to adopt Energy Management Systems? Energy Manage- ment Systems – ISO 50001 Energy Efficiency in Emerging Economies (E4) Training Week. Paris. Vivid Economics. (2012). Carbon taxation and fiscal consolidation: the potential of carbon pricing to reduce Europe’s fiscal deficits, report prepared for the European Climate Foundation and Green Budget Europe. Munich. Wageningen University & Research. (2016). Large differences in greenhouse horticulture energy taxes in North-West Europe. Wageningen. Retrieved from www.wur.nl/en/show/Large-differences-in-greenhouse-horticulture-energy-taxes-in-North-West-Europe.htm Widjaja, M. (2008). An Economic and Social Review on Indonesian Direct Cash Transfer Program to Poor Families 2005. Jakarta. World Bank. (2005). Environmental Fiscal Reform - What Should Be Done and How to Achieve It. Washington, DC. World Bank. (2012). BLT Temporary Unconditional Cash Transfer - Social Assistance Program and Public Expenditure Review 2. Jakarta. World Bank, Ecofys, Vivid Economics. (2016). State and Trends of Carbon Pricing 2016. Washington, DC. Green Industrial Policy - Concept, Policies, Country Experiences 120 CHAPTER 8 PROMOTING CIRCULAR ECONOMIES

Verena Balke, Steve Evans, Liazzat Rabbiosi, Sandra Averous Monnery Chapter 8 Promoting Circular Economies

1. INTRODUCTION 121 In 2017, the Earth Overshoot Day fell on August 2. resources are efficiently used and materials can It marks the date when human demand for envi- be reused or recycled at their highest possible ronmental resources and services exceeds what value, reducing waste–and keeping the extrac- nature can regenerate within that calendar year. tion of new resources–to a minimum. For this to August 2 is the earliest date since calculations happen, a product has to be managed differently, started in 1970 and means that the productivity throughout its whole life cycle, starting from the of not one, but 1.7 Earths would be needed to meet way it is designed and how its raw materials are humanity’s global demand sustainably (Earth extracted and covering manufacture, transporta- Overshoot Day 2017). tion and consumption, as well as how thoroughly it is recycled at the end (Wilts 2016). A circular In 2010, global material use amounted to 79.4 economy can reduce unsustainable exploitation billion tons (Schandl et al. 2016). Under a busi- of natural resources while increasing human ness-as-usual scenario, estimates suggest this well-being and economic wealth. annual consumption will rise to 180 billion tons by 2050 (Dittrich et al. 2012). This increase The structure of this chapter is as follows: The results from the predominance of a linear second section explains the concept of a circu- model of resource consumption in which ever lar economy in more detail. The third section new resources are extracted, used as inputs for provides some practical examples where the production processes and then discarded (Ellen principle of closed cycles has been applied at MacArthur Foundation 2013). With this economic various scales. Section 4 introduces a variety of model and its inbuilt accelerations, humanity is benefits that come with the implementation of a overstepping Earth’s bio-capacity boundary. circular economy. Section 5 explains measures to overcome barriers and encourage the transition To stay within planetary boundaries, economic towards a circular economy. Section 6 concludes. growth has to be decoupled from resource use Throughout this chapter, examples from compa- and environmental degradation (Rockström et nies and from countries at various levels of al. 2009). This can be achieved by a transition income and human development are used to illus- from the linear model of economic consumption trate how theory translates into reality. and production to circular economies in which

2. WHAT IS A CIRCULAR ECONOMY? In a circular economy, production and consump- be retained in a closed loop for as long as possi- tion systems are designed in a way that products ble, thereby creating new value (Ellen MacArthur and waste materials are reused and recycled Foundation 2016a). The energy powering this within the production and consumption system system should come from renewable sources, not (International Resource Panel n.d.). It builds on from fossil fuels (Ellen MacArthur Foundation reduce, reuse, recycle (3Rs) and explores oppor- 2016b). Non-toxic natural material that cannot tunities to promote closed material loops and be reused eventually returns to nature by being high levels of resource efficiency in a system- composted. Processed, non-toxic materials, such wide approach. The 3Rs serve as a framework for as glass, steel and plastics, are retained in the resource efficiency: Manufactured products need system and generate further value beyond the to be designed for long-term use and for reuse, time when they would have been discarded or which will encourage repair, refurbishment, and scrapped in the linear model (Evans et al. 2014). remanufacturing. Lastly, at the end of a product’s To be able to reuse most of the processed mate- lifetime, the input material has to be recycled and rial, products are designed in ways that their input returned to the industrial process (Jørgensen 2015). materials can be recovered at high quality levels With this approach, waste in production, supply, using minimal energy (Ellen MacArthur Founda- use and disposal is minimised, and a maximum tion 2016b). proportion of the original resource is recovered.

The circular economy is commonly described as a restorative and regenerative model that seeks to keep products, materials and resources at their highest value and utility at all times so they can Green Industrial Policy - Concept, Policies, Country Experiences

122

Raw materials Des ign

R e P m r a o g n d n u u i l f c c a t i Resid y c o t ua c n l u w e as r t R i e n Circular economy g

n o i t C u o ib l r le t c s t i io D n

n Consumptio air use, reuse, rep

Figure 8.1: Conceptual diagram of a circular economy17 (Adapted from European Commission 2014)

The circular economy model takes the full life increase in groundwater extraction. When more cycle of a product into account (Figure 8.1). Since efficient irrigation systems came into use, farm- some small amount of residual waste cannot ers shifted to crops that are water intensive and be avoided, new resources always need to be high yielding and they changed their cultivation added to the production process to some extent. patterns. In this case, greater irrigation efficiency However, the aim of the circular economy is aggravated the region’s problem of groundwater to keep this residual waste as small as possible depletion (Pfeiffer and Li 2014). and to rely on recycled resources for the major share of economic processes (European Commis- To ensure that natural capital is preserved as sion 2014). Therefore, it supports the objective of much as possible, that resources yields are opti- decoupling the economy from the use of non-re- mised and that system risks are minimised, the newable energy and of pristine resources. Decou- transition to a circular economy requires a whole pling refers to removing the links leading from systemic change reaching throughout value economic growth to resource use and to negative chains. Technological innovation is only part of environmental effects (International Resource the required transformation: innovative solutions Panel n.d.). In fact, most economies are now in organizational thinking, civil society partici- using fewer resources to produce one unit of GDP pation, financing methods and government poli- (UNEP 2016). This is a partial success because it cies are essential. The innovations necessary for extends the timeframe until economies reach, or a circular economy need to cover waste-min- further overstep, their environmental boundaries. imizing designs, new business models and Yet such relative decoupling can be negated by new recycling technologies as well as changes economic growth and increasing consumption in consumer behaviour. For example, product (Wijkman and Skånberg 2015). This case, where design should ensure that products have a longer resource consumption increases despite effi- lifespan and are easy to maintain. Also, products ciency gains, is referred to as the rebound effect. can be designed in a modular way so that broken components can be replaced and repair is easy. A good example of this rebound effect can be When technological change is fast, it should be illustrated by irrigation practices. Pfeiffer and possible to upgrade products by changing only Lin (2014) found that increases in the efficiency outdated components. At the end of the life of the of irrigation techniques, implemented in Kansas product, it should be easy to disassemble it and to between 1995 and 2005, have led to a significant reuse or recycle its parts (Bocken et al. 2016).

17 This is an illustrative example of the concepts which have been developed to visualise the circular economy. Chapter 8 Promoting Circular Economies

The circular economy is based on two important and remove waste as a category. Thus, it goes premises: at the company level, a closed loop busi- beyond simple efficiency gains and encompasses 123 ness model that reaches through a company’s the underlying consumption and production value chain; and at the societal level, a re-design patterns of society. As such, it requires holistic life of the entire system to use resources efficiently cycle systems thinking.

3. SHIFTING FROM LINEAR INDUSTRIAL SYSTEMS TO CIRCULAR ECONOMIES So far, no nation has made much progress ISKENDERUN BAY IN TURKEY: INDUSTRIAL towards achieving a circular economy. Yet, quite SYMBIOSIS AT THE REGIONAL LEVEL a number of local or sectoral experiments have The Iskenderun Bay Industrial Symbiosis been undertaken to explore advancing towards Programme established an industrial symbiosis circular economic implementation. These network in the Iskenderun Bay region of Turkey. include closed loop industrial systems with lines The programme is part of a wider national indus- of production designed to minimise waste and trial symbiosis strategy and aims at increasing with outputs from one production line used as collaboration between regional companies to inputs for other processes. As individual firms achieve environmental and economic improve- are rarely sufficiently diversified to reuse the ments (International Synergies Ltd. 2014). The entirety of their outputs within the company, implementation phase lasted from 2011 to 2014. firms can cluster in geographic proximity to make Within the industrial symbiosis network, 51 the exchange of materials easier. The concept of member companies from 28 different sectors industrial symbiosis describes networks of firms work together to benefit from synergies (Alkaya that share nontoxic by-products as resources to et al. 2014). gain mutual eco-innovative benefits while reducing harmful influences on the environment and An orange juice producer uses the waste heat of society (Tao 2017). In such networks, one company a lime producer to dry orange juice waste. As a can use an output of another company that would result, 12,000 tons of waste pulp no longer go otherwise have been wasted as an input to its to landfills, but are turned into 1,400 tonnes of own production process. Firms with such mutu- animal feed for reselling instead (International ally complementary productions may co-invest Synergies Ltd. 2014). A cooperation project of in eco-industrial parks. While less sophisticated seven companies discovered that lint, a waste versions of eco-industrial parks only share facil- product from cottonseed production, could be ities for pollution control, more sophisticated used for the bioremediation of petroleum-polluted parks build on industrial symbiosis (UNIDO soils. Analyses found that apart from its strong 2016). Denmark’s industrial symbiosis complex at absorptive capacity, the product also prevents Kalundborg is widely recognised as a best prac- petroleum from polluting the groundwater. tice case and its environmental and economic Currently, a bio-products company is working on benefits are well documented (Jacobsen 2006). the commercialisation of the new product, which is expected to conserve 6,500 m3 of water per year What is even more important, however, is the (Alkaya et al. 2014). application of closed cycles at the sector-wide or economy-wide level. Entire subsectors of the Full-scale implementation of ten symbiotic coop- economy, such as the automotive or the consumer erative systems will lead to an estimated reduc- electronics industry, should be designed to tion in CO2 emission of 36,700 tons annually and maximise value creation from existing resource the redirection of 327,250 tons of solid waste stocks and minimise waste. The following four from landfills to various forms of reuse. In finan- practical examples illustrate these points. While cial terms, annual savings of US$ 6.37 million the first three exemplify closed loop indus- balance against total programme investments trial systems, the example of a Korean national of nearly US$ 7 million, meaning the programme programme for reducing food waste in South will pay off in just over a year (Alkaya et al. 2014). Korea represents a more encompassing approach The success of the Iskenderun Bay Symbiosis integrating public and private actors on the Programme has led to the integration of Indus- production as well as the consumption side. trial Symbiosis into the 2014 to 2023 plans of 19 regional development agencies, some of which Green Industrial Policy - Concept, Policies, Country Experiences

have already been initiated. Subsequently, indus- ISL-TEDA industrial symbiosis project. The initi- 124 trial symbiosis was also included in the Ministry ative for this project was derived from China’s of Science, Industry and Technology’s National commitment to reduce its 2005 carbon intensity Efficiency Policy (Alkaya et al. 2014). by 40 to 45 per cent by 2020 (Tao 2017; Watts 2009). The local government in Tianjin responded to this YIXING ECONOMIC AND TECHNOLOGICAL commitment with a project to build an industrial DEVELOPMENT ZONE IN CHINA: INDUSTRIAL symbiosis match-up platform. Using this platform, SYMBIOSIS AMONG 1,200 COMPANIES factories can match their needs for input material with others’ waste in the surrounding area. This The Yixing Economic and Technological Devel- approach is replicating the UK’s National Indus- opment Zone was established in 2006 to support trial Symbiosis Programme (Wang et al. 2015). the industrial development of green technologies (Chen 2017). While optoelectronic materials, The ISL-TEDA project achieved 99 synergies for photovoltaics and advanced equipment manu- industrial symbiosis, exceeding the project target facturing are the leading industries, many types of 80 matches (Tao 2017). It started with the local of industries are located in the zone, including government’s strong ambition to foster indus- textile, mechanical and chemical manufacturers. trial symbiosis. The government took the lead More than 1,200 companies now form a complex and local private companies followed by imple- industrial symbiosis network with extensive menting concrete action. It shows that private cooperation (Li and Shi 2015). companies are willing to cooperate with governments when they can see economic benefits The electric industry uses liquid argon, a by-prod- such as increased value added for their products, uct of nitrogen fertilizer production. The textile and social benefits such as improved reputation, industry relies on acetic acid from citric acid and being long-listed for government procure- production. The by-product of an inorganic chem- ment contracts. It illustrates the importance of a ical company, sodium nitrite, is used to produce systemic approach from different actors to shift organic compounds. Concrete, cement and other towards circularity. construction products are produced with fly ash and gypsum from coal burning. The thermal A NATIONAL PROGRAMME FOR REDUCING FOOD power plant uses dried sludge from the textile WASTE IN SOUTH KOREA industry. Machinery components are produced with iron and steel scraps collected within the To tackle a significant food waste problem in zone (Li and Shi 2015). South Korea, several government ministries including the Ministry for Food, Agriculture, With the flourishing of industrial symbiosis came Forestry and Fisheries and the Ministry for Health, environmental problems. Due to water intensive Welfare and Family Affairs have put in place and polluting industries, water quality became a coordinated and comprehensive policy mix. a severe concern in the area, so stringent water The overall objective was to change food culture emission standards and strict governmental among the population, reduce waste and improve supervision were put in place. Two water treat- recycling. The measures include a National ment plants now recycle used water. In 2009, the Masterplan on Food Waste Reduction to set the total rate of water reuse was more than 90 per overall framework for implementation, a recy- cent. The power and chemical industries use cling programme requiring the collection of food the most water by a significant margin, but recy- waste in residential areas and catering industries, cle the majority of their discharge. However, the a landfill ban for food waste as well as voluntary textile industry only recycled 10 per cent of its agreements with the catering sector to encourage water, suggesting there is substantial need for reduction measures in food waste and loss. For analysis and innovative solutions (Li 2012). example, restaurants started offering eco-friendly COLLABORATION BETWEEN LOCAL GOVERNMENT menus and cut down the number of available AND THE PRIVATE SECTOR FOR INDUSTRIAL small side-dishes. Cafeterias in public institutions SYMBIOSIS IN TIANJIN, CHINA organized a no-leftover-day once a week. Another interesting approach was the introduction of a The UK’s National Industrial Symbiosis volume-based food waste fee system. It requires Programme, working through non-profit Inter- households to pay a fee based on the amount of national Synergies, Ltd. (ISL), and the Tianjin food waste they generate measured on a scale at Economic-Technological Development Area the bottom of the waste bin. Monthly data serve (TEDA) Administrative Commission in China have as a basis for charging fees to households (Kim established the government-led collaborative 2002). To promote recycling, the government has Chapter 8 Promoting Circular Economies

financed the development of public recycling food waste has risen from 2 per cent in 1995 to 95 facilities (Government of South Korea 2013). This per cent in 2009 (Kim et al. 2013). Remaining food 125 mix of food waste reduction measures has had waste is now being turned into compost and live- a great impact on raising consumer awareness stock feed, as well as biomass and biofuels. and on the environment. The recycling rate of

4. WHAT ARE THE SOCIAL AND ECONOMIC CO-BENEFITS OF GOING CIRCULAR? The benefits of the circular economy are not only every year and are sorted, cut, cleaned, pelletized limited to improving the environment and related and recycled by the informal sector to be returned health problems. There are also compelling, to the economy (WBCSD 2016). Prospects for the immediate social and economic benefits for such circular economy in India are promising: efforts a transition. These include employment gains and for a circular economy in construction, food and macro-economics. agriculture, as well as transport remanufacturing, could contribute US$ 624 billion by 2050, equiv- EMPLOYMENT EFFECTS alent to 30 per cent of India’s current GDP (Ellen MacArthur Foundation 2016c). The Ellen MacArthur Foundation (2013) estimates that in the EU manufacturing sector alone, net In the course of establishing better governmental material costs of up to US$ 630 billion per year recycling systems in developing countries, on the could be saved in a circular economy. Apart one hand, the informal waste picker community from significant environmental benefits, the has to be taken into account. To fight poverty, it is social gains from increased employment are essential not to destroy the livelihood of a large also substantial. In the UK manufacturing sector, number of poor people, especially women, who remanufacturing–the rebuilding of outdated or represent the majority of waste pickers. On the broken products with updated components and other hand, aspects of pollution, workers’ rights, modules–could potentially create an additional health and safety and the elimination of child 310,000 to 320,000 jobs (Next Manufacturing Revo- labour must be improved significantly. There- lution 2013). fore, public policy frameworks should be inclusive, clearly defining the role and contribution of Waste collection, sorting and recycling create waste pickers and their organizations, ensuring substantial sources of income for unskilled work- their access to protection, such as health and ers (Ellen MacArthur Foundation 2013). Some social programmes, and supporting safe prac- regions in industrialised countries and many tices. In Buenos Aires, Argentina, waste pickers developing countries do not yet have well-or- were recognised as specialised waste manage- ganized waste management infrastructure. In ment service providers and provided with offi- such conditions, a largely informal industry often cial certificates (Gutberlet et al. 2017). This shows provides jobs for waste scavengers, who search how governments can intervene to recognise for recyclables in urban settlements and landfills and certify jobs in newly developing areas of the that they can refurbish and resell. In Latin Amer- circular economy. ica between 500,000 and 4 million people work in this informal sector (Marello and Helwege 2014). MACRO-ECONOMIC EFFECTS The absence of centralized waste treatment has led to entrepreneurial activity. In some cities Investing in circular economies reduces the for example, waste pickers hold contracts with import bill of resource-importing countries and municipalities for their services, and waste picker can soften price shocks. In many resource-poor cooperatives have developed complex business countries, natural resources–particularly refined operations including the use of mechanisation for oil and coal–account for a large share of their some tasks (Marello and Helwege 2014). Although imports. Decreasing their domestic consumption the sector itself is still predominantly informal, 76 thus frees up substantial resources. Moreover, per cent of waste pickers in Africa, Asia and Latin world market prices of natural resource are often America sell mainly to formal businesses, proving characterised by enormous price fluctuations that there is demand for recycled goods (WIEGO that increase the importer’s vulnerability. When 2014). In India, approximately 4.7 million tons of China cut back the supply of aluminium in 2017, plastic bypass the public waste collection system world market prices skyrocketed to a six-year Green Industrial Policy - Concept, Policies, Country Experiences

high threatening the existence of aluminium-pro- circular economies aim to reduce their resource 126 cessing industries in import-dependent countries imports. That could have negative short-term (Guoping et al. 2017). Collecting and processing consequences for resource-exporting develop- waste aluminium, from electronics for instance, ing nations, by lowering their expected revenues could greatly reduce the dependence on alumin- from export of raw resources (De Jong et al. 2016). ium imports (Wilts 2016). Those exporting countries should develop mitigation strategies and diversify their economies away A possible repercussion involves effects on from resource extraction and export. resource-exporting countries: Countries currently importing extracted resources that move towards

5. HOW TO OVERCOME BARRIERS AND PROMOTE CIRCULAR ECONOMIES Despite the many exciting experiments and sector-specific targets based on well-defined potential benefits, figures for global produc- indicators. Japan, for example, uses three main tion and consumption show that no economy indicators to measure its progress towards circu- has achieved full circularity. Some promising larity: an indicator of resource productivity which achievements are worth mentioning, including measures material use as a share of total GDP; a exemplary regulations, such as the European cyclical usage indicator, quantifying the share of Union’s Ecodesign Directive and some compre- material reused of total material consumed by the hensive national recycling systems, such as economy; and an output indicator, which specifies Germany’s Green Dot system (EU 2009; Baughan how much waste is directed to landfills (Benton and Evale 2004). Yet overall, global production is and Hazell 2015). The targets derived from these still largely organized according to the unsustain- indicators then have to be further divided to give able linear resource-to-waste logic, and lacks an companies an objective for change. At the same organized, systemic approach to involve relevant time, this enables direct comparison of differ- actors to shift to circularity. What, then, is hold- ent companies and governmental programmes ing the circular economy back? This section looks (Anbumozhi et al. 2016). The EU has mapped out at the barriers and what policymakers can do to an action plan for the circular economy that is promote circular economies. supported by a concrete list of measures, divided by stages in product life cycle, with clear time- SETTING GOALS AND DEFINING POLICIES FOR THE lines. Located within the different stages are TRANSITION TO CIRCULARITY sub-strategies for the various industries and stakeholders involved in change (European Parlia- To achieve circularity at an economy-wide scale ment 2016). These examples highlight the impor- and make it the norm—the new business-as- tance of a multi-stakeholder approach including usual—the concept needs active government private and public sectors of the society. effort and strong coordination across various stakeholders as well as within industry itself. Measures to achieve the set objectives should Many governments have shown great leadership ideally include a coherent mix of policies. Provid- in setting out ambitions for green growth. Their ing information, creating standards and labels next challenge is implementation: learning how that make the material footprint of product and to achieve it in practical detail. processes transparent, educating consumers and encouraging voluntary initiatives can all The transition to circularity needs to be recog- help reduce wasteful resource consumption and nised as a key principle of green industrial policy environmental pollution. Yet massive change is and mainstreamed in all relevant sector policies. unlikely to happen unless producers and consum- An essential step is to have stakeholders–from ers have strong economic incentives to adopt different levels of government; the various play- circularity. The policy mix will critically depend ers from the private sector, including investors on the use of economic instruments, such as taxes, and social entrepreneurs; and especially from economic incentives and mandatory regulations, civil society’s cross generational and marginal- that can be combined in various ways. These are ized members–agree on overall national goals all important components of fiscal policy for envi- concerning resource use and pollution. These ronmental fiscal reform (Schlegelmilch et al. 2017, goals should be ambitious and broken down into this volume). Chapter 8 Promoting Circular Economies

NATIONAL REGULATIONS and Extended Producer Liability Law (Government of Chile 2016). 127 On a national level, extended producer responsibility is the prime concept to guide industry An extended producer responsibility approach is towards more circularity, because it incentivises both an incentive for sustainable product design producers to improve their products to increase and for waste-to-resource solutions; but in itself circularity. Additionally, waste collection charges, it is not sufficient to make economies circular. environmental taxes and mandatory deposit-re- Additional green industrial policies are needed, fund systems can be used to steer behaviour and such as subsidies and tax-relief, enforcement of discourage unwanted practices. fines and punishment for illegal landfill- dump ing, infrastructure development for the collec- The extended producer responsibility principle tion of secondary material and sustainable public was first implemented in Germany with - pack procurement. aging legislation in 1992, then adopted by the European Union in 2008. Also in 2008, it was intro- Waste collection charges and environmen- duced in Korea with the Act on Resource Recy- tal taxes, for example on landfill dumping and cling of Electrical and Electronic Equipment and incineration, can help to reduce waste and at the Vehicles (Ghisellini et al. 2016). Extended producer same time provide funds for better recycling and responsibility means that the costs of disposal treatment systems. In Russia, regulation used and recovery of input materials should be borne to limit the revenue generation from recycling. by producers at the end of a product’s lifetime. This prevented the emergence of a market for Companies can then either take products back or the treatment of municipal solid waste and led pay fees that cover the cost of recycling or waste to treatment rates of only 10 per cent (IFC 2012). management. Fees should vary, according to the Revised legislation that no longer limits the reve- treatment necessary for the input materials to be nue generation from recycling has been imple- recovered. This would provide an incentive for mented to encourage newcomers to enter the producers to improve their product design (Wilts market and increase transparency and compe- 2016). If companies use more resistant materi- tition (WasteTech 2017). In India, waste manage- als to make products reliable and durable, their ment is hampered by insufficient budgets for payments for product disposal would be reduced municipal authorities, which cannot afford to (Bocken et al. 2016). Advanced disposal fees would implement proper recycling systems (Kumar et al. also motivate producers to make their products 2017). Waste collection charges and environmen- repairable, so that the product is not immediately tal taxes could be used to improve the budget of returned to the producer once it is damaged. environmental agencies and give them financial leeway for changes towards circularity. Governments can also introduce standards to make the market more transparent. They can For reusable and recyclable packaging, deposit-re- request producers to provide information regard- fund systems can be used to improve collection ing repair and maintenance, and technical data rates (European Environmental Bureau n.d.). In for ease of repair should be standardised (RREUSE the US, many states have implemented a depos- 2015). This would also encourage competition it-refund program for lead-acid batteries. A US$ 10 and reduce prices for repair works. If companies deposit for any battery sold has resulted in recy- have to pay for disposal and input recovery, they cling rates of 97 per cent (Walls 2011). will also be motivated to enhance recyclability. Fewer input materials will be used and the INTERNATIONAL REGULATIONS ones integrated will have a lower environmental impact, as hazardous materials are more diffi- International regulation is also vital, because cult to be returned to the production process and resources and waste are traded across borders, should therefore entail a higher fee (Clark et al. sometimes illegally. For electronic waste alone, 2009). In France, producers of vacuum cleaners between 60 and 90 per cent is estimated to be with toxic flame retardant plastics pay a fee 20 handled informally or unregistered. E-waste per cent higher than for vacuum cleaners made worth US$ 18.8 billion per year is illegally traded from non-hazardous materials (European Envi- across borders and eventually dumped or unsafely ronmental Bureau n.d.). Emerging countries are recycled (Rucevska et al. 2015). Regional treaties also moving forward with the implementation of and international agreements, such as the Basel extended producer responsibility. In 2016, Chile Convention on the Control of Transboundary became the first country in Latin America to Movements of Hazardous Wastes and the Bamako implement legislation by adopting the Recycling Convention on the Ban of the Import into Africa Green Industrial Policy - Concept, Policies, Country Experiences

and the Control of Transboundary Movement UK, the profit-driven factories tend to send waste 128 and Management of Hazardous Wastes within directly to landfill, with little motivation to change Africa, provide frameworks on how to tackle (Tao 2017). Here, economic incentives could assign these cross-border issues. They aim at limit- value to recycled resources and motivate compa- ing the export of toxic waste and therefore have nies, particularly recyclers, to set up facilities direct implications for circular economies. These for recycling and establish a valuable business. conventions, however, provide some exemptions In providing these incentives, the government that have been abused to bypass regulation at should ensure that all actors in the value chain considerable scale (Khan 2016). Hence, additional can benefit from them, to create shared value out national regulations and monitoring are often of waste when it is treated as a resource. required to close existing loopholes. Under the Basel Convention, non-functioning electronic Foster life cycle thinking: being unfamiliar with devices which need to be repaired, refurbished life cycle thinking tools reduces opportunities to or upgraded can be defined as used electrical and have more viable solutions for resource efficiency. electronic equipment intended for direct reuse. The application of a life cycle approach and This category is exempted from the hazardous perspective is helpful to keep a systems perspec- waste definition and can be traded across borders. tive and have scientifically informed policy deci- National legislation complementing these agree- sions. It can be supported by tools, such as Life ments and treaties is necessary to provide further Cycle Assessment or Life Cycle Costing that definitions and standards. In addition to being integrate environmental costs in the evaluation party to the Basel Convention, countries should of barriers and support the effective choice and define what kind of material can be exported or use of policy instruments. A life cycle approach imported (Khan 2016). Regional regulation can can support policymakers in identifying the key be used to reap synergies and to avoid bureau- issues that need solutions and also in identifying cratic costs. In African states, national legislation whether specific policy proposals have any unin- complementing the Basel Convention is set to be tended consequences in other parts of the system. implemented on a regional level. At the Pan-Af- The approach displays existing trade-offs, and rican Forum on E-waste in 2012, twenty African maps key stakeholders along the value chain to countries pledged unified action for regional enable better decisions and choices. A life cycle regulation. They agreed to cooperate on the or systems perspective helps to target all parts management of transboundary movements and of the value chains of products such as product the environmentally sound treatment of elec- design; potential for repairing, refurbishing, recon- tronic waste (Pan-African Forum on E-waste ditioning, remanufacturing; resources extraction; 2012). Better regional regulation is also needed in transport and infrastructure needs, including the European Union where no uniform standard waste and recycling infrastructure, where many exists to define environmentally relevant charac- economic opportunities lie. teristics of recycled materials. The national rules Investment and finance availability: informing that have to be followed instead lead to additional entrepreneurs and small and medium enterprises bureaucratic costs (Barneveld et al. 2016). about investors and potential partners who are SUPPORT NEW BUSINESS MODELS willing to cooperatively start new business lines to close the cycle, partners that may be difficult To achieve a fully functioning circular economy, to find. Establishment of specific credit lines current business models need to change signif- for companies that want to innovate towards icantly. Most companies still base their sales sustainability will be essential to spur new busi- model on increasing the number of products sold, ness models, especially those that require tech- regardless of their material consumption. From a nology investment. sustainability perspective, more companies need Two particular emerging service trends illustrate to shift to new business models based on collect- how new business models can enable the tran- ing, reusing, recycling, refurbishing or sharing sition to a circular economy. The first is reverse products. New and innovative business models logistics, with the goal of maximising recov- that aim at closing the circle are underway using ery of material used in production. In cooper- a variety of tools. ation with manufacturing companies, reverse Price signals: it is difficult to motivate factories to logistics facilitates the flow of products at the change their way of dumping waste, such as into end of their lifetime from customers back to landfills, when dumping does not cost much. For the industry. Received products then have to be example, because of the extremely low landfill disassembled. Their input materials can be recov- fee in China, about 7 per cent of the cost in the ered, or repaired to be sold in second markets. Chapter 8 Promoting Circular Economies

Alternatively, the components can be remanu- PROMOTE REGIONAL COOPERATION AND factured to re-enter the first market or to be sold SPATIAL CLUSTERING 129 to second markets (Govindan et al. 2015). The To promote regional cooperation and spatial clus- lack of fiscal and economic incentives in Brazil tering, governments can assign designated areas hampered implementation of reverse logistics for eco-industrial parks. Policies and regulation under the National Solid Waste Policy. However, on e-waste management, pollution prevention, once reverse logistics became an important factor energy saving and emission reduction as well in competitiveness, companies started to invest as environmental taxes and fees on waste and in the necessary procedures (Finkler et al. 2017). pristine raw material can motivate companies to The second emerging service is product sharing. locate in eco-industrial parks and invest in indus- Instead of selling a product, a resource owner trial symbiosis networks. Additionally, guidelines provides a customer with a product or service, for green infrastructure development and cleaner while charging a transaction cost. Sharing production promotion can serve as support decreases the volume of products that have to be (UNIDO 2016). manufactured to supply demand and increases For designing eco-industrial parks and promot- the intensity of that one shared product’s utility, ing industrial symbiosis, a high level of coordi- so that it can be used to its full technical potential nation is necessary to ensure that companies (Daunorienė et al. 2015). Experience from Korea with complementary resource requirements shows that governmental support is important invest and that companies throughout the supply for such schemes, especially in the early phases chain receive a constant flow of the necessary of a sharing economy. Since 2012, Seoul’s Metro- inputs (Rahman et al. 2016). This is challenging, politan Government has pursued implementation not least because information about companies’ of its Sharing City Initiative. Sharing programmes waste and by-products is not common knowledge offer cars, books, music, tools, knowledge and and is treated with great secrecy (Fichtner et al. skills, education, lodging and meeting facili- 2005). In some countries, the success of eco-in- ties–all accessed through online platforms that dustrial parks has been hampered by a lack of help arrange matches. The initiative is designed specific knowledge and coordination capacity to provide public services to citizens, especially of public authorities. For example, the Andhra those who are economically and socially disad- Pradesh Special Economic Zone, India’s first vantaged, as the managers believe that the shar- planned eco-industrial park, has attracted much ing economy can be instrumental in solving less investment than anticipated, because many various urban problems. To trigger the diffusion companies withdrew their proposed plans for and increase the number of participants, the establishing a unit within in the park (UNIDO 2016). Metropolitan Government distributed subsidies of US$ 500,000 to initiatives led by district govern- Still, government agencies can fund material flow ments and promoted public-social partnerships. studies to identify potential industrial symbi- In the course of the initiative, new business osis projects. Under the Eco-Industrial Park models were fostered by creating a legal frame- Programme in China, administrative committees work in favour of the sharing economy to promote oversee the development and management of socially and environmentally beneficial activities eco-industrial parks. They give recommendations (Moon 2017). on how material and energy flows among tenants can be streamlined and which companies could The underlying idea of using products without be recruited to complement material flows while having to own them may involve a challenge for ensuring sufficient integrity of environmental service providers: They have to make the initial infrastructures. Subsidies for small and medi- investment in goods, but the flow of income starts um-sized enterprises can make eco-industrial later once usage fees are paid. To overcome this parks more diverse (Teh et al. 2014). obstacle, financial instruments, such as green bonds, may be encouraged to enable circular CONSUMER INVOLVEMENT business models (European Energy Agency 2014). Moreover, environmental fiscal reforms that shift Civil society must be involved to achieve a the tax burden from labour to resource consump- circular economy. Recycling and waste reduction and to pollution help to increase the compet- tion starts in the home and in schools, as does itiveness of circular business models that employ learning about sustainability choices. Consumer a large labour force (Wilts 2016; Schlegelmilch et choices play an essential role in the circular al. 2017, this volume). economy, because they determine the quality and quantity of goods and services as well as Green Industrial Policy - Concept, Policies, Country Experiences

the way those goods and services are produced. public procurement, environmental taxes, regula- 130 Goods and services that are produced in an envi- tions, bans and restrictions that expedite transi- ronmentally sound and sustainable manner can tion to a circular economy. be very competitive (Ambec 2017, this volume). Consumers are not driven by price alone, and Educational initiatives combined with product distorted price signals that ignore the cost of sustainability information further enable better environmental damage reduce consumers’ oppor- consumption choices (Anastasio 2016). When tunity to obtain sustainable products and services. designing educational programmes, policymak- Taxes and fees can be used to influence consumer ers need to address the gap between knowledge behaviour. Reward schemes for returned items and action and to explore smart means of moti- and deposit-refund systems can motivate vation: Even consumers with pro-environment consumers to return products at the end of their attitudes do not always engage in behaviour that life to the producer or to a recycling facility (Forlin is environmentally responsible. The reason for and Scholz 2017). this gap is that consumer behaviour can be influenced by rather irrational tendencies, such as Many citizens, in addition, want to be informed cognitive biases and force of habit (Frederiks et about the origins and sustainability of products, al. 2015). Also, convenience makes a big difference services and processes. Information tools such for consumer decisions. Research has found that as product sustainability information, awareness as convenience is improved in the design of recy- campaigns, or corporate sustainability reporting cling systems, participation rates increase (Samu- help consumers make informed choices. The elsen and Støyle 2016). efforts behind information-based instruments– environmental data collection and dissemina- Increasing numbers of people, as citizens and tion, development of indicators, environmental consumers, are aware of the benefits of closed valuation, energy audits, education and train- cycle, circular economies. Green consuming, eco-labelling or certification schemes, erism organizations are emerging in many public disclosure of enterprises’ sustainabil- countries and communities demanding more ity performance–not only provide information durable, well-designed, sustainable products and and knowledge to the end consumer. They also processes. Policymakers can support such move- complement and strengthen the effectiveness ments as important drivers of change towards of other policy instruments, such as sustainable circular economies.

6. CONCLUSION To stay within planetary boundaries and to So far, no single country has become fully circular. make today’s industrial production sustaina- That means that no known pathway to achieving ble, economies inevitably have to become circu- a fully circular economy exists. Although devel- lar. This chapter has shown that change comes oping countries can benefit from some of the with substantial benefits of increased economic experience in developed economies, such as the stability, natural resource conservation, inclusion technologies and implementation of successful of socially disadvantaged groups, job creation waste collection and recycling systems, measures and greener growth opportunities for all coun- to promote circularity are far more comprehen- tries. It has also become clear that to promote sive and must be tailored to realities in individual circular economies, no single instrument can communities and countries. Any governmental achieve the goal. Instead, a combination of coher- plan for promoting the circular economy should ent instruments is needed to fully optimise the include well-defined targets for each stakeholder environmental, social and economic benefits of group based on clear, measurable indicators and circular economies. This policy mix would typi- the mix of policy instruments selected should be cally include regulations and financial incentives coherent and socially inclusive. Any policy imple- to promote circular product design, set up waste mentation will be successful only if there is align- collection systems, support new business models ment, coherence and stakeholder coordination. and encourage change in consumer behaviour, among others. It requires a broad consensus among various actors, including the realignment of incentives. Chapter 8 Promoting Circular Economies

REFERENCES 131 Ambec, S. (2017). Gaining competitive advantage De Jong, S., van der Gaast, M., Kraak, J., Bergema, with green industrial policy. In Altenburg, T., & R., & Usanov, A. (2016). The circular economy Assmann, C. (Eds.). (2017). Green Industrial Policy. and developing countries: a data analysis of the Concept, Policies, Country Experiences (pp. impact of a circular economy on resource-de- 38–49). Geneva, Bonn: UN Environment; German pendent developing nations. The Hague Centre Development Institute (DIE). for Strategic Studies. Alkaya, E., Böğürcü, M., & Ulutaş, F. (2014). Indus- Dittrich, M., Giljum, S., Lutter, S., & Polzin, C. (2012). trial Symbiosis in Iskenderun Bay: A journey Green economies around the world. Implications from Pilot Applications to a National Program in of resource use for development and the environ- Turkey. Technology Development Foundation of ment. Vienna: SERI. Turkey. Earth Overshoot Day. (2017). Earth Overshoot Day Anastasio, M. (2016). The circular economy: practical 2017 is August 2, the earliest date since ecological steps to enhance the EU package. Green Budget overshoot began in the early 1970s. Press release Europe. 27/06/2017. Retrieved from www.overshootday. Anbumozhi, V., Baulraj, A., Mohanchezhian, A., & org/newsroom/press-release-english/ Rakhmah, T. F. (2016). Creating Integrated Busi- Ellen MacArthur Foundation. (2013). Towards the ness, Economic, and Environmental Value within Circular Economy. a Circular Economy in India. In Anbumozhi, V. Ellen MacArthur Foundation. (2016a). Circular Econ- and J. Kim (eds.), Towards a Circular Economy: omy Overview. Corporate Management and Policy Pathways (pp. Ellen MacArthur Foundation. (2016b). Growth within: 135–159). ERIA Research Project Report 2014–44, A circular economy vision for a competitive Jakarta: ERIA. Europe. Barneveld, J., Veen, G., Enenkel, K., Mooren, C., Ellen MacArthur Foundation. (2016c). Circular Econ- Talman-Gross, L., Eckartz, K., Ostertag, K., omy in India: Rethinking Growth for Long-Term Duque-Ciceri, N., Fischer, T., Gama, M., Scheidt, L., Prosperity. Wilts, H., Schäfer, L., & Fischer, S. (2016). Regula- tory barriers for the circular economy. Lessons European Commission. (2014). Towards a circular from ten case studies. Technopolis Group, Fraun- economy: A zero waste programme for Europe. hofer ISI, thinkstep, Wuppertal Institute. Communication from the Commission to the European Parliament, the Council, the Euro- Baughan, J. S., & Evale, C.M. (2004). The Green Dot pean Economic and Social Committee and the System: Promoting Recycling in the European Committee of the Regions. Union. WorldTrade Executive, Inc. European Energy Agency (EEA). (2014). Resource-Ef- Benton, D., & Hazell, J. (2015). The circular economy ficient Green Economy and EU Policies. EEA in Japan. Environmental SCIENTIST (Journal Report No 2/2014, Copenhagen. of the Institution of Environmental Sciences), March 2015, 55–59. European Environmental Bureau. (n.d.). Economic Instruments for a Circular Economy. Bocken, N. M., de Pauw, I., Bakker, C., & van der Grinten, B. (2016). Product design and business European Parliament. (2016). Closing the loop. New model strategies for a circular economy. Journal circular economy package. Briefing January 2016. of Industrial and Production Engineering, 33(5), European Union. (2009). Directive 2009/125/Ec of 308–320. the European Parliament and of the Council of Chen, J. C. (2017). Social-environmental dilemmas of 21 October 2009 establishing a framework for planning an ‘ecological civilisation’ in China. In the setting of Ecodesign Requirements for Ener- Bhan, G., Srinivas, S., & Watson, V. (Eds.). (2017). gy-Related Products. Journal of the European The Routledge Companion to Planning in the Union. Global South (pp.180–191). Routledge. Evans, S., Rana, P., & Short, S. (2014). Final Set of Tools Clark, G., Kosoris, J., Hong, L. N., & Crul, M. (2009). and Methods that Enable Analysis of Future Design for sustainability: current trends in Oriented, Novel, Sustainable, Value Adding Busi- sustainable product design and develop- ness-Models and Value-Networks. Deliverable ment. Sustainability, 1(3), 409–424. D2.6. SustainValue project. Daunorienė, A., Drakšaitė, A., Snieška, V., & Fichtner, W., Tietze-Stöckinger, I., Frank, M., & Rentz, Valodkienė, G. (2015). Evaluating sustainability of O. (2005). Barriers of interorganisational environ- sharing economy business models. Procedia-So- mental management: two case studies on indus- cial and Behavioral Sciences, 213, 836–841. trial symbiosis. Progress in Industrial Ecology, 2, 73–88. Green Industrial Policy - Concept, Policies, Country Experiences

Finkler, C., Teixeira, F. E., Costa, L. F., Ganzer, P. P., Khan, S. A. (2016). E-products, E-waste and the 132 Chais, C., Radaelli, A. A. P., Dorion, E. C. H., & Olea, Basel Convention: Regulatory Challenges and P. M. (2017). Motivations and Barriers on Reverse Impossibilities of International Environmental Logistics Solid Waste development practices. 4th Law. Review of European, Comparative & Inter- Responsible Management Education Research national Environmental Law, 25(2), 248–260. Conference, 2017, Curitiba, Brazil. Kim, I. C. (2002). Korea’s policy instruments for waste Forlin, V., & Scholz, E. (2017). Leveraging consumers’ minimization. Journal of Material Cycles and recycling incentives in a circular economy. CORE Waste Management, 4(1), 12–22. Discussion Papers, 2017/10. Kim, M. H., Song, H. B., Song, Y., Jeong, I. T., & Kim, Frederiks, E. R., Stenner, K., & Hobman, E. V. (2015). J. W. (2013). Evaluation of food waste disposal Household energy use: Applying behavioural options in terms of global warming and energy economics to understand consumer deci- recovery: Korea. International Journal of Energy sion-making and behaviour. Renewable and and Environmental Engineering, 4(1), 1. Sustainable Energy Reviews, 41, 1385–1394. Kumar, S., Smith, S.R., Fowler, G., Velis, C., Kumar, S.J., Ghisellini, P., Cialani, C., & Ulgiati, S. (2016). A review Arya, S., Kumar, R., & Cheeseman, C. (2017) Chal- on circular economy: the expected transition lenges and opportunities associated with waste to a balanced interplay of environmental and management in India. Open Science, 4(3), 160764. economic systems. Journal of Cleaner Produc- Li, R. (2012). Assessment of Water Footprint in Indus- tion, 114, 11–32. trial Park: A Case Study of Yixing Economic Government of Chile. (2016). President Bachelet Development Zone. enacts the Recycling and Extended Producer Li, Y., & Shi, L. (2015). The resilience of interdepend- Liability Law. 26/05/2016. Retrieved from www. ent industrial symbiosis networks: A case of gob.cl/president-bachelet-enacts-the-recy- Yixing Economic and Technological Develop- cling-and-extended-producer-liability-law/ ment Zone. Journal of Industrial Ecology, 19(2), Government of South Korea. (2013). South Korea’s 264–273. waste management policies. Legislative Council Marello, M., & Helwege, A. (2014). Solid waste Secretariat. INC04/12–13. management and social inclusion of waste Govindan, K., Soleimani, H., & Kannan, D. (2015). pickers: opportunities and challenges. Social-In- Reverse logistics and closed-loop supply chain: clusion-Working-Paper. Global Economic Govern- A comprehensive review to explore the future. ance Initiative, Paper, 7. European Journal of Operational Research, 240(3), Moon, M. J. (2017). Government-driven Sharing Econ- 603–626. omy: Lessons from the Sharing City Initiative of Guoping, L., Kaixi, H., & Wang, F. (2017). Aluminum the Seoul Metropolitan Government. Journal of Prices Surge as China Imposes Curbs. 27/09/2017. Developing Societies, 33(2), 223–243. Caixin. Retrieved from www.caixinglobal. Next Manufacturing Revolution. (2013). The Next com/2017-09-27/101151249.html Manufacturing Revolution: Non-Labour Resource Gutberlet, J., Carenzo, S., Kain, J. H., & Mantovani Productivity and its Potential for UK Manufactur- Martiniano de Azevedo, A. (2017). Waste Picker ing. Next Manufacturing Revolution. Retrieved Organizations and Their Contribution to the from www.nextmanufacturingrevolution.org/ Circular Economy: Two Case Studies from a wp-content/uploads/2013/09/Next-Manufactur- Global South Perspective. Resources, 6(4), 52. ing-Revolution-full-report.pdf International Resource Panel. (n.d.) Glossary. Pan-African Forum on E-waste. (2012). Call for Retrieved from http://internationalresourcepanel. Action on E-waste in Africa. Set of priority org/glossary actions. International Synergies Ltd. (2014). Iskenderun Pfeiffer, L., & Lin, C. Y. C. (2014). Does efficient irri- Bay Industrial Symbiosis. Retrieved from gation technology lead to reduced groundwater www.international-synergies.com/projects/ extraction? Empirical evidence. Journal of Envi- iskenderun-bay-industrial-symbiosis/ ronmental Economics and Management, 67(2), Jacobsen, N. B. (2006). Industrial symbiosis in 189–208. Kalundborg, Denmark: a quantitative assessment Rahman, M.F., Islam, K., & Islam, K.N. (2016). Industrial of economic and environmental aspects. Journal Symbiosis: A Review on Uncovering Approaches, of industrial ecology, 10(1–2), 239–255. Opportunities, Barriers and Policies. Journal of Jørgensen, S. E., Fath, B. D., Nielsen, S. N., Pulselli, F. Civil Engineering and Environmental Sciences, M., Fiscus, D. A., & Bastianoni, S. (2015). Flourish- 2(1), 011-019. ing Within Limits to Growth: Following Nature’s Reuse and Recycling EU Social Enterprises Way. Routledge. (RREUSE). (2015). Improving product reparability: Policy options at EU level. Chapter 8 Promoting Circular Economies

Rockström, J., Steffen, W., Noone, K., Persson, Å., Eisenmenger, A. Geschke, M. Lieber, H. P. Wieland, Chapin, F.S., Lambin, E.F., Lenton, T.M., Schef- A. Schaffartzik, F. Krausmann, S. Gierlinger, K. 133 fer, M., Folke, C., Schellnhuber, H.J., & Nykvist, Hosking, M. Lenzen, H. Tanikawa, A. Miatto, & B. (2009). A safe operating space for human- T. Fishman. Paris: United Nations Environment ity. Nature, 461(7263), 472–475. Programme. Rucevska, I., Nellemann, C., Isarin, N., Yang, W., Liu, United Nations Industrial Development Programme N., Yu, K., Sandnæs, S., Olley, K., McCann, H., (UNIDO). (2016). Global assessment of eco-indus- Devia, L., Bisschop, L., Soesilo, D., Schoolmeester, trial parks in developing and emerging countries. T., Henriksen, R., & Nilsen, R. (2015). Waste Crime Walls, M. (2011). Deposit-refund systems in practice – Waste Risks: Gaps in Meeting the Global Waste and theory. Discussion Paper, November 2011. Challenge. A UNEP Rapid Response Assessment. Resources for the Future. Nairobi and Arendal: United Nations Environ- Wang, Q.Z., Deutz, P., & Gibbs, D. (2015). UK-China ment Programme and GRID-Arendal. collaboration for industrial symbiosis: a multi- Samuelsen, A. G., & Støyle, R. V. (2016). The power of level approach to policy transfer analysis. In nudging; Using reverse logistics to improve recy- Deutz, P., Lyons, D., & Bi, J. (eds.) International cling behaviour in household waste manage- Perspectives on Industrial Ecology (pp.89-107). ment while taking the intention-action gap into Cheltenham, UK and Northampton, MA, USA: account (Master›s thesis, BI Norwegian Business Edward Elgar. School). WasteTech. (2017). Waste Management Market Schandl, H., Hatfield-Dodds, S., Wiedmann, T., in Russia. State. Legislation. Prospects. 10th Geschke, A., Cai, Y., West, J., Newth, D., Baynes, T., International Trade Fair for Waste Manage- Lenzen, M., & Owen, A. (2016). Decoupling global ment, Recycling, Environmental Technolo- environmental pressure and economic growth: gies and Renewable Energy. Retrieved from Scenarios for energy use, materials use and www.waste-tech.ru/RXRU/RXRU_Waste- carbon emissions. In Journal of Cleaner Produc- Tech/documents/2017/WasteMarket_eng. tion, 132, 45–56. pdf?v=636161093183299836 Schlegelmilch, K., Eichel, H., & Pegels, A. (2017). Pric- Watts, J. (2009). China sets first targets to curb ing environmental resources and pollutants and world’s largest carbon footprint, Copenhagen the competitiveness of national industries. In climate change conference 2009. The Guardian, Altenburg, T., & Assmann, C. (Eds.). (2017). Green 26/11/2009. Industrial Policy. Concept, Policies, Country Expe- Wijkman, A., & Skånberg, K. (2015). The circular econ- riences (pp. 102–119). Geneva, Bonn: UN Environ- omy and benefits for society. Jobs and Climate ment; German Development Institute (DIE). Clear Winners in an Economy Based on Renewa- Tao, Y. (2017). How Policies Work to Foster Industrial ble Energy and resource Efficiency. Club of Rome. Symbiosis: A Comparison between UK and China. Wilts, H. (2016). Deutschland auf dem Weg in die Krei- Cambridge: University of Cambridge. slaufwirtschaft? Friedrich-Ebert-Stiftung, Abtei- Teh, B. T., Ho, C. S., Matsuoka, Y., Chau, L. W., & Gomi, K. lung Wirtschafts-und Sozialpolitik. (2014). Determinant factors of industrial symbio- Women in Informal Employment: Globalizing and sis: greening Pasir Gudang industrial park. In IOP Organizing (WIEGO) (2014). The Urban Informal Conference Series: Earth and Environmental Workforce: Waste Pickers/Recyclers. Informal Science (Vol. 18, No. 1, p. 012162). IOP Publishing. Economy Monitoring Study. United Nations Environment Programme (UNEP). World Business Council for Sustainable Development (2016). Global Material Flows and Resource (WBCSD). (2016). Informal approaches towards a Productivity. An Assessment Study of the UNEP circular economy. World Resources Forum (WRF), International Resource Panel. H. Schandl, M. Swiss Federal Laboratories for Materials Science Fischer-Kowalski, J. West, S. Giljum, M. Dittrich, N. and Technology (EMPA). Green Industrial Policy - Concept, Policies, Country Experiences 134 CHAPTER 9 TRADE AND INVESTMENT LAW AND GREEN INDUSTRIAL POLICY

Aaron Cosbey Chapter 9 Trade and Investment Law and Green Industrial Policy

1. INTRODUCTION 135 Other chapters in this book explore questions most influential bodies of international law, with of what types of green industrial policies might the strongest mechanisms of dispute settlement work or have worked, and in what contexts. In and enforcement. this chapter we explore the ways in which such policies might be restricted by international trade The focus of this chapter is on describing where and investment law. that body of international law conflicts with green industrial policies, by compiling a list of problem- The question is salient because international atic policy areas. There are, as noted below, many trade and investment play a fundamental role in other types of green industrial policy that govern- sustainable development and trade is presented ments might pursue that are not trade-related and in the UN’s Sustainable Development Goals as a are not restricted by trade and investment law. means of implementation: “an engine for inclu- But these are not discussed at length. sive economic growth and poverty reduction [that] contributes to the promotion of sustainable This chapter first explores three cross-cutting development” (UN 2015: 29). As such, many tradi- issues that help elucidate the way in which trade tional industrial policy tools are related to trade. and investment law relate to green industrial At the same time, the body of international trade policy. This section will be optional for those with and investment law–contained in the General a solid understanding of trade law, but for others it Agreement on Tariffs and Trade (GATT) and will be an important introduction to the following World Trade Organization (WTO) agreements, in discussion. That discussion assesses five of the regional and bilateral trade agreements, in bilat- most widely used types of policies that govern- eral investment treaties and in investment provi- ments might employ to pursue green industrial sions of trade agreements–constitutes one of the policy, asking in each case how trade and investment law might treat them.

2. CROSS-CUTTING ISSUES To fully understand the relationship of trade 2.1. RELATIONSHIP BETWEEN WTO LAW AND and investment law to various green industrial NON-MULTILATERAL AGREEMENTS policy measures, it is necessary to review a few cross-cutting concepts: The WTO body of trade and investment law is a collection of legal agreements among the 164 ◼◼ How is WTO law related to non-multilateral members of the WTO, spanning various trade trade and investment law? This is important and investment-related subjects. One of the because it means nothing to simply say that most relevant to green industrial policy is the trade law prohibits or allows a given policy; GATT, but there are also more specialised agree- trade law exists at the WTO multilateral level ments covering how governments should create as well as in regional and bilateral agree- and implement trade-related measures such as ments; and there is trade-related investment subsidies, government procurement, standards law at both those levels as well as in dedicated and labelling, intellectual property rights, invest- investment agreements. The different bodies ment measures and others. For the most part, of law contain different rights and obligations. all WTO members are bound by the obligations ◼◼ How does WTO dispute settlement interpret of all agreements, but there are a few that have existing law? In many cases it is difficult to say been agreed among a sub-set of members only– with authority which policies are legal or not, called plurilateral agreements. Members that have and it is important to understand why that is. acceded to the WTO since it succeeded the GATT ◼◼ How do the exceptions work in trade and in 1995 are bound by the terms of their individual investment law and how might they be applied accession agreements, which may involve addi- in the context of green industrial policy? tional obligations. Exceptions are important to this discussion, since a number of green industrial policy measures seem to be in conflict with trade law obligations. But in some cases they can be saved by exceptions. Green Industrial Policy - Concept, Policies, Country Experiences

WTO members are also bound by commitments The analysis below focuses primarily on WTO law 136 they may have made in non-multilateral settings, when discussing trade-related law, since most of for example in regional trade agreements and the relevant non-multilateral commitments are bilateral investment treaties, and it is important similar to those found in the WTO. When discuss- to understand how the bodies of law interact. ing investment-related law, the analysis focuses primarily on bilateral investment treaties and the Regional trade agreements will not typically set investment chapters of regional trade agreements. out obligations that are less onerous than those When the analysis departs from these parameters, found in WTO law since, no matter what the this is noted. regional trade agreement parties agree among themselves, they are still bound by their obligations to other WTO members. Whether or not 2.2. INTERPRETATION AND APPLICATION OF these agreements explicitly refer to it, WTO law TRADE AND INVESTMENT LAW still binds the parties. Regional trade agreements Green industrial policy efforts can conflict with may clarify what the regional trade agreement trade and investment law. There are several parties understand to be WTO law, where there important caveats that apply when discussing are some doubts. For example, the Mexico-Pan- the potential conflicts. First, the law is not usually ama free trade agreement clarifies that the WTO spelled out to the level of detail that would allow it exceptions for measures designed to protect to be obvious how it applies in any given situation; human, animal, or plant life or health will cover were that possible, there would be few disputes. environmental measures, though the word envi- Rather, the applicable law must be interpreted in ronment does not appear in the original WTO the context of each case, taking into account the text (Mexico-Panama Free Trade Agreement 2014, specifics of the measures in question. Moreover, Article 19.2(1)). This is not an alteration of WTO law, some aspects of WTO law call for judgment in but merely a declaration of mutual understanding interpretation: What do ‘reasonable’ or ‘dispropor- as to how that law should be interpreted. tionately large’ mean in a specific context? And On other issues regional trade agreements go there are often dissenting interpretations issued beyond WTO law in setting out Parties’ obligations. on such questions by the different authorities: in In the Canada-EU Comprehensive Economic and the case of the WTO, panellists or appellate body Trade Agreement, for example (Article 8.5(1)(f)), members. As such, it is not always possible to say Canada and the EU prohibit the use of technology unequivocally that a certain type of policy is or is transfer requirements as a condition for investing not prohibited by WTO law. There is no principle in their territory–a commitment not found in the of precedent in WTO law, but in practice the prin- WTO’s Agreement on Trade-Related Investment ciple is well respected. Measures. Similarly, it is typical in regional trade Blanket statements become even more difficult in agreements for the parties to agree to intellectual the area of investment law. Ad hoc arbitral panels property right provisions that are stronger than interpret investment law and, while they will be those found in the WTO’s Agreement on Trade-Re- conscious of previous awards, they have been lated Intellectual Property, such as longer terms known to issue widely divergent interpretations for patent protection. In such cases the so-called on essentially similar questions of law (Jones WTO-plus provisions typically apply only between 2011; Ortino 2012; Nilsson and Englesson 2013). In the parties.18 WTO disputes the final interpreter of the law is In the area of investment, WTO law is contained the standing Appellate Body that tends to adhere in the Agreement on Trade-Related Investment strongly to case law. Measures, which has relatively light obligations. It should also be noted that most WTO members Non-multilateral agreements in this area are arguably have many WTO-illegal measures in found in bilateral investment treaties and in the place. The fact or probability of illegality only investment chapters of regional trade agreements. comes into play when some other WTO member The investment law in such agreements is almost formally complains about those measures, possi- all WTO-plus. As with trade-related provisions, bly taking the issue beyond consultations to WTO law is still in force and obligations in such dispute settlement. The decision to do so is not agreements will only apply between the parties. taken lightly by any government, and suspected non-conforming measures are often simply

18 Some regional trade agreements' (RTA) commitments apply more broadly. For example, a change in patent protection terms would affect all a country’s trading partners. Chapter 9 Trade and Investment Law and Green Industrial Policy

ignored as not damaging enough to warrant the are caveats to ensure that the measures to be cost and political friction engendered by dispute saved are only focused on achieving the specified 137 settlement proceedings. No such political or objectives, in this case environmental objectives. public interest filter exists in the case of invest- However, by definition green industrial policy ment law, where a private investor has the right to measures aim to achieve both environmental initiate binding arbitration with the host govern- objectives and economic objectives, and excep- ment, as investor-state dispute settlement. tions to the law are definitely not meant to save the latter.

2.3. EXCEPTIONS TO TRADE LAW In other words, if a green industrial policy measure is seeking the exception status of GATT Arti- Some forms of green industrial policy meas- cle XX (b) or (g), the objective of the measure will ures are likely to breach WTO or investment law have to match the environmental criteria in those obligations. For several areas of trade law this sub-paragraphs–and this should be straightfor- is not the end of the story. Some agreements ward. The measure will be aimed at preserving include carefully described exceptions that cover the environment. But the elements of the meas- agreed objectives, including environmental ones ure that simultaneously seek to promote indus- such as conservation of exhaustible resources trial development will also need to be justified as and protection of human, animal or plant life or environmental in nature, or face rejection under health (GATT, Article XX; General Agreement on the chapeau text. Trade in Services (GATS), Article XIV; Agreement on Trade-Related Investment Measures (TRIMs), So a measure discriminating in favour of a Article 3; Agreement on Government Procurement domestic green sector would have to provide (AGP), Article III). GATT Article XX serves as a sort some environmental rationale for that discrim- of template model for such exceptions: ination. If foreign competitors produce environmental goods or services more efficiently, there is Subject to the requirement that such meas- no such rationale at least in the short run: it would ures are not applied in a manner which would actually be better for the environment to use constitute a means of arbitrary or unjustifiable foreign manufactured environmental goods and discrimination between countries where the services. Because they are cheaper, such imported same conditions prevail, or a disguised restric- goods would yield more environmental benefit per tion on international trade, nothing in this dollar spent. Agreement shall be construed to prevent the adoption or enforcement by any contracting Cosbey and Mavroidis (2014) have suggested that party of measures: … such an environmental rationale might exist, describing an ‘environmental Bastable test’. The (b) necessary to protect human, animal or Bastable test asks whether the total costs of plant life or health; … an industrial policy measure are outweighed (g) relating to the conservation of exhaustible by the value of all the future benefits that the natural resources if such measures are made measure might produce. The cost and bene- effective in conjunction with restrictions on fits would be calculated as present discounted domestic production or consumption; … values of the stream of future costs and benefits. If the balance is positive, the measure is judged Some of the agreements discussed here have worthwhile. This is strictly a national economic these sorts of exceptions: the WTO and regional cost-benefit calculation. trade provisions on trade in goods, trade in services, trade-related investment measures and The environmental Bastable test, on the other government procurement. Others do not, such as hand, would be a global assessment and would the WTO Agreement on Subsidies and Countervail- look at total environmental costs and benefits. ing Measures, subsidy provisions of regional trade It would count as costs the lost environmental agreements and bilateral investment treaties. benefits from slower deployment of the technologies, and against this it would balance the future Governments will find it difficult, if not impos- environmental benefits. Those benefits would sible, to successfully use the general exceptions materialize only if the policy succeeded in creat- in trade law to ‘save’ non-conforming green ing new innovators and competitors in the envi- industrial policy measures. In GATT Article XX, ronmental technology space. the chapeau paragraph’s requirements that the measures not be unjustifiable discrimination or a disguised restriction on international trade Green Industrial Policy - Concept, Policies, Country Experiences

So if it could be convincingly argued that creating membership, including any countries that had 138 new firms in a green sector ultimately resulted withheld bilateral agreement. in environmental gains–that a measure passed the environmental Bastable test–then a green Section C allows a covered member to take other industrial policy measure might pass GATT Arti- sorts of unspecified measures aimed at the cle XX (b) or (g). But this would be difficult. The same objective, even where they conflict with environmental Bastable test, like the original test, most GATT obligations. Kuntze and Moeren- is mostly a heuristic device and impractical to hout (2014) argue that this might include feed-in use as a benchmark. The success or failure of any tariffs with domestic content requirements, for given green industrial measure is unpredictable example. The measures qualifying for exception and not easily quantified for the purposes of envi- status under GATT XVIII section C would have ronmental cost benefit analysis. Given the scep- to be notified to other WTO members, and if the tical audience such an argument would likely members agree that there is no more GATT-con- have in the WTO dispute settlement system, its sistent manner to achieve the same objectives, chances of success are not good. then the measure can proceed. However even if no such agreement can be reached, the notify- GATT’s Article XVIII provides another set of excep- ing member can, after 90 days, implement the tions to GATT rules, for balance of payments and notified measure. Affected members can request industrial policy purposes that apply to parties negotiations on compensation. whose economies “can only support low standards of living and are in the early stages of There has been limited case law to interpret Arti- development”. There has never been case law to cle XVIII, with only two cases and those concern- clarify exactly which countries are covered by ing balance of payments provisions, not the this description. industrial policy clauses (WTO 1999; WTO 2002a; WTO 2002b). As such it remains to be seen how These industrial policy exceptions are found in free a WTO member might be to take advantage GATT Article XVIII sections A and C. Section A of what seems to be a useful exception. Neither of allows a covered member to withdraw or modify the two balance of payments cases was decided a scheduled GATT concession, to raise a tariff, in favour of the implementing country. It is not to “promote the establishment of a particular obvious which countries might benefit. For exam- industry with a view to raising the general stand- ple, China is a developing country member of the ard of living of its people.” Such a move must WTO, but it is unclear whether it could be char- be preceded by notice to other members and acterized as ‘in the early stages of development’. negotiations with any member that is substan- It is also unclear whether the member would be tially affected to agree on compensatory adjust- required to demonstrate that the proposed meas- ment–likely the lowering of tariffs in other ures would actually raise the general standard of areas of interest to its partners. It is technically living of their people–something that might be possible but very difficult for such an effort to difficult in the case of domestic content require- proceed even when a member is unsuccessful in ments. And of course the requirements for negoti- getting negotiated agreement with its partners. ations on compensation to affected members may This would require agreement from the general make the proposition less viable.

3. LEGALITY OF TRADE- AND INVESTMENT- RELATED MEASURES This section explores the following types of This is not an exhaustive list of the policies that policies and measures, all of which might be aim to further green industrial policy, but it covers employed as elements of green industrial policy: the most commonly used policies that have trade and investment legal implications. ◼◼ Tariffs ◼◼ Subsidies ◼◼ Performance requirements 3.1. TARIFFS ◼ Service sector limitations ◼ Under certain conditions, tariff protection for ◼ Government procurement ◼ infant industries may result in welfare gains ◼ State-owned enterprises ◼ for the implementing economy (Greenwald and ◼ Competition policy. ◼ Stiglitz 2006). Tariffs are the most straightforward Chapter 9 Trade and Investment Law and Green Industrial Policy

tool of industrial policy. They were featured heav- of price support or a financial contribution by a ily in strategies for import substitution indus- government or public body. A financial contribu- 139 trialisation pursued by mostly Latin American tion can take several forms: economies between the 1930s and 1980s with a mixed record of success (Baer 1972). Tariff protec- ◼◼ direct transfer of funds tion for industrialisation purposes also served a ◼◼ non-collection of government revenue other- number of today’s developed economies through- wise due, such as taxes out their history of industrialisation (Baer 1972; ◼◼ provision of goods and services, other than Chang 2002). general infrastructure, or purchase of goods.

As well as simple tariff walls for infant industry Two final threshold tests of a subsidy in WTO protection, tariff levels can be manipulated to law are that the price support or financial contri- favour a greater degree of processing domesti- bution must confer some benefit on the recipi- cally. Tariff escalation is the practice of setting ent, for example, purchase of goods should be at tariffs increasingly higher for more heavily higher than market rates, and that the support or processed goods. contribution is specific. Specificity in this case means that it is targeted at, or primarily benefits, But there are limited opportunities to use tariffs a particular enterprise, industry, or group of enter- in these ways. Commitments to bind import prises or industries. According to the Agreement tariffs at agreed levels are the most basic of the on Subsidies and Countervailing Measures’ Article GATT obligations. Each member has a schedule 2, specificity can be found where there is explicit of commitments that outline the maximum tariff direction in the measure for a subsidy to go to a they may apply to various goods (GATT, Article narrow group, or can be found where the measure II:1). This maximum is the bound rate. Within the makes no such provisions but the final effect is space afforded by that rate, countries are free to such that specific groups benefit. vary their tariffs for any purpose and the resulting tariff is called the applied rate. But the progressive There are two types of subsidies in WTO law: liberalisation embodied in the GATT and WTO prohibited and actionable. A prohibited subsidy processes has seen average tariffs drop from an is conditioned on domestic content require- estimated 22 per cent pre-GATT in 1947 (Bown ments or on some form of export performance. and Irwin 2015) to 1.5 per cent in 2014 (World Bank If a subsidy is found to be prohibited, there is no 2017), meaning that space is limited for most need to demonstrate specificity–it is assumed. goods and most countries. Prohibited subsidies breach WTO commitments and must be removed. Actionable subsidies are all other measures that fit the WTO definition of a 3.2. SUBSIDIES subsidy. Actionable simply means they are open to complaint. To be found WTO-illegal, actionable Subsidies may be the most widespread tool of subsidies must be shown to be specific, and to industrial policy, and of green industrial policy cause adverse economic effects for the complain- more specifically. They can take many forms, ant’s producers. including: The SCM’s provisions constitute limits on the ◼◼ cash grants types of green industrial policy support that ◼◼ land grants countries might offer via subsidies. The most ◼◼ preferential tax treatment obvious limitation is against the use of prohib- ◼◼ concessional loans ited subsidies. Countries may not impose domes- ◼◼ loan guarantees tic content requirements as a condition for the ◼◼ export credit receipt of a subsidy. A number of countries have ◼◼ price support sought to build up their capacity in the renewa- ◼◼ mandated purchase regimes, such as feed-in ble energy space with such subsidies. Canada tariffs was taken to WTO dispute settlement over the ◼◼ public research and development province of Ontario’s use of domestic content ◼◼ provision of dedicated infrastructure, not of requirement-linked subsidies, as feed-in tariffs, use to the general public. to build up domestic solar and wind sectors (WTO The WTO definition of subsidies is used in this 2013); the EU was taken to consultations by China analysis. In the WTO Subsidies and Countervail- for similar measures in Greece and Italy (WTO ing Measures (SCM) Agreement’s Article 1, that 2012); and the US listed domestic content require- definition requires that there be either some form ment-linked subsidies as part of their complaint against the Indian National Solar Mission, though Green Industrial Policy - Concept, Policies, Country Experiences

ultimately it did not pursue the subsidy angle consideration of specificity, and then whether it 140 (WTO 2016). could cause adverse effects.

The other element of prohibited subsidies is a It was noted above that feed-in tariffs had been link to export performance. This means that the subject of a number of disputes. When condi- governments cannot make subsidies conditional tioned on domestic content requirements, such on a firm’s achieving a certain level of exports or tariffs will almost certainly be considered prohib- create tax exemptions that only apply to exported ited subsidies. But without the domestic content goods, for example. Many green industrial policy requirements, they fall into the grey zone of efforts involve an expectation that the supported uncertainty. The first question is whether they forms will contribute to the domestic economy are in fact subsidies. In the Canadian renewable through export activities. And best practice in energy dispute described above, the Panel ruled industrial policy involves assessing success and that they were, but the Appellate Body over- cutting off support to unsuccessful firms, which turned that ruling, arguing that it was neces- could be structured in such a way as to consti- sary to compare the prices offered under feed-in tute a conditional link to export performance. tariffs to prices offered in a hypothetical compet- So this aspect of subsidy law poses significant itive market for renewables—not to the prices of obstacles to some types of green industrial policy. conventional electricity—to gauge whether the Note, however, that least developed countries feed-in tariffs were too high (WTO 2013, paragraph and certain low-income countries are exempted 5.246). Being unable to identify such a hypothet- from the strictures on export-linked subsidies ical market, the Appellate Body was unable to (Agreement on Subsidies and Countervailing complete the analysis, and unable to find that Measures, Article 27.2). That exemption applies the feed-in tariffs were subsidies. In the end only until they have reached export competitive- the answer will depend strongly on the details ness–a threshold that is not defined (Agreement of the feed-in tariff regime in question. Among on Subsidies and Countervailing Measures, Arti- other things, the higher the rates paid, the more cle 27.5). likely the measure will be considered a subsidy. If feed-in tariffs are considered to be subsidies, the Export credit might also be considered a prohib- second question is whether they cause adverse ited subsidy. Export credit can be either the provi- effects, in this case to foreign producers of renew- sion of a loan to an exporter or the guarantee or able electricity. Renewable electricity being rarely insurance of such a loan, referred to as ‘pure cover traded, and almost universally subsidised, this support’. It can be used to help domestic firms second step is unlikely to occur. As such, feed-in achieve successful export market penetration, tariffs without domestic content requirements spanning the cash-flow gap between shipment attached are probably WTO-legal, and are used and actual payment. The Agreement on Subsidies in more than 100 jurisdictions worldwide, in 75 and Countervailing Measures includes disciplines national level schemes and 35 sub-national level on both types of export credit. The loans them- schemes (REN21 2016:109). selves cannot be made at rates of interest below the rates the government has had to pay for the Such feed-in tariffs may serve green industrial funds or the rates they would have to pay if they policy objectives by facilitating a green struc- borrowed commercially. The pure cover support tural transformation. But since they lack the cannot be made at premium rates that would domestic content requirements they may not be result in the support programme running at a as effective at fostering domestic sector develop- financial loss.19 ment. That said, feed-in tariffs without domestic content requirements could still have some bene- Moving beyond prohibited to actionable subsidies ficial effects on domestic firms, especially if they enters a ‘grey zone of uncertainty’ since, among are employed as one element in a suite of green other things, governments cannot predict whether industrial policy measures that target those firms. their measures are specific or whether they will Indeed, they were used in that way by Germany cause adverse effects for foreign producers (Char- in its Energiewende (Lütkenhorst and Pegels 2014; novitz 2014). Guidance in this zone must consider Pegels 2017, this volume). first whether a given green industrial measure might be considered a subsidy, including the

19 Item (k) paragraph 2 carves out any export credit that is covered by and conforming to “an international undertaking on export credit” – a clear reference to the OECD Arrangement on Guidelines for Officially Supported Export Credit. But this Arrangement has failed to evolve beyond a gentlemen’s agreement and so has no legal coverage of the sort envisioned in item (k) paragraph 2. As such, the carve out is not in effect. Chapter 9 Trade and Investment Law and Green Industrial Policy

Subsidies can also be used to support research environmental regulations. But the provisions were and development, a type of support that is central time-limited, and when they failed to be renewed 141 to green industrial policy efforts worldwide. in 1999 they expired. A number of authors have Research and development subsidies produced suggested reviving and modifying the non-action- strong success in the US with such breakthrough able category to cover subsidies with environmen- information technologies as the integrated circuit tal aims (Hufbauer et al. 2009; Cottier 2011; Rubini and the microprocessor and they contributed to 2012; Charnovitz 2014; Shadikhodjaev 2015). the establishment of Silicon Valley as a centre of excellence (Leslie 2000; Mazzucato 2013). These One of the most prevalent green industrial policy subsidies fall squarely in the grey zone of uncer- tools is tax policy, with preferences of various types tain legal status under WTO rules: while they may being accorded to favoured sectors and firms in an be actionable subsidies, it is inherently difficult effort to encourage the development of linkages in to assess whether they meet the test of causing the economy, to encourage export orientation, or to injury to foreign producers. In one of the largest address various market failures by lowering costs subsidy disputes ever handled in the WTO, the (Schlegelmilch et al. 2017, this volume). Export- US complained of a number of aspects of the linked subsidies are prohibited. Subsidies aimed EU’s support for Airbus’ production of large civil at encouraging linkages, for example tax pref- aircraft, including funding for research and tech- erences to firms that transfer technology or hire nology development. The Panel in that case found local workers, are considered under performance that the funding was indeed a subsidy, and that requirements–a separate category of policy tools it caused adverse effects to Boeing, its US-based and measures. competitor. The Appellate Body agreed that a For other types of tax preferences, the law of the subsidy existed, but overturned the ruling that WTO is that they are a subsidy if they represent it caused adverse effects, arguing that the Panel revenue forgone that is otherwise due (Agreement had not shown a direct link between the support on Subsidies and Countervailing Measures, Art. and specific technologies in Airbus products that 1.1(a)(1)(ii)). So if a tax preference is an exception demonstrably gave it a competitive advantage to the normal tax practice for a specific firm or (WTO 2011a, paragraph 1407). sector, it is a subsidy. Of course, determining what The most obvious point to draw from this is that, is normal practice is often difficult. Determining as noted above, it is not straightforward to predict adverse effects is not straightforward either, but what green industrial policy measures will run the Panel in an EU complaint against Boeing’s afoul of WTO law; on this specific question the research and development seemed to set the bar Panel and the Appellate Body disagreed. More low, using ‘commonsense reasoning’ to infer that important, the Appellate Body’s ruling seems to set lower taxes allowed Boeing to price at lower levels a fairly high bar for finding adverse effects from than would otherwise have been possible (WTO research and development subsidies. However in 2011b, paragraph 7.1820). a subsequent case, a complaint by the EU against Tax policy can also be used punitively, as a tool to US subsidies to Boeing, the Appellate Body agreed deliberately target sectors that should be phased with the Panel that research and development out or altered as part of a green structural trans- subsidies “contributed in a genuine and substan- formation. Pollution charges are a common tool tial way to Boeing’s development of technologies of environmental policy, and carbon taxes in for the 787” and as such caused adverse effects particular are becoming more so. This sort of (WTO 2011b, paragraph 1012). Ultimately the legal- disruptive use of green industrial policy tools ity of research and development subsidies is highly could be used to phase out sectors such as fossil context-specific, leaving an uncomfortable degree fuels (Cosbey et al. 2017, this volume). This use of uncertainty for governments. of tax policy is not limited by trade law, which is In the SCM Agreement there was a third cate- more concerned with states according favourable gory of subsidies in addition to prohibited and treatment to domestic firms and sectors. actionable: non-actionable (Agreement on Subsi- Several other types of potential green industrial dies and Countervailing Measures, Art.8). These policy measures are very likely candidates for were subsidies that served objectives that the conflicts with WTO subsidy law: members acknowledged as so important that they would be allowed to stand even if they did cause ◼◼ direct grants of land to a firm, where such adverse effects. These included subsidies for grants are not generally available to other research and development, as well as subsidies for firms/sectors a limited category of support to meet the costs of Green Industrial Policy - Concept, Policies, Country Experiences

◼◼ construction of infrastructure that will only be The purpose of most of these requirements is 142 used by the firm or sector, such as a railway to to better exploit the potential of foreign direct transport minerals from mine site to port investment to build up domestic capacity in ◼◼ mandated floor prices for the sector’s products, specific sectors or firms. where these exceed what would prevail in a competitive market. The WTO’s Agreement on Trade-Related Invest- ment Measures prohibits two types of perfor- Subsidies to services are not disciplined under mance requirements: requirements to purchase WTO law, so green industrial policy measures that domestic content and requirements regarding involve such subsidies are available to govern- certain export links and balance of payments. The ments. Governments could, for example, subsidise former was the point of law on which Canada lost domestic environmental engineering services to its case on renewable energy: its feed-in tariffs build up their capacity to export. were deemed prohibited investment measures because they conditioned investors’ price premi- There are a few types of measures commonly ums on the use of domestic content (WTO 2013). labelled subsidies that are not constrained by WTO law. In general, subsidies to consumers fit The prohibitions in the Agreement on Trade-Re- this mould, since the subsidies will not discrim- lated Investment Measures are augmented by inate between green goods from foreign and obligations in some international investment domestic producers; consumers can use the agreements–a group of agreements that Nikièma subsidies to purchase either, unless there are (2014) calls the ‘growing minority.’ Housed in aspects of the law that prevent such free choice. investment chapters of regional trade agreements This will mean no adverse effects. or in bilateral investment treaties, most modern agreements signed by the US, Canada and some The under-pricing of environmental externalities Asian nations also prohibit requirements to trans- is sometimes labelled subsidy, as when a govern- fer technology and to supply an identified interna- ment imposes weak environmental regulations tional market exclusively from the host country. on its firms, giving them a competitive advantage. This would not constitute a subsidy under The commitments in these agreements, even the definition of the Agreement on Subsidies and where they mirror those found in the WTO, argua- Countervailing Measures. bly have more deterring force. WTO commitments are enforced through state-to-state dispute settle- 3.3. PERFORMANCE REQUIREMENTS ment, which states will only initiate if they pass a significant threshold of public interest. Commit- A performance requirement is a condition that ments in international investment agreements, investors must meet to establish or operate a on the other hand, are enforced through inves- business, or to obtain some advantage offered by tor-state dispute settlement, which is triggered the host nation. Performance requirements could by the investor based on a much narrower calcu- include conditions to: lus. As well, a loss in the WTO dispute settlement system entails the withdrawal of the disputed ◼◼ export a certain percentage of total sales, or measure, while a loss in the investor-state dispute total production settlement system involves financial penalties ◼◼ enter into joint venture arrangements with ranging as high as billions of dollars.20 domestic partners ◼◼ transfer or share technology There is still a great deal of scope for the use of ◼◼ source a certain amount of inputs locally other forms of performance requirements even by ◼◼ expend a certain amount on research and Parties to these WTO-plus agreements. That scope development is clarified, for example, in the investment chap- ◼◼ hire a certain number or percentage of local ter of the US-Korea Free Trade Agreement [2012] employees. Article 11.8(3)(a):

Governments may impose performance require- Nothing in paragraph 2 [which prohibits ments as mandatory measures. Governments certain performance requirements] shall be may also provide investors with fiscal incentives construed to prevent a Party from condition- or other advantages in exchange for businesses’ ing the receipt or continued receipt of an compliance with the performance requirements. advantage, in connection with an investment

20 The largest award to date, awarded to Yukos under the Energy Charter Treaty over its expropriation by Russia, totalled US$ 50 billion (Brauch 2014). It was later set aside on jurisdictional grounds. Chapter 9 Trade and Investment Law and Green Industrial Policy

in its territory of an investor of a Party or of a Commitments will also specify what mode of non-Party, on compliance with a requirement service delivery is being covered: 143 to locate production, supply a service, train or employ workers, construct or expand particu- ◼◼ Mode 1: from a foreign country to the member. lar facilities, or carry out research and develop- ◼◼ Mode 2: consumed by member citizens in the ment, in its territory. foreign country. ◼◼ Mode 3: delivered by a foreign investor estab- That said, there is always some uncertainty. The lishing a commercial presence in the member Tribunal for a North American Free Trade Agree- country. ment (NAFTA) investment dispute ruled that ◼◼ Mode 4: by means of foreign nationals enter- new requirements for research and development ing the member country to work as service spending and education and training spend- providers. ing were prohibited performance requirements (Vis-Dunbar 2013; ICSID 2013). This ruling, however, Mode 3 is of particular interest in the context of was not related to NAFTA prohibitions against green industrial policy and trade law. The GATS requirements for research and development, specifies a number of restrictions to the sort of but relied on the prohibition against domestic regulations members may use to limit foreign content requirements, interpreting the new rules presence of service providers (GATS, Article XVI.2). as requirements for local spending. Members may not maintain or adopt: a. limitations on the number of service suppli- 3.4. SERVICE SECTOR LIMITATIONS ers whether in the form of numerical quotas, monopolies, exclusive service suppliers or the The WTO’s General Agreement on Trade in requirements of an economic needs test; Services governs trade in services at the multilat- b. limitations on the total value of service trans- eral level. Services are roughly defined as those actions or assets in the form of numerical purchased items that one cannot drop on one’s quotas or the requirement of an economic foot: banking, tourism, telecommunications or needs test; real estate services, for instance. Environmen- c. limitations on the total number of service tal services include environmental engineering, operations or on the total quantity of service environmental remediation, eco-tourism, and the output expressed in terms of designated services of actors such as energy service compa- numerical units in the form of quotas or the nies, wind/solar farm developers and residential requirement of an economic needs test; installers of renewable energy systems. Kommer- d. limitations on the total number of natural skollegium (2014) argues strongly that without an persons that may be employed in a particular underpinning of green services, there would be service sector or that a service supplier may few green goods. employ and who are necessary for, and directly related to, the supply of a specific service in WTO members each have a schedule of commit- the form of numerical quotas or the requirements in the GATS that describe what kind of ment of an economic needs test; treatment they will offer to foreign service-provid- measures which restrict or require specific ers. That treatment will vary from sector to sector, e. types of legal entity or joint venture through and there is no expectation that a member make which a service supplier may supply a service; commitments in all sectors. There are two types and of commitments possible: f. limitations on the participation of foreign capi- ◼◼ commitment to treat foreign service suppli- tal in terms of maximum percentage limit on ers no less favourably than domestic ones, or foreign shareholding or the total value of indi- national treatment vidual or aggregate foreign investment. ◼◼ openness to investment in the sector, or These obligations effectively constrain the abil- market access commitments. ity of governments to foster domestic capacity by It is possible to qualify the commitments in any limiting or putting conditions on entry of service way the members wishes. A member might, for sector investors. In the GATS, these obligations example, commit to offer market access to foreign only apply to sectors in which members have service-providers in the banking sector only to made some sort of commitments. However these those above a certain level of capitalisation. limitations appear in a number of recent regional trade agreements (US-Korea Free Trade Agree- ment [2012], Article 12.4; Trans-Pacific Partner- ship Agreement [2015], Article 10.5; Korea-Vietnam Green Industrial Policy - Concept, Policies, Country Experiences

Free Trade Agreement [2015]; Article 8.4; US-Cen- development. This has long been the case in the 144 tral America Free Trade Agreement [2004], Arti- aerospace sector, for example, where US and Euro- cle 11.4), mostly signed by the US, where they are pean governments award contracts for research horizontal commitments made with respect to all and development in areas of particular interest to sectors–a much more extensive restriction. their domestic firms (Eliasson 2010).

As of early 2017 there are negotiations ongoing The WTO strictures on government procurement among WTO members for a plurilateral Trade in are contained in a plurilateral deal, the Agreement Services Agreement. With the EU counting as one, on Government Procurement (GPA). As of Octo- the 23 members involved are responsible for over ber 2017 the Agreement covers 47 of the WTO’s 70 per cent of global trade in services. These talks 165 members, with another nine members in the have dragged on longer than expected—currently process of accession. Membership is strongly amounting to over 20 negotiating rounds—and tilted toward more developed economies, with may not be finalized any time soon. While the only six of the parties from outside the World drafts of the negotiating text are classified, leaked Bank high-income classification, and three of text as of June 2016 shows draft provisions on those being parties due to their EU membership localisation that would go well beyond what is (WTO 2017). in the GATS, including proposed prohibitions on requirements for: The GPA offers significant scope for governments to procure environmentally superior goods and ◼◼ a service provider’s board members to be services. Article X.6 makes it clear that Parties nationals are allowed to craft their technical specifications ◼◼ purchase of a set amount of domestic content, in such a way as to take environmental consider- or preferences for local goods ations into account. And the Agreement contains ◼◼ technology transfer an exception clause similar to GATT’s Article XX, ◼◼ the hiring of a certain number or percentage allowing for measures necessary to protect plant, of nationals animal or human life or health (Agreement on ◼◼ expenditure on research and development in Government Procurement, Article III.2). country. But while the scope for environmentally moti- On some of these provisions there is disagree- vated procurement exists, the scope for using ment among the negotiating members, so it those purchases to foster the development of is impossible to say what the final text might domestic capacity is extremely limited. The contain. It does seem clear that any final outcome exceptions clause makes it clear that the discrim- is likely to restrict the ability of governments to ination in question cannot be a disguised restric- use green industrial policy in the context of trade tion on international trade, and must be justifiable. in services. Other articles in the agreement make it clear that the drafters’ intent is such that building domes- 3.5. GOVERNMENT PROCUREMENT tic capacity would not be a justifiable form of discrimination. Article IV.6 prohibits the use of Government procurement can be an important offsets in the procurement practice, offsets being tool of green industrial policy. This can be signif- defined in Article 1 paragraph (l) as: icant as government procurement in 2010 aver- ...any condition or undertaking that encouraged 10 to 15 per cent of GDP globally (Anderson ages local development or improves a Party’s et al. 2011). The scale of demand means that this balance-of-payments accounts, such as the sort of assistance can be used to help firms attain use of domestic content, the licensing of tech- viable scale and learn by doing under difficult nology, investment, counter-trade and similar start-up conditions (Geroski 1990). This is particu- action or requirement. larly important for newly commercialized goods such as non-solar non-wind renewable energy Further, Article IV.2 mandates that there should technologies, new low-carbon transport and be no discrimination against locally established battery technologies, and the like. suppliers on the basis of nationality, or on the basis of the origin of the goods and services the Procurement for green industrial policy goes supplier might offer. beyond simply purchasing green goods. Procure- ment of services, and in particular services such There are certain exceptions provided for devel- as research and development, is a central part oping countries that accede to the Agreement of some governments’ strategies for industrial on Government Procurement. Among these are Chapter 9 Trade and Investment Law and Green Industrial Policy

that during an individually negotiated transition that reasoning, characterizing the contracts as period, such parties may adopt price preferences more akin to equity infusions in a joint venture, 145 for national suppliers or those of other develop- equity infusions being one type of subsidy as per ing countries. But for the majority of developing the WTO definition, wherein the return on the countries, not having acceded to the Agreement, government’s equity was expected to come in the there are no obligations that would prevent them form of useful scientific and technical informa- from using procurement as a tool to promote tion, discoveries and data. Either way, what the domestic enterprises. US characterized as government procurement of research services was ultimately found instead to The catch in some cases will be how procure- be a subsidy (WTO 2011b). ment is defined; it cannot be for the purposes of commercial resale (Agreement on Government Procurement, Article II.2(a)(ii)). This was one of 3.6. STATE-OWNED ENTERPRISES the legal downfalls of India in its WTO dispute Governments often use public ownership of with the US over its Jawaharlal Nehru National enterprises, whether full or partial ownership, to Solar Mission (WTO 2016). India awarded solar further industrial policy objectives. As of 2012, power developers with long-term power purchase state-owned enterprises included the world’s 13 agreements, but eligibility hinged on the use of biggest oil firms and biggest natural gas company, mandated amounts of domestic content. India and their penetration in emerging market sectors argued that the purchase of electricity from the was considerable: energy at more than 60 per cent developers amounts to government procurement. state-owned, utilities at more than 50 per cent, That argument was rejected, as the electricity telecoms at more than 30 per cent and finance at purchased was to be resold on a commercial basis more than 30 per cent (Wooldridge 2012). to Indian consumers. Governments are able to push such firms to Had India or Canada succeeded in arguing that achieve public policy objectives, as opposed to their purchases were for the purpose of govern- narrowly defined commercial objectives. State- ment procurement, they would have benefited owned enterprises might, for example, be pres- from a GATT carve-out for government procure- sured to source from local suppliers or hire ment measures (GATT, Article III.8(a)). That carve- and train local employees. Most or all of their out exempts government procurement from the research and development will be domestically GATT strictures of national treatment, and thereby conducted, leading to potential spillover benefits also exempts it from the obligations found in the and development of domestic sectoral expertise. Trade-Related Investment Measures Agreement, State-owned enterprises also may have greater under which both Canada and India lost their capacity and appetite than private sector actors to cases.21 GATT Article II.8(a) is a broad carve-out invest in ventures that require substantial capital that benefits any member that has not acceded outlays and/or risky ventures that aim to produce to the Agreement on Government Procurement, breakthrough technologies (Ciuriak and Singh allowing almost unlimited scope for the exercise 2015). State-owned enterprises as tools of green of procurement as green industrial policy. industrial policy are not widespread; their reach It was noted above that some governments use is mostly confined to extractive sectors and util- procurement of research services as indus- ities. But they could become green policy tools in trial policy support, and here too the definition the future, should governments decide to develop of procurement is central. The US argued that national firms in green sectors. its contracts for research services with Boeing There is broad scope within the WTO for such were government procurement and therefore policy choices (Singh and Jose 2016). GATT’s not covered under the Agreement on Subsidies Article XVII on State Trading Enterprises obliges and Countervailing Measures, which the US members that establish state-owned enterprises argued does not cover purchases of services. The to ensure that they operate as commercial enter- Panel disagreed with the claim as government prises in their purchases and sales, not discrim- procurement, arguing that the contracted work inating between national and international was for the sole benefit of Boeing, the contrac- suppliers or customers, for example. This does tor, and that therefore the contracts amounted to limit the potential use of local purchasing pref- direct transfer of funds (WTO 2011b, paragraphs erences, but does not limit the establishment of 7.978–7.1027). The Appellate Body overturned

21 The TRIMS Agreement’s prohibitions on domestic content requirement are in effect a clarification of what is prohibited under GATT’s Article III on national treatment. Green Industrial Policy - Concept, Policies, Country Experiences

state-owned enterprises. The not-yet-ratified in competition policy–the introduction of broader 146 Trans-Pacific Partnership goes much further than public policy objectives that may ultimately see WTO law, for example with signatories obliged to competition lessened to achieve non-competi- not cause injury to other signatories’ industries tion-related objectives. by providing direct or indirect non-commercial assistance to their state-owned enterprises There is no conflict between trade law and this (Trans-Pacific Partnership Agreement, Article sort of selective enforcement of competition 17.6). These provisions, which resemble subsidy policy. There is no discipline in WTO law that law, are not found in other regional trade agree- obliges members to enact or enforce competi- ments but their inclusion in the Trans-Pacific tion policy, though there has been discussion Partnership may signal a trend. on the subject for many years (Evenett 2003). There is a standing Working Group on the Inter- action between Trade and Competition Policy 3.7. COMPETITION POLICY that reports to the WTO’s General Council. The Trans-Pacific Partnership breaks new ground Competition policy is a framework of legislation for trade agreements in addressing competition and institutions that aims to ensure competition policy, but even that agreement allows significant among economic actors. The objectives of compe- leeway to national governments to exempt the tition policy vary with the national and historical application of competition law for public policy context (Motta 2004), but many of them focus reasons, asking only for consistency and trans- on a common set of problems raised by a lack of parency, and providing no dispute settlement competition, including: mechanism (Trans-Pacific Partnership Agree- ◼◼ distribution of wealth from consumers to ment 2015, chapter 16). monopolistic firms, with a resulting overall loss of welfare 3.8. CONCLUSIONS ◼◼ reduced incentives for firms to innovate and seek efficiencies, with possible environmental This section assesses whether there are areas impacts of trade and investment law that constrain the ◼◼ abuse of dominance in the market, including use of green industrial policy (Table 9.1). In many acting to reduce the ability of new firms to cases there do exist such constraints. For exam- enter the market. ple, tariff policy is strongly constrained by bound tariff rate commitments, subsidies and perfor- Competition policy usually consists of a national mance requirements are conditional on domestic competition law and institutions to enforce the content requirements and export performance are law and advocate for competitiveness among the prohibited. As well, governments that are party various related ministries. One of the major tasks to the Agreement on Government Procurement of any national competition body is to scrutinise cannot use offsets or discriminate in favour of mergers and acquisitions to determine whether domestic suppliers in their procurement. they meet competition-related public interest criteria. Where clear constraints do not exist, government policy may be discouraged by uncertainty. In some cases industrial policy can increase Governments will be understandably reluc- competition among firms in a national economy, tant to embark on programmes of green indus- but the intersection of competition law, industrial trial policy that take them into what we have policy and trade law lies in the practice of granting called the grey zone of uncertainty of trade and exemptions from competition law for industrial investment law, such as actionable subsidies for policy purposes. The history of industrial policy in research and development and other purposes, Korea, for example, involves state encouragement or the use of GATT’s Article XVIII exceptions for of mergers and cartels to achieve the economies of developing countries. scale and experience necessary to compete at the international level (Amsden 1989). Taiwan simi- That said, uncertainty could also be seen as larly encouraged mergers, particularly in sectors policy space. Most governments have laws that facing difficulties (Wade 1990). Cosbey and Mann are probably in breach of their trade and invest- (2014) discuss the widespread use of industrial ment obligations and this is only a problem when policy in the context of the mining sector to create another country complains about them. Formal national champions– domestic firms that domi- complaints are costly, in terms of both politi- nate the economic landscape in their respective cal capital and financial and human resources. sectors. These are examples of a common practice They will be launched only when those costs Chapter 9 Trade and Investment Law and Green Industrial Policy

are perceived as justified by significant adverse time a green industrial policy measure may have effects and when the odds of winning are suffi- achieved its goals. 147 ciently high. They may also be considered when domestic political concerns demand it, and this On the other hand, the mere threat of WTO may happen even in the face of low odds. Small disputes is often enough to steer small developing economies are less likely to be challenged, since country governments away from grey zone meas- their export streams tend to be less significant. ures, since the costs of managing a dispute are The time that it takes to finally come to resolution high and expert resources are scarce. In the case of a complaint can also be seen as policy space. In of non-WTO investment agreements, there is a rare cases, a decade might pass before the WTO’s very different calculus for launching a complaint. dispute mechanism renders final judgement on a This can be done by a private firm without consid- measure, its appeal, and subsequent complaints eration of wider political concerns, and there are over non-compliance with judgments. By that now funding models that encourage speculative forays into arbitration (Honlet 2015).

Table 9.1: Summary: Legality of Trade and Investment Measures

Status Conditions Applicable Law Notes

Protective tariffs unrestricted only permitted WTO - GATT bound tariffs are (conditions) within boundaries increasingly low - not much set by bound tariffs room for policy

Subsidies

linked to exports prohibited WTO - SCM But allowed for LDCs

conditional on LCR prohibited WTO - SCM, WTO - TRIMs

Feed-in tariffs unrestricted only when not WTO - SCM these may be actionable (conditions) accompanied by subsidies, but would likely LCR conditions never be challenged

Support for R&D cannot cause WTO - SCM adverse effectes for foreign competitors

Preferential tax treatment

subsidy removal unrestricted for example, fossil fuel (disruptive) subsidy reform

subsidies to services unrestricted

consumer subsidies unrestricted

Export credit cannot be offered on WTO - SCM non-market terms

Performance requirements

local content prohibited WTO - TRIMs

technology transfer allowed in WTO; prohibited in a few international investment agreements

joint venture allowed in WTO; prohibited in a few international investment agreements

employment targets unrestricted

R&D expenditure unrestricted But have been interpreted by a recent Tribunal as local content requirements

training of locals unrestricted Green Industrial Policy - Concept, Policies, Country Experiences

Status Conditions Applicable Law Notes 148 Service sector requirements

quantitative restrictions on prohibited WTO - GATS, many providers RTAs

limits on foreign prohibited WTO - GATS, many shareholding RTAs

domestic content unrestricted But possibly slated for prohibition in TISA

technology transfer unrestricted But possibly slated for prohibition in TISA

R&D expenditure unrestricted But possibly slated for prohibition in TISA

hiring of locals unrestricted But possibly slated for prohibition in TISA

Government procurement Highly restricted in GPA, but not all WTO members are parties to GPA

State-owned enterprises unrestricted no nationality-based WTO - GATT Article (conditions) discrimination in XVII purchases or sales

Exemptions from unrestricted competition policy

Punitive environmental taxes, unrestricted E.g., carbon taxes charges

That said, there are of course green industrial the as-yet untested industrial policy exceptions policy options still open to governments that are available through GATT Article XVIII. interested in promoting domestic green sectors. Feed-in tariffs can be used as part of a broader But ultimately it is clear that trade and investment suite of measures, as long as they do not contain law imposes substantial obstacles for many types domestic content requirements, for example. of green industrial policy. This is to be expected. Performance requirements for training of staff Trade and investment law have as core principles are also acceptable. Those governments that are the commitment to non-discrimination between not party to the GPA still retain broad scope for foreign and domestic producers and investors, using government procurement as green indus- and industrial policy is ultimately about benefiting trial policy. There are, moreover, a number of the latter, usually at the expense of the former. For types of green industrial policy to which trade and this reason domestic content requirements are so investment law do not apply, such as science and clearly and powerfully prohibited, for example. For education policies designed to build up domes- this reason consumer subsidies to purchase green tic capacity, general infrastructure investment, goods are probably acceptable while subsidies funded demonstration projects, legal reform to to domestic producers of those same goods are encourage investment, and so on. not. As a general rule, trade and investment law restricts ‘vertical’ policies–those that discriminate There are also a number of exceptions available between foreign and domestic firms in specific to developing countries, such as the ability of sectors. In contrast, ‘horizontal’ policies, such as least developed counties to use export-contin- generic infrastructure investment, legal reforms, gent subsidies, although they may lack the fiscal regulatory strengthening, science and education resources to make that exception meaningful, and policies, tend not to be affected. Chapter 9 Trade and Investment Law and Green Industrial Policy

4. CONCLUDING THOUGHTS 149 This analysis has found major impediments for new globally competitive entrants in green certain types of green industrial policies in trade sectors with global impacts, such as the renew- and investment law. It bears asking the question: able energy sector, then the benefits will be felt is this a problem? The answer depends, among globally. That is, not only will the implement- other things, on an assessment of whether the ing country improve environmental outcomes green industrial policies in question are actu- directly via its own structural transformation, but ally effective at achieving their stated objectives: it will also foster indirect global environmental enhancing environmental quality while contrib- improvement by lowering the costs, or improv- uting to domestic economic wellbeing by building ing the effectiveness, of green goods and services. up producers of green goods and services. Bahar If the goods and services in question address a et al. (2013) argue strongly that domestic content significant global environmental problem, such requirements are ineffective, wasting scarce as climate change, those global benefits will also dollars that could be more effectively spent on be significant. renewable energy technology dissemination and reducing global public welfare. If they are correct, This is an important consideration because the and this is a general truth, then it is not a prob- laws on trade and investment are founded in lem for domestic content requirements to be public welfare economics. If the restricted poli- constrained by trade and investment law. If, on cies can indeed be effectively used in ways that the other hand, domestic content requirements would increase global public welfare, then there is can be effective, then the limits imposed by trade a strong case for the reform of those restrictions. and investment law are problematic. Johnson In the context of the multilateral rules, however, (2013), for example, puts forward clear recom- a strong case may not be enough, at least in the mendations to policymakers for successful use of short term. The room for new negotiations in domestic content requirements, based on an anal- the WTO is highly restricted by the inability to ysis of India’s National Solar Mission. conclude the Doha Round of trade negotiations.

Assessing the potential effectiveness of the vari- Given the imperative importance of the two chal- ous measures surveyed here is beyond the scope lenges that green industrial policy addresses, of this paper, but the question needs to be asked if environment and development, it is urgent to find we are to judge the propriety of trade and invest- authoritative answers to the question of effec- ment law impediments to green industrial policy, tiveness. Of course it is impossible to say in the and to consider possible reforms. abstract whether certain policies will be effective. The specific context is decisive, as are the details A useful heuristic proposed earlier in this paper of implementation. It is likely that for many of the is the environmental Bastable test. For any given policies surveyed here the answer is that they green industrial policy intervention, does the are sometimes effective. The question then for present value of the environmental impacts of the trade and investment policymakers is whether measure, in terms of long-term increased compe- it is appropriate to prohibit policies that can be tition and innovation in a global green sector, effectively utilised, even if it is clear that success exceed the present value of the environmen- will be elusive. The alternatives to such a prohibi- tal costs–usually in terms of short-term higher tion are reform of the law to guide policies toward prices for environmental goods and services, effectiveness as mandatory sunset provisions, and their consequent curtailed dissemination? for example, or opening wide the doors to green It is noted above that this framework is useful industrial policy with the clear understanding for thinking about the problem, but less useful in that states can and will get it wrong, to the detri- calculating the value of any specific measure. ment of their trading partners. The crucial global public interest element to green industrial policy, It is also useful to bear in mind one of the key as opposed to traditional industrial policy, should differences between traditional industrial policy have an influence in the final balance point and green industrial policy. In the former context, among the many deciding factors in the mix. the benefits vest almost exclusively in the implementing country, and the costs are born by foreign producers–a traditional mercantilist outcome. In the latter context, the costs are much the same, but the benefits may be much different. If the green industrial policy is successful in fostering Green Industrial Policy - Concept, Policies, Country Experiences

150 REFERENCES Amsden, A. H. (1989). Asia’s next giant: South Korea Cottier, T. (2011). Energy in WTO Law and Policy. In and late industrialization. New York: Oxford T. Cottier & P. A. Delimatsis (Eds.), The prospects University Press. of international trade regulation. From fragmen- Anderson, R. D., Müller, A. C., Osei-Lah, K. P., Leon, J. tation to coherence (pp. 221–244). Cambridge P. D., & Pelletier, P. (2011). Government procure- University Press. ment provisions in regional trade agreements. In Eliasson, G. (2010). Advanced public procurement as S. Arrowsmith & R. D. Anderson (Eds.), The WTO industrial policy: The aircraft industry as a tech- regime on government procurement. Challenge nical university. Economics of science, technol- and reform (pp. 561–656). Cambridge University ogy and innovation: [v. 34]. New York, London: Press. Springer. Baer, W. (1972). Import Substitution and Industriali- Evenett, S. J. (2003). Can developing economies zation in Latin America: Experiences and Inter- benefit from WTO negotiations on binding disci- pretations. Latin America Research Review. (7), plines for hard core cartels? Aussenwirtschaft. (2), 95–122. 215–248. Bahar, H., Egeland, J., & Steenblik, R. (2013). Domestic Geroski, P. A. (1990). Procurement policy as a tool of Incentive Measures for Renewable Energy With industrial policy. International Review of Applied Possible Trade Implications. OECD Trade and Economics, 4(2), 182–198. Environment Working Papers: no. 2013/01. Paris: Greenwald, B., & Stiglitz, J. E. (2006). Helping Infant OECD Publishing. Economies Grow: Foundations of Trade Policies Bown, C. P., & Irwin, D. A. (2015). The GATT’s start- for Developing Countries. American Economic ing point: Tariff levels circa 1947. Working paper Review, 96(2), 141–146. series / National Bureau of Economic Research: Honlet, J.-C. (2015). Recent decisions on third-party Vol. 21782. Cambridge, MA: National Bureau of funding in investment arbitration. ICSID Review, Economic Research. 30(3), 699–712. Brauch, M. D. (2014). Yukos v. Russia: Issues and legal Hufbauer, G. C., Charnovitz, S., & Kim, J. (2009). reasoning behind US$ 50 billion awards. Invest- Global warming and the world trading system. ment Treaty News, 04/09/2014. Washington: Peterson Institute for international Chang, H.-J. (2002). Kicking away the ladder? economics. Development strategy in historical perspective. International Centre for the Settlement of Invest- London: Anthem. ment Disputes (ICSID). (2013). Award in the Charnovitz, S. (2014). Green Subsidies and the WTO arbitration under chapter eleven of the North (EUI Working Paper No. RSCAS 2014/93). American Free Trade Agreement between Mobil Ciuriak, D., & Singh, H. V. (2015). Mega-Regionals and Investments Canada Inc. & Murphy Oil Corpora- the Regulation of Trade: Implications for Indus- tion and Canada. Case No. ARB(AF)/07/4. trial Policy. E15 Initiative. Geneva. Johnson, O. (2013). Exploring the effectiveness of Cosbey, A., & Mann, H. (2014). Bilateral Investment local content requirements in promoting solar Treaties, Mining and National Champions: PV manufacturing in India. Discussion paper / Making it Work. Background paper for the Ad Deutsches Institut für Entwicklungspolitik: Vol. Hoc Experts Group Meeting: Bilateral Investment 2013,11. Bonn: Dt. Inst. für Entwicklungspolitik. Treaties and National Champions, 18th Meeting of Jones, D. (2011). Investor-State Arbitration: The Prob- the Inter-Governmental Committee of Experts: lem of Inconsistency and Conflicting Awards. “National Champions, Foreign Direct Investment Presented at the German-American Lawyers’ and Structural Transformation in Eastern Africa” Association Practice Group Day, Frankfurt, 26 Kinshasa, 17–20 February, 2014. Kinshasa. March 2011. Frankfurt. Cosbey, A., & Mavroidis, P. C. (2014). A Turquoise Kommerskollegium. (2014). Making Green Trade Mess: Green Subsidies, Blue Industrial Policy Happen. Environmental Goods and Indispensa- and Renewable Energy: The Case for Redrafting ble Services. Stockholm: Swedish National Board the Subsidies Agreement of the WTO. Journal of of Trade. International Economic Law, 17(1), 11–47. Kuntze, J., & Moerenhout, T. (2014). Are Feed-In Cosbey, A., Wooders, P., Bridle, R., & Casier, L. (2017). Tariffs Consistent with WTO Law? Legal Tools In with the good, out with the bad: Phasing out to Confront Interdisciplinary Challenges. Brill polluting sectors as green industrial policy. In Publishers. In F. Baetens & J. G. M. Caiado (Eds.), Altenburg, T., & Assmann, C. (Eds.). (2017). Green Frontiers of international economic law. Legal Industrial Policy. Concept, Policies, Country Expe- tools to confront interdisciplinary challenges riences (pp. 69–86). Geneva, Bonn: UN Environ- (pp. 151–180). Leiden: Martinus Nijhoff. ment; German Development Institute / Deutsches Institut für Entwicklungspolitk (DIE). Chapter 9 Trade and Investment Law and Green Industrial Policy

Leslie, S. (2000). The biggest ‘Angel’ of them all: The RES/70/1. military and the making of Silicon Valley. In M. Vis-Dunbar, D. (2013). Canada loses NAFTA arbitration 151 Kenney (Ed.), Understanding Silicon Valley. The over R&D performance requirements. Investment anatomy of an entrepreneurial region (pp. 48–67). Treaty News, 25/03/2013. Stanford, Calif.: Stanford University Press. Wade, R. (1990). Governing the market: Economic Lütkenhorst, W., & Pegels, A. (2014). Stable poli- theory and the role of government in East Asian cies--turbulent markets: Germany’s green industrialization. Princeton, NJ: Princeton industrial policy: the costs and benefits of University Press. promoting solar PV and wind energy. Research Wooldridge, A. (2012). The visible hand (Special report. Winnipeg, Manitoba, Geneva, Switzerland, Report: State Capitalism): The Economist, Beaconsfield, Quebec: International Institute for 21/01/2012. Sustainable Development; Global Subsidies Initi- ative; Canadian Electronic Library. World Bank. (2017). World Development Indicators database. Mazzucato, M. (2013). The entrepreneurial state: Debunking public vs. private sector myths. World Trade Organization. (1999). Appellate Body London/New York: Anthem Press. Report, India - Quantitative Restrictions on Imports of Agricultural, Textile and Industrial Motta, M. (2004). Competition policy: Theory Products. WT/DS90/AB/R, 23 August 1999. and practice. Cambridge, England, New York: Cambridge University Press. World Trade Organization. (2002a). Appellate Body Report, India – Measures Affecting the Automo- Nikièma, S. H. (2014). Performance Requirements in tive Sector. WT/DS146/AB/R, WT/DS175/AB/R, 19 Investment Treaties. Winnipeg. March 2002. Nilsson, A., & Englesson, O. (2013). Inconsistent World Trade Organization (2002b). Appellate Body Awards in Investment Treaty Arbitration: Is an Report, India – Measures Affecting the Automo- Appeals Court Needed? Journal of international tive Sector. WT/DS146/AB/R, WT/DS175/AB/R, 19 arbitration. (30(5)), 561–579. March 2002. Ortino, F. (2012). Legal Reasoning of International World Trade Organization. (2011a). Appellate Body Investment Tribunals: A Typology of Egregious Report, European Communities - Measures Failures. Journal of international dispute settle- Affecting Trade in Large Civil Aircraft. WT/DS316/ ment, 31–52. AB/R, 17 May 2011. Pegels, A. (2017). Germany: The energy transition World Trade Organization. (2011b). Panel Report, as a green industrial development agenda. In United States - Measures Affecting Trade in Large Altenburg, T., & Assmann, C. (Eds.). (2017). Green Civil Aircraft (second complaint). WT/DS353R, 31 Industrial Policy. Concept, Policies, Country Expe- March 2011. riences (pp. 166–184). Geneva, Bonn: UN Environ- ment; German Development Institute / Deutsches World Trade Organization. (2012). Request for consul- Institut für Entwicklungspolitk (DIE). tations by China, European Union and Certain Member States — Certain Measures Affecting REN21. (2016). Renewables 2016 Global Status Report. the Renewable Energy Generation Sector. WT/ Paris. DS452/1, G/L/1008, G/SCM/D95/1, G/TRIMS/D/34, Rubini, L. (2012). Ain’t Wastin’ Time No More: Subsi- 7 November 2012. dies for Renewable Energy, The SCM Agreement, World Trade Organization. (2013). Appellate Body Policy Space, and Law Reform. Journal of Inter- Report, Canada - Certain Measures Affecting the national Economic Law, 15(2), 525–579. Renewable Energy Generation Sector, Canada – Schlegelmilch, K., Eichel, H., & Pegels, A. (2017). Pric- Measures Related to the Feed-In Tariff Program. ing environmental resources and pollutants and WT/DS412/AB/R, WT/DS426/AB/R, 6 May 2013. the competitiveness of national industries. In World Trade Organization. (2016). Appellate Body Altenburg, T., & Assmann, C. (Eds.). (2017). Green Report, India - Certain Measures Relating to Solar Industrial Policy. Concept, Policies, Country Expe- Cells and Solar Modules. WT/DS456/AB/R, 16 riences (pp. 102–119). Geneva, Bonn: UN Environ- September 2016. ment; German Development Institute / Deutsches Institut für Entwicklungspolitk (DIE). World Trade Organization. (2017). Parties, observers and accessions. Agreement on Government Shadikhodjaev, S. (2015). Renewable Energy and Procurement. Government Support: Time to ‘Green’ the SCM Agreement? World Trade Review, 14(03), 479–506. Singh, H. V., & Jose, R. (2016). Industrial Policy and the WTO Rules-Based System. E15 Initiative. Geneva. United Nations. (2015). Transforming our world: The 2030 agenda for sustainable development. A/ Green Industrial Policy - Concept, Policies, Country Experiences 152 PART 4: COUNTRY EXPERIENCES Chapter 10 Renewable Energy as a Trigger for Industrial Development in Morocco CHAPTER 10 153 RENEWABLE ENERGY AS A TRIGGER FOR INDUSTRIAL DEVELOPMENT IN MOROCCO

Georgeta Vidican Auktor Green Industrial Policy - Concept, Policies, Country Experiences

154 1. MOROCCO: A REGIONAL PLAYER IN THE RENEWABLE ENERGY SECTOR22 The Middle East and North Africa (MENA) region embraced by Morocco appears to be more has been in the spotlight regarding investments comprehensive than that of other countries in the in renewable energy technologies. Both national region. In times of major turmoil across the MENA governments and international donors have region, efforts made by the Moroccan government started to view renewables as an important solu- to maintain political stability have reinforced the tion for the energy and development challenges in country’s status as the upcoming market for solar the region. Their interest has also been fuelled by and wind energy. The high level of investment increasing evidence that investments in renew- that already materialised for the first large-scale able energy technologies can open the path to renewable energy plants underlines this positive inclusive and green growth, through job crea- outlook. All these factors, if adequately chan- tion, private sector development, greening of the nelled, will contribute to positioning Morocco as a industrial sector and environmental sustainability regional player in this emerging sector. (UNEP 2011; World Bank 2012; IRENA 2013). In light of these early developments, this chap- Within the region, Morocco stands out as a ter aims to capture not only the strategic steps particularly interesting case. First, the country taken by the government to deploy renewable imported more than 95 per cent of its energy in energy technologies but, more importantly, to 2011, and its energy demand is expected to triple delve into how opportunities for entering new by 2030 (MEMEE 2011). This strong dependence sectors and developing skills have been exploited on imports places Morocco in a very precari- and how they could be further harnessed in the ous energy and financial situation over the next future. In particular, the chapter explores how decades. At the same time, due to its geograph- the government uses industrial policy to localize ical location, Morocco benefits from vast renew- parts of production for renewable power plants able energy resources that could be exploited to and support the development of a local indus- provide energy beyond the country’s domestic try. The chapter also examines how policymak- needs. Second, as in other countries of the region, ers balance high-tech concentrated solar power Morocco suffers from a high unemployment rate, plants and low-tech rooftop solar thermal and especially among the educated youth–a situation photovoltaic aspects of the renewable energy that calls for private sector development with a strategy. The chapter also explains how the strat- focus on job creation. Finally, Morocco’s manufac- egy pursues targeted skills development–from turing base is relatively low, industrial competi- technical vocational education and training to tiveness is deficient and entrepreneurship is higher education–to increase competitiveness. limited. However, promising signs of technologi- So far, progress has been slow due to several cal upgrading in some sectors, including automo- internal and external factors, but effort has been tive and aeronautics, suggest a new momentum consistent. Thus, assessing these early steps and in the private sector that could lead to new oppor- distilling lessons learned can only contribute to tunities (Hahn and Vidican Auktor 2017). solidifying further developments.

To respond to these challenges, the Moroc- The chapter is structured in the following way: It can government has engaged in an ambitious starts by highlighting why green industrial policy and highly dynamic process of developing the is crucial for Morocco in the context of pressing renewable energy sector. In a short period of challenges to development and energy security. It time, Morocco has gained a reputation as the then discusses three important elements of such most promising destination for solar and wind an industrial policy—market creation, supply side energy investment in the MENA region. What measures, and an integrated strategy for indus- distinguishes Morocco from other countries is its trial development—with regard to existing steps commitment to linking solar and wind electric- taken at the government and private sector level ity generation projects to industrial development, and the potential barriers to achieving the set employment creation and competitiveness more objectives. The chapter ends with some consider- generally, even if only small steps in this direc- ations on how to capitalise on these early steps. tion have been made so far. The overall approach

22 The empirical part of this chapter originates primarily from in-depth interviews conducted in 2012 with a large set of stakeholders involved in the solar energy sector development in Morocco (see Vidican et al. 2013). The data has been updated through subsequent research. Chapter 10 Renewable Energy as a Trigger for Industrial Development in Morocco

2. GREEN INDUSTRIAL POLICY: 155 A NECESSITY FOR MOROCCO Under the leadership of King Mohammed VI, While each of these initiatives is discussed in Morocco has proven to be one of the most stable greater detail in the following sections, it is note- economies in the MENA region. This stability, worthy to draw attention to a few distinctive however, could become fragile if insecurity and aspects. Morocco has become highly committed political unrest in neighbouring countries inten- to diversifying its energy mix and thus reducing sifies. Morocco’s late industrialisation process its vulnerability to changes in global fuel prices. and access to European markets following the This is mainly due to its dependence on energy signing of comprehensive trade agreements imports and its wealth in renewable energy enabled development of traditional sectors sources through high solar irradiation and wind such as agriculture and textiles, as well as new speed along the coast. Importantly, the govern- sectors such as automotive and aeronautics. In ment understood that high investments in the spite of such progress, the country faces deep clean energy sector must also deliver socio-eco- challenges to deliver much needed development nomic outcomes. Thus, what emerged from outcomes in terms of jobs and reduced inequal- these initiatives are the seeds of a comprehen- ity (AfDB 2011). Morocco realizes that it also sive policy aiming to advance structural change needs to identify global markets in which it may and competitiveness on the basis of low-carbon, have competitive advantages. resource efficient technologies. These actions, in effect, align with the definition this volume uses The high level of unemployment among the for green industrial policy. Even if the Moroccan educated youth is a persistent problem in government has not labelled their interventions Morocco, and across the entire MENA region, as as a green industrial policy, the direction is clear: is the low rate of female labour force participa- developing a market for renewable energy tech- tion (Nabli and Chamblou 2004). Unemployment nologies; supporting local companies to enter in Morocco reached 9.5 per cent in 2015; but it is this new sector; implementing environmental much higher among educated youth between 15 rules and standards in other sectors, such as and 24 years old, at 22 per cent (Haut Commis- agriculture and housing, to green the economy sariat au Plan du Maroc 2015). Coupled with high and thus aiming for an integrated industrial shares of employment in the informal sector– development strategy. Morocco’s commitment ranging from 15 to 82 per cent, depending on to green the economy is also illustrated by other concept definitions–and underemployment, the national strategies that were initiated since 2010— labour market problem in Morocco is magni- on Sustainable Development, on Environmental fied (Angel-Urdinola and Tanabe 2012; Saif 2013). Protection, on Improvements of the Environment, Moreover, Morocco’s relative competitiveness and the Green Morocco Plan for Agriculture. That has decreased over the past few years, mostly commitment is also demonstrated by the coun- due to a worsening performance with respect to try’s active engagement in the UN Framework business sophistication, innovation, and labour Convention on Climate Change (UNFCCC) COP21 market efficiency (WEF 2015). In response, the in 2015 and its decision to host UNFCCC COP22 government enacted extensive reforms to boost in Marrakesh in November 2016. While progress economic growth, reduce inequality, attract with respect to industrial development in the investment and increase efficiency in govern- sphere of renewable energy has not been as fast ment spending (IMF 2011). Three main initia- as observers may have preferred, advancement tives stand out: the National Energy Strategy; has been steadfast. the Moroccan Innovation Initiative; and the revi- talization of the national industrial development strategy, le Pacte National Pour l’Emergence Industrielle and its later revision le Plan d’Ac- célération Industrielle du Maroc. Green Industrial Policy - Concept, Policies, Country Experiences

156 3. EARLY STEPS TOWARDS RENEWABLE ENERGY DEVELOPMENT The renewable energy strategy was initially The types of technologies targeted for these framed to strictly address energy security investments are primarily concentrated solar concerns. In the process, however, Morocco’s power and a smaller share of solar photovoltaic. government recognised the opportunities that a The objective is to increase the share of solar green industrial policy would offer for its econ- energy in total electric capacity to 15 per cent omy and for its energy sector in particular. In by 2020 and to reduce emissions by 3.7 million

2008, ambitious renewable energy targets were tonnes of CO2 per year (Invest in Morocco 2016). set, aimed at diversifying the energy mix. The Along with the Plan Solaire, new laws and organ- link to industrial policy was also soon articulated. izations were created and large external invest- Although value chain localisation remains limited, ments, as well as development assistance, were Morocco’s strategy remains progressive with engaged. Donors, such as the World Bank and regards to investments, deployment, and value the European Investment Bank, saw vast oppor- creation at regional level and also among other tunities arising from supporting the large-scale developing countries. This section elaborates deployment of concentrated solar power tech- Morocco’s early steps with respect to renewable nologies in North Africa. In particular, cost reduc- energy market creation, supply side measures, tions for this more expensive clean technology and strategic steps towards linking energy and could be captured as well as the cumulative learn- industrial policy, with a focus on the solar energy ing from applying it, benefiting the solar indus- sector. Challenges towards achieving policy objec- try worldwide. Moreover, exploring the extent tives are also highlighted and discussed. to which social and economic benefits could be generated locally–in terms of jobs and local 3.1. MARKET CREATION manufacturing, for example–was also high on the agenda of development cooperation partners To address the very high and persistent depend- (World Bank 2012). ency on energy imports, in 2009 the government From the Moroccan side, the creation of a favour- elaborated the National Energy Strategy aiming to able legal framework has been an important step supply 42 per cent of electricity generation with to enabling the emergence of a market for renew- 2,000 MW from solar, 2,000 MW from wind, and able energy. In particular, Law 13-09 allowed 2,000 MW from hydro sources, thus harnessing medium and high-voltage projects to access the the vast renewable energy resources present in electricity network and opened up competition in Morocco. By 2017, Morocco expected to have 36 this new market. This law also permits projects per cent of installed capacity coming from renew- with capacities of maximum 50 MW to be built able sources: 1,800 MW from hydroelectric energy, and operated by private enterprises. Low-voltage 800 MW from wind, and 180 MW from solar. projects, those of less than 2 MW, would encour- Prior to the COP21 climate negotiations in Paris, age a market for solar rooftop electricity genera- the government renewed its renewable energy tion projects to emerge. They have been discussed targets to 52 per cent by 2030 (MEMEE 2015). Once since 2015 and further amendments of the current these targets are achieved it is expected that by law are still expected (Daouda 2016). 2030 3,440 MW of solar photovoltaic, 1,300 MW of concentrated solar power, and 4,997 MW of wind To ensure the implementation of the Plan Solaire energy will be installed (MEMEE 2015). the government committed US$ 9 billion and created the Energy Development Fund with an The Plan Solaire, the strategy for the solar energy equivalent of US$ 1 billion mostly based on dona- sector, was developed to provide a road map for tions from the Kingdom of Saudi Arabia, the achieving the solar energy targets by construct- United Arab Emirates and a contribution from the ing five large solar power plants for a total of 2,000 Hassan II Fund for Economic and Social Develop- MW. When completed in 2018, the Ouarzazate ment (Invest in Morocco 2016). It also created the Solar Complex of 500 MW will be the largest solar Société d’Investissements Energétique, endorsed farm in Africa. The first part of this complex, ‘Noor by the state and the Hassan II Fund, to support I’ of 160 MW is operational since February 2016, investments in small and medium scale renewa- developed by a consortium of foreign firms, while ble energy projects. `Noor II’ has already been commissioned. Chapter 10 Renewable Energy as a Trigger for Industrial Development in Morocco

A critical development has been the creation of et al. 2013). For instance, a feed-in tariff is lack- a separate agency in 2010, the Moroccan Agency ing and net-metering schemes have been under 157 for Solar Energy (MASEN), specifically tasked with discussion for a long time. To reach a resolution, project development and implementation of the at the end of 2015 the government adopted Law Plan Solaire. More recently, following an instruc- 58–15, which modifies and complements the tion of King Mohammed VI in December 2015, existing 13-09, introducing a net metering scheme MASEN’s role was extended to the entire renew- for solar and wind power plants, but only for those able energy sector, changing its formal name to connected to the high-voltage grid (Daouda 2016). the Moroccan Agency for Sustainable Energy, and A result of this law is that private renewable seeking to coordinate closely with the national power investors will be able to sell 20 per cent of utility Office National de l’ Électricité et de l’Eau their annual production to the grid. It is expected Potable on the implementation of both solar and that the law will be extended to also include the wind energy projects (Schinke and Klawitter 2016). middle and low-voltage projects, including solar MASEN started by organizing the competitive photovoltaic installations. bidding projects for the solar plants, coordinating the combination of grants and loans by multilat- Creating a market for small independent and eral and bilateral donors. MASEN has played an household electricity producers could have signif- important role in the development of the solar icant implications for private sector development sector, due to its high standards, professional- in the short and medium term. Distributed gener- ism and competencies, in terms of engaging both ation using solar photovoltaic has high poten- with international partners and with local policy- tial for local small and medium enterprises to makers and stakeholders. Through its integrated engage in the sector not only in installation and approach to the development of solar energy maintenance, but also in manufacturing parts 23 projects–with a focus on electricity production, and components for the system. A major hurdle industrial integration, training, and research and emerges exactly from the institutional structure development–MASEN also sought to harness that is often prescribed to developing countries, the co-benefits from power generation projects. specifically unbundling of the energy system. In However, its overarching power and stretch in Morocco’s larger urban centers, most electricity is terms of its mission, beyond project development distributed through private entities, called Régies. to foster industrial development and research Fearing loss of customers, these entities have so and development, could weaken the initiatives of far blocked regulatory advancement that would other actors in the larger system (Vidican 2015). support distributed electricity generation through By capturing a wide range of activities under their solar photovoltaics. powerful organizational umbrella, it is likely that Morocco does have previous experience with solar much-needed resources for more specialised energy technologies. In 1996 the government initi- actors could be diverted. More cooperation across ated a project for improving rural electrification, various stakeholders might magnify the positive called the Programme d’Électrification Rurale outcomes. To overcome this risk, MASEN should Globale (PERG). This programme increased the seek to further enhance its cooperation with other rural electrification rate from 15 per cent in 1996 local actors and to develop local competencies to 98 per cent in 2010 (RCREEE 2010: 19). Ten per across the renewable energy sector (Vidican 2015). cent of households, especially in off-grid rural While these developments enabled the imple- areas, received decentralized electrification by mentation of the first projects, several factors installing a solar photovoltaic system on their prevented further market expansion beyond the roofs. Up to 2011, around 60,000 of these systems landmark 160 MW Noor I, which had been mostly had been installed with a total capacity of 4 MW funded by bilateral and multilateral donors. These (PWMSP 2011: 10; RCREEE 2010: 19). This compo- factors include lack of sufficiently targeted incen- nent of the PERG aimed to expand the use of tives for small and medium enterprises, a missing renewable energy and enable local enterprises to regulatory framework for distributed generation gain experience with the production and installa- and small-scale projects such as rooftop photo- tion of small photovoltaic systems (Vidican et al. voltaics, and not clearly specifying the type of 2012). Indeed, most Moroccan enterprises active technologies targeted by the Plan Solaire (Vidican in the solar energy sector have their origins in this early initiative. Estimates of jobs created as a

23 The importance of a large market for incentivising private investments in manufacturing is illustrated by the wind energy sector in Morocco, where projects of more than 850 MW have been commissioned. As a result, Siemens Wind Power and Renewables is building a manufacturing plant for blades, committing an investment of EUR 100 million expected to generate 1,200 direct and indirect jobs (Innovant 2016). Green Industrial Policy - Concept, Policies, Country Experiences

result of the PERG are rare, but evidence suggests sector can emerge. Similarly, water pumping for 158 that 13,000 direct and indirect jobs were created irrigation in the agriculture sector is currently by 2006 and 30 subcontractors, especially system performed using engines powered by subsi- installers, benefited from the rural electrification dised butane gas. Yet, once the market distorting programme (George 2002; Allali 2011a). subsidies are removed, a market for solar water pumps can emerge, with similar gradual but Solar water heaters, a simple but effective tech- positive effects on the small and medium enter- nology, are currently deployed below capacity in prise sector in Morocco. Under increasing budg- Morocco. Yet, an integrated solar water heater and etary pressures, the Moroccan government made photovoltaic support programme was created in major progress in reducing subsidies, especially 2002. Although a number of companies emerged since 2014, with the goal of completely elimi- in manufacturing, installation and servicing, as nating consumer subsidies by 2017 (Verme and a result of a programme called PROMASOL, the El-Massnaoui 2015). market did not develop significantly. Problems in implementation, quality and servicing prevented Despite these institutional and organizational large-scale deployment. More recently, financial reforms, private companies have had limited support mechanisms and technical assistance involvement in manufacturing parts and compo- were established to help entrepreneurs and to nents and in providing services for the emerging trigger investment projects in local manufactur- renewable energy sector. Job creation has also ing of solar water heaters (Allali 2011b). In 2012 a lagged behind. new programme, called Shemsi, was created to capture this untapped potential. The objectives 3.2. SUPPLY SIDE MEASURES are to install 1.35 million m2 of solar water heaters by 2020, reduce the national energy bill by From the beginning of this move toward a green saving on imported and highly subsidised butane economy, the Moroccan government has made it gas, create jobs by supporting local producers of clear that its goal is to harness industrial develop- solar water heaters and improve product quality ment opportunities from its large investments in standards (ADEREE 2016). To avoid the pitfalls renewables. This section first reviews the meas- of the earlier programme, Shemsi has made ures taken to achieve that goal, provides a snap- significant effort in improving the quality of the shot of existing capabilities of manufacturing water heaters, through labelling and certifica- and services in the private sector, and critically tion programmes, accreditation of installers and discusses whether these measures contribute to developing product standards (ADEREE 2016). achieving the stated objectives of building local knowledge and capabilities. To foster further market creation for solar technologies, a particularly important effort is the Moroc- POLICIES AND PROGRAMMES can authorities’ focus on linking solar energy applications to other sectors, such as housing Morocco’s industrial development strategy, le Pacte with rooftop photovoltaic systems and solar water National pour l’Emergence Industrielle 2009–2015 heaters and agriculture with solar water pumps. (le Pacte), has been a successful instrument to The Agence Nationale pour le Développement des orient the development of strategic sectors for Energies Renouvelables et de l’Efficacité Energé- Morocco, such as agriculture, automotive and tique (ADEREE) initiated collaborations with the aeronautics, electronics, textiles and leather Ministry of Agriculture and Ministry of Housing and offshoring services. The strategy consists and Urban Development to integrate renewa- of measures that are both directly targeted at ble energy technologies in new projects. Such specific sectors, as well as cross-sectoral, specifi- initiatives are critical to introducing low-carbon cally improving the competitiveness of small and technologies into the Moroccan society more medium enterprises, improving the business envi- broadly. To address the social development goals, ronment, education and training and developing an important objective is the creation of new industrial parks for new sectors, including P2I– housing units for a growing population. If these Plateformes Industrielles Integrées. Le Pacte, and new housing units are fitted with solar water its 2014 revision Plan d’Accélération Industrielle du heater and photovoltaic systems, an important Maroc,24 has proven to be particularly effective for market for small and medium enterprises in this the automotive and aeronautics sectors, not only

24 An important component of the Plan d’Accélération Industrielle is its emphasis on local content creation through support programmes for the small and medium enterprises, and the formation of ecosystems in strategic sectors. In 2015 a dedicated industrial development fund (Fonds de Développement Industriel) has been created aimed at doubling the local value added in the automotive sector, for example (L’Usine Nouvelle 2014). Chapter 10 Renewable Energy as a Trigger for Industrial Development in Morocco

in terms of higher levels of investment but also in Imtiaz is a national investment competition for terms of moving up the technology ladder in local high-potential enterprises offering grants for up 159 manufacturing and exports (Achy 2015; Ait Ali and to 20 per cent of the total investment costs. The Msadfa 2016). Along with le Pacte, the Moroccan programme objective is to increase turnover, Innovation Initiative was launched in 2009, aiming support export activities and job creation and to to improve Morocco’s global competitiveness. One introduce new technologies or structural changes objective of this initiative is to set industrial clus- within the specific sector (ANPME 2013). ANPME ters and ‘Cities of Innovation’ by 2016 to facilitate also offers individual consulting and support the commercialisation of research and technology assisting companies to elaborate a development transfer in various sectors. strategy, a Plan de Progrès, customised to their individual needs and objectives. The energy sector, however, was not initially targeted by the larger industrial development While these different programmes have been strategy. As industrial development was seen to recognised to be highly beneficial, further improve- be a secondary objective to the energy security ments would be necessary, for example improving goal, it was the Ministry of Energy, Mining, Water standards, visibility, and processing time. Specif- and the Environment (MEMEE) and in particu- ically, many companies are not aware of the lar the new organization MASEN–instead of the services offered by ANPME. Others claim that the Ministry of Industry, Trade and New Technologies trade-off between the time invested to learn about (MCINET)–that were tasked to develop a concept these services and using them is too high, and that for an emerging renewable energy sector. Never- they could also benefit from further customised theless, the cross-sectoral incentives included in le incentives and support to the needs of the renewa- Pacte, and later in le Plan d’Accélération Industri- ble energy firms. Also, companies operating in the elle, were also applied to investments in the renew- renewables sector express need of a clear roadmap able energy sector. In addition, the MCINET notified for industrial upgrading, similar to the one applied potential investors and local companies interested in the automotive and aeronautics sector (Vidican in investing in the renewable energy sector about et al. 2013). a bundle of standard and specific support measures. The bundle includes support for training and MCINET has played an important role in channel- hiring through the Agence Nationale de Promotion ling support for the emerging renewable energy de l’Emploi et des Compétences; benefits stemming sector. But that has been only marginal, relative from free zone status; corporate tax incentives and to its role in the development of other sectors in free repatriation of profits and capital for non-res- Morocco. MASEN’s role has been much more prom- idents. It also offers investment grants from the inent not only by commissioning and developing Fonds de Développement Energétique for new new large scale projects, but also by elaborating a investments in manufacturing parts and compo- vision for the renewable energy sector, particularly nents for renewable energy and energy-efficiency solar. The Office National de l’ Électricité et de l’Eau technologies, up to 10 per cent of the acquisition Potable has been responsible for wind energy. Both cost of new capital goods, but not more than MAD are concerned with value chain localisation and (Moroccan dirham) 20 million. Other incentives enhancement of domestic local knowledge capa- for large-scale investments, larger than MAD 200 bilities. The perceived disconnect and limited coor- million, are also offered with respect to access to dination between these important stakeholders land, infrastructure, training, as well as exemption has been a source of concern for some companies from value added tax and customs duties for three active in this emerging sector (Vidican 2015). years (Invest in Morocco 2013). EXISTING CAPABILITIES The upgrading needs of the small and medium enterprise sector are served by the Agence Nation- Capabilities along the value chain of renewa- ale pour la Promotion de la Petite et Moyenne ble energy technologies are relatively low in the Entreprise (ANPME), which offers not only Moroccan private sector, existing mainly in steel financial incentives but also individual consult- products, civil works, electrical works, electronic ing services through their programmes Mous- equipment and cable manufacturing. Technolog- sanada and Imtiaz. Moussanada offers support ical capabilities required for knowledge-intensive programmes to accelerate the use of information goods and services, such as engineering, design technologies in the firms; to improve strategy, and project development skills, especially for large marketing and organization; and to improve busi- power plants, are currently missing or very weak. ness skills such as production processes, procure- Most companies active in the solar energy sector ment, design and research and development. are involved in the distribution and installation of Green Industrial Policy - Concept, Policies, Country Experiences

imported parts and components for solar technol- on Fraunhofer Research Institutes in Europe and 160 ogies. The local value added from these operations the National Renewable Energy Laboratory in is, therefore, low. Only a few companies operate the US, with the aim of bridging the gap between as manufacturers, and even fewer are active in academia and the private sector and enhancing project development and engineering. As for start- innovation and entrepreneurship in Morocco. To ups, only a couple of companies have emerged in understand and boost capabilities in Morocco’s this sector. small and medium enterprises, IRESEN first organized a series of calls for joint applied research As the technical characteristics of different solar proposals between partners in universities and energy technologies and their value creation local firms. One of the objectives of these calls was potential vary, it is crucial for Moroccan compa- to identify high potential firms and local experts. nies and policymakers to choose carefully what Then IRESEN, supported by the largest globally niche and what technology they work with (Vidi- competitive Moroccan company in the sector of can et al. 2013: 49–51). While solar photovoltaic phosphates, the Office Cherifien des Phosphates, is likely to generate more jobs for installed kWh, created the innovation and education platform concentrated solar power is more capital inten- Green Energy Park. The Green Energy Park is the sive but can address energy security issues more first platform in Africa to create synergies between effectively. Decision-makers have emphasized different research institutions and contribute concentrated power much more than photovol- to building knowledge capabilities in both the taic in the national deployment strategy; however, academic and private sectors. The international local companies have knowledge capabilities partners for the Green Energy Park include Fraun- primarily in the photovoltaic technology and hofer Center for Silicon Photovoltaics, the German its small-scale applications. The small market Aerospace Center, PVcomB Helmholtz Zentrum size for this technology prevents local small and Berlin and the Korean International Cooperation medium enterprises from scaling up their activ- Agency (IRESEN 2017). ities. Maximising the benefits from the large investments in concentrated solar power will Such initiatives and the large investments in the therefore require concerted effort to establish sector also mobilised local universities to coordi- linkages between technology providers and local nate their activities and boost cooperation efforts suppliers. nationally and internationally. The Moroccan Society of Renewable Energy Development and The industry association Association des the Mastère National en Energie Renouvelable et Industries Solaires et Eoliennes has played an en Efficacité Energétique are networking platforms important role both in terms of lobbying for the that emerged along with the national strategy interests of the small and medium enterprises as for renewable energy, not only bringing together well as in linking to other relevant stakeholders, academics and firms, but also finding ways to such as research and academia. But, the market adjust the university curriculum to the needs of the for photovoltaics has been small, until recently, private sector. and only a few manufacturers are present in the sector, so their leverage on policymakers has In a parallel effort, MASEN has been coordinating, remained limited. in collaboration with German development cooperation agencies, the creation of a solar industrial The level of technical education in Morocco is cluster, Cluster Solaire. The concept was launched among the highest in North Africa. However, in 2014 to complement activities at the large renewable energy technologies represent a rela- concentrated power plants in Ouarzazate Solar tively new field of knowledge for the country’s Complex and to support the enhancement of small universities. Basic research on renewable energy and medium enterprises’ capabilities. The Climate technologies started in the 1980s. However, applied Innovation Center in Morocco is also a supporter knowledge remains scattered due to the lack of of this cluster project, thus creating further syner- funding and low interest in the field research and gies between initiatives that were previously education programmes. Following the National disconnected. Among the members of the cluster, Energy Strategy adopted in 2009, curricula for which currently operates mostly as a coordinat- renewable energy education programmes have ing platform, are industry associations and higher been integrated in most universities, and research education institutions, as well as companies platforms have been developed, triggered primar- from relevant sectors such as energy, metallurgic ily by the newly created Institut de Recherche en and electro-mechanics. Its main objectives are Energie Solaire et en Energies Nouvelles (IRESEN). to enlarge the market, increase local value added, IRESEN is a research institute that was modeled develop local competencies, attract financing for Chapter 10 Renewable Energy as a Trigger for Industrial Development in Morocco

innovative projects and develop an attractive regu- technology and know-how between enterprises. latory framework (Cluster Solaire 2016). Local content requirements have also been used to 161 support the transfer of knowledge. In the Moroc- Most of these initiatives have emerged bottom-up, can renewable energy sector, joint ventures are without much coordination through strategic very limited. In the wind sector Nareva Holding, a interventions. For instance, the Ministry of Higher renewables company owned by Morocco’s Société Education, Scientific Research and Professional Nationale d’Investissement, established a joint Formation has been virtually missing from strate- venture with the French energy group GDF Suez to gic discussion about the development of this sector. be responsible for building and operating a wind However, without effective coordination and coop- farm. A more recent development is the UAE’s eration among all these initiatives, the impacts on clean energy company Masdar’s plan to set up a the local economy are likely to remain limited 70 per cent-owned joint venture to develop three solar power plants in Morocco (Yaneva 2015). The BUSINESS LINKAGES relatively small market for renewable energy tech- Given that most investments in renewable energy nologies in Morocco is viewed as the most limiting in Morocco are in large-scale projects, developed factor for developing business partnerships. and implemented by consortiums of technolo- Inter-company training can also be an effec- gy-intensive lead firms, establishing linkages tive mechanism for knowledge transfer. Several between technology providers and local small and companies in the solar energy sector indicate medium enterprises is essential for enhancing that they did benefit from such training activi- domestic capabilities through knowledge transfer. ties (Vidican et al. 2013). Training can materialize Thus they should be a core element of green indus- by sending employees to the ‘mother’ company trial policy. In their cooperation with the smaller abroad for hands-on education in manufacturing enterprises, lead firms or transnational corpora- facilities, by the creation of ‘learning rooms’ in tions define product and process standards along the manufacturing facility in Morocco together the supply chain, thus forcing their local partners with local universities and by operational training and suppliers to upgrade their industrial perfor- not only about the particular installation but also mance (Altenburg 2005). The large production about the technology. capability of transnational corporations, coupled with the flexibility and specialisation of small and Successful supplier development experiences medium enterprises, allows for successful tech- suggest that such programmes work best when nology transfer through spillover and trickle-down driven by lead firms rather than imposed by deci- effects across the entire sector. sion-makers. Yet, although lead firms should select their own partners and design adequate In Morocco, the experience with business linkages, supplier programmes, national agencies can act at both institutional and firm levels, is limited to as facilitators and support with various services few companies in sectors such as automotive, such as upgrading programmes, education and aeronautics and transportation infrastructure. training. The French transnational company, In the emerging solar sector, business linkages Alstom, has an elaborate supplier development remain scarce. The local entity with the direct programme in Morocco that is considered to be mandate of facilitating linkages between foreign the most effective and can be used as a learning companies and small and medium enterprises is platform for others as Alstom is seeking to expand the Bourse Nationale de Sous-traitance et de Parte- its investments in the renewable energy sector. nariat (BNSTP). Yet, due to understaffing, financing Alstom’s experience underlines how important gaps, and lack of a vision with respect to enhanc- it is for local companies to have a certain level ing competitiveness in the private sector, BNSTP of basic standards and processes in place to has remained ineffective and almost invisible to be considered as potential suppliers (Vidican foreign investors and most local companies. This et al. 2013: 133–134). Government-sponsored role has been indirectly played either by industry programmes channelled through ANPME or other associations or regional investment offices. organizations would, therefore, be necessary.

Business partnerships or joint ventures, firm-level Due to restrictions imposed by multilateral finan- training programmes, technology licensing and cial institutions,25 local content requirements supplier development programmes are among for renewable energy investments are set on a the most successful mechanisms to transfer voluntary basis. In the Noor I concentrated solar

25 Morocco has not signed the World Trade Organization plurilateral Agreement on Government Procurement from 1994/1996 and thus mandatory local content requirements are not generally prohibited (Vidican et al. 2013: 136). Green Industrial Policy - Concept, Policies, Country Experiences

power project at the Ouarzazate Complex, a 30 per for industrial development. Currently the indus- 162 cent local content requirement was set, achieved trial development part of the green industrial mainly in the construction phase. The expectation policy is the concern of various stakeholders with is that in the next phases, Noor II and Noor III, a 35 mandates in the renewable energy sector such per cent share of the goods and services will be as MASEN and IRESEN. The MCINET has played sourced locally (Oxford Business Group 2016). But, only an overarching role, not actively engaged as the local contribution in these large projects with the stakeholders of this emerging sector. Yet, comes mostly from civil works and steel structures, a unifying vision and a long-term perspective are opportunities for acquiring technological capabili- necessary and they should be capable of bring- ties remain limited. For local content requirements ing together these initiatives and orienting them to be effective, regardless of whether they are set towards a common goal. The role of MCINET in on a voluntary or mandatory basis, a sizable and this process, with a national mandate from the stable market is needed and a gradual increase Royal Court is critical. Altenburg and Lütkenhorst over time is also important, in consultation with (2015:162) call this vision a ‘national project of relevant stakeholders. The effectiveness of the industrial transformation’. local content requirements also closely depends on other measures aimed at reducing the techno- Such a strategy would need to select renewa- logical gap between lead firms and local small and ble energy technologies that can deliver a higher medium enterprises, such as industrial upgrading value added for the local economy and a higher programmes. A systemic approach to industrial number of jobs, to attract investment and support policy, and green industrial policy in particular, is the establishment of partnerships and joint therefore essential. ventures with local firms and to orient and adjust the education and research sector towards the needs of the new sector. Last but not least, green 3.3. INTEGRATED STRATEGY FOR INDUSTRIAL industrial policy should prepare local small and DEVELOPMENT medium enterprises to engage in the value chain of renewable energy technologies, aiming to All these institutional structures and interrelating strengthen their capabilities through industrial initiatives demonstrate the need for a compre- upgrading programmes, setting up a national hensive national strategy for the development of system of quality management and other targeted the renewable energy sector in Morocco. Without areas of intervention. As the Moroccan market is a strategy that targets both the creation of a siza- small compared to other countries, local compa- ble market for renewables and the development nies should have the opportunity to gain sufficient of a local industry–objectives of green industrial experience at home to then be able to seek busi- policy–stakeholders do not receive the necessary ness opportunities abroad and thus positioning signals to invest and engage in the emerging sector Morocco as a regional market leader. (Vidican et al. 2013). Since 2009, when Morocco’s National Energy Strategy was issued, important Weak coordination of green industrial policy meas- steps towards consolidating and aligning different ures is a reality in Morocco, but it can be addressed. initiatives have been taken. Yet, further steps are Various new agencies have been created to enable needed, as a concrete integrated strategy for indus- the attainment of the renewable energy targets, trial development in the realm of renewable energy agencies that start collaborating with established is still lacking. Le Pacte could serve as an example agencies previously engaged with other indus- in this regard. trial development goals. Bringing together and agreeing on a new vision, new objectives, and new The development of the renewable energy sector ways of thinking about how to integrate traditional requires the intervention of multiple govern- industrial and energy policy tools into a transfor- ment agencies: MEMEE, MCINET, ANPME; levels mational agenda is not an easy task. Nevertheless, of governance: national, regional, local; and vari- proactively seeking to streamline and harmonise ous non-governmental organizations: education, these initiatives and to coordinate across stake- research, civil society, private sector. This reality of holders can contribute to convergence towards green industrial policy calls for coordinating indus- a national vision. Vidican et al. (2013) argue that trial development objectives through a multi-stake- such an integration and coordination effort could holder governance process and ensuring a certain be steered by a new organization–specifically the level of development directionality (Mazzucato Moroccan Solar Energy Council–that would bring 2014; Altenburg and Lütkenhorst 2015). together key stakeholders. Further reflections on The development focus is important for Morocco which actors should take the lead in implement- to succeed in using renewable energy as a trigger ing such actions is necessary, in light of existing governance processes in Morocco. Chapter 10 Renewable Energy as a Trigger for Industrial Development in Morocco

Other aspects to consider are the integration reviews, monitoring, benchmarking, and periodic of learning mechanisms in policy-making and evaluation are desirable features of all incentive 163 strengthening of implementation capacity of the programmes”. But, the implementation capacity state bureaucracy. As OECD (2013:93f) argues, this of state bureaucracy can be, and often is, limited. requires a clear framework for policy evaluation Therefore, investing in improving the administra- and learning, so that incentive schemes and policy tive capacity of the central state bureaucracy is measures can be regularly monitored, evaluated also important. and adjusted. As Rodrik (2009:22) also mentions “conditionality, sunset clauses, built-in programme

4. CONSIDERATIONS FOR CAPITALISING ON EARLY STEPS Following investments in renewable energy, the sector can assist not only local companies but also potential of creating a new competitive indus- encourage lead firms to create linkages with local try in Morocco is significant. These early steps of companies and thus to localise knowledge. So far, linking energy and industrial policy have opened support measures have not been well targeted up opportunities for synergies between different to the specific needs of companies in this sector; initiatives for expanding the market for renewable therefore they have not been very effective. As energy and developing capabilities in the private consortia of technology-intensive lead firms sector. In effect, even if not labelled as such by the have developed the first large projects, critical national authorities, green industrial policy has steps could be made in developing a base of local been enabled in Morocco. Incentives for market suppliers of certain parts and components along creation made the creation of a market for renew- the value chain. Gaining a better understandable energy technologies possible and attracted ing of local manufacturing and service provision much needed investment. Training programmes capabilities is essential. The earlier experience made it possible for skilled and unskilled workers with attracting investments and developing a to participate in the first projects. Education and competitive automotive and aeronautics sector in research programmes contributed to increasing Morocco–through comprehensive supplier devel- awareness of this new sector and of future pros- opment programmes and synergetic initiatives in pects. Lastly, various support programmes for training, education, and research–could offer rele- the private sector triggered interest in exploring vant lessons to the development of a renewable investment opportunities in renewable energy. energy sector. To this end, a more detailed understanding of the factors that contributed to posi- To scale-up these initiatives and their outcomes, tive outcomes in these other sectors is necessary, it is necessary to embed them in a systemic followed by a reflection on lessons that could be approach. A large market offers incentives for transferred to the renewable energy sector. private investors to enter the market, as experienced in the wind energy sector in Morocco. Ultimately, an integrated national strategy is crit- Thus, eliminating regulatory barriers for low and ical, as is a long-term vision for developing this medium-voltage projects is critical in this regard. sector that stresses learning, not only from earlier Opening up this market segment will diversify the steps but also from Morocco’s experience with technology focus beyond concentrated solar power. the development of other sectors. Its favourable That should bring small and medium enterprises location and stable political environment, as well into the supply chain for renewable energy tech- as its consistent efforts to develop their renewa- nologies by manufacturing parts and components ble energy potential, could help position Morocco and providing associated services, in turn contrib- as a regional player in this sector. The success of uting to learning and improvement in competitive- such an effort is conditional on correcting earlier ness. A larger market also opens up opportunities shortcomings, especially with regards to enabling for a wider targeting of technologies with diverse further structural reforms. Areas for attention degrees of local value added potential. include the labour market and business climate, increasing public dialogue, fostering business and Moreover, targeted support measures for upgrading civic engagement and strengthening alignment capabilities in the small and medium enterprise across administrative functions. Green Industrial Policy - Concept, Policies, Country Experiences

164 REFERENCES Achy, L. (2015). Structural transformation and indus- International Monetary Fund (IMF). (2011). Regional trial policy in Morocco (Economic Research economic outlook: Middle East and Central Asia. Forum Working Paper No. 796). Washington, D.C. African Development Bank (AfDB). (2011). Poverty International Renewable Energy Agency (IRENA). and inequality in Tunisia, Morocco and Maurita- (2013). Renewable energy and jobs. Abu Dhabi. nia. Economic Brief. Invest in Morocco. (2016). Investment opportunities Agence Nationale pour le Développement des in solar energy. Retrieved from www.invest.gov. Energies Renouvelables et de l’Efficac- ma/?Id=24&lang=en&RefCat=2&Ref=145 ité Energétique (ADEREE). (2016). Shemsi Institut de Recherche en Energie Solaire et Program. Retrieved from: www.aderee.ma/ en Energies Nouvelles. (IRESEN). (2017). index.php/en/expertise/programmes-integres/ Green Energy Park has officially inaugu- programme-shemsi-en?showall=1&limitstart= rated. Retrieved from www.iresen.org/ Agence Nationale pour la Promotion de la Petite et green-energy-park-official-opening/?lang=en Moyenne Entreprise (ANPME). (2013). Programme L’Usine Nouvelle (2014). Plan d’Accélération Industri- Imtiaz. Retrieved from http://candidature.maro- elle du Maroc: Le Ministre Moulay Hafid Elalamy cpme.ma/imtiaz-croissance/. fait le SAV aupres du patronat. Retrieved from Ait Ali, A., & Msadfa, Y. (2016). Industrial policy, struc- www.usinenouvelle.com/article/plan-d-acceler- tural change and global value chains participa- ation-industrielle-du-maroc-le-ministre-mou- tion: Case study of Morocco, Tunisia and Egypt. lay-hafid-elalamy-fait-le-sav-aupres-du-patronat. (OCP Policy Paper). N29123 Allali, B. (2011a). PROMASOL: Democratizing access Mazzucato, M. (2014). The entrepreneurial state: to solar water-heaters: GIM initiative case study. Debunking public vs. private sector myths (Rev. Allali, B. (2011b). TEMASOL: Providing energy access ed.). New York, NY: PublicAffairs. to remote rural households in Morocco: UNDP Moroccan Ministry of Energy, Mines, Water and Growing Inclusive Markets Case Study. Environment (MEMEE). (2011). Moroccan energy Altenburg, T. (2005). Overview on international good strategy: overview. Rabat. practices in the promotion of business linkages.: Moroccan Ministry of Energy, Mines, Water and Paper prepared for the United Nations Confer- Environment (MEMEE). (2015). The role of solar ence for Trade and Development (UNCTAD). power in mitigation strategies in the MENA Altenburg, T., & Lütkenhorst, W. (2015). Industrial region: The Moroccan Solar Plan and its role in policy in developing countries: Failing markets, bridging the nexus between climate and energy weak states. Cheltenham, Northampton: Edward policies. Presentation given by Dr. Abderkader Elgar Publishing. Amara at the COP21 in Paris. Angel-Urdianola, D. F., & Tanabe, K. (2012). Micro-de- Nabli, M. K., & Chamlou, N. (2004). Gender and devel- terminants of informal employment in the opment in the Middle East and North Africa: Middle East and North Africa region. SP Discus- Women in the public sphere. MENA development sion Paper. Social Protection & Labor. report. Orientations in development series. Wash- ington, D.C.: The World Bank. Cluster Solaire. (2016). General concept. Retrieved from www.clustersolaire.ma/ OECD. (2013). Renewable energies in the Middle groupe-de-travail-collaboratifs/concept-general/ East and North Africa: Policies to support private investment. Competitiveness and private sector Daouda, T. (2016, January 4). Morocco amends development. Paris: OECD. renewable energy law. SeeNews. Retrieved from http://renewables.seenews.com/news/ Oxford Business Group. (2016). Morocco sets morocco-amends-renewable-energy-law-507698 bold targets to boost renewable energy generation capacity. Retrieved from www. George, G. (2002). Electrifying rural Morocco. Trans- oxfordbusinessgroup.com/analysis/moroc- mission and Distribution World. (54,1), 52–55. co-sets-bold-targets-boost-renewable-ener- Hahn, T., & Vidican Auktor, G. (2017). Industrial gy-generation-capacity upgrading in the automotive sector in Morocco Paving the Way for the Mediterranean Solar Plan (DIE Discussion Paper). Bonn. (PWMSP). Country report Morocco. Retrieved Haut Comissariat au Plan du Maroc. (2015). Taux from http://library.euneighbours.eu/sites/default/ de chomage trimestriel selon le sexe. Retrieved files/attachments/Country_report_-_Morocco_-_ from www.hcp.ma/Taux-de-Chomage-selon-le- activity_1_4-1.pdf sexe_a256.html Innovant. (2016). Siemens: Construction d’une usine de pales d’eoliennes au Maroc. Magazine Inno- vant. Energie & Environment. (Avril), 40–41. Chapter 10 Renewable Energy as a Trigger for Industrial Development in Morocco

Regional Center for Renewable Energy and Energy Efficiency (RCREEE). (2010). Provision of techni- 165 cal support/services for an economical, technological and environmental impact assessment of national regulations and incentives for renewable energy and energy efficiency: country report Morocco. Cairo. Rodrik, D. (2009). Industrial policy: Don’t ask why, ask how. Middle East Development Journal. (1(1)), 1–29. Saif, I. (2013). The bloated informal economies in Arab countries. Retrieved from: http://carneg- ie-mec.org/2013/02/12/bloated-informal-economies-in-arab-countries-pub-50966 Schinke, B., & Klawitter, J. (2016). Country Fact Sheet Morocco: Energy and development at a glance 2016. Background Paper. Middle East North Africa Sustainable Electricity Trajectories (MENA-SE- LECT). GermanWatch: Bonn. United Nations Environment Programme. (2011). PROSOL: Financing solar water hearing in Tunisia. Verme, P., & El-Massnaoui, K. (2015). An evaluation of the 2014 subsidy reforms in Morocco and a simu- lation of further reforms. Policy research working paper: Vol. 7224. Washington, D.C.: World Bank. Vidican, G. (2015). The emergence of a solar energy innovation system in Morocco: A governance perspective. Innovation and Development. (5(2)), 225–240. Vidican, G., Böhning, M., Burger, G., Sigueira Regue- ira, E. de, Müller, S., & Wendt, S. (2013). Achieving inclusive competitiveness in the emerging solar energy sector in Morocco. Studies / Deutsches Institut für Entwicklungspolitik: Vol. 79. Bonn: German Development Institute. World Bank. (2012). Inclusive green growth: The pathway to sustainable development. Washington: The World Bank. World Economic Forum. (WEF). (2015). The Global Competitiveness Report 2014–2015. Geneva. Yaneva, M. (2015). Masdar eyes wind and solar power investments in Morocco. Retrieved from http:// renewables.seenews.com/news/masdar-eyes- wind-solar-power-investments-in-morocco-report-496446. Green Industrial Policy - Concept, Policies, Country Experiences 166 CHAPTER 11 GERMANY: THE ENERGY TRANSITION AS A GREEN INDUSTRIAL DEVELOPMENT AGENDA

Anna Pegels Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

1. THE ENVIRONMENT AS A DRIVER 167 OF GERMAN INDUSTRY Generating about 20 per cent of Germany’s gross of quantitative targets for greenhouse gas reduc- domestic product (GDP) in 2014, manufacturing tion, renewable energy, and energy efficiency, value added plays a major role in the German which are discussed in more detail in section 2. economy (UNIDO 2016). Among EU countries, only the Czech Republic has a slightly higher share of The environmental targets form an integral part manufacturing value added in GDP. The EU aver- of the Energiewende, but a green industrial policy age stands at 14 per cent. strives not only for environmental objectives, but also for a complex set of economic and social Environmental goods and services have rapidly objectives. Typically, the latter include technolog- risen in importance for the German economy. ical innovation, competitiveness, value added, and According to Statistisches Bundesamt (2015), employment creation. These aims and the policy they provided about 260,000 jobs and generated measures of the Energiewende are discussed in a turnover of about EUR 66 billion in 2013. About section 3. 77 per cent of the jobs were located in manufacturing, producing about EUR 51 billion in turno- Other countries, among them numerous develop- ver. The Federal Ministry for the Environment, ing and emerging countries, have implemented using a broader definition, even calculates a or are planning similar sustainable energy poli- market volume for environmental technologies cies, such as renewable energy feed-in tariffs, and resource efficiency of EUR 344 billion in 2013, although with less ambitious technology deploy- which is 13 per cent of German GDP (BMUB 2014). ment objectives. Germany, and the successes It estimates the market will grow by a yearly aver- and setbacks of its Energiewende, are thus under age of 6.6 per cent to EUR 740 billion in 2025, at close international scrutiny. While the jury is that point accounting for more than 20 per cent still out on the success or failure of the Energie- of GDP. wende, the discourse in recent years has been vehement. Section 4 sets out to contribute some This development is supported strongly by objectivity to that discourse by analysing data on German policy making, most notably by the patents, trade, employment, and value added as Energiewende, or energy transition. With indicators for the success of the Energiewende as competitiveness as an explicitly stated aim, the Germany’s green industrial policy flagship. From Energiewende cannot be regarded as only an this analysis, and a subsequent discussion of the environmental policy, but needs to be seen as a achievements and remaining challenges of the green industrial policy. It certainly is one of the Energiewende, section 5 aims to derive lessons most ambitious energy transition ventures glob- for other countries’ energy transitions; and, by ally. It aims to steer the German energy sector presenting other countries’ lessons, for Germany. towards sustainability by meeting a combination

2. AIMS AND POLICY MEASURES OF THE ENERGY TRANSITION The stated policy aims of the Energiewende are Jungjohann (2016) sees this array of aims as (BMWi 2015a): building blocks for the success of the Energie- wende. Combining the various aims enables the ◼◼ Greenhouse gas emission reductions forging of coalitions that are vital for the imple- ◼◼ A complete phase-out of nuclear power by mentation of a national project of such order and 26 2022 socio-technological complexity as the Energie- ◼◼ Competitiveness wende, and the grounding of the Energiewende in ◼◼ Security of energy supply. society through the facilitation of energy cooperatives may have been one of the building blocks of much needed societal support (Schmitz 2016;

26 The decision to phase out nuclear power in Germany goes back to a longstanding societal discourse, which gained impetus after the nuclear catastrophe in Chernobyl in 1986. After the catastrophe in Fukushima in 2011, the German government decided to phase out all nuclear power generation by 2022. Green Industrial Policy - Concept, Policies, Country Experiences

Jungjohann 2016). Nonetheless, several windows the share of renewable energies and increas- 168 of opportunity were necessary to overcome resist- ing energy efficiency, with both aims tracked in ance from incumbent industries, among them the the electricity, heat and transport sectors. The German reunification, the discussions around an German government has set a number of quanti- environmental tax reform, and the nuclear inci- tative mid- and long-term targets for these sectors dent in Fukushima (Jungjohann 2016). (Table 11.1).

The policy aims of the Energiewende were sepa- rated into the two strategic aims of increasing

Table 11.1: Quantitative targets and status quo 2014 of the Energiewende

2014 2020 2030 2040 2050

Greenhouse gas emissions

Greenhouse gas emissions at least -80% -27% at least -40% at least -55% at least -70% (compared to 1990) -95%

Renewable energy

Share of gross final energy 13.5% 18% 30% 45% 60% consumption Share of gross electricity 27.4% at least 35% at least 50% at least 65% at least 80% consumption Renewable Renewable Energy Sources Energy Sources Act 2025: Act 2025: 40–45% 55–60% Share of heat consumption 12% 14% Share in transport sector 5.6%

Efficiency and consumption Primary energy consumption -8.7% -20% -50% (compared with 2008) Final energy productivity 1.6%/year 2.1%/year (2008–2050) (2008–2050) (2008–2014) Gross electricity consumption -4.6% -10% -25% (compred with 2008) Primary energy consumption in -14.8% -80% buildings (compared with 2008) Heat consumption in buildings -12.4% -20% (compred with 2008) Final energy consumption 1.7% -10% -40% (compared with 2005)

Source: BMWi (2015d: 4).

To achieve the aims of the Energiewende, the complement various loan programmes for renew- German government introduced a number of able energy, research and development support policy measures. The most significant and most through the Federal Energy Research Programme disputed measure is the system of feed-in tariffs and sector-specific innovation and cluster support for renewable energy, established in 2000 in programmes (Lütkenhorst and Pegels 2014). the Renewable Energy Sources Act, Erneuer- bare-Energien-Gesetz. These preferential tariffs Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

3. HOW TO MEASURE PROGRESS TOWARDS 169 ACHIEVING THE ENERGIEWENDE AIMS? Measuring progress on the targets set out in Table the target of reducing emissions to a maximum

11.1 is relatively easy, and the quantitative indica- of 750 Mt CO2e by 2020, in accordance with a 40 tors for renewable energy deployment, energy effi- per cent reduction from 1990 emissions levels. ciency, and greenhouse gas emission reductions One reason for the slow progress in emissions are helpful to assess the progress of the techno- reductions is the trading of electricity in the spot logical dimension of the Energiewende. market, combined with an electricity oversupply from unexpected levels of renewable electricity Progress on the targets has so far been mixed. generation. This oversupply led to a decrease in Renewable energy shares have been rising stead- electricity spot-market prices (Figure 11.2). The ily, reaching more than 30 per cent of gross elec- depressed price crowded the more expensive but tricity consumption in 2015 (BMWi 2015a). In this cleaner fossil fuel options out of the market first. regard, the Energiewende is well on track. The So natural gas generation, for example, slows first, other targets have seen less progress, though. while cheaper sources, such as lignite, remain– Greenhouse gas emissions declined rapidly in and lignite is the most polluting electricity source the first years after the German reunification in Germany. In sum, growth in renewable energy in 1989, but this decline was largely due to the sources triggers a market response that results dismantling of Eastern German heavy industry. in the increased use of the most polluting fossil Since then emission reductions have slowed, fuel source, lignite, because it is the cheapest. and since 2008 in particular no substantive So, emission levels remain stagnant, despite emissions reductions occurred. As shown in rising shares of renewables and improvements in Figure 11.1, efforts need to be reinforced to meet energy efficiency.

Figure 11.1: German greenhouse gas emissions 1990–2015, target 2020

(in million tonnes CO2 equivalents) 1,200

1,000

750 800

600

400

200

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020

Source: Umweltbundesamt (2017). Note: Figure includes emissions from land use, land-use change and forestry (LULUCF). Green Industrial Policy - Concept, Policies, Country Experiences

Figure 11.2: European Energy Exchange electricity spot-market prices in Germany (green, trend as 170 dotted line, EUR / MWh, left axis) against renewable electricity generation (orange, TWh, right axis)

100 250

80 200

60 150

40 100

20 50

0 0

2017

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

Source: based on BMWi (2017a), BMWi (2017b).

Improvements in energy efficiency have not been Second, they will draw from aggregate indices and sufficient either. Instead of the targeted 2.1 per cent survey data such as the Global Cleantech Inno- per year, final energy productivity improved only vation Index of the Cleantech Group, the Global by an annual 1.6 per cent between 2008 and 2014. Competitiveness Index of the World Economic Final energy consumption in transport saw a small Forum, and the Energiewende Barometer of the increase of 1.7 per cent between 2005 and 2014. German Chamber of Industry and Commerce.

While the targets are useful to measure the tech- The technical details of data use and sources are nological and environmental progress of the Ener- discussed in the respective subsequent sections. giewende, they do not capture the industrial policy Two caveats are in order: First, since ‘green energy’ aspects of Energiewende aims in their entirety. is not a category usually captured by conventional Measuring the success of the Energiewende as statistical classification, providing a harmo- an industrial policy agenda thus requires addi- nised picture across all indicators is notoriously tional consideration. Two questions need to be difficult. Depending on definitions, data -availa answered: First, has the Energiewende enhanced bility and database groupings, this assessment the competitive position of green industries, and will concentrate on a selection of typical green second, have effects on the performance of other energy technologies, such as renewable energy industries, for example through electricity price and energy storage technologies. Since most changes, been positive or negative? detailed data are available for renewable energy technologies, these will be assessed in subsequent To answer these questions, the subsequent sections, but other technologies are brought into sections will use various types of data. First, to the analysis where possible to provide a broader indicate progress on typical industrial policy aims picture. Second, effects on the competitiveness in green industries, they will use: of industries other than the selected ones, for example through electricity price changes, will be ◼◼ Trade data indicating competitiveness of green discussed in a more qualitative and descriptive industries on international markets way, since a comprehensive and detailed discus- ◼◼ Patent data and research expenditure indicat- sion of effects on all German industries would be ing innovativeness in green technologies beyond the scope of this chapter. ◼◼ Data on value added by green industries ◼◼ Employment data to assess the job creation potential of green industries. Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

4. THE ENERGIEWENDE AS INDUSTRIAL 171 DEVELOPMENT AGENDA

4.1. INNOVATION expenditure as the input indicator and patents as the output indicator of innovativeness in clean The World Wildlife Fund and Cleantech Group technologies. Data is sourced from the data ware- developed and biannually calculate the Global house of OECD, OECD.stat, which includes data Cleantech Innovation Index, which gives an and metadata for OECD countries and selected aggregated overview of countries’ innovation non-member economies (OECD 2015). These data- potential in clean technologies. Among 40 coun- bases provide pre-selected technology groupings tries, it seeks to identify those with “the greatest such as renewable energies or hydrogen and fuel potential to produce entrepreneurial cleantech cells, which will be used in the subsequent anal- start-up companies that will commercialise clean ysis. Energy efficiency technologies will not be technology innovations over the next 10 years” included in the analysis since drawing the bound- (Sworder et al. 2017:3). It uses 15 indicators related aries for this technology group is particularly chal- to the creation, commercialisation and growth lenging (Ekins et al. 2015:44). Any data presented of clean technology start-ups. According to this would thus be open to criticism, and comparability th index, Germany ranked 8 in 2017, after Denmark, across databases would be very limited. Finland, Sweden, Canada, the US, Israel, and the UK. Germany’s score in general innovation drivers Data on research and development expenditure in is relatively low, with a rank of 19. The 2014 report energy technologies refer to public expenditures attributed this weakness to a lack of “positive atti- (Figure 11.3). As can be seen, total German expend- tude towards entrepreneurship or level of early iture for research and development of renewable stage entrepreneurial activity” (Parad et al. 2014: energies has been following a clear upward trend 41). The 2017 report confirms the weakness in in recent years. Within OECD countries, Germa- early-stage entrepreneurship, ranking the coun- ny’s 2014 expenditure for renewable energies try second-to-last in this indicator, just ahead of research ranked at 3rd, for nuclear energy at 3rd, Italy. In contrast, Germany scores first in environ- for hydrogen and fuel cells at 3rd, and for energy mental patents, and is strong in clean technol- storage at 2nd, compared to a 6th rank in fossil fuel ogy exports and renewable energy jobs. Despite research (OECD 2015). Since 2010, renewable ener- the low level of early-stage entrepreneurship, its gies have been the most highly funded research strong industry and established manufacturing area. Nuclear energy research funding has stag- sectors allow it to efficiently convert innovation nated at a relatively high level, while funding for inputs into outputs. research on storage technologies has grown from a lower level. Hydrogen and fuel cells and fossil To add detail to this general picture, this section fuels have received comparatively little public will use disaggregated data on German research research funding. Green Industrial Policy - Concept, Policies, Country Experiences

Figure 11.3: Research, development and deployment budget energy technologies, Germany (million EUR) 172 300

250

200

150

100

0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Energy efficiency Fossil fuels Renewable energy sources Nuclear Hydrogen and fuel cells Other power and storage technologies

Source: OECD (2015).

While research budgets are useful as input indi- share in all other technology patents. Relative cators, they do not provide information on the patent share is normalized to assume values success of research activities. This success is between -100 and +100. A positive value indicates often measured by patent data, since patents can that the world share of the patent for a certain be considered as an output of research activities. technology is larger than that of all patents of the same country, and values higher than 20/lower One possible measure of the national innovative than -20 indicate an innovative strength/weak- strength in specific technologies is the relative ness in the assessed technology (Umweltbunde- patent share, which compares a country’s world samt 2014). patent share in the specific technology with its

Figure 11.4: Relative patent shares in relevant energy technologies, Germany

0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

-20

-40

-60 Renewable energy sources Energy storage Hydrogen and fuel cells

Source: OECD (2015). Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

While the German relative patent share values for 4.2. COMPETITIVENESS renewable energy generation have been declin- 173 ing, they are consistently positive at values WORLD MARKET SHARES IN CLEAN higher than 20 (Figure 11.4). Germany thus has an ENERGY TECHNOLOGIES innovative strength in these technologies, likely Many indicators are available to measure the mirroring the supportive policy environment of international competitiveness of industries and the Energiewende. The decline is likely attribut- most involve trade data. One of the most intuitive able to increasing competition from non-OECD indicators is the world market share, which illus- countries, in particular China (Ekins et al. 2015). trates the share each country has in world exports The growth of the Chinese domestic market, for a given product. This section concentrates on the location of international renewable energy Germany’s trade performance in selected renewa- firms in China and the increasing importance ble energy technologies compared to its most rele- of Chinese domestic firms have contributed to vant competitors in the same technologies. increased innovative activity. Germany’s world market share in wind energy Values for hydrogen and fuel cell technologies are converters has been fluctuating around 30 per cent positive as well, but below 20 in most years. The for the past ten years (Figure 11.5). The low world picture for energy storage is less positive: while market share before 2005 is explained by the fact relative patent share values have been increas- that in those years Germany represented a lead ing, they exceeded values of -20 only from 2007, market for wind energy—accounting for 45 per and have been oscillating around zero since. This cent of wind energy converter installations world- shows an innovation weakness in energy storage wide in 2002, but falling to 7 per cent in 2005. The technologies. Competition from Japan and China pioneering feed-in tariff introduction had created is particularly strong: between 2004 and 2013, 39 such a strong domestic market pull that early per cent of patent applicants were Japanese, 28 export efforts were effectively stifled (Lütken- per cent Chinese, and only 7 per cent German (UK horst and Pegels 2014). Since 2008 Germany has Intellectual Property Office 2014). been competing for world market leadership with Denmark, which has been a dominant player since the early stages of the technology.

Figure 11.5: World market shares wind energy converters

100

0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Spain Denmark India Germany

Source: WITS database, World Bank (2015). Green Industrial Policy - Concept, Policies, Country Experiences

Beyond the aggregate data presented in the chart, power sectors in general. Without the foundation 174 industry analysts underline the particularly of highly advanced manufacturing capabilities strong competitive position of German companies and skills across a whole range of industries, the when it comes to offshore turbines and offshore German wind energy sector would not have been wind parks in general, as well as large-scale able to achieve its current global position. Argu- onshore turbines above 5 MW capacity (Lema et al. ably, the North Western cluster represents an 2014; Lütkenhorst and Pegels 2014). Most recently internationally unique level of sophistication and the German company Siemens, which is leading comprehensiveness, with business players along in offshore wind technology, bought 59 per cent of the entire value chain exhibiting a high intensity the Spanish company Gamesa, which has a strong of interactions based on shared ambitions and position in emerging country markets (Spiegel quality standards. The cluster represents a genu- 2016). With about EUR 9.3 billion in annual sales ine public-private partnership and is co-funded by volume, the merger has created one of the world’s state resources and business membership fees. biggest multinational wind companies. In the field of solar photovoltaic cells, in contrast, A particular driver of competitive strength origi- Germany was not able to gain a strong market nates from a classic technology cluster in the four position. While the country’s global market share Northern states of Lower Saxony, Schleswig-Hol- increased from about 7 per cent in 2000 to 15 per stein, Bremen and Hamburg. This North Western cent in 2008, it has fallen back to 2000 levels since Wind Power Cluster has grown into a densely (Figure 11.6). Background data show that German interconnected web of more than 300 partners exports of solar photovoltaic cells were almost cut including globally leading turbine manufactur- in half between 2010 at US$ 8.1 million and 2012 ers, specialised component suppliers, wind park at US$ 4.5 million (Lütkenhorst and Pegels 2014). operators, local governments and cutting-edge Several German photovoltaic companies went research institutions. The cluster boasts some bankrupt during this time, such as Solon, Q-Cells of the industry’s major innovations, such as the and Odersun (Parad et al. 2014). At the same time, development of the 5 MW offshore turbines and the rapid increase of the Chinese world market the offshore test site Alpha Ventus. share is notable. Remaining below 5 per cent until 2004, it quickly rose to 45 per cent in 2014, making At the same time, the wind cluster also owes China the clear world market leader in the manu- some of its success to the long-standing track facture and export of photovoltaic cells. record of Germany’s engineering, machinery and

Figure 11.6: World market shares solar photovoltaic cells

0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Japan China Malaysia Germany

Source: WITS database, World Bank (2015). Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

Beyond the trends on photovoltaic cell manufac- PERCEIVED IMPACTS ON COMPETITIVENESS IN turing, the strong competitive position of German OTHER SECTORS 175 photovoltaic system component manufacturers When assessing the effects of the Energiewende and equipment suppliers must be emphasized. on competitiveness, implications for new green Data for 2011 show that the share held by German industries are a central factor. However, indirect firms in the global market for specialised photo- consequences for established industries need to voltaic equipment was as high as 50 per cent, and be considered as well to obtain a more holistic the market share of photovoltaic inverters that view of net effects. These consequences are noto- transform the direct cell current into alternating riously difficult to measure, since establishing grid current stood at 35 per cent (Germany Trade causality can be challenging. To provide a rough & Invest 2013). With a share of approximately 25 picture of possible effects, aggregate data from the per cent, the German company SMA Solar is the World Economic Forum’s Global Competitiveness world market leader for inverters. However, the Index and the results of surveys conducted by company has recently been suffering from over- the German Chamber of Industry and Commerce, capacity and had to announce significant job cuts Deutsche Industrie-und Handelskammer (DIHK) in the first quarter of 2013 (Lütkenhorst and Pegels help illuminate the effects of the Energiewende 2014). In total, employment in the German photo- on German industry. voltaic industry nearly halved in 2013 and 43,900 people lost their jobs (Lehr et al. 2015). On an aggregate level, the World Economic Forum’s Global Competitiveness Index does not In sum, Germany was successful in creating a confirm a clear upward or downward trend for competitive wind energy industry that can now German competitiveness over the past years. Out reap economies of scale, while the picture is of 140 countries, Germany ranked 6th in the 2012– less impressive for the solar photovoltaic indus- 13 Report, then 4th, then 5th, and 4th again in the try. Industrial policy investments in photovoltaic most recent Report of 2015–16 (World Economic cell manufacturing did not bear fruit, which is Forum 2015). Barriers to German competitive- not very surprising, since cell manufacturing is ness are mostly attributed to the complexity of not overly sophisticated and cells and modules tax regulations and high tax rates or to inefficient can easily be transported, making proximity to government bureaucracy and labour market rigid- markets a less important factor. A large part of ities, but not to energy price increases. cell manufacturing activity thus relocated to China. However, while some firms went bankrupt, Focusing on the Energiewende, the DIHK presents other parts of the German photovoltaic industry the results of annual company and energy expert concentrated on more sophisticated activities surveys in its publication ‘Energiewende-Barom- upstream as cleanroom production facilities for eter’. The first surveys were conducted in 2012, the manufacturing cells, for example, and down- most recent ones in 2017. In contrast to the results stream as system components such as invert- of the Global Competitiveness Index, the overall ers, mounting structures, and cabling. What we perception of Energiewende effects on compet- observe in the photovoltaic industry is a fairly itiveness seems to be slightly negative, with a typical transition: when cell manufacturing was clearly negative perception in the industrial sector still a cutting edge technology, Germany was a (Figure 11.7). Between 2012 and 2014, about half of leading producer. Low cost assembly in China the companies in this sector replied that the Ener- crowded German producers out when the tech- giewende affects them negatively or very nega- nology matured, but German industry was able to tively. However, there seems to be a trend towards build competitive advantages in specialised high- a more positive perception. The share of compa- value niches of that industry. German industrial nies reporting a positive or very positive effect is policy could have been more successful had it rising, and the share reporting negative effects is supported this transition earlier by focusing on declining (DIHK 2012, 2013, 2014, 2015). In 2017, the technologically sophisticated steps in the photo- average assessment of Energiewende effects on voltaic value chain. competitiveness was very slightly positive. Green Industrial Policy - Concept, Policies, Country Experiences

Figure 11.7: Perception of Energiewende impacts on competitiveness 176 (in per cent of company responses, 2012–2015)

P o s i t i v e / v ery positive

ors

All sect N e g a t i v e / v e r y n e g a t i v e 2012 2013 2014 2015 2016 P o s i t i v e / v ery positive

Indust Negative/very negative

0 10 20 30 40 50 60

Sources: DIHK (2012), DIHK (2013), DIHK (2014), DIHK (2015). Note: Values not adding up to 100 per cent indicate missing or neutral responses.

The main reason for negative assessments seems from this effect are estimated at about EUR 500 to be price. While electricity stock market prices million (Kemfert et al. 2015). A recent study on the have decreased (see Figure 11.2), grid and renew- competitiveness effects of energy price changes able energy fees have increased. DIHK calculates due to the Energiewende concludes that many that fees and taxes account for 81 per cent of energy-intensive companies offering premium electricity costs to those companies that are not products, such as high quality steel or alumin- eligible for price reductions (DIHK 2015). Energy ium, would be able to stay profitable when having security, which is crucial for the industrial sector, to pay the renewable energy surcharge (Ecofys, is seen as high. Fraunhofer ISI 2015). However, many companies producing lower quality products, and thus having It is unclear, however, whether companies see to compete globally on the basis of price, would the ‘Energiewende-Barometer’ report as a means face economic losses. These companies might to lobby government for lower electricity prices in the medium term consider relocation to other or further industry exemptions to the renewable countries if they had to pay the full surcharge. It electricity surcharge. Past attempts have been should be noted, however, that companies’ loca- successful, and it would be in the interest of tional decisions depend on many additional industry to see these exemptions prolonged and factors, including network effects and proximity broadened. Currently, the industrial sector has to consumers. Nonetheless, the study comes to to bear only 30 per cent of the surcharge, while it the conclusion that in total the exemptions have a accounts for almost 50 per cent of German elec- positive net effect on the German economy. Using tricity consumption. Private households with an a macroeconomic model, the authors estimate electricity consumption share of roughly 25 per an average increase in production costs of 3.5 cent have to bear 35 per cent of the surcharge per cent in 2020, were all exemptions abolished. (Lütkenhorst and Pegels 2014). In 2013, the elec- Positive effects from cost relief of households tricity surcharge–the rise in electricity price and currently non-privileged industry would not attributable to the feed-in tariffs–for consumers suffice to outweigh the negative effects on the who did not benefit from exemptions amounted economy: Exports in 2020 would be 0.3 per cent to EUR 5.3 ¢/kWh, rising to EUR 6.24 ¢/kWh in or EUR 4.7 billion lower than in the reference case 2014 (Lütkenhorst and Pegels 2014; Kemfert et with current exemptions, and GDP would be 0.15 al. 2015). In addition, exempted industries bene- per cent or EUR 4 billion lower (Ecofys and Fraun- fit from the decrease of the wholesale electricity hofer ISI 2015). spot-market price caused by increasing amounts of renewable electricity fed into the grid, the ‘merit order effect’ (Figure 11.2). The annual benefits Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

4.3. VALUE ADDED and renewable energy industries. EUR 5 billion are directly attributable to renewable energies, EUR 177 If data indicating innovation or competitiveness 15 billion to conventional energies. The remain- needs to be treated with care since the boundaries ing EUR 48 billion, mostly added in the value of the category ‘clean energy technologies’ are not chains of electricity and district heating, cannot clearly defined, this caveat is even more relevant be directly attributed. in the case of value added. Not only is it necessary to draw technology boundaries, but also bounda- Compared to conventional energies, the value ries between economic activities. added by renewable energies calculated by Prog- nos (2015) seems to be low. One reason is the This may be one reason why only a few stud- relation of forecasted investment cost and oper- ies exist on the Energiewende‘s effects on value ation cost of renewable energies. Most renewa- added in Germany, and why they mostly concen- ble energies, such as solar or wind energy, do not trate on renewable energies as a relatively clearly require constant fuel inputs. They are, however, defined technology group. Two such studies merit capital-intensive. Compared to conventional particular attention, Aretz et al. (2013) and Prog- energies, a higher share of value added is there- nos (2015). fore generated in the installation phase. Another Aretz et al. (2013) calculate value added in reason is the conservative calculation method Germany in 2012 by renewable energies, includ- applied by Prognos, only considering clearly ing wood fuel, heat grid, heat pump, solar power, attributable value added for the group of renewa- bioenergy, deep geothermal, hydro power and ble energies. Value added in the statistical cate- wind energy. In addition to municipal value added gory for Energieversorgung, or energy supply, is they calculate tax revenues at state and federal allocated to the group of conventional-renewable levels, and they calculate direct as well as indi- energy mix. With EUR 45.8 billion, this category rect effects from upstream activities by using an contributes the main share to the non-attributa- input-output model. Direct value added comes to ble group, and indeed to total value added in the EUR 16.9 billion, EUR 11.1 billion of which accrue energy sector. With more fine-grained statistical to the municipal level. Of the remaining EUR 5.8 methods, part of this value added could be attrib- billion, EUR 4.5 billion are federal and EUR 1.1 utable to renewable energies. When taking the billion state taxes. share of renewables in the electricity mix in 2011, which was 25 per cent, as a proxy for the share In terms of value added along the value chain, of renewables in the value added of the mixed the operation-related steps of plant and system category, the total value added of renewables in operation and maintenance contribute the largest Prognos (2015) would correspond to about EUR 16 share with a total of EUR 8.0 billion, or 47 per cent billion, which is comparable to the value calcu- of total value added. They are followed by systems lated by Aretz et al. (2013). and components manufacturing with EUR 6.6 billion, and planning and installation with EUR 4.4. EMPLOYMENT AND 2.3 billion. INCLUSIVENESS EFFECTS Indirect value added by upstream activities add Distinguishing employment effects due to clean another EUR 8.5 billion in 2012, EUR 6 billion of energy technologies is another difficult task, simi- which accrue to the municipal level, with the rest lar to the problems with value added (Haider et al. accruing to federal and state levels. Total direct 2017, this volume). Accordingly, most studies refer and indirect value added thus adds up to EUR 25.4 to renewable energies. Study results vary with billion. About EUR 16 billion stem from wind and the methodologies applied, in particular with the solar energy. consideration of indirect employment effects in Prognos (2015) calculates value added for the upstream sectors. An overview of the results of entire German energy sector. In 2011, a total value three recent studies on employment created in of EUR 68 billion was added by the conventional the German renewable energy sector offers possible insight (Table 11.2). Green Industrial Policy - Concept, Policies, Country Experiences

Table 11.2: German gross employment in renewable energies, overview of study results 178 Study Aretz et al. (2013) Lehr et al. (2015) Prognos (2015)

Gross 166,000 in 2012, 371,400 in 2013, 104,000 in renewable energies in 2011; a employment only direct effects direct and indirect further 293,000 not clearly attributable to effects conventional or renewable energies

Lehr et al. (2015) provide a valuable overview of in the solar photovoltaics industry. While the job developments from 2004 to 2013 (Figure 11.8). number of people employed in solar cell manufac- While the number of people employed in the turing nearly halved in 2013, employment in the different renewable energy technologies grew other technologies remained relatively constant considerably from 2004 to 2012, it declined in 2013. or increased. This is almost entirely attributable to job losses

Figure 11.8: German gross employment in renewable energies, 2004–2013

450

400

housan T 350

300 2004

250 2007 2012 200 2013 150

100

l ta

Wind

Sola

Hydr

admin.

thermal

lic R&

omas Bi

Pub Source: Lehr et al. (2015). Note: Values for 2013 are estimates. Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

Nestle et al. (2016) contrast total employment However, the results of the comparison strongly created by renewable energies with total employ- depend on attribution methodology and indus- 179 ment created by the German coal industry. Their try boundaries. Prognos (2015) calculate a total of analysis indicates that renewable energies create 140,000 jobs created by conventional energies in considerably higher numbers of jobs, despite 2011, including jobs in hard coal, lignite, natural gas recent job losses in the photovoltaic industry and mineral oil products, as compared to 104,000 (Figure 11.9). Furthermore, it shows declining jobs clearly attributable to renewable energies. employment numbers in German hard coal mining.

Figure 11.9: Employment in renewable energy and coal industries, Germany, 2002–2015

400

350

housan

300

250

200

150

100

2015

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Renewables Hard coal Lignite

Source: Adapted from Nestle et al. (2016: 9), data from Lehr et al. (2015), O’Sullivan et al. (2015), Statistik der Kohlewirtschaft e.V. (2016a), Statistik der Kohlewirtschaft e.V. (2016b).

When assessing the effects on inclusiveness, not only employment effects but also households’ price increases merit consideration. These are increasingly causing concern in the German case, and the distribution of additional costs of the Energiewende has become subject to a heated debate. The principal idea of the German support scheme is to subsidise renewable energy technologies by adding a surcharge to the price of electricity. This surcharge entails a net cost factor and a distributional factor (Figure 11.10). Green Industrial Policy - Concept, Policies, Country Experiences

Figure 11.10: Feed-in tariff components 180

Costs per kWh

FiT rat e Markup as compensation for a d d e d i n v e s t m e n t risk Generat ion costs of r e n e w a b le electricity Difference in generat ion costs

Generat ion costs of conventional electricity Generat ed r e n e w a b le electricity

Source: Anna Pegels. Note: The areas do not reflect the actual size of cost components.

This differentiation is essential, although it is not costs–the hatched area in Figure 11.10. In the case always made explicit in the literature. The addi- of Germany, the re-distributive effects are rein- tional net costs arise from the fact that electricity forced by the exemptions granted to energy-in- production from most renewable sources is still tensive enterprises, which raise the burden on more expensive than from conventional sources. the remaining consumer groups. These groups These costs can be measured as the difference include households and most small and medium between the levelized cost of electricity generated enterprises, which are not beneficiaries of the from renewable sources and the levelized cost exemptions. The number of German households from non-renewable sources. If a feed-in tariff is to and enterprises that have been cut off their elec- induce investments in renewable energy, it needs tricity supply for payment default has been rising to cover these additional costs plus a reasonable from about 312,000 in 2012 to about 350,000 in 2014, markup as compensation for the added risks of indicating rising energy poverty (Bundesnet- such investments. The risk markup, however, does zagentur and Bundeskartellamt 2015). There is a not add to net macroeconomic costs. It is rather a trade-off between the protecting the competitive- re-distribution of funds from electricity consum- ness of energy-intensive industries on the one ers to producers of renewable electricity. The hand and the competitiveness of non-exempted surcharge thus includes an additional component enterprises and distributional effects for house- that cannot be counted as net macroeconomic holds on the other hand.

5. LESSONS LEARNED AND REMAINING CHALLENGES The Energiewende–with its facets of renewa- industrial policy. The answer crucially depends ble energy and energy efficiency in electricity, on the analysed section of industry. Renewable transport and heating–is a comprehensive and energies provide the most accessible and compa- complex undertaking. Regarding the officialEner - rable databases, which is why this chapter mostly giewende targets, Germany has made consid- concentrates on them and takes other technolo- erable progress particularly in the deployment gies only as complementary, although somewhat of renewable energies for electricity generation, anecdotal, evidence. while other areas, such as energy efficiency, are lagging behind expectations. In renewable energy technologies, Germany has been partly successful in building innovative It is not possible, however, to clearly categorise advantages. Although competition by emerging the Energiewende as successful or unsuccessful economies is increasing, it was able to build on Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

its strong technological capacities to create an When interpreting the above assessment, two innovative edge. In wind energy it has succeeded caveats are in order. First, such economic effects 181 in gaining considerable global market shares, but as reduced health costs from improved local air the picture for solar photovoltaics is less inspir- quality, avoided costs from climate change miti- ing. A stronger focus on sophisticated steps in the gation and other factors do not enter into the global value chain may have proven effective in picture. Second, the above assessment can only avoiding bankruptcies and job losses in the photo- provide a snapshot and neglects dynamic and voltaic cell industry. While Germany persistently secondary effects, such as technological spill- ranks high in the Global Competitiveness Index of overs that could be expected from a smart grid, the World Economic Forum, indicating no nega- innovative materials or energy storage technol- tive overall impact of the Energiewende, compa- ogies. Positive economic effects of renewable nies in sectors other than clean technologies energies are thus likely to be underestimated, take a critical stance towards the policy’s reper- particularly in the longer term. Furthermore, cussions on their competitiveness. They attribute the effects of the German Energiewende as an electricity price increases to the Energiewende inspiration for other countries’ energy policies, and express concerns about the security of elec- and subsequent global technological progress tricity supply. in renewable energies, are considerable. The Chinese success in building a highly competi- In terms of value added, renewable energies are tive solar photovoltaic industry has led to stark becoming an economic factor in municipalities declines in photovoltaic technology costs and as well as in the form of state and federal taxes. thus to a virtuous cycle of scale economies and While renewable energies create considerable technology diffusion. The phenomenon is partly employment, these positive effects are at least based on German pioneering of photovoltaic tech- partly countered by job losses in conventional nology. However, the detailed analysis of interna- energies. An estimate of net employment effects tional competitiveness effects is beyond the scope is challenging, since available data often does not of this chapter. allow clear attribution: first, drawing clear boundaries to clearly define relevant sectors is a chal- Despite these caveats, the German case provides lenge in itself, and second, available data is not a wealth of experience and lessons. Success yet sufficiently fine-grained. factors, barriers, and subsequent policy amendments will be discussed below, including impli- Alongside job effects, the distribution of additional cations for–and lessons from–developing and costs and the impacts on energy poverty are emerging countries. central to assessing the inclusiveness dimension of the Energiewende. The fairness of cost distri- Three fundamental principles established in bution in Germany has indeed become an issue Erneuerbare-Energien-Gesetz, the German renew- of vigorous public debate. The distributive effects able energy law, can be considered the basis for are reinforced by exemptions from the electricity the establishment of renewable energy industries price markups granted to energy-intensive enter- and for the rapid growth of renewable energy prises. These exemptions raise the burden on the capacity in Germany, while safeguarding compet- remaining consumer groups, including house- itiveness of industries other than renewables. holds and non-privileged small and medium These principles are: enterprises. The resulting price increases contribute to energy poverty: the number of electricity ◼◼ Energy source-specific feed-in tariffs covering payment defaulters has been rising recently. a 20-year time horizon ◼◼ Grid connection, transmission and distribution While effects on electricity access are not rele- priority for electricity supplied from decentral- vant in the German case–electricity access is at ized renewable energy sources 100 per cent–they can be a particularly important ◼◼ Burden sharing of additional costs by all elec- aspect of inclusiveness and a major policy aim in tricity consumers through an electricity price developing countries, particularly in Africa. Aver- surcharge, with exceptions in particular for age African electrification rates are at 43 per cent, energy-intensive manufacturing enterprises. but considerably lower in rural areas, where they average 26 per cent (IEA 2016). In Sub-Saharan Until the renewable energy amendment in 2014, rural Africa they average only 17 per cent. Renew- the feed-in tariffs for renewable energy sources able energy off-grid solutions can make valuable were guaranteed and fixed, constituting a and cost-efficient contributions to electrification, purchasing guarantee for unlimited volumes. This particularly in remote rural areas. provided unprecedented investment security for project developers and, combined with rapid cost Green Industrial Policy - Concept, Policies, Country Experiences

digression especially for solar photovoltaics, led renewable energy projects, in particular when it 182 to strong growth of renewable energy industries needs to be implemented by an actual competi- and national installed capacity. However, while tor of the renewable energy project developers. only some industries could firmly establish their This is the case in many developing countries competitive edge, capacity growth increased the where electricity supply and grid operation is in cost burden on consumers, which led to criticism the hands of state-owned and often monopolistic of the scheme. The inflexible feed-in tariff rates companies. These companies face a conflict of and windfall profits of the Energiewende’s early interest when obliged to undertake grid invest- years are a reason why the total policy costs are ments to connect their competitors. Morris and not a suitable orientation for similar policies in Martin (2015) describe the South African state- other countries: on the one hand, technology costs owned enterprise Eskom to show behaviour that have considerably decreased, and on the other can be interpreted as ‘malicious compliance’–not hand, policies themselves have become more actively opposing the government’s renewable efficient. The renewable energy amendment of energy policies, but using its own control over the 2014 reacted to the increasing cost of the Energie- electricity value chain to obstruct independent wende by establishing technology-specific, yearly power producer engagement. Public opposition, caps for additional capacity (Deutsche Bundesr- too, can constitute a barrier to grid expansion. egierung 2014). In addition, the government took While fossil fuel power plants can be built in the decision to change the determination method close proximity to centres of electricity demand, for feed-in tariffs from pre-determination to renewable energy capacity needs to be located tendering. Following a pilot tender for large solar at favourable generation sites. This necessi- photovoltaic installations in 2015, this method tates transmission lines to bring the electricity will take effect from 2017 for 80 per cent of new to centres of demand. Between 2013 and 2015, wind and photovoltaic installations (BMWi 2015c). 72 to 78 per cent of respondents to the Energie- To avoid a competitive disadvantage for small- wende-Barometer surveys saw grid expansion scale installations, those smaller than 1 MW are as the main priority of German energy policy exempted from the tendering requirement. In in the context of the Energiewende (DIHK 2013, turning to competitive elements in the support 2014, 2015). The construction of transmission of renewable energy, the German government lines, however, can be in conflict with local resi- follows the example of many emerging coun- dents’ interests. Building the lines below ground tries, such as China, India and South Africa. These is one option to mitigate such conflicts, but it is countries have successfully experimented with considerably costlier than building them above tendering schemes and have gathered a wealth ground. In their agreement of 01 July 2015 on the of experience in the technical design of efficient Energiewende implementation, the chairpersons and effective feed-in tariffs (Pegels 2014). Lessons of Germany’s ruling parties nonetheless stipu- learned from these countries include the inte- lated that underground cables will be favoured in gration of a systematic policy learning process the planning of new high voltage direct current through successive amendments of the scheme transmission lines, thus hoping to increase the in the course of various bidding windows; the chances of public acceptance (BMWi 2015b). integration of safeguards against adventurous bidding, such as penalty payments in the case of Where competitiveness concerns forestall assign- project defaults; and the integration of measures ment of the additional costs of such measures to secure local economic and social benefits, such to eligible industries, these costs will increase as domestic content requirements and social the electricity price for non-eligible consumers, development requirements for the eligibility of thereby increasing the risk of energy poverty. projects for support. These effects can constitute a barrier to public acceptance of feed-in tariff schemes, especially Guaranteed grid connection and transmission in developing countries. Where social security and distribution priority for German renewable systems fail to cushion those effects, additional energy provide further investment security, since compensation for people living in poverty needs project developers can be sure that the generated to be implemented. electricity can be fed into the grid and sold at the feed-in tariff rate. Grid connection can otherwise constitute a major barrier for the feasibility of Chapter 11 Germany: The Energy Transition as a Green Industrial Development Agenda

REFERENCES 183 Aretz, A., Heinbach, K., Hirschl, B., & Schröder, A. Deutsche Industrie- und Handelskammer (DIHK). (2013). Wertschöpfungs- und Beschäftigungsef- (2015). Anpassung statt Aufbruch. IHK-Energie- fekte durch den Ausbau Erneuerbarer Energien: wende-Barometer: Vol. 2015. Berlin. Studie im Auftrag von Greenpeace Deutschland, Ecofys & Fraunhofer ISI (2015). Stromkosten der Hamburg. Berlin. energieintensiven Industrie. Ein internationaler Bundesministerium für Umwelt, Naturschutz, Bau Vergleich. Berlin. und Reaktorsicherheit (BMUB). (2014). GreenTech Ekins, P., Bradshaw, M., & Watson, J. (2015). Global made in Germany 4.0. Umwelttechnologie-Atlas Energy: Issues, Potentials, and Policy Implica- für Deutschland. Berlin. tions. Oxford: Oxford University Press. Bundesministerium für Wirtschaft und Energie Esposito, M., Haider, A., Semmler, W., & Samaan, D. (BMWi). (2015a). Die Energie der Zukunft: Ein (2017). Enhancing job creation through green gutes Stück Arbeit ; vierter Monitoring-Bericht transformation. In Altenburg, T., & Assmann, C. zur Energiewende (Stand: November 2015). Berlin. (Eds.). (2017). Green Industrial Policy. Concept, Bundesministerium für Wirtschaft und Energie Policies, Country Experiences (pp. 50–67). (BMWi). (2015b). Eckpunkte für eine erfolgreiche Geneva, Bonn: UN Environment; German Umsetzung der Energiewende. Politische Verein- Development Institute / Deutsches Institut für barungen der Parteivorsitzenden von CDU, CSU Entwicklungspolitk (DIE). und SPD vom 1. Juli 2015. Germany Trade & Invest. (2013). Industry overview. Bundesministerium für Wirtschaft und Energie The photovoltaic market in Germany. Berlin. (BMWi). (2015c). EEG-Novelle 2016 - Eckpunkte- International Energy Agency (IEA). (2015). World papier. Berlin. Energy Outlook 2015: Energy access database. Bundesministerium für Wirtschaft und Ener- Jungjohann, A. (2016). The Energiewende - A Success gie (BMWi). (2015d). The Energy of the Future. Story at a Crossroads. In Bertelsmann Foundation Fourth “Energy Transition” Monitoring Report – (Ed.), Newpolitik (pp. 51–60). Washington D.C. Summary: BMWi. Berlin. Kemfert, C., Opitz, P., Traber, T., & Handrich, L. Bundesministerium für Umwelt, Naturschutz, Bau (2015). Deep Decarbonisation in Germany: und Reaktorsicherheit (BMUB). (2017a). Entwick- A Macro-Analysis of Economic and Politi- lung von monatlichen Energiepreisen zu nomi- cal Challenges of the ‘Energiewende’ (Energy nalen Preisen Deutschland: Energiedaten Tabelle Transition). DIW Berlin: Politikberatung 26a. Berlin. kompakt: Vol. 93. Berlin: Deutsches Institut für Bundesministerium für Wirtschaft und Energie Wirtschaftsforschung. (BMWi). (2017b). Stromerzeugungskapazitäten, Lehr, U., Ulrich, P., Lutz, C., Thobe, I., Edler, D., O’Sulli- Bruttostromerzeugung und Bruttostromver- van, M., Simon, S., Naegler, T., Pfenning, U., Peter, brauch Deutschland: Energiedaten Tabelle 22. F., Sakowski, F., & Bickel, P. (2015). Beschäftigung Berlin. durch erneuerbare Energien in Deutschland: Bundesnetzagentur and Bundeskartellamt. (2015). Ausbau und Betrieb, heute und morgen. Monitoringbericht 2015: Bundesnetzagentur für Lema, R., Nordensvärd, J., Urban, F., & Lütkenhorst, W. Elektrizität, Gas, Telekommunikation, Post und (2014). Innovation paths in wind power: Insights Eisenbahnen; Bundeskartellamt. Monitoringberi- from Denmark and Germany. Discussion paper / cht gemäß § 63 Abs. 3 i. V. m. § 35 EnWG und § 48 Deutsches Institut für Entwicklungspolitik: Vol. Abs. 3 i. V. m. § 53 Abs. 3 GWB, Stand: 10. Novem- 2014,17. Bonn: DIE. ber 2015. Bonn. Lütkenhorst, W., & Pegels, A. (2014). Stable poli- Deutsche Bundesregierung. (2014). Gesetz zur cies--turbulent markets: Germany’s green grundlegenden Reform des Erneuerbare-En- industrial policy : the costs and benefits of ergien-Gesetzes und zur Änderung weiterer promoting solar PV and wind energy. Research Bestimmungen des Energiewirtschaftsrechts report. Winnipeg, Manitoba, Geneva, Switzerland, vom 21 Juli 2014. Bonn: Bundesanzeiger. Beaconsfield, Quebec: International Institute for Deutsche Industrie- und Handelskammer (DIHK). Sustainable Development; Global Subsidies Initi- (2012). IHK-Energiewende-Barometer 2012. Noch ative; Canadian Electronic Library. überwiegt die Skepsis. Berlin. Morris, M., & Martin, L. (2015). Political Economy of Deutsche Industrie- und Handelskammer (DIHK). Climate-relevant Change Policies: The Case of (2013). Unternehmen packen’s an – Skepsis bleibt. Renewable Energy in South Africa. Final Report IHK-Energiewende-Barometer: Vol. 2013. Berlin. for The Political Economy Analysis of Climate Deutsche Industrie- und Handelskammer (DIHK). Change Policies (PEACH) Project, IDS Evidence (2014). Mehr Verlierer – weniger Gewinner: Report. Brighton. Fakten, Trends, Forderungen (IHK-Energie- Nestle, U., & Morris, C. (2016). Das EEG: Besser wende-Barometer No. 2014). Berlin. als sein Ruf. WISO direkt: 2016/14. Bonn: Green Industrial Policy - Concept, Policies, Country Experiences

Friedrich-Ebert-Stiftung. Statistisches Bundesamt. (2015). Umsatz mit 184 O’Sullivan, M., Lehr, U., & Edler, D. (2015). Brutto- Umweltschutzgütern und Umweltschutzleistun- beschäftigung durch erneuerbare Energien in gen 2013. Wiesbaden: Statistisches Bundesamt. Deutschland und verringerte fossile Brenn- Sworder, C., Salge, L., Van Soest, H. (2017). The Global stoffimporte durch erneuerbare Energien und Cleantech Innovation Index 2017. World Wildlife Energieeffizienz. Makroökonomische Wirkungen Foundation, Cleantech Group. und Verteilungsfragen der Energiewende, Zulief- UK Intellectual Property Office. (2014). Eight Great erung für den Monitoringbericht 2015. Technologies. Energy Storage: A patent overview. OECD. (2015). OECD Statistics. Paris: OECD. Newport. Parad, M., Henningsson, S., Currás, T. A., & Youngman, Umweltbundesamt. (2014). Wirtschaftsfak- R. (2014). Global Cleantech Innovation Index 2014. tor Umweltschutz. Produktion – Außenhan- Pegels, A. (2014). Green Industrial Policy in Emerg- del – Forschung – Patente: Die Leistungen der ing Countries. Routledge studies in ecological Umweltschutzwirtschaft in Deutschland. Umwelt, economics: Vol. 34. London: Routledge. Innovation, Beschäftigung 01/2014. Dessau- Roßlau: Umweltbundesamt. Prognos. (2015). Wertschöpfungs-und Beschäfti- gungseffekte der Energiewirtschaft: Studie im Umweltbundesamt. (2017). Nationale Trendta- Auftrag des Bundesministeriums für Wirtschaft bellen für die deutsche Berichterstattung und Energie, Projektnummer 49/13. Basel. atmosphärischer Emissionen 1990–2015. Dessau- Roßlau: Umweltbundesamt. Schmitz, H. (2016). Who Drives Climate-relevant Poli- cies in the Rising Powers? (IDS Evidence Report United Nations Industrial Development Organization No. 180). (UNIDO). (2016). International Yearbook of Indus- trial Statistics 2016 (2016 ed.). International year- Der Spiegel (2016, July 17). Spanische Gamesa. book of industrial statistics. Vienna: UNIDO. Siemens schmiedet weltgrößten Windradbauer. Der Spiegel. Retrieved from www.spiegel.de/ World Bank. (2015). World integrated Trade Solutions wirtschaft/unternehmen/siemens-und-game- (WITS) database. Washington, D.C. sa-schmieden-weltgroessten-windradbau- World Economic Forum. (2015). The Global Competi- er-a-1098337.html tiveness Report 2015–2016. Geneva. Statistik der Kohlewirtschaft e.V. (2016a). Braunkohle. Statistik der Kohlewirtschaft e.V. (2016b). Steinkohle. Chapter 12 Electric Mobility and the Quest for Automobile Industry Upgrading in China CHAPTER 12 185 ELECTRIC MOBILITY AND THE QUEST FOR AUTOMOBILE INDUSTRY UPGRADING IN CHINA

Tilman Altenburg, Kaidong FENG, Qunhong SHEN Green Industrial Policy - Concept, Policies, Country Experiences

186 1. INTRODUCTION China promotes an ambitious electric mobility of China’s ability to achieve its environmental programme, with two main objectives: to reduce objectives, while at the same time reaping early urban air pollution and to enhance the compet- mover advantages in an emerging global indus- itiveness of its national automotive industry. A try. Section 2 informs the reader about the current range of industrial policies is being applied, and technological shift from internal combustion several pathways to electric mobility are being engines to electric automobile technology and explored in parallel. On the one hand, China’s how this may affect the competitive positioning industrial policy promotes the development of of countries and companies globally. Section 3 battery-electric and plug-in hybrid passenger cars, discusses China’s objectives, highlighting two: buses and trucks using a wide range of subsidies enhancing national competitiveness in the auto- and regulations. In addition to encouragement of mobile industry and reducing urban air pollution. public and private Chinese carmakers, the govern- Section 4 provides an overview of China’s most ment puts pressure on international investors to important electric vehicle policies, revealing an co-develop modern electric cars in joint ventures enormous political commitment for this transi- with their Chinese partners. On the other hand, in tion. Section 5 takes stock of where China stands the shadow of national industrial policy and with in terms of technological achievements and very little government support, low-cost electric emerging competitive advantages, differentiating vehicles experience an unprecedented boom. the four main technological areas of high-speed These include the 200 to 230 million electric cars and buses, low-speed cars, two-wheelers and two-wheelers now circulating on China’s roads battery manufacturing. Section 6 then discusses and the recent addition of simple low-speed elec- to what extent these developments actually tric cars. Although the latter are not permitted on address the environmental problems related to highways, their popularity in rural areas is enor- the automobile industry. Section 7 summarizes mous, with 600,000 produced in 2015 in Shandong the technological and environmental achieve- Province alone. ments of China’s move towards electric mobility.

This paper explores the direction of the electric vehicle industry’s evolution and the extent

2. FROM INTERNAL COMBUSTION ENGINES TO ELECTRIC DRIVING: A TECHNOLOGICAL PARADIGM SHIFT Globally, road transport is at the beginning of a economic importance of the respective industries, technological paradigm shift. According to inter- as well as their huge environmental impact. national scenarios, transport in urban areas has to be largely decarbonised by 2050 to keep global Cars and two-wheelers can be fully or partly elec- warming within manageable limits (Teng et al. trified. In a fully electrified battery-electric vehicle, 2015). The old technology of internal combus- the internal combustion engine is replaced with tion engine automobiles is incompatible with the an electric engine powered by a huge, in most imperative of decarbonising the world economy. cases lithium-ion, battery. There are, however, New carbon-efficient transport technologies are different intermediate stages of hybrid vehicles required and electrification is the most prominent that combine electric and fuel-engine driving. alternative option, provided that the electric- Mild hybrids are propelled mainly by internal ity used by cars comes from low-carbon energy combustion engines–they use batteries as a sources. Furthermore, in contrast to conventional complementary temporary power source. Batter- traffic, electric vehicles produce almost no local ies recharge with the electricity generated by the emissions and therefore reduce air pollution in combustion engine and the energy recuperated the cities. from braking. Such vehicles can drive on electric only in the short term and their range is very Electric vehicles include a range of different trans- limited. Plug-in hybrids are designed for elec- port technologies, including electric ships, trains, tric propulsion and have only a small auxiliary buses, cars, and two-wheelers. The focus here is combustion engine. They use larger batteries that on cars and two-wheelers, given the enormous can be plugged into the electricity grid. Chapter 12 Electric Mobility and the Quest for Automobile Industry Upgrading in China

Electric passenger cars, defined as fully have set similar targets. Physical limits to fuel battery-electric vehicles and plug-in hybrids, efficiencies in combustion engines mean they 187 still account for only a small share of the global cannot be fully decarbonised (Ferguson and car market: in 2016, 0.2 per cent of all passenger Kirkpatrick 2015). Since the required reductions cars (OECD and IEA 2017). In the market segment cannot be achieved by improving combustion of two-wheelers, the world market share of elec- engines, carmakers must substantially increase tric motors is much higher, about 25 per cent, a the share of hybrid and electric vehicles in their proportion almost exclusively due to the popular- fleets. In China, political pressure for electrifi- ity of electric bikes and scooters in China (OECD cation is similarly high, although for a different and IEA 2016). Overall, electric vehicle deployment reason. Chinese cities suffer unbearable air pollu- has been held back by technological problems, tion and, given the combination of strong public mainly the disappointing performance and high discontent with air pollution and a massive price of batteries: Most electric vehicles can travel increase of car ownership, reducing car-based only 100–200 miles without recharging and fully pollution is a political top priority. recharging a battery takes 4–8 hours. Progress on these fronts however is quickening. Battery The automotive industry is one of the most performance is improving, particularly in terms important manufacturing industries globally. In of the possible driving range, and their cost has 2015 it produced 68.5 million passenger cars and decreased rapidly in recent years, from US$ 1,000 90.7 million buses and trucks, with the corre- per kWh in 2010 to US$ 300–400 per kWh in 2014 sponding employment benefits (OICA 2016). In (Nykvist and Nilsson 2015). These authors forecast the EU, for example, 2.3 million people were that the price will go down to US$ 200 per kWh directly employed in car manufacturing in 2014, by 2020 and calculate that electric vehicles will increasing to 12 million if related services are become cost competitive with conventional cars included (ACEA 2016). A similar order of magni- at a price of US$ 150 per kWh. In the same vein, a tude applies for the United States and China. recent study by Bloomberg New Energy Finance The car industry creates important secondary (2016) states that battery price reductions “will effects in industries such as steel, machine tools, bring the total cost of ownership of electric vehi- electronics and chemicals, explaining why all cles below that for conventional-fuel vehicles major developing countries try to create domes- by 2025, even with low oil prices.” At that stage, tic automotive industries as a backbone of their uptake is likely to accelerate very fast. The same national industrial development strategies. At source predicts that sales of electric vehicles will the same time, the minimum scale requirements reach 41 million by 2040 and include 35 per cent and technological sophistication of the automo- of light duty vehicle sales. This is nearly 90 times tive industry have continuously increased over the equivalent 2015 figure, when electric vehicle the past decades. Since South Korea joined the sales may have reached 462,000, about 60 per cent exclusive club of car-producing nations in the more than 2014 (Bloomberg 2016). In fact, global 1970s, not a single newcomer has been able to car manufacturers have already begun a race for catch up with the technological frontier. China, new and higher-performance battery-electric and India and other emerging economies are quite plug-in hybrid models. successful in producing cars and two-wheelers for their domestic markets, but none of them has At the same time, political pressure to elec- been able to seriously challenge the technological trify road transport is increasing, as countries leaders, so far. committed to ambitious decarbonisation agendas in the 2015 Paris Agreement. Even before that, The paradigm shift to electric driving will likely many OECD countries had defined roadmaps cause a major industry shake-up. When a new for the gradual reduction of GHG emissions that set of technologies and the necessary supporting force carmakers to successively reduce their institutions emerge, much of the accumulated average fleet emissions. The European Commis- knowledge and physical assets accumulated sion requires carmakers to ensure average emis- by incumbents loses its value, making it much sions of their fleets do not exceed a maximum easier for newcomers to compete (Perez 1988). of 130 g CO /km for newly registered cars in 2015, Battery-electric vehicles require different kinds 2 of auto parts: electric engines with integrated and this threshold will be lowered to 95 g CO2/km by 2021. Although further reductions have not yet powertrains, magnets, powerful traction batteries, been specified, the regulations ask “the Commis- different power electronics, new software, invert- sion to maintain a clear emissions-reduction ers, charging devices, new lightweight materials trajectory comparable to that achieved up to 2020” such as carbon fibre-reinforced polymers–and (EC 2017). United States and Japanese authorities totally different thermo management systems Green Industrial Policy - Concept, Policies, Country Experiences

because there is no combustion engine that can leapfrogging refers to newcomers adopting a new 188 be used for heating and cooling. Conversely, technology more quickly and thereby overtaking combustion engines and their parts, such as the formerly leading firms (Fudenberg et al. 1983). pistons, crankshafts and alternators as well as Tesla, a complete newcomer to the automotive exhaust systems and fuel tanks, will no longer be industry based in California, started production required. As a result, supplier industries are likely of an all-electric sports car in 2005 and within a to go through a major restructuring and see the few years leapfrogged into the segment of lead- emergence of newcomers to the auto industry ing luxury carmakers. Between June 2012 and (McKinsey & Company 2011). Similarly, the archi- September 2017, it sold almost 200,000 Model S tecture of electric cars can be radically different: luxury vehicles and the company aims to deploy The combustion engines are no longer needed, at least 1 million electric cars by 2020 (OECD and but heavy batteries need to be placed in the IEA 2017). auto bodies in ways that allow for good driving performance. Electric vehicles can have centrally Other, albeit not yet equally successful, newcom- placed electric engines, but there may also be ers have sprung up in other countries. In China, two motors attached to the front and rear axles BYD and Zotye were the first movers in produc- respectively, or four small motors placed in the ing electric cars in China–each arriving from a wheels (e-mobil BW 2010). Chassis can be made of different industry background. BYD was a battery carbon fibre rather than steel. All these changes manufacturer that ventured into car manufactur- challenge the existing carmaker’s competitive ing by 2003 and Zotye was an auto-parts dealer advantages, which are currently rooted in their that began producing cars in 2005. Both started capabilities to master internal combustion engine with traditional internal combustion engines but and transmission technology, to design cars and took up electric vehicle production earlier than the to manage multi-tiered production networks. established carmakers. At a national scale, China’s aim is to take advantage in a similar way. With a In fact, worldwide, many firms from non-au- strong effort to become a leading manufacturer of tomotive backgrounds are now venturing into electric vehicles, as well as a leading market for electric vehicle manufacturing, trying to take them, the Chinese government hopes to become advantage of the techno-organizational change more successful in the global car industry than it to leapfrog into the car industry. The concept of was in the era of combustion engines.

3. CHINA’S OBJECTIVES: ENHANCED COMPETITIVENESS AND CLEANER URBAN AIR China is one of the most ambitious promoters of Since private car ownership was allowed, sales electric mobility and, clearly, the most important and production soared. China has by far the larg- one outside the OECD. China promotes electric est market in terms of passenger car sales world- vehicles for various reasons, but two stand out. wide (Euler Hermes 2014). In 2015, China produced 21 million passenger vehicles (OICA 2016). The first reason is that the country regards the However, technological progress has been slow technological shift from internal combustion and foreign brands still capture almost 60 per engines to electric propulsion as an opportu- cent of the Chinese market (EU SME Centre 2015). nity to catch up with the global leaders in auto- Fuel engines and transmissions are among the motive technology and production. Although technological fields where China clearly is behind China’s automobile production goes back to the international competitors; but these technologies early years of the People’s Republic of China, the are not needed in battery-electric vehicles. At the industry took off only in the 1990s after being same time, electric engines are less complex and declared officially strategic for China’s economic less expensive. Finally, China has the second larg- development. Car production was seen as a est reserves of lithium in the world and is a lead- cornerstone of technological development and ing global manufacturer of lithium-ion batteries, national policies aimed at indigenous innovation so there seems to be a solid basis for seizing this (Chu 2011). China’s provinces tried to create state- opportunity. For these reasons, policymakers are owned automobile companies with local supplier optimistic that China’s industry can take advan- networks (Thun 2006). tage of the technological change and, leapfrogging Chapter 12 Electric Mobility and the Quest for Automobile Industry Upgrading in China

into the era of electric mobility, become a leading compared to 0.9 per cent in the US, 1.4 per cent player in the global automotive industry (Wang in the EU and -0.5 in Japan (Gao et al. 2014). At 189 and Kimble 2011). such growth rates, urban air pollution clearly will become unmanageable without a radical change The second reason for shifting to electric vehi- in vehicle technologies. cles is that they are emission-free locally and therefore promise a solution for urban air pollu- Therefore, notably, climate change mitigation is tion–one of China’s most pressing problems. not the main rationale behind China’s electric Concentrations of particulate matter in big mobility programmes. In fact, China’s grid-pow-

Chinese cities are far beyond what the World ered electric cars emit even more CO2 than aver- Health Organization considers safe, and respira- age petrol cars when life cycle emissions are tory health problems are among the main causes calculated. This is because electric vehicle emis- of death. Zheng et al. (2015) estimate that every sion totals include those from the fossil fuels year between 2001 and 2012 in Beijing’s central burned to produce that electricity. Also, emissions area, over 5000 people died prematurely due to from the manufacture of electric cars are higher increased PM2.5 concentrations.27 Estimates than those of conventional petrol cars. Taking on the costs of health attribute US$ 1.4 trillion these factors into account, electric cars would losses to outdoor air pollution in China in 2010 have a better performance in terms of CO2 emis- (OECD 2014). It is difficult to isolate how much of sions, except in heavily coal-based economies this cost is due to road traffic, but for Beijing it is such as China. In 2016, coal-fired power plants still estimated that in 2013 motor vehicles accounted accounted for more than 65 per cent of China’s for 21 per cent of PM2.5 and estimates for Shang- electricity demand (China Energy Portal 2017). As hai 2014 attribute 22 per cent to motor vehicles, a result, based on 2009 data, grid-powered electric ships and airplanes (Beijing Municipal Environ- cars in China emitted 258 g CO2e/km, equivalent mental Protection Bureau 2014; Shanghai Munic- to a fairly inefficient petrol car with a 9 l/100 km ipal Environmental Protection Bureau 2015). fuel consumption (Wilson 2013). This share is likely to increase with the globally unprecedented boom in combustion engine car However, a new study suggests that China’s coal sales in China. By 2015, the registered passenger consumption peaked in 2014, much sooner than cars in civil use surpassed 95 million, making scenarios had predicted. Intervening years are China by far the world’s biggest automobile verifying the trend and while coal will remain the market (National Bureau of Statistics of China primary source of energy in China for the next 2016). Moreover, between 2004 and 2014, car sales decades, the declining trend may accelerate (Qi et increased by 11.4 per cent annually in China al. 2016).

4. CHINA’S ELECTRIC VEHICLE POLICIES China’s electric vehicle policies are driven mainly (State Council 2012). To encourage electric vehicle by the desire to leapfrog into a new promising development and deployment, a wide spectrum of field of technological specialisation and the need policies were enacted by the central government to reduce urban air pollution. To achieve these (Table 12.1). In addition, provinces offered their goals, China’s central government set ambitious own policy packages, which in part added to the targets for electric vehicle deployment: 500,000 central government’s incentives, but also pushed cars to be sold by 2015 and 5 million by 2020 in different directions.

27 Particulate matter consists of microscopic parts suspended in the atmosphere. PM2.5 refers to particulate matter with a mean aerodynamic diameter of 2.5 μm. These are highly carcinogenic as they penetrate deep into lungs and blood streams. Green Industrial Policy - Concept, Policies, Country Experiences

Table 12.1: Main polices for the promotion of electric vehicles in China 190 Phase 1 (1990–98): ◼◼ First research and development projects for electric vehicles in the 8th and Research and development 9th National Key Technology Research and Development Programmes and small-scale demonstration projects Phase 2 (1999–2008): ◼◼ Deployment of Cleaner Vehicles programme: city trials in 12 major cities Scale up systemic research ◼◼ 10th National Key Technology Programme for electric vehicles: scale up and development projects research and development for electric vehicles by about factor ten with small-scale local demonstration projects Phase 3 (2009–2012): ◼◼ New Energy Vehicles declared strategic emerging industry Large-scale demonstration ◼◼ Ten Cities, Thousand Vehicles Demonstration and Deployment Programme: projects and first major testing and subsidies for public fleets in 25 cities incentive programmes ◼◼ Use of conventional motorcycles banned in major cities: totally in 13 and partially in 16 cities ◼◼ Energy saving and New Energy Vehicles industry development plan 2012– 2020: preferential treatment of electric vehicles ◼◼ Subsidy policy for purchasing private electric vehicles in 6 pilot cities ◼◼ Public procurement regulations and subsidies ◼◼ 400,00 charging stations and 2000 batteries swapping station target Phase 4 (2013–2015): ◼◼ Guidelines for New Energy Vehicles Deployment with refined incentives: New round of incentives, purchase tax waiver, trade barrier removal, price regulation of electricity systematic policy for charging, infrastructure subsidy etc. charging ◼◼ Pilot region expanded to 88 cities ◼◼ National Plan for Charging Infrastructure Phase 5 (2016–ongoing): ◼◼ Phase out of subsidies by 2020 announced Reduced subsidies, stricter ◼◼ Electric vehicle quota for carmakers announced performance standards ◼◼ Stricter fuel economy requirements ◼◼ A general ban on the production and sale of fossil fuel cars under study for the near future ◼◼ Use of stricter technology standards and licensing policies to overcome industry fragmentation ◼◼ Enforcement of technological and environmental regulations for low speed electric four-wheelers

Sources: Adapted from Xu and Su (2016), Zhang et al. (2014).

As a key element of electric mobility support, free vehicle registration. For the city of Shang- the ‘Ten Cities, Thousand Vehicles’ programme hai, effective subsidies reached US$ 27,600 per launched demonstration projects in 13 pilot cities vehicle (Wan et al. 2015). Public procurement was in 2009 and added 12 more later. Public electric another key policy. The central and local govern- grid companies were told to build an infrastruc- ments purchased electric vehicles for government ture of charging stations for electric vehicles. In fleets and made their procurement - manda parallel, 29 major cities fully or partially banned tory for bus and taxi companies. However, local the use of petrol motorcycles, thereby boosting procurement policies come at a price. In many demand for electric two-wheelers (ADB 2009). cases, local governments focused subsidies onto Later on, especially with the launch of the ‘Energy locally manufactured car brands, thereby blocking saving and New Energy Vehicles industry devel- competition and creating a geographically frag- opment plan 2012–2020’ generous incentives were mented industry (Zhang et al. 2014). offered for the whole country, including purchase subsidies and tax exemptions for electric vehicles More recently, the government announced plans as well as large dedicated research programmes to phase out purchase subsidies by 2020 that with emphasis on lithium-ion battery research. motivated local governments and investors to Purchase subsidies are probably the most gener- adopt many electric vehicle projects but at the ous ones worldwide (Mock and Yang 2014; Alten- same time reduced automakers and battery burg et al. 2016). In addition to US$ 9,200 per car suppliers motivation to develop higher-quality from the central government, regional govern- products (Bloomberg News 2016). Instead, the ments offered additional cash subsidies and government is now augmenting pressure on Chapter 12 Electric Mobility and the Quest for Automobile Industry Upgrading in China

carmakers to accelerate electrification. Since from a certain date (The Guardian 2017). Moreo- June 2017, new draft legislation imposes an elec- ver, stricter technology standards and licensing 191 tric vehicle quota on carmakers. By 2018, 8 per requirements are imposed to consolidate the cent of all newly built cars should have an electric currently fragmented auto industry (Bloomberg engine, going up to 12 per cent in 2020. Companies New Energy Finance 2016). that fail to meet their quota would have to reduce their sales of conventional cars or purchase credit In parallel, pressure is increased on foreign points from other carmakers. Whether the quota carmakers to share technologies. As China is by can actually be reached, however, also depends far the world’s largest and most dynamic market on supply-side constraints, such as technological for automobiles, all major global carmakers are progress and availability of charging infrastruc- keen to increase their market presence in China. ture. Yet, such pressure clearly has a strong effect China pursues a barter-type policy allowing inter- on company strategies, and various carmakers national carmakers to build new or expand exist- already announced new investments in electric ing production facilities, but only if they bring in vehicle manufacturing for the Chinese market, their latest technology and co-develop cars with including Volvo and Ford (The Guardian 2017). local partners (Holmes et al. 2015). A development plan issued in 2010 clearly states that for any Furthermore, the government defined a roadmap joint venture manufacturing key components with gradually increasing fuel economy require- of electric vehicles–such as batteries, motors ments for conventional cars which by 2020 will be and controllers–the Chinese partner must hold among the most ambitious worldwide. The Minis- at least 51 per cent of the capital. Again, such try of Industry and Information Technology even requirements can only yield the expected results announced their work on a timetable to totally if local partners manage to enhance their absorp- ban the production and sale of fossil fuel cars tive capacity for the new technologies.

5. TECHNOLOGICAL ACHIEVEMENTS AND EMERGING COMPETITIVE ADVANTAGES Within a few years, China has become a leading while individual consumers are still hesitant global market for electric vehicles. This section to buy highway-capable personal vehicles provides an overview of achievements in four despite the enormous subsidies offered. So far, different segments of the electric vehicle indus- Chinese products range at the low end of tech- try in terms of market development, technological nological complexity and do not meet the quality upgrading and competitiveness.28 demanded on international markets (Bloomberg News 2016). To assess the potential for techno- 5.1. HIGHWAY-CAPABLE PASSENGER CARS logical upgrading, a differentiated view of China’s AND BUSES automotive companies is needed. State-owned enterprises, especially those Sales of highway-capable battery-electric and engaged in joint ventures with international plug-in cars and buses remained marginal until carmakers, have benefited from booming markets 2013, with less than 18,000 in total, leading some for traditional vehicles with internal combustion analysts to pessimistic forecasts and critical engines. Therefore, they have had little incentive assessments of China’s electric vehicle policy to enter the risky field of electric vehicles. When package (Wan et al. 2015). Since then, however, the large state-owned car companies finally deployment powerfully took off, surpassing entered electric vehicle production, not least to 330,000 vehicles in 2015 (Xu and Su 2016). At about benefit from the generous government subsidies, 1.4 per cent, this is still a minor fraction of all car they relied on retrofitting the internal combus- and bus sales, but the increase is impressive and tion car models rather than developing models indicates a real breakthrough for electric driving optimised for electric vehicle technology. Gener- in China. ally, their strength is in producing technologically Purchases by bus and taxi companies are a unsophisticated, affordable cars. For example, major force for electric vehicle deployment, one of China’s best-selling electric vehicles, the

28 Information on specific car models and their technological innovations in this section has been collected from specialised press reports and company websites. Green Industrial Policy - Concept, Policies, Country Experiences

Chery QQ3EV, used cheap but environmentally 5.2. AFFORDABLE LOW-SPEED 192 harmful lead-acid batteries until production ELECTRIC CARS stopped in 2016. But Chinese producers can adapt: Hundreds of Chinese companies have started to lithium cells power Chery’s follow-up model, the develop technologically simple small low-speed, EQ-S15EV. low-voltage, low-range electric cars. These cars Recently the government increased its pressure are built for a top speed between 40 and 70 km/h, on international joint venture partners to launch sufficient to drive in cities or rural roads, but these electric vehicles in the country and share the rele- low-speed electric vehicles are not allowed on vant knowledge with Chinese firms. In response highways. Their driving range without refilling to this, several international carmakers, includ- the battery is up to 70 kms, which is sufficient for ing Toyota and BMW, have recently developed most Chinese customers. battery-electric vehicles that fit Chinese market Low-speed electric vehicles are very cheap. conditions and are in some cases exclusively Depending on their sophistication, they are sold at produced for this market. How this will affect a price of US$ 2,000 to US$ 8,000. Hence the lowest technological learning and competitiveness of cost low-speed electric vehicles are affordable Chinese firms is not yet clear. for many households unable to purchase a new While state-owned enterprises long remained gasoline car, which start at around US$ 5,000 in hesitant to venture into electrification, some China. The fact that low-speed models are exempt technologically ambitious newcomer firms in from registration adds to their attraction; some car production became electric vehicle pioneers. big cities impose quotas on the registration of These include Zotye Auto, the auto-part producer automobiles to limit the number of cars circu- that started to produce battery-electric vehicles lating in their jurisdictions. Prices are so low for in 2010 and initiated an interesting battery-swap- two reasons. First, designs are very basic. These ping project with the State Grid Company in vehicles run on simple 1.5 to 4 kW direct current Hangzhou; and BYD, the battery producer for the motors, they use low-cost components, such as electronics industry. BYD first moved into car lead-acid batteries, and all non-essential features battery production and then produced the first have been removed. Low-speed electric vehicles plug-in hybrids in 2008 followed by a battery-elec- lack sophisticated battery management or motor tric vehicle in 2010. In 2014, BYD launched the control systems and in most cases do not meet battery-electric model Denza, a car developed basic safety standards. Second, manufacturers with Daimler in China and the first car Daimler combine very simple low cost techniques, such as ever developed outside Germany. traditional stamping dies, with hi-tech elements, including three-dimensional laser cutting robots It is not yet clear which electric vehicle devel- (Autoblog 2014). opments in China are most promising in terms of technological leapfrogging and interna- The future of China‘s low-speed electric vehicle tional competitiveness. Chinese carmakers industry is unclear. Unlike highway-capable elec- may successfully specialise in simple, no-frills tric vehicles, the low-speed models do not receive low-cost electric vehicles that meet the needs of any government support. In particular, customers emerging global middle classes. However, based are not entitled to receive any of the generous on a rapidly expanding internal market for elec- purchase subsidies offered for highway-capable tric vehicles, China could become the global plat- cars. Quite the opposite: national traffic regula- form where automotive multinationals develop tions discourage the emergence of this industry. and produce large portions of their electric port- Only some provincial governments, especially the folios, with minor or major contributions from one in Shandong, encouraged low-speed electric their Chinese joint venture partners. Also, electric vehicle production. bus manufacturing could become China’s field of Still, demand is high for simple and affordable competitive specialisation. The rollout of elec- cars for families in smaller cities and rural areas. tric buses has been impressive, with more than Tapping into this demand, dozens of relatively 300,000 buses circulating in China in 2015 (OECD small carmakers emerged particularly in Shan- and IEA 2017). This provides economies of scale dong province where 600,000 low-speed electric and potential early mover advantages. Chongqing vehicles were sold in 2015. According to OECD/IEA Hengtong Bus Power System Co. Ltd., for example, (2017) estimates, three to four million low-speed has developed the world’s first rapidly recharging electric vehicles were circulating in China in 2016. battery-electric public bus. Policymakers had not expected this popularity. In fact, Shandong’s industry had emerged in the Chapter 12 Electric Mobility and the Quest for Automobile Industry Upgrading in China

shadow of national law. According to national existed in 2011, of which the 50 largest account law, low-speed electric vehicles were illegal until for half of the production (Fu 2013). The bulk of 193 very recently, yet some provincial and municipal production is sold in China, but an estimated 5 regulations provided for exceptions on certain million electric bikes are exported, mainly to other roads. Only in 2016, the central government Asian countries (INSG 2014). Electric two-wheeler started to develop legislation to legalise and regu- production and maintenance thus became a late low-speed electric vehicles (Paglee 2017). So major and unique industry in China, even though far, many central and local government policies they are technologically quite unsophisticated. remain contradictory with regard to minimum With this scale of market success, electric bikes speed requirements and phasing out lead-acid are the first mass-produced and massively popu- batteries, among other issues. Once this legal lar alternative-fuel vehicles in the history of uncertainty is overcome the market is likely to motorised vehicles (Cherry 2010). receive an additional boost; by 2020, low-speed electric vehicles’ annual sales are projected to 5.4. BATTERY TECHNOLOGY DEVELOPMENT reach about 2 million (OECD and IEA 2016). China-specific technological solutions are also 5.3. ELECTRIC TWO-WHEELERS being developed in the field of lithium-ion battery technology. Chinese module production is close China’s market for electric two-wheelers is boom- to the international technological frontier, but ing. Two-wheelers include electric bicycles–bikes the country lags far behind in the fields of battery with a small electric motor that still retain the chemistry, membranes and battery management ability to be pedalled by the rider–as well as more systems. About 200 battery manufacturers are powerful scooters and motorcycles. Sales started operating in China, but most cannot compete with gradually during the 1990s and by 2004 40,000 global brands (Bloomberg News 2016). Technolog- were sold (Cherry 2010). At around US$ 230 to ical mastery of battery management systems in US$ 290 for simple versions and US$ 650 for high- particular is important to produce premium cars end versions, electric two-wheelers are affordable and to differentiate brands: each system needs to for many Chinese families (Fu 2013). To keep their be specifically tailored to the energy requirements cost low, Chinese e-bikes are generally low-tech of each car; moreover, they are necessary to opti- and powered by lead-acid batteries. Only a small mise battery performance and prevent failures. fraction of two-wheelers are equipped with lith- Specific research and development programmes ium-ion batteries, which increase the cost by have been established to build capacity, but reach- US$ 160 (Fu 2013). Also, electricity costs are low: ing the level of global leaders will take time. The cost of powering a two-wheeler is about 0.2¢ per kilometer. Battery replacement amounts to The lack of technological capabilities in battery about 1¢ per km, depending on the size of the technology has led to two China-specific develop- battery and fluctuating prices. These estimates ments: First, some Chinese companies, including compare to an average of 8¢ per kilometer for the Chinese state-owned automotive manufac- cars and 3¢ per kilometer for motorcycles. An turer BAIC and Zotye, have built remote battery average bus trip costs 3¢ per kilometer as well monitoring systems that allow them to monitor (Cherry 2010). the status of entire car fleets in real time through wireless networks. Monitoring centres can then Two policy decisions make electric bikes so popu- send messages to drivers assisting them to lar: The government classifies electric two-wheel- improve battery performance and avoid emer- ers–with pedals, a maximum speed of 20km/h gency situations. Compared to mature onboard and maximum weight of 40 kg–as bicycles, so systems, remote monitoring is clearly a second- riders do not need a driver’s license or registra- best option, but it can substitute some of the function. As well, many city governments restrict tions of advanced battery management systems. conventional motorcycle use in their inner cities (Cherry 2010). Second, Chinese firms are investing in battery swapping operations so cars can exchange their About 200 to 230 million electric bicycles were discharged battery with a charged one, rather circulating in China in 2015, according to various than recharging their own battery. This, in prin- sources, with 37 million produced annually (OECD ciple, saves time and provides flexibility to the and IEA 2016). This makes China by far the most user. The practical problem is that only identi- important global market with a share of about cal batteries can be swapped. However, carmak- 85 per cent (INSG 2014). About 2,600 licensed ers design battery management systems to whole-vehicle manufacturers and assemblers meet the specific requirements of each model, Green Industrial Policy - Concept, Policies, Country Experiences

and swapping stations cannot hold stocks of all Corporation owning the batteries. The experiment 194 battery models. In Hangzhou, electric vehicle targets mainly taxis of the Zotye brand and bus manufacturers, battery manufacturers and the fleets, which keeps the number of battery types State Grid Corporation of China set up a citywide manageable. Extending the experiment to other battery swapping experiment with the State Grid car brands will be quite difficult, however.

6. TO WHAT EXTENT DOES THE SHIFT TO ELECTRIC VEHICLES SOLVE ENVIRONMENTAL PROBLEMS? The shift to electric vehicles has one big posi- vehicles use lead-acid batteries, and even in tive effect on the environment: Electric vehicles the segment of highway-capable cars, Chinese themselves produce almost no local pollutants, so firms offer cheap lead-acid versions. Electric they help to reduce the enormously harmful smog two-wheelers contribute greatly to releasing lead levels in Chinese cities. into the environment, given that they require a new battery containing 10–20 kg of lead every In terms of greenhouse gas emissions, in contrast, 12–18 months (Cherry 2010). In Beijing alone, China’s efforts to promote electric mobility do not 30,000 to 50,000 tons of lead-acid batteries need make any positive contribution, yet. As discussed, to be recycled every year (Fu 2013). However, with China’s current energy mix, use of an aver- Chinese battery recycling is not well regulated age electric car emits as much CO as a fairly 2 and as much as a quarter or more of the used inefficient petrol car. This, however, will change lead is not captured in the process (Cherry 2010). over time. On one hand, petrol cars are becom- After frequent cases of lead poisoning in China, in ing more fuel-efficient. On the other hand, China 2011 the government decided to close 80 per cent is undertaking considerable efforts to decar- of all the registered lead-acid battery production, bonise its power sector: The government set the assembly and recycling companies (Fu 2013). target for peak CO2 emissions around 2030 and is committed to steadily reduce them afterwards Lithium-ion batteries are less harmful, but at and carbon intensity will decrease by 60 to 65 per this stage in the industry’s growth they are too cent below 2005 levels by 2030 (He 2014). China’s expensive for general use. As the technology coal consumption peaked in 2014, much sooner matures, however, costs are decreasing quickly than scenarios had predicted, and will decline by (Nykvist and Nilsson 2015). Government regula- degrees, even as coal remains the primary source tions will then have an important role in ensuring of energy for the next decades (Qi et al. 2016). that manufacturers adopt lithium-technology as Thanks to the energy-saving targets established soon as possible and that lead-acid batteries are th in the 2005–2010 11 Five-Year Plan, CO2 emis- phased out. While lithium-ion batteries also pres- sions declined by 1.46 billion tons during that ent a number of environmental challenges, there period (Teng et al. 2015). Between 2011 and 2015, is hardly any alternative to using this technology the share of low-carbon sources grew from 19 to at a much larger scale (USEPA 2013). Progress on 28 per cent of the national energy mix (OECD and lithium-ion batteries is critical to increase the IEA 2016). Whether electrification of China’s trans- driving range of electric vehicles and decrease port is good for the global climate will depend on their cost, which in turn makes electric vehicles the relationship between fuel efficiency improve- attractive to more customers. ments and the rate of electricity mix decarbonisation. Physical limits to fuel efficiencies in Last but not least, rebound effects matter: As elec- combustion engines mean they cannot be fully tric vehicles become cheaper and better, and the decarbonised (Ferguson and Kirkpatrick 2015). population grows and becomes wealthier, more Therefore, over the long run, transport electrifica- consumers can afford more advanced technol- tion is the only alternative for China. ogies. People who used to pedal their traditional bicycles are shifting to electric bikes and scoot- The manufacture, use and disposal of batteries ers. People who used to drive electric two-wheel- produce another major environmental problem. ers start buying low-speed electric cars, and those China has made electric driving affordable. But who drove such cars in the past may now be able this has been achieved by using the lead-acid to purchase a comfortable highway-capable car. batteries that are very harmful for the environ- Cherry (2010) cites studies conducted in Shang- ment, especially when they are not properly hai, Kunming and Shijiazhuang cities showing recycled. Nearly all two-wheelers and low-speed that most electric bike users used to travel by bus Chapter 12 Electric Mobility and the Quest for Automobile Industry Upgrading in China

or bicycle, and only a few substituted them for in higher total emissions. Low-carbon strategies cars. Furthermore, as people become wealthier, need to account for these potential repercussions. 195 they tend to travel longer distances. Even if the Still, complementary reforms are necessary, such technologies in use become more resource-effi- as the promotion of public transport systems, as cient and the share of renewables in the energy well as urban designs that minimise the need mix increases, the sheer expansion of modern to commute between housing areas, business vehicle ownership and mileage is likely to result districts and shopping centres.

7. CONCLUSIONS

China’s ambitious programme to electrify road ◼◼ In electric bus production, China has become transport is an exemplary case of green indus- one of the first volume manufacturers globally trial policy pursuing technological upgrading and should pursue this opportunity, especially and greater competitiveness in parallel with to meet the demand from other countries for environmental improvements. China’s compre- reducing urban air pollution. hensive policy package includes huge research ◼◼ China produces 85 per cent of electric and development efforts, technology-sharing two-wheelers globally and exports about 5 agreements with global investors, strategic public million units annually to other Asian markets, procurement, purchase subsidies and city trials. although it only covers the low-end of markets, The result is that electric-powered two-wheelers, so far. passenger cars, trucks and buses are becoming a ◼◼ Low-speed electric vehicles have emerged as major alternative to fuels-based driving. an interesting niche market in which there is hardly any international competition because The Chinese government set hopes on electric Chinese demand conditions are different from vehicles as a technology that would allow the those in high-income countries. Offering such country to catch up with the global leaders in simple affordable products may be a viable the automobile industry–something it had not business model for exports to other developing achieved under the old paradigm of traditional countries, especially as their decarbonisation combustion engine cars. Multinational carmak- efforts progress. ers maintained their significant advantage in ◼◼ China’s industry has a long tradition at the combustion engine vehicle production through level of lithium battery cell manufacturing and accumulated specialist knowledge in all the a strategic position due to its enormous lith- domains needed to build good engines and trans- ium reserves. missions and to integrate them in the car design. However, manufacturing electric vehicles requires So there are various promising opportunities for a different range of skills and, in principle, this industrial development related to electric mobil- will allow advantages for China. ity. In all the technological fields mentioned above China has become a major player. The challenge China’s leapfrogging aspirations are high, but the lies in technological upgrading. country has made significant progress in various domains: Across the board, Chinese producers cover the low-end product range, so far; whereas Japanese, ◼◼ In the mainstream market of highway-capa- Korean, European and American competitors ble electric vehicles many Chinese companies cater to the more complex high-value vehicle are now offering battery-electric and plug-in market. But there is cause for optimism about hybrid models. By international comparison, the prospects for upgrading due to at least three these are all still low-end, but they are also reasons: First, the Chinese government has low-cost and suitable for the rapidly expand- recognised the challenge and redesigned its poli- ing domestic market. cies with a stronger emphasis on research and ◼◼ Some automobile multinationals are now development, stricter technology standards and producing new battery-electric models exclu- consolidation of the fragmented auto and battery sively for the Chinese market. Given the size of industries. Second, China is the largest market the Chinese markets and government pressure for most of the related supply chain products, to locate research and development operations giving the country an advantage in terms of scale in China, this may promote enhanced techno- of production as well as potential foreign direct logical capabilities. investment. Third, exploiting the export potential Green Industrial Policy - Concept, Policies, Country Experiences

of simple but reliable low-priced goods to coun- automotive sector and simultaneously pursue the 196 tries with similar demand conditions is a prom- decarbonisation of the electricity supply. Another ising option. problem lies in the increased demand for environmentally harmful but inexpensive batteries. What about the hoped-for environmental Due to the preference for low-cost electric vehi- improvements? After a few years of disappoint- cles, the particularly harmful lead-acid batteries ingly slow uptake, a tipping point has now been dominate the market. Gradual transitions to lith- reached where the share of electric vehicles in ium-ion battery installation, as well as appropri- the Chinese fleet of two-wheelers, cars, buses ate recycling policies, are needed to minimisrbe and trucks is gaining traction. With each tradi- environmental harm. tional fuel vehicle being replaced by an electric one, local emissions go down and, with enor- Transition to sustainability requires complex mous ramifications for public health, China’s systemic shifts rather than just technologi- most pressing environmental problem, urban cal substitutes. Policymakers need to ensure air pollution, is reduced. In terms of greenhouse that the electricity mix gets decarbonised; that gas emissions, the shift to electric vehicles will low-carbon public transport systems become not improve the state of affairs in the short term. more attractive; that cities are designed so Instead, emissions will increase as electric vehi- people do not have long commutes between the cles run on power generated by coal-fired plants. places where they live, work and shop; and that But China’s power sector is slowly shifting away materials are reused or recycled to the greatest from coal. As this trend continues and accel- extent possible. What inspires hope is the speed erates, the carbon footprint of China’s electric at which some countries, including China, are vehicles is bound to improve. With a substantial greening their economies and the realisation share of renewables in the power mix, electric that emerging economies can reap the double cars are clearly preferable to conventional cars. dividend of environmental improvement and An integrated programme for greening the auto- enhanced competitiveness. motive industry must go beyond changes in the Chapter 12 Electric Mobility and the Quest for Automobile Industry Upgrading in China

REFERENCES 197 ACEA (2016): Facts about the Automobile Ferguson, C. R., & Kirkpatrick, A. T. (2015). Internal Industry. Retrieved from www.acea.be/ combustion engines: applied thermosciences. automobile-industry/facts-about-the-industry John Wiley & Sons. Altenburg, T., Schamp, E. W., & Chaudhary, A. (2016). Fu, A. (2013). The Role of Electric Two-Wheelers in The emergence of electromobility: Comparing Sustainable Urban Transport in China: Market technological pathways in France, Germany, analysis, trends, issues, policy options. Retrieved China and India. Science and Public Policy, 43(4), from https://sustainabledevelopment.un.org/ 464–475. content/documents/3792fu2.pdf Asian Development Bank (ADB). (2009). Electric Fudenberg, D., Gilbert, R., Stiglitz, J., & Tirole, J. (1983). Bikes in the People’s Republic of China: Impact Preemption, leapfrogging and competition in on the Environment and Prospects for Growth patent races. European Economic Review, 22(1), (ADB economics working paper series). Manila, 3–31. Philippines. Gao, P., Hensley, R., & Zielke, A. (2014). A road map to Autoblog. (2014). A window into China’s low-speed the future for the auto industry (McKinsey Quar- electric vehicle revolution. Retrieved from www. terly No. October 2014). autoblog.com/2014/07/25/a-window-into-chinas- He, J.-K. (2014). An analysis of China’s CO2 emis- low-speed-electric-vehicle-revolution/ sion peaking target and pathways. Advances in Beijing Municipal Environmental Protection Bureau. Climate Change Research, 5(4), 155–161. (2014). The analysis of sources of PM2.5 in Beijing. Holmes, T. J., McGrattan, E. R., & Prescott, E. C. (2015). Retrieved from http://search.bjepb.gov.cn/bjepb/ Quid Pro Quo: Technology Capital Transfers for 413526/331443/331937/333896/396191/index.html Market Access in China. The Review of Economic Bloomberg New Energy Finance. (2016). Electric Studies, 82(3), 1154–1193. vehicles to be 35% of global new car sales by 2040. International Nickel Study Group (INSG). (2014). The Retrieved from http://about.bnef.com/press-re- Global E-bike Market (Briefing Paper No. 23). leases/electric-vehicles-to-be-35-of-global-new- Lisbon. car-sales-by-2040/ McKinsey & Company. (2011). Boost! Transforming Bloomberg News (2016, July 11). China Has Too Many the powertrain value chain: a portfolio challenge. Mediocre Electric Carmakers, Researcher Says. Retrieved from: http://actions-incitatives.ifsttar. Retrieved from www.bloomberg.com/news/ fr/fileadmin/uploads/recherches/geri/PFI_VE/ articles/2016-07-11/china-has-too-many-medio- pdf/McKinsey_boost.pdf cre-electric-carmakers-researcher-says Mock, P., & Yang, Z. (2014). Driving electrification. A Cherry, C. (2010). Electric Two-Wheelers in China: global comparison of fiscal incentive policy for Promise, Progress and Potential. Retrieved from electric vehicles. (White Paper). Washington, D.C. www.accessmagazine.org/articles/fall-2010/elec- National Bureau of Statistics of China. (2016). Statis- tric-two-wheelers-china-promise-progress-po- tical Communique of the People’s Republic of tential/ China on the 2015 National Economic and Social China Energy Portal. (2017). 2016 Detailed Electricity Development. Retrieved from www.stats.gov.cn/ Statistics. Retrieved from https://chinaenergypor- english/PressRelease/201602/t20160229_1324019. tal.org/en/2016-detailed-electricity-statistics/ html Chu, W.-W. (2011). How the Chinese government Nykvist, B., & Nilsson, M. (2015). Rapidly falling costs promoted a global automobile industry: VW and of battery packs for electric vehicles. Nature Toyota vying for pole position. Industrial and Climate Change, 5(4), 329–332. Corporate Change, 20(5), 1235–1276. OECD. (2014). The cost of air pollution: Health e-mobilBW. (2010). Strukturstudie BWe mobil − impacts of road transport. Paris: OECD. Baden-Württemberg auf dem Weg in die Elektro- OECD & International Energy Agency (IEA). (2016). mobilität. Stuttgart. Global EV Outlook 2016: Beyond one million elec- EU SME Centre. (2015). The Automotive Market in tric cars. Paris, Paris: OECD Publishing; IEA. China. Brussels. Retrieved from www.cham- OECD & International Energy Agency (IEA). (2017). berelancs.co.uk/wp-content/uploads/2015/10/ Global EV Outlook 2017: Two million and counting. EU-SME-Centre-Sector-Report-The-Automotive- Paris: OECD/IEA. Market-in-China.pdf Organisation Internationale des Constructeurs d’Au- Euler Hermes. (2014). Economic Outlook No.1210: tomobiles (OICA). (2016). Production and Sales Special Report August-September 2014. Statistics. Retrieved from www.oica.net/category/ European Commission (EC). (2017). Reducing CO2 production-statistics emissions from passenger cars. Retrieved from Paglee, C. (2017): China’s huge hidden electric vehicle https://ec.europa.eu/clima/policies/transport/ market. China Automotive Review, (12)4, 22–23. vehicles/cars_en Green Industrial Policy - Concept, Policies, Country Experiences

Perez, C. (1988). New technologies and develop- United States Environmental Protection Agency 198 ment. In C. Freeman & B.-ê. Lundvall (Eds.), Small (USEPA). (2013). Application of Life-Cycle Assess- countries facing the technological revolution ment to Nanoscale Technology: Lithium-ion (pp. 85–97). London, New York: Pinter Publishers. Batteries for Electric Vehicles. Washington, D.C. Qi, Y., Stern, N., Wu, T., Lu, J., & Green, F. (2016). China’s Wan, Z., Sperling, D., & Wang, Y. (2015). China’s electric post-coal growth. Nature Geoscience, 9(8), car frustrations. Transportation Research Part D: 564–566. Transport and Environment, 34, 116–121. Shanghai Municipal Environmental Protection Wang, H., & Kimble, C. (2011). Leapfrogging to elec- Bureau. (2015). 70% of PM2.5 pollutants come from tric vehicles: Patterns and scenarios for China’s a local source. Retrieved from www.shanghai. automobile industry. International Journal of gov.cn/nw2/nw2314/nw2315/nw17239/nw17252/ Automotive Technology and Management, 11(4), u21aw968232.html 312–325. State Council. (2012). Energy Saving and New Energy Wilson, L. (2013). Shades of Green: Electric Cars’ Auto Industry Development Plan (2012–2020). Carbon Emissions Around the Globe. Retrieved from www.gov.cn/zwgk/2012-07/09/ Xu, L., & Su, J. (2016). From government to market and content_2179032.htm from producer to consumer: Transition of policy Teng, F., Gu, A., Yang, X., & Wang, X. (2015). Pathways mix towards clean mobility in China. Energy to deep decarbonization in China. Sustainable Policy, 96, 328–340. Development Solutions Network (SDSN) and Zhang, X., Rao, R., Xie, J., & Liang, Y. (2014). The Institute for Sustainable Development and Inter- Current Dilemma and Future Path of China’s Elec- national Relations (IDDRI): Paris, France. tric Vehicles: The current dilemma and future The Guardian (2017, September 11). China to ban path of China’s electric vehicles. Sustainability, production of petrol and diesel cars ‘in the near 6(3), 1567–1593. Sustainability, 6(3), 1567–1593. future’. The Guardian. Retrieved from www. Zheng, S., Pozzer, A., Cao, C. X., & Lelieveld, J. (2015). theguardian.com/world/2017/sep/11/china-to- Long-term (2001–2012) concentrations of fine ban-production-of-petrol-and-diesel-cars-in-the- particulate matter (PM2.5) and the impact on near-future human health in Beijing, China. Atmospheric Thun, E. (2006). Changing lanes in China foreign Chemistry and Physics, 15(10), 5715–5725. direct investment, local government, and auto sector development. Cambridge University Press. Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident? CHAPTER 13 199 ETHANOL POLICY IN BRAZIL: A GREEN INDUSTRIAL POLICY BY ACCIDENT?

Pedro da Motta Veiga, Sandra Polónia Rios Green Industrial Policy - Concept, Policies, Country Experiences

200 1. INTRODUCTION Brazil pioneered policies to stimulate the status in environmental and climate change consumption of renewable energy sources, negotiations (MME 2016). providing incentives for the production of hydro- electricity and ethanol. Policies that foster the Can these policies be considered ‘green’? On one industrial transformation of sugar cane into etha- hand, they actually stimulated the supply and nol gained attention during a period of strong demand for renewable energy. On the other hand, economic growth in Brazil under the 1964 to 1985 analysis of these policies’ implementation reveals authoritarian political regime. a limited perception on the part of the government in any green dimension. This is particu- The adoption of these policies was not due to larly notable in the case of ethanol policy, which environmental concerns, which were not wide- evolved in accordance with criteria and priorities spread at the time. In fact, their implementation only marginally related to green concerns. created considerable social and environmental problems, especially in the case of constructing This paper describes and analyses the case of large hydroelectric plants. Instead, the motiva- Brazilian policies for the development of etha- tions for these policies in Brazil were directly nol as a fuel competing with and replacing an oil related to limiting the rise of the oil imports derivative, gasoline. Brazil now has more than 40 after the oil price shocks of the 1970s and to years of experience with sectorial policies in this mitigating the macro-economic impacts of any area, which involve policies aimed at the differ- further shocks. In other words, the rationale for ent stages of the ethanol’s production chain. The these policies related to energy security and to antecedents and characteristics of the first phase reduce the economic repercussions of growing of the ethanol policy adopted by Brazil in the 1970s oil imports. Nevertheless, due to these policies, are described in section 2. The second phase of Brazil now runs one of the world’s most success- the ethanol policy is the topic of section 3, which ful programmes for using renewable energies in examines changes to the ethanol policies in the its transport sector (Farina et al. 2013). Moreover, last decade and consequences for the sector’s these policies have produced a high proportion of performance. Section 4 presents some policy renewables in Brazil’s energy matrix, more than lessons that can be drawn from the Brazilian expe- 40 per cent in 2015, giving the country significant rience with ethanol policies. Section 5 concludes.

2. THE FIRST PHASE OF THE ETHANOL POLICY: 1975 TO 1990

2.1. BACKGROUND According to Moreira (2008), IAA’s state intervention took over the production chain, including the grow- The production of sugarcane is one of the orig- ing, commercialisation, price setting, and assigning inal large-scale commercial activities in Brazil, quotas for production, exporting and associated preceded only by extraction of the brazilwood that imports with the explicit aim of balancing produc- produces a red dye. Since the Portuguese coloniza- tion and consumption in the sector. The IAA also tion in 1500, Brazil has been one of the world’s prin- sought to manage tensions and conflicts involv- cipal producers and exporters of sugar. However, ing different regions and actors linked to distinct th since the 17 century Brazilian sugar met increased stages in the production chain. Decisions about competition from nearby regions, principally the productive capacity and its expansion, or about Antilles, and later from sugar beet produced in the destination of products, for example, were the th Europe, particularly at the end of the 19 century responsibility of the IAA. (Moreira 2008). The first governmental measures aimed at encour- As a response, there were unsuccessful attempts aging the production and consumption of ethanol to modernise the production of sugar in Brazil. added to gasoline dated from the 1930s, establish- Frequent crises of overproduction and the insta- ing a proportion reaching 5 per cent of ethanol to be bility of prices led the federal government to inter- mixed with gasoline. The rationale for this measure vene in the market at the beginning of the 1930s, was principally to reduce gasoline imports to Brazil. creating the Institute of Sugar and Alcohol, in Portuguese, Instituto do Açúcar e do Álcool (IAA). Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

Between the 1930s and 1960s, the sugarcane sector sectorial pressures to alleviate problems of excess diversified geographically to the South East region production of sugar and the instability of the 201 around São Paulo, the most economically devel- prices of this product also contributed to acceler- oped State in Brazil, and expanded productively ate the government response to the new scenario. with the growth of ethanol output. Until the end For Meyer et al. (2012), the Proálcool programme of the 1960s, the principal economic function of “was intended to both reduce Brazil’s dependence ethanol production, for the government and for on imported oil and make use of idle capacity in producers, was to avoid oversupplying sugar and the sugar industry by converting sugar to ethanol.” depressing its prices. The aims of Proálcool were linked to the logic of import substitution, which commanded Brazil’s 2.2. PROÁLCOOL: DEVELOPING A NEW industrialisation strategy, not only because it INDUSTRIAL SECTOR aimed to replace oil imports. The strategy also intended to develop internal suppliers of capital The strong rise of international oil prices at the goods for the sugar and ethanol industry. As noted beginning of the 1970s led to a radical change in by Moreira (2008), although the programme was the role of ethanol in the national economy. Brazil not restricted to producing ethanol from sugar- had been going through a period of accelerated cane, this did become the principal raw material economic growth driven by an expansion of the for ethanol production. Compared to alterna- production and consumption of industrialised tive raw materials, sugarcane presented better goods. Among the industrial sectors that emerged, marketing conditions, high availability, and short- automobile production stood out, as did its signif- term expansion potential. icant and growing dependency on oil imports. With the sudden increase in oil prices, between The instruments used by the programme were 1973 and 1974 the value of fuel imports jumped diverse and supported by extensive government from US$ 600 million to more than US$ 2 billion subsidies (Box 13.1). The government made subsi- (Moreira 2008). dised lines of credit available for investment in both crops and industrial application, maintaining Given the new scenario and the anticipation that a balance between the prices of sugar and alco- oil prices would remain high, the federal govern- hol to eliminate a deciding factor when producers ment created the National Alcohol Programme, chose either sugar or ethanol, as well as guaran- Proálcool, in November 1975. Concerns with the teed purchases (Moreira 2008). At the same time, rapid deterioration of the trade balance, and the a consortium of both public and private research macro-economic repercussions, were at the institutions worked to improve sugarcane varie- root of the decision to create Proálcool. However, ties (Meyer et al. 2012).

Box 13.1: Proálcool era: mid 1970s to mid 1980s Context: During this period, the State had the monopoly of oil exploration and production and the control of prices of all types of fuel.

Main policy instruments:

On the demand side:

◼◼ Reduction of taxes on the purchase and sale of alcohol powered automobiles ◼◼ Establishment of a limit price of ethanol at 65% of the price of gasoline ◼◼ Increase of the compulsory mixing to 22% of anhydrous ethanol in gasoline; ◼◼ Tax incentives (at the federal and state level) for the purchase of ethanol powered automobiles; ◼◼ Credit subsidies for the purchases of ethanol powered automobiles; ◼◼ Differentiated rates for the highways maintenance tax (TRU); ◼◼ Authorization for gas stations to open during weekends if providing ethanol fuel services exclusively. Green Industrial Policy - Concept, Policies, Country Experiences

202 On the supply side:

◼◼ Subsidised lines of credit for investment in crops and industries; ◼◼ Annual crop plans set by Governmental agencies, setting quotas for ethanol and sugar in each production unit in the country; ◼◼ Definition of a price parity between sugar-cane, ethanol and sugar; ◼◼ Purchase guarantees. Results: Ethanol was firmly introduced in the Brazilian energy matrix.

These policies caused a significant growth of managed by the Brazilian Development Bank investments in the construction or modernisa- (BNDES) fostered the production of machines and tion of more than 200 sugar mills, as well as an equipment used in the harvesting of sugarcane as increase in the production of low moisture, or well as in the industrial phase of ethanol produc- anhydrous, ethanol and, during its initial years, tion, with the aim of substituting for imported the programme was aimed at encouraging the capital goods. addition of anhydrous alcohol to gasoline to replace imported tetraethyl lead. From 1975 Backed by BNDES’ long-term loans at low or nega- to 1979, ethanol production grew from 580,000 tive real interest rates, Brazilian-owned compa- to 3.676 million m³, surpassing its target by 15 nies traditionally operating in the capital goods per cent (BNDES and CGEE 2008). However, this sector diversified their portfolios, heavily invest- increase was not sufficient to compensate for the ing in ethanol-producing equipment. At the same rise in the import costs of oil derivatives, espe- time, new companies were established targeting cially following a new international price shock specifically the production of capital goods used in 1979. in ethanol production.

In response to the new shock, the government According to Pereira (2015), until the early 1990s, expanded Proálcool, now tasked with the large- “the sugarcane agro-industry was one of the sectors scale production of ethanol for use as the exclu- in which state intervention was most intense. sive fuel for automobiles in the country. A set Official credit resources covered more than 80 per of measures was adopted to encourage the cent of the fixed investment in distilleries and the consumption of ethanol and, at the same time, production of sugarcane.” During the initial years guarantee a stable base for its supply. of the programme, the expansion of sugarcane production principally occurred through the exten- The main consumption incentives were the sion of planted areas, including in regions previ- reduction of taxes on the purchase and sale of ously occupied by other economic activities, such ethanol-powered automobiles, the establishment as coffee cultivation or livestock raising. of a price limit on ethanol at 59 per cent of the price of gasoline and the compulsory blending The existence of Proálcool was contested despite of 22 per cent of ethanol in gasoline. At the same the fact that Brazil was living under an authori- time, the government pushed and subsidised auto tarian political regime for much of that period. companies to develop engines capable of running Defenders of the programme pointed to its on ethanol alone (Meyer et al. 2012). In addition, success in reducing dependence on oil imports, the state-owned oil company Petrobras built a its principal objective. Critics pointed to the high distribution and pump infrastructure for pure, as cost of the programme, especially the amount opposed to blended, ethanol (Meyer et al. 2012). of subsidies and its environmental and social impacts that involved both the expansion of the In response to these measures, the sale of auto- agricultural land area used for sugarcane and the mobiles fully powered by ethanol grew signifi- pollution created by industrial ethanol production. cantly in the 1980s, representing around 80 per cent of the total automobiles sold between 1983 Despite its success in triggering the growth of and 1988. On the production side, the government ethanol production, Proálcool was weakened by encouraged the expansion of autonomous distill- a set of events that began in 1985. On one side, eries, designed for exclusive ethanol produc- crude oil prices began to fall, while domestic tion, as a form of assuring the sustained supply production of oil was growing. At the same time, of ethanol. Moreover, the financing programmes sugar prices increased, raising ethanol production’s opportunity costs (Moreira 2008). On the Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

other side, the new civilian government, elected By the end of the 1980s the ethanol sector had expe- in 1985, revisited incentive and subsidy policies rienced a supply crisis that affected the domes- 203 to ethanol resulting in the reduction of average tic market through shortages, even forcing Brazil sugarcane agro-industry returns and stimulat- to import ethanol to fuel all its ethanol-only cars, ing even more switching of the raw sugarcane and undermined the confidence of producers and to sugar exports (BNDES and CGEE 2008). Facing consumers in Proálcool (Box 13.2). growing inflation rates, the government reduced ethanol price guarantees below the cost of As a result of this first Proálcool crisis, the sale of production. In 1986, ethanol output fell for the ethanol-powered cars fell sharply in just two years: first time since the programme began (Meyer et in 1988, the share of these vehicles in total automo- al. 2012). bile sales in Brazil was 88 per cent. Two years later, this share had fallen to 12 per cent. Between 1995 and 2001, the market share of ethanol-powered cars 2.3. FROM GOVERNMENT INTERVENTIONISM in new sales was close to zero. Domestic ethanol TO DEREGULATION AND LIBERALIZATION supply stagnated and imports grew to meet the demand created by the first life cycle of the ethanol From the second half of the 1980s, the worsening of policy (Meyer et al. 2012). the macro-economic situation, and the exhaustion of the government’s fiscal capacity, led to important changes in governmental ethanol policy and to reduced intervention in the sector.

Box 13.2: Deregulation and liberalization in the 1990s Context: During this period, there was a fall in international oil prices, an increase in the domestic production of oil and a rise in the international prices of sugar. Brazilian Government implemented an encompassing set of modernizing structural economic reforms.

Main policy instruments:

◼◼ Dismantling of IAA; ◼◼ Phase-out of tax and credit incentives both to production and to the purchase of ethanol powered cars; ◼◼ End of price administration; ◼◼ End of the State monopoly in the production and exploration of oil. Results: Dramatic fall in the market-share of ethanol-powered automobiles combined with an ethanol supply crisis. Market incentives to increase productivity and decrease production costs across the entire production chain.

This worsening of the macro-economic situation kept gasoline prices above ethanol prices at a fixed after the second half of the 1980s, expressed in margin. In the early phase of the ethanol policy, skyrocketing inflation rates, and the policy shift funds were transferred, through Petrobras with implemented in the early 1990s, towards market the monopoly on ethanol and gasoline distribution, liberalisation and deregulation, strongly affected from gasoline consumers to ethanol producers ethanol production. At the institutional level, many through pricing. This intrasectorial subsidy ended that had been established to regulate markets and with price deregulation, though the government to control the supply of exports and substitution of continued to support ethanol by taxing it favoura- imports were dismantled. In 1990, the IAA itself– bly relative to gasoline. symbol of government stewardship of the sugar and ethanol sectors–ceased to operate. During From 1985 until 2003, annual ethanol produc- that decade, various measures further liberalised tion stayed within the 10 to 15 billion litres range. the market and prices in the sugar sector (Meyer However, loss of subsidies motivated the industry et al. 2012). to consolidate and increase production efficiency during this period (Goldemberg 2009). Productiv- Deregulation and market liberalisation led to the ity doubled—from 3,000 to 6,000 litres per hectare— 1999 dismantling of the subsidy mechanism that while production costs declined significantly as Green Industrial Policy - Concept, Policies, Country Experiences

focus on efficiencies worked along the entire value opposed to those of the more traditional areas, 204 chain, moving from research on yield increases to mainly located in the poorest Northeast region adopting economies of scale through ever-larger of the country (Pereira 2015). Broadly, the entities distilleries. The search for pragmatic solutions led representing the São Paulo region preferred an to the utilisation of bagasse, the stalks and pulp end to government regulation and market liber- residue of the canes, to generate electricity (Meyer alisation measures. By contrast, the Northeast et al. 2012). interests asked for price regulation and other measures of market control. Hence, despite the fact that the 1990s are identified as a period of crisis for sugar and ethanol, It was also in the 1990s that ethanol policy and the decade was also a period of intense institu- the environment agendas intersected for the first tional change affecting the sectors’ complete time when the environmental aspect began to gain shared and separate production chain. Besides relevance in the Brazilian public policy debate. To institutional shifts within the federal government a certain extent, this resulted from the 1992 Rio bodies associated with market deregulation and Conference that Brazil hosted and the commit- liberalisation, the sectors’ business organizations ments it made there. In 1993 the Federal Ministry and representation structures were profoundly of the Environment was created and the more transformed, with the emergence of new enti- developed States, such as São Paulo, at the time ties linked to the most modernised and produc- the largest ethanol producer, adopted environmen- tive areas of the sugarcane industry (BNDES and tal legislation that directly effected the sugar and CGEE 2008). These were based in the São Paulo ethanol sectors and that encouraged the mechani- region and defended interests that, at times, were sation of harvesting.29

3. THE SECOND PHASE OF THE ETHANOL POLICY: 2002 TO 2015

3.1. ETHANOL IN TIMES OF Bosch recognised the potential for engines that CLIMATE CONCERN could replace the ethanol-only version that was declining in popularity. This engine operates The sugar and ethanol sectors only began to with gasoline or ethanol, or with a mix of the two recover from the long crisis of the 1990s in 2002. fuels. It was well received by Brazilian consum- However, the recovery could not count on the ers, completely replacing engines operated only incentives and subsidies characterizing the with ethanol. Brazilian consumers appreciated a first policy cycle. In the second cycle, the stim- car that would not become obsolete as fuel prices, ulus came from demands of the international and government policies, fluctuated. This attrac- market and from the main industrial sector that tion was matched by the promise that the coun- consumed ethanol, automobiles, and any associ- try’s wide-ranging infrastructure would adapt and ated government incentives. that consumers would have increased security through the availability of gasoline, ethanol, or Two factors explained this recovery. One is any mixture of the two. It would also represent that various countries decided to replace their savings for car manufacturers, as they would no consumption of fossil fuels with the use of prod- longer have to supply duplicate models to meet ucts from renewable energy sources, in particular, demand for alcohol and gasoline powered vehi- biofuels such as ethanol. The other factor is a tech- cles. For producers, it would allow greater flexibil- nological innovation, the launch of flex-fuel auto- ity to offer their product to either the ethanol or mobiles that can be powered by gasoline or ethanol sugar end use according to which opportunity is or by combinations of the two (Moreira 2008). more advantageous at the time (Szwarc 2002). The technological game changer at the domestic level, directly affecting the most relevant demand for Brazilian ethanol, was the introduction and diffusion of the flex-fuel engine, developed by Bosch, a multinational present in Brazil. In 1994

29 The State of São Paulo prohibited the burning of the sugarcane before its harvest and set a timetable for the adoption by the producers of the mechanical harvesting. Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

In 2004, around half of the automobiles sold after oil. From 2007, only cars powered by gaso- in Brazil had flex-fuel engines, a share which line and flex-fuel cars were manufactured. By 205 reached 90 per cent in 2007 (Figure 13.1). That year, 2012, half the Brazilian fleet was already flex-fuel sugarcane became the second most important (Pereira 2015). source of energy in the Brazilian energy matrix,

Figure 13.1: Number of licensed vehicles per type of fuel, 1979–2015

3 Millions

Gasoline Ethanol 2 F l e x F u e l Tot al

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

2015

Source: Adapted from UNICADATA (n.d.-a).

The political game changer between the first and second policy cycles was the growing international perception of climate risks and the enormous responsibility of fossil fuel burning for the greenhouse gas emissions that cause global warming. Some consider the signing of the Kyoto Protocol in 1997 as the first effective and coordinated move to confront the problem at the global level. For the first time, developed countries assumed commitments to reduce emissions. Although these commitments were non-binding, they led several countries to adopt measures encouraging the use of biofuels in the transport sector. Ethanol, especially from sugarcane, was identified as one of the most promising options to replace fossil fuels (Box 13.3). Green Industrial Policy - Concept, Policies, Country Experiences

206 Box 13.3: The energy and environmental balance of sugarcane ethanol Sugarcane ethanol is recognised as a product with very positive environmental and energy attributes, even in comparison with other biofuels. The energy balance of ethanol registers an average energy output/input of around 8.3, against less than 2.0 for the main biofuel alternatives of corn, sugar beet, wheat straw and timber. All the energy necessary for the production of ethanol comes from the bagasse or residue of the cane. The direct consumption of fossil fuels in the production process is limited to transport trucks, agricultural machinery, and fertilisers. Moreover, since sugarcane ethanol production costs are lower than those of other biofuels, the net energy return on investments for ethanol from sugarcane is much higher than competing biofuels. Due to these characteristics, greenhouse gas emissions from the production and consumption of ethanol are much lower compared not only to gasoline but also to other biofuels. In 2004, the ethanol sector avoided emissions equivalent to 13 per cent of the total greenhouse gases emitted by the industrial and commercial sectors in Brazil. More recent data indicates that, combined with the required mixture of 25 per cent of biofuels in gasoline, the consumption of ethanol by flex-fuel vehicles allowed reductions in greenhouse gas emissions exceeding 40 million tons

of CO2 equivalent annually. Assad et al. (2012) show that in a scenario of production expansion, the GHG emission reductions obtained through the use of ethanol in 2020 would be sufficient to reach and even surpass Brazil’s targets stipulated in the National Climate Change Policy for the transport sector.

An assessment of the ethanol life cycle considered its environmental impacts on air, water, soil, the use of land including competition with food production, and biodiversity. Also included in the assessment were social aspects, such as impacts on employment, wages, income distribution and landholding structure, as well as working conditions. The balance of the assessment is very positive, although it does recognise problems and challenges in some areas, especially in relation to working conditions. The assessment of ethanol presented by Koizumi (2014) also registered the highly favourable energy and environmental balance of ethanol, as well as the benefits of its production on work and economic variables, such as oil imports.

Nevertheless, ethanol development policies in Brazil have been questioned by critics in relation to their direct and indirect impact on the use of land, and more specifically about the production of food and Amazonia deforestation. According to various evaluations, this questioning found scarce support in reality, since the direct impact of the expansion of the area occupied by sugarcane was on degraded pasture, while its indirect impacts only contributed marginally to deforestation. However, there are less optimistic assessments, especially in relation to the indirect impacts of the expansion of sugarcane: For example, De Sá et al. (2013) did find evidence of a positive relationship between cattle ranching in Amazonia and sugarcane expansion in the south of the country. The effect is that sugarcane production moving into degraded pasture pushes ranches into the Amazonia forest frontier, materialised dynamically over 10 to 15 years periods.

Source: Farina et al. 2013; Goldemberg et al. 2008; Koizumi 2014; Walter et al. 2008.

When the international context shifted more The introduction of a new tax, levied on the towards environmental concerns, Brazil had vast import and sale of fuels, helped diffuse this flex- industrial and agricultural installed capacity for fuel innovation. Contribuição de Intervenção no the production of ethanol and a wide infrastruc- Domínio Econômico–Combustíveis (CIDE–Fuel) ture for distribution and sale of the product all is a tax that falls on contractual obligations over the country. However, it was still recovering implying the transfer of technology: the payment from the crisis that marked the end of the first of royalties. Although the import and the sale of life cycle of its ethanol policy. As well, it is worth fuels do not fit into the activities under the scope noting that the introduction of the flex-fuel engine of the CIDE tax, the Brazilian government used as an option for consumers by automobile produc- this instrument to regulate the relative prices of ers in Brazil, all multinationals, was not directly fuels that compete with each other, especially linked to any governmental incentive programme. gasoline versus ethanol. According to the law Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

creating the tax, the funds raised were to be used fuels or to improve the transport infrastructure in environmental programmes to reduce the (Box 13.4). 207 effects of pollution caused by the use of these

Box 13.4: Flex-fuel automobiles and international commodity prices from 2003 to 2008 Context: During this period, the main drivers were: (i) in the domestic context, the launching of flex-fuel automobiles; (ii) in the international arena, the decision of many relevant economies to increase the share of renewables (and biofuels, in particular) in their energy matrix and the rise in the oil prices after 2004–2005.

Main policy instruments:

On the demand side:

◼◼ Tax incentive on the purchase of flex-fuel automobiles; ◼◼ Introduction of CIDE fuel in 2001—a typical regulatory tax on the sales and imports of fuels, with rates over fossil fuel representing, in the beginning, more than twenty times the level incurred on ethanol. On the supply side:

◼◼ Increase the public financing (BNDES) for long term investment in new plants.

The taxation level was fixed, in the case of gaso- concerns about climate change and defined a set line, at a value equivalent to more than twenty of economic incentives to accelerate the shift to times the value of the tax rate imposed on ethanol, renewable energy. so it had a huge impact on the relative price of the two products. Moreover, the reduction of the tax Various developed countries adopted policies on industrialised products or on the sale of auto- to increase the proportion of biofuels in their mobiles powered by ethanol or with a flex-fuel energy matrices, with the US being especially motor was extended. The increase in oil prices active in this process and overtaking Brazil as after 2004 was another incentive to spread flex- the largest global producer of ethanol in 2005. In fuel models and the use of ethanol as a fuel for the response to these evolutions in the domestic and Brazilian automobile fleet. foreign spheres, the ethanol sector entered a new investment cycle, characterised by the expan- After 2003, propelled by strong growth in demand, sion of sugarcane production through productive ethanol production went through a period of improvements and in areas already occupied by consolidation, marked by a concentration in large human activity, rather than exploiting frontiers or companies, and by the entrance into the sector of pristine landscapes (Nassar et al. 2008). Instead, business groups from the food and energy sectors production improvements resulted from replant- (Nastari 2012). An immediate consequence of this ing areas previously used for other crops, the process was the modernisation of the manage- establishment of new alcohol production mills, ment and the professionalisation of many and the development of sub-products such as companies, until then controlled by the families electric energy and alcohol-chemistry (Moreira traditionally established in sugarcane production. 2008). In the expectation of strong growth in ethanol production, some large companies from the sector The nominal value of disbursements by BNDES opened their capital and resorted to the capital multiplied fivefold between 2004 and 2007, in market to fund their investments. In the inter- support of sugarcane production and the etha- 30 national sphere, increasing oil prices renewed nol industry. The area planted with sugar- concerns about energy security. These added to cane increased 65 per cent between 2003 and 2009, while ethanol production rose 87 per cent

30 Between 2001-02 and 2007-08, Brazilian production of ethanol grew at an annual rate of 10.2 per cent, due to modern production units established in the South East regions of the country. Green Industrial Policy - Concept, Policies, Country Experiences

between the harvests of 2003-04 and 2008-09, an expanding large landholders and peasants, the 208 increase completely attributable to the production environmental pollution generated by the produc- of innovative hydrous ethanol, which expanded tion of sugarcane and ethanol, and the destruction by more than 210 per cent in this period (UNICA- of native vegetation (Duarte et al. 2013). DATA n.d.-b). However, the institutional reaction of the sector As external demand for the product grew, Brazil- was not limited to refuting this critique on the ian ethanol exports to developed countries, princi- basis of technical arguments. Indeed, the Brazil- pally the European Union, expanded considerably. ian Sugarcane Industry Association (UNICA) Between 2003 and 2008, exports increased by opened an office in Brussels and increasingly about 650 per cent. Scenarios on the issue, such became involved in international discussions on as an OECD and FAO (2007) projection to 2016 as the socio-environmental certification of products its time horizon, predicted a promising future for in the ethanol production chain (Chiodi et al. 2014). global ethanol consumption, as well as brilliant prospects for Brazilian ethanol exports. According 3.2. THE SECOND ETHANOL to those optimistic projections, Brazil would have POLICY CRISIS been responsible for 85 per cent of global ethanol exports by 2017. The international economic crisis that was triggered in 2008 reversed the positive trends in Dating from this period are the first attempts in Brazilian ethanol exports and investments in Brazil, in cooperation with the US, to define an the sector. Exports had a significant recovery agenda to transform ethanol into an international in 2011 and 2012. However, this export perfor- commodity, a precondition for the consolida- mance did not reflect the dynamism of the sector, tion of a global market for the product. The 2007 but instead showed the weakness of domestic Brazil–USA Memorandum of Understanding demand, largely due to changes in policies applied allowed for cooperation between the two coun- to fossil fuel prices. Since 2006, consumer prices tries to establish technical norms and standards of oil derivatives no longer varied periodically for ethanol, while at the same time stimulating and, with the exception of 2013, all price increases production and exports from other countries, incurring at the refinery phase were offset by the notably in Central America and the Caribbean reduction of federal taxes imposed on the product, (ICTSD 2007). to keep the gasoline pump prices stable and avoid Nevertheless, in this period of growth in produc- economic inflation. The compensatory mech- tion, domestic consumption and exports, Brazil- anism used to prevent increases in consumer ian ethanol was criticised, especially in European gasoline prices was the CIDE-Fuels tax (Farina et Union countries, by arguments based on criteria al. 2013), with values gradually reduced during the of social and environmental sustainability. Thus period and reaching zero in July 2012. During this biofuels, that had been seen as one of the solu- post 2008 financial crisis phase, the production of tions to mitigate the effects of global warming in sugarcane and ethanol also suffered from unfa- the 1990s, were blamed in the 2000s–in scientific vourable climatic conditions, as droughts affect- articles, organizations and forums, and by the ing three successive harvests. media–for causing deforestation, increasing food As a result of these policy shifts, between Janu- prices, and even threatening global food security ary 2006 and December 2012 the average price of (Moreira 2008). The Brazilian ethanol industry gasoline for Brazilian consumers increased by saw this critique as a manifestation of agricultural 10.5 per cent in nominal terms, in contrast with protectionism from European countries. Accord- an accumulated inflation rate during the same ing to the producers’ perception, it ignored the period of 42.1 per cent, which implied a fall in the superiority, in terms of social and environmental price, in real terms, of 22.2 per cent. On the other performance, of Brazilian ethanol in relation to hand, between June 2006 and December 2012 the other biofuels and the possibility of the expan- consumer price of ethanol rose by 56.3 per cent sion of sugarcane production in Brazil without in nominal terms, and a little more than 10 per generating any deforestation of tropical forests cent in real terms, according to Agência Nacional or compromising food products. At the same time, do Petróleo, Gás Natural e Biocombustíveis, the the international critique echoed with Brazilian governmental agency in charge of regulating the stakeholders, such as environmental NGOs and activities in the field of oil and gas (ANP 2016). those concerned with social and labour aspects. These denounced the labour conditions found As noted by Farina et al. (2013), the spread of vehi- in sugarcane production, the conflicts between cles with flex-fuel engines allowed the consumer Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

to decide the type of fuel to be used at the moment then most, people drove flex-fuel autos. of filling the tank at the pump, thereby increasing 209 consumers’ sensitivity to the relative prices of The domestic fuel price policy generated many ethanol and gasoline. Traditionally, drivers choose negative effects for the ethanol production sector. ethanol to fill their tanks when it cost less than These effects materialised at a moment when 70 per cent of the value of gasoline, since the fuel the impacts of the international crisis had hurt a extracted from sugarcane produces around 30 per significant part of the sector and produced a sharp cent less energy per liter, compared to gasoline. retraction in external demand for ethanol. Para- Hence, the sensitivity of the consumers to govern- doxically, the policy and its effects materialised ment policies affecting the relative prices of gaso- during a period when the international prices of line and ethanol grew substantially as more, and oil rose significantly (Box 13.5) (Kutas 2015a).

Box 13.5: Back to the crisis: since 2008 Context: The discovery of huge offshore oil deposits and the expectations created around the economic development effects of their exploitation led to a downgrade of the ethanol position in the ranking of the Brazilian energy policy priorities. Besides, the international economic crisis had severe repercussions on the financial/corporate structure of the firms operating in the sector.

Main policy instruments:

On the demand side:

◼◼ Use of CIDE-Fuel as a compensatory mechanism to prevent inflation, reducing its rates to zero in 2012; ◼◼ Gasoline price administration to reduce its impacts on inflation; ◼◼ New set of heavy incentives to the automotive industry as part of the new industrial policy; ◼◼ Gradual increase in the share of ethanol in mixed gasoline, reaching 27 per cent in March, 2015; ◼◼ From 2015 onwards, in the mid of a fiscal crisis, Brazilian central government and some of the states, as well, began to increase taxes (CIDE, PIS/COFINS, ICMS) on gasoline. On the supply side:

◼◼ BNDES introduced new lines that finance the stocking of ethanol and a support programme for the renovation and expansion of sugarcane plantations aimed at increasing their productivity. Results: Closing of 14 per cent of sugarcane processing mills, growth of only 3 per cent in the production of ethanol, while planted area rose 23 per cent, with a significant productivity decrease in the period 2008–2009 to 2014–2015. Despite the launch of new financing lines by BNDES, the amount of credits for the sector fell to their lowest level in almost ten years.

In response to the slowdown of the economy from Ethanol’s waning focus for governmental policies 2011 onwards, the government adopted a series of is also due to immense offshore oil deposits that measures aimed at stimulating industrial produc- Petrobras discovered of Southeast Brazil in 2006. tion and investments, giving special emphasis to The government presented the discovery as the the automobile sector. The automobile fleet grew beginning of a new economic and social develop- significantly, around 42 per cent between 2008 ment phase for Brazil. It also inspired a new cycle and 2013, but the proportion of ethanol in the of policies aimed at the import substitution of energy matrix fell considerably, reaching 33 per goods and services, now in the oil and gas sector. cent in 2012. The fall is even more significant for A vast political and business coalition was estab- light vehicles: in 2012 only 17 per cent of the total lished around the aim of exploring the recently energy consumed by them were ethanol (Farina discovered oil deposits and sharing the royalties et al. 2013). In turn, the volume of hydrous etha- and profits deriving from them. nol produced fell from 19 billion litres in 2010–11 to 12.7 billion litres in 2013–14 (UNICADATA n.d.-b). Green Industrial Policy - Concept, Policies, Country Experiences

In this scenario, it was not a surprise that the 8.7 thousand megawatts of installed capacity, 210 environmental implications of the policies giving equivalent to 6.2 per cent of the Brazilian electri- a high priority to fossil fuels were not part of the cal matrix (Pereira 2016). The fact that the govern- debate. Against this background, ethanol poli- ment had identified a relevant potential for the cies lost relevance in the public policy agenda. import substitution of machinery and equipment In contrast, during the same period, wind energy helps to explain the focus placed on wind energy gained attention in Brazil, with the support of by public policies aimed at stimulating invest- public financing administered by BNDES. The ment and local production. government held the first wind energy bid in the country in 2009 and, to encourage domestic The effects of these developments on the etha- production of wind energy generation equipment, nol industry were significant. Between the 2008- it raised the import tax for wind turbines from 09 and 2014–15 harvests, production of ethanol zero to 14 per cent, though the tariff remained grew only 3 per cent in contrast to the 87 per cent at zero for equipment with a superior genera- between 2003-04 and 2008-09. The area planted tion capacity (Oliveira 2012). In ten years, wind with sugarcane rose by 23 per cent in this period energy generation rose from 27.1 megawatts to in contrast to 65 per cent, between 2003-04 and 2008-09 (Figure 13.2) (UNICADATA n.d.-c).

Figure 13.2: Ethanol production in Brazil, 1980–2015 (Thousand m³)

30,000

25,000

20,000

15,000

10,000

5,000

1980/1981

1982/1983

1984/1985

1986/1987

1988/1989

1990/1991

1992/1993

1994/1995

1996/1997

1998/1999

2000/2001

2002/2003

2004/2005

2006/2007

2008/2009

2010/2011

2012/2013

2014/2013

Ethanol production - Brazil

Source: Adapted from UNICADATA (n.d.-b).

Investments in the complete production chains for ethanol and a support programme for the reno- sugarcane and ethanol were also affected. In 2015, vation and expansion of sugarcane plantations, BNDES credits for the production chain fell to their since their productivity had been falling. lowest level in almost ten years. Repercussions were felt on investments for the planting and Again, a side effect of the crisis has been a new renovation of sugarcane plantations, as well as on cycle of corporate restructuring and consolidation. industrial expansion projects and the construc- On the one hand, about one third of the compa- tion of new units, which continued to fall in 2015. nies in the sector faced difficulties as many mills Furthermore, this occurred despite BNDES having were in debt and operating at a loss (Duarte et al. introduced new lines of financing dedicated to the 2013). From 2008 onwards, nearly 15 per cent of sugar and ethanol production sector in 2010. These the sugarcane processing mills closed and many included a programme for financing the storage of others faced financial difficulties and high debt Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

(Kutas 2015b). In January 2016, there were 85 mills In more recent years, the federal government under judicial bankruptcy protection and many adopted measures favourable to ethanol, such 211 others negotiating with creditors to restructure as the increase in 2013 of the share of ethanol in their debts (Batista 2016). On the other hand, gasoline from 20 per cent to 25 per cent. In March investors with a background in agro-industry, 2015, the percentage was increased again, reach- oil, and chemicals entered the sugar and energy ing 27 per cent in the case of common gasoline. sector, bringing in new entrepreneurial expertise. Also in 2015, some Brazilian States increased This led an analyst of the sector to estimate that taxes on fossil fuels or reduced those on etha- in the first half of 2011, more than 70 per cent of nol while the federal government raised the the sector had good assets, capital structures and value of CIDE-Fuel and federal taxes on gasoline governance, operational performance and access and diesel. The sugar-producing sector pressed to high quality capital (Jank 2011). for a new and significant increase of CIDE for gasoline, but did not obtain this decision from Another positive development, from a strate- the federal government. Even more important, gic point of view, is that the ethanol producers from 2015 onwards, the federal government adopted an agenda for dealing with new chal- began to distance itself from its gasoline price lenges, including the concept of social/environ- management policy, and consumer gasoline mental certification. This initiative is part of the prices increased substantially. In January, as a sector’s reaction to the increased questioning by consequence, the average price of gasoline for some developed countries on the effects of sugar- consumers was 21.2 per cent higher than one year cane production on deforestation, food security earlier. This set of recent policy measures led to and working conditions, among other concerns increased consumption of ethanol in Brazil, but (Chiodi et al. 2014). did not foster a new cycle of investment. Invest- ments in capacity expansion in the ethanol sector A study done by UNICA, carried out in 2011, stated have been frozen since 2008. (UNICA 2012):

"the Better Sugarcane Initiative (Bonsucro), 3.3. THE PAISS PROGRAMME: PUSHING a global certification launched in July 2011, TECHNOLOGICAL INNOVATION which assessed the sustainability of products manufactured from sugarcane, is currently At the same time, there has been significant public the model of certification most used in Brazil investment aimed at fostering fundamental green covering 20 sugar energy companies and more innovation in the industrial sectors of the sugar- than 437,000 hectares of sugarcane, equiva- cane production chain. In 2011, BNDES and the lent to more than 1.7 per cent of the sugarcane public agency responsible for financing innova- planted in the world". tion projects, FINEP, joined efforts to launch what became ‘PAISS Industry’, a programme supporting Bonsucro certification system includes two stand- technological innovation in the sugarcane and ards (Bonsucro product standard and Bonsucro ethanol industries. The resources formally allo- mass balance chain of custody standard) and a cated by both institutions to the programme for certification protocol. The certificate holder is the period 2011–2014 amounted to 600 million. the mill, which holds the ultimate responsibility for compliance. The Bonsucro product standard The objectives of the programme were to support contemplates six principles referring to respect the development of second-generation bioetha- to the laws, human rights and labour standards, nol that exploits sugars converted from cellulosic management of input, production and processing materials; to develop high-biomass sugarcane, efficiencies to enhance sustainability. Each prin- energy cane; and to foster new gasification tech- ciple has its own criteria, indicators and stand- nologies required for treating sugarcane biomass ards defined for the different phases of production (CGEE 2016). (agricultural and industrial). In 2016, 124 out of the 400 members of Bonsucro are Brazilian compa- According to Milanez et al. (2015), ethanol produc- nies, including producers of sugar and ethanol tivity could increase up to 50 per cent with the (the large majority), multinationals operating in new technologies. CGEE (2016) concludes, “The the country in sectors supplying raw materials perspective to improve the potential yield of and machinery for the production of sugar and bioethanol to almost 25,000 litres per hectare is real.” As a consequence, global CO emissions ethanol, and associations of sugarcane suppliers 2 to the processing industries of the sector. and pressure on land and food prices could be significantly reduced (Nyko et al. 2010). CGEE Green Industrial Policy - Concept, Policies, Country Experiences

(2016) estimates that ethanol from energy cane assessments of the design and implementa- 212 alone could replace 10 per cent of the total gaso- tion have been published (Milanez et al. 2015; line consumed globally on less than 10 million Bomtempo 2014). In broad terms, these assess- hectares of land. ments highlight the main positive elements of the PAISS Programme as: The policy instruments used by PAISS were financial: credit, equity, non-reimbursable support for ◼◼ The high level of coordination between BNDES cooperation projects between private companies and FINEP, the public agencies responsible for and centers of research, and economic subven- the implementation of the programme, allow- tions. It was operationalised through a call for ing for synergies between different instru- projects to be presented by the companies, which ments and modalities of financial support were then selected on their merits. In February ◼◼ The building and operation of three second 2014, BNDES announced that this PAISS Industry generation ethanol plants, with two at programme had disbursed the amount of US$ 1 commercial scale, and taking the projects billion to support 35 innovation projects in 57 beyond lab or pre-operational levels companies, with a total investment that should ◼◼ The incentives for cooperation schemes reach US$ 1.4 billion. The production of cellulosic between companies and between companies ethanol for 2015 in Brazil was estimated at 140 and research centers, through specific mech- million litres (Milanez et al. 2015). anisms of financing and/or non-reimbursable funding. In 2014, a new phase of the PAISS programme began, with the launching of ‘PAISS Agriculture’ As to the weaknesses of the programme, Milanez that targets innovation and development of new at al. (2015) criticise that the incentives target technologies in the agricultural sectors and activ- expansion of supply only, in contrast to policies ities of the production chain. At its launching, that also cover demand-side incentives. Accord- the new initiative was allocated around US$ 620 ing to Bomtempo (2014), other problematic million in public resources available for the same aspects of the programme include poorly defined financing modalities that were operationalised criteria to select the projects and the absence of through PAISS Industry. mechanisms to effectively monitor implementation of the projects, as well as instruments that The main focus of PAISS Agriculture is innovation condition the continuation of government support in areas directly relating to genetic improvements on performance. and the enhancement of productivity of sugarcane in Brazil. However, it also targets industrial Though these programmes are significant in segments producing machines and equipment fostering green innovation in Brazilian etha- for agriculture, emphasizing innovation geared at nol, Brazil was a latecomer in the development fostering the use and diffusion of precision agri- of second-generation ethanol (Nyko et al. 2010). culture techniques. In 2014, 35 business plans had According to Kutas (2015b), this can be attrib- been approved with the framework of PAISS Agri- uted to the lack of pressure to go beyond the first culture, totalling US$ 849 million for the period generation because of land availability, no need from 2014 to 2018. to trade off between food production and biofuels and the good environmental and climatic perfor- Although it is certainly too early to assess the mance of first generation ethanol. success of the PAISS Programme, some initial

4. POLICY LESSONS Due to the long history of ethanol policies, Brazil 25 per cent of global production and 20 per cent of now has one of the most successful programmes global exports. In 2014, sugarcane was the second in the world for the use of renewable energy in the principal source in the Brazilian energy matrix, transport sector. No other country in the world with 15.7 per cent of the total, compared to 39.4 has a fleet of almost 20 million light vehicles that per cent for oil and gas (Kutas 2015b). can use any combination of gasoline and sugarcane ethanol the consumer wants to use (Farina As described in the previous sections, a wide set et al. 2013). Despite recent difficulties, Brazil has of policy instruments, including investments, maintained its position as the second largest price subsidies, and demand incentives, were producer of ethanol in the world, accounting for deployed in the 1980s to promote suppliers in the agricultural and industrial sector, especially in Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

sectors producing capital goods such as agricul- characteristics of this industry, at the supply as tural machinery and equipment for sugarcane well as at the demand level. From the point of 213 processing, ethanol production and stocking view of its final use, ethanol competes with fossil fuel and the demand for it depends basically on While, in the last decades, the producers of agri- the relative prices of the two products, besides the cultural machinery have benefited from the regulation related to the gasoline mix with ethanol. strong performance of the Brazilian agricultural sector as a whole–and particularly from the Industrial policies aimed at fostering the demand commodities boom of the past decade–compa- for ethanol in Brazil combined incentives for the nies engaged in the production of industrial development of ethanol-powered cars and later capital goods for ethanol mills suffered from the for flex-fuel cars. Associated price controls or use ups and downs of ethanol production and of the of a specific tax, CIDE-Fuels, kept relative prices associated policies. As a consequence, industrial of ethanol attractive to consumers. The launch- spillovers arising from ethanol policies to supply ing of flex-fuel cars, stimulated by tax and credit sectors, although relevant, were limited by the subsidies, created a potentially large and grow- volatility of the demand that the ethanol-producing market for ethanol and helped to increase ing firms had for dedicated capital goods. the price elasticity of the demand for ethanol in Brazil (Farina et al. 2013). Consumption of ethanol In principle, the pattern of government interven- depended strongly on the price of gasoline. tion applied to the ethanol sector and the instruments mobilised to foster its emergence do not An additional factor that played a relevant, differ from the ones that guided the import-sub- although secondary, role in stimulating ethanol stitution industrialisation that Brazil went production is the external demand. Although through in the previous decades: High tax and industrial policies can do little to increase foreign credit subsidies combined with instruments and demand, trade policies can help. Brazil engaged in incentives adapted to the specific features of the trade negotiations with developed countries in the sector. The direct and indirect subsidies mobi- mid-1990s at the Doha Round and other venues, lised by the ethanol policies have been substan- with the liberalisation of trade barriers to etha- tial compared to the size of the Brazilian economy, nol as one of its top priorities in terms of market but were significantly reduced at the end of the access demands. Furthermore, the country signed 1980s, forcing the industry to consolidate (Meyer a memorandum with the US to define an agenda et al. 2012). for disseminating the use of ethanol around the world, through the establishment of technical There is no doubt that the ability to foster the standards. None of these initiatives resulted in emergence of new producers in the Southeast, relevant achievements and ethanol exports still the most developed region of Brazil, and to attract face relevant barriers to enter foreign markets. investors and large companies from other activities to the ethanol industry has been one of the On the supply side, ethanol competes with sugar main successes of the ethanol policies. This was for sugarcane as a raw material. Furthermore, essential for the technological and organizational being an agricultural commodity, sugarcane modernisation of the firms acting in the sector production and productivity depend on climate and ultimately for their survival despite the ups instabilities, soil conditions and other variables. and downs in its performance and the swings in While processing facilities exist that produce the public policies. It created the conditions for ethanol exclusively, the large majority of produc- reducing the production cost of ethanol, which ers process the sugarcane to produce sugar and decreased by 65 per cent from 1976 to 2005 for ethanol, in a mix that depends on relative prices a number of reasons, including innovation in of both products. Therefore, while on the demand research and development, economies of scale side the incentives to produce ethanol depend and the effects of learning by doing (Meyer et al. strongly on the domestic policies related to 2012). The crisis that the ethanol sectors under- fossil fuels, on the supply side these incentives went from 2008 was largely due to a combination are strictly conditioned by the relative prices of policies geared at controlling the consumer of ethanol and sugar, and so by the fluctuations prices for gasoline in the domestic market, in a and trends in the international prices of sugar. period of high international oil prices, with the These specific features of ethanol production reduction of the value of CIDE-Fuels incurred on tend to create a structural volatility in its supply, gasoline sales, reaching zero in July 2012. leaving it vulnerable to the vagaries of policies and markets for sugar and for fossil fuels. With The assessment of ethanol industrial policies in new research on energy cane varieties geared Brazil should consider the nature and specific Green Industrial Policy - Concept, Policies, Country Experiences

specifically for ethanol production, this linkage blending of gasoline and ethanol and using CIDE- 214 between ethanol and sugar may be broken in the Fuel to influence the price ratio in favour of etha- future, thereby stabilising the ethanol market. nol consumption. Especially in a period of low international oil prices, these instruments seem Ethanol industrial policies have to deal with these to be best fitted to drive the performance of this sector-specific characteristics if they aim to foster industry in the next years, while at the same time the development of the sector, with or without avoiding the introduction of artificial distortions environmental motivations. During the period in the management of gasoline’s domestic prices. of strong government intervention, a wide set of policy instruments were used. As already stressed, In recent years under the PAISS programme, the main instruments were subsidised credit lines Brazil has moved in a different direction, away for investment and production, price adminis- from consumption mandates and demand-driv- tration and arbitration between sugar and etha- ing pricing schemes, toward supply-side inter- nol, and purchase guarantees. Except for BNDES’ ventions designed to boost the productivity credit lines for investments, which can include a of the agricultural resource. While there are subsidies component, and the recent support for encouraging signs, it is too early to tell whether innovation in second-generation ethanol, most of this new approach ultimately will be successful. these instruments have been phased out. This is partly because we have too few years of data, and partly because the commercialisa- Actually, the subsidy component of the etha- tion of second-generation ethanol is still in its nol industrial policy has lost relevance. As more early stages. Success in such a dynamic field, so interventionist policies tend to be unfeasible for crowded with international competitors, is inher- fiscal reasons, government policies are likely to ently uncertain. concentrate on two instruments: the mandatory

5. CONCLUSIONS Reference was made, in the introduction, to the pasture, while its indirect impacts only contrib- ambivalence of the ethanol policies in Brazil, uted marginally to deforestation. Other authors when assessed according to green criteria. disagree with this view, pointing to the indirect impacts on the Amazon region of the expansion On one hand, these policies can be considered of sugarcane production. These discussions will green, to the extent that, irrespective of their continue, as sustainability becomes a main- rationale and motivations, they actually promoted stream concern (Box 13.3). changes in the structure of the energy matrix, in a direction compatible with the objectives of On the other hand–and this is the ambivalence– sustainable development. Nowadays, more than the implementation of ethanol policies reveals a 15 per cent of Brazil’s energy matrix has sugar- narrow perception on the part of the government cane as its source and the balance of greenhouse of their green dimension. This appears clearly in gas emissions from the production and consump- the first phases of Proálcool, during the military tion of ethanol is clearly favourable compared regime, when the expansion of sugarcane produc- to that of gasoline and other biofuels. Therefore, tion did not consider the environmental effects from the point of view of their aggregate result, and the social consequences. During the initial industrial policies targeting the ethanol produc- years of the Programme, the expansion of sugar- tion, besides being successful as import substitu- cane production principally occurred through tion policies, can also be considered green. the extension of the planted areas, including in regions previously occupied by other economic When the entire life cycle of ethanol produc- activities, such as coffee or livestock raising tion is considered, the picture is more nuanced. (Oliveira 1982). In this period, the working condi- Specifically, ethanol development policies in tions in the production and harvesting of sugar- Brazil have been criticised for both their direct cane, activities that involved a large number of and their indirect consequences for land use, temporary workers due to the low level of mech- specifically their effect on food production and anisation, were far from adequate even when Amazonia deforestation. According to various assessed according to the Brazilian legislation. evaluations, this critique is largely unfounded, since the direct impact of the expansion of the But the narrow perception of the green dimen- area occupied by sugarcane was on degraded sion of the ethanol policies is also made explicit Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

by a more recent episode: the shift in domestic and climate goals. As a matter of fact, until the policies that hurt the sector in the late 2000s. It is advent of the relatively recent PAISS programme, 215 striking that the recent crisis in the ethanol sector, the green dimension has never played a major mostly driven by domestic policy choices related role in the design of ethanol development poli- to the price of fossil fuels, followed a period when cies in Brazil. Conceived to promote ethanol as ethanol was praised in Brazil and in other coun- a substitute for fossil fuels during the oil crisis tries for its green attributes. At this time, Brazil of the 1970s, these policies ended up linking the supported the use of ethanol as a biofuel, while ventures of this sector to the volatility of the pushing for the inclusion of ethanol in the list domestic policies for the fossil fuels sector. Thus, of environmental goods during the Doha Round PAISS in this sense represents a distinct depar- negotiations at the WTO. It was impossible to ture from previous policies. imagine that a few years after this period of ethanol-euphoria the sector would go through a deep Finally, the ethanol policies have been oriented crisis that is still ongoing. only partly towards industrial upgrading and technological learning. The main policy focus was If ethanol policies in Brazil had been linked more on increased agricultural productivity. Some tech- explicitly to environmental objectives, the sector nological spillovers, however, are worth mention- would have been a strong candidate for receiving. These include the development of innovative ing continued support from the government. flex-fuel engines; processing of ethanol-based Government policies, however, generated a high chemicals; and sophisticated sugar and ethanol level of uncertainty and unpredictability that has processing plants, which are likely to receive a damaged the sector and its performance. The technological boost with the shift to processing past changes of ethanol policies in Brazil indi- cellulose-based ethanol. cate how vulnerable these can be with regard to economic cycles and government priorities that are not directly linked to environmental Green Industrial Policy - Concept, Policies, Country Experiences

216 REFERENCES ANP (Agência Nacional do Petróleo, Gás Natural International Centre for Trade and Sustainable e Biocombustíveis). (2016). Historical series of Development (ICTDS) (2007). US, Brazil Agree To prices and fuel sales margins. Retrieved from Cooperate On Biofuels - But Leave Out Tariffs. anp.gov.br/wwwanp/precos-e-defesa/234-pre- Bridges, Volume 11, No.9, 1/03/2007. cos/levantamento-de-precos/868-serie-his- Jank, M. S. (2011). Etanol-Novo ciclo de cresci- torica-do-levantamento-de-precos-e-de-mar- mento. In Correio Braziliense, Brasília, 27. gens-de-comercializacao-de-combustiveis Koizumi, T. (2014). Biofuels and food security: Biofuel Assad, E. D., Pinto, H. S., & Sousa, Z. (2012). Estudo impact on food security in Brazil, Asia and major sobre os impactos ambientais do programa producing countries. Springer Briefs in Applied sucroenergético brasileiro. Sciences and Technology. iii-iv. Batista, F. (2016). Abengoa Bioenergia reestrutura Kutas, G. (2015a). New bio-refineries in Brazil. Future dívidas de usinas no Brasil. Valor Econômico, production potential and market development. published 27/01/2016. Advanced Biofuels Conference, Stockholm, BNDES & CGEE. (2008). Sugarcane-based bioetha- 16–17/09/2015. nol: energy for sustainable development. Rio de Kutas, G. (2015b). US$ 50 por barril: o fim do sonho do Janeiro, BNDES. etanol? Portal ÚNICA, 10 de junho. Bomtempo, J. V. (2014). Bioeconomia em construção Meyer, D, Mytelka, L., Press, R., Dall’Oglio, E.L., de II – Os grants e subvenções às empresas: Souza Jr., P.T., & Grubler, A. (2012). Brazilian comparando o Biomass Program do DOE e o Ethanol: unpacking a success story of energy PAISS do BNDES / FINEP. Grupo De Econo- technology innovation. In Grubler A., Aguayo, F., mia Da Energia. Blog Infopetro, posted in Gallagher, K.S., Hekkert, M., Jiang, K., Mytelka, L., Biocombustíveis, 30/06/2014. Retrieved from Neij, L., Nemet, G., & Wilson, C., Historical case https://infopetro.wordpress.com/2014/06/30/ studies of energy technology innovation, chapter bioeconomia-em-construcao-ii-os-grants-e- 24, Cambridge University Press. subvencoes-as-empresas-comparando-o-bio- Milanez, A.Y., Nyko, D., Valente, M.S., Sousa, L.C. mass-program-do-doe-e-o-paiss-do-bndesfinep/ de, Bonomi, A., Jesus, C.D.F. de, Chagas, M.F., CGEE. (2016). Second Generation Sugarcane Bioen- Watanabe, M. D. B., Rezende, M. C. A. F., Cavalett, ergy & Biochemicals: Advanced Low-Carbon O., Junqueira, T. L., & Gouveia, V.L.R de (2015). De Fuels for Transport and Industry. Final Version. promessa a realidade: como o etanol celulósico Brasília: Center for Strategic Studies and Manage- pode revolucionar a indústria da cana-de-açú- ment (CGEE). car – uma avaliação do potencial competitivo e Chiodi, L., Nassar, A., Antoniazzi, L., & Zambianco, W. sugestões de política pública. BNDES Setorial 41, (2014). The Business Case for Mill Compliance Rio de Janeiro, (41). with and Certification to the Bonsucro Production Ministry of Mines and Energy (MME). (2016). Standard. Agroicone, November. Participação de renováveis na matriz De Sa, S. A., Palmer, C., & Di Falco, S. (2013). Dynamics energética brasileira é três vezes superior of indirect land-use change: empirical evidence ao indicador mundial. Published 01/12/2016. from Brazil. Journal of Environmental Economics Retrieved from www.mme.gov.br/web/ and Management, 65(3), 377–393. guest/pagina-inicial/outras-noticas/-/ Duarte, C.G., Gomes, F., Soriano, E., & Malheiros, T.F. asset_publisher/32hLrOzMKwWb/content/ (2013). Ensaio sobre a evolução da proteção ambi- participacao-de-renovaveis-na-matriz-energet- ental no setor sucroenergético. Essay on the envi- ica-brasileira-e-tres-vezes-superior-ao-indica- ronmental protection evolution in the sugarcane dor-mundial industry. Revista Brasileira de Ciências Ambien- Moreira, M. M. R. (2008). Análise prospectiva do tais, 29. 1–16. padrão de expansão do setor sucroenergético Farina, E., Rodrigues, L., & Sousa, E. D. (2013). A brasileiro: uma aplicação de modelos probabilísti- Política de Petróleo e a Indústria de Etanol no cos com dados georeferenciados (Doctoral disser- Brasil. Interesse Nacional, 64–75. tation, Universidade de São Paulo). Goldemberg, J. (2009). The Brazilian Experience with Nassar, A. M., Rudorff, B. F., Antoniazzi, L. B., Aguiar, Biofuels (Innovations Case Narrative). Innova- D. A. D., Bacchi, M. R. P., & Adami, M. (2008). tions, 4(4), 91–107. Prospects of the sugarcane expansion in Brazil: impacts on direct and indirect land use Goldemberg, J., Coelho, S. T., & Guardabassi, P. (2008). changes. In Zuurbier, P., & van de Vooren, J. (eds), The sustainability of ethanol production from Sugarcane ethanol: contributions to climate sugarcane. Energy policy, 36(6), 2086–2097. change mitigation and the environment, 63–93. Nastari, P. (2012). The Brazilian Experience with Biofuels. São Paulo: Fórum das Américas. Chapter 13 Ethanol Policy in Brazil: A Green Industrial Policy by Accident?

Nyko, D., Garcia, J. L. F., Milanez, A. Y., & Dunham, F. B. (2010). A corrida tecnológica pelos biocom- 217 bustíveis de segunda geração: uma perspectiva comparada. BNDES Setorial, Rio de Janeiro, (32), 5–48. OECD & FAO (2007). Outlook 2007–2016. Organization for Economic Co-operation and Development, & Food and Agriculture Organization of the United Nations. Oliveira, A. de (2012). Transição para a economia verde: a agenda energética. Breves CINDES 67. Oliveira, F. B. de (1982). As disfunções do Programa Nacional do Álcool em função de excessiva ênfase na cana de açúcar. Revista de Adminis- tração Pública 16(3), FGV. Pereira, R. (2016) – Energia eólica ganha competitivi- dade, O Estado de São Paulo, 07/03/2016. Pereira, W. S. (2015) - A participação do Estado no fomento ao etanol como uma oportunidade estratégica de desenvolvimento econômico: as políticas federais de estímulo ao etanol no Brasil e nos EUA. (Doctoral dissertation, Universidade Federal do Paraná). Szwarc, A. (2002). A opção pelo veículo flex fuel. Portal ÚNICA, 1/09/2002. ÚNICA. (2012). Bonsucro já é certificação mais usada para comprovar sustentabilidade da indústria da cana no Brasil. Retrieved from www. unica.com.br/noticia/36949486920333453814/ bonsucro-ja-e-certificacao-mais-usada-para-comprovar-sustentabilidade-da-indus- tria-da-cana-no-brasil/ UNICADATA (n.d.-a). Estimated Brazilian automobile and light vehicle fleet (Otto cycle). Retrieved from www.unicadata.com.br/listagem.php?idMn=55 UNICADATA (n.d.-b). Hydrous ethanol production, 1980/1981 - 2016/2017. Retrieved from www.unicadata.com.br/ historico-de-producao-e-moagem.php?id- Mn=31&tipoHistorico=2&acao=visualizar&id- Tabela=1883&produto=etanol_hidratado&sa- fraIni=1980%2F1981&safraFim=2016%2F2017&estado=RS%2CSC%2CPR%2CSP%2CRJ%2C- MG%2CES%2CMS%2CMT%2CGO%2CD- F%2CBA%2CSE%2CAL%2CPE%2CPB%2CRN%2CC- E%2CPI%2CMA%2CTO%2CPA%2CAP%2CRO%- 2CAM%2CAC%2CRR UNICADATA (n.d.-c) Various entries for database access. Retrieved from www.unicadata.com.br/historico-de-area-ibge. php?idMn=33&tipoHistorico=5 Walter, A., Dolzan, P., Quilodrán, O., Garcia, J., Da Silva, C., Piacente, F., & Segerstedt, A. (2008). A sustainability analysis of the Brazilian ethanol. Report Submitted to the United Kingdom Embassy, Brazil. Green Industrial Policy - Concept, Policies, Country Experiences

218 ABOUT THE AUTHORS

Tilman Altenburg is Head of the Department of Verena Balke is a recipient of the Carlo-Schmid- Sustainable Economic and Social Development Programme, working on green economy and at the German Development Institute, the govern- green industrial policy with UN Environment. ment-funded think tank for development policy in Her academic background is in Sustainability Germany. He received his doctorate in Economic Economics, Management, and East-Asian Studies. Geography in 1991. Since 1986 Mr. Altenburg has done empirical research on economic development in Latin America, Asia and Africa, with a Richard Bridle is a Senior Policy Advisor and Lead focus on competitiveness, industrial and inno- in Renewable Energy at the International Institute vation policy. He has published about 100 papers for Sustainable Development (IISD) supporting on these issues, and he is regularly advising the the Global Subsidies Initiative. His research inter- German government and international agencies. ests include efficient design of renewable energy support policies, biofuels, green industrial policy and energy subsidies. Richard has a background Stefan Ambec is INRA Research Professor at in the renewable energy industry with experience Toulouse School of Economics in France where he in policy analysis, project management, develop- teaches environmental economics and policy. His ment and procurement. research focuses on the choice of policy instruments in environmental policy, how it impacts firm’s strategy and consumer’s behaviour, about Liesbeth Casier is a Project Officer in IISD’s their efficiency and fairness properties. He has Economic Law and Policy program. Liesbeth is a worked on water use, air pollution, climate change, Sustainable Public Procurement specialist who common-pool resources, electricity provision has done consulting work with the Inter-Ameri- and pest resistance. He is currently Editor for can Network for Government Procurement, the Resource and Energy Economics and visiting Inter-American Development Bank and the Inter- professor at the University of Gothenburg. national Development Research Centre. She has served as an Environment and Trade Consult- ant with UN Environment, where she focused Claudia Assmann leads work on green econ- on research at the intersection of international omy and green industrial policies at UN Envi- economic law, trade policy and the environment. ronment, both at global and country level. In her former role with the German Development Coop- eration, she was an advisor on climate finance to Aaron Cosbey is Senior Associate at the Inter- governments and country offices. She has also national Institute for Sustainable Development, specialised in climate change adaptation and where he has worked on trade and investment climate-resilient development, while working law and policy since 1991. He has consulted for a with the World Bank and the UN system. wide range of governments, inter-governmental organizations, development banks and UN agencies, and he has published extensively in research Sandra Averous Monnery is a Programme Officer areas including trade law and investment law as at UN Environment. Since June 2015, she has been they relate to sustainable development, invest- in charge of the Responsible Production Portfolio, ment policy, trade policy, green industrial policy, engaging all actors, especially the private sector and climate change law and policy. and experts towards sustainable value chains and circular economy, in sectors related to chemicals such as textiles and plastics. Sandra is a development economist, and worked previously at UNESCO and the UN Statistics Division. About The Authors

Hans Eichel is the former German Finance Pedro da Motta Veiga is the Director of CINDES— 219 Minister and co-founder of the G20. He was Centre for Studies in Integration and Develop- Chairman of the G7 in 1999 and Chairman of the ment, a think tank based in Rio de Janeiro and G20 in 2004. Eichel currently leads the expert a Partner at Ecostrat Consultores. He works group on sustainable structural development on trade policies and negotiations as well as for the Friedrich Ebert Foundation, the world’s on industrial policy issues and has acted as a oldest and largest foundation to promote democ- consultant to different international organiza- racy and political education. He is also leading tions (OECD, The World Bank, IADB etc). the Advisory Committee of the think tank Green Budget Germany. René Kemp is Professorial fellow at the Maas- tricht Economic and Social Research Institute Michela Esposito is working as Junior Economist on Innovation and Technology at United Nations in the ILO Research Department. She holds a PhD University-MERIT and Professor of Innovation and in Economics and Public Finance from the Univer- Sustainable Development at the Centre for Inte- sity of Genoa. Before joining the ILO, she worked as grated assessment and Sustainable development Research Economist at WTO. Her research inter- (ICIS) of Maastricht University. He is well-known ests include climate change, international trade, for his work on eco-innovation, environmental development economics and labour. policy and transition management. His research interests are environmental policy, innovation for sustainable development, sustainability transitions, Steve Evans is the Director of Research in Indus- urban labs and humanisation of the economy. trial Sustainability in the Department of Engi- neering at University of Cambridge. He holds a PhD from the University of Bath. Prof Steve Evans’ Babette Never is a researcher in the Department research seeks a deep understanding of how of Sustainable Economic and Social Development industry develops solutions that move us towards at the German Development Institute. She holds a sustainable future. a PhD in political science from the University of Hamburg. In the past four years, her research focused on various aspects of energy transitions in Kaidong Feng is Associate Professor at the developing countries. She currently leads an inter- School of Government at Peking University. disciplinary research group that analyses carbon He obtained a Master’s degree in Management consumption patterns of the new middle classes. Science & Engineering from Tsinghua Univer- sity and a PhD in Science & Technology Policy Research from SPRU, University of Sussex. His Emilio Padilla is an Associate professor in the research is focused on technological learning Department of Applied Economics at the Univer- and industrial catching-up of developing coun- sitat Autónoma de Barcelona and coordinator of tries, science and technology policy and Chinese the PhD program in Applied Econonomics. His industrialisation history since 1949. research is focused on environmental economics, public economics, ecological economics, intergen- erational equity, sustainable development and the Alexander Haider is a PhD student at the New economics of climate change. School for Social Research, New York, in the Department of Economics. He focuses on macroeconomics, environmental economics, and labor economics. Green Industrial Policy - Concept, Policies, Country Experiences

220 Anna Pegels is Senior Researcher at the German Daniel Samaan is a Senior Economist and Development Institute (DIE). She received her Researcher at the International Labour Organ- doctorate in economics in 2007 and has since ization (ILO) in Geneva. He holds a PhD from the then focused on the interconnections between New School for Social Research in New York. His economic development and climate change research focuses on globalization and its impact on mitigation. At DIE, Anna has led several research labour markets as well as on labour markets effects projects on ecologically sustainable development. of climate change. He is also part of a research Her current research focuses on the economic group at the ILO that focuses on the Future of Work and political aspects of climate change mitigation, and analyses the impacts of digitalization and arti- in particular the transition to clean energy. ficial intelligence on labour markets.

Sandra Polónia Rios is a director at Centre for Kai Schlegelmilch is a political economist and Studies in Integration and Development and President of Green Budget Germany and Vice- partner at Ecostrat Consultores. She holds a Chair of the Advisory Committee of Green Budget Master’s degree in Economics. Her areas of exper- Europe. He is an expert on Environmental Fiscal tise include international trade negotiations Reforms and has worked for 24 years in research, and trade policy. She teaches at the Pontifical policy and government institutions, focusing Catholic University of Rio de Janeiro and acts on energy, renewable and climate policies and as a consultant at the National Confederation of required fiscal incentives. He also advised several Industry of Brazil. Asian Governments regarding their implementation. Since 1999, he worked for the German Federal Ministry for the Environment, Nature Liazzat Rabbiosi is a Programme Officer for the Conservation, Building and Nuclear Safety. Montreal Protocol Ozone Action Programme at the UN Environment. She is responsible for work related to compliance assistance to 13 countries Willi Semmler is the Arnhold Professor of Inter- in South Asia region. Prior, she worked on capac- national Cooperation and Development at New ity building for SMEs including the eco-innova- School, New York. In 1976, he earned a PhD from tion project implemented in 9 countries around Free University of Berlin. He specialises in Empir- the world. She holds a Master’s degree in Envi- ical Macroeconomics, Macroeconomics of the US ronment and Development from London School and EU, Financial Markets, Economics of Climate of Economics and a Master of Science in Envi- Change, Business Cycles and Macro Dynamics. ronmental Studies from Strathclyde University in Scotland. Qunhong Shen is Associate Professor at the School of Policy and Management at Tsinghua Dani Rodrik is an economist whose research University. She received her master degree from covers globalization, economic growth and devel- Southeast University and Ph.D. from Nanjing opment, and political economy. He is the Ford University. Shen’s areas of research include Foundation Professor of International Political innovation and knowledge management, human Economy at Harvard's John F. Kennedy School resource management, and organization theory. of Government. He was previously the Albert Meanwhile she also carries out studies on O. Hirschman Professor in the School of Social Chinese transition of social regulation. Science at the Institute for Advanced Study in Princeton from 2013 to 2015. Professor Rodrik is currently President-Elect of the International Economic Association. Chapter 13 About The Authors

Georgeta Vidican Auktor is an associate 221 researcher at the German Development Insti- tute and development policy consultant. She holds a PhD in international development and regional planning from the Massachusetts Insti- tute of Technology. She works on issues related to industrial policy, energy policy, innovation systems, science and technology policy, especially as they relate to the process of transition to a low-carbon economy in developing countries. She has published extensively in peer-reviewed journals, taught at university level, and worked with development policy organizations to advise both development cooperation agencies and national governments.

Peter Wooders is IISD's Group Director for ‘Energy’. He has 20 years of experience across the energy sector. He currently contributes to various IISD’s programmes, including Trade and Climate Change, notably Border Carbon Adjustment and the GHG impacts of possible Environmental Goods and Services agreements, the Global Subsidies Initia- tive, Post–2012 Architecture of GHG Agreements, Carbon Markets and Climate Change Adaptation. Initially trained as an engineer, Peter first worked in technology research. He then spent 15 years as an Energy and Environment consultant. THE GERMAN DEVELOPMENT INSTITUTE / DEUTSCHES INSTITUT FÜR ENTWICKLUNGSPOLITIK (DIE) The German Development Institute / Deutsches Every Monday, the German Development Institute Institut für Entwicklungspolitik is one of the / Deutsches Institut für Entwicklungspolitik (DIE) leading Think Tanks for development policy comments the latest news and trends of develop- world-wide. It is based in the UN City of Bonn. DIE ment policy in The Current Column. builds bridges between theory and practice and works within international research networks. The German Development Institute / Deutsches The key to DIE’s success is its institutional inde- Institut für Entwicklungspolitik (DIE) is pendence, which is guaranteed by the Institute’s headed by Dirk Messner (Director) and Imme founding statute. Since its founding in 1964, DIE Scholz (Deputy Director). DIE is member of the has based its work on the interplay between Johannes-Rau-Forschungsgemeinschaft (JRF). research, consulting and training. These three areas complement each other and are the factors responsible for the Institute’s distinctive profile. THE PARTNERSHIP FOR ACTION ON GREEN ECONOMY (PAGE) The Partnership for Action on Green Economy conditions to meet their sustainability commit- brings together the expertise, convening power ments, in particular the Sustainable Development and networks of five UN agencies – UN Environ- Goals and the Paris Agreement, through inclusive ment, the International Labour Organization, the green economy approaches. UN Development Programme, the UN Industrial Development Organization and the UN Institute PAGE is a direct response to the Rio+20 Decla- on Training and Research – to support coun- ration, The Future We Want, which called upon tries and regions in addressing one of the most the United Nations system and the international pressing challenges of the 21st century: trans- community to provide assistance to interested forming their economies and financial systems countries in developing, adopting, and imple- into drivers of sustainability and social equity. menting green economy policies and strategies. The Partnership supports nations and regions Working closely with national governments, the in reframing economic policies and practices private sector, and civil society, PAGE offers a around sustainability to foster economic growth, comprehensive, coordinated, and cost-effective create income and jobs, reduce poverty and package of analytical support, technical assis- inequality, and strengthen the ecological founda- tance, and capacity building tools and services. tions of their economies. PAGE provides technical and financial assistance PAGE works to build capacity within partner coun- in 13 countries as of 2017. tries so that they are able to provide favourable “The world is facing profound and inter-related “This publication highlights how countries across challenges, from increasing inequalities and sluggish the world are using green industrial policies to growth to crowded cities, pollution and climate promote higher productivity and competitiveness, change. The current development and growth while increasing resource efficiency and decoupling models are unsustainable and require a structural economic growth from environmental degradation. As change geared towards a big environmental push. such, it represents an important addition to the green This report must be an ongoing reference for policy economy knowledge base and our common efforts to makers seeking to foster domestic development by achieve the Sustainable Development Goals.” embracing more sustainable patterns of production ACHIM STEINER and consumption.” Administrator, United Nations Development Programme, and ALICIA BÁRCENA Head of the United Nations Development Group Executive Secretary of the Economic Commission for Latin America and the Caribbean (ECLAC)

"Excellent coverage of both theory and practical "Industrialization in developing countries is essential applications, anchored by country studies. This book for generating higher productivity jobs for eradicating shows the next step in the harmonious co-evolution of poverty, eliminating hunger and enhancing well-beings, linked socio-economic and ecological systems, based which are UN’s first three Sustainable Development on resource-efficient, industrial ecologies of the 21st Goals. Industrialization, however, may increase the century.” energy intensity and carbon dioxide emissions, unless MOHAN MUNASINGHE governments adopt an innovative, green industrial policy. This Report is a timely intellectual contribution Vice Chairman of the United Nations Intergovernmental with a practical guidance for the governments around Panel on Climate Change (IPCC) in Geneva, co-winning the the world to achieve green industrialization.” 2007 Nobel Peace Prize JUSTIN YIFU LIN Director, Center for New Structural Economics, former Chief Economist, the World Bank

PAGE also gratefully acknowledges the support of all its funding partners:

◼◼ European Union ◼◼ Federal Ministry for the Environment, Nature Conservation, Building and Nuclear Safety, Germany ◼◼ Ministry for Foreign Affairs of Finland For further information: ◼◼ Norwegian Ministry of Climate and Environment ◼◼ Ministry of Environment, Republic of Korea PAGE Secretariat ◼◼ Government Offices of Sweden UN Environment, Resources & Markets Branch, ◼◼ Swiss Confederation, State Secretariat for Economic 11–13 Chemin des Anémones, CH-1219, Affairs (SECO) Chatelaine, Geneva, Switzerland

un-page.org [email protected] un-page.org/newsletter

@PAGExchange /GreenEconomyUNEP Job No: DTI/2150/GE